September 2009
NSF 09/26/EPADWCTR
EPA/600/R-09/075
Environmental Technology
Verification Report
Removal of Microbial Contaminants in
Drinking Water
Koch Membrane Systems, Inc.
Targa® 10-48-35-PMC™ Ultrafiltration
Membrane, as Used in the Village Marine
Tec. Expeditionary Unit Water Purifier
Prepared by
NSF International
Under a Cooperative Agreement with
U.S. Environmental Protection Agency
-------
-------
THE ENVIRONMENTAL TECHNOLOGY VERIFICATION
PROGRAM
oEPA
ET
V^lVl
V
U.S. Environmental Protection Agency
NSF International
ETV Joint Verification Statement
TECHNOLOGY TYPE: ULTRAFILTRATION MEMBRANE MODULE
APPLICATION: REMOVAL OF MICROBIAL CONTAMINANTS IN
DRINKING WATER
PRODUCT NAME: TARGA® 10-48-35-PMC™ ULTRAFILTRATION
MEMBRANE MODULE, AS USED IN THE VILLAGE
MARINE TEC. EXPEDITIONARY UNIT WATER PURIFIER
VENDOR: KOCH MEMBRANE SYSTEMS, INC.
ADDRESS: 850 MAIN STREET
WILMINGTON, MA 01887
PHONE: MAIN - 888-677-5624
CUSTOMER SERVICE - 800-343-0499
FAX: 978-657-5208
EMAIL: INFO@KOCHMEMBRANE.COM
NSF International (NSF) manages the Drinking Water Systems (DWS) Center under the U.S.
Environmental Protection Agency's (EPA) Environmental Technology Verification (ETV) Program. The
DWS Center recently evaluated the performance of the Koch Membrane Systems, Inc. Targa® 10-48-35-
PMC™ Ultrafiltration (UF) Membrane, as used in the Village Marine Tec. Expeditionary Unit Water
Purifier. NSF performed all of the testing activities and also authored the verification report and this
verification statement. The verification report contains a comprehensive description of the test.
EPA created the 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, 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 that are responsive to the needs of stakeholders, 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.
NSF 09/26/EPADWCTR The accompanying notice is an integral part of this verification statement. September 2009
VS-i
-------
ABSTRACT
Testing of the Koch Membrane Systems, Inc. Targa® 10-48-35-PMC™ Ultrafiltration (UF) Membrane
was conducted as part of the ETV verification of the US Navy Office of Naval Research's (ONR)
Expeditionary Unit Water Purifier (EUWP), manufactured by Village Marine Tec. The EUWP uses the
Targa 10-48-35-PMC membrane module in the UF treatment step. During field verification testing of the
EUWP, removal of Bacillus endospores was measured as a surrogate for removal of Cryptosporidium
parvum oocysts (see the full verification report for a discussion about the appropriateness of using
Bacillus endospores as a surrogate for C. parvum). The observed log reductions were below what had
previously been observed during lab challenge testing of the same UF membrane fibers, indicating that
either there were membrane integrity problems, or that there were endospores present on the filtrate side
of the UF modules that were sloughing off. To test whether there was poor membrane integrity within the
UF modules, NSF and EPA had the field testing organization randomly select two UF modules from the
field tested EUWP and send them to NSF to conduct additional microbial challenges under controlled
laboratory conditions.
The UF modules were challenged with approximately 4 logic per milliliter (mL) of B. atrophaeus
endospores, and 5 logio per liter (L) of formalin-fixed C. parvum oocysts. Each challenge test was 30 or
45 minutes in length, and was conducted at a target flux of 38 gallons per day per square foot (gfd), which
is the target flux for UF module operation in the EUWP. The membranes removed a minimum of 2.4
logio per mL of B. atrophaeus, and 4.3 logio per L of C parvum.
PRODUCT DESCRIPTION
The following technology description was provided by the manufacturer and has not been verified.
The UF modules used in the EUWP are Koch Targa 10-48-35-PMC membrane modules, with endcaps
designed and manufactured by Village Marine Tec. The Targa 10-48-35-PMC is a 10.75 inch x 48 inch
module (not including the endcaps). The membrane fibers are made of polysulfone, with a nominal fiber
inner diameter of 0.9 millimeters. The nominal membrane surface area for the module, using the fiber
inner diameter, is 554 square feet. The nominal molecular weight cutoff rating for the membrane is
100,000 Daltons.
VERIFICATION TESTING DESCRIPTION
Selection of Modules
After completion of field testing of the EUWP UF system at Selfridge Air National Guard Base in July
and August of 2007, two UF modules from the EUWP were chosen at random for the lab challenge tests.
The modules chosen were serial numbers KM840643-4015 and KM849697-5021. Prior to the summer
2007 field test, each UF module was individually integrity tested using a pressure decay test. The
pressures were measured from 0 to 10 minutes, with a starting applied pressure of approximately 15 psig.
KM840643-4015 had a pressure decay rate of 0.21 psig/min. This module was checked for compromised
fibers; one was found and plugged. KM840643-4015 was then retested, and the new pressure decay rate
was 0.17 psig/min. KM849697-5021 had a pressure decay rate of 0.13 psig/min. No fibers were plugged
for this module. For the tests described in this VS, KM840643-4015 was designated as Module 1, and
KM849697-5021 was designated as Module 2.
Test Site
The testing site was the Drinking Water Treatment Systems Laboratory at NSF in Ann Arbor, Michigan.
A description of the test apparatus can be found in the verification report.
NSF 09/26/EPADWCTR The accompanying notice is an integral part of this verification statement. September 2009
VS-ii
-------
Methods and Procedures
The testing methods are detailed in the document Test/QA Plan for the Microbial Seeding Challenge
Study of the Koch Membrane Systems Targa 10-48-35-PMC UF Membrane. Two UF membrane modules
were tested for removal of pathogenic protozoa using two different surrogate organisms - endospores of
the bacteria Bacillus atrophaeus (ATCC 9372, deposited as B. subtilis var. nigef), and formalin-fixed C.
parvum oocysts. Bacillus endospores were chosen as a challenge organism because field testing of the
EUWP also examined Bacillus endospore removal. Note that the test protocol was not designed to
achieve the regulatory requirements for membranes under the Long-Term 2 Enhanced Surface Water
Treatment Rule (LT2ESWTR). This verification did not address long-term performance, membrane
cleaning, or full-scale field maintenance and operation issues. These items are addressed in the
verification reports for the full EUWP system.
The testing was conducted in December 2007 and February 2008. In December 2007 the UF membranes
were challenged with both Bacillus endospores and C. parvum. In February 2008, the membranes were
challenged again with C. parvum to confirm that the oocysts found in one filtrate sample from the
December 2007 test was not due to sample contamination.
The UF modules were not sanitized immediately prior to testing. The UF modules were cleaned in
September 2007 following EUWP field testing. The cleaning procedure used was that prescribed in the
EUWP operation and maintenance manual. Prior to the challenge tests, the modules were flushed for
approximately 15 minutes using deionized water.
Before and after testing, the membranes underwent a pressure decay membrane integrity test following
the procedure in ASTM Standard D6908 - Standard Practice for Integrity Testing of Water Filtration
Membrane Systems.
Each UF module was tested individually. The membranes were challenged with both organisms
simultaneously. In the EUWP, the Targa 10-48-35-PMC is operated at a target flux of 38 gfd, with a
reject flow rate of 10% of the feed flow. To approximate these operation conditions, the target feed flow
rate was set at 16.2 gallons per minute (gpm), and the target filtrate flow rate was 14.6 gpm. For the
December 2007 tests, the membranes were challenged with each organism for 30 minutes, with feed and
filtrate samples collected at start-up, 15 minutes, and 30 minutes. For the February 2008 C. parvum
retest, the membranes were challenged for 45 minutes, with feed and filtrate samples collected at 15, 30,
and 45 minutes. All samples were analyzed for the challenge organism(s) in triplicate.
VERIFICATION OF PERFORMANCE
For presentation of the challenge organism data, the observed triplicate feed and filtrate counts were
averaged by calculating geometric means. Non-detect results were treated as one organism per unit
volume for the purpose of calculating the means.
Table VS-1 presents the B. atrophaeus endospores data. Note that endospores were found in the module
flush samples, despite the UF system chemical cleaning that was conducted after the August 2007 field
test of the EUWP UF system. The modules were forward flushed for 15 minutes on December 10 using
deionized water, and the flush samples were collected at the end of this flush. The modules were flushed
again on December 11 for approximately one minute immediately prior to conducting the microbial
challenges. The module flush samples had no C. parvum, but greater than 1 logio of endospores (25 and
15 CPU/100 mL). Tryptic Soy Agar (TSA) was supposed to be substituted for nutrient agar in the
SM9218 enumeration method for the endospores, in order to be able to distinguish the challenge
endospores from wild-type endospores already present in the membrane modules from the field testing.
B. atrophaeus gives orange colonies with a distinctive morphology on TSA. However, due to
NSF 09/26/EPADWCTR The accompanying notice is an integral part of this verification statement. September 2009
VS-iii
-------
miscommunication between the DWS Center and the NSF Microbiology Lab, the B. atrophaeus
endospores were enumerated on nutrient agar, so they could not be distinguished from the wild-type
endospores.
The log removal value (LRVtest) for the endospore challenges show log removals between 2 and 3, but
this data cannot be considered a true picture of UF module performance due to the flush sample counts. It
is possible that many of the endospores in the filtrate samples did not come through the membranes, but
rather were already present on the filtrate side due to contamination from the previous field tests. At time
0 the endospore counts for both modules were higher than those at 15 and 30 minutes, indicating that the
endospores continued to be rinsed out of the filtrate side after the start of the challenges. The UF modules
were chemically cleaned at the end of the August 2007 field test, but it is possible that the cleaning
procedure did not completely remove all of the endospores.
Table VS-1. December 2007 B. atrophaeus Endospores Reduction Data
Module 1
Module 2
Sample Point
Flush
Start-Up
15 Minutes
30 Minutes
Overall Geometric Mean
Flush
Start-Up
15 Minutes
30 Minutes
Overall Geometric Mean
Feed
Geometric Mean
(CFU/mL)
1.74xl04
1.57xl04
1.66xl04
1.66xl04
2.02xl04
1.65xl04
1.75xl04
l.SOxlO4
Log10
4.24
4.20
4.22
4.22
4.31
4.22
4.24
4.26
Filtrate
Geometric Mean
(CFU/mL)
24.8
69
13
14
23
15
175
57
47
78
Log10
.4
.8
.1
.2
.4
.2
2.2
.8
.7
.9
Log
Reduction
2.4
3.1
3.0
2.8
2.1
2.4
2.5
2.4
Table VS-2 presents the December 2007 C. parvum challenge data, and Table VS-3 the February 2008 C.
parvum challenge data. For the December 2007 test, all filtrate samples were below the detection limit,
except for the Module 2 30-minute sample. Because oocysts were found in this sample, C. parvum retests
were conducted in February 2008. No C. parvum was detected in the Module 1 filtrate samples from the
December 2007 challenge, but it was found in both the 30-minute and 45-minute samples from the retest.
C. parvum was also found in the Module 2 30-minute filtrate sample, as was the case with the December
2007 challenge. However, no C. parvum was detected in the Module 2 45-minute filtrate sample. In spite
of the C. parvum filtrate counts, the UF membrane still removed greater than 4 logs of the oocysts.
Table VS-2. December 2007 C. parvum Reduction Data
Module 1
Module 2
Sample Point
Flush
Start-Up
15 Minutes
30 Minutes
Overall Geometric Mean
Flush
Start-Up
15 Minutes
30 Minutes
Overall Geometric Mean
Feed
Geometric Mean
(Cysts/L)
1.2xl05
7.5xl04
7.1xl04
8.6xl04
l.lxlO5
8.4xl04
8.4xl04
9.2xl04
Log™
5.1
4.9
4.9
5.0
5.0
4.9
4.9
4.9
Filtrate
Geometric Mean
(Cysts/L)
<1
<1
<1
<1
<1
<1
<1
<1
47
3.6
Log™
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.7
0.6
Log
Reduction
5.1
4.9
4.9
5.0
5.0
4.9
3.2
4.3
NSF 09/26/EPADWCTR
The accompanying notice is an integral part of this verification statement. September 2009
VS-iv
-------
Table VS-3. February 2008 C. parvum Reduction Retest Data
Module 1
Module 2
Sample Point
Flush
Start-Up
30 Minutes
45 Minutes
Overall Geometric Mean
Flush
Start-Up
30 Minutes
45 Minutes
Overall Geometric Mean
Feed
Geometric Mean
(Cysts/L)
6.3xl04
6.2xl04
7.9xl04
6.8xl04
5.7xl04
5.6xl04
5.1xl04
5.5xl04
Log10
4.8
4.8
4.9
4.8
4.8
4.8
4.7
4.7
Filtrate
Geometric Mean
(Cysts/L)
<1
<1
2
1
0.7
<1
<1
4
<1
1.6
Log10
0.0
0.0
0.4
0.0
0.0
0.0
0.0
0.6
0.0
0.2
Log
Reduction
4.8
4.4
4.9
4.7
4.8
4.2
4.7
4.5
The December 2007 and February 2008 pre-test and post-test pressure decay rate calculations are shown
in Tables VS-4 and VS-5, respectively. Note that two pressure decay rates were calculated, one for the
entire test, and another for just the span of 10 to 20 minutes. The 10 to 20 minute calculation was
included because ASTM D6908 suggests allowing the pressure decay rate to stabilize before conducting
the official pressure decay test. The higher pressure decay rate was not reflected in the Bacillus
endospore and C.parvum reduction data. It is possible that the higher Module 1 pressure decay rate was
due to air leaks out of the temporary plumbing on the test rig.
Table VS-4. December 2007 Pressure Decay Rates
Time (min.)
10-20 Minute Pressure
Decay Rate (psig/min)
0-20 Minute Pressure
Decay Rate (psig/min)
Pre-Test
Module 1
0.3
0.35
Module 2
0.08
0.09
Post-Test
Module 1
0.45
0.74
Module 2
0.08
0.1
Table VS-5. February 2008 Pressure Decay Test Data
Time (min.)
10-20 Minute Pressure
Decay Rate (psig/min)
0-20 Minute Pressure
Decay Rate (psig/min)
Pre-Test
Module 1
0.3
0.6
Module 2
0.2
0.25
Post-Test
Module 1
0.4
0.4
Module 2
0.2
0.2
QUALITY ASSURANCE/QUALITY CONTROL (QA/QC)
NSF provided technical and quality assurance oversight of the verification testing as described in the
verification report, including a review of 100% of the data. NSF QA 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.
NSF 09/26/EPADWCTR
The accompanying notice is an integral part of this verification statement. September 2009
VS-v
-------
Original signed by Sally Gutierrez 09/29/09 Original signed by Robert Ferguson 09/11/09
Sally Gutierrez Date Robert Ferguson Date
Director Vice President
National Risk Management Research Water Systems
Laboratory NSF International
Office of Research and Development
United States Environmental Protection
Agency
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 test protocol, the verification statement, and the verification report (NSF
report # NSF 09/26/EPADWCTR) are available from the following sources:
1. ETV Drinking Water Systems Center Manager (order hard copy)
NSF International
P.O. Box 130140
Ann Arbor, Michigan 48113-0140
2. Electronic PDF copy
NSF web site: http://www.nsf.org/info/etv
EPA web site: http://www.epa.gov/etv
NSF 09/26/EPADWCTR The accompanying notice is an integral part of this verification statement. September 2009
VS-vi
-------
September 2009
Environmental Technology Verification Report
Removal of Microbial Contaminants in Drinking Water
Koch Membrane Systems, Inc.
Targa® 10-48-35-PMC™ Ultrafiltration Membrane, as Used in the
Village Marine Tec. Expeditionary Unit Water Purifier
Prepared by:
Mike Blumenstein, Senior Project Manager
Bruce Bartley, Technical Manager
NSF International
Ann Arbor, Michigan 48105
and
Jeffrey Q. Adams, Project Officer
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
Under a cooperative agreement with the U.S. Environmental Protection Agency
-------
Notice
The U.S. Environmental Protection Agency, through its Office of Research and Development,
funded and managed, or partially funded and collaborated in, the research described herein. It
has been subjected to the Agency's peer and administrative review and has been approved for
publication. Any opinions expressed in this report are those of the author (s) and do not
necessarily reflect the views of the Agency, therefore, no official endorsement should be inferred.
Any mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
11
-------
Foreword
The EPA is charged by Congress with protecting the nation's air, water, and land 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, the EPA's Office of Research and
Development provides data and science support that can be used to solve environmental
problems and to build the scientific knowledge base needed to manage our ecological resources
wisely, to understand how pollutants affect our health, and to prevent or reduce environmental
risks.
The Environmental Technology Verification (ETV) Program has been established by the EPA to
verify the performance characteristics of innovative environmental technology across all media
and to report this objective information to permitters, buyers, and users of the technology, thus
substantially accelerating the entrance of new environmental technologies into the marketplace.
Verification organizations oversee and report verification activities based on testing and quality
assurance protocols developed with input from major stakeholders and customer groups
associated with the technology area. ETV consists of six environmental technology centers.
Information about each of these centers can be found on the internet at http://www.epa.gov/etv.
Under a cooperative agreement, NSF International has received EPA funding to plan, coordinate,
and conduct technology verification studies for the ETV "Drinking Water Systems Center" and
report the results to the community at large. The DWS Center has targeted drinking water
concerns such as arsenic reduction, microbiological contaminants, particulate removal,
disinfection by-products, radionuclides, and numerous chemical contaminants. Information
concerning specific environmental technology areas can be found on the internet at
http://www.epa.gov/nrmrl/std/etv/verifications.html.
in
-------
Acknowledgments
The authors wish to acknowledge the support of all those who helped plan and conduct the
verification test, analyze the data, and prepare this report. We sincerely appreciate the
involvement and support of all staff from the NSF testing laboratory who were involved with
testing activities for this verification. In particular, we would like to thank Rob Herman,
Director of the NSF Drinking Water Treatment Systems Laboratory, William Allen III, the lead
laboratory technician for this project, Robert Donofrio, PhD, Director of the NSF Microbiology
Laboratory, Robin Bechanko, Senior Microbiologist, and Joe Terrell, Supervisor of QA and
Safety.
Finally, we would like to thank Richard Sakaji, PhD, East Bay Municipal Utility District, and
Craig Patterson, U.S. EPA, for their reviews of this verification report.
IV
-------
Table of Contents
Verification Statement VS-i
Title Page i
Notice ii
Foreword iii
Acknowledgements iv
Table of Contents v
List of Tables vi
List of Figures vi
Abbreviations and Acronyms vii
Chapter 1 Introduction 1
1.1 ETV Program Purpose and Operation 1
1.2 Purpose of Verification 1
1.3 Testing Participants and Responsibilities 2
1.3.1 NSF International 2
1.3.2 U.S. Environmental Protection Agency 3
1.3.3 U.S. Navy ONR 3
1.3.4 U.S. Army TARDEC 3
1.3.5 U.S. Bureau of Reclamation 3
1.3.6 Koch Membrane Systems, Inc 4
Chapter 2 Product Description 5
2.1 UF Membranes General Description 5
2.2 Targa 10-48-35-PMC UF Membrane Description 5
2.3 Modules Chosen for Testing 6
Chapter 3 Methods and Procedures 7
3.1 Introduction 7
3.2 Challenge Organisms 7
3.3 Test Apparatus 7
3.4 Test Rig andUF Membrane Module Sanitization 9
3.5 UF Module Integrity Tests 9
3.6 Membrane Module Operation 9
3.7 Test Water Composition 10
3.8 Challenge Test Procedure 11
3.9 Analytical Methods 11
3.9.1 Water Quality Analytical Instruments 11
Chapter 4 Results and Discussion 12
4.1 December 11, 2007 Challenges 12
4.2 February 14, 2008 C. parvum Retest 18
4.3 Conclusions 21
Chapter 5 Quality Assurance/Quality Control 22
5.1 Introduction 22
5.2 Test Procedure QA/QC 22
5.3 Sample Handling 22
5.4 Chemistry Laboratory QA/QC 22
-------
5.5 Microbiology Laboratory QA/QC 22
5.5.1 Growth Media Positive Controls 22
5.5.2 Negative Controls 23
5.6 Documentation 23
5.7 Data Review 23
5.8 Data Quality Indicators 23
5.8.1 Representativeness 23
5.8.2 Accuracy 23
5.8.3 Precision 24
5.8.4 Completeness 24
Chapter 6 References 26
Appendices
Appendix A Bacillus Endospores as a Surrogate for Cryptosporidium parvum Oocysts
Appendix B Test/Quality Assurance Project Plan
Appendix C Triplicate Challenge Organism Counts
List of Tables
Table 2-1. Targa 10-48-35-PMC Specifications 5
Table 3-1. Water Quality Requirements 10
Table 3-2. Analytical Methods for Laboratory Analyses 11
Table 4-1. December 2007 B. atrophaeus Endospores Reduction Data 13
Table 4-2. December 2007 C. parvum Reduction Data 13
Table 4-3. December 2007 Pressure Decay Test Data 14
Table 4-4. December 2007 Pressure Decay Test LRVDrr 17
Table 4-5. December 2007 Operational Data and Water Chemistry Data 17
Table 4-6. February 2008 C. parvum Reduction RetestData 18
Table 4-7. February 2008 Pressure Decay Test Data 19
Table 4-8. February 2008 Pressure Decay Test LRVDrr 20
Table 4-9. February 2008 Operational Data and Water Chemistry Data 20
Table 5-1. Completeness Requirements 24
List of Figures
Figure 2-1. UF modules in the EUWP UF skid 6
Figure 3-1. UF modules plumbed to test station in NSF testing laboratory 8
Figure 3-2. Schematic diagram of tank rig test station 9
VI
-------
Abbreviations and Acronyms
ALCR air-liquid conversion ratio
ASTM ASTM International
ATCC American Type Culture Collection
°C degrees Celsius
CPU colony forming units
cm centimeter
Da Daltons
DWS Drinking Water Systems
EPM electrophoretic mobility
ETV Environmental Technology Verification
EUWP Expeditionary Unit Water Purifier
°F degrees Fahrenheit
ft2 square feet
FTO Field Testing Organization
gfd gallons per square foot per day
gpd gallons per day
gpm gallons per minute
h hours
HC1 hydrochloric acid
FIPC heterotrophic plate count
in inch(es)
L liter
Lpm liters per minute
LRV log removal value
LT2ESWTR Long Term 2 Enhanced Surface Water Treatment Rule
MFGM Membrane Filtration Guidance Manual
mg milligram
mL milliliter
mm millimeter
mM milliMolar
MWCO molecular weight cutoff
NaOH sodium hydroxide
ND non-detect
NRMRL National Risk Management Research Laboratory
NSF NSF International (formerly known as National Sanitation Foundation)
NSWCCD Naval Surface Warfare Command - Carderock Division
NTU Nephelometric Turbidity Unit
ONR Office of Naval Research
ORD Office of Research and Development
psi pounds per square inch
psig pounds per square inch, gauge
QA quality assurance
QC quality control
vn
-------
QMP Quality Management Plan
RPD relative percent difference
SM Standard Methods for the Examination of Water and Wastes
SNL S andi a Nati onal Lab oratory
SOP standard operating procedure
TARDEC Tank Automotive Research, Development, and Engineering Center
TDS total dissolved solids
TOC total organic carbon
ISA tryptic soy agar
UF ultrafiltration
ug microgram
jam micrometer
uS microSiemens
USER United States Bureau of Reclamation
USEPA United States Environmental Protection Agency
VCF volumetric concentration factor
Vlll
-------
Chapter 1
Introduction
1.1 ETV Program Purpose and Operation
The U.S. Environmental Protection Agency (USEPA) 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 that are responsive to the needs of stakeholders;
conducting field or laboratory testing, collecting and analyzing data; and by 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 USEPA has partnered with NSF International (NSF) under the ETV Drinking Water
Systems (DWS) Center to verify performance of drinking water treatment systems that benefit
the public and small communities. It is important to note that verification of the equipment does
not mean the equipment is "certified" by NSF or "accepted" by USEPA. Rather, it recognizes
that the performance of the equipment has been determined and verified by these organizations
under conditions specified in ETV protocols and test plans.
1.2 Purpose of Verification
Testing of the Koch Membrane Systems, Inc. Targa® 10-48-3 5-PMC™ Ultrafiltration (UF)
Membrane was conducted as part of the ETV verification of the US Navy Office of Naval
Research's (ONR) Expeditionary Unit Water Purifier (EUWP), manufactured by Village Marine
Tec. The EUWP uses the Targa 10-48-35-PMC membrane module in the UF treatment step.
During field verification testing of the EUWP, removal of Bacillus endospores was measured as
a surrogate for removal of Cryptosporidiumparvum (see Appendix A for a discussion about the
appropriateness of using Bacillus endospores as a surrogate for C. parvum). The observed log
reductions were below what had previously been observed during lab challenge testing of the
same UF membrane fibers, indicating that either there were membrane integrity problems, or that
there were endospores present on the filtrate side of the UF modules that were sloughing off. To
test whether there was poor membrane integrity within the UF modules, NSF and EPA had the
field testing organization randomly select two UF modules from the field tested EUWP and send
them to NSF to conduct additional microbial challenges under controlled laboratory conditions.
-------
Note that the test protocol was not designed to achieve the regulatory requirements for
membranes under the Long-Term 2 Enhanced Surface Water Treatment Rule (LT2ESWTR).
Also, this verification does not address long-term performance over the life of the membrane,
cleaning of the membranes, nor any other maintenance and operation. These items are covered
under verification testing of the full-scale EUWP.
1.3 Testing Participants and Responsibilities
EUWP design, construction, and testing was overseen by a federal multi-agency team composed
of representatives from Office of Naval Research (ONR); Army Tank-Automotive Research,
Development, and Engineering Center (TARDEC); Naval Surface Warfare Command -
Carderock Division (NSWCCD); United States Department of Interior Bureau of Reclamation
(USER); and Sandia National Laboratories (SNL). The manufacturer, Village Marine Tec., was
contracted to design and build the EUWP to the team's Generation 1 specifications using the
above requirements and 2004 state-of-the-art technology.
The organizations involved with verification testing were:
• NSF
• USEPA
• ONR
• TARDEC
• USER
• Village Marine Tec.
The following is a brief description of each of the ETV participants and their roles and
responsibilities.
1.3.1 NSF International
NSF is an independent, not-for-profit organization dedicated to public health and safety, and to
protection of the environment. Founded in 1944 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. The USEPA partnered with NSF to verify the performance of drinking
water treatment systems through the USEPA's ETV Program.
NSF performed all verification testing activities at its Ann Arbor, MI location. NSF prepared the
test/QA plan, performed all testing, managed, evaluated, interpreted, and reported on the data
generated by the testing, and reported on the performance of the technology.
Contact Information: NSF International
789 N. Dixboro Road
Ann Arbor, MI 48105
Phone: 734-769-8010
Contact: Mr. Bruce Bartley, Project Manager
Email: bartley@nsf.org
-------
1.3.2 U.S. Environmental Protection Agency
USEPA, through its Office of Research and Development (ORD), 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 USEPA, and recommended for public release.
1.3.3 U.S. Navy ONR
The U.S. Navy Office of Naval Research provided oversight of the EUWP development program
which involved developing high productivity water treatment units for land and shipboard
military and civilian emergency preparedness applications. The Office of Naval Research also
provided funding for the EUWP ETV testing project.
Contact Information: Office of Naval Research
Logistics Thrust Program
Operations Technology Division
800 N. Quincy St.
Arlington, VA 22217
Contact: Major Alan Stocks
Phone: 703-696-2561
Email: stocksa@onr.navy.mil
1.3.4 U.S. Army TARDEC
TARDEC served as the field testing organization (FTO) for the full EUWP verifications at
Selfridge Air National Guard Base, MI and Port Hueneme, CA.
Contact Information: U.S. Army TARDEC
c/o NFESC, ESC32
110023rd Avenue
Point Hueneme, C A 93043
Contact: Mr. Mark Miller
Phone: 805-982-1315
Email: mark.c.miller@navy.mil
1.3.5 U.S. Bureau of Reclamation
USER was the FTO for the full EUWP verification in Gallup, NM. USER also provided field
operations and technical support for the other field verification tests.
Contact Information: U. S. Bureau of Reclamation
Denver Federal Center (D-8230), PO Box 25007
Denver, CO 80225
Contact: Ms. Michelle Chapman
Phone: 303-445-2264
Email: mchapman@do.usbr.gov
-------
1.3.6 Koch Membrane Systems, Inc.
Koch Membrane Systems, Inc. supplies the UF membranes for the EUWP. Koch Membrane
Systems, Inc. was responsible for providing logistical and technical support, as needed.
Contact Information: Koch Membrane Systems, Inc.
850 Main Street
Wilmington, MA 01887
-------
Chapter 2
Product Description
2.1 UF Membranes General Description
UF membranes remove contaminants from water through sieving based on the size of the
membrane pores relative to the physical size of the contaminant. A common arrangement for the
membranes is in hollow fibers, with the fibers "potted" in a resin. The flow of water through the
fibers is typically "inside-out," where the water flows into the inside of the fibers at one end of
the module and then flows through the fiber wall leaving contaminants behind. UF membranes
can be classified by pore size or the molecular weight cutoff (MWCO) point. Pore sizes
generally range from 0.01 to 0.05 micrometer (jam). Typical MWCO points are 10,000 to
500,000 Daltons (Da), with 100,000 being a common MWCO rating for drinking water
treatment. With these specifications, UF membranes can remove viruses, bacteria, and
protozoan cysts, as well as large molecules such as proteins, and suspended solids.
2.2 Targa 10-48-35-PMC UF Membrane Description
The UF modules used in the EUWP are Koch Targa 10-48-35 PMC modules with end caps
designed and manufactured by Village Marine Tec. The Targa 10-48-35-PMC is a 10.75 inch
(in) x 48 in UF membrane module. The membrane fibers are made of polysulfone. The module
specifications are listed below in Table 2-1.
The UF membranes in the EUWP are operated at a target flux of approximately 38 gallons per
day per square foot (gfd), based on the inner diameter surface area. The membranes were
operated at a similar flux during the laboratory microbial challenges.
Table 2-1. Targa 10-48-35-PMC Specifications
Parameter
Dimensions:
Nominal Fiber Inner Diameter
Module Outside Diameter
Module Length
Nominal Membrane Surface Area (Inner)
Nominal MWCO
Operating Limits:
Max. Inlet Pressure
Max. Temperature
Min. Temperature
Max. Production Transmembrane Pressure
Max. Backflush Transmembrane Pressure
Specification
0.035 in (0.9 millimeters (mm))
10.75 in (273 mm)
48 in (1219 mm)
554 square feet (ft2) (51.5 square meters (m2))
100,000 Da
45 pounds per square inch (psi)
104 °F (40 °C)
32 °F (0 °C)
30 psi
20 psi
-------
2.3 Modules Chosen for Testing
Some of the UF modules, as installed in the EUWP UF skid, are shown in Figure 2-1. The two
UF modules tested were chosen at random from the EUWP. The UF modules had been operated
in the EUWP system for 1,520 hours over three field tests, as of August 24, 2007.
The modules chosen were serial numbers KM840643-4015 and KM849697-5021. Prior to field
testing of the EUWP UF system at Selfridge Air National Guard Base in July and August of
2007, each UF module was individually integrity tested using a pressure decay test. The
pressures were measured from 0 to 10 minutes, with a starting applied pressure of approximately
15 psig. KM840643-4015 had a pressure decay rate of 0.21 psig/min. This module was checked
for compromised fibers; one was found and plugged. KM840643-4015 was then retested, and
the new pressure decay rate was 0.17 psig/min. KM849697-5021 had a pressure decay rate of
0.13 psig/min. No fibers of this module were plugged.
For the tests described in this report, module KM840643-4015 was designated as Module 1, and
KM849697-5021 was designated as Module 2.
Figure 2-1. UF modules in the EUWP UF skid.
-------
Chapter 3
Methods and Procedures
3.1 Introduction
The challenge tests were conducted in December of 2007 and February of 2008. The tests
followed the procedures described in the Test/QA Plan for the Microbial Seeding Challenge
Study of the Koch Membrane Systems Targa® 10-48-35-PMC™ UF Membrane Module, as
Used in the Village Marine Tec. Expeditionary Unit Water Purifier (EUWP), Gen. 1. The
challenge protocol was adapted from the ETVProtocolfor Equipment Verification Testing for
Physical Removal of Microbiological and Paniculate Contaminants. The test/QA plan is
included in this report as Appendix B.
3.2 Challenge Organisms
Two UF membrane modules were tested for removal of pathogenic protozoa using two different
surrogate organisms - endospores of Bacillus atrophaeus (American Type Culture Collection
(ATCC) 9372, deposited as B. subtilis var. niger), and formalin-fixed C. parvum oocysts.
Note that no virus or bacteria challenges were conducted, and neither of the challenge organisms
is a suitable surrogate for viruses or vegetative bacteria.
As discussed in Section 1.2, Bacillus endospores were used as a challenge organism because
endospore removal was measured during previous field testing as a surrogate for C. parvum. See
Appendix A for Bacillus endospores as a surrogate for C. parvum. For these laboratory
challenge tests, formalin-fixed C. parvum oocysts were also used because it was not cost
prohibitive to do so (versus during field testing), and challenging with Bacillus endospores and
C. parvum side-by-side allowed a direct comparison of the removal efficiency of one versus the
other.
B. atrophaeus was purchased from Presque Isle Cultures of Erie, PA. The C. parvum oocysts
were purchased from Sterling Parasitology Lab of Tuscon, AZ.
3.3 Test Apparatus
The modules were plumbed to a test station in the NSF Drinking Water Treatment Systems
Laboratory. The test station uses a 1,200-gallon stainless steel tank or a 1,200-gallon
polyethylene tank to hold the challenge water. Figure 3-1 shows the UF modules plumbed to the
test rig. Figure 3-2 is a schematic diagram of the test rig.
-------
Figure 3-1. UF modules plumbed to test station in NSF testing laboratory.
-------
Any suitable pressure or delivery system
Water supply
Tank
fill Mechanica
-^_^_. valve filter
1 1 ^ O
Back flow preventer \_J
Mixer
CD?
^^
•^s f ~\ ^^"
/\ i y
| Pump
^h Tank
Drain ine
Pressure
gauge
„ 9
/ N
Diaphragm
Pressure
regulator
(
Inf
san
P
s
Influent
sampling
point
o-
Water meters
Pressure gauges
-0
Cycling
solenoid A
Product water
sampling points
Figure 3-2. Schematic diagram of tank rig test station
\) Test units Q)
. Cycling
solenoid I
3.4 Test Rig and UF Membrane Module Sanitization
The test rig was sanitized prior to testing according to NSF standard operating procedure (SOP).
The process is proprietary, and uses multiple chemicals as sanitizers. After sanitization, the test
rig was flushed until a less-than-detectable concentration of sanitizing agent was present. The
UF modules were not sanitized immediately prior to testing. The UF modules were cleaned in
September 2007 following EUWP field testing. The cleaning procedure used was that prescribed
in the EUWP operation and maintenance manual. Prior to the laboratory challenge tests, the
modules were flushed for approximately 15 minutes using deionized water.
3.5 UF Module Integrity Tests
Before and after testing, the membranes underwent a pressure decay membrane integrity test
following the procedure in ASTM Standard D6908 - Standard Practice for Integrity Testing of
Water Filtration Membrane Systems.
3.6 Membrane Module Operation
As discussed in Section 2.2, the TARGA 10-48-35-PMC UF membranes are operated at a flux of
approximately 38 gfd in the EUWP system, so this flux was targeted for the microbial
challenges. A flux of 38 gfd equals a filtrate flow rate of approximately 14.6 gpm. The
membranes were operated with a retentate flow of approximately 10% of the feed, as they are in
the EUWP. Therefore, to achieve a filtrate flow rate of 14.6 gpm, the target feed flow rate was
16.2 gpm.
-------
The modules were challenged individually. The same tank of challenge water was used for both
tests.
3.7 Test Water Composition
Local tap water was treated by carbon filtration, reverse osmosis, and deionization to make the
base water (RO/DI water) for the tests. This water was low in particulates, thus representing a
worse case for testing because there were few suspended particles to which the challenge
organisms could attach. Note that suspended particle concentrations were not analyzed during
testing.
The RO/DI water has the following QC requirements for use in the NSF testing laboratory:
• Conductivity < 2 microSiemens per centimeter (|j,S/cm) at 25°C;
• TOC < 100 micrograms per liter (ng/L);
• Total chlorine < 0.05 milligrams (mg)/L; and
• Heterotrophic bacteria plate count (HPC) < 100 colony forming units per milliliter
(CFU/mL).
Of the above parameters, only total chlorine was measured specifically for this verification. The
other parameters are measured periodically by NSF as part of the internal QA/QC program for
test water quality.
The base water was adjusted to meet the requirements of Table 3-1.
Table 3-1. Water Quality Requirements
Parameter
Alkalinity
pH
Temperature
Value
100 ± 10 mg/L
7.5 ±0.5
20±2.5°C
Chemical for Adjustment
sodium bicarbonate (NaHCO3)
hydrochloric acid (HC1) or sodium hydroxide (NaOH)
none
Grab samples were collected at the start of each challenge period for analysis of total chlorine,
alkalinity, pH, temperature, total dissolved solids, and turbidity. The pH was also measured at
the end of the challenge period.
The challenge organisms were added to the tank of water at a sufficient titer to achieve the
following target challenge organism concentrations:
• B. atrophaeus - approximately IxlO4 CFU/mL
• C.parvum - approximately 1x105 oocysts/L
Note that both organisms were added to the same tank of water, so that there was a simultaneous
challenge with both organisms.
10
-------
3.8 Challenge Test Procedure
Immediately prior to beginning the tests, the influent challenge holding tank was mixed for a
minimum of 10 minutes using a recirculation pump.
The initial inlet water pressure was set as necessary to deliver a feed flow rate of approximately
16.2 gpm. For the December 2007 tests, the modules were operated for 30 minutes. Feed and
filtrate samples for challenge organism enumeration were collected at start-up, after 15 minutes
of operation, and after 30 minutes of operation. For the February 2008 C. parvum retests, the
modules were operated for 45 minutes, and feed and filtrate samples were collected at 15, 30,
and 45 minutes.
3.9 Analytical Methods, and Accuracy and Precision Limits
A list of laboratory analytical methods can be found in Table 3-2. All samples for B. atrophaeus
and C. parvum were analyzed in triplicate.
Table 3-2. Analytical Methods for Laboratory Analyses
Parameter
Alkalinity (total, as CaCO3)
PH
IDS
Total Chlorine
Turbidity
Bacillus Endospores
Cryptosporidium Oocysts
•a
o
-*^
1
EPA 3 10.2
SM 4500-H+(2)
SM 2540 C
SM 4500-C1 G
SM2130
SM9218
EPA 1623
'!*
>*j o. a
i/5 0) .5
Z Ctf hJ
5 mg/L
NA
5 mg/L
0.05 mg/L
0.1NTU(4)
1 CFU/100 mL
1 oocyst/L
*% /-v
O >%
s5 '-
•- 0
3 £
3 o
^fl
JS o
^E
90-110
NA
90-110
90-110
95-105
—
—
=
o
x ^~--
11
^ ^
sS "•?
hJ €^
<13
< 10
< 10
< 10
—
—
O
=
H ^^
^ ^
MS
14
(3)
7
(3)
(3)
30 h(5)
72 h
a>
•«•!
&-2
£ =
£3
1 L plastic
NA
1 L plastic
NA
NA
1 L plastic
1 L plastic
=
IS
^ &
8. OJ
O
-------
Chapter 4
Results and Discussion
As stated in Section 2.3, module KM840643-4015 was designated as Module 1, and KM849697-
5021 was designated as Module 2.
The challenge tests were conducted on December 11, 2007 and February 14, 2008. The
December 2007 tests were conducted with both B. atrophaeus and C. parvum. During the
December 2007 test, the Module 2 30-minute filtrate sample had a mean C. parvum count of 47
oocysts/L, so C. parvum challenges were conducted again on February 14, 2008 to confirm the
apparent C. parvum breakthrough. For the February 14 retests, it was decided to operate the UF
modules for 45 minutes to see whether C. parvum breakthrough, if observed, was a function of
time. The results for the retest indicate that breakthrough was not time dependent.
For presentation of the challenge organism data in this chapter, the observed triplicate feed and
filtrate counts were averaged by calculating geometric means. Non-detect results were treated as
one organism per unit volume for the purpose of calculating the means. The triplicate counts for
each sample are presented in Appendix C.
4.1 December 11, 2007 Challenges
The modules were forward flushed for 15 minutes on December 10 using deionized water, and
the flush samples were collected at the end of this flush. The modules were flushed again on
December 11 for approximately one minute immediately prior to conducting the microbial
challenges. Table 4-1 presents the B. atrophaeus endospores challenge data, and Table 4-2 the
C. parvum challenge data. The module flush samples had no C. parvum, but greater than 1 logic
of endospores (25 and 15 CFU/100 mL). Tryptic Soy Agar (TSA) was supposed to be
substituted for nutrient agar in the SM9218 enumeration method for the endospores, in order to
be able to distinguish the challenge endospores from wild-type endospores already present in the
membrane modules from the field testing. B. atrophaeus gives orange colonies with a distinctive
morphology on TSA. However, due to miscommunication between the DWS Center and the
NSF Microbiology Lab, the B. atrophaeus endospores were enumerated on nutrient agar, so they
could not be distinguished from the wild-type endospores.
The log removal value (LRVtest) for the endospore challenges show log removals between 2 and
3, but this data cannot be considered a true picture of UF module performance due to the flush
sample counts. It is possible that many of the endospores in the filtrate samples did not come
through the membranes, but rather were already present on the filtrate side due to contamination
from the previous field tests. At time 0 the endospore counts for both modules were higher than
those at 15 and 30 minutes, indicating that the endospores continued to be rinsed out of the
filtrate side after the start of the challenges. The UF modules were chemically cleaned at the end
of the August 2007 field test, but it is possible that the cleaning procedure did not completely
remove all of the endospores.
12
-------
C. parvum was found in one filtrate sample, that for Module 2 at 30 minutes. Because of the C.
parvum breakthrough, retests for oocyst removal only were conducted on February 14, 2008.
See Section 4.2 for the retest data and discussion.
Table 4-1. December 2007 B. atrophaeus Endospores Reduction Data
Module 1
Module 2
Sample
Point
Flush
Start-Up
15 Minutes
30 Minutes
Overall
Geometric
Mean
Flush
Start-Up
15 Minutes
30 Minutes
Overall
Geometric
Mean
Feed
Geometric Mean
(CFU/mL)
1.74xl04
1.57xl04
1.66xl04
1.66xl04
2.02xl04
1.65xl04
1.75xl04
l.SOxlO4
Log10
4.24
4.20
4.22
4.22
4.31
4.22
4.24
4.26
Filtrate
Geometric Mean
(CFU/mL)
24.8
69
13
14
23
15
175
57
47
78
Log10
1.4
1.8
1.1
1.2
1.4
1.2
2.2
1.8
1.7
1.9
Log
Reduction
2.4
3.1
3.0
2.8
2.1
2.4
2.5
2.4
Table 4-2. December 2007 C. parvum Reduction Data
Module 1
Module 2
Sample
Point
Flush
Start-Up
15 Minutes
30 Minutes
Overall
Geometric
Mean
Flush
Start-Up
15 Minutes
30 Minutes
Overall
Geometric
Mean
Feed
Geometric Mean
(Oocysts/L)
1.2xl05
7.5xl04
7.1xl04
8.6xl04
l.lxlO5
8.4xl04
8.4xl04
9.2xl04
Log10
5.1
4.9
4.9
5.0
5.0
4.9
4.9
4.9
Filtrate
Geometric Mean
(Oocysts/L)
<1
<1
<1
<1
<1
<1
<1
<1
47
3.6
Log10
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.7
0.6
Log
Reduction
5.1
4.9
4.9
5.0
5.0
4.9
3.2
4.3
Immediately after the 15-minute flushes on December 10, the modules were subjected to a
pressure decay test. A post-challenge pressure decay test was run the next day. The pre-test and
post-test pressure decay data is shown in Table 4-3. Note that two pressure decay rates were
calculated, one for the entire test, and another for just the span of 10-20 minutes. The 10-20
13
-------
minute calculation was included because ASTM D6908 suggests allowing the pressure decay
rate to stabilize before conducting the official pressure decay test. Module 1 had higher pressure
decay rates than Module 2, as it did in July of 2007 when pressure decay tests were conducted
prior to the field test (see Section 2.3 for further discussion). However, the higher pressure
decay rate for Module 1 was not reflected in the endospore and oocyst reduction data.
Table 4-3. December 2007 Pressure Decay Test Data
Time (min.)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
0-20 Minute
Pressure Decay
Rate (psig/min)
10-20 Minute
Pressure Decay
Rate (psig/min)
Pre-Test
Module 1
17
17
16
16
15.5
15
15
14.4
14
13.4
13
13
12.5
12
12
11.5
11
11
11
11
10
0.35
0.30
Module 2
18
17.8
17.8
17.8
17.8
17.8
17.2
17.1
17.1
17.1
17
17
17
17
17
17
17
16.8
16.5
16.2
16.2
0.09
0.08
Post-Test
Module 1
20
17.5
16.5
15.2
14.5
13.5
12
11
11
10
9.7
9
8.5
8
7.7
7.2
7
6.2
6
5.5
5.2
0.74
0.45
Module 2
20
19.8
19
19
19
19
19
19
18.9
18.8
18.8
18.8
18.5
18.5
18.5
18.5
LE(1)
18.5
18.5
18.2
18
0.10
0.08
(1) Lab data recording error
It is possible that the higher Module 1 pressure decay rates were due to air leaks out of the
temporary plumbing on the test rig. The pressure decay rates can be translated into an expected
log removal values (LRVoix) for Cryptosporidium using the equations in Chapter 4 of the
USEPA Membrane Filtration Guidance Manual (MFGM). The LRVDix ranged from 2.22 to 3.14
for Module 1, and from 3.09 to 3.78 for Module 2. These LRV are much lower than those
observed for C. parvum, with the exception of the Module 2 30-minute sample. The LRVoix
calculations are presented in Table 4-4.
14
-------
The equation used here to calculate LRVoix is Equation 4.7 of the MFGM, which is expressed as
follows:
LRVDIT = log
•ALCR"
(MFGM Equation 4.7)
where:
LR.VDIT = log removal value equating to measured pressure decay rate;
Qp = membrane unit design capacity filtrate flow rate (liters per minute, Lpm);
ALCR = air-liquid conversion ratio (dimensionless);
Qair = flow of air through a critical membrane breach during a pressure decay test
(Lpm); and
VCF = volumetric concentration factor (dimensionless).
For equation 4.7, Qp was set at 55.26 Lpm to equal the target flow rate given in Section 3.6,
while ALCR and Qa;r were calculated as described below.
The VCF is the ratio of the concentration of suspended solids on the feed side of the membrane
at point x in the membrane unit (in this case, point X along each membrane fiber), relative to that
of the influent feed to the membrane unit. Systems with higher VCF will allow for increased
passage of pathogens through a membrane breach as opposed to a system with a VCF of 1.
Depending on the system design, the VCF may vary spatially, and/or temporally. The VCF will
increase temporally if the concentrate stream is recycled back into the feed water. For this ETV
verification, and for the field testing ETV verifications of the EUWP UF system, the concentrate
stream was not recycled, so the UF membrane was considered to be a plug-flow reactor. For a
plug-flow reactor, the VCF only increases spatially as the feed water travels along the length of
the UF fibers. Table 2.1 in Section 2.5.2.1 of the MFGM gives flow-weighted averages and
maximum values for the VCF for various recovery rates. Both the flow-weighted average of
2.56 and maximum of 10 for a 90% recovery rate were input into Equation 4.7. The maximum
VCF of 10 could be expected at or near the outlet end of the feed side of the membrane, as the
90% recovery setting has allowed the suspended solids concentration to increase ten-fold as a
slug of the feed water travels through the UF fibers.
For the ALCR, Equation C.4 in the MFGM was used. This equation is for the Darcy pipe flow
model for turbulent flow through a membrane breach. The equation is expressed as follows:
HP -BP}*(P + P}
ALCR = 170 •Y»J±-*27 J—^ ^ (MFGM Equation C.4)
15
-------
where:
ALCR = air-liquid conversion ratio (dimensionless);
Y = net expansion factor for compressible flow through a pipe to a larger area
(dimensionless);
Ptest = direct integrity test pressure (psig)
BP = backpressure on the system during the integrity test (psi)
Patm = atmospheric pressure (psia)
T = water temperature (°F)
TMP = maximum transmembrane pressure during normal operation
To calculate ALCR, the following values were used for the variables:
Y =0.612;
Ptest = initial applied pressure in Table 4.3 for each integrity test;
BP = 1 .7 psi (4 feet of water at 0.43 psi/ft);
Patm = 14.7 psi;
T =68 °F; and
TMP = 30 psig.
Using these values, the ALCR was calculated for each pressure decay test by inputting the initial
applied pressure. The ALCR value was then input into Equation 4.7.
Qair also needed to be calculated for the LRVoix equation. Qa;r was calculated using Equation 4.8
of the MFGM, which is expressed as follows:
,
Qair = — ™ - 2L (MFGM Equation 4.8)
atm
where:
Qair = flow of air (Lpm);
APtest = rate of pressure decay during the integrity test (psig/min);
Vsys = volume of pressurized air in the system during the integrity test (L); and
Patm = atmospheric pressure (psi).
To calculate Qa;r, the following values were used for the variables:
APtest = the pressure decay rates in Table 4.3 for each integrity test (psig/min);
Vsys = 13. 8 L; and
Patm 14.7 psi.
Note that the LRVoix associated with the maximum VCF of 10 are approximately 0.6 logic lower
than those for the flow-weighted average VCF of 2.56. This indicates that a membrane breach at
the outlet end of a fiber could be expected to allow up to 0.6 logic more of Cryptosporidium
through the breach, as opposed to a breach at the point where the VCF is 2.56.
16
-------
Table 4-4. December 2007 Pressure Decay Test LRVoix
Pressure Decay Test
Module 1 Pre-Test
Module 1 Post-Test
Module 2 Pre-Test
Module 2 Post-Test
Pressure Decay
Timeframe
0 to 20 minutes
10 to 20 minutes
0 to 20 minutes
10 to 20 minutes
0 to 20 minutes
10 to 20 minutes
0 to 20 minutes
10 to 20 minutes
Pressure Decay
Rate (APtest)
0.35
0.30
0.74
0.45
0.09
0.08
0.10
0.08
VCF
2.56
10
2.56
10
2.56
10
2.56
10
2.56
10
2.56
10
2.56
10
2.56
10
LRVDiT (logio)
3.08
2.49
3.14
2.55
2.81
2.22
3.03
2.44
3.69
3.10
3.74
3.15
3.68
3.09
3.78
3.19
Table 4-5 displays the UF module operational data and water chemistry data for the December
11, 2007 challenges. The filtrate flow rates were above the target of 14.6 gpm, thus giving
fluxes above 38 gfd. These higher fluxes, plus the low-particle test water, gave a more
conservative test than the field test conditions for the full EUWP.
Table 4-5. December 2007 Operational Data and Water
Sample
Feed Flow at Start-up (gpm)
Filtrate Flow at Start-up (gpm, calculated)
Reject Flow at Start-up (gpm)
Membrane Flux at Start-up (gfd)
Feed Flow at 30 Minutes (gpm)
Filtrate Flow at 30 Minutes (gpm, calculated)
Reject Flow at 30 Minutes (gpm)
Membrane Flux at 30 Minutes(gfd)
Inlet Pressure at Start-up (psig)
Inlet Pressure at 30 Minutes (psig)
Feed Water Chemistry at Start-up:
Alkalinity (mg/L CaCO3)
PH
Temperature (°C)
Total Chlorine (mg/L)
TDS (mg/L)
Turbidity (NTU)
pH at End of Challenge
Chemistry
Module 1
17.15
15.42
1.73
40.08
16.99
15.16
1.83
39.41
14
15
Data
Module 2
17.04
15.40
1.64
40.03
17.71
16.05
1.66
41.72
14
NR(1)
110
7.67
19.9
O.05
120
0.14
7.72
(1) not recorded
17
-------
4.2 February 14, 2008 C. parvum Retest
As discussed in Section 4.1, C. parvum challenges were conducted again to confirm the
breakthrough observed for Module 2. The results of these challenges are presented in Table 4-6.
NSF and USEPA decided not to run the B. atrophaeus challenges over again because funding
was not available to run both challenge organisms. The retests were carried out to 45 minutes to
determine if breakthrough was correlated to the time of operation. Prior to the retest, the
modules were backflushed for approximately 5 minutes, then forward flushed for 10 minutes.
Deionized water was used for both the backflush and forward flush. The technician conducting
the tests noted that during the backflush, the effluent contained high levels of suspended solids.
No C. parvum was detected in the Module 1 filtrate samples from the December 2007 challenge,
but it was found in both the 30-minute and 45-minute samples from the retest. C. parvum was
also found in the Module 2 30-minute filtrate sample, as was the case with the December 2007
challenge. However, no C. parvum was detected in the Module 2 45-minute filtrate sample.
These results do not indicate the C. parvum breakthough was related to the time of operation. In
spite of the C. parvum filtrate counts, the UF membrane still removed greater than 4 logs of the
oocysts.
Table 4-6. February 2008 C. parvum Reduction Retest Data
Module 1
Module 2
Sample Point
Flush
Start-Up
30 Minutes
45 Minutes
Overall
Geometric
Mean
Flush
Start-Up
30 Minutes
45 Minutes
Overall
Geometric
Mean
Feed
Geometric Mean
(Oocysts/L)
6.3xl04
6.2xl04
7.9xl04
6.8xl04
5.7xl04
5.6xl04
5.1xl04
5.5xl04
Logio
4.8
4.8
4.9
4.8
4.8
4.8
4.7
4.7
Filtrate
Geometric Mean
(Oocysts/L)
<1
<1
2
1
0.7
<1
<1
4
<1
1.6
Logio
0.0
0.0
0.4
0.0
0.0
0.0
0.0
0.6
0.0
0.2
Log
Reduction
4.8
4.4
4.9
4.7
4.8
4.2
4.7
4.5
The pre-test and post-test pressure decay test data is shown in Table 4-7. As with the December
2007 pressure decay data, two pressure decay rates are given. Again, the higher pressure decay
rate for Module 1 is not reflected in the filtrate C. parvum counts.
18
-------
Table 4-7. February 2008 Pressure Decay Test Data
Time (min.)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
10-20 Minute
Pressure Decay
Rate (psig/min)
0-20 Minute
Pressure Decay
Rate (psig/min)
Pre-Test
Module 1
20
18
17
16
15
15
14
13.5
12
11
11
10
10
9
9
9
9
8
8
8
8
0.3
0.6
Module 2
20
19
19
19
18
18
18
18
18
18
17
17
17
16
16
16
16
15
15
15
15
0.2
0.25
Post-Test
Module 1
20
20
20
20
19.5
19.5
19
18
NR
17
16
16
15
15
14
14
13
13
12.5
12
12
0.4
0.4
Module 2
20
20
20
19.5
19.5
19
19
19
19
18.5
18
NR
18
18
18
17.5
17
17
17
16.5
16
0.2
0.2
As with the December 2007 pressure decay data, the February 2008 decay data was used to
calculate LRVoix for each pressure decay test. These calculations are presented in Table 4.8.
The same variable values were used for these calculations as for the December 2007
calculations. The LRVoix for Module 1 ranged from 2.31 to 3.20 logio, while those for Module 2
ranged from 2.69 to 3.38. As with the December 2007 LRVDix, the February 2008 LRVDix are
approximately 1 logic or more below the observed C. parvum LRVtest.
19
-------
Table 4-8. February 2008 Pressure Decay Test LRVDix
Pressure Decay Test
Module 1 Pre-Test
Module 1 Post-Test
Module 2 Pre-Test
Module 2 Post-Test
Pressure Decay
Timeframe
0 to 20 minutes
10 to 20 minutes
0 to 20 minutes
10 to 20 minutes
0 to 20 minutes
10 to 20 minutes
0 to 20 minutes
10 to 20 minutes
Pressure Decay
Rate (APtest)
0.60
0.30
0.40
0.40
0.25
0.20
0.20
0.20
VCF
2.56
10
2.56
10
2.56
10
2.56
10
2.56
10
2.56
10
2.56
10
2.56
10
LRVDiT (logio)
2.90
2.31
3.20
2.61
3.08
2.49
3.08
2.49
3.28
2.69
3.38
2.79
3.38
2.79
3.38
2.79
Table 4-9 displays the UF module operational data and water chemistry data for the February 14,
2008 C. parvum challenges. As with the December 2007 challenges, the filtrate flow rates were
above the target of 14.6 gpm, thus giving fluxes above 38 gfd.
Table 4-9. February 2008 Operational Data and Water
Sample
Feed Flow at Start-up (gpm)
Filtrate Flow at Start-up (gpm, calculated)
Reject Flow at Start-up (gpm)
Membrane Flux at Start-up (gfd)
Feed Flow at 45 Minutes (gpm)
Filtrate Flow at 45 Minutes (gpm, calculated)
Reject Flow at 45 Minutes (gpm)
Membrane Flux 45 Minutes (gfd)
Inlet Pressure at Start-up (psig)
Inlet Pressure at 45 Minutes (psig)
Feed Water Chemistry at Start-up:
Alkalinity (mg/L CaCO3)
pH
Temperature (°C)
Total Chlorine (mg/L)
TDS (mg/L)
Turbidity (NTU)
pH at End of Challenge
Module 1
18.05
16.26
1.79
42.3
16.5
14.85
1.65
38.6
20.4
18.0
Chemistry Data
Module 2
17.60
15.92
1.68
41.4
17.79
16.12
1.67
41.9
18.8
18.6
87
7.55
19
0.05
100
0.1
7.59
20
-------
4.3 Conclusions
These results indicate that the UF modules in the EUWP are not capable of providing a sole
barrier to microorganisms by themselves. However, it is important to note that the modules had
1,520 hours of operation prior to the laboratory challenges. The modules had been through three
field tests, one using tertiary wastewater as the source, and two tests using fresh surface water as
the source. Also, the EUWP as a whole includes RO treatment downstream of the UF modules,
and then chlorination of the RO permeate to provide finished drinking water.
21
-------
Chapter 5
Quality Assurance/Quality Control
5.1 Introduction
An important aspect of verification testing is the quality assurance and quality control (QA/QC)
procedures and requirements. Careful adherence to the procedures ensured that the data
presented in this report was of sound quality, defensible, and representative of the equipment
performance. The primary areas of evaluation were representativeness, accuracy, precision, and
completeness.
Because this ETV was conducted at the NSF testing lab, all laboratory activities were conducted
in accordance with the provisions of the NSF International Laboratories Quality Assurance
Manual (NSF 2007).
5.2 Test Procedure QA/QC
NSF testing laboratory staff conducted the tests by following a USEPA-approved test/Q A plan
created specifically for this verification. NSF QA Department staff performed an audit during
testing to ensure the proper procedures were followed. The audit yielded no significant findings.
5.3 Sample Handling
All samples analyzed by the NSF Chemistry and Microbiology Laboratories were labeled with
unique ID numbers. All samples were analyzed within allowable holding times.
5.4 Chemistry Laboratory QA/QC
The calibrations of all analytical instruments and the analyses of all parameters complied with
the QA/QC provisions of the NSF International Laboratories Quality Assurance Manual (NSF,
2007).
The NSF QA/QC requirements are all compliant with those given in the EPA method or
Standard Method for the parameter. Also, every analytical method has an NSF SOP governing
the procedure.
5.5 Microbiology Laboratory QA/QC
5.5.1 Growth Media Positive Controls
All media were checked for sterility and positive growth response when prepared and when used
for microorganism enumeration. The media was discarded if growth occurred on the sterility
check media, or if there was an absence of growth in the positive response check.
22
-------
5.5.2 Negative Controls
For each sample batch processed, an unused membrane filter and a blank with 100 mL of
buffered, sterilized dilution water was filtered through the membrane were also placed onto the
appropriate media and incubated with the samples as negative controls. No growth was observed
on any blanks.
5.6 Documentation
All laboratory activities were documented using specially prepared laboratory bench sheets and
NSF laboratory reports. Data from the bench sheets and laboratory reports were entered into
Excel spreadsheets. These spreadsheets were used to calculate the geometric means and logic
reductions for each challenge. One hundred percent of the data entered into the spreadsheets was
checked by a reviewer to confirm all data and calculations are correct.
5.7 Data Review
NSF QA/QC staff reviewed the raw data records for compliance with QA/QC requirements. As
required in the ETV Quality Management Plan (QMP), NSF ETV staff checked at least 10% of
the data in the NSF laboratory reports against the lab bench sheets.
5.8 Data Quality Indicators
The quality of data generated for this ETV is established through four indicators of data quality:
representativeness, accuracy, precision, and completeness.
5.8.1 Representativeness
Representativeness refers to the degree to which the data accurately and precisely represent the
expected performance of the UF membranes under normal use conditions. The membranes were
to be operated at fluxes similar to those they encounter in the full EUWP system. The test fluxes
were higher than the typical EUWP flux, which gave a more conservative test. The test water
was of very low turbidity, and low particle count to minimize the potential for microbial
adhesion to suspended particles, which could enhance log reduction.
Representativeness was ensured by consistent execution of the test protocol for each challenge,
including timing of sample collection, sampling procedures, and sample preservation.
Representativeness was also ensured by using each analytical method at its optimum capability
to provide results that represent the most accurate and precise measurement it is capable of
achieving.
5.8.2 Accuracy
Accuracy was quantified as the percent recovery of the parameter in a sample of known quantity.
Accuracy was measured through use of both matrix spikes of a known quantity and certified
standards during calibration of an instrument.
23
-------
The following equation was used to calculate percent recovery:
Percent ReCOVery = 100 X [(Xknown - Xmeasured)/Xknown]
where: Xkn0wn = known concentration of the measured parameter
Xmeasured = measured concentration of parameter
Accuracy of the benchtop chlorine, pH, and turbidity meters was checked daily during the
calibration procedures using certified check standards. Alkalinity and IDS were analyzed in
batches. Certified QC standards and/or matrix spikes were run with each batch.
The percent recoveries of all matrix spikes and standards were within the allowable limits for all
analytical methods.
5.8.3 Precision
Precision refers to the degree of mutual agreement among individual measurements and provides
an estimate of random error. One sample per batch was analyzed in duplicate for the TDS
measurements. At least one out of every ten samples for alkalinity was analyzed in duplicate.
Duplicate municipal drinking water samples were analyzed for pH, total chlorine, and turbidity
as part of the daily calibration process. Precision of duplicate analyses was measured by use of
the following equation to calculate RPD:
RPD =
:200
where:
Sl = sample analysis result; and
^ = sample duplicate analysis result.
All RPDs were within NSF's established allowable limits for each parameter. Please note that
samples from this evaluation for alkalinity and TDS were batched with other non-ETV samples.
The duplicate analysis requirements apply to the whole batch, not just the samples from this
ETV.
5.8.4 Completeness
Completeness is the proportion of valid, acceptable data generated using each method as
compared to the requirements of the test/QA plan. The completeness objective for data
generated during verification testing is based on the number of samples collected and analyzed
for each parameter and/or method, as presented in Table 5-1.
Table 5-1. Completeness Requirements
Number of Samples per Parameter and/or Method
0-10
11-50
>50
Percent Completeness
80%
90%
95%
24
-------
Completeness is defined as follows for all measurements:
%C = (V/T)X100
where:
%C = percent completeness;
V = number of measurements judged valid; and
T = total number of measurements.
All planned samples were collected, and all had valid results. One inlet pressure recording was
missed, that for Module 2 at 30 minutes during the December 2007 challenges. A total of eight
inlet pressure measurements were planned, including the February 2008 C. parvum retest. The
one missed recording gives a completeness percentage of 87.5% for this parameter, which
exceeds the completeness requirements in Table 5-1.
25
-------
Chapter 6
References
APHA, AWWA, and WEF (1999). Standard Methods for the Examination of Water and
Wastewater, 20th Edition.
ASTM International (2003). D 6908-03, Standard Practice for Integrity Testing of Water
Filtration Membrane Systems.
USEPA (2005). Membrane Filtration Guidance Manual (EPA 815-R-06-009).
USEPA and NSF International (2005). ETVProtocolfor Equipment Verification Testing for
Physical Removal of Microbiological and Paniculate Contaminants.
26
-------
Appendix A
Bacillus Endospores as a Surrogate for Cryptosporidiumparvum Oocysts
The EPA LT2ESWTR allows the use of a surrogate for C. parvum, provided the surrogate is
conservative. The EPA MFGM specifically discusses Bacillus subtilis as a surrogate, but states
"Because there is limited data currently available regarding the use of Bacillus subtilis in
membrane challenge studies, a characterization of this organism would be necessary to
determine whether it could be used as a Cryptosporidium surrogate..." The MFGM also states
"Based on the size.. .Bacillus subtilis could potentially be considered a conservative
surrogate.. .pending a comparison of other characteristics (e.g., shape, surface charge, etc.)..."
1. Organism Size and Shape
C. parvum is spherical in shape, while Bacillus endospores are ellipsoidal in shape (football
shaped). C. parvum has a diameter of 4-6 jim. Bacillus endospores are approximately 0.8 jim in
diameter, and 1.8 jim in length. Therefore, Bacillus endospores are a conservative surrogate for
C. parvum, no matter what the orientation of the endospore is when it impacts the test
membrane.
Baltus et. al. (2008) studied membrane rejection of bacteria and viruses with different length vs.
diameter aspect ratios. They theorized, based on a transport model for rod-shaped particles, that
rejection would improve as the aspect ratio (length vs. diameter) increased for a fixed particle
volume. However, their experimental results contradicted this, with similar rejection rates for
particles with a range of aspect ratios. The model assumed that particles would impact the
membrane with equal frequency for all particle orientations. They theorize that instead, an end-
on orientation was favored for transport of the particles in the water stream. They concluded that
microorganism removal by membranes could be conservatively estimated using only the rod
diameter in transport models. These findings add an additional safety factor to using Bacillus
endospores as a surrogate for C. parvum.
2. Electrophoretic Mobility and Isoelectric Point
A suitable surrogate should have a surface charge similar to C. parvum, as measured through the
isoelectric point and electrophoretic mobility (EPM). The isoelectric point is the pH at which the
particle has a neutral surface charge in an aqueous environment. Below this point the particle
has a net positive charge, above it a net negative charge. Many studies have pegged the
isoelectric point of C. parvum between pH values of 2 and 4, thus it would have a negative
surface charge in the neutral pH range. The isoelectric point can be found by measuring the
EPM of the particle at various pH values. The pH where the EPM is zero is classified as the
isoelectric point.
Lytle et. al. (2002) measured the EPM of both C. parvum and B. subtilis endospores in solutions
of increasing buffer concentration (0.915 millimolar (mM) 9.15 mM, and 91.5 mM KH2PO4).
They found that increasing the buffer concentration also increases the EPM toward a positive
value. The buffer concentration of the test water for the Siemens tests was approximately 1 mM.
Therefore, the 0.915 mM data from this study should be the most accurate representation of the
A-l
-------
C. parvum and B. subtilis EPM for the ETV tests. In 0.915 mM solutions at pH values between
7 and 8, they observed EPM of approximately -2.2 to -2.6 jim cm V"1 s"1 for C. parvum, and -1.9
to -2.2 |im cm V'V1 for B. subtilis. For B. subtilis, the researchers did not measure an isoelectric
point at any buffer concentration. For C. parvum, they did find an isoelectric point at a pH
around 2.5, but only for the 9.15 mM solution. For both organisms, the 0.915 mM solution
generally gave lower (more negative) EPM values than the solutions with higher buffering
capacity.
3. Aggregation
The NSF Microbiology Laboratory microscopically examined a sample of the B. atrophaeus
stock solutions purchased for the tests. The sample was suspended in sterile, buffered, deionized
water and stirred at moderate speed for 15 minutes. The estimated cell density was IxlO9
CFU/100 mL, which is approximately 100 times higher than the suspensions injected into the
pilot units to challenge the UF membranes. Figure 1 is a photograph of the B. atrophaeus
endospores in the sample. The magnification is lOOOx oil immersion with differential
interference contrast microscopy. No evidence of endospore aggregation was found.
Figure B-l. Mono-dispersed B. atrophaeus endospores used for challenge tests.
A-2
-------
References
Baltus, R. E., A. R. Badireddy, W. Xu, and S. Chellam (2009). Analysis of Configurational
Effects on Hindered Convection of Nonspherical Bacteria and Viruses across Microfiltration
Membranes. Industrial and Engineering Chemistry Research. In press.
Brush, C. F., M. F. Walter, L. J. Anguish, and W. C. Ghiorse (1998). Influence of Pretreatment
and Experimental Conditions on Electrophoretic Mobility and Hydrophobicity of
Cryptosporidium parvum Oocysts. Applied and Environmental Microbiology. 64: 4439-4445.
Butkus, M. A., J. T. Bays, and M. P. Labare (2003). Influence of Surface Characteristics on the
Stability of Cryptosporidium parvum Oocysts. Applied and Environmental Microbiology. 69:
3819-3825.
Lytle, D. A., C. H. Johnson, and E. W. Rice (2002). A Systematic Comparison of the
Electrokinetic Properties of Environmentally Important Microorganisms in Water. Colloids and
Surfaces B: Biointerfaces. 24: 91-101.
A-3
-------
Appendix B
Test/Quality Assurance Project Plan
Contact Mr. Bruce Bartley at 734-769-5148 or bartley@nsf.org for a copy of this document.
B-l
-------
Appendix C
Challenge Organism Triplicate Counts
C-l
-------
Table C-l. Module 1 B. atrophaeus Triplicate Counts
Sample
Start-up
15 Minutes
30 Minutes
Triplicate Counts (CFU/mL)
Feed
2.02xl04, 1.42xl04, 1.85xl04
1.90xl04, 1.28xl04, 1.59xl04
1.80xl04,2.04xl04, 1.24xl04
Filtrate
94, 52, 67
12, 15, 12
13, 16, 14
Table C-2. Module 2 B. atrophaeus Triplicate Counts
Sample
Start-up
15 Minutes
30 Minutes
Triplicate Counts (CFU/mL)
Feed
2.22xl04, 2.05xl04, l.SlxlO4
2.02xl04, 1.64xl04, 1.36xl04
1.70xl04, 1.91xl04, 1.65xl04
Filtrate
217, 190, 129
51,63,57
56, 46, 41
Table C-3. December 2007 Module 1 C. parvum Triplicate Counts
Sample
Start-up
15 Minutes
30 Minutes
Triplicate Counts (Oocysts/L)
Feed
1.36xl05, 1.44xl05, 9.8xl04
l.OlxlO5, 7.0xl04, 5.9xl04
7.9xl04, 5.4xl04, 8.5xl04
Filtrate
------- |