August 2004
                         NSF 04/08/EPADWCTR
Environmental Technology
Verification Report

Removal of Arsenic in Drinking Water

Kinetico Inc. and Alcan Chemicals
Para-Flo™ PF60 Model AA08AS
with Actiguard AAFS50
              Prepared by
           NSF International
        Under a Cooperative Agreement with
 <>EPA U.S. Environmental Protection Agency

-------
        THE ENVIRONMENTAL TECHNOLOGY VERIFICATION
                                    PROGRAM
       SEPA
                                  ET
                          LA1V1   S\

                          
-------
NSF International (NSF), in cooperation with the EPA, operates the Drinking Water Systems (DWS)
Center, one of seven technology areas under the ETV Program. The DWS Center recently evaluated the
performance of an adsorption media filter technology for the reduction of arsenic in drinking water. This
verification statement provides a summary of the test results for the Kinetico Inc. and Alcan Chemicals
Para-Flo™ PF60 Model AA08AS with Actiguard AAFS50  System. Gannett Fleming,  Inc.,  an NSF-
qualified field testing organization (FTO),  performed the  verification  testing. The  verification report
contains a comprehensive description of the test.

ABSTRACT
Verification testing of the Kinetico he. and Alcan Chemicals Para-Flo™ PF60  Model AA08AS  with
Actiguard AAFS50 arsenic adsorption media filter system  was conducted at the  Orchard Hills Mobile
Home Park (MHP) Water Treatment Plant (WTP) in Carroll  Township, Pennsylvania from April 22, 2003
through October 28, 2003.   The source  water was  untreated groundwater  from one of the MHP's
groundwater supply wells.  The source water, with an average total arsenic concentration of 14 |jg/L and a
pH of 7.6, received no treatment or chemical addition prior to entering the treatment unit.  When operated
under the manufacturers' specified site conditions at a flow  rate of 1.9 gpm ±0.1 gpm, the Kinetico Inc.
and Alcan Chemicals Para-Flo™  PF60 Model AA08AS with Actiguard  AAFS50  arsenic adsorption
media filter  system removed arsenic from the feed water to less than the detection  limit (2  |Jg/L) for
approximately 8,000 bed volumes, to less than 10 |jg/L for approximately 25,000 bed volumes, and to
less than the predetermined test endpoint (11 |Jg/L) after approximately 2,350 hours of total equipment
operation for a total of approximately 29,000 bed volumes.

TECHNOLOGY DESCRIPTION
The following technology description was provided by the manufacturer and has not been verified.
The arsenic  adsorption media filter system  included Kinetico  Inc.'s  Para-Flo™ PF60 Model AA08AS
filter unit, which includes two  pressure filter tanks and a  filter  control module.  The control module
houses  water-driven  gears and  mechanically  interconnected  pulse-turbine meter and valves  to
automatically initiate and control filter backwashes. The movement of the gears determines the position
of the filter  valves. Following the throughput of a set total volume of water, the pulse-turbine meter
triggers the water-driven gears to manipulate valves, so that the operating mode of one filter is  switched
from service to backwash, to purge, and finally returns to service.  During a backwash event, one filter
supplies treated water for the backwashing filter and treated water effluent. The filter tanks operate in
parallel when both are in service.  Each filter was loaded  with Alcan Chemicals' Actiguard AAFS50
media, a proprietary granular iron-enhanced activated alumina media. Literature  for Alcan Chemicals'
Actiguard AAFS50 media states that it is certified to NSF/ANSI 61.
The treatment unit is intended for use on groundwater supplies not under the influence of surface water
serving small communities having limited manpower and operating skills.  However, the technology is
also scalable for serving larger systems. The filter system does not require electricity to operate and can
operate continuously or intermittently.  The filter components are modular in nature and can be installed
by a qualified plumber. The tanks are freestanding, requiring only a level surface capable of supporting
the weight of the unit, maintenance of ambient temperature above 35°F (1.7°C), and a feed water pressure
between 30 and 125 psi.

VERIFICATION TESTING DESCRIPTION

Test Site
The verification testing site was the Orchard Hills MHP WTP in Carroll Township, Pennsylvania.  The
source water was untreated groundwater from the WTP Well No.l, which is one of three wells  currently
04/08/EPADWCTR      The accompanying notice is an integral part of this verification statement.           August 2004
                                              VS-ii

-------
used to supply the MHP.  The source water was of generally good quality, with relatively low turbidity,
slightly basic pH, and moderate hardness of about 99 mg/L.  The source water had a high concentration of
manganese, 144 |jg/L on average; an average total arsenic concentration of 14  |Jg/L, ranging from a
minimum concentration of 12 ug/L to a maximum of 17 ug/L; an average iron concentration of 34 ug/L;
an average silica concentration of 19.0 mg/L; and an average alkalinity concentration of 89 mg/L.

Methods and Procedures
Operations, sampling, and analyses were performed to provide an accurate evaluation of the treatment
system under the field conditions. The verification testing was conducted in two phases.  The first phase,
the Integrity Test, was designed to evaluate equipment operation reliability under the environmental and
hydraulic conditions at  the WTP site during the initial two weeks of testing.  The second phase, the
Capacity Test, included testing designed to evaluate the capacity of the arsenic adsorption media filter
system to remove arsenic from the Well No. 1 feed water.
The Integrity Test ran for 13 full days plus 8 hours, during which the field test operator was on-site to
record test data twice per day. The treatment system was  operated continuously using the manual mode of
operation for Well No. 1 2 hours each day and operated  intermittently  during the remainder of each day.
During the Capacity Test, the treatment unit operated intermittently  in  concert  with the  WTP well
operation.  The Capacity Test continued  until an arsenic concentration of 11  |Jg/L  was detected in the
treated water for a minimum of 3 consecutive samples.
Flow rate, production volume, and pressure were monitored and recorded twice per day. Grab samples of
feed and treated water  samples were analyzed for pH,  temperature, turbidity, alkalinity, calcium,
magnesium, hardness, and fluoride by the field test operator. Grab samples were collected and delivered
to the  PADEP Laboratory for analysis of silica,  aluminum, iron, manganese,  chloride, sulfate, and total
phosphorus.  Arsenic samples were  collected and sent to the NSF Laboratories for analyses.   Sample
collection for some water quality parameters was more frequent during the initial two-week Integrity Test
period.  Arsenic  samples were also collected more  frequently  as  the  treated  water  total arsenic
concentration approached the predetermined  end-point  concentration  for a total number of 47 arsenic
samples.  Three sets of samples were speciated for  arsenic during the Integrity Test, to determine the
relative proportion of the total  arsenic concentration that was soluble, that was in the As III species, and
that was in the As V species.   Samples for arsenic speciation were also collected periodically during the
Capacity Test.
Complete descriptions of the verification  testing  results and quality assurance/quality control  procedures
are included in the verification report.

VERIFICATION OF PERFORMANCE

System Operation
The verification testing was conducted under the manufacturers' specified operating  conditions.  Contact
time is a critical parameter for arsenic adsorption efficiency and is dependent upon maintaining the flow
rate within the design range of 1.9 gpm ±0.1 gpm.  A non-integral pressure regulating valve and
diaphragm valve on  the treated water line were used to control and maintain the flow rate.  A relatively
constant flow rate was maintained with minimal flow rate adjustments required.
The system was operated continuously for a 2-hour period each day for the first 13 days plus 8 hours as
part of the Integrity Test using the manual mode of operation for Well No. 1.   The system operated
intermittently in concert with the Well No. 1 operation during the  remainder of the Integrity Test and
throughout the Capacity Test.  The filter unit operated for a total of 14.2 hours per day, on average.
04/08/EPADWCTR      The accompanying notice is an integral part of this verification statement.           August 2004
                                              VS-iii

-------
The filter control module automatically initiates and controls backwashes based on a preset throughput
volume.  The treatment unit was set to backwash one filter following the throughput of approximately
10,500 gallons, plus or minus ten percent.  A single filter was backwashed at a time.  Therefore, each
filter was backwashed every 21,000 gallons. Using the setscrew on the control module, filter backwashes
were manually initiated at the end of the Integrity Test and monthly throughout the Capacity Test for the
purpose of measuring backwash volume and  testing backwash water quality.  These manually initiated
backwashes were performed  for verification testing purposes  only.   Headless across the filter  unit
averaged 1.1  psi during the test period, an amount only slightly greater than the 1.0 psi average headless
during the first two weeks of the test.

Water Quality Results

The feed water  arsenic concentration averaged  14  |Jg/L, with approximately  4 |jg/L as the arsenic III
species and 10  |jg/L as the arsenic V species.  Treated water arsenic concentrations were less than or
equal to  the 2 |jg/L detection limit during the  initial 5 weeks  of testing, or approximately 8,000 bed
volumes  of treated water.  At the end of the verification test, the treated water arsenic concentration
reached  11  ug/L following  approximately  2,350 hours  of equipment operation and treatment  of
approximately 28,800 to 29,200 bed volumes of water, based on the calculated media bed volume of 1.20
cubic feet.  A steep breakthrough curve, which is typical with ion exchange processes, did not occur, as
presented in Figure VS-1. The arsenic breakthrough curve may have been slowed by mixing of the filter
media during filter backwashes.

                            Figure VS-1. Arsenic Breakthrough Curve
                                     (Detection Limit = 2 (ig/L)
                                          15,000         20,000
                                         Treated Water Bed Volumes

                                          I  *  Feed ^  Treated I


At the beginning of the test, the treatment process reduced the pH from 7.3 in the feed water to 6.8 in the
treated. As the media became conditioned by the feed water, the treated water pH increased such that, by
the end of the first week of testing, the pH of the treated water was 7.5 compared to a pH of 7.7 in feed
water. This pH reduction corresponded with a removal of alkalinity during the first two weeks of the test.
Initially, the feed water alkalinity of 88 mg/L was reduced by 43%. However, by the end of the first week
04/08/EPADWCTR
The accompanying notice is an integral part of this verification statement.
                         VS-iv
August 2004

-------
of testing, the feed and treated alkalinity levels were essentially equal. The initial reduction in these water
quality parameters was likely due to the acidic character of the coating on the virgin media.
Fluoride and silica were removed from the feed water initially,  but as the total adsorption  site area
decreased, the preferentially favored arsenic ions out-competed the ions of fluoride and silica for the
remaining adsorption sites. Initially, the feed water fluoride level of around 0.17 mg/L was reduced by up
to 88%. Removal of this ion rapidly declined, so that by the end of the first two weeks of operation,
fluoride was no longer  being  adsorbed  by the media. Similarly,  the initial feed water silica level of
approximately 18 mg/L was reduced by up to 83%. Silica removal decreased within the first two weeks of
operation to a range of 10% to 15% and remained at that level for  approximately one month. Thereafter,
levels of feed water and treated water silica were essentially equal.
The average feed water manganese level  of 144  ug/L,  which  is almost three times the secondary
maximum contaminant level of 50  ug/L, was reduced by an average 92% by the adsorption media.  The
initial treated water sulfate level (29.2 mg/L) exceeded the feed water sulfate level by  180%. Presumably,
this was due to rinsing of excess coating from the media, which apparently contained a sulfate compound.
After the first week of operations, the treated level of sulfate was only approximately 10% higher than the
feed water sulfate. Thereafter, the feed and treated levels of sulfate were essentially equal.
The feed water total phosphorus level,  which averaged 0.032 mg/L, was reduced during the entire period
of verification testing. During the first  6 weeks of testing, between  60%  and 70% of the total phosphorus
was removed. Total phosphorus removal became more erratic thereafter, ranging between 20%  and 68%.
Turbidity was  also  reduced  during  the treatment  process.   However,  concentrations  of calcium,
magnesium, hardness, aluminum,  iron,  and  chloride  were not significantly affected by the  treatment
process. Data tables presenting the on-site and laboratory water quality parameters collected during the
Integrity Test and Capacity Test can be found in the verification report.

Operation and Maintenance Results
The two-phase verification test began on April 22, 2003 and ended  following the conclusion  cf the
Capacity  Test  on  October 28,  2003.    The  treatment  unit, including backwash  cycles,  operated
automatically throughout the test. However, manually  initiated backwashes were also performed as  part
of the testing process.  Operator attention was required to verify  and maintain a constant flow rate, to
check for leaks in the piping and filter unit, and to verify that backwashes occurred as required based on
throughput.  Equipment operation required minimal operator attention.

Consumables and Waste Generation
No chemicals or  electrical power were required.  Wastewater from filter backwash,  purge, and control
module drive water was discharged to a sanitary sewer.  The total water  usage of approximately 83
gallons per backwash cycle represents less than 1  percent of the total finished water production.
Toxicity Characteristic Leaching Procedure (TCLP) and California Waste Extraction Tests (CA WET)
were performed on spent Actiguard AAFS50 media. All concentrations of analyzed parameters  were less
than the current regulatory limits. A complete summary of the TCLP and CA WET results are provided in
the verification report.

Quality Assurance/Quality Control
NSF provided technical and quality assurance oversight of the verification testing as described in the
verification  report, including an audit of nearly 100% of the  data. NSF personnel also conducted a
technical systems audit  during testing to ensure the testing was in compliance with the test  plan.  A
complete description of the QA/QC procedures is provided in the verification report.
04/08/EPADWCTR      The accompanying notice is an integral part of this verification statement.           August 2004
                                              VS-v

-------
      Original Signed by
      Lawrence W. Reiter
               09/08/04
Original Signed by
Gordon Bellen
09/23/04
   Lawrence W. Reiter                 Date
   Acting Director
   National Risk Management Research Laboratory
   Office of Research and Development
   United States Environmental Protection Agency
                                     Gordon Bellen
                                     Vice President
                                     Research
                                     NSF International
                           Date
    NOTICE:  Verifications are based on an evaluation of technology performance under specific,
    predetermined criteria and the appropriate quality assurance procedures.  EPA and NSF make no
    expressed or implied warranties as to the performance of the technology and do not certify that a
    technology will always operate as verified.  The end-user is  solely responsible for complying with
    any and all  applicable federal, state, and local requirements. Mention of corporate names, trade
    names, or commercial products does not constitute endorsement or recommendation for use of
    specific products.   This report is not an NSF Certification of the specific product mentioned
    herein.
       Availability of Supporting Documents
       Copies of the ETV Protocol for Equipment Verification Testing for Arsenic Removal
       dated April 2002, the verification statement, and the verification report (NSF  report
       #04/08/EPADWCTR) are available from the following sources:
       (NOTE: Appendices are not included in the verification report.  Appendices are available
       from NSF upon request.)

       1.  ETV Drinking Water Systems Center Manager (order hard copy)
           NSF International
           P.O. Box 130140
           Ann Arbor, Michigan 48113-0140
       2.  NSF web site: http://www.nsf. org/etv (electronic copy)
       3.  EPA web site: http://www.epa.gov/etv (electronic copy)
04/08/EPADWCTR
The accompanying notice is an integral part of this verification statement.
                         VS-vi
                           August 2004

-------
                                                         August 2004
     Environmental Technology Verification Report
         Removal of Arsenic in Drinking Water
           Kinetico Inc. and Alcan Chemicals
          Para-Flo™ PF60 Model AA08AS with
                    Actiguard AAFS50
                         Prepared for:

                       NSF International
                   Ann Arbor, Michigan 48105
                         Prepared by:

                      Gannett Fleming, Inc.
               Harrisburg, Pennsylvania 17106-7100
Under a cooperative agreement with the U.S. Environmental Protection Agency
                 Jeffrey Q. Adams, Project Officer
           National Risk Management Research Laboratory
               U.S. Environmental Protection Agency
                     Cincinnati, Ohio 45268

-------
                                       Notice

The  U.S. Environmental Protection  Agency  (EPA)  through its  Office  of Research  and
Development  has financially  supported and collaborated with NSF International (NSF) under
Cooperative Agreement No. R-82833301.  This verification effort was supported by the Drinking
Water  Systems (DWS) Center,  operating under the Environmental  Technology Verification
(ETV)  Program. This document has been peer reviewed, reviewed by NSF and EPA, and
recommended for public release.

-------
                                       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 subsurface resources; protection of water quality in public water
systems;  remediation of contaminated sites, sediments and ground water;  prevention and control
of indoor air pollution; and restoration of ecosystems.  NRMRL  collaborates with both public
and private  sector partners to foster technologies that reduce the  cost  of compliance and to
anticipate emerging problems. NRMRL's research provides solutions to environmental problems
by: developing and promoting technologies that protect and improve the environment; advancing
scientific and engineering information to support regulatory and policy decisions;  and providing
the technical  support  and  information transfer to ensure  implementation  of environmental
regulations and strategies at the national, state, and community levels.

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.
                                         Lawrence W. Reiter, Acting Director
                                         National Risk Management Research Laboratory
                                           in

-------
                                   Table of Contents

Section                                                                           Page
Verification Statement	VS-i
Title Page	i
Notice	ii
Foreword	iii
Table of Contents	iv
Abbreviations and Acronyms	ix
Acknowledgements	xi

Chapter 1    Introduction                                                             1
1.1    ETV Purpose and Program Operation	1
1.2    Testing Participants and Responsibilities	2
       1.2.1  NSF International	2
       1.2.2  Field Testing Organization	2
       1.2.3  Manufacturers	3
       1.2.4  Analytical Laboratories	4
       1.2.5  PA Department of Environmental Protection	4
       1.2.6  U.S. Environmental Protection Agency	5
1.3    Verification Testing Site	5
       1.3.1  Source Water	6
       1.3.2  Pilot Filter Discharges	7

Chapter 2    Equipment Description and Operating Processes                         8
2.1    Equipment Description	8
       2.1.1  Basic Scientific and Engineering Concepts of Treatment	8
       2.1.2  Filter System Components	12
       2.1.3  Photographs of Equipment	13
       2.1.4  Drawing of Equipment	14
       2.1.5  Data Plate	14
2.2    Operating Process	17
       2.2.1  Operator Requirements	18
       2.2.2  Required Consumables	18
       2.2.3  Rates of Waste Product!on	19
       2.2.4  Equipment Performance Range	19
       2.2.5  Applications of Equipment	19
       2.2.6  Licensing Requirements Associated with Equipment Operation	19

Chapter 3    Methods and Procedures	20
3.1    Experimental Design	20
       3.1.1  Objectives	20
       3.1.2  Equipment Characteristics	20
             3.1.2.1  Qualitative Factors	20
             3.1.2.2  Quantitative Factors	21
3.2    Equipment Operations and Design	21
3.3    Field Test Equipment	22
                                          iv

-------
                              Table of Contents (continued)

Section                                                                            Page
3.4    Communications, Documentation, Logistics, and Equipment	22
3.5    Equipment Operation and Water Quality Sampling for Verification Testing	23
3.6    Recording Data	23
3.7    Recording Statistical Uncertainty for Assorted Water Quality Parameters	24
3.8    Verification Testing Schedule	24
3.9    Task 1:  System Integrity Verification Testing	25
       3.9.1  Introduction	25
       3.9.2  Experimental Objectives	25
       3.9.3  Work Plan	25
       3.9.4  Analytical Schedule	27
       3.9.5  Evaluation Criteria and Minimum Reporting Requirements	29
3.10   Task 2:  Adsorption Capacity Verification Testing	30
       3.10.1 Introduction	30
       3.10.2 Experimental Objectives	30
       3.10.3 Work Plan	30
       3.10.4 Analytical Schedule	30
       3.10.5 Evaluation Criteria and Minimum Reporting Requirements	33
3.11   Task 3:  Documentation of Operating Conditions and Treatment Equipment
       Performance	34
       3.11.1 Introduction	34
       3.11.2 Experimental Objectives	34
       3.11.3 Work Plan	34
       3.11.4 Schedule	34
       3.11.5 Evaluation Criteria	35
3.12   Task 4:  Data Management	35
       3.12.1 Introduction	35
       3.12.2 Experimental Objectives	35
       3.12.3 Work Plan	35
3.13   Task 5:  Quality Assurance/Quality Control (QA/QC)	36
       3.13.1 Introduction	36
       3.13.2 Experimental Objectives	36
       3.13.3 Work Plan	36
       3.13.4 Analytical Methods	37
       3.13.5 Samples Shipped Off-Site for Analysis	38
3.14   Operations and Maintenance	39

Chapter 4    Results and Discussion	40
4.1    Introduction	40
4.2    Task 1:  System Integrity Verification Testing	40
       4.2.1    Equipment Installation, Startup, and Shakedown	40
       4.2.2    Experimental Objectives	43
       4.2.3    Integrity Test Operational Data	43
       4.2.4    Integrity Test On-Site Water Quality Analyses	45

-------
                             Table of Contents (continued)

Section                                                                            Page
       4.2.5   Integrity Test Laboratory Water Quality Analyses	50
       4.2.6   Integrity Test Arsenic Analyses	52
4.3    Task 2: Adsorption Capacity Verification Testing	54
       4.3.1  Experimental Objectives	54
       4.3.2  Capacity Test Operational Data	54
       4.3.3  Capacity Test On-Site Water Quality Analyses	56
       4.3.4  Capacity Test Laboratory Water Quality Analyses	62
       4.3.5  Capacity Test Arsenic Analyses	67
4.4    Equipment Operation	69
4.5    Backwash Water Quality, Quantity, and Flow Rate	69
4.6    Spent Media Analyses	71
4.7    Task 3: Documentation of Operating Conditions and Treatment Equipment	72
       4.7.1  Introduction	72
       4.7.2  Experimental Objectives	72
       4.7.3  Operations and Maintenance	73
             4.7.3.1   Operations	73
             4.7.3.2   Maintenance	74
4.8    Task 4: Data Management	74
4.9    Task 5: Quality Assurance/Quality Control (QA/QC)	74
       4.9.1  Introduction	74
       4.9.2  Data  Quality Indicators	75
             4.9.2.1   Representativeness	75
             4.9.2.2   Accuracy	75
                      4.9.2.2.1     Split Samples	76
                      4.9.2.2.2     Performance Evaluation Samples for Water Quality
                                  Testing	76
                      4.9.2.2.3     Spike Sample Analyses	77
             4.9.2.3   Precision	77
             4.9.2.4   Statistical Uncertainty	78
             4.9.2.5   Completeness	78

Chapter 5    References	79

Chapter 6    Vendor Comments	80

Tables                                                                             Page
1-1    Feed Water Quality during Testing	7
2-1    Manufacturing and Procedures Specific to Alcan Chemicals' Actiguard AAFS50
       Adsorptive Media	9
2-2    Equipment Design Criteria	10
2-3    Alcan Chemicals' Actiguard AAFS50 Media Specifications	12
3-1    Field Analytical and Calibration Equipment	22
3-2    On-Site Equipment Operating Parameter Monitoring and Data Collection Schedule	27
                                           VI

-------
                              Table of Contents (continued)

Tables                                                                              Page
3-3    Water Quality Sampling Schedule for System Integrity Verification Testing	28
3-4    Arsenic Sampling Plan	29
3-5    Water Quality Sampling Schedule for Media Adsorption Capacity Verification
       Testing	32
3-6    Monitoring, Sampling, and Analyses for Backwash Wastewater, Purge Water, and
       Control Module Drive Water	33
3-7    Schedule for Observing and Recording Equipment Operation and Performance Data	34
3-8    Water Quality Sampling Protocol	39
4-1    Preliminary Arsenic Speciation	41
4-2    Weight of Media Installed and Freeboard in Each Filter Tank	42
4-3    Integrity Test Operational Data	44
4-4    Integrity Test On-Site Water Quality Data	45
4-5    Integrity Test Laboratory Water Quality Data	50
4-6    Integrity Test Laboratory Arsenic Data	53
4-7    Capacity Test Operational Data	55
4-8    Capacity Test On-Site Water Quality Data	56
4-9    Capacity Test Laboratory Water Quality Data	62
4-10   Capacity Test Laboratory Arsenic Data	68
4-11   Backwash Water Characteristics	70
4-12   Purge Water Characteristics	70
4-13   Control Module Drive Water Characteristics	71
4-14   Spent Media Characterization	72
4-15   Field Instrument Calibration Schedule	75
4-16   Split-Samples (April 22, 2003)	76
4-17   Split-Samples (April 23, 2003)	76

Figures                                                                             Page
2-1    Kinetico Inc. and Alcan Chemicals Para-Flo™ PF60 Model AA08AS with Actiguard
       AAFS50, as installed at the Orchard Hills MHP WTP	13
2-2    Kinetico Inc. and Alcan Chemicals Para-Flo™ PF60 Model AA08AS with Actiguard
       AAFS50, as installed at the Orchard Hills MHP WTP	14
2-3    Treated water line showing auxiliary flow control equipment as installed at the Orchard
       Hills MHP WTP	14
2-4    Schematic of Kinetico Para-Flo™ PF60 Model AA08 AS with Actiguard AAFS50 and
       appurtenances at Orchard Hills MHP	16
4-1    Integrity Test headloss and pressure as a function of cumulative run time	44
4-2    Integrity Test pH (4/22/03 to 5/5/03)	46
4-3    Integrity Test temperature (4/22/03 to 5/5/03)	46
4-4    Integrity Test turbidity (4/22/03 to 5/5/03)	47
4-5    Integrity Test alkalinity concentration (4/22/03 to 5/5/03)	48
4-6    Integrity Test fluoride concentration (4/22/03 to 5/5/03)	49
4-7    Integrity Test silica concentration (4/22/03 to 5/5/03)	51
4-8    Integrity Test aluminum concentration (4/22/03 to 5/5/03)	52
4-9    Integrity Test arsenic concentration (4/22/03 to 5/5/03)	53
                                           vn

-------
                             Table of Contents (continued)

4-10   Capacity Test headless and pressure as a function of cumulative run time	55
4-11   Capacity Test pH	57
4-12   Capacity Test temperature	57
4-13   Capacity Test turbidity	58
4-14   Capacity Test alkalinity concentration	59
4-15   Capacity Test fluoride concentration	60
4-16   Capacity Test calcium, magnesium, and total hardness	61
4-17   Capacity Test silica concentration	63
4-18   Capacity Test aluminum concentration	63
4-19   Capacity Test iron concentration	64
4-20   Capacity Test manganese concentration	65
4-21   Capacity Test chloride concentration	65
4-22   Capacity Test sulfate concentration	66
4-23   Capacity Test phosphorus concentration	67
4-24   Capacity Test arsenic concentration	68
                                      Appendices

A     Alcan Chemicals' Technical Bulletin for Actiguard AAFS50 and Media Marketing
       Brochure
B     AAFS50 Media MSDS
C     Equipment Photographs
D     Media Bed Volume Calculations
E     Protocol for Arsenic Speciation
F      Copies of Original Logbooks, Operational Data, and On-Site Water Quality Data
G     PADEP Laboratory Water Quality Data and Sample Submission Forms
H     PADEP Laboratory QA/QC Data
I      On-Site Arsenic Analyses Procedure
J      Spent Media TCLP and CA Wet Analyses
K     Procedure for Media Replacement
L     Media Gradation Analyses
M     TCLP and CA Wet Methods
N     Kinetico Owner's Manual and Installation Guide
O     Preliminary Arsenic Speciation Analyses Reports and Sample  Submission Forms
P      Signed Media Installation Certification
Q     NSF Laboratory Arsenic Data, Sample Submission Forms, and QA/QC Data
R     Performance Evaluation Results
                                         Vlll

-------
                            Abbreviations and Acronyms
ANOVA
ANSI
AWWA
AA
BET
CAWET
cm
°C
c.u.
D
DQO
EBCT
EPA
ETV
°F
FRP
FTO
gpm
H
HazMat
HOPE
ICR
ISE
L
Ib
LCD
LED
m
M
MCL
MCLG
mg/L
MHP
mL
mm
MDL
MSDS
N/A
NA
ND
NEMA
NIST
Analysis of Variance
American National Standards Institute
American Water Works Association
Activated Alumina
Brunauer, Emmett and Teller
California Waste Extraction Tests
Centimeter
Degrees Celsius
Platinum-Cobalt Color Units
Depth
Data Quality Objectives
Empty Bed Contact Time
U. S. Environmental Protection Agency
Environmental Technology Verification
Degrees Fahrenheit
Fiberglass Reinforced Plastic
Field Testing Organization
Gram
Gallons per Day
Gallons per Minute
Height
Hazardous Material
High Density Polyethylene
Information Collection Rule
Ion Selective Electrode
Liter
Pound
Liquid Crystal Diode
Liquid Emitting Diode
Meter
Mole
Maximum Contaminant Level
Maximum Contaminant Level Goal
Milligram per Liter
Mobile Home Park
Milliliter
Millimeter
Method Detection Level
Material Safety Data Sheets
Not Applicable
Not Analyzed
Not Detected
National Electrical Manufacturers Association
National Institute of Standards and Technology
                                         IX

-------
                       Abbreviations and Acronyms (continued)

NPDES            National Pollution Discharge Elimination System
NR                Not Reported
NSF               NSF International (formerly known as National Sanitation Foundation)
NTU               Nephelometric Turbidity Units
O&M              Operation and Maintenance
OSHA             Occupational Safety and Health Administration
PA                Pennsylvania
PADEP            PA Department of Environmental Protection
PE                Performance Evaluation
PRV               Pressure Reducing Valve
PSM               Process Safety Management
psi                Pounds per Square Inch
PSTP              Product Specific Test Plan
PVC               Polyvinyl  Chloride
QA                Quality Assurance
QC                Quality Control
QA/QC            Quality Assurance/Quality Control
QAPP              Quality Assurance Project Plan
RCRA             Resource and Recovery Act
RMP               Risk Management Plan
SM                Standard Methods for the Examination of Water and Wastewater
SOP               Standard Operating Procedure
SS                Stainless Steel
TCLP              Toxicity Characteristic Leaching Procedure
TSTP              Technology Specific Test Plan
UPS               Uninterruptible Power Supply
|j,g/L               microgram per liter
W                 Width
WTP               Water Treatment Plant
WWTP            Wastewater Treatment Plant
                                         x

-------
                                  Acknowledgements

The Field Testing Organization, Gannett Fleming, Inc., was responsible for all elements in the
testing sequence, including collection of samples,  calibration and verification of instruments,
data  collection and analysis, data management, data interpretation, and the preparation of this
report.

       Gannett Fleming, Inc.
       P.O. Box 67100
       Harrisburg, PA  17106-7100
       (717) 763-7212, Ext. 2109
       (717) 763-1808 FAX
       Contact: William Allis, Project Manager
       E-mail: wallis@gfnet.com

The  laboratory  selected for laboratory  analyses of all of the ETV water quality parameters
(except arsenic) that were scheduled to be conducted by an EPA accredited and PADEP certified
laboratory was:

       Pennsylvania Department of Environmental Protection Laboratories
       1500 North 3rd Street
       Harrisburg, PA  17102
       (717)705-2197
       (717) 783-1502 FAX
       Contact: Ted Lyter, Inorganic Services Division Chief
       E-mail: plyter@state.pa.us

Spent media toxicity analyses were performed by:

       TriMatrix Laboratories, Inc.
       5555 Glenwood Hills Parkway, SE
       Grand Rapids, MI 49588
       (616)975-4500
       Contact: Michael W. Movinski, Vice President, Sales and Marketing
       Email: mmtrimatrix@comcast.net

Arsenic analyses were performed by the NSF Laboratory:

       NSF International
       789 N. Dixboro Road
       Ann Arbor, MI 48105
       (734)769-8010
       (734) 769-0109 FAX
       Contact: Bruce Bartley, Project Manager
       E-mail: bartley@nsf.org
                                          XI

-------
The manufacturers of the equipment (joint venture) were:

       Kinetico Inc.
       10845 Kinsman Road
       P.O. Box 193
       Newbury, OH 44065
       (440) 564-9111 Ext. 233
       (440) 564-4222 FAX
       Contact: Mark Brotman, Research Scientist
       E-mail: mbrotman@kinetico.com

       Alcan Chemicals
       525 S. Washington Street
       Suite No. 9
       Naperville, IL 60540-6641
       (630)527-1213
       (630) 527-1229 FAX
       Contact: William Reid
       E-mail: bill.reid@alcan.com

Gannett Fleming wishes to thank the following participants:

NSF International,  especially Bruce Bartley, Dale Scherger, and Angela Beach, for providing
guidance and program management.

The Orchard Hills MHP WTP owner, Robert Goodling.
                                         xn

-------
                                       Chapter 1
                                      Introduction

1.1    ETV Purpose and Program Operation

The U.S. Environmental Protection Agency (EPA) has created the Environmental Technology
Verification  (ETV)   Program  to facilitate   the  deployment of  innovative   or  improved
environmental technologies through performance verification and dissemination of information.
The  goal  of the ETV  Program  is to  further environmental protection by  accelerating the
acceptance and use of improved and more cost-effective technologies. ETV seeks to achieve this
goal by providing high-quality, peer-reviewed data on technology performance to  those involved
in the design, distribution, permitting, purchase, and use of environmental technologies.

ETV works in partnership with recognized standards and testing organizations;  with stakeholder
groups consisting of buyers, vendor organizations, and  permitters; and with the  full participation
of individual technology developers.  The  program evaluates the performance  of innovative
technologies by  developing test plans responsive to the needs of  stakeholders, by conducting
field or laboratory tests (as appropriate), collecting and analyzing  data, and preparing peer-
reviewed reports. All evaluations are conducted in accordance with  rigorous quality assurance
protocols to ensure that data of known and adequate quality are generated and that the results are
defensible.

The EPA has partnered with NSF International (NSF)  under the ETV Drinking Water Systems
(DWS) Center to verify the performance  of  small drinking water systems that serve small
communities.  A goal of verification testing is to enhance and facilitate the acceptance of small
drinking water treatment equipment by state drinking water regulatory officials and consulting
engineers,  while reducing  the need  for testing  of  equipment at  each  location where the
equipment's use is contemplated. NSF meets this goal by working with manufacturers and NSF-
qualified Field Testing Organizations (FTOs) to conduct verification  testing under the approved
protocols.  It is important to note that verification of the equipment does not mean the equipment
is "certified" by NSF or "accepted" by EPA.  Rather,  it recognizes that the performance of the
equipment has been determined and verified by these organizations for those conditions tested by
the FTO.

The DWS  Center evaluated the performance  of the Kinetico Inc.  and Alcan  Chemicals Para-
Flo™ PF60 Model AA08AS with Actiguard AAFS50  System, which is an arsenic adsorption
media filter used in drinking water treatment system applications. The verification test evaluated
the ability of the absorptive media to remove arsenic from  drinking water.  This document
provides the verification test results for the Kinetico Inc. and Alcan Chemicals  Para-Flo™ PF60
Model AA08AS with Actiguard AAFS50 System.

-------
1.2    Testing Participants and Responsibilities

The ETV testing of the Kinetico Inc. and Alcan  Chemicals Para-Flo™ PF60 Model AA08AS
with Actiguard AAFS50 System was a cooperative effort between the following participants:

       NSF International
       Gannett Fleming, Inc.
       Kinetico Inc.
       Alcan Chemicals
       PA Department of Environmental Protection
       U.S. Environmental Protection Agency
       Orchard Hills Mobile Home Park (MHP)

The following is a brief description of each ETV participant and their roles and responsibilities.

1.2.1   NSF International

NSF is an independent, not-for-profit testing and certification organization dedicated to public
health and safety and to the protection of the environment.  Founded in 1946 and located in Ann
Arbor,  Michigan, NSF  has been instrumental in the development of consensus standards for the
protection of public health and the environment. NSF also provides testing and certification
services to  ensure products bearing the NSF Name, Logo, and/or Mark meet those standards.
The EPA partnered with  NSF to verify the performance of drinking water treatment systems
through the EPA's ETV Program.

NSF provided technical oversight of the verification testing.  An audit of the field analytical, data
gathering, and recording procedures was conducted.  NSF also performed all laboratory arsenic
water quality  analyses and provided review of the Product Specific Test Plan (PSTP) as well as
this report.

Contact Information:
       NSF International
       789 N. Dixboro Rd.
       Ann Arbor, MI 48105
       Phone: (734) 769-8010
       Fax:  (734)769-0109
       Contact: Bruce Bartley, Project Manager
       Email: bartley@nsf.org

1.2.2   Field  Testing Organization

Gannett Fleming, Inc., a consulting engineering firm located in Harrisburg,  Pennsylvania,
conducted the verification testing of the Kinetico Inc.  and Alcan Chemicals arsenic removal
system. Gannett Fleming is an NSF-qualified FTO for the ETV Drinking Water Systems Center.

-------
Gannett Fleming was responsible for conducting the Integrity Verification testing for 14 calendar
days (13 full days plus 8 hours) and for conducting Capacity Verification testing until a pre-
determined arsenic breakthrough concentration was achieved.  Gannett Fleming provided all
needed logistical support, established a communications network, and scheduled and coordinated
activities of all participants.  Gannett Fleming was responsible for ensuring the testing location
and feed water conditions were such that the verification testing could meet its stated objectives.
Gannett Fleming prepared the PSTP; oversaw the pilot testing; managed, evaluated, interpreted,
and reported on the data generated by the testing; and evaluated and reported on the performance
of the technology.

The Gannett Fleming field engineer conducted the  on-site  analyses (on-site or at the Gannett
Fleming Treatability Lab) and data recording during the testing. Oversight of the daily tests was
provided by Gannett Fleming's Project Manager.

Contact Information:
       Gannett Fleming, Inc.
       P.O. Box 67100
       Harrisburg, PA 17106-7100
       (717) 763-7212, Ext. 2109
       (717) 763-1808 FAX
       Contact:  William Allis, Project Manager
       E-mail:  wallis@gfnet.com

1.2.3   Manufacturers

The treatment system is a joint venture, with the Para-Flo™  PF60 Model AA08AS filter unit
manufactured by Kinetico Inc.  and the Actiguard AAFS50 adsorption filter media manufactured
by Alcan Chemicals.

The manufacturers were responsible for supplying a field-ready arsenic adsorption media filter
system equipped with all necessary components, including treatment equipment, instrumentation
and  controls,  and  an operations  and  maintenance  manual. The  manufacturers were  also
responsible for  providing logistical and technical  support as needed,  as  well  as providing
technical assistance to the FTO during  operation and monitoring of the equipment undergoing
field verification testing.

Contact Information:
       Kinetico Inc.
       10845 Kinsman Road
       P.O. Box 193
       Newbury, OH 44065
       (440) 564-9111 Ext. 233
       (440) 564-4222 FAX
       Contact:  Mark Brotman, Research Scientist
       E-mail:  mbrotman@kinetico.com

-------
       Alcan Chemicals
       525 S. Washington Street
       Suite No. 9
       Naperville, IL 60540-6641
       (630)527-1213
       (630) 527-1229 FAX
       Contact: William Reid
       E-mail: bill.reid@alcan.com

1.2.4  Analytical Laboratories

The  PADEP  Laboratories performed all  of the  laboratory  water quality analyses, excluding
arsenic.

Contact Information:
       Department of Environmental Protection Laboratories
       Inorganic Services Division
       1500 North 3rd Street
       Harrisburg, PA  17102
       (717)705-2197
       (717) 783-1502 FAX
       Contact:  Ted Lyter, Inorganic Services Division Chief
       E-mail: plyter@state.pa.us

NSF laboratories performed all laboratory arsenic water quality analyses.

Tri-Matrix Laboratories performed TCLP and CA WET analyses on the spent media.

Contact Information:
       TriMatrix Laboratories, Inc.
       5555 Glenwood Hills Parkway, SE
       Grand Rapids, MI  49588
       (616)975-4500
       Contact: Mr. Michael W. Movinski, Vice President, Sales and Marketing
       Email: mmtrimatrix@comcast.net

1.2.5  PA Department of Environmental Protection

The  PADEP's mission  is to protect Pennsylvania's air,  land and water from pollution and to
provide for the health and safety of its citizens through a cleaner environment.

The  PADEP  is  the  state  agency  largely responsible  for  administering Pennsylvania's
environmental laws and regulations.  Its responsibilities include:  reducing air pollution, making
sure  Pennsylvania's drinking water is  safe, protecting water quality in Pennsylvania's rivers and
streams,  making  sure waste is handled  properly, managing the Commonwealth's recycling
programs,  and  helping citizens  prevent pollution  and  comply with the  Commonwealth's

-------
environmental regulations. PADEP is committed to providing general environmental education
and encouraging effective public involvement in setting environmental policy.

The roles and responsibilities of PADEP included laboratory analyses for all of the ETV water
quality parameters (except arsenic) that were scheduled  to be conducted by an EPA accredited
and PADEP certified laboratory.

The PADEP was also responsible for reviewing the test plan  and final report because this testing
may also  serve as a pilot  study component  of a water  supply  permit  application  for  the
installation  of a full-scale version of this type of process at  this  site.  Also, because the site is
already  a  permitted  public  water  supply,  the PADEP needed  to  be involved  with  any
modifications.

1.2.6   U.S. Environmental Protection Agency

The EPA,  through its Office of Research and  Development, has financially supported  and
collaborated with NSF under Cooperative Agreement No. R-82833301.  This verification effort
was supported by the DWS Center operating under the ETV Program. This document has been
peer reviewed, reviewed by NSF and EPA, and recommended for public release.

1.3    Verification Testing  Site

The verification testing site was Orchard Hills MHP Water Treatment Plant (WTP) located off of
Windy Hill Road in Carroll Township, PA.  The WTP is  housed within a masonry block building
located within the MHP.  The building is heated to a minimum temperature of 50°F.  Bordering
the MHP  boundary,  in  close proximity to the back  of the WTP building, is land under
cultivation.    The  WTP,  with  a permitted capacity  of  30  gpm,  supplies approximately
200 domestic connections. The sources of supply for the WTP are Well Nos. 1, 11, and 12,  of
which a portion of Well No.  1  discharge was used as the source water for the arsenic adsorption
media  filter  verification  testing.   Well  No.  1  is  located near the entrance to the MHP,
approximately 100  yards north  of the WTP.    The  WTP  process consists  of five pressure
manganese  greensand filters, two chlorine  contact/finished  water storage tanks, two finished
water pumps, and six hydropneumatic tanks.

Two chemicals are fed at the WTP:  sodium hypochlorite for oxidation and  disinfection,  and
polyphosphate for sequestration and corrosion  control.  The chemical feed points are  located
downstream of the arsenic  adsorption media  filter supply  connection.  The control of the
wells/filtration process is based on a  level control system in two finished water storage tanks,
located within the WTP building.  The well pumps operate based  on level sensors in the finished
water storage tanks. Water from the finished water storage tanks is pumped to hydropneumatic
tanks  via   finished water pumps.   Low-  and  high-pressure switches associated with  the
hydropneumatic tanks activate and deactivate the finished water pumps. The hydropneumatic
tanks supply the distribution system and provide backwash water for the greensand filters.

The frequency and duration of well pump  operation depends  on distribution  system demand and
well water level/production capacity.  Average daily well run time, as observed during this test,

-------
was approximately  14 hours per day.   The total combined WTP flow range for  all wells, as
reported by the operator, is 10 to 20 gpm.

During the ETV test, a portion of Well No.l discharge, prior to any treatment, was diverted to
the arsenic adsorption media filter.  The arsenic adsorption media filter was set up inside the
WTP building, directly in front of several of the manganese greensand filters.  The treated water,
control module water, and backwash wastewater from the  arsenic adsorption media  filter were
discharged to  an existing drainpipe inside the building and subsequently conveyed to the MHP
Wastewater Treatment Plant (WWTP).

1.3.1   Source Water

The source water for the verification test was untreated groundwater from Orchard Hills WTP
Well No. 1.

Well No. 1 source water is generally of good quality, with relatively low turbidity, slightly basic
pH,  and  moderate  hardness.    The  source  water average  manganese  concentration  of
approximately 144 (ig/L is almost three times the Secondary Standard for drinking water.  Black
particles  were frequently observed in  the  feed  water samples.   The feed water  total arsenic
concentration averaged approximately 14 |J,g/L, approximately 4  [ig/L of which was in the form
of Arsenic III.  The source water total arsenic  concentration is below the current  maximum
contaminant level (MCL) of 50 |J,g/L, but exceeds the future MCL of 10  ng/L that will become
effective  in January 2006.   A  summary of the  feed water quality  information is  presented in
Table 1-1 below.  Additional feed water quality data are presented in Chapter 4.

Alcan Chemicals indicated that no pretreatment would be required for the arsenic  adsorption
media system.  Alcan stated: "Manganese  is very far down on the selectivity series,  and Alcan
Chemicals  does not  expect that it will be an issue.   [Ion selectivity  series is included in
Table 2-3.] Additional work has shown media adsorption capacity for arsenic to be independent
of the manganese in the water.  In addition, iron is really only a problem if it is present in very
high amounts  as it  precipitates and clogs the bed.  This  is easily rectified with a  backwash or
other type of agitation.  This is a  mechanical function that would be common to  any granular
bed, not  a chemical interference.  Again,  there is  no indication that iron in solution has any
negative impact whatsoever on the media's  ability to adsorb arsenic."

-------
Table 1-1. Feed Water
Parameter Units
Arsenic
pH
Temperature
Turbidity
Alkalinity
Calcium
Magnesium
Hardness
Fluoride
Silica
Aluminum
Iron
Manganese
Chloride
Sulfate
Total
Phosphorus
Mg/L
-
°C
NTU
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
Mg/L
Mg/L
Mg/L
mg/L
mg/L
mg/L
Quality during Testing
Number of
Samples Mean Minimum
47
198
184
184
84
27
27
27
39
40
40
28
28
28
28
28
14
7.6
13.8
0.25
89
26.0
8.3
99
0.17
19.0
203
34
144
18.7
10.5
0.032
12
7.3
11.5
0.10
84
24.8
7.3
96
0.13
17.4
<200
<20
36
16.8
10.1
0.024
Maximum
17
7.8
15.5
3.9
92
28.0
8.7
104
0.27
21.1
339
116
1481
20.4
11.2
0.043
Standard
Deviation
1.1
N/A
0.94
0.30
1.5
0.92
0.50
1.7
0.03
0.80
22.0
24
286
0.85
0.26
0.005
95%
Confidence
Interval
14-
7.6-
13.6-
0.20-
89-
25.6-
8.1-
98-
0.16-
18.7-
<200(1)
23-
16-
18.3-
10.4-
0.029 -
14
7.6
13.9
0.30
89
26.4
8.5
100
0.18
19.3
-212
45
272
19.1
10.6
0.034
TTT
    The lower confidence interval level was calculated below the detection limit for this parameter.
 1.3.2   Pilot Filter Discharges

 The treated water, control module drive water, and backwash water from the arsenic adsorption
 media filter unit were discharged to an existing  drainpipe inside the building and subsequently
 conveyed to the Orchard Hills WWTP. No discharge permits were required.  At the request of
 PADEP,  backwash wastewater, purge water, and  control module drive water were monitored,
 sampled,  and  analyzed every  second month to  evaluate the  quantity and  quality of water
 discharged to the WWTP.  Treated water quality and the quantity, as well as the quality of all
 backwash water discharged from the pilot filter unit to the MHP WWTP, are discussed in detail
 in Chapter 4.

-------
                                        Chapter 2
                     Equipment Description and Operating Processes

2.1    Equipment Description

The equipment tested was Kinetico Inc.'s  and Alcan Chemicals' arsenic adsorption media filter
system. The model tested was the Para-Flo™ PF60 Model AA08AS filter unit with Actiguard
AAFS50 media. The major system components include two pressure filter tanks with adsorptive
filter media,  a control module, filter media, feed water pipe,  treated water pipe,  feed  water
sample tap, treated water sample tap, and two wastewater  ports (rinse and backwash).  The
system configuration and major components are described in more detail in the following
sections.

After the verification test, Kinetico  renamed the tested model to reflect the use of a larger tank
inlet and outlet facilitating faster flow rates. Please refer to  Chapter 6, Vendor Comments, for
additional details concerning these modifications.

2.1.1   Basic Scientific and Engineering Concepts of Treatment

The conceptual treatment process for the  arsenic adsorption media filter is based  on passing
arsenic-contaminated feed water through a  bed of media having a strong affinity for arsenic.

Activated  alumina media historically  has provided cost-effective, reliable performance as  a
material for  producing  a granular  adsorbent media for removal of arsenic from feed water.
Actiguard AAFS50 is an iron-enhanced activated alumina media, which  has been determined to
significantly  promote the adsorption effectiveness of conventional activated alumina. As water
passes  down  through a filter vessel containing this media, the arsenic concentration declines until
it is no longer detectable. As the upper portion of the media becomes saturated, the treatment
band (mass transfer zone) progresses downward until all adsorptive capacity is used  and arsenic
breakthrough occurs.

Adsorption is the attachment of the adsorbate (arsenic) to the  surface of the adsorbent media
grains  (activated alumina). The removal capacity and effectiveness of the arsenic removal media
is  dependent on a number of factors,  of  which surface area is of primary  importance.  The
surface area is a function of the porosity of the media grains.  Adsorbent media contains a large
quantity of very small pores throughout the media grains.  Other factors determining the  capacity
and effectiveness of adsorbent media are  accessibility of the pore  sites  for arsenic ions, time
available for  arsenic ions to migrate to pore sites, ions competing for pore sites, concentration of
arsenic in the feed  water,  pH of  the  feed  water, oxidation state  of  arsenic,  and  flow
characteristics of the feed water conveying  the arsenic into the bed of adsorbent media.

The Kinetico/Alcan Chemicals arsenic adsorption media filter system uses Actiguard AAFS50, a
proprietary,  granular,  iron-enhanced,  activated alumina media. Tests  performed by  Alcan
Chemicals indicate that  AAFS50  has up to five times(1) the  arsenic  adsorption capacity of
 ' As stated in the Alcan AAFS50 marketing brochure (see Appendix A).

-------
standard activated alumina and that iron enhancement also  enables the  removal of As (III).
Tables 2-1, 2-2, and 2-3 present information specific to this equipment and media.
  Table 2-1.  Manufacturing and Procedures Specific to Alcan Chemicals' Actiguard
  AAFS50 Adsorptive Media

  Item	Manufacturing/Procedures	
  Raw Material (used to make adsorptive      Activated Alumina and Iron
  media)

  Method of Manufacture                  Chemical Processes: Proprietary
                                       Thermal Processes: Proprietary
                                       Sizing/Screening Methods:  Proprietary
                                       Packaging Methods: Proprietary

  Preconditioning Procedure                Wetting Requirements: 10 Bed Volumes of Feed Water

  Regeneration Procedure                  N/A

  Regeneration Results	N/A	

Filter operations are automatically controlled by the filter control module.  The control module
houses water-driven gears and mechanically interconnected  pulse-turbine meter and valves.  The
movement of the gears determines the position of the filter valves. Following the throughput of
a set total volume of water, the pulse-turbine meter triggers the water-driven gears to manipulate
valves so that the operating mode of one  filter is switched  from service to backwash, to purge,
and finally returns to service.  The other filter remains in service, providing treated water for the
backwashing filter.

-------
Tables 2-2 and 2-3 present design criteria for the arsenic adsorption process and appurtenances.
 Table 2-2. Equipment Design Criteria
         Para-Flo™ PF60 Model AA08AS
                No. of Filter Tanks
                Filter Tank Dimensions
                        Inside Diameter (ID)
                        Height (including integral control module)
                        Height (vessel only)
                Mode of Operation
                Design Flow, Total
                Flow Range, Total
                Design Capacity, Total
                Empty Bed Contact Time (EBCT) at 2 gpm
                Minimum Recommended Feed Pressure
                Filter Media
                        Depth
                        Freeboard Above Media

                        Volume Per Tank

                        Weight Per Tank
                        Volume, Total (2 tanks)

                        Mesh Size  (Tyler mesh series)
                        Media Expansion during Backwash
                Filter Tank Material
                Backwash Control
                Backwash
                        Flow Rate
                        Duration
                Time Between Backwash and Rinse
                Purge
                        Flow Rate
                        Duration
         Pressure Gauges
                Manufacturer
                Type
                Pressure Range
8 inches
46 inches
40 inches
Parallel
1.9±0.1 gpm
1.8 to 2.0 gpm
2.0 gpm
4.6 minutes
30psi

21 inches
17.5 inches
(Actual 18.25 inches)
0.70 cu. ft.
(Actual -0.60 cu. ft.)
39.76 Ibs
1.4cu. ft.
(Actual ~1.20 cu. ft.)
28x48
50%
Polyester,  Vinylester
Automatic based on total
throughput of 10,500 gallons ±
10%

4.0 gpm
13 minutes
3 minutes

1.9gpm± 0.1 gpm
5 minutes

Ashcroft® Duralife
1084,  Grade 2A
0-100 psi (accuracy of
±0.5%)	
                                                10

-------
Table 2-2. Equipment Design Criteria (continued)
        Totalizer Meters
               Manufacturer
               Type
               Series
               Accuracy
        Rotameter
               Manufacturer
               Model
               Maximum Reading
               Accuracy
               Pressure, max
        Treated Water Throttling Valve
               Manufacturer
               Type
               Material of Construction
               Size
               Control
        Three Way Regulating Valve
               Manufacturer
               Model No.
               Maximum Inlet Pressure
               Reduced Pressure Range
        Y-Check Valve
               Manufacturer
                Size Code/Size
               Material of Construction
ABB
Positive displacement
VI00 (feed)/C700 (filtrate)
± 1.5%

Blue-White
F-50376N
2.0 gpm
No Data
250 psi

George Fischer
Diaphragm
Type 304, DN25,PVC-U
1 inch
Manual

Watts Industries, Inc.
2A645
300 psi
3 to 50 psi

George Fischer
1 inch
Type 304, DN25, PVC-U
                                                11

-------
Table 2-3.  Alcan Chemicals' Actiguard AAFS50 Media Specifications
Chemical Constituents                                                        Weight, %
       A12O3 + Proprietary Additive                                                 83
       Silicon as SiO2                                                           0.020
       Titanium as TiO2                                                         0.002
       Loss on Ignition                                                           17
Physical Properties
       Bulk Density                                                     0.91 g/cm3 (56.8 lbs/ft3)
       BET(1) Area                                                            220 nf/g
       Attrition                                                                 0.3%
       Voids                                                                   48%
       Pore Size                                                              No Data
       Pore Volume                                                          <0.35 cm3/g
       Abrasion Loss                                                <5% (due to spray coating fines,
                                                                        smaller than 48 mesh)
       Moisture (weight)                                                     0-300°C: 25%
                                                                         300-1000°C:  10%
       Sieve sizes, US sieve series                                                28 x 48
       Particle Size                                                           No Data
       Effective Size                                                          0.37 mm
       Uniformity Coefficient                                                     1.48
Ionic Preference Series
       •       Amons: OH->HAsO4>Si(OH)3>O->F>HSeO3>SO42>CrO42>HCO3>Cr>NO3
       •       Cations: Th>Al>U(4)>Zr>Ce(4)>Fe(3)>Ce(3)>Ti>Hg>UO2>Pb>Cu>Ag>Zn>Co>
                        Fe(2)>Ni>Tl>Mn
Approvals
       •       Certified to NSF/ANSI 61
       •       Passed U.S. EPA TCLP
       NSF/ANSI 61 and TCLP approvals are indicated in Alcan Chemicals' Technical Bulletin for AAFS50
       Media and Media Marketing Brochure, included in Appendix A.
MSDS  (See Appendix B)	

2.1.2  Filter System Components

The arsenic adsorption media filter is a modular equipment process consisting of the following
components:

•      Two pressure filter tanks (main and remote) piped for parallel operation;
    The BET theory is used to estimate the number of molecules required to cover the absorbent surface with a
    monolayer of adsorbed molecules, N^.  Multiplying N^, by the cross-sectional area of an adsorbate molecule
    yields the sample's surface area.

                                               12

-------
•      One  control module situated on top of the main filter tank and consisting  of a pulse-
       turbine meter,  water-driven gear mechanism, and valves to control the filter modes of
       operation;
•      One feed water sample tap and one treated water sample tap;
•      One influent pipe and one effluent pipe connecting the main filter tank to the remote filter
       tank;
•      One feed water pipe connected to the control module;
•      One treated water pipe connected to the control module;
•      Alcan Chemicals' Actiguard AAFS50 media in each filter tank;  and
•      Two waste ports incorporated in the control module for backwash wastewater and gear
       mechanism drive water discharge.

The following equipment was provided by Kinetico specifically for the ETV and is not normally
included with the arsenic adsorption media filter:

•      Two pressure gauges, one located on the feed water pipe and one located  on the treated
       water pipe;
•      One Y-check valve located on the feed water pipe, just upstream of the pilot filter;
•      Two totalizer water meters, one located on the feed water pipe and one located on the
       treated water pipe;
•      One diaphragm valve for flow regulation located on the treated water pipe  just upstream
       of the rotameter;
•      One rotameter located on the treated water pipe downstream of the diaphragm valve; and
•      One pressure regulating valve located just upstream of the diaphragm valve on the treated
       water pipe.

2.1.3   Photographs of Equipment

Photographs of the equipment installed at the WTP are included below.  Additional photographs
are included in Appendix C.
                                            Figure 2-1.  Kinetico  Inc. and  Alcan Chemicals
                                            Para-Flo™  PF60 Model AA08AS with Actiguard
                                            AAFS50, as installed  at the Orchard  Hills MHP
                                            WTP.
                                           13

-------
                                           Figure  2-2.  Kinetico Inc.  and Alcan Chemicals
                                           Para-Flo™  PF60 Model AA08AS with Actiguard
                                           AAFS50, as installed at  the  Orchard Hills MHP
                                           WTP.
                                           Figure 2-3.  Treated water line showing  auxiliary
                                           flow control equipment, as installed at the Orchard
                                           Hills MHP WTP.
2.1.4  Drawing of Equipment

A schematic drawing of the equipment is shown in Figure 2-4.

2.1.5  Data Plate

A data plate  was installed on  the arsenic  adsorption  media filter main  tank to provide the
following information:

      Equipment Name:
      Para-Flo™ PF60 with Actiguard Media
      Model Number: AA08AS
      Media Number: AAFS50

      Manufacturers' Names and Addresses:
      Kinetico Incorporated        Alcan Chemicals
      10845 Kinsman Road        525 S. Washington Street
      Newbury, Ohio 44065       Suite #9
                                 Naperville, Illinois 60540
                                          14

-------
Additional Information:
Serial Number: 0052690
Service flow:  1.8 - 2.0 gpm
Unit installed for NSF and EPA Environmental Technology Verification Program.
Call (440) 564.4233 for more information.

Warning and Caution Statements:
Testing in progress, please do not disturb.
This unit is designed to operate with minimum and maximum inlet pressures of 30 psi
and 125 psi, respectively.
                                   15

-------
r
                                                             0




1

— : ••! '
f
                 Figure 2—4,  Schematic of Kineifco  Para-Flo™ PF60
    Model  .ViOHAS  with Aatiguard AAF35D and  appurtenances  at Orchard  Hills MHP.
                                    16

-------
2.2    Operating Process

This modular filter system  consists  of dual,  pressurized filter tanks designed  for  parallel
operation in the downflow mode.  The filter system does not require electricity to operate.  Both
filter tanks are in service except when  one filter tank is off-line for backwashing.  During a
backwash event,  one filter tank supplies  the treated  water production, control module drive
water, and  treated water for backwashing the other  filter.  The filter system can operate either
intermittently  or  continuously.  Modes of operation are automatically controlled, based  on
volume  of throughput, using  a  proprietary control module  containing a  pulse-turbine meter.
Valve operation is controlled by a water-driven gear mechanism within the control module that is
mechanically interconnected with the pulse-turbine meter.  The gear mechanism drive water is
required only  during backwash and purge  and  is supplied by the filter remaining in service.
There are no other triggers for automatic initiation of operating modes.  The control module has
a set-screw for manually adjusting the actuator to conduct a manual backwash; this procedure is
described  in the  proprietary  Technical Manual, which was on  file  at NSF International and
Gannett Fleming during the test.

The combined total  flow and flow rate from the filter tanks was  monitored with two accessory
totalizer meters and a rotameter.  Flow rate was adjusted with a nonintegral diaphragm valve,
located on the treated water side of the filter tanks.  There are no flow gauges to monitor the rate
of backwash  wastewater.   This was checked using  the "bucket and  stopwatch"  method.
Collection  of backwash and purge water for volume  determination  and water quality analyses
was performed once during the  Integrity Test and once every other month during the system
Capacity Test.  The incremental throughput readings  from  each totalizer  meter were used to
estimate the quantity of water used in backwash cycles for  the instances when backwashes
occurred  and the wastewater was not collected.  The incremental  feed water  totalizer meter
reading minus the incremental treated water totalizer meter reading equals the estimated volume
of backwash,  purge, and  control module drive water used for both filter tanks.  Also, two
totalizer meters provided redundancy.  If one totalizer meter had failed, the other  meter would
have served as a backup.  The difference in feed  water and treated water pressure readings
provided the determination of loss of head across both filters.

Grab samples for on-site and laboratory analyses were collected from the feed water and treated
water sample taps, located immediately upstream and downstream of the adsorption media filter
tanks, as shown on Figure 2-1. Samples from these taps were collected following the opening of
their respective ball valves and a flush period of  approximately five seconds.

The  manufacturer states  that Actiguard AAFS50  is regenerable.   However, the additional
adsorption  capacity  of this  media  compared  to conventional activated alumina  offers  an
advantage,  because regeneration may not be economical for a  small system. Alternatively, the
media  may be removed  and replaced  with new media prior  to  breakthrough,  based  on a
predetermined life of media for  a specific site water quality.   The manufacturer indicates the
media has passed the U.S. EPA TCLP test and is landfillable.  Regeneration was not considered
for this test.
                                           17

-------
2.2.1   Operator Requirements

The arsenic adsorption media filter was operated with Well No.  1 in an automatic on-demand
mode during the adsorptive media Integrity and Capacity Verification Tests.  The MHP WTP
well pumps were controlled based on the finished water storage tank level, started on a low level
setpoint, and stopped on a high level setpoint.  Therefore, operator attention was minimal during
the tests and consisted mainly of monitoring the equipment to confirm proper operation and data
collection.

Because Well No. 1 normally operated for only brief periods in automatic mode, the well pump
was operated manually by the Gannett Fleming field engineer during the 13-day plus  8 hour
Integrity Test for the required minimum of 2 hours of continuous operation  on a daily basis. The
well supply and arsenic adsorption media filter operated automatically for the remainder of the
six-month Capacity Test, except during the backwashes observed by  Gannett Fleming.  During
the observed and monitored backwashes, Well No. 1  was operated manually by the Gannett
Fleming field  engineer  to produce  continuous   operation and to  provide  more  accurate
measurement of backwash, purge, and drive water flow rates.

Spent Actiguard AAFS50  media can  be removed and replaced by the operator following
breakthrough of arsenic.   After  the conclusion of the  Capacity Test,  data were generated
representing the volume  of water treated by  the Actiguard AAFS50 media and the resultant
treated water arsenic concentrations.  The results of Capacity Testing are included in Chapter 4.

The system was designed to backwash automatically after a throughput of 10,500 gallons ± 10%.
Operator  initiation was not  required during automatic  backwashes.    The  system  also
automatically re-initiated service operation of the backwashed filter. The position of an indicator
dot on top of the control module actuator (see Figure 2-1) provided evidence that a backwash had
occurred during those periods when the plant was not staffed.

The  manually  initiated backwash required  approximately 1.5  to 2.0  hours of operator time.
Operator time included setup, approximately 25 minutes of backwash time, on-site water  quality
analyses, sample collection for laboratory water quality analyses, documentation, and equipment
cleanup. The manually initiated backwash, monitoring, and data collection  were requested by
PADEP as special conditions of the test  plan   and  are not general  equipment operating
requirements.

2.2.2  Required Consumables

The system does not use electricity or chemicals during normal treatment operations and requires
only treated water for each backwash cycle.   The required consumables are limited  to the
adsorption media and treated water for backwash use, as described below:

•     Actiguard AAFS50 activated alumina media: approximately 0.7 cubic feet per filter tank
      (-1.4 cubic feet total)  per  manufacturer specifications.  Approximately 1.20 cubic feet
      were installed  in  the  2-filter test unit,  based  on volumetric calculations included  in
      Appendix D.
                                           18

-------
•      Treated water:   62  gallons  of backwash  and  rinse  per  cycle per  manufacturer
       specifications.   The actual treated water usage during backwash (including purge and
       control module drive water) averaged approximately 83 gallons.

2.2.3   Rates of Waste Production

The manufacturer indicated approximately 62 gallons of filter backwash wastewater and purge
(rinse) wastewater would be  generated for every  10,500 gallons ±  10% of throughput.  The
observed wastewater  volume was  approximately  83 gallons,  including  approximately  9.75
gallons of control module drive water.  The total volume of water used per filter unit backwash
was consistent for each manually initiated and observed backwash.  Backwash water quantity
and water quality characteristics are described in more detail in Chapter 4.

2.2.4   Equipment Performance Range

The equipment flow range and minimum recommended pressure are presented in Table 2-2. The
manufacturer has  stated their  arsenic adsorption media system may not  be  appropriate for feed
water  quality containing high  levels  of potentially interfering  ions,   such as sulfate, silica,
fluoride, and phosphate, depending on the feed water pH. However, the  manufacturer has stated
these interferences can be mitigated by pretreatment, if necessary.

2.2.5   Applications of Equipment

The manufacturer stated the process is  appropriate  for groundwater not under the influence  of
surface water at "very  small" and "small" systems having limited manpower and operating skills.
It is also appropriate for "medium"  systems.  The EPA defines "very small" systems as those
systems serving a population of 25-500  people, "small" systems as those systems serving a
population  501-3,300  people, and "medium-size"  systems as  those serving 3,301 to  10,000
people.

MHP  Well No. 1  has relatively high manganese levels that were not treated  prior to  passing
through the system.   However, the manufacturers  indicate the arsenic  adsorption  capacity  is
independent of the manganese concentration in the feed water.

2.2.6   Licensing Requirements Associated with Equipment Operation

States generally require a specific grade  of waterworks operator permit in order to operate a filter
process on  a public water supply. However, this requirement did not apply for the ETV because
all treated water was discharged to waste.

In Pennsylvania, to operate a full-scale version of this treatment technology for the  Orchard  Hills
MHP  public drinking water supply, a D9 license would be required;  "D" refers to a capacity  of
0.1 mgd or less and "9" refers  to inorganics removal.
                                           19

-------
                                        Chapter 3
                                Methods and Procedures
3.1    Experimental Design
This verification test was developed to provide verifiable information related to the performance
of the Kinetico Inc. and Alcan Chemicals arsenic adsorption media system.  Field  operations,
sampling, and analytical methodologies were performed in a manner assuring the quality of data
collected  would  provide an  accurate  evaluation  of the  treatment  system  under the field
conditions.

The ETV testing was conducted in two phases.  The first phase, the Integrity Test, was designed
to evaluate the reliability of equipment operation  under  the  environmental and hydraulic
conditions at the MHP WTP site during the initial two weeks of testing.  The  second phase, the
Capacity Test, included testing designed to evaluate the capacity of the arsenic adsorption system
to remove arsenic from the Well No. 1 feed water.

3.1.1   Objectives

The objectives of the verification test were:

•      Produce data to meet the Data Quality Objectives  (DQOs) shaped by the manufacturers'
       performance objectives;
•      Present data on the impact of variations in feed water quality, such as  turbidity, arsenic,
       pH, silica, fluoride, iron, and manganese on equipment performance;
•      Evaluate the  logistical,  human, and  economic  resources necessary  to  operate  the
       equipment;
•      Evaluate the reliability,  ruggedness,  cost factors,  range  of usefulness,  and ease  of
       operation of the equipment; and
•      Evaluate the arsenic adsorption capacity of the equipment under field conditions.

3.1.2  Equipment Characteristics

3.1.2.1  Qualitative Factors.  The equipment was  operated in such a  way as to  maintain its
operating parameters  within the manufacturers' recommendations.  Contact time is a critical
parameter for arsenic adsorption efficiency  and is  dependent on maintaining flow within the
design range.  The nature and  frequency of the changes  required to maintain the operating
conditions were used in the qualitative evaluation of the equipment.

Frequent and significant adjustments would have indicated  a relatively lower reliability and
higher susceptibility to environmental conditions, as well as the degree  of operator experience
that may be required.  However, as discussed in more detail in Chapter 4, flow rate adjustments
were minimal. The effect of operator experience on the treatment results was evaluated.

The modular nature of the filter components,  similar to a residential ion exchange water softener,
makes equipment installation easy and  straightforward.  The equipment can be installed by  a
                                           20

-------
qualified plumber.  The equipment is also easy to move and reinstall at another location.  The
filter tanks are freestanding, requiring only a level surface capable of supporting 210 pounds and
maintenance of ambient temperature above 35°F.

3.1.2.2 Quantitative Factors. The following factors were quantified for site-specific conditions,
based upon data collected during this testing program:

•      Backwash water quantity and quality;
•      Backwash and purge duration and frequency; and
•      Estimated labor hours for operation and maintenance.

These quantitative factors were used  as an initial benchmark to assess equipment performance
and to develop operation and maintenance costs.

3.2    Equipment Operations and Design

The EPA/NSF ETV Protocol for Equipment Verification Testing for Arsenic Removal,  including
Chapter 6: Testing Plan - Adsorptive Media Processes for the Removal of Arsenic, specifies the
procedures used to ensure the  accurate documentation of  both equipment performance  and
treated water quality.  Strict adherence to these procedures result in the  definition of verifiable
performance of the equipment.  Chapter 5 includes information on the ETV Protocol  and other
documents used in the preparation of this report.
                                          21

-------
3.3    Field Test Equipment

Table 3-1 presents the analytical and calibration equipment used on-site.
Table 3-1.  Field Analytical and Calibration Equipment

Equipment                                       Manufacturer/Model/Specs
Turbidimeter


pH/ISE Meter
Thermometer

Arsenic Field Test Kit


Dead Weight Pressure Gauge Tester


Burettes (for analytical titrations)


Stopwatch and "Bucket"
Platform Scale
Hach Model 2100P Portable Ratio™ Optical System
(meets or exceeds USEPA Method 180.1 criteria)

Orion Model 290A with Tnode pH Electrode Model 91-
578N (resolution 0.1/0.01/0.001, accuracy ± 0.005); and
Fluoride Combination Electrode Model 96-09
(reproducibility ± 2%)

Miller & Weber (range 0-32°C; NIST traceable)

Industrial Test Systems (ITS), Inc. Model QUICK Low
Range II (optimum accuracy below 6 (ig/L)

Amthor Testing Instrument Co. Inc. (Type No. 460;
range 0-6000 psi)

50 mL capacity with 0.1 mL subdivisions and 1000 mL
reagent reservoir

Digital stopwatch and 2.0 L graduated cylinder with 10
mL increments for rotameter, totalizer meters, and
control module drive water calibration checks. Fifty
gallon container for backwash wastewater flow
calibration

Triner Scale Model 303, Serial No. 87D-065, Capacity
202 Ibs.
3.4    Communications, Documentation, Logistics, and Equipment

It was Gannett Fleming's responsibility to coordinate communication  between all verification
testing participants.  Gannett Fleming maintained all  field documentation. Bound field logbooks
were used to record all water treatment equipment operating data.  Each page was sequentially
numbered  and labeled with the  project name and number.  Completed pages were signed  and
dated by the individual responsible for the entries.  Errors had one line  drawn through them and
this line was initialed  and dated. Any deviations from the approved final PSTP were thoroughly
documented in the field logbook. Copies of the logbook pages are included in the appendices of
this report.

All field activities were thoroughly documented using the following forms of record:

•      Field Logbook
•      Field Data Sheets
•      Photographs
•      Laboratory Submission Sheets and Reports
                                             22

-------
Laboratory  submission  forms  accompanied  all  samples shipped to  the  PADEP and  NSF
Laboratories.   Copies of laboratory  submission forms  for all  samples are included in the
appendices of this verification report.

3.5    Equipment Operation and Water Quality Sampling for Verification Testing

The  field activities  conformed  to requirements in the PSTP developed and approved for this
verification test. The sampling and sample analyses that occurred during this verification testing
program  were performed according to the procedures detailed by Gannett Fleming in the PSTP.
Any unanticipated or unusual situations that  altered the  plans for equipment operation, water
quality sampling, or data quality were discussed with the NSF technical lead and PADEP.  Any
deviations from the approved final PSTP were  documented.

During routine operation, the following were documented daily:

•      The number of hours the arsenic adsorption media filter was operated;
•      The number of hours the operator was working at tasks at the treatment plant related to
       the operation of the arsenic adsorption media filter; and
•      Description of tasks performed during arsenic adsorption media filter operation.

3.6    Recording Data

The following information was recorded on-site:

•      Experimental run number
•      Water type (feed, treated, waste type)
•      Hours of operation (calculated)
•      Feed water flow rate
•      Treated water flow rate
•      Feed water production
•      Treated water production
•      Feed water pressure
•      Treated water pressure
•      Feed water temperature
•      Treated water temperature
•      Feed water turbidity
•      Treated water turbidity
•      Feed water pH
•      Treated water pH
•      Feed water arsenic concentration (qualitatively with field test kit)
•      Treated water arsenic concentration (qualitatively with field test kit)
•      Occurrence of a backwash
•      Backwash water flow rate (when field engineer is present)
•      Backwash duration (when field engineer is present)
•      Backwash total volume  (measured directly when field engineer is present)
                                           23

-------
3.7    Recording Statistical Uncertainty for Assorted Water Quality Parameters

For the  analytical  data obtained during verification testing, 95%  confidence intervals were
calculated by Gannett Fleming for arsenic data and for all other water quality data where the
sample set contains eight or more values.

The  consistency and precision of water  quality  data were  evaluated  with the  use of the
confidence interval.  A confidence interval describes a population range in which any individual
population measurement may exist with a specified percent confidence. The following formula
was used for confidence interval calculation:
                         confidence interval = X±tn-i, \.-  \SI-Jnj
where:    X is the sample mean;
          S is the sample standard deviation;
          n is the number of independent measures included in the data set;
          t is the t distribution value with n-1 degrees of freedom; and
          a is the significance level, defined for 95% confidence as:  1 - 0.95 = 0.05.

According to  the 95%  confidence interval approach,  the a term is defined to have a value of
0.05, thus simplifying the equation for the 95% confidence interval in the following manner:
                        95% confidence interval = X + tn -1,0.975 (S /Jn
Results of these calculations were expressed as the sample mean, plus or minus the width of the
confidence interval.

pH statistics were calculated on a log basis.

3.8    Verification Testing Schedule

Verification testing  activities  included  equipment set up and shakedown, equipment Integrity
Verification Testing, Adsorption Capacity Testing, and water quality sampling and analysis.  The
test schedule was developed to encompass all of these activities.

The Integrity Test began on April 22, 2003.  The Integrity and Adsorption Capacity Verification
Tests were initiated simultaneously.  The Integrity Verification Test ran for a 2-week (13 full
days plus 8 hours)  period, ending  May 5,  2003.  The Adsorption Capacity Verification test
continued until 11 ng/L*^ of arsenic was detected in the treated water for a minimum of three
consecutive samples.   Three consecutive treated water samples with  arsenic concentrations
greater than or equal to 11 (ig/L were required to ensure the predefined endpoint had in fact been
   Kinetico/Alcan Chemicals originally requested that 12 (ig/L be used  as the  stopping point to ensure the
   threshold of 10  (ig/L had actually been crossed and the reading was not due to analytical error or method
   variability.  Due to relatively slow arsenic breakthrough and reduced feed water arsenic concentrations, the
   manufacturer, NSF, and Gannett Fleming agreed to revise the stopping point to 11 (ig/L.


                                            24

-------
reached and the end of the test would not be influenced an analytical  error.  The capacity test
ended on October 28,  2003.  The equipment was disassembled by the manufacturer and filter
core samples were taken by Gannett Fleming on November 4, 2003.

3.9    Task 1:  System Integrity Verification Testing

3.9.1  Introduction

During Task 1, Gannett Fleming evaluated the  reliability of the equipment operation under the
environmental and hydraulic conditions at Orchard Hills MHP WTP Well No. 1.  The Integrity
Verification Test  was  performed  to  determine whether  the treatment objectives could  be
achieved for arsenic removal at the design operating parameters for the  arsenic adsorption media
system.    The  adsorption media  filter was  operated for  Integrity Test purposes within the
operational range presented in the equipment design criteria.

3.9.2  Experimental Objectives

The  experimental  objectives  for  the Integrity Test phase  of the  verification  testing are
summarized below:

•      Evaluate equipment operational reliability under field conditions;
•      Document feed water quality and arsenic concentration;  and
•      Collect operational and water quality data under field conditions.

3.9.3  Work Plan

Initial  shakedown  testing was  performed  on  the  adsorption filter  unit  to  establish basic
operability.  Two sets  of feed and treated speciated arsenic samples were used to establish the
capability of the  filter unit to remove  arsenic  from the feed water.  Following the initial
shakedown  testing,  a pressure-reducing  valve was  added  to the system upstream  of the
diaphragm valve to maintain a constant flow rate under variable feed water pressures.

Prior to  beginning the Integrity and  Capacity Test phases,  the  manufacturer  installed new
Actiguard AAFS50 media in each of the two adsorption  filter tanks. A platform scale was used
to weigh the media prior to installation into  each filter tank. The weight of the media and the
measurement of "freeboard" from the top of the media to the top of the unit (top of the opening
in each filter tank where the media is added) were recorded.

Following the protocol for startup,  as  detailed  in the Alcan Chemicals'  Technical Bulletin for
Actiguard AAFS50 in Appendix A, the initial 10 bed volumes of treated water  (flushing water)
should be  discounted  prior to recording the totalizers'  startup readings.  The  manufacturer
actually used approximately 350 gallons, or 36  bed volumes,  during startup to  wash the media
and to verify the operation of the filter control module. This water volume, used for startup, was
documented when recording the initial totalizer reading prior to initiation of the Integrity and
Capacity Tests.
                                           25

-------
The Integrity Test monitoring and on-site data collection were performed at frequencies shown
in the schedule presented in Table 3-2.  The treatment system primarily operated intermittently
due to the intermittent operation of Well No. 1.  However, the treatment system was required to
operate  continuously for at least 2 hours each day during the Integrity Test, as specified in  the
test plan.   The 2-hour  continuous operation each day  was performed  and witnessed by the
Gannett Fleming field engineer and used the manual mode of operation for Well No. 1 at  the
WTP well pump control panel.

Grab  samples for  on-site  and laboratory  analyses were collected according to the  sampling
schedule presented in Table 3-3.  The feed water and treated water sample taps were flushed for
at least  five seconds prior to sample collection.  A sampling plan for arsenic that includes  the
Integrity Verification Test is presented in Table 3-4.  Three days of the daily feed water and
treated water samples were collected  to speciate arsenic,  as specified in Table 3-4. The protocol
for arsenic speciation (from the TSTP) is presented in Appendix E.  Daily and weekly samples
collected for on-site analysis were  analyzed immediately  after collection during the 2-hour
period of continuous operation. Alkalinity, total hardness, calcium hardness, and fluoride were
analyzed in the Gannett Fleming Treatability Lab within two hours of leaving the site.  Sample
collection  and handling procedures  followed Standard Methods  3010 B.  Daily and weekly
samples for laboratory analysis were  collected during the 2-hour period of continuous operation.
At least one hour of operation occurred prior to sample collection for arsenic.

All of the samples were collected by the Gannett Fleming field engineer in appropriate sample
bottles prepared with preservatives, as  required,  specific to the analytical methods to be used.
Additionally, the  samples  were stored  and  shipped in accordance with appropriate procedures
and holding times, as specified by the PADEP and NSF. A water quality sampling protocol for
PADEP Laboratory analysis, describing volumes, preservation, holding times, and laboratory
sample identification for each water quality parameter, is presented in Table 3-8.   The methods
used by the laboratory for the analytical procedures are presented in Section3.13.4 and described
in Task 5, Quality Assurance/Quality Control. All on-site data and observations were recorded
by the Gannett Fleming field engineer in a series of bound logbooks.  Copies of the original
logbooks and on-site Water Quality Data are included as Appendix F. All PADEP Laboratory
water quality data and sample  submission forms  are  included  in Appendix  G.   PADEP
Laboratory QA/QC Summary Tables are  included  in  Appendix  H.   Complete QA/QC
documentation is on file at NSF.

Two backwashes occurred during the  System Integrity Verification Test, one  of which was
manually  initiated and witnessed  by the field engineer.  Backwash water flow,  duration,  and
volume  were monitored volumetrically and recorded. Backwash water quality was analyzed as
listed in Table 3-6. Complete results and data analysis are presented in Chapter 4.
                                           26

-------
3.9.4  A nalytical Schedu le

The arsenic  adsorption  media  filter  system operational data were monitored following the
procedures and at  the frequencies prescribed in the test  plan, as  summarized below  and in
Table 3-2.

•      The treated  water flow rate was monitored and adjusted, as needed, using the rotameter
       and diaphragm valve located on the treated water pipe.  The treated water flow rate was
       recorded twice per day, before and after any necessary adjustment.  The flow rate was set
       and maintained at 1.9 gpm ± 0.10 gpm.
•      The feed water and treated water production  were monitored and recorded twice per day
       at the totalizer meters located on the feed water and treated water pipes.
•      Well pump run time is not totalized at the WTP motor control  center. Therefore, run time
       was back-calculated from the totalizer readings and flow rate.
•      The feed water pressure was monitored twice per day at the  pressure gauge located on the
       feed water pipe.  Minimum and maximum operating pressures for the filter tanks are 30
       psi  and 125  psi, respectively.
•      The treated  water pressure was monitored twice per day at the pressure gauge located on
       the  treated water pipe.  This reading was performed at the same time as the feed water
       pressure measurement.   The difference  between these values represents tie headloss
       through the  system.

Table 3-2.  On-Site Equipment Operating Parameter Monitoring and Data Collection
Schedule
  Parameter	Monitoring Frequency	Monitoring Method	
  Treated Water Flow Rate            Check & record twice per day (adjust  Rotameter
                                 when 5% above or belo w target record
                                 before and after adjustment)
  Feed Water and Treated Water       Check & record twice per day        Feed and treated totalizer
  Production                                                     meters
  Hours of Production                Calculate & record once per day      Calculated from totalizer meter
                                                                and flow rate data
  Feed Water Pressure                Check & record twice per day        Feed water pressure gauge
  Treated Water Pressure             Check & record twice per day        Treated water pressure gauge


Water quality data were collected as described below:

•      The water quality of the feed water and treated water were characterized by  analysis of
       the  water quality  parameters listed in Table 3-3.  The water quality analyses presented in
       Table  3-3 were  conducted to provide  state drinking  water regulatory agencies  with
       background  data  on the quality of the feed water being treated and the quality of the
       treated water.
•      Samples were collected during the 2-hour period of continuous operation, following a
       minimum of 1 hour of operation.
•      Temperature, pH, turbidity, and qualitative arsenic were analyzed on-site.
                                            27

-------
Table 3-3.  Water Quality Sampling Schedule for System Integrity Verification Testing


                                   Test Streams to be Sampled
Parameter
 Sampling
Frequency
Standard
Method(1)
  EPA
Method(2)   Hach Method
 On-Site Analyses
 Arsenic               ^

 pH              Twice Daily

 Temperature         Daily

 Turbidity           Daily

 Alkalimty(4)         Daily

 Calcium''0          Weekly

 Magnesium1-4-1       Weekly

 Hardness™         Weekly

 Fluonde(4)           Daily


 Laboratory Analyses

 Arsenic'-5'1           Daily
  Silica

  Aluminum

  Iron
                    Daily

                    Daily

                   Weekly
 Manganese        Weekly

 Chloride           Weekly

 Sulfate             Weekly

 Total              Weekly
 Phosphorus	
                                      Adsorptive Media
                                 Feed Water & Treated Water
                                      Adsorptive Media
                                 Feed Water & Treated Water
                                      Adsorptive Media
                                 Feed Water & Treated Water
                                      Adsorptive Media
                                 Feed Water & Treated Water
                                      Adsorptive Media
                                 Feed Water & Treated Water
                                      Adsorptive Media
                                 Feed Water & Treated Water
                                      Adsorptive Media
                                 Feed Water & Treated Water
                                      Adsorptive Media
                                 Feed Water & Treated Water
                                      Adsorptive Media
                                 Feed Water & Treated Water
                      Adsorptive Media
                 Feed Water & Treated Water
                      Adsorptive Media
                 Feed Water & Treated Water
                      Adsorptive Media
                 Feed Water & Treated Water
                      Adsorptive Media
                 Feed Water & Treated Water
                      Adsorptive Media
                 Feed Water & Treated Water
                      Adsorptive Media
                 Feed Water & Treated Water
                      Adsorptive Media
                 Feed Water & Treated Water
                      Adsorptive Media
                 Feed Water & Treated Water
                                                          (See Appendix I)
                                                4500-H+ B

                                                 2550 B

                                                 2130 B
                                                                            8221

                                                                            8222

                                                                          Calculated
                                                                         (8226-8222)
                                                                            8226
                                                4500-F'C
              200.8

              200.7

              200.7

              200.7

              200.7

              300.0

              300.0

              365.1
TOAPHA, AWWA and WPCF (1995). Standard Methods for Examination of Water and Wastewater. 19th ed.
    Washington, D.C. APHA.
(2)  EPA Methods Source: EPA Office of Ground Water and Drinking Water. EPA Methods are available from the
    National Technical Information Service (NTIS).
^  See Table 3-4. An arsenic field test kit was used for periodic qualitative arsenic checks.
'-4-1  Analyzed on-site or at the Gannett Fleming Treatability Lab.
'-5-1  The NSF Laboratory performed laboratory arsenic analyses.  The PADEP Laboratory performed all other
    laboratory analyses during the Integrity Test.
                                                   28

-------
Table 3-4. Arsenic
Test Period
Laboratory Analyses
Shakedown
Integrity
Verification
Adsorption
Capacity
Verification
Adsorption
Capacity
Verification
Sampling
Sample
Sources
Feed,
Treated
Feed,
Treated
Feed,
Treated
Feed,
Treated
Plan
Sample
Frequency

Daily
Daily
Weekly
Daily
Sampling
Period

2 days
13 days
8 hours
First 6 months'-1-1
Final
2 months(1)
No. of Days
Samples
Speciated1-2-1 Hold Samples

2 None
3 None
2(2) None
1(2) 12 per week
Total No.
Analyses

12
40
56
min: 20
max: 124
 On-Site Qualitative Analyses^'
     Integrity
    Verification

    Adsorption
     Capacity
    Verification
                 Feed,
                Treated
                 Feed,
                Treated
Weekly
13 days
8 hours
Weekly    First 6 months(1)
N/A
               N/A
N/A
             N/A
              48
Adsorption
Capacity
Verification

Feed,
Treated


3 /week

Final
2 months'-1'1


N/A


N/A


48
(3)
The estimated sampling period was 8 months.  If breakthrough did not occur within 8 months, the test and
sampling plan would have continued until breakthrough occurred.
This was considered the minimum number of days samples are speciated during the capacity verification
testing. If arsenic was detected in the treated water, feed and treated water samples collected the following week
would have been speciated and analyzed.
Method procedure presented in Appendix I.
3.9.5  Evaluation Criteria and Minimum Reporting Requirements

A table and time  series plots were produced to present all feed water and treated water quality
data which varied with time from the system Integrity  Verification test. The system Integrity
Verification test demonstrates the initial ability of the adsorptive media to remove the feed water
arsenic concentration  to below  detectable levels in the  treated water.   All water quality
parameters, operational parameters,  backwash flow  rates,  and  quantities were tabulated and
plotted, as appropriate.  The backwash waste stream and control  module discharge flow rates
were tabulated.  A plot of feed  and treated water pressure  and system headloss  is presented.
System headloss information was used to  infer power requirements for a system that will  pump
directly through the treatment unit.  No direct measurement of power was possible because the
system does  not  require electricity.    Test  results are summarized, plotted, and  discussed in
Chapter 4. All raw data are included in appendices, as referenced in Chapter 4.
                                              29

-------
3.10   Task 2: Adsorption Capacity Verification Testing

3.10.1  Introduction

The objectives of the Adsorption Capacity Test were to produce operational and water quality
data up through and including what Kinetico Inc. and  Alcan Chemicals have defined as the
breakthrough arsenic level  for their  arsenic adsorption system.   The  performance of the
adsorptive  media is a function of feed water quality, contact time,  rest time, and type of
adsorptive  media used.   Arsenic breakthrough is highly dependent  on the concentration and
adsorptive characteristics (isotherm) of the arsenic to be treated by the adsorptive media.  Design
and empty  bed contact time (EBCT) will help define the performance of the media for a given
feed water quality.   Adsorption capacity verification testing  was performed one time  for the
arsenic adsorption media system, using the feed water from Well No. 1 at Orchard Hills MHP.

3.10.2  Experimental Objectives

The experimental objective was to provide equipment operating and water quality data related to
the adsorptive media capacity to remove arsenic from the feed water to the pre-defined arsenic
breakthrough concentration.

3.10.3  Work Plan

Task 2 Adsorption  Capacity Verification Testing began simultaneously with Task  1, System
Integrity  Verification Testing.  The operating conditions were as stated under 3.9.3 Work Plan
for Task  1:  System  Integrity Verification Testing.

3.10.4  A nalytical Schedu le

•      Operational Data Collection
       o      The treated water flow rate was monitored and  adjusted, as needed, using the
              rotameter and diaphragm valve located on the treated water pipe. The treated
              water flow rate  was recorded  twice per day, before and  after any necessary
              adjustment. The flow rate was set and maintained  at 1.9 gpm ±0.10 gpm.
       o      The feed  water  and treated water production was monitored  and recorded twice
              per day at the totalizer meters, located on the feed water and treated water pipes.
       o      Well  pump run time is not totalized at the WTP motor control center.  Therefore,
              run time was back-calculated from the totalizer readings and flow rate.
       o      The  feed water  pressure was  monitored twice per day  at the pressure gauge
              located on the feed water pipe.  Minimum and  maximum operating pressures for
              the filter tanks are 30 psi and 125 psi, respectively.
       o      The treated water pressure was monitored twice  per day at  the  pressure gauge
              located on the treated water pipe.  This was performed at the same time as the
              feed water pressure measurement.  The difference between these values represents
              the headloss through the system.
                                           30

-------
Sample Holding
o      As indicated in Table 3-4, as the media approached 70% of its predicted capacity,
       samples for laboratory arsenic analyses were collected on a daily basis and held
       (approximately 2 weeks) pending the results of the weekly arsenic samples.  This
       was done in the event arsenic breakthrough was missed with the weekly sampling.
       Arsenic hold samples for the final 4 weeks  of the Capacity Test were submitted
       for  analysis.  Fluoride,  silica, iron, manganese, and aluminum samples were
       collected weekly  during Task 2.

Water Quality Data Collection
o      The adsorptive media feed water quality, treated water quality,  and wastewater
       quality were characterized by the analysis of the water quality parameters listed in
       Tables 3-5 and 3-6.  The sampling frequency is also described in Tables 3-5 and
       3-6.   This frequency was intended to  provide sufficient water quality data to
       effectively characterize  the breakthrough profile of arsenic and to develop a
       representative wastewater quality profile.
o      Grab samples of backwash wastewater were collected for water  quality analyses
       at the frequency presented in Table 3-6. The backwash and purge  water collection
       procedure is for one of the two filter tanks.  The samples were mixed to maintain
       a relatively homogenous  suspension during sample collection.

Arsenic Speciation
       The minimum arsenic speciation frequency is presented on Table 3-4.

Spent Media Analysis
o      TCLP  and  CA WET were performed on spent Actiguard AAFS50 media,  as
       required by the test plan. The physical condition of the spent media was noted and
       reported, along with the result of the TCLP and CA WET testing in Chapter 4 and
       Appendix J.
o      A 1.5-inch thin-walled copper tube,  4 feet in length, was used to  core one sample
       of spent Actiguard AAFS50 adsorption media from each of the  two  filter tanks.
       The Kinetico procedure  for media  replacement in  Appendix K was  followed
       through Step 8a.  (with the exception of emptying the media  into the bucket) to
       gain access to the media contained in each filter tank and to decant the water out
       of each tank. Following  decanting, the copper tube was used to obtain a core
       sample through the  entire depth  of the media from each tank.  Each core was
       discharged into a large plastic bag.  The bag was vigorously shaken to provide a
       homogenous media sample.  The  sample was used for TCLP and CA WET
       analyses.
o      A media gradation analysis was performed on the spent Actiguard AAFS50 media
       and compared to the gradation  analysis of new media, presented in Appendix L,
       to determine the extent of media physical degradation, if any.
o      The result of all testing on spent media are discussed in Chapter 4.

-------
Table 3-5. Water Quality Sampling Schedule for Media Adsorption Capacity Verification
Testing
 Parameter
 Sampling
Frequency
Test Streams to be Sampled
Standard      EPA        Hach
Method(1)   Method(2)     Method
 On-Site Analyses
 Arsenic

 pH

 Temperature

 Turbidity

 Alkalimty(4)

 Calcium'4)

 Magnesium1-4-1

 Hardness(4)

 Fluonde(4)

 Laboratory Analyses
    (3)              Adsorptive Media
               Feed Water & Treated Water
  Daily            Adsorptive Media         4500-FT" B
               Feed Water & Treated Water
  Daily            Adsorptive Media           2550 B
               Feed Water & Treated Water
  Daily            Adsorptive Media           21 SOB
               Feed Water & Treated Water
 3/Week           Adsorptive Media
               Feed Water & Treated Water
 Weekly           Adsorptive Media
               Feed Water & Treated Water
 Weekly           Adsorptive Media
               Feed Water & Treated Water
 Weekly           Adsorptive Media
               Feed Water & Treated Water
 Weekly           Adsorptive Media          4500-F" C
               Feed Water & Treated Water
                                       (See Appendix I)
                                                        8221

                                                        8222

                                                      Calculated
                                                     (8226-8222)
                                                        8226
Arsenic

Silica

Aluminum

Iron

Manganese

Chloride

Sulfate

Total Phosphorus

TCLP(6)

CA WET(6)
Weekly (5)

Weekly

Weekly

Weekly

Weekly

Weekly

Weekly

Weekly

Once

Once
Adsorptive Media
Feed Water & Treated Water
Adsorptive Media
Feed Water & Treated Water
Adsorptive Media
Feed Water & Treated Water
Adsorptive Media
Feed Water & Treated Water
Adsorptive Media
Feed Water & Treated Water
Adsorptive Media
Feed Water & Treated Water
Adsorptive Media
Feed Water & Treated Water
Adsorptive Media
Feed Water & Treated Water
Spent Actiguard AAFS50
Adsorptive Media
Spent Actiguard AAFS50
Adsorptive Media
200.8

200.7

200.7

200.7

200.7

300.0

300.0

365.1

SW-846
EPA 13 11
(See Appendix M)
    Standard Methods for Examination of Water and Wastewater. 19th ed. Washington, D.C. APHA.
    EPA Methods Source: EPA Office of Ground Water and Drinking Water. EPA Methods are available from the National
    Technical Information Service (NTIS).
    An arsenic field test kit was used for periodic qualitative arsenic checks, as specified in Table 3-6.
    Analyzed on-site or at the Gannett Fleming Treatability Lab.
    See arsenic sampling plan in Table 3-4.
    TriMatrix Laboratories, Inc. performed the TCLP and CA WET analyses.
                                                  32

-------
Table 3-6. Monitoring, Sampling,
and Control Module Drive Water
Purge and Backwash
Wastewater
Parameter Sample Type
Flow Rate
Volume
Duration
Turbidity
pH
Arsenic
Manganese
Iron
Aluminum
Volumetric
Direct
measurement
Manually timed
Grab«
Grab(1)
Grab(1)
Grab(1)
Grab«
Grab(1)
and Analyses for Backwash Wastewater, Purge Water,
Control Module
Drive Water
Sample Type Frequency® Method
Volumetric
Direct
measurement
Manually timed
Grab(1)
Grab'1'
Grab(1)
Grab(1)
Grab(1)
Grab(1)
Every second month
Every second month
(directly)
Every second month
Every second month
Every second month
Every second month
Every second month
Every second month
Every second month
"Bucket"(3)(4) &
stopwatch
Graduated
container1-3-1
Stopwatch
SM2130-B
SM4500-H+
EPA 200. 8
EPA 200. 7
EPA 200. 7
EPA 200.7
TTT
 (4)
Grab samples were collected using a 2-liter beaker from a continuously mixed batch tank. Backwash and purge
wastewaters were collected in 50- and 30-gallon containers, respectively. Grab sample for control module drive
water were collected with a 2-liter beaker.
Frequencies indicated per request of PADEP.
The "buckets" were 50- and 30-gallon containers for calibrating backwash and purge flow rates, respectively.
Increments in liters were marked on the sides of these  containers, based on incrementally filling the containers
beforehand with a 2-liter graduated cylinder.
A 2.0 graduated cylinder was the "bucket" for determining control module drive water discharge flow rate.
 3.10.5 Evaluation Criteria and Minimum Reporting Requirements

 The results of Adsorption Capacity Testing are presented in Chapter 4 and include the following:

 •      Record of Arsenic Removal:
        o      An arsenic breakthrough curve was plotted showing the adsorptive media treated
               water  concentrations versus volumes treated.  Feed water arsenic concentrations
               were included on the same plot.
        o      A spreadsheet of arsenic feed water concentrations and calculations of the average
               feed water arsenic concentration was tabulated.

 •      Process Control:
        o      The adsorptive media feed water and treated water arsenic, pH, pressure,  and
               water  production  were tabulated and used  to  calculate  incremental feed  and
               treated water production,  differential pressure, and  cumulative arsenic removed.
               The adsorptive media  feed  water  average,  standard deviation,  and confidence
               interval were included for each parameter, when appropriate.
                                               33

-------
3.11   Task  3:    Documentation  of Operating Conditions  and  Treatment Equipment
       Performance

3.11.1 Introduction

During each day of verification testing, arsenic adsorption media filter operating conditions were
documented, including the rate of headloss gain.  The volumetric flow rate through an adsorptive
media vessel is a critical parameter and was thoroughly monitored and documented.  Adsorptive
media performance is affected  by the EBCT,  which varies directly with volumetric  flow rate
through the vessel.

3.11.2 Experimental Objectives

The objective of this task was to accurately and fully document the operating conditions and
performance of the equipment.

3.11.3 Work Plan

During the verification test, treatment equipment operating  parameters were monitored  and
recorded on a routine basis.  This included a complete description of all applicable data.

3.11.4 Schedule

Table  3-7  presents the  schedule  for  observing and  recording  equipment  operation  and
performance data.
Table 3-7. Schedule for Observing and Recording Equipment Operation and Performance
Data
Operational Parameter
Action
Treated water flow rate


Filter  system feed water and treated  water
pressures


Total hours operated per day



Tasks performed during equipment operation

Numb er of hours per day operator attends to all
tasks related to the treatment process

Totalizer Meter Readings	
Check and record in logbook twice per day; adjust when >5%
above or below target. Record before and after adjustment.

Record in logbook: initial clean bed feed water and treated
water pressure at the start of the run; thereafter, record twice
per day.

Record at  end  of day  or  at beginning  of  the following
workday, as calculated from totalizer meter readings and flow
rate.

Record tasks performed daily in logbook.

Record number of hours required by operator  to accomplish
all tasks.

Record totalizer meter readings twice daily.	
                                             34

-------
3.11.5 Evaluation Criteria

The data developed from the Integrity and Capacity Tests were used to evaluate the performance
of the adsorption media filter.  An objective evaluation of the difficulty of operations was based
on an assessment of time required for process monitoring and hydraulic control.

3.12   Task 4: Data Management

3.12.1  Introduction

The data management system that was used  in this verification involved the use of computer
spreadsheet software and manual recording of system operating parameters.

3.12.2 Experimental Objectives

The objective of this task was to establish a viable structure for the recording and transmission of
field-testing data by Gannett Fleming, such that NSF  received sufficient and reliable data for
verification purposes.

3.12.3 Work Plan

The following procedures were implemented for data handling and data verification by Gannett
Fleming:

The field-testing operator recorded operating and water quality data and calculations by hand in a
laboratory notebook.

•      Daily measurements were recorded on specially prepared data log sheets.
•      The logbook is permanently bound with consecutively numbered pages.
•      The logbook indicates the starting and ending dates that apply to entries in the logbook.
•      All pages have appropriate headings to avoid entry omissions.
•      All logbook entries were made in black water-insoluble ink.
•      All corrections in the logbook were made by  placing one line through the erroneous
       information and were initialed by the field-testing operator.
•      The pilot operating  logs include a description  of the adsorptive  media  equipment,
       description of test run(s), names of visitors, description of any problems or issues,  etc;
       such descriptions were provided in addition to experimental calculations and other items.

The  original logbook was  photocopied  at  least once  per  week and  copies forwarded  to  the
Gannett Fleming project engineer.  This protocol not only eased referencing of the original data,
but offered protection of the original record of results.

The  database  for this verification test program  was  set up in the form of custom-designed
spreadsheets.  The spreadsheets were capable of storing and manipulating each monitored water
quality and operational  parameter from each task, each sampling location, and each sampling
time.   All  data from the laboratory notebooks and  data log sheets were entered into  the
                                           35

-------
appropriate spreadsheets.   Data entry  was conducted  off-site by  the  designated field-testing
operator.  All  recorded calculations were also checked at this time.  Following data entry, the
spreadsheet was printed out and the printout was checked against the handwritten data sheet by
another individual.  Any corrections were noted on the hard copies  and  corrected on the screen;
then a corrected version of the spreadsheet was printed out. Each step of the verification process
was initialed by the field-testing operator or supervisor performing the entry or verification step.

Each experiment (e.g., each test run) was assigned a run number that was then tied to the data
from the experiment through  each  step of data entry and analysis.   As  samples were collected
and sent to the PADEP and NSF Laboratories, the data were tracked by use of a system of run
numbers.  Data from the PADEP and NSF Laboratories were received and reviewed by the field-
testing operator. These data were  entered into the data spreadsheets, corrected,  and verified in
the same manner as the field data.

3.13   Task 5: Quality Assurance/Quality Control (QA/QC)

3.13.1  Introduction

Quality assurance and quality  control for the operation of the arsenic adsorption media filter and
the measured water quality parameters were maintained during the verification testing program,
as described in this section.

3.13.2  Experimental Objectives

The  objective  of this task was  to  maintain strict QA/QC methods and procedures during this
verification test. Maintenance of strict QA/QC procedures was important in that, if a  question
were to arise  when  analyzing or interpreting data collected for the arsenic adsorption media
filter, it would be possible to verify the exact conditions at the time of testing.

3.13.3  Work Plan

Equipment flow rates were verified and recorded on a routine basis.  A routine  daily  walk-
through during testing was established to verify each piece of equipment or instrumentation was
operating properly.    The  items listed below are in addition to any  specified checks outlined in
the analytical methods.

It  was  extremely  important that system flow rates be maintained at set values  and monitored
frequently. Doing so allowed maintenance of a constant and known EBCT in the adsorptive
media.   Adsorptive  media performance is  directly affected by the EBCT, which, in turn, is
proportional to the volumetric flow rate through the media.   Therefore, an important QA/QC
objective was  the maintenance of a constant volumetric flow rate through the adsorptive media
by frequent monitoring and  documentation for possible needed adjustment.   Documentation
included an average and standard deviation of recorded flow rates through the adsorptive media.
                                           36

-------
 •      Weekly QA/QC Verifications
       o      In-line  rotameter (clean any foulant  buildup, as needed, and verify flow rate
              volumetrically);
       o      In-line  totalizer meters (clean any foulant buildup, as needed,  and verify flow
              rate); and
       o      Tubing  (verify good  condition of  all  tubing  and  connections;  replace  as
              necessary).

3.13.4 Analytical Methods

The analytical methods utilized in this study for on-site and laboratory monitoring of adsorptive
media feed and treated water quality are described in the section below.

•      Arsenic
       Arsenic analyses were performed at the NSF Laboratory according to EPA Method
       200.8. These analyses were the most critical for the entire ETV test. Minimum analytical
       turnaround time was required to achieve optimum process control. This method required
       ultra-pure  (optimum) grade  nitric acid  be used,  not reagent grade,  to avoid the trace
       amounts of arsenic, which can be present in reagent grade nitric acid.

       Arsenic analyses were also performed on-site for qualitative  purposes.  These used the
       Model  QUICK Low Range  II field test  kit from Industrial Test Systems (ITS), Inc. The
       arsenic field test kit has an optimum accuracy below 6 (ig/L  and a reaction time of less
       than 15 minutes.  The complete method procedure is presented in Appendix I.

•      pH
       Analyses for pH were performed on-site according to Standard Method  4500-FTf B
       (Electrometric Method). A three-point calibration of the pH meter used in this study was
       performed once per day.  Certified  pH buffers 4.0, 7.0, and 10.0 were used. The pH
       electrode was stored in an appropriate solution, as  defined in the instrument manual.

•      Alkalinity
       Analyses for  alkalinity were  performed at  the  Gannett   Fleming Treatability  Lab
       according to Hach Method 8221 (Buret Titration Method).

•      Fluoride
       Analyses for fluoride were performed at the Gannett Fleming  Treatability Lab according
       to Standard Method 4500-F" C (Ion-Selective Electrode Method).

•      Chloride
       Analyses for chloride were  performed  at the PADEP Lab according to EPA Method
       300.0.

•      Sulfate
       Analyses for sulfate were performed at the PADEP Lab according to EPA Method 300.0.
                                           37

-------
•      Silica
       Analyses for silica were performed at the PADEP Lab according to EPA Method 200.7.

•      Aluminum
       Analyses for aluminum were performed at the PADEP Lab according to EPA Method
       200.7.

•      Total Phosphorus
       Analyses for phosphate were performed at the PADEP Lab according to EPA Method
       365.1.

•      Calcium
       Analyses for calcium were performed at the Gannett Fleming Treatability Lab according
       to Hach Method 8222 (Buret Method), with 0.020 N titrant.

•      Hardness
       Analyses for hardness were performed at the Gannett Fleming Treatability Lab according
       to Hach Method 8226 (ManVer 2 Buret Titration), with 0.020 N titrant.

•      Magnesium
       Magnesium results  were calculated by subtracting the calcium result (Hach Method
       8222) from the Hardness result (Hach Method 8226).

•      Iron
       Analyses for iron were performed at the PADEP according to EPA Method 200.7.

•      Manganese
       Analyses  for  manganese  were performed  at  the  PADEP Lab  according to EPA
       Method 200.7.

•      Turbidity
       Turbidity analyses were performed on-site according to Standard Method 2130 B using a
       portable turbidimeter.

•      Temperature
       Temperature was analyzed on-site according to Standard Method 2550 B.

3.13.5  Samples Shipped Off-Site for Analysis

Samples  for inorganic analysis,  including arsenic, chloride, sulfate,  silica, aluminum, total
phosphorus, iron, and manganese, were collected and preserved in accordance with Standard
Method 3010 B.  Particular  attention was paid to the sources of contamination as  outlined in
Standard Method  3010 C.   The samples were  refrigerated  at approximately  2° to  8°C
immediately upon  collection  (except  for  the arsenic  samples),  shipped  in  a cooler,  and
maintained  at a temperature of approximately 2° to 8°C.  The PADEP Lab maintained  the
                                          38

-------
samples at approximately 2° to 8°C until initiation of analysis.  Table 3-8 presents the sampling
protocol followed during the ETV for samples analyzed by the PADEP Laboratory.

Table 3-8. Water Quality Sampling Protocol
Sequence
Niimhpr
Paramftfar
Laboratory
Aluminum &
Silica
Iron&
Manganese
Sulfate &
Chloride
Total
Phosphorus
TCLP
Sample

125 mL
HDPE
125 mL
HDPE
500 mL
HDPE
125 mL
HDPE
Plastic
Sample Sample

100 mL Nitric Acid to
pH <2.0; iced
100 mL Nitric Acid to
pH <2.0; iced
250 mL Iced
lOOmL SulfuncAcid
to
pH <2.0; iced
N/A
Sample

6 months 101

6 months 201

28 days 201
28 days 201

N/A N/A
SA£)

102 107

202 106

202 106
202 106

N/A 242
NSF
Tfiit Tracking TP
Bottle Collector Date/Time
Par* TTl Wn PnlWtp H Tnt^nrih7 Psn^it^

M 1749 I II

M 1749 * I II

N/A 1749 * I II
P 1749 * I II

N/A 1749 * N/A II
'-1-1 Information also required on sample bottle.

3.14   Operations and Maintenance

Gannett Fleming reviewed Kinetico's O&M Manual; comments related to the applicability of the
manual are included in Chapter 4.  The Owner's Manual and Installation Guide are included in
Appendix N; the technical sheets  are on  file  at Gannett Fleming and NSF.   These manuals
present specific information on the mechanical operation of the filter tanks for a variety of media
types, including Actiguard AAFS50.
                                           39

-------
                                        Chapter 4
                                  Results and Discussion
4.1    Introduction
The ETV testing of Kinetico Inc.'s and Alcan Chemicals'  arsenic adsorption filter system was
conducted in two phases, including an Integrity Verification Test and an Adsorption Capacity
Test.   Prior to  initiation of the Integrity and Capacity Testing, equipment shakedown  was
performed; this included collection and analysis of two days of speciated feed and treated water
samples.  The two-week (13 full days plus 8 hours) Integrity Verification Test was initiated on
April  22, 2003 and concluded on May 5, 2003.  The initiation of the Adsorption Capacity Test
coincided with  the Integrity  Verification Test and continued  until an  arsenic breakthrough
concentration of 11 |J,g/L was detected in three consecutive treated water samples.  Following
confirmed breakthrough of arsenic, the treatment unit was shutdown on October 28, 2003.  Spent
media samples were collected on November 4, 2003, which  concluded the verification test.

This section of the ETV report presents  a summary of the equipment startup and preliminary
arsenic speciation sample analyses, results of the Integrity Verification Test,  results of the
Adsorption Capacity Test, and a discussion of the results. The results and discussion encompass
the concentration and speciation of arsenic in the feed and treated water, analysis of other key
feed and treated water  quality parameters, the quantity and rate  of treated water production,
backwash water quantity and water quality, spent media  analyses,  and equipment operation
characteristics, as well as quality assurance and quality control procedures.

4.2    Task 1:  System Integrity Verification Testing

The verification  test  site was the  Orchard  Hills MHP WTP,  located  in Carroll  Township,
Pennsylvania.  The WTP and arsenic adsorption filter system are described in detail in Chapter 2.

4.2.1   Equipment Installation, Startup, and Shakedown

The arsenic adsorption media filter system equipment was installed by Kinetico Inc. personnel in
September 2002. Initial arsenic speciation tests were performed on the feed and treated water in
December 2002,  prior to PSTP fmalization.   These initial arsenic tests were used to make a
preliminary assessment of the ability of the system to remove arsenic under the existing water
quality conditions at the site and to  evaluate the speciation of arsenic in the feed water.  During
the Integrity Verification Test, Gannett Fleming evaluated the reliability of equipment operation
under the environmental and hydraulic conditions  at the Orchard Hills MHP WTP site, while the
equipment was supplied feed water by Well No. 1. The  adsorption media filter was operated for
Integrity  Verification testing purposes for 13 days plus 8 hours within the  operational range
presented in the equipment design criteria.

Preliminary arsenic speciation analyses indicated a total feed  water arsenic concentration of
approximately 17  ng/L.   Arsenic III was not detected in the feed water above the  4  |j,g/L
detection limit.  Arsenic was not detected in the  treated water.  Preliminary arsenic speciation
results are presented in Table  4-1.  Analytical test reports and sample submission forms for the
                                            40

-------
preliminary arsenic speciation analyses are included in Appendix O.  The anion exchange resin
columns used for  these  preliminary  arsenic  speciations  were  later  found,  during  later
performance evaluation  testing, to be only approximately  70%  effective in the recovery of
arsenic  III.  Laboratory  arsenic analyses for the preliminary samples with an arsenic method
detection limit of 4  |j,g/L were performed at the PADEP Laboratory.  Subsequent speciations
were  made during the Integrity and Capacity Verification  Tests, with  a  new  batch of ion
exchange  columns  (prepared by  NSF).  The arsenic analyses were performed with a method
detection  limit of 2 [ig/L  at the  NSF Laboratory. These analyses indicated an arsenic  III
concentration  of  approximately 4  |j,g/L in the feed water  as  described later in this chapter.
Arsenic speciation using NSF-prepared ion exchange columns resulted in a 100% recovery of
arsenic III in performance evaluation  testing. Performance evaluation testing results for arsenic
speciation  and on-site water quality  analyses  are  presented later in this chapter.   The NSF-
prepared anion exchange resin columns were  used for  arsenic speciation during the Integrity
Verification and Adsorption Capacity testing.
Table 4-1. Preliminary Arsenic Speciation
Feed Water

Sample
Date
12/10/2002
12/11/2002
Total
Arsenic
(|jg/L)
16.7
17.2
Soluble
Arsenic
(|jg/L)
15.4
16.2

Arsenic III
(|jg/L)
<4.0
<4.0
Calculated
Arsenic V-1-1
(|jg/L)
>11.4
>12.2
Treated Water
Total
Arsenic
(|ig/L)
<4.0
<4.0
Soluble
Arsenic
(|jg/L)
<4.0
<4.0

Arsenic III
(|ig/L)
<4.0
<4.0
Calculated
Arsenic V
(|jg/L)
<4.0
<4.0
('   The laboratory minimum reporting limit is used for all statistical calculations. For preliminary (i.e., Shakedown)
    arsenic analyses only, the laboratory minimum reporting limit is 4 |ig/L.

Several physical modifications were made to the system prior to the initiation of testing on
April 22, 2003.  Modifications included installation  of a  second totalizer meter ahead of the
treatment unit, a Y-check valve,  and a pressure regulating valve located  downstream of the
treatment unit, but upstream of the diaphragm flow control valve.  The pressure regulating valve
was added in response to the widely variable WTP  pressures in order to maintain a constant
pressure at the diaphragm valve.  A constant pressure at the diaphragm valve allows a constant
and adjustable flow rate to be maintained through the treatment unit. The second totalizer meter
was added to function as a backup to the treated water totalizer meter and to allow calculation of
the estimated volume of water used during a  backwash cycle.  The manufacturer also replaced
the treatment unit control  module  with  a control module calibrated at their lab to automatically
initiate a filter backwash cycle at an interval of approximately 11,230 gallons of treated water.

The manufacturer installed new Actiguard AAFS50 media on February  11,  2003, prior to
initiation  of the  Integrity and  Capacity  Verification Testing.   The media installation  was
witnessed by Gannett Fleming. A platform scale and 5-gallon bucket were used to measure and
install  39.76 pounds of media in  each  of the two treatment unit tanks.  Following  the media
installation,  the manufacturer certified that the media installation, including the total weight of
media  installed into each tank,  met the manufacturer's requirements.   A copy of the  signed
certification is included in Appendix P.  The 39.76 pounds of dry, uncompacted media per unit
resulted in a "freeboard," or depth  to  the wetted, compacted media,  of approximately 18-1/4
inches  from the top of the media to the  top of the opening in each filter tank, as summarized in
                                            41

-------
 Table 4-2. The optimum freeboard, based on the manufacturer's specifications, is 17-1/2 inches.
 The freeboard was measured again following the Adsorption Capacity Test.  At the end of the
 testing, the depth to the wetted, compacted media was approximately 18-1/2 inches in the main
 tank and 19-1/2 inches in the remote tank.

 Table 4-2. Weight of Media Installed and Freeboard in Each Filter Tank
             Tank                     Media Weight (Ib)                  Freeboard (in.)
           Primary                          39.76                          18.25
	Remote	39.76	18.25	

 Based on the reported media density of 56.8  pounds per cubic  foot,  the 39.76 pounds of media
 installed per  unit should have resulted in a bed volume of approximately 0.7 cubic feet per tank,
 for a total bed volume of 1.4 cubic feet.  However, given a total tank height of 40 inches and a
 tank diameter of  8 inches, as reported by the  manufacturer, the actual bed volume was estimated
 to be approximately 0.63 cubic feet per tank, or approximately  1.27 cubic feet total.  The media
 volume was  calculated without accounting for the tank wall thickness, the round bottom of the
 tank, or subtraction of the volume of the internal flow distribution apparatus,  all of which could
 be significant. Therefore,  the bed volume was more  accurately measured following the test by
 sealing the internal flow distributor and carefully measuring the  amount of water required  to
 achieve the originally measured freeboard of  18-1/4 inches to the top  of the tank. The media bed
 volume, as determined by liquid measure, was  0.60  cubic feet per  tank for a  total media bed
 volume of 1.20 cubic feet.  Media bed volume  calculations are included  in Appendix D.  The
 PSTP indicated, "Data will be generated that will represent the actual volume of water treated by
 the 1.4 cubic feet of Actiguard AAFS50 media...".  This  difference in bed volume could make a
 significant difference in the apparent media capacity.  Therefore, the more  accurate total bed
 volume of 1.20 cubic feet was used for media capacity calculations, included later in this chapter.

 Equipment startup was performed by the manufacturer and witnessed by Gannett Fleming.  The
 protocol for startup is included in Alcan Chemicals' Technical Bulletin for  Actiguard AAFS50 in
 Appendix A. The  manufacturer specified the initial 10 bed volumes of treated water should  be
 used  as media flushing water and,  therefore,  should be  discounted prior to  recording the
 totalizers'  startup readings.  The treated water totalizer meter reading during  media installation,
 prior to any  flow through the newly installed media, was 471,665 gallons.  Prior to initiation  of
 the Integrity  Verification and Adsorption Capacity Testing on April 22, 2003, the totalizer meter
 reading was  472,015 gallons,  indicating 350  gallons had been used by the manufacturer during
 startup.  The corresponding feed water totalizer reading was 342 gallons.  The initial feed water
 totalizer reading  at installation was 0.0 gallons.  Based on an approximate media bed volume  of
 1.20 cubic feet, the actual volume of water wasted during startup was equal to approximately  39
 bed volumes, which is 3.9 times the stated 10 bed volumes required to pre-wash the media.  The
 manufacturer indicated the additional water was  used to verify proper operation of the filter unit
 control module.  Water used during startup was not included in  the treated water volume used to
 assess the capacity of the media.
                                            42

-------
4.2.2  Experimental Objectives

As established in the PSTP, the experimental objectives for Integrity Verification testing were as
follows:

•      Evaluate equipment operational reliability under field conditions;
•      Document feed water quality and arsenic concentration; and
•      Collect operational and water quality data under field conditions.

4.2.3  Integrity Test Operational Data

Following initiation of testing, the arsenic adsorption media filter system operated intermittently
in concert with the operation of Well No. 1.  However, during the Integrity Verification Test, the
treatment system was operated continuously for at least 2 hours daily and operated intermittently
during the  remaining 22 hours each day, as required in the ETV protocol.   The 2 hours of
continuous  operation per day were initiated using the  manual mode of operation for Well No.  1
at the WTP control panel and were witnessed by the Gannett Fleming field engineer.  During the
2-hour continuous operation period, a ball valve on the Well No. 1 discharge pipe was throttled
by the field engineer to provide the required minimum feed water pressure of 30 psi. Throttling
was necessary when only Well No. 1 was operating, because the low flow rate from a single well
resulted  in Ittle headloss through the WTP piping and treatment process.  The backpressure
measured at the treatment unit would have been  less than 20 psi, which  is less than  the required
minimum operating pressure for the Kinetico treatment unit, without throttling the well discharge
ball valve. .

Monitoring and on-site data collection were performed,  as scheduled, to verify the equipment
performance.  Table 4-3 summarizes the arsenic adsorption media filter unit  operational  data
during the  Integrity Verification Test.  Copies of the original logbook data sheets and compiled
Integrity Test operational data are included in Appendix F.

The  treatment unit operated for an average of 14 hours per day during  the Integrity Test.  The
combination  of the  pressure  regulating  valve  and  diaphragm valve  maintained  a relatively
constant flow rate, as shown.   However,  flowmeter calibration at the end  of the  Integrity
Verification Test indicated an  actual flow rate  of 2.0 gpm was produced when the rotameter
(flow rate meter) indicated a flow rate of 1.9 gpm.  Therefore, the average flow rate during the
Integrity Test was higher than the target of 1.9 gpm, but was within the manufacturer's specified
range of 1.8 to  2.0 gpm.  Following the Integrity Test, the flow rate set-point was adjusted and
verified  to produce  a rate  of 1.9  gpm.   The  adjusted set-point was maintained  during the
Adsorption Capacity Test.

The  feed water pressure averaged 56.4 psi during the Integrity Test, which was well within the
manufacturer's  specified operating  pressure  range of 30 psi to 125 psi.   Headloss across the
treatment unit was relatively low, with a pressure differential averaging 1.0 psi, and  did not
appear to vary  significantly as a function of run time during  the two-week test, as shown in
Figure 4-1.  This indicates  that, despite the particulate manganese and turbidity observed in the
                                            43

-------
feed water as discussed in Section 4.2.4, headless did not significantly accumulate between filter
backwashes.  Therefore, the production volume between backwashes could have been increased.
Table 4-3.
Integrity Test Operational
Flow Rate
Before Flow Rate After
Adjustment Adjustment
(gpm) (gpm)
Number of
Samples
Mean 1.99
Minimum 1.90
Maximum 2.00
Standard
Deviation
95% Confidence
T , 1 l.Vo ~ 2* \j\j
Interval
39
2.00
1.90
2.00
0.02
1.99-2.00
( ' During 2-hour continuous operation.


60 -
bU |
55 -

P.
2 1°





10 <

Data
Daily Run
Feed Treated Pressure Time
Pressure Pressure Differential Average
(psi) (psi) (psi) (hours/day)
30 30 30 39
56.4 55.4 1.0 13.8
53.0 51.5 0.5 12.6
60.0 59.0 1.5 24.0(1)
2.1 2.1 0.4 1.7
55.5-57.3 54.5-56.3 0.8-1.2 13.2-14.5







's^~\
r^f^^^
*^*^
•L •A'-.A'^dls. yfts. ..
-------
4.2.4  Integrity Test On-Site Water Quality Analyses

The results of on-site water quality analyses are summarized in Table 4-4.  Copies of the original
logbook data sheets and compiled Integrity  Test on-site  water quality data are  included in
Appendix F.
Table 4-4. Integrity Test On-Site Water Quality Data
Number of
Parameter Units Samples Mean Minimum
Feed Water
pH
Temperature
Turbidity
Alkalinity
Calcium
Magnesium
Hardness
Fluoride
Treated Water
pH
Temperature
Turbidity
Alkalinity
Calcium
Magnesium
Hardness
Fluoride

-
°C
NTU
mg/L
mg/L
mg/L
mg/L
mg/L

-
°C
NTU
mg/L
mg/L
mg/L
mg/L
mg/L

28
14
15
14
3
3
3
14

28
14
15
14
3
3
3
14

7.6
12.0
0.55
87
28.0
7.6
101
0.19

7.3
12.0
0.20
81
26.4
8.3
100
0.05

7.3
11.5
0.15
84
28.0
7.3
100
0.15

6.8
11.6
0.10
50
26.4
8.3
100
0.02
Maximum

7.7
12.3
3.9
90
28.0
8.3
104
0.27

7.6
12.3
0.75
90
26.4
8.3
100
0.12
Standard
Deviation

N/A
0.2
0.96
2 2
N/A
N/A
N/A
0.03

N/A
0.2
0.17
11
N/A
N/A
N/A
0.03
95%
Confidence
Interval

7.5-7.7
11.9-12.2
0- 1.2
86-89
N/A
N/A
N/A
0.17-0.21

7.2-7.5
11.9-12.2
0.10-0.30
74-89
N/A
N/A
N/A
0.03-0.07
N/A = Not Applicable. Standard Deviation and 95% Confidence Intervals were not calculated for parameters with
      data sets of fewer than 8 values.
                                              45

-------
The pH was reduced within the treatment unit during the two-week Integrity Verification Test, as
shown in  Figure 4-2.  This  reduction in pH is  a function of the ion  exchange process and
consumption of alkalinity, as shown in Figure 4-5.
      7.80
      7.60
       4/21/03     4/23/03     4/25/03     4/27/03
                                               4/29/03

                                                Time
                                                          5/1/03
                                                                   5/3/03
                                                                              5/5/03
                                                                                       5/7/03
                                            • Feed •
                                                     - Treated
Figure 4-2.  Integrity Test pH (4/22/03 to 5/5/03).

Due to a relatively  short hydraulic detention time, the feed and treated water temperatures were
nearly equal throughout the  test.  Feed water temperature varied less than  1°C during the two-
week test period, as shown in Figure 4-3.
      12.40
      12.30
      12.20
   cr
   ^
      12.10
      12.00
    g  11.90
    S,
    |  11-80
   H
      11.70
      11.60
      11.50
      11.40
                                                                X
        4/21/03     4/23/03     4/25/03     4/27/03     4/29/03      5/1/03

                                                Time
                                                                    5/3/03      5/5/03      5/7/03
                                        |   ^  Feed  ^  Treated |

Figure 4-3.  Integrity Test temperature (4/22/03 to 5/5/03).
                                              46

-------
Figure 4-4 shows the Integrity Test feed and treated water turbidity as a function of time.  The
feed water turbidity was generally low, averaging approximately 0.55 NTU, but was somewhat
variable.  The variability in feed water turbidity appeared to result from black particles, possibly
oxidized manganese, which periodically appeared in the feed water.  Treated water turbidity was
consistently very low, with a 95% confidence  interval of 0.10 to 0.30 NTU.  The lower treated
water turbidity likely was due to  physical removal or filtering by the filter unit media.
     4.50
     4.00
     3.50
   _
   H
     3.00
     2.50
   H  1.50
      1.00
     0.50
       4/21/03
                 4/23/03
                          4/25/03
                                    4/27/03
                                              4/29/03

                                               Time
                                                        5/1/03
                                                                  5/3/03
                                                                            5/5/03
                                                                                     5/7/03
                                           • Feed •
                                                   • Treated
Figure 4-4. Integrity Test turbidity (4/22/03 to 5/5/03).
                                             47

-------
As shown in Figure 4-5, the media consumed  approximately 38  mg/L as  CaCCb of alkalinity
during the initial day of the test.  Alkalinity consumption gradually decreased to nearly zero by
the end of the first week of operation.
       4/21/03
                 4/23/03
                          4/25/03
                                    4/27/03
                                              4/29/03

                                               Time
                                                        5/1/03
                                                                  5/3/03
                                                                            5/5/03
                                                                                     5/7/03
                                           • Feed •
                                                   • Treated
Figure 4-5. Integrity Test alkalinity concentration (4/22/03 to 5/5/03).
                                             48

-------
Initially, fluoride was nearly entirely removed through the treatment process, as shown in Figure
4-6. However, treated water fluoride levels gradually increased during the Integrity Test period.
The manufacturer has  indicated that fluoride competes with HAsO/f for adsorption.  However,
the media has a  lower affinity for fluoride than for arsenic.  Therefore, fluoride  breakthrough
should be observed prior to arsenic breakthrough as the total adsorption site area  is reduced,
resulting in arsenic out-competing fluoride for the remaining sites.  Integrity Test results indicate
fluoride removal  efficiency was decreasing as the Integrity Test ended.
     0.30
     0.25
     0.20
   "a
   v
   7s
   o
   I
         7      V
0.15
     0.10
       4/21/03
                4/23/03
                          4/25/03
                                   4/27/03
                                             4/29/03

                                              Time
                                                       5/1/03
                                                                5/3/03
                                                                          5/5/03
                                                                                    5/7/03
                                          • Feed •
                                                  • Treated
Figure 4-6. Integrity Test fluoride concentration (4/22/03 to 5/5/03).

Water quality analyses results indicate calcium, magnesium, and total hardness concentrations in
the feed water were apparently unaffected by the treatment process. However, Integrity Testing
included only three tests  for these parameters.   The Capacity Test provided additional data.
Therefore,  detailed analyses for hardness,  calcium,  and magnesium  are included only in the
Capacity Test results (Section 4.3).
                                            49

-------
4.2.5      Integrity Test Laboratory Water Quality Analyses

The results of Integrity Test water quality analyses performed at the PADEP Laboratory  are
summarized in Table 4-5.   Compiled data, copies of the  analytical test  reports, and sample
submission forms are included in Appendix G.  The raw data are on file at NSF.
Table 4-5. Integrity Test Laboratory Water Quality Data
Number of
Parameter Units Samples Mean Minimum
Feed Water
Silica
Aluminum
Iron
Manganese
Chloride
Sulfate
Total
Phosphorus
Treated Water
Silica
Aluminum
Iron
Manganese
Chloride
Sulfate
Total
Phosphorus

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

14
14
2
2
2
2
2

14
14
2
2
2
2
2

18.9
210
23
306
18.9
10.3
0.027

10.1
<200
<20
42
18.9
20.3
0.010

17.9
<200
<20
79
18.8
10.3
0.024

3.00
<200
<20
15
18.5
11.3
0.010
Maximum

19.7
339
26
532
19.0
10.3
0.030

14.3
<200
<20
69
19.2
29.2
0.010
Standard
Deviation

0.50
37.1
N/A
N/A
N/A
N/A
N/A

3.42
0
N/A
N/A
N/A
N/A
N/A
95%
Confidence
Interval

18.6-19.3
<200-235
N/A
N/A
N/A
N/A
N/A

7.82-12.4
<200 -<200
N/A
N/A
N/A
N/A
N/A
N/A = Not Applicable.  Standard Deviation and 95% Confidence Intervals were not calculated for parameters with
      fewer than 8 values.
Note:  The laboratory minimum reporting limit was used for statistical calculations for sample results less than the
      laboratory minimum reporting limit.
                                              50

-------
The analyses indicate silica was initially removed from the feed water by the treatment process.
However, silica concentrations in the treated water increased, as shown in Figure 4-7, during the
two-week Integrity Test.  Like fluoride, as discussed previously, silica competes with arsenic for
adsorption sites on the media. However, the media has a lower affinity for silica than for arsenic.
Therefore, the  increasing treated water silica concentration indicates that, as the total adsorption
site  area decreases,  the arsenic ions out-compete silica ions for the remaining sites. The ionic
preference series for Actiguard AAFS50 media is included in Table 2-3.
      25
      20
   3 15
   £
   2
   I  10
       o
     4/21/2003   4/23/2003   4/25/2003   4/27/2003   4/29/2003   5/1/2003    5/3/2003

                                              Time
                                                                          5/5/2003    5/7/2003
                                          • Feed —•— Treated
Figure 4-7. Integrity Test silica concentration (4/22/03 to 5/5/03).
                                             51

-------
Aluminum concentrations were apparently unaffected by the treatment process.  Only one feed
water sample result was greater than the MDL of 200 ng/L and no aluminum was detected in the
treated water.   These data indicate the media was not releasing aluminum to the treated water
above  detectable levels.  The feed and treated water aluminum concentrations are shown in
Figure 4-8.
     400

     350


     300
   ^25°

   I 200
      100
      50
       0

                                       \
.    .   .    /.\
      4/21/2003   4/23/2003
                         4/25/2003   4/27/2003
                                            4/29/2003

                                             Time
                                                      5/1/2003
                                                               5/3/2003
                                                                         5/5/2003
                                                                                  5/7/2003
                                          Feed
                                                   Treated
Figure 4-8. Integrity Test aluminum concentration (4/22/03 to 5/5/03).

Only two samples were collected for laboratory analyses for iron, manganese, chloride, sulfate,
and phosphorus  during  the Integrity Test.   Therefore,  the  description of  results  for these
parameters is included in the Capacity Test analyses (Section 4.3).

4.2.6 Integrity Test Arsenic Analyses

Feed  water  and treated water  arsenic  samples  were  collected  daily  during the  Integrity
Verification Test.  Three of the sample sets were speciated to determine the distribution of the
total soluble arsenic between the arsenic III and the arsenic V species.  The fraction of arsenic III
in the feed water affects the treatability of the water, because  arsenic III is non-ionic at normal
drinking water pH ranges  and is therefore generally more difficult to remove by ion exchange
treatment processes.   The results  of the laboratory  arsenic  analyses  performed  at the NSF
Laboratory are summarized in Table 4-6.   During the Integrity Test, the feed water total arsenic
concentration averaged 15 ng/L, with approximately 5  [ig/L as the arsenic III  species and 10
Hg/L as the arsenic V species.  Treated water arsenic concentrations were all less than or equal to
the 2 |j,g/L method detection limit during  the Integrity Test.  Approximately 2,337 bed volumes
were treated during approximately  178 hours of equipment run time.  Feed and treated arsenic
concentrations, as a function of treated water bed volumes, are shown in Figure  4-9.  Complete
arsenic  analyses results including a summary  table, analytical test reports, sample submission
forms, and raw data are included in Appendix Q.
                                            52

-------
Table 4-6.
Integrity
Test Laboratory
Arsenic Data
Feed Water


Number of
Samples
Mean
Minimum
Maximum
Standard
Deviation
Total
Arsenic
Qig/L)
14
15
14
17
0.83
95% Confidence . .
T , , 15-16
Interval
Soluble
Arsenic
Qig/L)
3
15
14
17
N/A
N/A
Arsenic
(HR/L;
3
5
4
6
N/A
N/A
Calculated
III Arsenic V
((jg/L)
3
10
8
12
N/A
N/A



Treated Water
Total
Arsenic
(|jg/L)
14
<2
<2
2
0
<2-<2
Soluble
Arsenic
((jg/L)
3
<2
<2
<2
N/A
N/A
Arsenic III
Qig/L)
3
<2
<2
<2
N/A
N/A
Calculated
Arsenic V
Qig/L)
3
<2
<2
<2
N/A
N/A
N/A =  Not Applicable. Standard Deviation and 95% Confidence Intervals were not calculated for parameters with
       fewer than 8 values.
Note:   The laboratory minimum reporting limit was used for statistical calculations for sample results less than the
       laboratory minimum reporting limit.
     10
                                                                            2,000
       0                500               1,000              1,500
                                          Treated Water Bed Volumes

                                           F*~Feed -•- Treated I
Figure 4-9.  Integrity Test arsenic concentration (4/22/03 to 5/5/03).


Field arsenic analyses were  also performed  using  the  ITS QUICK Low Range  II test kit to
monitor the feed and treated water arsenic concentrations onsite. On-site arsenic data is included
in the logbook copies in Appendix F.
                                                53

-------
4.3  Task 2: Adsorption Capacity Verification Testing

Adsorption Capacity Testing began on April 22, 2003, coinciding with the initiation of Integrity
Verification Testing.   Water quality  sampling and analysis, system  monitoring,  and  data
collection were performed  as scheduled in the test plan  and as described in Chapter 3.   The
treated water arsenic  concentration reached the pre-defined breakthrough concentration of 11
Hg/L on October 3, 2003.  The treatment system was shutdown on October 28, 2003, following
receipt of laboratory arsenic analyses results indicating more than three consecutive treated water
arsenic samples with an arsenic concentration greater than or equal to 11 ng/L. The treated water
arsenic  concentration reached 11  (ig/L following  approximately  2,350 hours  of equipment
operation and treatment of approximately 28,800 to 29,200  bed volumes of water, based on the
calculated media bed volume of 1.20 cubic feet. Spent media samples were collected by Gannett
Fleming, and the treatment unit was disassembled and removed by Kinetico Inc., on November
4, 2003.  The results of the Adsorption Capacity Testing are detailed in the following sections.
Adsorption Capacity Test data include data collected during the Integrity Test.

4.3.1 Experimental Objectives

The experimental objective of the Adsorption Capacity Testing is to provide operating and water
quality data relative to the ability of the arsenic adsorption media filter system to remove arsenic
from feed water under field conditions.

4.3.2  Capacity Test Operational Data

The treatment unit operated intermittently in concert with the operation of Well No. 1 during the
Capacity Test. Well No. 1 was operated in manual mode only to provide continuous flow for the
filter backwashes, which were observed and sampled by Gannett Fleming. Monitoring  and on-
site data collection were performed as scheduled to verify the equipment performance. Table 4-7
summarizes the arsenic adsorption media filter unit operational  data during the Capacity Test.
Copies of the  original logbook data  sheets  and  compiled operational  data are included  in
Appendix F.  The non-integral flow control  system, consisting of a pressure regulating valve and
diaphragm valve, maintained a relatively constant flow rate averaging 1.9 gpm.
                                           54

-------
 Table 4-7.  Capacity Test Operational Data



Number of
Samples
Mean
Minimum
Maximum
Standard
Deviation
95% Confidence
Interval
Before
Flow Rate
Adjustment
(gpm)
384
1.90
0.00
2.00
0.11
1.89- 1.91
After
Flow Rate
Adjustment
(gpm)
385
1.91
1.80
2.00
0.04
1.90- 1.91

Feed
Pressure
(psi)
375
51.2
27.0
60.0
5.1
50.6-51.8

Treated
Pressure
(psi)
375
50.1
24.0
59.0
5.2
49.5-50.7

Pressure
Differential
(psi)
375
1.1
0.5
3.0
0.3
1.0- 1.1

Daily Run Time
Average
(hours/day)
384
14.2
12.6
24.0(1)
0.6
14.1 - 14.3
   During 2-hour continuous operation.

The equipment operated approximately  14 hours per day, on average.  The feed water pressure
was maintained within the manufacturer's recommended pressure limits of 30 to 125 psi, with
the exception of one day during which the recorded feed water pressure was only 27 psi.  The
filter bed headloss did not accumulate significantly as a function of run time, as shown in Figure
4-10.  Headloss across the treatment unit averaged 1.1 psi, only slightly greater than the 1.0 psi
average headloss  observed during the two-week Integrity Test. However, the headloss became
more  variable and reached the maximum pressure differential observed during the test as  the
media capacity for arsenic removal reached exhaustion.  The clean-bed headloss,  observed at the
initiation of testing, was approximately 0.5 psi.
     75
                   500
                               1000         1500         2000

                                   Cumulative Run Time (hours)
                                                                   2500
                                                                               3000
                             "Feed Pressure
                                          • Treated Pressure
                                                         •Headloss
Figure 4-10. Capacity Test headloss and pressure as a function of cumulative run time.
                                            55

-------
4.3.3 Capacity Test On-Site Water Quality Analyses

The results of on-site water quality analyses are summarized in Table 4-8. Copies of the original
logbook  data sheets and compiled Integrity  Test  on-site water quality data are included in
Appendix F.
Table 4-8. Capacity Test On-Site Water Quality Data
Number of
Parameter Units Samples Mean Minimum
Feed Water
pH
Temperature
Turbidity
Alkalinity
Calciun
Magnesium
Hardness
Fluoride
Treated Water
pH
Temperature
Turbidity
Alkalinity
Calciun
Magnesium
Hardness
Fluoride

-
°C
NTU
mg/L
mg/L
mg/L
mg/L
mg/L

-
°C
NTU
mg/L
mg/L
mg/L
mg/L
mg/L

198
184
184
84
27
27
27
39

198
184
184
84
27
27
27
39

7.6
13.8
0.25
89
26.0
8.3
99
0.17

7.5
13.8
0.15
88
25.8
8.4
99
0.12

7.3
11.5
0.10
84
24.8
7.3
96
0.13

6.8
11.6
0.05
50
24.0
7.3
96
0.02
Maximum

7.8
15.5
3.9
92
28.0
8.7
104
0.27

7.8
15.7
0.75
92
26.4
9.2
100
0.17
Standard
Deviation

N/A
0.94
0.30
1.5
0.92
0.50
1.7
0.03

N/A
0.94
0.10
5.4
0.58
0.41
1.3
0.05
95%
Confidence
Interval

7.6-
13.6-
0.20-
89-
25.6-
8.1-
98-
0.16-

7.5-
13.6-
0.10-
87-
25.6-
8.2-
99-
0.10-

7.6
13.9
0.30
89
26.4
8.5
100
0.18

7.6
13.9
0.15
89
26.1
8.6
100
0.14
As  discussed in the Integrity  Test results  (Section 4.2.4), the treatment process significantly
reduced the pH from the feed  water compared to the treated water, at the beginning of the test.
The pH reduction is likely a function of the removal of alkalinity. Following the initial period of
approximately two weeks  of  significant pH reduction, the feed and  treated water pH were
essentially  equal for the remainder of the Capacity Test, as shown in Figure 4-11.  On average,
the  treated water pH was nearly equal to the feed water pH.
                                           56

-------
      6.80
      6.60
       3/20/03
                        5/9/03
                                        i/28/03           8/17/03
                                                Time
                                                                       10/6/03
                                                                                       11/25/03
                                            "Feed
                                                    -Treated
Figure 4-11.  Capacity Test pH.
Due to the relatively short hydraulic detention time, the feed and treated water temperatures were
nearly equal throughout the test, as shown in Figure 4-12.  Due to seasonal temperature changes,
the water temperature varied by approximately 4°C during the test.
      18.00
      16.00
      14.00
      12.00
      10.00
    e.
    g
      4.00
      2.00
        3/20/03
                        5/9/03
                                       6/28/03           8/17/03
                                                Time
                                                                       10/6/03
                                                                                       11/25/03
                                         |  '  Feed   '  Treated |
Figure 4-12.  Capacity Test temperature.
                                              57

-------
With the exception of several  brief feed water turbidity spikes, the feed water turbidity  was
generally low,  averaging  less  than  0.25  NTU.   Black particles,  believed to  be  oxidized
manganese particles, were  often observed in the  feed water during the turbidity spikes.   The
treated water turbidity was  also consistently low, averaging 0.15 NTU.  The lower treated water
turbidity was likely due to  filtering by the treatment unit.  The feed water turbidity and treated
water turbidity observed during the Capacity Test are shown in Figure 4-13.
     4.50
     4.00
     3.50
   _
   H
   3
     3.00
     2.50
   •a 2.00
      1.50
       3/20/03
                       5/9/03
                                      6/28/03
                                                     8/17/03
                                                                     10/6/03
                                                                                    11/25/03
                                              Time
                                           • Feed •
                                                   • Treated
Figure 4-13. Capacity Test turbidity.
                                             58

-------
As discussed in the Integrity Test results, the treatment process consumed alkalinity during the
first  week of operation.   Following  the  first week  of operation, the feed and treated water
alkalinity was essentially equal, as shown in Figure 4-14.
     100
   I
       3/20/03
                                                                     10/6/03
                                                                                    11/25/03
                                      \  *  Feed   *  Treated \

Figure 4-14. Capacity Test alkalinity concentration.
                                             59

-------
Initially, fluoride was almost entirely  removed through the treatment process.  However,  as
shown in Figure 4-15, treated water fluoride levels gradually increased during the Integrity Test
period.   The  manufacturer has  indicated  fluoride  competes  with  HAsO/f for  adsorption.
However, the media has a  lower affinity  for  fluoride than for arsenic.  Therefore, fluoride
breakthough  should be  observed prior to  arsenic breakthrough, as  arsenic ions out-compete
fluoride  ions  for the  remaining sites.   Capacity Test results indicate that  complete fluoride
breakthrough  occurred  by  the  end of the third  week  of  testing, following treatment  of
approximately 3,600 bed volumes.
     0.30
       3/20/03
                      5/9/03
                                     6/28/03          8/17/03

                                              Date
                                                                    10/6/03
                                                                                   11/25/03
                                          • Feed  "  Treated |
Figure 4-15. Capacity Test fluoride concentration.
                                            60

-------
Capacity  test  water  quality  analyses  indicate  calcium,  magnesium,  and  total  hardness
concentrations in the  feed water were relatively  consistent during the test period and were
apparently unaffected by the treatment process, as shown in Figure 4-16.
      120.0
      100.0
   •S  60.0
   g
   o
      40.0
      20.0
        3/20/03
                        5/9/03
                                       6/28/03          8/17/03
                                                Time
                                                                       10/6/03
                                                                                      11/25/03
                         • Calcium Feed
                          Magnesium Treated
• Calcium Treated
• Hardness Feed
• Magnesium Feed
 Hardness Treated
Figure 4-16.  Capacity Test calcium, magnesium, and total hardness.
                                              61

-------
4.3.4 Capacity Test Laboratory Water Quality Analyses

The results of water  quality analyses performed at the PADEP Laboratory are summarized in
Table  4-9.  Laboratory water quality data are  summarized and the analytical test reports and
sample submission forms are included in Appendix G. The raw data are on file at NSF.
Table 4-9. Capacity Test Laboratory Water Quality Data
Number of
Parameter Units Samples Mean Minimum
Feed Water
Silica
Aluminum
Iron
Manganese
Chloride
Sulfate
Total
Phosphorus
Treated Water
Silica
Aluminum
Iron«
Manganese
Chloride
Sulfate
Total
Phosphorus

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

40
40
28
28
28
28
28

40
40
27
28
28
28
28

19.0
203
34
144
18.7
10.5
0.032

15.3
<200
21
12
18.8
11.3
0.014

17.4
<200
<20
36
16.8
10.1
0.024

3.00
<200
<20
<10
17.0
10.3
0.010
Maximum

21.1
339
116
1481
20.4
11.2
0.043

20.4
<200
32
69
20.2
29.2
0.023
Standard
Deviation

0.80
22.0
24
286
0.85
0.26
0.005

4.46
0
4
11
0.82
3.5
0.004
95%
Confidence
Interval

18.7-
<200-
23-
16-
18.3-
10.4-
0.029 -

13.7-
<200-
<20-
<10-
18.4-
9.7-
0.012-

19.3
-212
45
272
19.1
10.6
0.034

17.0
<200
-23
- 17
19.2
12.9
0.016
([)   The treated water iron concentration of 666 (ig/L on 7/3/03, as reported by the laboratory, was believed to be in
    error and was not included in the statistical analyses.

The analyses indicate silica was initially removed from the feed water by the treatment process.
However, as shown in Figure 4-17, silica concentrations in the treated water increased during the
capacity test, until a complete breakthrough was achieved  and the feed and treated water silica
concentrations were equal.  Like fluoride, as discussed above,  silica competes with arsenic for
adsorption on the media.  The media has a lower affinity for silica than for arsenic.   Therefore,
the increasing  treated water  silica concentration  indicates the total adsorption site area  has
decreased to the point where arsenic ions out-compete silica ions for the remaining media sites.
The ionic preference series for Actiguard AAFS50 media is included in Table 2-3.
                                             62

-------
      3/20/2003
                      5/9/2003
                                     6/28/2003
                                                    8/17/2003
                                                                    10/6/2003
                                                                                   11/25/2003
                                              Time
                                           •Feed
                                                   -Treated
Figure 4-17.  Capacity Test silica concentration.

Aluminum concentrations were apparently unaffected by the treatment process.  Only one feed
water sample result was greater than the MDL of 200 ng/L and no aluminum was detected in the
treated water.  These data indicate that the media was not releasing aluminum to the treated
water above detectable levels.  The feed and treated water aluminum concentrations are shown in
Figure 4-18.
      400
      350
      3/20/2003
                      5/9/2003
                                     6/28/2003         8/17/2003

                                              Time
                                                                    10/6/2003
                                                                                   11/25/2003
                                      |  +  Feed   ^  Treated!

Figure 4-18.  Capacity Test aluminum concentration.
                                             63

-------
Treated water iron levels were reduced in the treatment process to at or near the MDL of 20
     , as shown in Figure 4-19.
    140
    120
          Note: The treated water iron concentration of 666 ug/L on 7/3/03
          was believed to be erroneous and was not included in
          the statistical calculations or in this data plot.
     3/20/2003
                      5/9/2003
                                      6/28/2003           8/17/2003
                                                Time
                                                                        10/6/2003
                                                                                        11/25/2003
                                             -Feed -*- Ti
reated I
Figure 4-19. Capacity Test iron concentration.

The feed water manganese concentration averaged 144  (ig/L during the Capacity Test.  Feed
water manganese  concentrations were somewhat variable, with concentrations spiking during
periods when particles  of oxidized manganese were observed in the feed water.  As shown in
Figure 4-20,  manganese  in the  feed water was  removed  in the  treatment process to  a
concentration at or below the MDL of 10 |j,g/L for most of the weekly water quality samples. A
portion of the manganese may have been removed as a result of physical removal (i.e., filtration)
of paniculate manganese.    During the filter backwashes observed by Gannett Fleming,  the
backwash water was black in color and had manganese concentrations  of 5,620 to 17,500 ng/L.
These high levels of manganese in the backwash water indicate some manganese was physically
filtered from the water  and was easily removed during backwash, rather than adsorbed  onto the
filter media.
                                            64

-------
      1600
      1400
       3/20/2003
                        5/9/2003
                                       6/28/2003         8/17/2003

                                                 Time


                                           *  Feed —B—Treated
                                                                        10/6/2003
                                                                                       11/25/2003
Figure 4-20. Capacity Test manganese concentration.

Chloride concentrations  were apparently unaffected by  the treatment process,  as shown in
Figure 4-21.
       25
       20 '
       15
    o
   3   10
       0
      3/20/2003
                       5/9/2003
                                       6/28/2003         8/17/2003

                                                 Time
                                                                        10/6/2003
                                                                                       11/25/2003
                                        |   *  Feed ~  ~Treated]

Figure 4-21. Capacity Test chloride concentration.
                                                65

-------
Sulfate concentrations were apparently unaffected by the treatment process during most of the
Capacity Test, with an increase in the average sulfate concentration in the treated water of less
than 1 mg/L.  However, as shown in Figure  4-22, during the first few weeks of operation, the
treated water sulfate  concentration was  greater than  the  feed water  concentration, possibly
indicating  the treatment equipment  or  media were  contributing to the  treated water sulfate
concentration.
   1
      3/20/2003
                      5/9/2003
                                    6/28/2003
                                                                   10/6/2003
                                                                                  11/25/2003
                                              Time
                                           Feed  ^  Treated |
Figure 4-22. Capacity Test sulfate concentration.
                                             66

-------
As  shown in Figure 4-23, phosphorus was initially removed from the feed water to below the
MDL (0.010 mg/L).  As the media adsorption capacity was consumed, phosphorus removal
efficiency decreased and the treated water phosphorus concentration began to approach the feed
water concentration.
      0.05
       3/20/2003
                      5/9/2003
                                    6/28/2003        8/17/2003

                                             Time
                                                                 10/6/2003
                                                                                11/25/2003
                                         • Feed •
                                                 • Treated
Figure 4-23. Capacity Test phosphorus concentration.

4.3.5 Capacity Test Arsenic Analyses

The results of arsenic analyses performed by the NSF Laboratory are summarized in Table 4-10.
Feed  water  and treated  water arsenic samples were collected  daily  during  the Integrity
Verification  Test  and  weekly  during the  Capacity  Test.   As  the treated water  arsenic
concentration approached the pre-defined breakthrough concentration of 11 ng/L,  samples were
collected  three times per week.  Seven of the sample sets were  speciated to determine the
distribution of total soluble arsenic between the arsenic  III and the arsenic V species.  The
fraction of arsenic III in the feed water affects the treatability of the  water,  because arsenic III is
generally  more difficult to remove by known treatment  processes.   The feed water total arsenic
concentration averaged approximately  14  ng/L, with approximately 4 |j,g/L as the arsenic III
species and 10 ng/L as the  arsenic V  species.  As described in the previous section, the feed
water manganese   concentration  was  significant and  was  observed to  include  particulate
manganese, which  could impact  the apparent  arsenic  removal  capacity  of the media  by
enhancing arsenic removal.

Treated water arsenic concentrations were all less than or  equal to the 2 |j,g/L method detection
limit during the initial  5 weeks of testing, which included approximately 621  to 727 hours of
equipment operation and  approximately 8,000 to 9,113 bed volumes of water treated.  The
treated water arsenic  concentration reached  11  |j,g/L following  2,350  hours of equipment
                                            67

-------
operation and treatment of approximately 28,800 to 29,200 bed volumes of water, based on the
calculated media  bed volume  of 1.20  cubic feet. The treated water  arsenic  concentration
increased slowly to the pre-defined breakthrough  concentration.  A steep breakthrough curve,
which is typical with ion  exchange process, did not occur.  The  arsenic breakthrough may have
been slowed by mixing of the filter unit media during filter backwashes.  Feed and treated water
arsenic  concentrations as a function of treated water bed volumes are shown in Figure  4-24.
Complete arsenic  analyses results, including a summary table, analytical test reports, raw data,
and sample  submission forms, are included in Appendix Q.
Table 4-10.
Capacity
Test Laboratory Arsenic Data
Feed Water



Number of
Samples
Mean
Minimum
Maximum
Standard
Deviation
Total
Arsenic
0±g/L)
47

14
12
17

1.1
95% Confidence
T, -I IT- IT-
Interval

Soluble
Arsenic
(HgflO
7

14
13
15

N/A
N/A


Arsenic III
0±g/L)
7

4
<2
6

N/A
N/A

Calculated
Arsenic V
0±g/L)
7

10
8
12

N/A
N/A

Treated Water
Total
Arsenic
0±g/L)
47

6
<2
13

4
5-7

Soluble
Arsenic
0±g/L)
7

4
<2
10

N/A
N/A


Arsenic III
0±g/L)
7

<2
<2
<2

N/A
N/A

Calculated
Arsenic V
0±g/L)
7

3
1
8

N/A
N/A

N/A = Not Applicable.  Standard Deviation and 95% Confidence Intervals were not calculated for parameters with
      fewer than 8 values.
                                        15,000        20,000
                                       Treated Water Bed Volumes

                                        I * Feed ^  Treated I

Figure 4-24. Capacity Test arsenic concentration.
                                            68

-------
4.4  Equipment Operation

During  the  Verification  Test,  minimal time  and/or  attention were  required to  operate the
equipment, although significant time was spent on-site for testing purposes.  The time required
for daily  operation of the treatment unit  included  a 5-minute check of the flow rate  and
verification there  were no leaks in the system.  Permanent installation of the equipment would
also require periodic on-site arsenic analyses (requiring approximately  15  to 20 minutes to
perform),  and/or collection of samples for laboratory arsenic analyses.  The filter unit control
module automatically initiated filter backwashes, with no operator attention required.

4.5  Backwash Water Quality, Quantity, and Flow Rate

The filter unit control  module automatically initiated  filter backwashes,  with no  operator
attention required.  The  unit  backwashed  a  single  filter  unit  at  an interval  of 11,000 to
12,000 gallons of treated  water.  Therefore,  each filter was backwashed at an interval of 22,000
to 24,000 gallons. The filter unit not being backwashed continued to operate and produce treated
water (for consumption, but discharged to waste for this  test), treated water used for the filter
backwash and purge, and treated water used for control module drive water.  During the filter
backwashes, which were witnessed by Gannett Fleming, it was observed that the high combined
flow rate  through the unit resulted in a headloss of approximately 10 psi.  During  manually
initiated backwashes,  Well No. 1 was operated in manual mode, with the well discharge ball
valve set to maintain a minimum pressure of 30 psi.  Due to the additional headloss during the
backwash cycle, the treated water pressure was reduced to  less than 20 psi, which was the setting
of the non-integral pressure-regulating valve used in the  flow control system.  Therefore, the
treated water production was reduced to approximately 1.2 gpm.  Four filter backwashes were
initiated and witnessed  by the Gannett Fleming field-test engineer.   Backwash, purge,  and
control  module drive water flow  rate,  total quantity of flow, and water quality results are
summarized in Tables 4-11 through 4-13. The backwash  water was generally highly turbid and
black in color, which correlates with the very  high concentration of manganese detected in the
laboratory samples.  The elevated level of iron in the backwash water was unexpected given that
feed and treated water iron analyses results were primarily less than the 20  (ig/L detection limit.
The backwash water iron concentration could be a result of the buildup of particulate iron from
the feed water on the media and/or the result of media attrition.

The backwash water arsenic concentration averaged 24 |J,g/L, which is  significantly greater than
the average feed water arsenic concentration of approximately 14 ng/L. The increased arsenic
concentration in the backwash water could have resulted  from the removal of adsorbed arsenic
buildup within the filter unit or, more likely, from the removal  of arsenic associated with the iron
and manganese in the backwash.  The source of arsenic  in the backwash  could also  be media
attrition. The aluminum concentration in the backwash water was greater than concentrations in
the feed water for the first two backwashes sampled, indicating the media may have contributed
to the level of aluminum  in the backwash water.  The third and fourth  sampled backwash water
aluminum samples had concentrations less than the MDL.

During  the backwash cycle, a high flow rate  of more  than 4 gpm (backwash and production
flow) was passing through a single treatment unit tank.  This flow rate is much greater than the
normal production rate of 1.9 gpm; the minimal contact time of less than 1 minute could be the


                                           69

-------
cause of arsenic concentrations in the purge and drive water samples greater than those in the
treated water samples.

The  automatic filter backwash process occurred regularly, as described in the manufacturer's
literature. However, the total volume of backwash water, flow rate, and time varied somewhat
from  the manufacturer's  specifications.    The filter  backwash duration  was  approximately
18.3 minutes at a flow rate of 3.0 gpm, compared to the specified filter backwash of 13 minutes
at 4.0  gpm.  Similarly, the purge cycle spanned approximately 5.9 minutes  at a flow rate of
approximately 3.0 gpm, compared to a specified flow rate of 1.9 gpm for a 5-minute period.  The
time between the backwash and rinse (purge) stages was just under 1 minute, as opposed to the
specified 3-minute interval. Also, the total volume of backwash and rinse water was indicated in
the equipment specifications as 62 gallons.   The actual water use for backwash and rinse was
approximately 73 gallons, with an additional 9.8 gallons used for control module drive water, for
a total  usage of approximately 83 gallons for the entire backwash cycle.  Given a total production
of approximately  11,000 to 12,000  gallons between  filter backwash  cycles, the quantity of
backwash water used represents less than 1% of the total production.  Backwash water quality
characteristics are sourcewater-dependent. The impact of this backwash water on the wastewater
treatment plant NPDES permit requirements was not evaluated.
Table 4-11. Backwash Water Characteristics
Date/Time
5/5/2003
7/3/2003
9/25/2003
10/28/2003
Number of
Samples
Average
Minimum
Maximum
Volume
(gallons)
55.5
55.0
56.0
55.0
4
55.4
55.0
56.0
Duration
(min.)
16.50
18.58
18.08
20.17
4
18.33
16.50
20.17
Flow Rate
(gpm)
3.1
3.0
3.1
2.7
4
3.0
2.7
3.1
pH
(unit)
7.48
7.55
7.56
7.50
4
7.5
7.5
7.6
Turbidity
(MTU)
42.9
31.0
15.5
15.0
4
26.1
15.0
42.9
Arsenic
(|ig/L)
24
23
27
21
4
24
21
27
Iron
(|ig/L)
2,690
1,250
1,111
1,440
4
1,623
1,111
2,690
Manganese
(|jg/L)
17,500
5,650
5,751
5,620
4
8,630
5,620
17,500
Aluminum
(|jg/L)
658
201
<200
<200
4
315
200
658
Silica
(mg/L)
17.1
19.3
18.7
18.4
4
18.4
17.1
19.3

Table 4-12. Purge Water Characteristics

           Volume Duration Flow Rate    pH
Date/Time   (gallons)  (min.)	(gpm)     (unit)
Turbidity  Arsenic
 (NTU)
                                                         Iron
Manganese  Aluminum   Silica
                     (mg/L)
5/5/2003
7/3/2003
9/25/2003
10/28/2003
Number of

Samples
Average
Minimum
Maximum
18.0
17.5
18.0
17.8

4

17.8
17.5
18.0
5.58
5.75
5.55
6.67

4

5.89
5.55
6.67
3.2
3.0
3.2
2.7

4

3.0
2.7
3.2
7.55
7.64
7.53
7.50

4

7.6
7.5
7.6
0.40
0.45
0.40
0.44

4

0.40
0.40
0.45
3
7
11
12

4

8
3
12
<20
25
44
46

4

34
<20
46
42
39
48
39

4

42
39
48
<200
<200
<200
<200

4

<200
<200
<200
15.0
19.1
17.9
18.2

4

17.5
15.0
19.1
                                           70

-------
Table 4-13.  Control Module Drive Water Characteristics

Date/Time
5/5/2003
7/3/2003
9/25/2003
10/28/2003
Number of
Samples
Average
Minimum
Maximum
Volume
(gallons)
9.75
9.75
9.75
9.80
4
9.8
9.8
9.8
Duration
(min.)
24.50
26.60
25.80
29.00
4
26.48
24.50
29.00
Flow Rate
(gpm)
0.4
0.37
0.38
0.34
4
0.4
0.3
0.4
pH
(unit)
7.56
7.57
7.54
7.54
4
7.6
7.5
7.6
Turbidity
(MTU)
0.14
0.50
0.14
0.17
4
0.25
0.15
0.50
Arsenic
(|jg/L)
3
6
11
12
4
8
3
12
Iron
(|jg/L)
<20
368
<20
<20
4
107
<20
368
Manganese Aluminum
(Hg/L) (M-g^L)
<10 <200
<10 <200
<10 <200
<10 <200
4 4
<10 <200
<10 <200
<10 <200
Silica
(mg/L)
16.2
19.3
19.2
18.0
4
18.2
16.2
19.3
    The original backwash operational and on-site water  quality data are included in the logbook
    copies in Appendix F.  Laboratory water quality analyses reports are included in Appendix G and
    Appendix Q.

    4.6  Spent Media Analyses

    Following completion of the Adsorption Capacity Test, spent media core samples were extracted
    from each filter tank, for the purposes of verification testing, using  a 1.5-inch  diameter, thin-
    walled copper tube. The core samples were combined and thoroughly mixed in a large plastic
    bag, then divided into two separate samples, one for TCLP  and CA WET analyses to verify the
    spent media exhibits no toxicity characteristics, and one for a media gradation analysis.

    The complete results of TCLP and CA WET analyses, including QA/QC data, are included in
    Appendix J.  The results are summarized in Table 4-14.  Arsenic was not detected in the TCLP
    analysis of the spent media. Only barium and cadmium were detected in TCLP analyses, both at
    concentrations less than the regulatory limit (RCRA).  The arsenic concentration detected by CA
    WET  analyses was 0.25 mg/L (250 ng/L), well below the regulatory limit of 5 mg/L.  Other
    metals detected by CA WET analyses included barium,  cadmium, copper, and zinc.  All
    concentrations were less than the regulatory limits.
                                              71

-------
Table 4-14. Spent Media Characterization
Parameter
Arsenic
Barium
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Silver
Zinc
Result
(mg/L)
ND
4.12(3)
0.015
ND
ND
ND
ND
ND
ND
ND
TCLP
Reporting Limit
(mg/L)
0.20
0.20
0.010
0.080
0.020
0.10
0.0004
0.010
0.010
0.20
CA
Result
(mg/L)
0.25
6.30(3)
0.032(3)
ND
0.13
ND
ND(3)
ND(3)
ND
0.32
WET
Reporting Limit
(mg/L)
0.20
0.20
0.010
0.050
0.010
0.10
0.0040
0.010
0.010
0.05
TCLP(1)
Regulatory Limit
(mg/L)
5.0
100.0
1.0
5.0
N/A
5.0
0.2
N/A
5.0
N/A
CAWET(2)
Regulatory Limit
(mg/L)
5.0
100.0
1.0
5.0
25
5.0
0.2
20
5.0
250
(1)  40 CFR 261.24 Toxicity Characteristics.
(2)  California Regulations 66261.24.
^  Laboratory data qualifications included in Appendix J.
ND =  Non-Detect.
N/A= Not Applicable.

Visual observation and comparison  of the spent media and new media revealed no observable
physical  degradation. This  observation was supported by  gradation  analyses performed by
Gannett Fleming, the results of which indicated almost identical new and spent media particle
size distributions. Gradation analyses reports are included in Appendix L.

4.7    Task 3:  Documentation of Operating Conditions and Treatment Equipment

4.7.1   Introduction

During each day of verification testing, arsenic  adsorption media filter operating conditions and
treatment equipment performance  were  documented,  as  described  in  Section  3.11.  The
volumetric flow  rate through  an adsorptive  media  vessel is a  critical parameter  and was
thoroughly monitored and documented. Adsorptive media performance is affected by the EBCT,
which varies directly with volumetric flow rate through a vessel.

4.7.2   Experimental Objectives

The objective of this task was  to accurately and fully document the operating conditions and
performance of the equipment.   This task was  performed in conjunction with both the system
Integrity  Verification Testing and the Adsorption Capacity Verification Testing.
                                           72

-------
4.7.3   Operations and Maintenance

The following are recommendations for criteria to be included in the Para-Flo™ Operation and
Maintenance  (O&M)  Manual  for  adsorptive media removal of arsenic, as described  in the
Technology Specific Test Plan (TSTP) within the ETV Protocol.

4.7.3.1 Operations.  Kinetico  Inc. provided an Owner's Manual and Installation Guide, which
provided much of the data and information needed to conduct the test. Technical sheets intended
for Gannett Fleming and NSF review only  and  not for publication were also submitted. The
Owner's  Manual and Installation Guide are included in Appendix N; the technical sheets are on
file at Gannett Fleming and NSF. These manuals  present specific information on the mechanical
operation of the filter tanks for a variety of media types, which include Actiguard AAFS50.

Kinetico  Inc.  and Alcan Chemicals provided readily understood  information on the required or
recommended procedures (task-specific SOPs) related  to the proper operation of the arsenic
adsorption media filter. Gannett Fleming  discussed the following issues with Kinetico Inc. and
Alcan Chemicals prior to testing:

•      Monitoring of Preconditioning of Adsorptive Media
       o      Utilizing the manufacturer's specific procedure for Actiguard AAFS50 adsorptive
              media, including backwashing initially with at least  10  bed volumes to remove
              fines;
       o      Backwash parameters (flow rate and time);
       o      Volume of wastewater; and
       o      Wastewater disposal requirements.

•      Monitoring of Operation
       o      Use of an arsenic field test kit for the  purpose of monitoring  feed and treated
              arsenic levels;
       o      Feed water pressure;
       o      Treated water flow rate;
       o      Treated water pressure;
       o      Maintenance and operator labor requirements; and
       o      Spare parts requirements.

•      Operability
       During verification testing and during compilation  of process operating data, attention
       was given to the arsenic adsorption media filter operability aspects.  Among the factors
       that were considered are:
       o      Fluctuation of flow rates, as well  as the  time interval at which flow adjustment
              was needed;
       o      Ease of adjusting the flow rate when outside the design range; and
       o      Contacting the state regulatory agency to acknowledge the volumes and nature of
              wastewater residue  from  the preconditioning   of  the  media and backwash
              wastewater.
                                           73

-------
4.7.3.2 Maintenance.  Kinetico Inc.  and  Alcan  Chemicals provided  readily  understood
information on the required or recommended maintenance schedule for each piece of operating
equipment including, but not limited to:

•      manual valves
•      on-line measuring instruments
•      control module

Kinetico Inc. and Alcan Chemicals provided readily understood information on the required or
recommended maintenance schedule for non-mechanical or non-electrical equipment including,
but not limited to:

•      adsorptive media vessels
•      feed lines

4.8  Task 4: Data Management

The data management plan was executed as presented in Section 3.12.3.  Data were entered into
computer spreadsheets  and  submitted in electronic and  hard copies.   QA/QC forms, field
notebooks,  and photographs are included in the appendices of this report.

4.9  Task 5:  Quality Assurance/Quality Control (QA/QC)

4.9.1   Introduction

Appropriate quality assurance and quality control measures were performed to ensure the quality
and integrity  of all  measurements of operational and water quality parameters during the ETV
testing.   QA/QC  procedures for the operation of the arsenic adsorption media filter and  the
measured water quality parameters were maintained during the verification testing program as
specified in the test plan, and as described in Section 3.13.

On-site QA/QC  activities were recorded in the logbooks and are  included  as Appendix F.
QA/QC activities  included fluoride electrode, pH meter, turbidimeter,  flow meter, and rotameter
calibrations, as well  as  collection and analysis of duplicate, blank, and  spike  samples,  as
specified in the PSTP.

QA/QC  efforts also included  review of laboratory raw data (run  logs and bench sheets);
calibration  of on-site analytical instrumentation; calibration of totalizer meters; calibration of the
flow meter; analyses of split samples to verify Hach Test Kit analyses for alkalinity, calcium,
and hardness; pressure gauge calibration;  collection  of duplicate  samples  for  on-site and
laboratory  analyses;  and spiked sample analyses.  Performance evaluation analyses were also
performed  by Gannett  Fleming to  demonstrate proficiency  and accuracy  of the  analytical
equipment  and of the laboratory techniques required for all on-site water quality analyses.  All
data entry performed by the field engineer was checked by a second person.
                                           74

-------
An on-site system inspection and audit for sampling activities and field  operations were
conducted by NSF.  The Gannett Fleming QA officer also conducted an on-site inspection during
the first two weeks of operation.

4.9.2   Data Quality Indicators

Data quality indicators include the following:

•      Representativeness
•      Accuracy
•      Precision
•      Statistical Uncertainty
•      Completeness

4.9.2.1 Representativeness. Representativeness refers to the degree to which the data accurately
and precisely represent the conditions or characteristics of the parameter represented by the data.
Representativeness was ensured by executing a consistent sample collections protocol, by using
each method to its optimum capability to achieve a high level of accuracy and precision, and by
collecting sufficient data to be  able to detect a change in operations.

4.9.2.2 Accuracy. Accuracy refers to the difference between a sample result  and the true or
reference value.    Accuracy  was  optimized  through  equipment  calibrations,  performance
evaluation sample analysis, collection  of  split  samples, analysis of duplicate samples,  and
analysis  of spiked samples, as  specified in the PSTP. Periodic calibration of field test equipment
included calibration of  pressure gauges,  rotameter, totalizer meters,  portable turbidimeter, pH
meter, and fluoride meter/electrode, as specified in Table 4-15.
Table 4-15 Field
Instrument
Pressure Gauges
Rotameter
Instrument Calibration Schedule
Calibration Method
Dead weight calibration tester
Volumetric "bucket & stop watch"
Frequency
Biannual
Weekly
Acceptable
Accuracy
± 10%
± 10%
Totalizer Meters

Portable Turbidimeter
Portable pH/ISE Meter with Combination
pH/Temperature Electrode

Thermometer (NIST-traceable)

Portable pH/ISE Meter with Fluoride Ion
Selective Electrode
Volumetric "bucket & stop watch"       Weekly      ± 1.5%

Secondary turbidity standards           Daily        PE sample
Primary turbidity standards             Weekly

Three-point calibration using 4.0,7.0      Daily        ±5%
and 10.0 buffers
Calibration not required                N/A

0.1 mg/L or 0.5 mg/L fluoride standard,   Daily        ±2%
and 10.0 mg/L fluoride standard	
                                             75

-------
 4.9.2.2.1  Split Samples. Split samples for alkalinity, calcium, and total hardness were
 collected and analyzed by the PADEP Laboratory during each of the first two days of the
 Integrity Test to verify the accuracy of the Hach methods for on-site analyses of these
 parameters.  The results of the split sample analyses by the PADEP Laboratory, as shown
 in Tables 4-16 and 4-17, were within the allowable 30% limit of difference established by
 NSF.  As a result, the Hach methods were utilized for the remainder of the verification
 test.

Table 4-16. Split-Samples (April 22, 2003)
Feed Water
Parameter
Alkalinity (mg/L as CaCO3)
Calcium (mg/L)
Hardness (mg/L as CaCOs)
GFLab
88.0
28.0
100.0
PADEP Lab
94.8
24.5
95
% Difference
7.
14
5.
2%
.3%
3%
Treated Water
GFLab
50.
26.
100
0
4
.0
PADEP Lab
53.4
24.8
96
% Difference
6.4%
6.5%
4.2%

Table 4-17.  Split-Samples (April 23, 2003)

Parameter
Alkalinity (mg/L as CaCO,)
Calcium (mg/L)
Hardness (mg/L as CaCO3)

GFLab
88.0
28.0
104.0
Feed Water
PADEP Lab %
97.8
24.0
93

Difference
10.0%
16.7%
12%

GFLab
66.0
26.4
100.0
Treated Water
PADEP Lab %
73.8
23.0
90

Difference
10.6%
14.8%
11%
 4.9.2.2.2  Performance Evaluation Samples for Water Quality Testing. Performance
 evaluation (PE) samples are samples of known concentration prepared by an independent
 performance evaluation laboratory and provided as unknowns to an analyst to evaluate
 his or her analytical performance.  Analyses of laboratory PE samples were conducted
 before the initiation of verification testing.  The control limits for the  PE samples were
 used to evaluate the field analytical method performance.

 A PE sample comes with statistics derived from the analysis of the sample by a number
 of laboratories using EPA-approved methods. These statistics include a true value of the
 PE sample, a mean of the laboratory results obtained from the analysis  of the PE sample,
 and  an  acceptance  range for  sample values.   The field  laboratory  and the PADEP
 Laboratory provided  results from  the analysis of the PE  samples, which meet  the
 performance objectives of the verification testing.

 PE sample results for the PADEP Laboratory and the results of PE checks for on-site
 water quality parameters are included in Appendix R.

 The results of arsenic speciation column performance evaluation tests are also included in
 Appendix R. The initial  speciation column test produced  less than acceptable accuracy
 for arsenic  III  recovery.   It  was  determined  that the speciation columns were  not
 functioning properly and  a second batch of columns (prepared by NSF) were tested and
                                     76

-------
       provided acceptable accuracy.  This second batch of columns was used  for Integrity
       Verification Test arsenic speciation.

       4.9.2.2.3  Spike  Sample Analyses. Matrix spikes were not performed for the on-site
       water quality parameters, including alkalinity, calcium, hardness, and fluoride; however,
       analysis of spiked blanks for each  parameter were analyzed for accuracy at a  10%
       minimum  frequency.  A summary  of on-site  water quality spike  sample analyses,
       including calculated percent recoveries for each test, is included in Appendix F. Percent
       recoveries for all of the spiked  blanks for the on-site  water quality  parameters  were
       within the acceptable accuracy range of 30%.

       The results of spike sample analyses performed by the PADEP Laboratory are included
       in  the laboratory analysis  summary tables included  in Appendix H.  Spike sample
       analyses were performed by the PADEP Laboratory at a frequency  of 10%. Spike sample
       analysis percent recoveries for iron,  manganese, aluminum,  and silica were within the
       acceptable accuracy range of 30%.  Spike sample results for chloride  and sulfate  were
       within the acceptable accuracy range of 20%, while total phosphorus was within the
       acceptable accuracy range of 10%.

       The results of NSF Laboratory spike sample analyses for arsenic are included in the
       laboratory QA/QC data in Appendix Q. Percent Kcoveries for arsenic were within the
       acceptable accuracy range of 30%.

4.9.2.3 Precision.  Precision  refers  to  the  degree of  mutual  agreement among individual
measurements.  It provides an estimate of random error and can be measured by replication of
analyses.  The precision levels for all duplicate analyses were calculated.

On-site water quality relative percent deviation calculations are included with the  on-site water
quality data  in Appendix F.   Duplicate analyses for on-site water quality parameters  were
performed at a 10% minimum frequency. One set of duplicates for turbidity had a precision level
of 31%; all other precision levels for the on-site water quality data were  within the acceptable
precision level of 30%.

PADEP Laboratory relative percent deviation calculations for field duplicates are included in
Appendix G.  Field duplicates of PADEP Laboratory samples were collected at a 10% minimum
frequency. A single duplicate  sample for iron was not within the acceptable level of precision of
30%, but  all other field duplicate analyses performed by the  PADEP Laboratory were well
within acceptable precision levels.

PADEP Laboratory relative percent  deviation calculations for laboratory duplicates are included
in Appendix H. The PADEP Laboratory performed duplicates analyses at a 10% minimum
frequency. All PADEP Laboratory duplicate analyses were within the acceptable levels of
precision.
                                           77

-------
NSF relative percent deviation calculations  for  field and  laboratory arsenic  duplicates are
included in Appendix Q.   All NSF  Laboratory  arsenic duplicate  analyses were within the
acceptable precision level of 30%.

4.9.2.4 Statistical Uncertainty. Statistical uncertainty of water quality parameters (for data sets
of eight or more parameters) was  evaluated through the calculation of the 95% confidence
interval around the sample mean.

4.9.2.5 Completeness. Completeness refers to the amount of valid, acceptable data collected
from a measurement process compared to the amount expected to be obtained. The completeness
objective for data generated during  this verification test was based on the number of samples
collected and analyzed for each parameter and/or method. Completeness was defined as the
following for all measurements:

       %C = (V/T) X 100

where:    %C = percent completeness;
          V = number of measurements judged valid; and
          T = total  number of measurements.

Calculation  of data completeness was made  for on-site water quality measurements, PADEP
Laboratory  water quality measurements, and arsenic  measurements. These calculations are
presented in Appendices F, G, and Q of this report, respectively. During the Integrity Test, no
duplicates were  collected for the  on-site water quality parameters, including pH,  temperature,
turbidity, alkalinity,  and fluoride;  however, the level  of completeness fir these  parameters was
deemed acceptable for the amount of data collected during the Capacity Test,  which included
Integrity Test data.  94%  completeness was achieved for the feed and treated water alkalinity
measurements during the Capacity Test, which  is  below  the  95% completeness objective
outlined in the ETV protocol.  The level of completeness for all  other parameters either met or
exceeded the completeness objectives.
                                           78

-------
                                      Chapter 5
                                      References

The following references were used in the preparation of this report:

Standard Methods for Examination of Water and Wastewater. 19th ed. Washington, D.C. APHA.
1995.

U.S. EPA/NSF  International. ETV Protocol for Equipment  Verification Testing for Arsenic
Removal, April 2002.
                                          79

-------
                                      Chapter 6
                                  Vendor Comments

Kinetico Inc. submitted the following comments concerning the ETV test and report.  These
statements were not validated in the verification test and are the opinion of Kinetico, Inc.:

"The Para-Flo™ PF60 Model AA08AS was tested in the ETV process. In the time  between
submitting the equipment and the writing of this report, our marketing department has re-named
much of Kinetico's product line.  The new  model name for this arsenic treatment system is the
2060f-OD (UltrAsorb-A) with Actiguard AAFS50.  Although the new name reflects the use of a
larger tank inlet and outlet  to facilitate faster flow rates, the fact that the flow must be restricted
to obtain a minimum empty bed contact time means that the arsenic treatment process will not be
materially affected in any way."
                                          80

-------