January 2011
NSF10/34/EPADWCTR
EPA/600/R-11/004
Environmental Technology
Verification Report
Removal of Microbial Contaminants in
Drinking Water
Dow Chemical Company - Dow Water
Solutions
SFD-2880 Ultrafiltration Module
Prepared by
NSF International
Under a Cooperative Agreement with
U.S. Environmental Protection Agency
-------
-------
January 2011
Environmental Technology Verification Report
Removal of Microbial Contaminants in Drinking Water
Dow Chemical Company - Dow Water Solutions
SFD-2880 Ultrafiltration Module
Prepared by:
NSF International
Ann Arbor, Michigan 48105
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, 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 U.S. Environmental Protection Agency (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 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
-------
Table of Contents
Verification Statement VS-i
Title Page i
Notice ii
Foreword iii
List of Tables v
List of Figures v
Abbreviations and Acronyms vi
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
Chapter 2 Product Description 3
2.1 UF Membrane General Description 3
2.2 SFD-2880 Membrane Module Description 3
Chapter 3 Methods and Procedures 5
3.1 Introduction 5
3.2 Organisms and Challenge Concentrations 5
3.3 Test Apparatus 6
3.4 Test Water Composition 8
3.5 UF Module Conditioning 9
3.6 Test Rig Sanitization 9
3.7 UF Module Integrity Tests 9
3.8 Microbial Challenge Test Procedure 9
3.9 MS2 Reduction vs. Flux 10
3.10 Analytical Methods 10
Chapter 4 Results and Discussion 12
4.1 Introduction 12
4.2 C. parvum Challenge Test 13
4.3 B. atrophaeus Endospores Challenge Tests 13
4.4 MS2 Challenge Tests 15
4.5 Pressure Decay Test Results 16
4.6 MS2 Reduction vs. Flux 17
4.7 Operational Data and Water Quality Data for All Challenges 19
Chapter 5 Quality Assuarance/Quality Control 21
5.1 Introduction 21
5.2 Test Procedure QA/QC 21
5.3 Sample Handling 21
5.4 Chemistry Laboratory QA/QC 21
5.5 Microbiology Laboratory QA/QC 21
5.6 Documentation 22
5.7 Data Review 22
5.8 Data Quality Indicators 22
Chapter 6 References 25
IV
-------
Appendices
Appendix A Test/Quality Assurance Proj ect Plan
Appendix B Bacillus Endospores as a Surrogate for C. parvum Oocysts
Appendix C Triplicate Challenge Organism Counts
List of Tables
Table 2-1. SFD-2880 Specifications 3
Table 2-2. Serial Numbers of Tested Modules 4
Table 3-1. Analytical Methods for Laboratory Analyses 11
Table 4-1. C. parvum Challenge Results 13
Table 4-2. B. atrophaeus Endospores Challenge Results 14
Table 4-3. B. atrophaeus LRVs with the Feed Capped at 6.5 Logic 15
Table 4-4. MS2 Challenge Results 16
Table 4-5. Pressure Decay Data 17
Table 4-6. MS2 vs. Flux Results 18
Table 4-7. MS2vs. Flux Study Pressure Decay Data 19
Table 4-8. Operation Data 19
Table 4-9. Water Chemistry Data 20
Table 5-1. Completeness Requirements 23
Table C-l. MS2 Triplicate Count Data C-l
Table C-2. B. atrophaeus Triplicate Count Data C-2
Table C-3. C., parvum Triplicate Count Data C-2
List of Figures
Figure 2-1. Diagram of the SFD-2880 UF module 4
Figure 3-1. Schematic diagram of the test rig used for verification testing 6
Figure 3-2. Photo of test rig 7
Figure 4-1. MS2 LRVs at Lower Fluxes 18
Figure B-l. Mono-dispersed B. atrophaeus endospores used for challenge tests B-2
-------
Abbreviations and Acronyms
ATCC American Type Culture Collection
ฐC degrees Celsius
CPU colony forming units
cm centimeter
EPM Electrophoretic Mobility
ETV Environmental Technology Verification
ฐF degrees Fahrenheit
ft foot(feet)
gfd gallons per square foot per day
gpm gallons per minute
in inch(es)
L liter
LRV log removal value
LT2ESWTR Long Term 2 Enhanced Surface Water Treatment Rule
m meter
MFGM Membrane Filtration Guidance Manual
mg milligram
min minute
mL milliliter
mm millimeter
mM millimolar
MWCO molecular weight cutoff
NRMRL National Risk Management Research Laboratory
NSF NSF International (formerly known as National Sanitation Foundation)
NTU Nephelometric Turbidity Unit
ORD Office of Research and Development
PFU plaque forming unit
psig pounds per square inch, gauge
PVDF polyvinylidene fluoride
QA quality assurance
QC quality control
RPD relative percent difference
SM Standard Methods for the Examination of Water and Wastewater
TDS total dissolved solids
TOC total organic carbon
UF ultrafiltration
ug microgram
jam microns
uS microsiemens
USEPA U. S. Environmental Protection Agency
VI
-------
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 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 Dow Chemical Company SFD-2880 Ultrafiltration (UF) membrane module was
conducted to verify microbial reduction performance under the membrane challenge
requirements of the USEPA Long Term 2 Enhanced Surface Water Treatment Rule
(LT2ESWTR). This report meets the "Membrane Challenge Test Requirements" in section
IV.D.ll.a. of the LT2ESWTR. The report does not address the following section 11.a.
LT2ESWTR requirements: Membrane Direct Integrity Testing; Continuous Indirect Integrity
Monitoring, nor any non-testing requirements such as product modifications, or assuring that the
membrane product sold conforms to the established quality control release value.
Please also note that this verification does not address long-term performance, or performance
over the life of the membrane. This verification test did not evaluate cleaning of the membranes,
nor any other maintenance and operation.
-------
While the LT2ESWTR only addresses Cryptosporidium, the EPA Membrane Filtration Guidance
Manual states that virus reduction can be tested under the same framework. Therefore, reduction
of the coliphage virus MS2 was also evaluated during this study, using the same test protocol as
that for Cryptosporidium reduction.
1.3 Testing Participants and Responsibilities
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: 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 Dow Chemical Company
The Dow Chemical Company supplied the tested membrane modules, and also provided
logistical and technical support, as needed.
Contact: The Dow Chemical Company - Dow Water Solutions
169 IN. Swede Road
Midland, MI 48674
Contact: Daryl Gisch
Phone: +1 989-636-9254
Email: dgisch@dow.com
-------
Chapter 2
Product Description
2.1 UF Membrane 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 can be either "inside-out" or "outside-in". 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
microns (urn). Typical MWCO points are 10,000 to 500,000 Daltons, 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 SFD-2880 Membrane Module Description
The Dow SFD-2880 is a polyvinylidene fluoride (PVDF) hollow fiber ultrafiltration membrane
module. The module specifications and operating parameters are listed in Table 2-1. The SFD-
2880 is a pressure driven module, with the normal operating flow orientation from the outside to
the inside of the fibers. The SFD-2880 is certified to NSF/ANSI Standard 61, which establishes
minimum public health related requirements for drinking water system components.
Table 2-1. SFD-2880 Specifications
Parameter
Dimensions:
Module outside diameter
Module length
Module volume
Nominal membrane pore size
Maximum membrane pore size
Average active membrane area (outer)
Operating Limits:
Filtrate flux range at 25ฐC
Flow range
Operating temperature range
Max. inlet module pressure
Max. transmembrane pressure (TMP)
Operating pH range
Max. NaOCl
Max. TSS
Max. Turbidity
Specification
8.9 inches (in) (225 millimeters (mm))
92.9 in (2360 mm)
10.3 gallons (gal) (39 liters (L))
0.03 urn
0.05 urn
829 square feet (ft2) (77 square meters (m2))
24-70 gallons per square foot per day (gfd) (40-120 L/m2/hr)
13.6-40.9 gallons per minute (gpm) (3.1-9.3 nrVhr)
34-104 Fahrenheit (ฐF) (1-40 Celcius (ฐC))
44 pounds per square inch (psi) (3.0 bar)
30psi(2.1bar)
2-11
2,000 milligrams per L (mg/L)
100 mg/L
300 Nephelometric Tuibidity Units (NTU)
-------
A diagram of the SFD-2880 module is pictured in Figure 2-1. The module design allows for an
optional reject line connection, but this port will be closed off for the challenge tests. The
modules will be operated in dead-end mode.
Figure 2-1. Diagram of the SFD-2880 UF module.
Dow supplied five new UF modules for testing. There was no seasoning period, other than that
specified by Dow to sufficiently rinse out the membrane preservative and wet the membranes.
See Section 3.5 for a description of the UF module conditioning procedure. The serial numbers
of the tested modules are listed in Table 2-2. The first five modules submitted for testing were
randomly selected by Dow personnel from existing inventory. For submission of the 6th module,
Dow provided NSF with the serial numbers of three modules on hand in their Edina, MN
warehouse, and NSF randomly selected a module for Dow to submit. The module numbers in
the first column are the numbers used in Chapter 4 to identify each module.
Table 2-2. Serial Numbers of Tested Modules
Module
1
2
3
4
5
6
Serial Number
PEO9B00016
PEO9B00010
PEO9B00007
PEO9B00028
PEO9B00017
PEO9B00007
-------
Chapter 3
Methods and Procedures
3.1 Introduction
The tests followed the procedures described in the Test/QA Plan for the Microbial Seeding
Challenge Study of the Dow Chemical Company SFD-2880 Ultrafiltration Module. The
challenge protocol was adapted from the ETV Protocol for Equipment Verification Testing for
Physical Removal of Microbiological and Particulate Contaminants, and the USEPA Membrane
Filtration Guidance Manual (MFGM). Note that the MFGM references the ETV protocol as an
acceptable protocol for testing membrane products according the to the USEPA requirements.
The test/QA plan is available from NSF upon request.
A total of six modules were submitted for testing. The test plan called for testing only five
modules, but the module tested for Cryptosporidium parvum reduction developed an apparent
membrane breach during the test. As a result, Dow chose to submit a sixth module for testing so
they could have a five module data set demonstrating the performance of fully integral modules.
The tests were conducted in October of 2009, February of 2010, and May of 2010. See Table 4-
8 for the dates of each individual challenge test. In between tests, the modules were stored wet,
and without any preservative, at NSF.
3.2 Organisms and Challenge Concentrations
The SFD-2880 modules were tested for removal of microorganisms using endospores of the
bacteria Bacillus atrophaeus (American Type Culture Collection (ATCC) number 9372,
deposited as Bacillus subtilis var. niger), and MS2 coliphage virus (ATCC 15597-B1). In
addition, one module was challenged with live Cryptosporidium parvum oocysts in order to
experimentally establish the B. atrophaeus endospores as a surrogate for Cryptosporidium. B.
atrophaeus was selected as a surrogate for C. parvum, due to the high cost and lack of
availability of suitable numbers of C. parvum for challenging all five modules. The strain of B.
atrophaeus used for testing yields orange colonies with a distinctive morphology on trypicase
soy agar (TSA), so it can be distinguished from wild-type endospores that could be present as
contamination. B. atrophaeus endospores are ellipsoidal (football shaped), with an average
diameter of 0.8 jim, and an average length of 1.8 jim. See Appendix B for further discussion
regarding the use of Bacillus endospores as a surrogate for Cryptosporidium.
Virus removal testing was conducted using MS2 for possible virus removal credits. MS2 is
considered a suitable surrogate for pathogenic viruses because of its small size, at approximately
24 nanometers in diameter.
The following were the target challenge concentrations for each organism:
MS2 - 5xl05 plaque forming units per milliliter (PFU/mL);
B. atrophaeus - IxlO7 colony forming units (CFU) per 100 mL; and
C. parvum - 5xl05 oocysts/L.
-------
The LT2ESWTR calls for the maximum challenge concentration to be 6.5 logic above the
organism's detection limit. The detection limit of all challenge organisms was 1 per unit
volume. The goal for the B. atrophaeus challenges was to be able to measure log reductions
greater than six, so NSF elected to target IxlO7 CFU/100 mL in order to account for less than
100% recovery of spiked challenge organism concentration. After all six modules were tested,
and the feed concentrations were found to be above 6.5 logic, NSF learned that the maximum 6.5
logio challenge level is not just guidance, but rather the maximum allowed in the rule language in
the Federal Register. Therefore, NSF decided to retest two modules with lower challenge levels
to provide a data set that meets the requirements of the rule.
The MS2 stock suspension was purchased from Biological Consulting Services of North Florida,
Inc. B. atrophaeus was purchased from Presque Isle Cultures. The C. parvum oocysts were
purchased from Sterling Parasitology Lab.
3.3 Test Apparatus
The modules were tested in a test rig constructed specifically for these tests. The test rig
construction conformed to the requirements of the MFGM. See Figure 3-1 for a schematic
diagram of the test rig, and Figure 3-2 for a photo of the test rig.
Filtrate
Pressure Gate
Sensingฎ \Val"e
Buffered Ol
Water Tank
Centrifugal Pump
Figure 3-1. Schematic diagram of the test rig used for verification testing.
-------
Figure 3-2. Photo of the test rig.
-------
The challenge organisms were introduced into the feed water by intermittent injection during the
challenge tests. Injection and mixing of the organisms followed the guidelines of the MFGM,
except for the suggested distance between the injection point and feed sample tap. Specifically,
the total stock solution volume injected into the feed stream during each challenge test was
between 0.5 and 2 percent of the total spiked test solution volume, a chemical metering pump
that delivered a steady flow of the challenge solution was used, and the injection port included a
quill extending into the middle of the feed pipe. The MFGM also calls for a static mixer to be
placed downstream of the injection point, and that the feed sample tap be located at least ten pipe
diameters downstream of the static mixer. NSF misread this suggestion as ten pipe diameters
including the static mixer. The inlet and outlet fittings on the SFD-2880 module are 2 inches
(DN50), so the test rig plumbing was also 2 inches in diameter. For this test rig, the distance
between the injection point and the feed sample tap, including the static mixer, was
approximately 27 inches. The distance between the static mixer and the feed sample tap was not
measured.
The filtrate grab samples were also collected from a sample tap with a quill extending into the
middle of the pipe. Both the feed and filtrate sample taps were metal so they were able to be
flame-sterilized prior to sample collection. The feed and filtrate sample ports were located
immediately upstream and downstream of the membrane module.
3.4 Test Water Composition
Local tap water was further treated by carbon filtration, reverse osmosis, ultraviolet disinfection,
and deionization at the NSF Laboratory to make the base water for the tests. The base water has
the following quality control (QC) requirements for use in the NSF testing laboratory:
Conductivity <2 microsiemens (|J,S) per centimeter (cm) at 25ฐC;
Total organic carbon <100 micrograms (|j,g) per L;
Total chlorine <0.05 mg/L; and
Heterotrophic bacteria plate count <100 CFU/mL.
Of the above parameters, only total chlorine and total organic carbon were measured specifically
for this verification. The other parameters are measured periodically by NSF as part of the
internal quality assurance (QA)/QC program for test water quality.
A 4,000-gallon water supply tank was filled with the base water. For the first round of challenge
tests in October 2009, sodium bicarbonate was added to the base water in sufficient quantity to
provide alkalinity at a target of 100 ฑ 10 mg/L as calcium carbonate. The pH was then lowered
with hydrochloric acid to a target range of 7.5 ฑ 0.5. For the retests in February 2010 and again
in May 2010, NSF elected to switch to phosphate buffering at 0.1 milliMolar, because phosphate
is called for in the ETV membrane challenge testing protocol.
Feed water samples were collected prior to each challenge period for analysis of total chlorine,
alkalinity, pH, temperature, total dissolved solids, total organic carbon, and turbidity. These
samples were collected prior to addition of the challenge organism.
-------
3.5 UF Module Conditioning
Prior to testing, the modules were conditioned following a proprietary procedure supplied by
Dow. Immediately prior to testing, each module was forward flushed at 40 gpm for one minute,
then backflushed for one minute at 40 gpm.
3.6 Test Rig Sanitization
The Dow module conditioning procedure included an hour long flush with a bleach solution.
This procedure was sufficient to sanitize the test rig prior to testing.
3.7 UF Module Integrity Tests
Before and after each challenge test, each module was subjected to a 20-minute pressure decay
test to satisfy the non-destructive performance test requirement in Section 3.6 of the MFGM.
The test procedure followed ASTM D6908-03 Standard Practice for Integrity Testing of Water
Filtration Membrane Systems. The water was drained from the feed side of the membrane, but
not the filtrate side. Approximately 20 psig of pressure was applied to the feed side and the
remaining pressure was recorded every minute to chart the pressure decay. This applied pressure
met the resolution requirement of Section 4.2.1 of the MFGM. The baseline decay rate of the
pressurized portion of the test rig was also measured over 20 minutes immediately prior to each
pre-challenge pressure decay test. This value was added to the expected UF module pressure
decay rate to ensure that the final applied pressure at the end of the 20-minute test still met the
applied pressure resolution requirement.
3.8 Microbial Challenge Test Procedure
Each of the SFD-2880 modules submitted for testing was challenged individually, as shown in
the photo of the test rig (Figure 3-2). The target flux for membrane operation was Dow's
maximum recommended value of 70 gfd at 25 ฐC, which equals a flow rate of approximately 40
gpm.
Separate tests were conducted for each challenge organism, so each module was tested twice
over the course of the testing activities. In addition, two modules were tested a third time with
live C. parvum oocysts. The modules chosen for the C. parvum challenges were the ones with
the highest filtrate counts from the Bacillus endospores challenges. For most of the modules,
both the MS2 and B. atrophaeus challenges were conducted on the same day.
At the end of the forward flush described in Section 3.5, a filtrate sample was collected to serve
as the negative control flush sample.
Each challenge test was approximately 35 minutes in length. As discussed in Section 3.3, the
challenge organisms were intermittently injected into the feed stream prior to, and during sample
collection. Sections 3.10.2, 3.10.4, and 3.12.4 of the MFGM describe the requirements for the
challenge test sampling plan. The MFGM requires that feed and filtrate samples not be collected
until at least three hold-up volumes of water containing the challenge organism have passed
through the membrane, to allow for establishment of equilibrium (equilibrium volume). The
hold-up volume is defined as the "unfiltered test solution volume that would remain in the
-------
system on the feed side of the membrane at the end of the test." Dow's specification sheet for
the SFD-2880 gives the module volume as 10.3 gal. It is assumed that this volume is the total
water holding volume of the module, not just the volume of the feed side of the membranes. As
such, its use as the module hold-up volume added a safety factor to the holdup volume
calculation.
The total hold-up volume also needs to include the pipe volume between the injection port and
the module inlet. As discussed in Section 3.3, the test rig used 2-inch diameter pipe, and the
injection port was approximately 45 inches upstream from the module inlet. Forty-five inches of
2-inch diameter pipe has a volume of approximately 141 in3, which translates into 0.61 gal. The
pipe volume plus the module volume gives a total hold-up volume of approximately 10.9 gal,
which can be rounded up to 11 gal. If the hold-up volume is 11 gal, then the equilibrium volume
is 33 gal. The challenge flow rate approximately 40 gpm, so the challenge organisms needed to
be injected for about 1 minute prior to sampling to meet the requirement of passing the
equilibrium volume. In practice, the injection times prior to sampling approximately two
minutes, because the test engineer had to adjust the injection flow rate at the start of each
injection period to ensure the proper challenge concentration.
The challenge organisms were injected into the feed stream at start-up, after 15 minutes of
operation, and after 30 minutes of operation. After at least one minute of injection, grab samples
were collected from the feed and filtrate sample taps. The sample taps were flame sterilized
prior to sample collection. Also, at least 100 mL was collected and discarded prior to collection
of each sample to flush the taps. After each round of sample collection, injection of the
challenge organism suspension was turned off, and clean feed water was pumped through the
modules at 40 gpm until the next sampling point.
3.9 MS2 Reduction vs. Flux
After the MS2 reduction data was shared with Dow, they requested that NSF conduct three more
MS2 reduction challenges on one module at lower fluxes to identify whether MS2 reduction
would increase as the flux was lowered, and to generate a curve of MS2 reduction vs. flux.
The challenge test procedure was the same as described above in Section 3.8. Module 5 was
randomly chosen for testing by the laboratory engineer. The tests were conducted at the
following target flow rates specified by Dow: 13.6 gpm, 25.4 gpm, and 35.6 gpm. These flow
rates translate into fluxes of 23.6, 44.1, and 61.8 gfd, respectively.
3.10 Analytical Methods
A list of laboratory analytical methods can be found in Table 3-1. Single samples of adequate
volume were collected for challenge organism enumeration, and were analyzed in triplicate.
10
-------
Table 3-1. Analytical Methods for Laboratory Analyses
Parameter
Alkalinity (total, as CaCO3)
pH
Total Dissolved Solids (TDS)
Total Chlorine
Total Organic Carbon (mg/L)
Turbidity
MS2
B. atrophaeus Endospores
Crypto sporidium Oocysts
Method
USEPA310.2
SM1 4500-H+
SM 2540 C
SM 4500-C1 G
SM5310C
SM2130
NSF 554
SM92185
USEPA 1623
NSF
Reporting Limit
5 mg/L
NA2
5 mg/L
0.05 mg/L
0.1 mg/L
0.1 NTU
1 PFU/mL
1 CFU/100 mL
1 oocyst/L
Hold Time
14 days
none3
7 days
none3
28 days
none3
30 hours
30 hours
72 hours
(1) SM = Standard Methods
(2) Not Applicable
(3) Immediate analysis required
(4) Method published in NSF/ANSI Standard 55 - Ultraviolet Microbioloง
EPA Method 1601.
(5) TSA was substituted for nutrient agar in SM 9218 so that the challen;
endospores. TSA gives orange colonies with a distinctive morphology.
ical Water Treatment Systems. Method is similar to
;e endospores could be distinguished from wild-type
11
-------
Chapter 4
Results and Discussion
4.1 Introduction
For presentation of the challenge organism data in this chapter, the observed triplicate counts
were averaged by calculating geometric means, as suggested for microbial enumeration data in
SM 9020. Geometric means <1 were rounded up to 1, unless all three triplicate analyses had no
organisms found. The mean counts were logic transformed for the purpose of calculating log
removal values (LRV). The triplicate counts for each sample are presented in Appendix C.
The LT2ESWTR and MFGM specify that an LRV for the test (LRVc-iEsi) be calculated for each
module tested, and that the LRVs for each module are then combined to yield a single LRVc-iEsi
for the product. If fewer than 20 modules are tested, as was the case for this verification, the
LRVc-iEST is simply the lowest LRV for the individual modules. However, the rule does not
specify a method to calculate LRVc-iEsi for each module. Suggested options in the MFGM
include:
1. Calculate a LRV for each feed/filtrate sample pair, then calculate the average of the
individual sample point LRVs;
2. Average all of the feed and filtrate counts, and then calculate a single LRV for the
module; or
3. Calculate a LRV for each feed/filtrate sample pair, select the LRV for the module as the
lowest (most conservative of the three options).
Options 1 and 2 give LRVc-iEST values that are either identical, or within a few hundredths of
each other, so in this report, options 1 and 3 are used to calculate the LRV for each module.
Since the triplicate counts were averaged by calculating geometric means, so too do the LRV
calculations use geometric mean.
Each module was challenged with both B. atrophaeus and MS2 on the same day. After all of the
modules were tested, the B. atrophaeus data was examined to choose the module to undergo the
C. parvum challenge test. Modules 2 and 3 were the only ones with B. atropheaus CFU found in
all three triplicate counts of a filtrate sample. For Module 2, 1 CFU was found in each of the
triplicate measurements for the 2-minute filtrate sample. For Module 3, the 30-minute filtrate
sample triplicate counts were 3,1, and 1 CFU, so Module 3 was chosen over Module 2 for the C.
parvum test. During the C. parvum test, there was a possible integrity breach that developed,
because the post-test pressure decay rate was approximately double that measured immediately
before the challenge test (See Table 4-5 for the pressure decay data). When the filtrate samples
were analyzed, one C. parvum oocyst was found in one of the triplicate counts for the 30-minute
filtrate sample. As a result, Dow decided to submit a sixth module for testing. This sixth
module was first challenged with B. atrophaeus in order to compare its performance to the other
modules. The B. atrophaeus data set was re-examined, omitting Module 3, and Module 2 was
chosen for a second C. parvum challenge test.
12
-------
As discussed in Section 3.2, after the tests were conducted, NSF learned that the maximum 6.5
logic challenge level is not just guidance, but rather the maximum allowed in the rule language in
the Federal Register. Therefore, NSF decided to randomly pick two modules to retest with lower
challenge levels to provide a data set that meets the requirements of the rule. NSF also learned
from EPA that the States could accept the data from the high feed challenge tests, provided that
the feed concentrations were capped at 6.5 logio for the purpose of calculating the LRV.
Therefore, two sets of LRV calculations are presented, one set using the observed feed counts,
and a second set with the feed concentration set at 6.5 logic.
4.2 C. parvum Challenge Test
The C. parvum challenge data is presented in Table 4-1. As discussed in Section 4.1, based on a
review of the B. atrophaeus challenge data, Module 3 was challenged with live C. parvum
oocysts. One oocyst was found in one of the three triplicate analyses of the 30-minute filtrate
sample from the Module 3 challenge test, so Dow requested that a second module be challenged
with C. parvum, and they submitted a sixth module for testing. The sixth module was tested with
B. atrophaeus, and the data set was reviewed again, excluding Module 3. NSF chose to test
Module 2 for the second C. parvum challenge. No oocysts were found in any filtrate samples
from the Module 2 challenge test.
In Table 4-1, the LRVs in the "Overall Mean" rows are the geometric means of the individual
sample point LRVs. The LRVc-iEST using the overall means is 6.20. The LRVc-iEST based on
the individual sample pairs is 5.97.
Table 4-1. C. parvum Challenge Results
Module
Number
Module 3
Module 2
Sample
Point
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Feed
Geometric Mean
(Oocysts/L)
1.74xl06
2.4xl06
9.4x10"
1.6xl06
2.02xl06
1.5xl06
1.92xl06
l.SxlO6
Log10
6.24
6.38
5.97
6.20
6.31
6.18
6.28
6.26
Filtrate
Geometric Mean
(Oocysts/L)
<1
<1
<1
1
1
<1
<1
<1
<1
<1
Log10
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
LRV
6.24
6.38
5.97
6.20
6.31
6.18
6.28
6.26
4.3 B. atrophaeus Endospores Challenge Tests
The B. atrophaeus endospore challenge results are displayed in Table 4-2. As discussed
previously, the challenge concentrations for the first round of tests were above the allowable
maximum of 6.5 logic, so two modules were retested with lower challenge concentrations. The
results of these two retests are displayed at the bottom of Table 4-2. NSF has also learned from
EPA that the States can accept the test data with the feed concentrations capped at 6.50 logic.
Therefore, the LRV calculations with the feed concentrations capped at 6.50 logic are presented
in Table 4-3. Excluding the Module 2 and 4 lower challenge data, all modules had mean B.
13
-------
atrophaeus LRVs of 6.50, except for Module 4, which was 6.40 due to one filtrate sample that
was above 0.0. The LRVc-iEsi from this subset would be 6.40 based on the overall mean LRVs,
or 6.20 based on the lowest individual sample point LRVs. If the Module 2 and 4 lower
challenge retest data is included in the dataset for determination of the LRVc-iEsi, it is 5.90 based
on the overall mean LRVs, or 5.77 based on the lowest individual sample point LRVs.
Table 4-2. B. atrophaeus Endospores Challenge Results
Challenge
Test
Module 1
Module 2
Module 3
Module 4
Module 5
Module 6
Module 2
Retest with
Lower
Challenge
Module 4
Retest with
Lower
Challenge
Sample
Point
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Feed
Geometric Mean
(CFU/100 mL)
1.13x10'
1.1x10'
1.17x10'
1.1x10'
1.36x10'
1.20x10'
1.22x10'
1.26x10'
1.07x10'
1.0x10'
1.35x10'
1.1x10'
1.16x10'
1.0x10'
7.3xl06
9.5xl06
1.24x10'
1.28x10'
1.44x10'
1.32x10'
1.46xl07
1.63xl07
1.43xl07
1.50xl07
9.4xl05
9.5xl05
l.OxlO6
9.6xl05
1.29xl06
LlSxlO6
1.29xl06
1.25xl06
Log10
7.05
7.04
7.07
7.05
7.13
7.08
7.09
7.10
7.03
7.00
7.13
7.05
7.06
7.00
6.86
6.97
7.09
7.11
7.16
7.12
7.16
7.21
7.16
7.18
5.97
5.98
6.00
5.98
6.11
6.07
6.11
6.10
Filtrate
Geometric Mean
(CFU/100 mL)
<1
<1
1
<1
1
2
1
1
1
1
<1
1
<1
1
1
1
<1
2
<1
1
1
<1
<1
<1
<1
1
1
1
<1
1
<1
1
<1
1
1
1
2
2
1
2
Log10
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.3
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.3
0.3
0.0
0.2
LRV
7.05
7.04
7.05
7.05
7.13
7.08
7.09
7.10
7.03
7.00
7.13
7.05
7.06
6.70
6.86
6.87
7.09
7.11
7.16
7.12
7.16
7.11
7.16
7.18
5.97
5.98
6.00
5.98
5.81
5.77
6.11
5.90
14
-------
While an oocyst was found in only one of the filtrate samples from the C. parvum challenges, B.
atrophaeus endospores were found in many filtrate samples. This provides experimental
evidence that endospores are indeed a conservative surrogate for Cryptosporidium.
Table 4-3. B. atrophaeus LRVs with the Feed Capped at 6.5 Logio
Challenge Test
Module 1
Module 2
Module 3
Module 4
Module 5
Module 6
Sample Point
2 Minutes
15 Minutes
30 Minutes
Overall Mean
2 Minutes
15 Minutes
30 Minutes
Overall Mean
2 Minutes
15 Minutes
30 Minutes
Overall Mean
2 Minutes
15 Minutes
30 Minutes
Overall Mean
2 Minutes
15 Minutes
30 Minutes
Overall Mean
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Log10 of Feed
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
Log10 of Filtrate
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.3
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
LRV
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.20
6.50
6.40
6.50
6.50
6.50
6.50
6.50
6.50
6.50
6.50
4.4 MS2 Challenge Tests
Table 4-4 displays the MS2 challenge data. As with the B. atrophaeus challenge levels, the MS2
challenges above 6.50 logic were capped to calculate the LRVs. There was a wide range of MS2
reduction observed, from a mean LRV of 4.51 logic for Module 1, down to 2.54 logic for Module
6. Under both LRVc-iEST calculation methods, Module 6 gives the LRVc-iEST of 2.54 or 2.37
logic.
15
-------
Table 4-4. MS2 Challenge Results
Module 1
Module 2
Module 3
Module 4
Module 5
Module 6
Sample
Point
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Feed
Geometric Mean
(PFU/mL)
3.0xl06
3.0xl06
3.2xl06
3.07xl06
2.3xl06
S.lxlO6
3.4xl06
2.9xl06
2.41xl06
.34xl06
.14xl06
.54xl06
.31xl06
.22xl06
l.lxlO6
1.2xl06
7.6xl05
9.3xl05
9.3xl05
8.7xl05
1.15x10'
1.2x10'
1.40x10'
1.2x10'
Log10
6.48
6.48
6.51
6.49
6.36
6.49
6.53
6.46
6.38
6.13
6.06
6.19
6.12
6.09
6.04
6.08
5.88
5.97
5.97
5.94
7.06
7.08
7.15
7.10
Capped
Log10
6.50
6.50
6.50
6.50
6.50
6.50
Filtrate
Geometric Mean
(PFU/mL)
1
80
93
l.lxlO2
94
<1
5.8xl02
5.3xl02
4.4xl02
5.1xl02
<1
4.8xl02
5.7xl02
4.7xl02
5.0xl02
<1
l.lxlO3
5.0xl02
3.0xl02
5.5xl02
<1
7.7xl02
4.6xl02
3.3xl02
4.9xl02
<1
5.2xlOj
l.lxlO4
1.35xl04
9.2x10"
Log10
1.90
1.97
2.04
1.97
2.76
2.72
2.64
2.71
2.68
2.76
2.67
2.70
3.04
2.70
2.48
2.74
2.89
2.66
2.52
2.69
3.72
4.04
4.13
3.96
LRV
4.58
4.51
4.47
4.52
3.60
3.77
3.89
3.75
3.70
3.37
3.39
3.48
3.08
3.39
3.56
3.34
2.99
3.31
3.45
3.25
2.78
2.46
2.37
2.54
4.5 Pressure Decay Test Results
The pre-test and post-test pressure decay test results are displayed in Table 4-5. Immediately
prior to each pre-test pressure decay measurement, the background pressure decay rate of the
pressurized test rig plumbing was measured, and the observed background decay rate, if any, was
recorded. The background pressure decay rates were subtracted from the measured decay rates,
and the corrected pressure decay rates are displayed in the last column of Table 4-5. For most
challenge tests, the post-test pressure decay rate was lower than the pre-test decay rate. The
membranes were not backwashed prior to measuring the post-test pressure decay rate, so the
lower post-test decay rates could be a result of accumulation of particulate matter, including the
challenge particulates, on the membrane surface. However, the challenge test results do not
indicate that any accumulation of particulates improved membrane performance over the test
periods.
16
-------
Table 4-5. Pressure Decay Data
Module
#1
#2
#3
#4
#5
#6
Test
MS2 Pre-test
MS2 Post-test
Bac. Pre-test
Bac. Post-test
MS2 Pre-test
MS2 Post-test
Bac. Pre-test
Bac. Post-test
C.parvum Pre-test
C. parvum Post-test
Bac. Pre-retest
Bac. Post-retest
MS2 Pre-test
MS2 Post-test
Bac. Pre-test
Bac. Post-test
C. parvum Pre-test
C. parvum Post-test
MS2 Pre-test
MS2 Post-test
Bac. Pre-test
Bac. Post-test
Bac. Pre-retest
Bac. Post-retest
MS2 Pre-test
MS2 Post-test
Bac. Pre-test
Bac. Post-test
MS2 Pre-test
MS2 Post-test
Bac. Pre-test
Bac. Post-test
Date
10/05/2009
10/05/2009
10/05/2009
10/05/2009
10/14/2009
10/14/2009
10/14/2009
10/14/2009
02/26/2010
02/26/2010
05/14/2010
05/14/2010
10/02/2009
10/02/2009
10/02/2009
10/02/2009
10/23/2009
10/23/2009
10/07/2009
10/07/2009
10/07/2009
10/07/2009
05/14/2010
05/14/2010
10/09/2009
10/09/2009
10/09/2009
10/09/2009
02/11/2010
02/11/2010
02/11/2010
02/11/2010
Starting
Pressure
(psig)
16.86
16.40
16.96
16.76
20.88
20.70
20.90
20.26
20.91
20.59
21.42
21.39
15.62
17.09
16.29
16.70
20.85
22.40
21.27
20.76
20.55
20.55
21.46
21.68
20.27
20.07
20.19
21.17
20.26
20.39
20.31
20.29
Final
Pressure
(psig)
15.97
15.68
16.19
16.63
19.55
20.58
19.83
20.15
20.32
19.83
20.54
20.17
14.57
15.82
15.26
15.46
19.50
19.78
19.84
20.60
19.29
19.13
20.39
20.12
19.17
19.32
18.98
20.21
19.17
19.53
19.60
19.58
Elapsed
Time
(min)
20.67
20.00
20.00
20.00
20.00
20.00
20.00
20.00
20.00
20.00
20.00
20.00
20.00
20.13
20.00
20.00
20.00
20.66
21.33
20.00
20.00
20.00
20.00
22.00
20.00
20.00
20.00
20.00
20.00
20.00
20.00
20.00
Decay Rate
(psig/min)
0.0431
0.0360
0.0385
0.0065
0.0665
0.0060
0.0535
0.0055
0.0295
0.0380
0.0440
0.0610
0.0525
0.0631
0.0515
0.0620
0.0675
0.1268
0.0670
0.0080
0.0630
0.0710
0.0535
0.0709
0.0550
0.0375
0.0605
0.0480
0.0545
0.0430
0.0355
0.0355
Background
Decay Rate
(psig/min)
0.0000
0.0000
0.0000
0.0000
0.0085
0.0085
0.0085
0.0085
0.0000
0.0000
0.0110
0.0110
0.0000
0.0000
0.0000
0.0000
0.0180
0.0180
0.0255
0.0255
0.0255
0.0255
0.0700
0.0700
0.0170
0.0170
0.0170
0.0170
0.0095
0.0095
0.0095
0.0095
Corrected
Decay
Rate
(psig/min)
0.0431
0.0360
0.0385
0.0065
0.0580
-0.0025
0.0450
-0.0030
0.0295
0.0380
0.0330
0.0500
0.0525
0.0631
0.0515
0.0620
0.0495
0.1088
0.0415
-0.0175
0.0375
0.0455
-0.0165
0.0009
0.0380
0.0205
0.0435
0.0310
0.0450
0.0335
0.0260
0.0260
4.6 MS2 Reduction vs. Flux
Dow requested that NSF conduct three extra MS2 challenge tests at lower flows to determine
whether MS2 reduction increased as the flux decreased. Module #5 was chosen for these tests
because it was the worst performing module of the five that had been tested at the time.
The data for these tests is displayed in Table 4-6. The LRV numbers for each challenge test are
also displayed graphically in Figure 4-1. The data does indicate that MS2 reduction is inversely
proportional to the flux, but the observed LRVs for the lower flow rate tests are all within the
range of LRVs from the maximum flux tests, except for the first sampling point from the 13.6
gpm test. The feed concentrations for these challenges are not capped at 6.5 logic because the
17
-------
intent of this study was not to provide regulatory compliance data, but rather only to supply
comparative data on membrane performance at lower fluxes.
Table 4-6. MS2 vs. Flux Results
Test Flow
13.6 gpm
25.4 gpm
35.6 gpm
Sample Point
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Flush
2 Minutes
15 Minutes
30 Minutes
Overall Mean
Feed
Geometric Mean
(PFU/mL)
1.31xl07
l.lxlO7
l.lxlO7
1.2xl07
l.OxlO7
l.lxlO7
1.2xl07
l.lxlO7
1.1x10'
1.0x10'
8.5xl06
1.0x10'
Logio
7.12
7.04
7.04
7.07
7.00
7.04
7.08
7.04
7.04
7.00
6.93
6.99
Filtrate
Geometric Mean
(PFU/mL)
<1
223
690
1.35xl03
590
44
1.52xl03
3.4xl03
5.4xl03
3.0xl03
50
2.9xl03
6.5xl03
1.41xl04
6.5xl03
Logio
2.35
2.84
3.13
2.77
3.18
3.53
3.73
3.48
3.46
3.81
4.15
3.81
LRV
4.77
4.20
3.91
4.29
3.82
3.51
3.35
3.56
3.58
3.19
2.78
3.18
-=
I
3
it
e
J
0
Start
13.6gpm
25.4 gpm
35.6 gpm
15 Minutes
30 Minutes
Figure 4-1. MS2 LRVs at lower flow rates.
18
-------
The pressure decay data for the lower flux tests is displayed in Table 4-7. The data does not
indicate that there were any membrane integrity issues during these challenge tests.
Table 4-7. MS2 vs. Flux Study Pressure Decay Data
Module
#5
Test
13.6 gpmPre-test
13.6 gpm Post-test
25.4 gpmPre-test
25.4 gpm Post-test
35.6 gpmPre-test
35.6 gpm Post-test
Date
02/24/2010
02/24/2010
02/24/2010
02/24/2010
02/25/2010
02/25/2010
Starting
Pressure
(psig)
20.26
21.03
20.96
20.97
21.14
21.02
Final
Pressure
(psig)
19.55
20.33
20.04
20.17
20.34
20.15
Elapsed
Time
(min)
20.00
20.00
20.00
20.00
20.00
20.00
Decay Rate
(psig/min)
0.0355
0.0350
0.0460
0.0400
0.0400
0.0435
Background
Decay Rate
(psig/min)
0.0005
0.0005
0.0005
0.0005
0.0305
0.0305
Corrected
Decay
Rate
(psig/min)
0.0350
0.0345
0.0455
0.0395
0.0095
0.0130
4.7 Operational Data and Water Quality Data for All Challenges
The pilot unit operational data is presented in Table 4-8. The filtrate flows, and feed and filtrate
pressure readings were recorded onto bench sheets. The fluxes were calculated from the flow
data. The target flux for the tests was 70.0 gfd. The recorded flows translated into fluxes
ranging from 69.5 to 71.9 gfd. Except for the 41.4 gpm filtrate flow measurement, all recorded
flows were within 0.5 gpm or less of the target flow of 40.3 gpm for the tests.
Table 4-8. Operation Data
Module #
Date
Filtrate Flow Rate
(gpm)
OMin.
30 Min.
Flux
(gfd)
OMin | 30 Min
Feed Pressure
(psig)
OMin.
30 Min.
Filtrate Pressure
(psig)
OMin.
30 Min.
MS2 Challenges
Module 1
Module 2
Module 3
Module 4
Module 5
Module 6
Mod. 5 13.6 gpm
Mod. 5 25.4 gpm
Mod. 5 35.6 gpm
10/05/09
10/14/09
10/02/09
10/07/09
10/09/09
02/11/10
02/24/10
02/24/10
02/25/10
40.0
40.3
40.2
41.4
40.0
40.7
13.7
25.7
35.7
40.1
40.1
40.0
40.7
40.0
40.4
13.7
25.4
35.7
69.5
70.0
69.8
71.9
69.5
70.7
23.8
44.6
62.0
69.7
69.7
69.5
70.7
69.5
70.2
23.8
44.1
62.0
25.75
21.62
21.90
24.29
25.06
24.95
13.08
18.52
24.86
25.05
20.90
20.97
23.31
24.38
23.80
12.65
18.02
24.20
2.10
0.55
1.06
2.52
2.16
0.84
2.07
1.40
2.68
2.02
0.51
1.05
2.32
2.19
0.74
2.01
1.43
2.72
B. atrophaeus Challenges
Module 1
Module 2
Module 3
Module 4
Module 5
Module 6
Mod. 2 Retest
Mod. 4 Retest
10/05/09
10/14/09
10/02/09
10/07/09
10/09/09
02/11/10
05/14/10
05/13/10
40.3
40.1
40.1
40.1
40.4
40.3
40.0
40.8
40.0
40.0
40.1
40.1
40.0
40.3
40.4
40.4
C.i
Module 3
Module 2
10/23/09
02/26/10
40.3
40.1
40.1
40.3
70.0
69.7
69.7
69.7
70.2
70.0
69.5
70.9
69.5
69.5
69.7
69.7
69.5
70.0
70.2
70.2
25.88
21.83
22.17
23.65
25.38
25.92
27.42
27.78
25.35
21.29
21.42
22.97
24.33
25.12
27.01
26.80
1.97
0.25
2.08
1.93
2.09
1.86
6.72
6.04
2.01
0.32
2.13
1.95
2.01
1.92
6.86
5.82
oarvum Challenge
70.0
69.7
69.7
70.0
21.83
26.69
20.96
24.85
1.39
4.05
1.36
3.16
19
-------
The feed water chemistry data is displayed in Table 4-9. As discussed in Section 3.4, the water
recipe was changed between the first round of the tests in October 2009, and the second and third
rounds of tests in February and May of 2010. The phosphate buffered water had very low
alkalinity and TDS compared to the calcium carbonate buffered water. The phosphate buffered
water also had a lower pH. In fact, the calcium carbonate buffered water slightly exceeded the
target pH range of 7.0 to 8.0 for three of the eight tests with that water.
Note that the measured alkalinity for the Module 3 C. parvum challenge was <5 mg/L, when it
should have been around 100 mg/L. There was a preliminary alkalinity of 99 mg/L recorded on
the bench sheet for this test, but the pH was measured at 6.70, which is similar to the other pH
measurements for the phosphate buffered waters with no carbonate alkalinity. Therefore, NSF
believes that the water did not contain any calcium carbonate. NSF does not think this is a
significant issue though, since the Module 3 C. parvum challenge water was similar to the
phosphate buffered water with respect to alkalinity and pH.
Table 4-9. Feed Water Chemistry Data
Module #
Alkalinity
(mg/L
CaCO3)
PH
Temp. (ฐC)
Total
Chlorine
(mg/L)
TDS (mg/L)
TOC
(mg/L)
Turbidity
(NTU)
MS2 Challenges
Module 1
Module 2
Module 3
Module 4
Module 5
Module 6
Mod. 5 13.6 gpm
Mod. 5 25.4 gpm
Mod. 5 35.6 gpm
83
99
99
100
99
ND(5)
6
6
6
7.98
7.97
7.95
7.96
8.00
7.30
6.73
6.72
6.83
22.0
21.9
22.2
22.5
21.9
19.7
18.5
18.9
17.5
O.05
<0.05
O.05
<0.05
O.05
O.05
O.05
<0.05
O.05
86
120
100
110
110
16
28
22
20
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.10
0.11
0.07
0.09
0.07
0.20
0.15
0.25
0.09
B. atrophaeus Challenges
Module 1
Module 2
Module 3
Module 4
Module 5
Module 6
Mod. 2 Retest
Mod. 4 Retest
97
100
92
100
97
5
7
8
8.08
8.01
7.96
7.96
8.05
7.36
7.26
7.41
22.4
22.0
22.1
19.9
22.2
20.0
19.7
19.0
O.05
O.05
O.05
<0.05
O.05
<0.05
O.05
<0.05
100
120
92
120
110
12
23
27
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.11
0.14
0.21
0.10
0.08
0.15
0.11
0.28
C. parvum Challenge
Module 3
Module 2
<5
6
6.70
7.13
21.5
17.4
<0.05
O.05
<5
11
0.3
<0.1
0.14
0.18
20
-------
Chapter 5
Quality Assurance/Quality Control
5.1 Introduction
An important aspect of verification testing is the 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.
5.2 Test Procedure QA/QC
NSF testing laboratory staff conducted the tests by following a USEPA-approved test/QA 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 identification 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.
The NSF QA/QC requirements are all compliant with those given in the USEPA method or
Standard Method for the parameter. Also, every analytical method has an NSF standard
operating 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.
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.
21
-------
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
Microsoft Excelฎ spreadsheets. These spreadsheets were used to calculate the geometric
means and logic reductions. One hundred percent of the data entered into the spreadsheets was
checked by a reviewer to confirm all data and calculations were 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, 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 is a qualitative term that expresses "the degree to which data accurately and
precisely represent a characteristic of a population, parameter variations at a sampling point, a
process condition, or an environmental condition." 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.
The following equation was used to calculate percent recovery:
Percent Recovery = 100 X [(X^owa -
where: Xknown = 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 TDS were analyzed in
batches. Certified QC standards and/or matrix spikes were run with each batch.
The NSF Laboratory Quality Assurance Manual establishes the frequency of spike sample
analyses at 10% of the samples analyzed for chemical analyses. Laboratory control samples are
also run at a frequency of 10%. The recovery limits specified for the parameters in this
22
-------
verification, excluding microbiological analyses, were 70-130% for laboratory-fortified samples
and 85-115% for laboratory control samples. The NSF QA department reviewed the laboratory
records and found that all recoveries were within the prescribed QC requirements. Calibration
requirements were also achieved for all analyses.
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 IDS
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 =
x200
where:
Sl = sample analysis result; and
^ = sample duplicate analysis result.
Acceptable analytical precision for the verification test was set at an RPD of 30%. Field
duplicates were collected at a frequency of 1 out of every 10 samples for each parameter, to
incorporate both sampling and analytical variation to measure overall precision against this
objective. In addition, the NSF Laboratory also conducted laboratory duplicate measurements at
10% frequency of samples analyzed. The laboratory precision for the methods selected was
tighter than the 30% overall requirement, generally set at 20% based on the standard NSF
Chemistry Laboratory method performance.
All RPD 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%
23
-------
Completeness is defined as follows for all measurements:
%C = (V/T)x 100
where:
%C = percent completeness;
V = number of measurements judged valid; and
T = total number of measurements.
One hundred percent completeness was achieved for all aspects of this verification. All planned
testing activities were conducted as scheduled, and all planned samples were collected for
challenge organism and water chemistry analysis.
24
-------
Chapter 6
References
APHA, AWWA, and WEF (1999). Standard Methods for the Examination of Water and
Wastewater, 20th Edition.
NSF International (2007). NSF/ANSI Standard 55 - Ultraviolet Microbiological Water
Treatment 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 P articulate Contaminants
25
-------
Appendix A
Test/Quality Assurance Project Plan
Contact Mr. Bruce Bartley at 734-769-5148 or bartley@nsf.org for a copy of this document.
A-l
-------
Appendix B
Bacillus Endospores as a Surrogate for C parvum 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 KH^PO/i).
They found that increasing the buffer concentration also increases the EPM toward a positive
value. The buffer concentration of the test water for the Dow verification tests was
approximately 1 mM for the carbonate buffered test water, and 0.1 mM for the phosphate
B-l
-------
buffered test water. Therefore, the 0.915 mM data from this study should be the most accurate
representation of the 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
7-1 -1
C. parvum., and -1.9 to -2.2 jim cm V" s" 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 B-l 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.
B-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.
B-3
-------
Appendix C
Challenge Organism Triplicate Counts
Table C-l. MS2 Triplicate Count Data
Module
Module 1
Module 2
Module 3
Module 4
Module 5
Module 6
Module 5 at
13.6 gpm
Module 5 at
25.4 gpm
Module 5 at
35.6 gpm
Sample
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Feed (PFU/mL)
Count 1
3.4xl06
2.8xl06
3.4xl06
S.OxlO6
2.8xl06
4.7xl06
2.37xl06
1.49xl06
l.OSxlO6
1.28xl06
1.22xl06
9.7xl05
7.9xl05
1.19xl06
9.4xl05
1.26x10'
1.48x10'
1.60x10'
1.42xl07
1.58xl07
1.45xl07
1.08x10'
1.28x10'
1.49x10'
1.33x10'
1.39x10'
9.9xl06
Count 2
2.77xl06
S.lxlO6
2.9xl06
2.5xl06
4.2xl06
2.6xl06
2.44xl06
LlSxlO6
1.21xl06
1.53xl06
1.35xl06
1.33xl06
7.6xl05
S.OxlO5
9.1xl05
1.13x10'
1.34xl06
1.37x10'
1.36xl07
8.5xl06
1.04xl07
9.8xl06
9.8xl06
9.6xl06
9.7xl06
7.9xl06
7.7xl06
Count 3
2.97xl06
S.OxlO6
3.4xl06
1.64xl06
2.5xl06
3.2xl06
2.42xl06
1.37xl06
LlSxlO6
1.14xl06
1.09xl06
l.llxlO6
7.2xl05
8.5xl05
9.5xl05
1.06x10'
9.1xl06
1.25x10'
1.16xl07
LlSxlO7
S.lxlO6
9.9xl06
1.06x10'
1.10x10'
9.5xl06
8.8xl06
S.OxlO6
Filtrate (PFU/mL)
Count 1
<1
82
l.llxlO2
1.06xl02
<1
e.ixio2
S.OxlO2
4.6xl02
<1
5.6xl02
5.4xl02
5.3xl02
<1
1.34xl03
5.8xl02
2.8xl02
<1
8.8xl02
4.6xl02
2.9xl02
<1
5.6xl03
9.8xl03
l.SOxlO4
<1
2.00x1 02
6.8xl02
1.39xl02
58
1.35xl03
3.7xl03
6.8xl03
51
3.2xl03
6.4xl03
1.46xl04
Count 2
<1
81
92
93
<1
5.5xl02
6.6xl02
4.5xl02
<1
4.1xl02
5.6xl02
4.7xl02
<1
1.22xl03
4.8xl02
3.3xl02
<1
S.OxlO2
S.lxlO2
3.6xl02
<1
5.3xl03
l.llxlO4
1.17xl04
<1
2.05x1 02
6.2xl02
1.24xl02
34
1.66xl03
2.9xl03
4.1xl03
42
2. 39x1 03
6.5xl03
1.52xl04
Count 3
1
77
79
1.42xl02
<1
5.8xl02
7.6xl02
4.1xl02
<1
4.7xl02
6.2xl02
4.3xl02
<1
S.lxlO2
4.5xl02
S.OxlO2
<1
6.5xl02
4.2xl02
3.3xl02
<1
4.8xl03
1.24xl04
1.40xl04
<1
2.71x1 02
7.8xl02
1.42xl02
43
1.56xl03
3.7xl03
5.5xl03
59
3.3xl03
6.5xl03
1.27xl04
C-l
-------
Table C-2. B. atrophaeus Triplicate Count Data
Module
Module 1
Module 2
Module 3
Module 4
Module 5
Module 6
Module 2
Retest
Module 4
Retest
Sample
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Feed (CFU/mL)
Count 1
1.04x10"
9.9xl04
1.23x10"
.18x10"
.15xl05
.26xl05
.06x10"
.07x10"
.29x10"
1.20x10"
9.9xl04
6.3xl04
.21x10"
.30x10"
.46x10"
.36x10"
.64x10"
.33x10"
8.5xlOj
8.9xlOj
1.06xl04
1.26xl04
LlSxlO4
1.26xl04
Count 2
1.01x10"
1.14x10"
1.00x10"
1.53x10"
1.16x10"
1.14x10"
1.15x10"
1.06x10"
1.40x10"
1.23x10"
1.03x10"
V.lxlO4
1.23x10"
1.28x10"
1.45x10"
1.53x10"
1.58x10"
1.59x10"
9.7xlOj
1.05xl04
9.8xlOj
1.32xl04
1.21xl04
1.23xl04
Count 3
1.39x10"
1.20x10"
1.30x10"
1.40x10"
1.28x10"
1.25x10"
1.00x10"
9.8xl04
1.36x10"
1.07x10"
1.05x10"
8.6xl04
1.29x10"
1.26x10"
1.40x10"
1.50x10"
1.66x10"
1.38x10"
1.02xl04
9.2x10"
LlOxlO4
1.29xl04
1.16xl04
1.37xl04
Filtrate (CFU/lOOmL)
Count 1
<1
<1
2
<1
2
1
1
1
<1
1
<1
3
2
<1
11
<1
<1
<1
<1
<1
1
1
2
<1
<1
1
<1
1
<1
2
2
1
Count 2
<1
<1
1
<1
4
1
2
<1
<1
<1
<1
1
1
<1
<1
<1
<1
<1
<1
<1
1
<1
<1
<1
<1
<1
<1
1
<1
1
<1
<1
Count 3
<1
<1
<1
<1
2
1
<1
<1
<1
1
<1
1
1
<1
<1
<1
<1
<1
<1
<1
2
<1
<1
<1
<1
<1
<1
<1
1
2
2
1
Table C-3. C. parvum Triplicate Count Data
Module
Module 3
Module 2
Sample
Flush
2 Minutes
1 5 Minutes
30 Minutes
Flush
2 Minutes
1 5 Minutes
30 Minutes
Feed (oocysts/L)
Count 1
2.63xl06
1.39xl06
1.74xl06
1.49xl06
8.4x10"
2.02xl06
Count 2
1.38xl06
6.9xl06
6.1x10"
2.14xl06
2.05xl06
1.78xl06
Count 3
1.46xl06
1.38xl06
7.8x10"
2.58xl06
2.06xl06
1.97xl06
Filtrate (oocysts/L)
Count 1
<1
<1
<1
1
<1
<1
<1
<1
Count 2
<1
<1
<1
<1
<1
<1
<1
<1
Count 3
<1
<1
<1
<1
<1
<1
<1
<1
C-2
------- |