September 2000
NSF 00/13/EPADW395
Environmental Technology
Verification Report
Physical Removal of Microbiological
and Particulate Contaminants in
Drinking Water
Ionics
UF-1-7T Ultrafiltration Membrane
System
Escondido, California
Prepared by
NSF International
Under a Cooperative Agreement with
AEPA U.S. Environmental Protection Agency
etVElVElV
-------
THE ENVIRONMENTAL TECHNOLOGY VERIFICATION
PROGRAM
«EF% ETIr
U.S. Environmental Protection Agency NSF International
ETV Joint Verification Statement
TECHNOLOGY TYPE:
MEMBRANE FILTRATION USED IN PACKAGED DRINKING
WATER TREATMENT SYSTEMS
APPLICATION:
PHYSICAL REMOVAL OF MICROBIOLOGICAL &
PARTICULATE CONTAMINANTS IN DRINKING WATER
IN ESCONDIDO, CALIFORNIA
TECHNOLOGY NAME:
UF-1-7T ULTRAFILTRATION MEMBRANE SYSTEM
COMPANY:
IONICS
ADDRESS:
65 GROVE STREET PHONE: (617)926-2510
WATERTOWN, MA 02472-2882 FAX: (617)926-4304
WEB SITE:
EMAIL:
http://www.ionics.com
avongottberg@ionics.com
The U.S. Environmental Protection Agency (EPA) has created the Environmental Technology Verification
(ETV) Program to facilitate the deployment of innovative or improved environmental technologies through
performance verification and dissemination of information. The goal of the ETV program is to further
environmental protection by substantially 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; stakeholders groups which
consist 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 International (NSF) in cooperation with the EPA operates the Drinking Water Treatment Systems
(DWTS) Pilot, one of 12 technology areas under ETV. The DWTS Pilot recently evaluated the
performance of an ultrafiltration membrane system used in package drinking water treatment system
applications. This verification statement provides a summary of the test results for the Ionics UF-1-7T
Ultrafiltration (UF) Membrane System. Montgomery Watson, an NSF-qualified field testing organization
(FTO), performed the verification testing.
00/13/EPADW395 The accompanying notice is an integral part of this verification statement. September 2000
VS-i
-------
ABSTRACT
Verification testing of the Ionics UF-1-7T Ultrafiltration membrane system was conducted over two test
periods at the Aqua 2000 Research Center in Escondido, California. The first test period, from December
7, 1999 to January 11, 2000 represented winter conditions. The second test period, from March 6, 2000 to
April 6, 2000 represented spring conditions. The source water was a blend of Colorado River and State
Project Water. Verification testing was conducted at manufacturer specified operating conditions. The
membrane unit was operated in dead-end mode at a constant flux of 33 gfd (57 L/hr-m2) with feedwater
recovery of 92-93 percent. Test Period 1 consisted of one filtration run. The membrane was completely
fouled at the end of Test Period 1. Between test periods, modifications were made to the backwash
protocol. As a result, the system completed Test Period 2 without appreciable loss of specific flux. The
system experienced one incident of fiber breakage during Test Period 1 and three incidents of fiber
breakage in Test Period 2. The manufacturer recommended cleaning procedure was effective in recovering
membrane productivity. The membrane system achieved significant removal of particulate contaminants
and bacteria and seeded MS2 bacteriophage as described later.
TECHNOLOGY DESCRIPTION
The Ionics UF-1-7T unit is comprised of seven hollow fiber UF membrane modules inside an aluminum
pressure vessel and mounted on a transportable skid. The skid is constructed of steel, and can be shipped
by truck. The Ionics UF unit is completely self-contained, including all the components required for
operation. The only connections are a raw water connection to the feed pump, drain lines for filtrate tank
overflow and backwash waste, and electrical power. The unit requires approximately 35 ft2 (3.2 m2) of
floor space.
The UF-1-7T unit has an Allen Bradley programmable logic controller (PLC). The PLC controls the
opening and closing of pneumatic valves and the operation of pumps required for filtration and backwash.
The backwash frequency and the length of time the system spends in each backwash phase are set by
entering values into the appropriate screen on the PLC. The PLC maintains a constant filtrate flow during
filtration by automatically adjusting feed pump speed. The Ionics UF unit has digital flow, pressure and
temperature measurement and a data logger to acquire operating information digitally.
The Ionics UF-1-7T unit has two alternating operating modes. These are filtration and backwash. During
filtration, raw water is driven under pressure through pores in the UF membrane. Treated water is
collected from the filtrate side (inside) of the membrane. At the end of the filtration cycle, the system
initiates a backwash. During backwash, the feed pump shuts down, valves are repositioned, and the
backwash pump starts. The backwash pump draws treated water from the filtrate storage tank, chlorinates
it, and forces the water under pressure in the reverse direction through the fibers. This reverse flow
removes solids and organics, which have accumulated on the membrane surface. Chlorine is added to the
backwash water to assist in oxidizing organics that have accumulated on the membrane surface. Air is also
added during backwashing to scour the membrane for more effective cleaning. The long-term operation of
the Ionics UF unit frequently results in the accumulation of materials on the membrane surface which are
not effectively removed by backwash. This is called membrane fouling and is quantified by a gradual
increase in the pressure required to maintain the desired flux. Once a critical upper pressure has been
reached, normal operation is discontinued and the membrane undergoes chemical cleaning. Chemical
cleaning involves the use of a citric acid solution, followed by a high pH solution and pH 2 backwash to
restore membrane productivity.
The pressure vessel of the Ionics UF unit contains seven Toray Model TP-TE07-S membrane modules.
These 3.5 inch (8.8 cm) diameter modules each contain approximately 3,600 fibers. The Toray module is a
00/13/EPADW395 The accompanying notice is an integral part of this verification statement. September 2000
VS-ii
-------
hollow fiber configuration, manufactured from polyacrylonitrile, with nominal molecular weight cut-off of
100,000 Daltons. This corresponds with a pore diameter of approximately 0.01 micron. At this pore size,
the membrane is expected to remove particulates, including protozoa, bacteria and virus.
VERIFICATION TESTING DESCRIPTION
Test Site
The verification test site was the City of San Diego's Aqua 2000 Research Center at 14103 Highland
Valley Road in Escondido, California. The Research Center includes office and lab trailers, a covered
concrete test pad and a dedicated operations staff with substantial membrane experience. The source water
for testing was Lake Skinner water via the San Diego Aqueduct. Lake Skinner water consists of Colorado
River water and State Project water, which are two of the major raw drinking water supplies in Southern
California.
Methods and Procedures
Turbidity, pH, chlorine and temperature analyses were conducted daily at the test site according to
Standard Methods for the Examination of Water and Wastewater, 19th Ed. (APHA, et. al., 1995).
Standard Methods, 19th Ed. (APHA, 1995) and Methods for Chemical Analysis of Water and Wastes
(EPA, 1979) were used for analyses conducted at The City of San Diego Laboratory. These included
alkalinity, total and calcium hardness, total dissolved solids (TDS), total suspended solids (TSS), total
organic carbon (TOC), ultraviolet absorbance at 254 nanometers (UV254), total coliform and heterotrophic
plate count (HPC). Total and calcium hardness analyses were conducted every other week. All other
analyses were conducted weekly. MS2 bacteriophage analysis was conducted by EPA Information
Collection Rule (ICR) Method for Coliphage Analysis (Sobsey, et al. 1990). Online Hach 1900 WPC
particle counters and 1720D turbidimeters continuously monitored these parameters in both the raw water
and membrane system filtrate. The particle counters were set up to enumerate particle counts in the
following size ranges: 2-3 um, 3-5 um, 5-7 um, 7-10 um, 10-15 um and > 15 um. Data from the online
particle counters and turbidimeters were stored at one-minute intervals on a computer.
VERIFICATION OF PERFORMANCE
System Operation
Verification testing was conducted at manufacturer specified operating conditions. The membrane unit was
operated at a constant flux of 33 gfd (57 L/hr-m2) with feedwater recovery of 92 percent. Filtrate flow rate
was set by entering the target flow in a screen on the PLC. Backwash frequency was every 60 minutes.
Backwash volume averaged 55 gallons (208 liters) for Test Period 1 and 75 gallons (283 liters) for Test
Period 2. Backwash chlorine concentration was in the range 5 to 10 mg/L. The reverse flow backwash
volume was increased from 15 gallon (57 liters) in Test Period 1 to 30 gallon (113 liters) in Test Period 2.
The system was operated during Test Period 1 with moderate fouling until it reached the maximum
recommended operating pressure towards the end of the testing period. During this period specific flux
decreased from 6.0 to 1.2 gfd/psi at 20°C (148 to 39 L/hr-m2 at 20°C). The system, however, ran all of
Test Period 2 without appreciable fouling. During Test Period 2 the specific flux decreased from 5.9
gfd/psi at 20°C (145 L/hr-m2 at 20°C) over three days before stabilizing at 4.0 gfd/psi at 20°C (98 L/hr-m2
at 20°C) for remainder of testing.
Membrane cleaning was performed according to manufacturer recommended procedure. A citric acid
solution followed by a high pH cleaning solution was prepared in the feed storage tank and recirculated
through the feed side of the membrane at approximately 330 gpm (1250 L/min) for 60 minutes. A pH 2
acid rinse was used after the high-pH cleaning step to remove potential precipiates. Flux-pressure profiles
00/13/EPADW395 The accompanying notice is an integral part of this verification statement. September 2000
VS-iii
-------
were performed after each cleaning step to evaluate recovery of specific flux. The manufacturer
recommended cleaning procedure was effective in recovering specific flux. Loss of original flux was 4.8
percent after the cleaning at the end of Test Period 1 and 17 percent after the cleaning at the end of Test
Period 2.
One incident of broken fibers occurred during Test Period 1 and three incidents of broken fibers occurred
during Test Period 2. Air pressure-hold tests were conducted near the beginning and end of each test
period as well as before and after fiber repairs to assess membrane integrity. Air pressure-hold tests were
conducted by selecting the integrity test from the appropriate PLC screen. During the air pressure-hold test
the pressure vessel is first drained, then the feed side of the membrane is pressurized with air and the
filtrate side of the membrane is opened to atmosphere. Once pressurized, the loss of held pressure on the
feed side was monitored over 10 minutes. A loss of > 1 psi every five minutes of held pressure typically
would indicate the membranes were not intact. The air pressure-hold test was inconsistent in identifying
fiber breaks based on this performance criterion. The pressure decay before repair was less than 2 psi for
two fiber breakages and just over 2 psi for the other two fiber breakage incidents. Particle counting was a
reliable indicator of broken fibers, and all incidents of broken fibers were identified by visual observation of
filtrate particle counts. Typically, one or two broken fibers produced a increase in permate particle counts
(> 2 um) of from one-half to one log.
Source Water
The source water for the ETV testing consisted of a blend of Colorado River water and State Project water
delivered to the test site via the San Diego Aqueduct. The source water had the following average water
quality during the two test periods: TDS 500/470 mg/L, total hardness 240/220 mg/L as CaC03, alkalinity
120/120 mg/L as CaC03, TOC 3.2/3.6 mg/L, pH 8.3/8.2, temperature 15/19 °C and turbidity 1.2/1.4
NTU.
Particle Removal
Total suspended solids in the filtrate were removed to below the detection limit for the analysis (1 mg/L),
for all samples analyzed. Filtrate turbidity was 0.05 NTU or less 95 percent of the time. The test system
removed greater than 3 logs of both Cryptospordium-sized (3-5 um) particles and Giardia-sized (5-15 um)
particles, 95 percent of the time. Filtrate levels of particles in these size ranges were elevated and particle
removal was decreased during periods of operation with compromised fibers that occurred during Test
Period 2. Four hour average raw water and filtrate particle levels and daily average particle removal in
these size ranges for Test Periods 1 and 2 are presented in the following table:
Ionics UF-1-7T UF System Particle Counts and Particle Removals for Test Periods 1/2
3-5 um Particles
5-15 um Particles
Raw Water
Filtrate
Log
Raw Water
Filtrate
Log
(#/mL)
(#/mL)
Removal
(#/mL)
(#/mL)
Removal
Average
1700/1400
0.19/0.63
4.2/3.4
900/680
0.16/0.37
3.9/3.3
Standard Deviation
230/310
0.35/0.46
0.40/0.30
170/200
0.26/0.24
0.40/0.24
95% Confidence Interval
1700-1700/
0.14-0.24/
4.1-4.3/
880-920/
0.12-0.20/
3.8-4.0/
1400-1400
0.56-0.70
3.3-3.5
650-710
0.34-0.40
3.2-3.4
Minimum
1200/690
0.04/0.15
3.0/2.9
530/270
0.04/0.11
3.0/2.8
Maximum
2300/2400
1.7/3.4
4.6/3.9
1400/1500
1.9/2.3
4.4/3.7
Microbial Removal
Total Coliforms and HPC were analyzed on a weekly basis during both ETV test periods. Raw water total
coliforms averaged 25 and 8 MPN/lOOmL during Test Periods 1 and 2, respectively. No total coliforms
00/13/EPADW395 The accompanying notice is an integral part of this verification statement. September 2000
VS-iv
-------
were detected in the filtrate. HPC averaged 83 and 310 cfu/mL in the raw water for Test Periods 1 and 2
while filtrate levels of HPC averaged 100 and 200 cfu/mL, respectively. The relatively high levels of HPC
in the filtrate are possibly due to contamination of the filtrate side with HPC during periods of operation
with compromised fibers. Challenge experiments with MS2 bacteriophage were conducted at the end of
Test Period 1 and beginning of Test Period 2, immediately after membrane cleaning (worst case for virus
removal). Virus were continuously added to the membrane feed water. The membrane was allowed to
operate for 1 filtration cycle to come to equilibrium and then paired samples were taken from the feed and
filtrate within 1-minute of completion of backwash, at the middle and at the end of the filtration cycle, over
the next two filtration cycles. Specific flux during the seeding conducted at the end of Test Period 1 was
4.9 gfd/psi (119 L/hr-m2-bar), while specific flux for the seeding conducted at the beginning of Test Period
2 was 6.2 gfd/psi (152 L/hr-m2-bar). Feed virus concentration ranged from 7.4 x 106 to 2.8 x 107 plaque
forming units (pfii)/100mL for the first virus seeding and from 3.5 x 107 to 6.0 x 107 pfii/lOOmL for the
second virus seeding. Log removal of virus ranged from 4.0 to 5.7 for Test Period 1 and from 2.9 to 4.3
for Test Period 2.
Operation and Maintenance Results
Operation was initiated by entering target filtrate flow rate, backwash frequency and time of each
backwash phase in the appropriate PLC screen. Backwash flow rate was adjusted manually using a valve.
As the membrane system fouled, the feed pump speed was automatically readjusted to maintain a constant
filtrate flow rate. The sodium hypochlorite dosing pump required initial manual adjustment to achieve a
target chlorine dose in the backwash water of 5 to 10 mg/L. Chlorine concentration in the backwash
feedwater was checked twice daily.
Operation of the membrane unit consumed 0.12 gal (0.46 L) of 10% sodium hypochlorite per day to
chlorinate backwash water. No other chemicals were consumed during routine operation of the system.
During a typical chemical cleaning, 17.0 pounds (7.7 kg) of citric acid, 1.8 gallon (7.0 liter) of high pH
cleaning solution and 200 milliliters of muriatic acid (40% hydrochloric acid) were consumed. The
manufacturer supplied an Operations and Maintenance manual that was extremely helpful in explaining the
setup, operation and maintenance of the ETV test system.
Original Signed by
E. Timothy Oppelt
10/10/00
E. Timothy Oppelt Date
Director
National Risk Management Research Laboratory
Office of Research and Development
United States Environmental Protection Agency
Original Signed by
Tom Bruursema
10/17/00
Tom Bruursema Date
General Manager
Environmental and Research Services
NSF International
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 a NSF Certification of the specific product mentioned herein.
00/13/EPADW395 The accompanying notice is an integral part of this verification statement. September 2000
VS-v
-------
Availability of Supporting Documents
Copies of the ETV Protocol for Equipment Verification Testing for Physical Removal of
Microbiological and Particulate Contaminants, dated April 20, 1998 and revised May
14, 1999, the Verification Statement, and the Verification Report (NSF Report
#00/13/EPADW395) are available from the following sources:
(NOTE: Appendices are not included in the Verification Report. Appendices are available
from NSF upon request.)
Drinking Water Systems ETV Manager (order hard copy)
NSF International
P.O. Box 130140
Ann Arbor, Michigan 48113 -0140
NSF web site: http://www.nsf.org/etv (electronic copy)
EPA web site: http://www.epa.gov/etv (electronic copy)
00/13/EPADW395 The accompanying notice is an integral part of this verification statement. September 2000
VS-vi
-------
September 2000
Environmental Technology Verification Report
Physical Removal of Microbiological and Particulate Contaminants
in Drinking Water
Ionics UF-1-7T Ultrafiltration Membrane System
Escondido, California
Prepared for:
NSF International
Ann Arbor, MI 48105
Prepared by:
Samer Adham, Ph.D.
&
Karl Gramith
Montgomery Watson
Pasadena, CA 91101
Under a cooperative agreement with the U.S. Environmental Protection Agency
Jeffrey Q. Adams, Project Officer
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
Notice
The U.S. Environmental Protection Agency (EPA) through its Office of Research and
Development has financially supported and collaborated with NSF International (NSF) under
Cooperative Agreement No. CR 824815. This verification effort was supported by Drinking
Water Treatment Systems Pilot operating under the Environmental Technology Verification
(ETV) Program. This document has been peer reviewed and reviewed by NSF and EPA and
recommended for public release.
11
-------
Foreword
The following is the final report on an Environmental Technology Verification (ETV) test
performed for the NSF International (NSF) and the United States Environmental Protection
Agency (EPA) by Montgomery Watson, in cooperation with Ionics. The test was conducted in
1999 and 2000 at the Aqua 2000 Research Center in San Diego, California.
Throughout its history, the EPA has evaluated the effectiveness of innovative technologies to
protect human health and the environment. The ETV Program has been instituted to verify the
performance of innovative technical solutions to environmental pollution or human health threats.
ETV was created to substantially accelerate the entrance of new environmental technologies into
the domestic and international marketplace. Verifiable, high quality data on the performance of
new technologies are made available to regulators, developers, consulting engineers, and those in
the public health and environmental protection industries. This encourages more rapid availability
of approaches to better protect the environment.
The EPA has partnered with NSF, an independent, not-for-profit testing and certification
organization dedicated to public health, safety and protection of the environment, to verify
performance of small drinking water systems that serve small communities under the ETV
Drinking Water Treatment Systems (DWTS) Pilot Project. A goal of verification testing is to
enhance and facilitate the acceptance of small drinking water treatment equipment by state
drinking water regulatory officials and consulting engineers while reducing the need for testing of
equipment at each location where the equipment's use is contemplated. NSF will meet this goal
by working with manufacturers and NSF-qualified Field Testing Organizations (FTO) to conduct
verification testing under the approved protocols.
NSF is conducting the ETV DWTS with participation of manufacturers, under the sponsorship of
the EPA Office of Research and Development, National Risk Management Research Laboratory,
Water Supply and Water Resources Division, Cincinnati, Ohio. It is important to note that
verification of the equipment does not mean that the equipment is "certified" by NSF or
"accepted" by EPA. Rather, it recognizes that the performance of the equipment has been
determined and verified by these organizations for those conditions tested by the FTO.
in
-------
Table of Contents
Section Page
Verification Statement VS-i
Title Page i
Notice ii
Foreword iii
Table of Contents iv
List of Tables vi
List of Figures vii
Abbreviations and Acronyms viii
Acknowledgements ix
Chapter 1 - Introduction 1
1.1 Environmental Technology Verification (ETV) Purpose and Program Operation 1
1.2 Project Participants 1
1.3 Definition of Roles and Responsibilities of Project Participants 1
1.3.1 Field Testing Organization Responsibilities 1
1.3.2 Manufacturer Responsibilities 2
1.3.3 Operator and Test Site Staff Responsibilities 2
1.3.4 Water Quality Analyst Responsibilities 2
1.3.5 NSF Responsibilities 3
1.3.6 EPA Responsibilities 3
Chapter 2 - Equipment Description and Operating Processes 4
2.1 Description of the Treatment Train and Unit Processes 5
2.2 Description of Physical Construction/Components of the Equipment 7
Chapter 3 - Materials and Methods 8
3.1 Testing Site Name and Location 8
3.1.1 Site Background Information 8
3.1.2 Test Site Description 8
3.2 Source/Feed Water Quality 9
3.3 Environmental Technology Verification Testing Plan 10
3.3.1 Task 1: Characterization of Membrane Flux and Recovery 10
3.3.2 Task 2: Evaluation of Cleaning Efficiency 10
3.3.3 Task 3: Evaluation of Finished Water Quality 12
3.3.4 Task 4: Reporting of Membrane Pore Size 12
3.3.5 Task 5: Membrane Integrity Testing 12
iv
-------
Table of Contents, continued
Section Page
3.3.6 Task 6: Data Management 13
3.3.7 Task 7: Quality Assurance/Quality Control 13
3.3.8 Task 8: Microbial Removal (Optional) 17
3.4 Calculation of Membrane Operating Parameters 18
3.4.1 Filtrate Flux 18
3.4.2 Specific Flux 18
3.4.3 Transmembrane Pressure 18
3.4.4 Temperature Adjustment for Flux Calculation 18
3.4.5 F eedwater Sy stem Recovery 19
3.4.6 Rejection 19
3.5 Calculation of Data Quality Indicators 19
3.5.1 Precision 19
3.5.2 Relative Percent Deviation 20
3.5.3 Accuracy 20
3.5.4 Stati stical Uncertainty 20
3.6 Testing Schedule 21
Chapter 4 - Results and Discussion 22
4.1 Task 1: Characterization of Membrane Flux and Recovery 22
4.2 Task 2: Evaluation of Cleaning Efficiency 23
4.3 Task 3: Evaluation of Finished Water Quality 23
4.3.1 Turbidity, Particle Concentration and Particle Removal 23
4.3.2 Indigenous Bacteria Removal 25
4.3.3 Other Water Quality Parameters 25
4.4 Task 4: Reporting Membrane Pore Size 26
4.5 Task 5: Membrane Integrity Testing 26
4.6 Task 6: Data Management 26
4.6.1 Data Recording 26
4.6.2 Data Entry, Validation, and Reduction 27
4.7 Task 7: Quality Assurance/Quality Control (QA/QC) 27
4.7.1 Data Correctness 27
4.7.2 Stati stical Uncertainty 27
4.7.3 Completeness 27
v
-------
Table of Contents, continued
Section Page
4.7.4 Accuracy 28
4.7.5 Precision and Relative Percent Deviation 28
4.8 Task 8: Microbial Removal 28
4.9 Additional ETV Project Requirements 28
4.9.1 Operation and Maintenance (O&M) Manual 28
4.9.2 System Efficiency and Chemical Consumption 29
4.9.3 Equipment Deficiencies Experienced During the ETV Project 29
Chapter 5 - References 31
Tables
Table 2-1. Characteristics of the Ionics UF 1-7T ultrafiltration membrane 33
Table 3-1. Water quality analytical methods 34
Table 4-1. Ionics UF membrane system operating conditions 35
Table 4-2. Evaluation of cleaning efficiency for Ionics UF membrane 36
Table 4-3. Onsite lab water quality analyses for Ionics UF membrane system 36
Table 4-4. Summary of online particle and turbidity data for Ionics UF membrane system 37
Table 4-5. Summary of microbial water quality analyses for the Ionics UF membrane system.... 39
Table 4-6. Summary of general water quality analyses for the Ionics UF membrane system 40
Table 4-7. Comparison of calculated and measured total suspended solids for Ionics UF
membrane system 41
Table 4-8. Feed and filtrate concentrations of MS2 virus for the Ionics UF membrane system... 42
Table 4-9. Review of manufacturer's operations and maintenance manual for the Ionics UF
membrane system 43
Table 4-10. Efficiency of the Ionics UF membrane system 45
Table 4-11. Chemical consumption for the Ionics UF membrane system 45
vi
-------
Table of Contents, continued
Figures
Figure 1-1. Organizational chart showing lines of communication 46
Figure 2-1. Photographs of ETV test unit 46
Figure 2-2. Spatial requirements for the Ionics UF unit 47
Figure 2-3. Schematic diagram of the Ionics UF membrane process 48
Figure 3-1. Schematic of Aqua 2000 Research Center test site 48
Figure 3-2. Lake Skinner raw water quality 49
Figure 3-3. Lake Skinner raw water quality 50
Figure 3-4. Response of online particle counters to Duke Monosphere solution 51
Figure 3-5. Membrane verification testing schedule 52
Figure 4-1. Transmembrane pressure and temperature profiles for the Ionics UF membrane
system 53
Figure 4-2. Operational flux and specific flux profiles for the Ionics UF membrane system 54
Figure 4-3. Clean water flux profile during membrane chemical cleanings - Test Period 1 55
Figure 4-4. Clean water flux profile during membrane chemical cleanings - Test Period 2 56
Figure 4-5. Turbidity profile for raw water and Ionics UF membrane system - Test Period 1.... 57
Figure 4-6. Turbidity profile for raw water and Ionics UF membrane system - Test Period 2.... 57
Figure 4-7. Particle counts for raw water and Ionics filtrate - Test Period 1 58
Figure 4-8. Particle counts for raw water and Ionics filtrate - Test Period 2 60
Figure 4-9. Particle removal for Ionics UF membrane system - Test Period 1 62
Figure 4-10. Particle removal for Ionics UF membrane system - Test Period 2 64
Figure 4-11. Probability plots of filtrate turbidity and log removal of particles for the Ionics UF
membrane system 66
Figure 4-12. Air pressure hold test results for the Ionics UF membrane system 67
Figure 4-13. Log removal of seeded MS2 virus by Ionics UF membrane system 70
Appendices
Appendix A - Additional Documents and Data Analyses
Appendix B - Raw Data Sheets
Appendix C - Hardcopy Electronic Data
vii
-------
Abbreviations and Acronyms
c
Celsius degrees
cfu
Colony forming unit(s)
CIP
Clean in place
Cf
Feed concentration
Cp
Filtrate concentration
cm
Centimeter
CRW
Colorado River water
d
Day(s)
DI
Deionized
DOC
Dissolved organic carbon
DWTS
Drinking Water
Treatment System
EPA
Environmental Protection
Agency
ETV
Environmental Technology
Verification
FOD
Field Operations Document
ft2
Square foot (feet)
FTO
Field Testing Organization
gfd
Gallon(s) per day per square
foot of membrane area
gpd
gallon per day
gpm
Gallon(s) per minute
HPC
Heterotrophic plate count
hr
Hour(s)
ICR
Information Collection Rule
in Hg
Inch(es) of Mercury
JS;
Initial specific
transmembrane flux
JSf
Final specific
transmembrane flux
Js
Specific flux
JSi0
Initial specific
transmembrane flux at t=0 of
membrane operation
Jt
Filtrate flux
Jtm
Transmembrane flux
kg
Kilogram(s)
L
Liter(s)
m2
Square meter(s)
m3/d
Cubic meter(s) per day
mgd
Million gallons per day
mg/L
Milligram(s) per liter
min
Minute(s)
mL
Milliliter(s)
MPN
Most probable number
NIST
National Institute of Standards
and Technology
NSF
NSF International (formerly
known as the National
Sanitation Foundation)
NTU
Nephelometric turbidity unit(s)
O&M
Operations and Maintenance
Pi
Pressure at inlet of
membrane module
Po
Pressure at outlet of
membrane module
PP
Filtrate pressure
Ptm
Transmembrane pressure
PC
Personal computer
PLC
Programmable Logic
Controller
ppm
Parts per million
psi
Pound(s) per square inch
PVC
Polyvinyl chloride
Qf
Feed flow
QP
Filtrate flow
Qr
Recycle flow
QA
Quality assurance
QC
Quality control
s
Membrane surface area
scfm
Standard cubic feet per minute
slpm
Standard liter per minute
sec
Second(s)
SPW
State Project water
T
Temperature
TC
Total coliform bacteria
TOC
Total organic carbon
TDS
Total dissolved solids
TSS
Total suspended solids
um
Micron(s)
UF
Ultrafiltration
UV254
Ultraviolet light absorbance
at 254 nanometer
Vlll
-------
Acknowledgements
The authors would like to thank the EPA, for sponsoring the ETV program and providing partial
funding for the study. In particular, the authors would like to thank Jeffrey Q. Adams, project
officer with the EPA, for his continuous support throughout the project.
The authors would also like to thank NSF, for administrating the ETV pilot project. The time and
continuous guidance provided by the following NSF personnel is gratefully acknowledged: Bruce
Bartley, Carol Becker, and Kristie Wilhelm.
The time and outstanding efforts provided by the manager of Aqua 2000 Research Center, Paul
Gagliardo with the City of San Diego is gratefully acknowledged. The authors would also like to
thank Jeff Williams from the Aqua 2000 Center operation team for his assistance in operating the
membrane system. The authors would also like to thank Dana Chapin from the City of San Diego
Water Laboratory for facilitating most of the water quality analyses in the study. In addition, the
authors would like to thank Yildiz Chambers from the City of San Diego Marine Microbiology
Laboratory for co-ordinating the microbial analyses in the study.
The author would also like to acknowledge the manufacturer of the equipment employed during
the ETV project (Ionics, Watertown, MA) for their continuous assistance throughout the ETV
test operation periods and for providing partial funding to the project. In particular, the authors
would like to thank Antonia von Gottberg from Ionics and Mark Thompson from Advanced
Membrane Systems for their continuous support.
The authors gratefully acknowledge the contributions of the following co-workers from
Montgomery Watson: Anthony Huang, Rion Merlo, Lina Boulos, Natalie Flores, and Rene
Lucero.
IX
-------
Chapter 1
Introduction
1.1 Environmental Technology Verification (ETV) Purpose and Program Operation
The U.S. Environmental Protection Agency (EPA) has 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 substantially 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; stakeholders
groups which consist 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 (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 International (NSF) in cooperation with the EPA operates the Drinking Water Treatment
Systems (DWTS) Pilot, one of 12 technology areas under ETV. This DWTS Pilot evaluated
the performance the Ionics UF-1-7T ultrafiltration (UF) system used in drinking water
treatment system applications.
This report provides the ETV results for the Ionics UF-1-7T membrane system.
1.2 Project Participants
Figure 1-1 is an organization chart showing the project participants and the lines of
communication established for the ETV. The Field Testing Organization (FTO) was Montgomery
Watson, a NSF-qualified FTO, which provided the overall management of the ETV through the
project manager and project engineer. The ultrafiltration membrane manufacturer for the ETV
was Ionics. The operations management and staff were from the test site at the City of San Diego
Metropolitan Wastewater Department, Aqua 2000 Research Center in Escondido, California.
The City of San Diego laboratory, a State-certified laboratory, provided water quality analyses.
Data management and final report preparation were performed by the FTO, Montgomery Watson.
1.3 Definition of Roles and Responsibilities of Project Participants
1.3.1 Field Testing Organization Responsibilities
The specific responsibilities of the FTO, Montgomery Watson, were to:
1
-------
• Provide the overall management of the ETV through the project manager and the project
engineers.
• Provide all needed logistical support, the project communication network, and all scheduling
and coordination of the activities of all participants.
• Manage, evaluate, interpret and report on data generated in the ETV.
• Evaluate the performance of the ultrafiltration membrane technology according to the Field
Operating Document (FOD) and the testing, operations, quality assurance/quality control
(QA/QC), data management and safety protocols contained therein.
• Provide all quality control (QC) information in the ETV report.
• Provide all data generated during the ETV in hard copy and electronic form in a common
spreadsheet or database format.
1.3.2 Manufacturer Responsibilities
The specific responsibilities of the ultrafiltration membrane manufacturer, Ionics, were to:
• Provide complete, field-ready equipment for the ETV at the testing site.
• Provide logistical and technical support as required throughout the ETV.
• Provide partial funding for the project.
• Attend project meetings as necessary.
1.3.3 Operator and Test Site Staff Responsibilities
The specific responsibilities of the operations and test site staff from the City of San Diego
Metropolitan Wastewater Department were to:
• Provide set-up, shake-down, operations, maintenance and on-site analytical services according
to the FOD and the testing, operations, QA/QC, data management and safety protocols.
• Provide the necessary and appropriate space for the equipment to be tested in the ETV.
• Provide all necessary electrical power, feedwater and other utilities as required for the ETV.
• Provide all necessary drains to the test site.
1.3.4 Water Quality Analyst Responsibilities
The specific responsibilities of the water quality analytical staff from the City of San Diego
Laboratory were to:
• Provide all off-site water quality analyses prescribed in the FOD according to the QA/QC
protocols contained therein.
• Provide reports with the analytical results to the data manager.
• Provide detailed information on the analytical procedures implemented.
2
-------
1.3.5 NSF Responsibilities
NSF was responsible for administration of the testing program. Specific responsibilities of the
NSF were to:
• Develop test protocols and qualify FTOs.
• Review and approve FODs.
• Conduct inspections and make recommendations based on inspections.
• Conduct financial administration of the project.
• Review all project reports and deliverables.
1.3.6 EPA Responsibilities
The specific responsibilities of EPA were to:
• Initiate the ETV program.
• Provide significant project funding.
• Review final reports.
3
-------
Chapter 2
Equipment Description and Operating Processes
The equipment tested in the ETV is Ionics UF-1-7T package ultrafiltration membrane system.
The test unit is comprised of seven membrane modules in an aluminum pressure vessel mounted
on a transportable skid. The 3.5 inch (8.8 cm) diameter ultrafiltration modules are model TP-
TE07-S, manufactured by Toray in Japan. The skid is constructed of steel, and can be shipped by
truck. A photograph of the ETV test unit is shown in Figure 2-1. The figure shows the front of
the UF-1-7T system in the photo on the left, and the back of the system (including the aluminum
membrane pressure vessel) on the right. The skid includes all major equipment elements and
controls and requires approximately 35 square feet (ft2) or 3.2 square meters (m2) of floor space.
The Ionics UF unit is shipped with an oil-free air compressor that is used for air scour during
backwash as well as for operating pneumatic valves. The spatial requirements and locations of
major components and instruments of the Ionics UF unit are shown in Figure 2-2.
The Ionics UF unit is completely self-contained, including all the components required for
operation. The only connections are a raw water connection to the feed pump, drain lines for
filtrate tank overflow and backwash waste, and electrical power.
The Ionics UF unit includes an Allen Bradley programmable logic controller (PLC) with
PanelView display. Operating parameters such as filtrate flow rate, backwash frequency and time
spent in each backwash phase are set using the PLC. The PLC automatically controls feed pump
speed to maintain a constant filtrate flow and controls pumps and valves during backwash.
The Ionics UF unit has two alternating operating modes: filtration and backwash. During
filtration, feed water is driven under pressure from the feed side of the hollow fibers (outside of
fibers), through pores in the UF membrane. Filtrate is collected from the inside of the fibers. The
filtration cycle typically lasts from 15 to 30 minutes. At the end of the filtration cycle, the system
initiates a backwash. During backwash, the feed pump shuts down, valves are repositioned, and
the backwash pump starts. The backwash pump draws treated water from the filtrate storage
tank, chlorinates it, and forces the water under pressure in the reverse direction through the fibers.
With the flow of water now from the inside of the fiber to the outside of the fiber, the backwash
water exits to the outside of the fibers, carrying with it particulate material which has accumulated
on the membrane surface during filtration. Chlorine added to the backwash water assists in
oxidizing organics that have accumulated on the membrane surface. Air is also added to the feed
side during the backwashing step to scour the membrane for more effective cleaning. The
backwash cycle typically lasts from 45 to 90 seconds, after which the unit returns to filtration
mode. Filtrate storage tank overflow and backwash waste streams were directed to drain.
The long-term operation of the Ionics UF unit frequently results in the accumulation of materials
on the membrane surface, which are not effectively removed by backwash. This is called
membrane fouling and is quantified by a gradual increase in the pressure required to force water
through the membrane pores. Once a critical upper pressure has been reached, normal operation
is discontinued and the membrane undergoes chemical cleaning. Chemical cleaning involves the
use of acid and caustic solutions to restore efficient operation of the membrane.
4
-------
The Ionics UF package unit uses seven model TP-TE07-S membrane modules manufactured by
Toray. Table 2-1 provides the specification of membranes used in the Ionics UF membrane
system. The information in Table 2-1 is taken from a letter supplied by the system manufacturer,
Ionics (see Appendix A). The Toray module is a hollow-fiber, outside-in configuration membrane
with nominal molecular weight cut-off of approximately 100,000 daltons. This corresponds with
a pore diameter of approximately 0.01 micron. At this pore size, the membrane is expected to
remove particulate material, including protozoa, bacteria and virus.
2.1 Description of the Treatment Train and Unit Processes
Figure 2-3 presents a schematic diagram of the Ionics UF system. The test system has two
alternating operation modes: filtration and backwash.
The operation of the UF membrane system is summarized in the following steps:
1. The feed pump provides the pressure (up to approximately 44 pounds per square inch (psi) or
3.0 bars) needed to filter the water through the membranes at a constant flow rate. Feed
pump speed is automatically adjusted to achieve the desired filtrate flow. Feed water is
pumped into the base of the pressure vessel.
2. The pressure in the 45 gallon pressure vessel forces water through the pores of the membrane
fibers to the inside of the fibers. The filtrate water travels up the fibers to the top of the
pressure vessel, which is sealed from the feed side of the pressure vessel, and into the
approximately 80 gallon (300 liter) filtrate storage tank. Overflow from the filtrate storage
tank was directed to drain. The modules filter on a cycle of 15 to 30 minutes, after which a
backwash is initiated.
3. Backwash is initiated automatically based on a timer. The objective of the backwash is to
remove solids and organics that have accumulated on the feed side (outside) of the membrane
surface during filtration. A PLC automatically operates pumps and valves to accomplish a
backwash.
4. There are two distinct backwash cycles. These are referred to as "B" backwash and "C"
backwash. During both Test Periods, nine "B" backwashes were performed, followed by a
"C" backwash. The number of "B" backwashes before a "C" backwash is selected by entering
the desired number of "B" backwashes in the appropriate screen on the PLC. The "B"
backwash consisted of air scour, fast flush with feed water, and reverse flow with chlorinated
filtrate. A "C" backwash consisted of fast flush with air scour, reverse flow, emptying and
then refilling the pressure vessel with feed water. In an effort to reduce fouling of the system,
the "B" backwash sequence was modified to include more reverse flow and less fast flush
water consumption between Test Periods 1 and 2. The "B" backwash sequence from Test
Period 1 is described below:
4.1 Fast Flush 1. During fast flush, feed water is pumped into the bottom of the pressure
vessel and exits the top of the pressure vessel removing some of the accumulated solids.
This step lasts approximately 10 seconds and consumes 12 gallons of feed water.
4.2 Reverse Flow. During this backwash phase, filtrate from the filtrate storage tank is
pumped under pressure in the reverse direction through the membrane and exits into the
pressure vessel. The backwash feed water is chlorinated to approximately 5-10
5
-------
milligrams per liter (mg/L). This step lasts approximately 15 seconds and consumes 15
gallons of filtrate.
4.3 Air Scour. During air scour, 1 standard cubic foot per minute (scfm) or 3.8 standard
liters per minute (slpm) air is introduced to the base of each module. The air flow
agitates the fibers and assists in removal of accumulated solids. This step lasts
approximately 20 seconds.
4.4 Intermediate Flush. The intermediate flush phase consists of a fast flush cycle with air
scour. This cycle lasts 10 seconds and consumes 12 gallons of feed water.
4.5 Fast Flush 2. This final fast flush with feed water removes air from the top of the
pressure vessel. This phase lasts approximately 5 seconds and consumes 7 gallons of
feed water.
Overall, the "B" backwash lasted 60 seconds. Waste from the backwash cycle is routed to
drain.
5. Backwash wastewater was directed to drain during ETV testing. At the completion of
backwash, the PLC stops the backwash pump, readjusts the appropriate valves and restarts
the system in filtration mode.
After extended periods of operation, typically on the order of weeks to months, the pressure
required to force water through the membrane pores increases because some material is not
effectively removed by backwash. This process is called membrane fouling. Once the system
reaches a critical pressure, the system is shut down and a chemical cleaning is performed to
restore membrane efficiency. The Ionics ETV test system was considered fouled when the
transmembrane pressure reached a critical pressure of 15 to 20 psi (1.0 - 1.4 bar). Cleaning the
Ionics unit is a two-step process. A citric acid solution with pH between 2.0 and 2.5 is used first.
This is followed by a high pH (pH in the range 10 to 12) cleaning step. Finally, a pH 2
hydrochloric acid rinse is performed to remove any metals that may have precipitated on the
membrane.
Each step in the cleaning process involves preparing approximately 100 gallons of cleaning
solution, preheated to 35 °C, in the feed storage tank contained on the membrane system skid.
The feed tank includes a heating element to maintain the cleaning solution temperature. Valves
are reconfigured so the fast flush drain flow and filtrate flow are returned to the feed/cleaning
tank. The feed pump is started, and the system is adjusted to recirculate cleaning solution on the
feed side of the membrane at 330 gallon per minute (gpm) or 1250 liter/min, with no filtrate flow,
for 60 minutes. After this, the filtrate valve is adjusted to allow a filtrate flow of 21 gpm (80
L/min), with the same concentrate flow for 15 minutes. After the cleaning step is complete, the
cleaning solution is directed to drain, the filtrate storage tank is filled with tap water and the
contents of the filtrate tank are backwashed through the system to remove residual cleaning
solution from the membrane modules. After the high pH cleaning step, a pH 2 hydrochloric acid
rinse is conducted to remove possible metal precipitates.
Filtration, in the Ionics UF unit, is accomplished with seven Model TP-TE07-S UF membrane
modules manufactured by Toray. The Toray membrane is a hollow fiber configuration with fibers
potted at the top only. Each fiber runs from the potting material down the inside of a clear acrylic
tube. The fiber then wraps around a plastic cross pipe near the base of the acrylic tube, and runs
back up the tube and the other fiber end is potted in the opposite half of the potting at the top.
6
-------
Each fiber has an inside diameter of approximately 0.016 inch (0.4 mm), an outside diameter of
0.026 inch (0.68 mm) and is 5.2 feet (1.6 m) long from end to end (see Table 2-1). With
approximately 3,600 fibers per module, the active surface area of each module is approximately
129 square feet (12 square meters). The membrane material is polyacrylonitrile. The surface of
the membrane has a neutral charge and is hydrophilic. The membrane is chlorine tolerant, has an
operating pH range of 2 - 10, and can operate to a maximum transmembrane pressure of 44 psi
(3.0 bar).
The fibers hang from the potting material inside a clear acrylic cylinder that is open at the bottom
and has approximately eight holes V2 inch (1.3 cm) in diameter around the top. The top of each
module has two o-rings that seal the feed side of the pressure vessel below, from the filtrate side
of the pressure vessel above. The modules are installed by removing the top of the pressure
vessel and slipping the modules into precision-machined holes in an approximately 3/4 inch (1.9
cm) thick aluminum plate.
2.2 Description of Physical Construction/Components of the Equipment
The Ionics UF unit is skid-mounted with a footprint of approximately 8 feet 9 inches (2.7 m) long
by 4 feet (1.2 m) deep. The unit is 7 feet 2 inches (2.2 m) in height with a base and frame
constructed of steel. At a weight of 2,800 pounds (1,270 kg), including air compressor, the unit
can be moved with a forklift and transported by truck. The Ionics UF unit is self contained,
requiring only connections to feedwater, drain and electrical. The electrical requirements of the
system are 50 amps of 480 volt three-phase, 60 Hz power.
The major components of the Ionics ETV test unit included:
• Seven 129 ft2 (12 m2) Toray TP-TE07-S UF modules housed in an 18 inch (46 cm) diameter
aluminum pressure vessel
• PLC-based control system
• Backwash pump
• Feed pump
• Feed storage / cleaning tank
• Filtrate storage tank
• Air compressor
• Pneumatic valves
• Sodium hypochlorite tank and metering pump
• Rotameter and magmeter flow meters
• Digital pressure gauges
• Digital feed thermometer.
Figure 2-2 presents the spatial requirements and layout of the major components of the Ionics UF
unit.
7
-------
Chapter 3
Materials and Methods
3.1 Testing Site Name and Location
The test site selected for the ETV project is the City of San Diego's Aqua 2000 Research Center
at 14103 Highland Valley Road in Escondido, California.
3.1.1 Site Background Information
The Aqua 2000 Research Center was established in 1995 to conduct most of the research work
related to the Water Repurification Project of the City of San Diego. The Center has dedicated
full time operators with substantial experience in operating membrane systems. This site is also
connected to San Diego County Water Authority's Aqueduct System. Sufficient raw water
supply, electrical power, and proper drainage lines to a wastewater treatment plant were provided
to the ETV test system treatment train.
3.1.2 Test Site Description
Figure 3-1 is a schematic diagram of the test site and the location of the Ionics UF unit. Below is
a list of the facilities and equipment that were available at the test site.
Structural
• 5,000 square foot concrete pad.
• Semi-permanent shading to protect from sunlight.
• Potable water connections.
• San Diego County Water Authority's Aqueduct System connections.
• Drainage system connected to a wastewater plant.
• Chemical containment area.
• Sufficient lighting for 24-hour operation.
• Full electrical supply.
• Chemical safety shower and eyewash.
• An operations trailer with conference room, offices, and computers.
• A laboratory trailer for on-site water quality analyses.
Instrumentation/Equipment
On-Site Laboratory
• DR 4000 Spectrophotometer by Hach
• Ratio/non-ratio 2100N Turbidimeter by Hach
• pH/Temperature meter by Accumet Research (AR-15)
• Portable conductivity meter by Fisher (No. 09-327-1)
• Two total organic carbon (TOC) Analyzers (Sievers Model No. 800)
8
-------
Concrete Pad
• Feed, filtrate, backwash, and waste storage tanks.
• Chemical Cleaning Skid with hot water supply.
• Chemical Feed Systems.
• Micro 2000 On-line Chlorine Analyzer
• Four 1720D On-line Hach Turbidimeters
• Four 1900WPC On-line Hach Particle Counters
Raw Water Intake
The raw water was delivered to the test site through schedule 80 PVC pipe. The San Diego
Aqueduct connection was approximately one mile away from the test site. The available water
flow rate was 150 gpm.
Collection of Raw Water
The raw water was directed to a covered tank with an overflow system. The feedwater pipe of
the test unit was connected to the covered raw water tank.
Handling of Treated Water and Residuals
The Aqua 2000 Research Center has a drainage system that connects to a wastewater treatment
plant. All of the filtrate, backwash water, and any chemicals used were directed to waste.
3.2 Source/Feed Water Quality
The source of feedwater for the ETV testing is San Diego Aqueduct Water. The aqueduct is
supplied primarily from Lake Skinner which receives Colorado River Water (CRW) from the
West Portal of the San Jacinto Tunnel, and State Project Water (SPW) from Lake Silverwood. A
typical blending ratio of these two waters in Lake Skinner is 70 percent CRW and 30 percent
SPW. The lower total dissolved solids (TDS) SPW is added to maintain the TDS of Lake
Skinner at approximately 500 mg/L or less (depending on availability of SPW). The aqueduct
water is characterized by relatively high levels of total dissolved solids, hardness and alkalinity,
with moderate levels of organic material and relatively low turbidity.
Figure 3-2 illustrates Lake Skinner water quality for the period of November 1997 through
November 1998, which is typical for this source water. The stable quality of the water is apparent
in all parameters illustrated in the figure. Hardness ranged from 200 through 298 mg/L as CaC03,
alkalinity ranged from 108 to 130 mg/L as CaC03 and calcium ranged from 47 to 75 mg/L as Ca
(118 to 188 mg/L as CaC03). The hardness levels are quite high, with relatively high alkalinity as
well. TDS ranged from 429 to 610 mg/L, indicating the relatively high level of salinity in this
source water. pH ranged from 8.26 to 8.45 during the year.
Figure 3-3 illustrates turbidity, temperature and TOC for Lake Skinner water. Turbidity was
relatively low with a range of 1.10 to 3.50 nephalometric turbidity units (NTU). Lake Skinner
exhibits relatively warm temperatures throughout the year, typical of many water supplies in the
southwestern and southeastern United States. The temperature range was 13 to 27°C. Annual
9
-------
low temperatures on the order of 10°C are typical of this supply. The levels of organic material,
as quantified by TOC, are moderate in this supply. The TOC range was 2.33 to 2.94 mg/L.
3.3 Environmental Technology Verification Testing Plan
This section describes the tasks completed for the ETV. The test equipment was operated 24
hours a day, seven days a week, with operations staff on-site Monday through Friday for one 8-
hour shift each day. Tasks that were performed by the operations and engineering staff are listed
below:
Task 1: Characterization of Membrane Flux and Recovery
Task 2: Evaluation of Cleaning Efficiency
Task 3: Evaluation of Finished Water Quality
Task 4: Reporting of Membrane Pore Size
Task 5: Membrane Integrity Testing
Task 6: Data Management
Task 7: Quality Assurance/Quality Control
Task 8: Microbial Removal (optional)
An overview of each task is provided below.
3.3.1 Task 1: Characterization of Membrane Flux and Recovery
The objective of this task is to evaluate the membrane operational performance. Membrane
productivity was evaluated relative to feedwater quality. The rates of transmembrane pressure
increase and/or specific flux decline were used, in part, to evaluate operation of the membrane
equipment under the operating conditions being verified and under the raw water quality
conditions present during the verification testing period.
Work Plan
After set-up and shakedown of the membrane equipment, membrane operation was established at
the flux condition being verified in this ETV. Testing took place over two 30-day test periods.
When substantial specific flux decline occurred before the end of the 30-day test period, chemical
cleaning was performed and (if necessary) adjustments to the operational strategy were made.
Measurement of the membrane system flows, pressures and temperatures were collected at a
minimum of twice a day.
3.3.2 Task 2: Evaluation of Cleaning Efficiency
An important aspect of membrane operation is the restoration of membrane productivity after
specific flux decline has occurred. The objective of this task is to evaluate the effectiveness of
chemical cleaning for restoring finished water productivity to the membrane system. The
recovery of specific flux and the fraction of original specific flux lost were determined after each
chemical cleaning.
10
-------
Work Plan
The membrane was operated at the flux condition being verified in this ETV until such time as the
termination criteria were reached. The two criteria for cleaning of the membrane were: 1)
reaching the minimum specific flux operational limit of the membrane (specific flux < 0.85
gfd/psi), or, 2) completing the 30-day test period. The membrane was chemically cleaned when
either of these termination criteria was reached. Chemical cleaning was performed in accordance
to the manufacturer procedure (see Appendix A). For the feedwater utilized in this ETV, the
manufacturer recommended their typical chemical cleaning procedure using citric acid and caustic
cleaning solutions.
The first cleaning step uses a two percent citric acid solution in tap water preheated to 35 °C,
with pH in the range 2.0 to 2.5. This is followed by a high pH cleaning step using 0.5 percent
caustic solution in tap water preheated to 35 °C, with pH in the range 11 to 12. On the
recommendation of Ionics, a proprietary high pH cleaning agent, ROClean L211, manufactured
by Avista, was used instead of caustic. This cleaning agent contains the metal chelating agent
ethylenediamine tetraacetic acid. The high-pH cleaning step includes a final pH 2 hydrochloric
acid rinse to remove potential metal precipitates.
To determine cleaning efficiency, flux-pressure profiles were developed at each stage of the
chemical cleaning procedure (i.e., before cleaning, after first chemical solution, after second
chemical solution). The slope of the flux-pressure profile represents the specific flux of the
membrane at each cleaning stage and was used to calculate the cleaning efficiency indicators. Two
primary indicators of cleaning efficiency and restoration of membrane productivity were examined
in this ETV:
1. The immediate recovery of membrane productivity, as expressed by the ratio between the final
specific flux value of the current filtration run (Jsf) and the initial specific flux (Js;) measured
for the subsequent filtration run:
Recovery of Specific Flux = 100 x [1 - (Jsf Js; )]
where: Jsf = specific flux (gallon/ft2/day (gfd)/psi, L/(hr-m2)/bar) at end
of current run (final)
Js; = specific flux (gfd/psi, L/(hr-m2)/bar) at
beginning of subsequent run (initial)
2. The loss of specific flux capabilities is expressed by the ratio between the initial specific flux
for any given filtration run (Js;) and the specific flux (Js;0) at time zero, as measured at the
initiation of the first filtration run in a series:
Loss of Original Specific Flux = 100 x [1 - (Jsf Js;0)]
where: Js;0= specific flux (gfd/psi, L/(hr-m2)/bar) at
time t = 0 of membrane testing
11
-------
3.3.3 Task 3: Evaluation of Finished Water Quality
The objective of this task is to evaluate the quality of water produced by the ETV test system.
Many of the water quality parameters described in this task were measured on-site. Analyses of
the remaining water quality parameters were performed by the City of San Diego Laboratory, a
State-certified analytical laboratory.
Work Plan
The parameters monitored during this ETV and the methods used for their measurement are listed
in Table 3-1. Finished water quality was evaluated relative to feedwater quality and operational
conditions.
3.3.4 Task 4: Reporting of Membrane Pore Size
Membranes for particle and microbial removal do not have a single pore size, but rather have a
distribution of pore sizes. Membrane rejection capabilities are limited by the maximum membrane
pore size.
Work Plan
The manufacturer was asked to supply the 90 percent and the maximum pore size of the
membranes being tested in the ETV. The manufacturer was also asked to identify the general
method used in determining the pore size values.
3.3.5 Task 5: Membrane Integrity Testing
A critical aspect of any membrane process is the ability to verify that the process is producing a
specified water quality on a continual basis. For example, it is important to know whether the
membrane is providing a constant barrier to microbial contaminants. The objective of this task is
to evaluate one or more integrity monitoring methods for the membrane system.
Work Plan
The selected methods for monitoring of membrane integrity of the Manufacturer's UF system
during this study are described below:
Air Pressure-Hold Test
The air pressure-hold test is one of the direct methods for evaluation of membrane integrity. This
test can be conducted on several membrane modules simultaneously; thus, it can test the integrity
of a full rack of membrane modules used for full-scale systems. The test is conducted by
pressurizing the filtrate side of the membrane after which the pressure is held and the decay rate is
monitored over time. Minimal loss of the held pressure (generally less than 1 psi every 5 minutes)
at the filtrate side indicates a passed test, while a significant decrease of the held pressure indicates
a failed test.
12
-------
Particle Counting
On-line particle counting in the size ranges of 2-3 microns (um), 3- 5 um, 5-7 um, 7-10 um, 10-15
um and >15 um was used in this ETV as an indirect method of monitoring membrane integrity.
Turbidity Monitoring
On-line turbidity monitoring was also used in this ETV as an indirect method of monitoring
membrane integrity.
3.3.6 Task 6: Data Management
The objective of this task is to establish the protocol for management of all data produced in the
ETV and for data transmission between the FTO and the NSF.
Work Plan
According to EPA/NSF ETV protocols, a data acquisition system was used for automatic entry of
on-line testing data into computer databases. Specific parcels of the computer databases for
online particle and turbidity were then downloaded for importation into Excel as a comma
delimited file. These specific database parcels were identified based on discrete time spans and
monitoring parameters. In spreadsheet form, data were manipulated into a convenient framework
to allow analysis of membrane equipment operation. For those parameters not recorded by the
data acquisition system, field-testing operators recorded data and calculations by hand in
laboratory notebooks. Daily measurements were recorded on specially-prepared data log sheets as
appropriate.
The database for the project was set up in the form of custom-designed spreadsheets. The
spreadsheets were capable of storing and manipulating each monitored water quality and
operational parameter from each task, each sampling location, and each sampling time. Data from
the log sheets were entered into the appropriate spreadsheet. Following data entry, the
spreadsheet was printed out and the printout was checked against the handwritten data sheet.
Any corrections were noted on the hard-copies and corrected on the screen, and then a corrected
version of the spreadsheet was printed out. Each step of the verification process was initialed by
the field testing operator or engineer performing the entry or verification step.
Data from the outside laboratory were received and reviewed by the field testing operator. Data
from the on-site lab and City of San Diego Microbiology lab were entered into the data
spreadsheets, corrected, and verified in the same manner as the field data. Data from the City of
San Diego Water Quality lab were received both electronically and in hardcopy printouts
generated from the electronic data.
3.3.7 Task 7: Quality Assurance/Quality Control
An important aspect of verification testing is the protocol developed for quality assurance (QA)
and quality control (QC). The objective of this task is to assure the high quality of all
measurements of operational and water quality parameters during the ETV.
13
-------
Work Plan
Equipment flow rates and pressures were documented and recorded on a routine basis. A routine
daily walk-through during testing is performed each morning to verify that each piece of
equipment or instrumentation is operating properly. On-line monitoring equipment, such as flow
meters, are checked to confirm that the read-out matches the actual measurement and that the
signal being recorded is correct. Below is a list of the verifications conducted:
Monitoring Equipment
System Pressure Gauges
Pressure and vacuum gauges supplied with the membrane systems tested were verified against
grade 3A certified pressure or vacuum gauges purchased at the start of ETV testing. The
certified pressure and vacuum gauges were manufactured by Ashcroft and have an accuracy of
0.25% over their range (0-30 psi pressure). Where possible, system gauges were removed and
tested over the expected range of operating pressures against the verification gauge, using a
portable hand pump. The Ionics system feed and differential pressure gauges were consistently
accurate to within five percent or less over their range. The filtrate pressure gauge was accurate
to within 0.2 psi over its range.
System Flow Rates
Membrane system flow rates were verified volumetrically on a monthly basis near the beginning
and end of each test period. System flows were diverted to a 55 gallon graduated tank for
approximately two minutes. The measured flow rate was compared with flows indicated on the
rotameter and magmeter. Measured and indicated flows agreed to within three percent for the
filtrate rotameter. The magmeter consistently measured 1 gpm lower than actual flow. Since the
filtrate flow rate was automatically adjusted based on the magmeter reading, this was
compensated for during testing by entering a filtrate flow setpoint of 20 gpm into the PLC,
resulting in an actual flow rate of 21 gpm.
Analytical Methods
pH
An Accumet Research Model AR15 laboratory pH meter was used to conduct routine pH
readings at the test facility. Daily calibration of the pH meter using pH 4, 7 and 10 buffers was
performed. The slope obtained after calibration was recorded. The temperature of the sample
when reading sample pH was also recorded.
Temperature
Accuracy of the feed water inline thermometer was verified against an National Institute for
Standards and Technology (NIST)-certified thermometer on 12/12/99 and 4/7/00. Comparisons
were made at three temperatures covering the range of anticipated raw water temperatures. In all
cases, the raw water thermometer compared to within ± 0.2°C of the NIST-certified
thermometer.
14
-------
Turbidity
On-line turbidimeters were used for measurement of turbidity in the raw and filtrate waters, and a
bench-top turbidimeter was used for measurement of the feedwater and backwash waste water.
On-line Turbidimeters: Hach 1720D on-line turbidimeters were used during testing to acquire raw
and filtrate turbidities at one-minute intervals. The following procedures were followed to ensure
the integrity and accuracy of these data:
• a primary calibration of the on-line turbidimeters (using formazin primary standards) was
performed near the beginning of the test periods.
• Aquaview + data acquisition software was used to acquire and store turbidity data. Data were
stored to the computer database each minute. After initial primary calibration of the
turbidimeters, zero, mid-level and full-strength signals (4, 12 and 20 mA) were output from
each turbidimeter to the data acquisition software. The signals received by the data
acquisition software from all four on-line turbidimeters had less than one percent error over
their range of output (0, 1 and 2 NTU for filtrate, and 0, 10 and 20 NTU for feed) as stored in
the Aquaview database.
• the manufacturer's specified acceptable flow range for these turbidimeters is 250 to 750
mL/min. The flow range initially targeted during testing was 500 mL/min +/- 100 mL/min.
On-line turbidimeter flows were verified manually with a graduated cylinder and stopwatch
daily.
• turbidimeter bodies were drained and sensor optics cleaned approximately every week on an
as-needed basis.
• on-line turbidities were compared to desktop turbidities when turbidity samples were
collected. Comparative calibrations of the raw water on-line turbidimeter against the Hach
2100N desktop turbidimeter were conducted on an as-needed basis during the course of the
testing when the difference between on-line and desktop turbidity readings were greater than
approximately 10 percent.
• Approximately 50 part per million (ppm) free chlorine solution was pumped through turbidity
sample lines as needed to clean potential buildup from these lines.
Desktop Turbidimeters: A Hach 2100N desktop turbidimeter was used to perform onsite turbidity
analyses of raw water, backwash and filtrate samples. Readings were recorded in non-ratio
operating mode. The following quality assurance and quality control procedures were followed to
ensure the integrity and accuracy of onsite laboratory turbidity data:
Primary calibration of turbidimeter according to manufacturer's specification was conducted on a
weekly basis. Secondary standard calibration verification was performed on a daily basis. Three
secondary standards (approx. 0.8 NTU, 1.8 NTU and 20 NTU) were recorded after primary
calibration and on a daily basis for the remaining 6 days until the next primary calibration.
Proficiency samples with a known turbidity of 0.8 NTU were purchased from a commercial
supplier. Turbidity proficiency samples were prepared and analyzed every two weeks.
15
-------
Particle Counting
Hach 1900 WPC light blocking particle counters were used to monitor particles in raw and filtrate
waters. These counters enumerate particles in the range 2 to 800 microns (um).
The particle counters were factory calibrated. Factory calibrations took place on May 25, 1999.
The manufacturer recommends factory calibration on a yearly basis. The following procedures
were followed to ensure the integrity and accuracy of the on-line particle data collected:
• The Aquaview software was configured to store particle counts in the following size ranges:
2-3 um, 3-5 um, 5-7 um, 7-10 um, 10-15 um and >15 um.
• To demonstrate the comparative response of the particle counters, NIST traceable
monospheres were purchased from Duke Scientific in the following sizes: 2 um, 4 um, 10 um
and 20 um. Duke monospheres were added to constantly stirred deionized (DI) water and
pumped to one of the constant head flow controllers using a peristaltic pump. The flow from
this controller was then directed to each of the particle counters for approximately 10 minutes.
The same solution was used for each particle counter (raw water and filtrate).
The precise concentration of each monosphere was not known, but based on Duke Scientific
estimates the following approximate concentration of each monosphere was present in the test
solution:
A typical response of the particle counters to this monosphere solution near both test periods is
presented in Figure 3-4. The figures show a good comparative response of the raw water and
filtrate particle counters to the same monosphere solution.
Flows through the particle counters were maintained at 200+/- 10 mL/min with constant head
devices. Flows were verified on a daily basis with a graduated cylinder and stop watch. Flows
were observed to be extremely consistent (typically within 2 mL/min of the target flow rate).
Fifty mg/L free chlorine was run through particle counters for on an as-needed basis to remove
potential buildup.
Chemical and Microbial Water Quality Parameters
The analytical work for the study was performed by the City of San Diego Laboratory, which is a
State of California certified water laboratory. All water samples were collected in appropriate
containers (containing preservatives as applicable) prepared by the City of San Diego laboratory.
Samples for analysis of Total Coliforms (TC) and Heterotrophic Plate Count (HPC) analysis were
collected in bottles supplied by the City of San Diego laboratory and transported with an internal
cooler temperature of approximately 2 to 8°C to the analytical laboratory. All samples were
preserved, stored, shipped and analyzed in accordance with appropriate procedures and holding
times. All reported results had acceptable QA and met method-specific QC guidelines, which was
confirmed by letters from the City of San Diego Water Quality and Marine Microbiology
• 2 um
• 4 um
• 10 um
• 20 um
1,000 - 10,000/mL
100 - 1,000/mL
10 - 100/mL
1 - 10/mL
16
-------
Laboratories (Appendix A). For the Marine Microbiology Laboratory, these QC procedures
included the use of positive / negative controls, blanks and sterility checks.
3.3.8 Task 8: Microbial Removal (Optional)
The objective of this task is to evaluate microbial removal capabilities by seeding the membrane
system with selected virus. Removal capabilities were evaluated under the worst case scenario for
the membrane system operation (in this case, directly after chemical cleaning of the membranes).
Work Plan
The seeding experiments were performed at the test site and the samples collected during the
seeding experiments were submitted to the City of San Diego Marine Microbiology Lab, a State-
certified laboratory, for analysis of the seeded microorganisms.
Organisms for Seeding Experiments
The organism selected for seeding experiments is MS2 bacterial virus. MS2 virus is not a human
pathogen; however, this organism is similar in size (0.025 microns), shape (icosahedron) and
nucleic acid (RNA) to polio and hepatitis virus. Since MS2 is not a human pathogen, live MS2
virus was used in the seeding experiments. Organism stocks received from the suppliers were
stored refrigerated at 4°C in the dark until use in the seeding experiments.
Microbial Seeding Protocols
The virus were added to approximately 100 gallons (380 liter) of dechlorinated tap water in a 55
gallon polyethylene tank. A peristaltic pump was used to continuously add this virus stock
solution to the membrane feed water. During the MS2 seeding experiment, three samples from the
membrane feed water and three samples from the filtrate water were collected during the second
and third service cycles after the initiation of seeding. The first filtrate sample during each
filtration cycle was collected within the first minute of filtration after completion of backwash.
The last filtrate sample during each filtration cycle was collected within 3 minutes of the end of
the cycle. Each sample was collected in sterile 250-mL bottles, was stored at 1°C and processed
within 24 hours. The microorganism concentration in the feed water was sufficient to
demonstrate a minimum of 4 logs of removal of the seeded organism.
The MS2 seeding experiments were conducted at the end of Test Period 1 and the beginning of
Test Period 2. The experiments were conducted under the operating conditions in which the
microorganisms would most likely penetrate the membrane; when the membrane is clean, and at a
high flux rate (Jacangelo et al. 1995, Montgomery Watson, 1997). Therefore, the membrane was
cleaned immediately prior to MS2 seeding.
17
-------
3.4 Calculation of Membrane Operating Parameters
3.4.1 Filtrate Flux
The average filtrate flux is the flow of filtrate water divided by the surface area of the membrane.
Filtrate flux is calculated according to the following formula:
Jt = QP - S
where Jt = filtrate flux at time t (gfd, L/(hr-m2))
Qp = filtrate flow (gallon per day (gpd), L/hr)
S = membrane surface area (ft2, m2)
Flux is expressed only as gfd and L/(hr-m2) in accordance with EPA/NSF ETV protocol.
3.4.2 Specific Flux
The term specific flux is used to refer to filtrate flux that has been normalized for the
transmembrane pressure. The equation used for calculation of specific flux is:
Jtm Jt ~ Ptm
where Jtm = specific flux at time t
(gfd/psi, L/(hr-m2)/bar)
Jt = filtrate flux at time t (gfd, L/(hr-m2))
Ptm = transmembrane pressure (psi, bar)
3.4.3 Transmembrane Pressure
The average transmembrane pressure is calculated as follows:
Ptm = [(Pi + Po) - 2] - Pp
where Pta = transmembrane pressure (psi, bar)
P; = pressure at the inlet of the membrane
module (psi, bar)
P0 = pressure at the outlet of the membrane
module (psi, bar)
Pp = filtrate pressure (psi, bar)
3.4.4 Temperature Adjustment for Flux Calculation
Temperature corrections to 20°C for transmembrane flux were made to account for the variation
of water viscosity with temperature. The following equation was employed:
18
-------
where Jt = instantaneous flux (gfd, L/(hr-nf))
Qp = filtrate flow (gpd, L/hr)
T = temperature, (°F, °C)
S = membrane surface area (ft2, m2)
3.4.5 Feedwater System Recovery
The recovery of filtrate from feedwater is the ratio of filtrate flow to feedwater flow:
% System Recovery = 100 x (Qp/Qf)
where Qp = filtrate flow (gpd, L/hr)
Qf = feed flow to the membrane (gpd, L/hr)
3.4.6 Rejection
The rejection of contaminants by membrane process was calculated as follows:
Cp
R = (1 —) *100%
Cp
where: R = Rejection, %
Cp = Filtrate water concentration, (mg/L)
Cp = Feed water concentration, (mg/L)
3.5 Calculation of Data Quality Indicators
3.5.1 Precision
As specified in Standard Methods (Method 1030 C), precision is specified by the
standard deviation of the results of replicate analyses. An example of replicate analyses
in this ETV is the biweekly analysis of turbidity proficiency samples. The overall
precision of a study includes the random errors involved in sampling as well as the errors
in sample preparation and analysis.
n
Precision = Standard Deviation = V[£ (x, - x)2 ^ (n - 1)]
i= 1
where: X = sample mean
X, = z'th data point in the data set
n = number of data points in the data set
19
-------
3.5.2 Relative Percent Deviation
For this ETV, duplicate samples were analyzed to determine the overall precision of an analysis
using relative percent deviation. An example of duplicate sampling in this ETV is the daily
duplicate analysis of turbidity samples using the bench-top turbidimeter.
Relative Percent Deviation = 100 x [(xi - x2) ^ x]
where x = sample mean
xi = first data point of the set of two duplicate
data points
x2 = second data point of the set of two
duplicate data points
3.5.3 Accuracy
Accuracy is quantified as the percent recovery of a parameter in a sample to which a known
quantity of that parameter was added. An example of an accuracy determination in this ETV is
the analysis of a turbidity proficiency sample and comparison of the measured turbidity to the
known level of turbidity in the sample.
Accuracy = Percent Recovery = 100 x [XmeaSured Xkn0wn]
where Xkn0wn = known concentration of
measured parameter
XmeaSured = measured concentration of
parameter
3.5.4 Statistical Uncertainty
For the water quality parameters monitored, 95 percent confidence intervals were calculated. The
following equation was used for confidence interval calculation:
Confidence Interval Xi [tn-i;i - (a/2) * (S/Vn)]
X =
sample mean
S =
sample standard deviation
n =
number of independent measurements
included in the data set
t =
Student's t distribution value with n-1
degrees of freedom
a =
significance level, defined for 95 percent
confidence as: 1 - 0.95 = 0.05
20
-------
According to the 95 percent confidence interval approach, the a term is defined to have the value
of 0.05, thus simplifying the equation for the 95 percent confidence interval in the following
manner:
95 Percent Confidence Interval = x ± [tn-i,0.975 x (S/Vn)]
3.6 Testing Schedule
The ETV schedule is illustrated in Figure 3-5. The testing project took place starting in
December 1999 and finishing by the beginning of April 2000. Test Period 1 represented the
winter season and Test Period 2 represented the spring season.
21
-------
Chapter 4
Results and Discussion
This chapter presents the data obtained under each task of the ETV project of the Ionics UF
system.
4.1 Task 1: Characterization of Membrane Flux and Recovery
The operating conditions for the Ionics UF membrane system are provided in Table 4-1. The
manufacturer established the operating parameters for the ETV testing. The membrane system
ran at a target flux of 33 gfd (57 L/hr-m2). Filtration cycle length was 30 minutes followed by a
60 to 65 second "B" backwash or 130 to 155 second "C" backwash. Nine "B" backwashes were
performed before a "C" backwash was performed and the pressure vessel was drained. Filtrate
consumed during backwash was 15 gallons (57 liters) for Test Period 1 and 30 gallons (113 liters)
for Test Period 2. Feed water consumed during backwash was 30 gallons (113 liters) during Test
Period 1 and 7 gallons (26 liters) during Test Period 2. The backwash feed water was chlorinated
at 5 - 10 mg/L during reverse flow. The feed water recovery was 93 percent during Test Period 1
and 92 percent during Test Period 2.
Figure 4-1 (A and B) provides the membrane transmembrane pressure and temperature profiles
for Test Periods 1 and 2. Operational readings were taken approximately 5 minutes before and
after backwash. These are displayed on the figures as pairs of data points at nearly the same point
in time. The data point taken before backwash has the higher transmembrane pressure value. For
Test Period 1, the clean membrane transmembrane pressure began at approximately 6 psi. The
transmembrane pressure stabilized at 7 to 10 psi for approximately 2.5 weeks and then fouled
more rapidly over the remainder of the filter run. Transmembrane pressure at the beginning of
Test Period 2 was 6 psi. The transmembrane pressure remained between 7 and 10 psi for the
remainder of Test Period 2. The changes made to the backwash conditions between Test Periods
1 and 2 are likely responsible for the improved fouling performance during Test Period 2.
Figure 4-2 (A and B) provides the membrane flux and specific flux profiles for Test Periods 1 and
2. The target flux during Test Periods 1 and 2 was 33 gfd (57 L/hr-m2). For Test Period 1, the
average temperature adjusted membrane flux was 37 gfd at 20°C (63 L/hr-m2 at 20°C). Due to
the relatively higher water temperatures during Test Period 2, a lower average temperature
adjusted membrane flux of 34 gfd at 20°C (58 L/hr-m2 at 20°C) was observed. The temperature
adjusted specific flux decreased from 6 to 1 gfd/psi at 20°C (148 to 25 L/hr-m2-bar at 20°C) over
the 35 days of Test Period 1. Temperature adjusted specific flux decreased from 5.8 to 4.8
gfd/psi at 20°C (143 to 118 L/hr-m2-bar at 20°C) over the first 3 days of operation in Test Period
2. Temperature corrected specific flux then gradually decreased to 3.8 gfd/psi at 20°C (94 L/hr-
m2-bar at 20°C) by the end of Test Period 2.
The same data in Figures 4-1 and 4-2 are also provided in Appendix A of this report, but with
metric units.
22
-------
4.2 Task 2: Evaluation of Cleaning Efficiency
Chemical cleanings were performed when the membrane fouled (transmembrane pressure in the
range 15 to 20 psi [1.0 to 1.4 bar]), or the end of a test period was reached. The manufacturer's
cleaning procedure was a two step process. A citric acid cleaning solution was used first,
followed by a high pH cleaning solution. The 2 percent citric acid cleaning solution was prepared
by dissolving 17 pounds (37 kg) of citric acid in approximately 30 gallons of tap water preheated
to 35 °C. The pH of this solution was in the range 2 to 2.5. The citric acid solution was placed in
the feed tank and recirculated through the feed side of the membrane for 60 minutes at a flow of
330 gpm (125 L/min) with a feed pressure of approximately 15 psi. After this, filtrate flow was
adjusted to 21 gpm (79 L/min) and the cleaning solution was allowed to recirculate for an
additional 15 minutes. After discarding the cleaning solution and rinsing the system with tap
water, the same cleaning procedure was followed using a high pH cleaning solution. The high pH
cleaning solution was made by adding 1.8 gallon (7 liters) of Avista ROClean L211 to 100 gallons
tap water preheated to 35 °C. The pH of this solution was in the range 10.5. Since the high pH
cleaning solution was prepared in tap water, the caustic cleaning step included a pH 2
hydrochloric acid rinse to remove any precipitates that potentially formed under these conditions.
The flux-pressure profiles of the membrane system at different stages of the chemical cleaning
procedure for Test Periods 1 and 2 are shown in Figures 4-3 and 4-4, respectively. The slope of
the flux-pressure profile represents the specific flux of the membrane at each cleaning stage and
was used to calculate the cleaning efficiency indicators. These are listed in Table 4-2. The
recovery of specific flux for the cleanings at the end of Test Period 1 and 2 were 78 percent and
23 percent, respectively. The lower recovery after the cleaning at the end of Test Period 2 was
due to the fact the membrane was not completely fouled when the cleaning was conducted and
because the specific flux after cleaning was not as high as during the previous cleaning.
The membrane lost 4.8 percent of original specific flux after cleaning at the end of Test Period 1.
The loss of original specific flux increased to 17 percent after the cleaning at the end of Test
Period 2. Because of the limited number of cleanings, the usable membrane life can not be
estimated.
The same data in Figures 4-3 and 4-4 are also provided in Appendix A of this report, but with
metric units. In addition, the manufacturer's detailed cleaning procedure is included in Appendix
A.
4.3 Task 3: Evaluation of Finished Water Quality
Several water quality parameters were monitored during testing. Below is a summary of the
water quality data.
4.3.1 Turbidity, Particle Concentration and Particle Removal
23
-------
Figures 4-5 and 4-6 present the on-line turbidity profile for the Ionics UF membrane system
during Test Periods 1 and 2, respectively. The figures show online turbidity for raw and filtrate
water and desktop turbidity for raw water, filtrate and backwash waste. The desktop turbidity
data are summarized in Table 4-3 and the online turbidity data are summarized in Table 4-4. For
both testing periods, the raw water turbidity was in the range of 1-3 NTU. The turbidity of the
backwash waste water averaged about 17 NTU for Test Period 1 and 12 NTU for Test Period 2.
The filtrate turbidity was typically below 0.1 NTU.
Figures 4-7 and 4-8 present the particle count profile (2-3 um, 3-5 um, 5-7 um, 7-10 um, 10-15
um and >15 um) collected during Test Periods 1 and 2, respectively. The data presented
represent 4-hour average values of data collected at one-minute intervals. For both testing
periods, the feed particle concentration of the Cryptosporidium-sized particles (3-5 um) were in
the range of 1,000 to 10,000 particle/mL while the combined Giardia-sized particles (5-7 um, 7-
10 um and 10-15 um) were in the range 300 to 1,500 particle/mL. The filtrate concentration in
these size ranges was typically in the range of 0.04 to 3 particle/mL during Test Period 1. The
gap in the particle data near December 25, 1999 was due to an electrical power failure to the
particle counters. The gap at January 1, 2000 was due to a Y2K related software failure. The
sudden increase in filtrate particle concentration on January 3, 2000 was due to a single broken
fiber. After removing the pressure vessel cover, a bubble-point test was conducted and one
compromised fiber was detected and repaired. Since both ends of the fiber are potted, two air
bubbles would be expected from both ends. However, the fibers tended to break near the potting
material and the fiber end with the long fiber attached frequently did not produce visible bubbles.
Filtrate particle levels during Test Period 2 exhibit a general increasing trend. During the course
of Test Period 2, a number of incidents of broken fibers were detected by visual observation of
real time filtrate particle counts. The repairs near the beginning of Test Period 2 were generally
successful in decreasing particle counts to near pre-break levels. Two compromised fibers were
detected, via the bubble point test, after the break that occurred on March 13, 2000 and were
repaired on March 15, 2000. One compromised fiber was detected from the break that occurred
on March 20, 2000 and was repaired on March 22, 2000. One compromised fiber was detected
from a break that occurred on March 26, 2000 and was repaired on March 28, 2000. It was
observed during the repair of March 28, 2000 that the potting material around the leaking fiber
end was damaged and that the stainless steel pin used to make the repair was not seating tightly.
During a repair of the same leaking fiber end on March 31, 2000, a larger pin was installed, but
filtrate particle counts never recovered to previous levels. It was believed that this repaired fiber
end continued to leak, to lesser degrees, for the remainder of Test Period 2. Particle removals
were lower during periods of operation with compromised fibers.
Figures 4-9 and 4-10 present the log removal of particles (2-3 um, 3-5 um, 5-7 um, 7-10 um, 10-
15 um, and >15 um) based on raw and filtrate particle count data collected during Test Periods 1
and 2, respectively. Data presented on this plot represent one-day average values of data
collected at one minute intervals. Removal ranged from 3.0 to 4.6 logs for the Cryptosporidium-
sized particles (3-5 um) and from 2.8 to 4.2 logs for the Giardia-sized particles (5-7 um, 7-10 um
and 10-15 um) during Test Period 1. Removals decreased to between 2.9 to 3.9 logs for the
24
-------
Cryptosporidium-sized particles and to between 2.6 to 3.9 logs for the Giardia-sized particles
during Test Period 2. The online turbidity and particle removal data are summarized in Table 4-4.
To assist in assessing test system performance, Figure 4-11 presents the probability plots of the
membrane system filtrate turbidity and particle removal data for the Cryptosporidium-sized
particles (3-5 um) and Giardia-sized particles (5-15 um). The figure shows that the filtrate
turbidity was 0.05 NTU or below 95 percent of times and that removal of particles (3-5 um and 5-
15 um) was greater than 3 logs 95 percent of times.
4.3.2 Indigenous Bacteria Removal
The removal of naturally occurring bacteria was also monitored during the ETV study (see Table
4-5). The raw water total coliform bacteria ranged from <2 to 80 most probable number
(MPN)/100mL during Test Period 1 and from <2 to 17 MPN/lOOmL during Test Period 2. Total
coliforms bacteria were not detected in the filtrate of the Ionics UF membrane system during
either test period. HPC bacteria were not reduced by membrane filtration. This could be due to
the fact some fibers were compromised and repaired during the testing, which may have
contaminated the filtrate side. Previous studies (Jacangelo et al., 1995) have demonstrated that
HPC bacteria can be introduced on the filtrate side of the membrane rather than by penetration
through it. HPC bacteria in the raw water ranged from 2 to 120 colony forming units (cfu)/mL
during Test Period 1 and from 2 to 1400 cfu/mL in Test Period 2. HPC bacteria in the Ionics UF
filtrate ranged from 11 to 140 cfu/mL during Test Period 1 and from 48 to 580 cfu/mL during
Test Period 2.
4.3.3 Other Water Quality Parameters
Table 4-6 presents the results of general water quality parameters across the Ionics UF system for
Test Periods 1 and 2. As expected, no change was observed in the alkalinity, total dissolved
solids, total hardness, and calcium hardness of the water across the membrane system. No
reduction in organic material in the filtrate was observed.
The total suspended solids (TSS) in the backwash waste reached as high as 68 mg/L (during Test
Period 1), while the filtrate TSS remained consistently below the detection limit (1 mg/L).
Table 4-7 presents the mass balance conducted on total suspended solids across the membrane
system. Only the result of December 8, 1999 falls within the calculated range of TSS assuming
either the first or ninth "B" backwash following a "C". The relatively poor predictions of actual
backwash TSS may in part be due to the TSS in the raw water being near the detection limit of
the analysis, 1 mg/L.
25
-------
4.4 Task 4: Reporting Membrane Pore Size
A request was submitted to the membrane Manufacturer to provide the 90 percent and maximum
pore size of the membrane being verified. In their response letter, Ionics stated the 90 percent
pore size is 0.01 um as determined by Field Emission Scanning Electron Microscope. The letter
further stated the 100 percent pore size was 0.04 um as determined by latex.
The above information are taken from a letter supplied by the Ionics which is included in
Appendix A of this report. This is provided for informational purposes only and the results were
not verified during the ETV testing.
4.5 Task 5: Membrane Integrity Testing
Figure 4-12 shows the results of the air pressure-hold tests conducted on the UF membrane
during Test Periods 1 and 2. The air pressure-hold test on the Ionics system was conducted by
pressurizing the feed side of the membrane. If any of the membrane fibers were compromised,
one would expect significant loss of held pressure (> 1 psi every 5 minutes) across the membrane
element. The air pressure-hold tests conducted before and after repairs of broken fibers did not
consistently support this test criteria. During the air pressure-hold test before the first repair
conducted on January 4, 2000, the pressure decay was only 0.6 psi over 10 minutes. While two
bubble points were observed, on March 15, 2000, the air pressure-hold tests before and after
repair showed identical pressure decays of 1.6 psi.
During other fiber breakage incidents, pressure decay slightly over 2 psi over 10 minutes was
observed in the system with compromised fibers. In the air pressure-hold tests conducted before
repairs on March 22, 28 and 31 the pressure decay was slightly greater than 2 psi in 10 minutes.
While there were a number of fiber breakage incidents over the course of testing, the turbidity
profiles shown in Figures 4-5 and 4-6 show consistently low filtrate turbidity levels. Thus,
turbidity monitoring was not useful in detecting 2 or less compromised fibers.
Filtrate particle counts would be expected to noticeably increase if the membrane modules were
compromised (Adham et. al., 1995, Montgomery Watson, 2000). During testing of the Ionics UF
system, particle counting was a reliable method of determining membrane integrity. Every fiber
breakage was displayed in real time on the computer display. Air pressure-hold tests were
conducted after this to verify loss of membrane integrity. The air pressure-hold test was followed
by bubble-point testing to identify leaking fibers.
4.6 Task 6: Data Management
4.6.1 Data Recording
Data were recorded manually on operational and water quality data sheets prepared specifically
for the study. In addition, other data and observations such as the system calibration results were
recorded manually on laboratory and QC notebooks. Data from the particle counters and
26
-------
turbidimeters were also recorded via data acquisition system. All of the raw data sheets are
included in Appendix B of this report.
4.6.2 Data Entry, Validation, and Reduction
Data were first entered from raw data sheets into similarly designed data entry forms in a
spreadsheet. Following data entry, the spreadsheet was printed and checked against handwritten
datasheets. All corrections were noted on the electronic hard copies and then corrected on the
screen. The hardcopy of the electronic data are included in Appendix C of this report.
4.7 Task 7: Quality Assurance/Quality Control (QA/QC)
The objective of this task is to assure the high quality and integrity of all measurements of
operational and water quality parameters during the ETV project. Below is a summary of the
analyses conducted to ensure the correctness of the data.
4.7.1 Data Correctness
Data correctness refers to data quality, for which there are five indicators:
• Representativeness
• Statistical Uncertainty
• Completeness
• Accuracy
• Precision
Calculation of the above data quality indicators were outlined in the Materials and Methods
section. All water quality samples were collected according to the sampling procedures specified
by the EPA/NSF ETV protocols, which ensured the representativeness of the samples. Below is a
summary of the calculated indicators.
4.7.2 Statistical Uncertainty
Ninety-five percent confidence intervals were calculated for the water quality parameters of the
Ionics UF system. These include turbidity, particle count, particle removal, and indigenous
bacteria. Ninety-five percent confidence intervals were presented in summary tables in the
discussion of Task 3 - Finished Water Quality.
4.7.3 Completeness
Data completeness refers to the amount of data collected during the ETV study as compared to
the amount of data that were proposed in the FOD. Calculation of data completeness was made
for onsite water quality measurements, laboratory water quality measurements, and operational
data recording. These calculations are presented in Appendix A of this report. Nearly all
parameters were 100 percent complete. Overall, the database of laboratory water quality data and
27
-------
operational readings was more than 85 percent complete, which met the objective of the ETV
project.
4.7.4 Accuracy
Accuracy is quantified as the percent recovery of a parameter in a sample to which a known
quantity of that parameter was added. An example of an accuracy determination in this ETV is
the analysis of a turbidity proficiency sample and comparison of the measured turbidity to the
known level of turbidity in the sample. Calculation of data accuracy were made to ensure the
accuracy of the onsite desktop turbidimeter used in the study. Accuracy ranged from 102 to 105
percent of the proficiency sample known values. Comparative calibration of online turbidimeters
with the desktop turbidimeters were performed as corrective actions as needed. All accuracy
calculations are presented in Appendix A.
4.7.5 Precision and Relative Percent Deviation
Duplicate water quality samples were analyzed to determine the consistency of sampling and
analysis using relative percent deviation. Calculations of relative percent deviation for duplicate
samples are included in Appendix A of this report. The relative percent deviation for analyses not
near the lower detection limit were within 15 percent for onsite analyses, within 41 percent for
other general water quality analyses, and within 75 percent for microbial analyses. Relative
percent deviation for online and desktop turbidimeter results were also conducted.
4.8 Task 8: Microbial Removal
To demonstrate microbial removal by the Ionics UF system, two seeding experiments with MS2
bacterial virus were conducted. The two seeding experiments were conducted during each test
period, immediately after a membrane cleaning. The clean membrane condition provides worst
case conditions for virus removal (Jacangelo et al. 1995, Montgomery Watson, 1997).
The feed and filtrate concentrations and log removal of virus during this seeding are presented in
Table 4-8 and Figure 4-13. The membrane virus rejection ranged from 4.0 to 5.7 logs for the
seeding conducted at the end of Test Period 1 and from 2.9 to 4.3 logs for the seeding conducted
at the beginning of Test Period 2.
4.9 Additional ETV Project Requirements
4.9.1 Operation and Maintenance (O&M) Manual
The O&M manual for the Ionics UF system supplied by the manufacturer was reviewed during
the ETV testing project. The review comments for the O&M manual are presented in Table 4-9.
The review found the O&M manual to be an extremely useful resource. The manual is very well
organized, well written, clear and complete. The manual makes excellent use of tables and
graphics to organize and clarify the presentation of material. The manual includes a complete set
of manufacturer information sheets for components used on the membrane system.
28
-------
4.9.2 System Efficiency and Chemical Consumption
The efficiency of the small-scale Ionics UF system was calculated based on the electrical usage
and water production of the system. The data are presented in Table 4-10. Overall, an efficiency
of only 20 percent was calculated for the system. This system, however, is significantly more
efficient than many small-scale low pressure membrane systems.
The chemical consumption of the system was also estimated based on the operating criteria used
during the ETV project. Table 4-11 provides a summary of the chemical consumption of the
small-scale Ionics UF system.
4.9.3 Equipment Deficiencies Experienced During the ETV Project
Test Period 1
Ionics UF Membrane System
At the beginning of Test Period 1, frequent problems of inconsistent output from the positive
displacement pump used to chlorinate the backwash feed water were encountered. On December
10, 1999 a pump from a different manufacturer was installed. After this the chlorine feed to the
backwash water was consistent.
On January 3, 2000 at 11:30, the minimum filtrate > 2 um particle counts increased from 0.04/mL
to approximately 2/mL. On January 5, 2000 the top of the pressure vessel was removed and upon
pressurization, 1 bubble point was identified. This leaking fiber end was repaired with a stainless
pin. Filtrate particle counts returned to previous low levels by January 6, 2000.
Online Turbidimeters and Particle Counters
The raw water online turbidimeter failed on December 8, 1999. A spare turbidimeter was used to
record raw water turbidity while this unit was returned to the manufacturer for repair.
On January 1, 2000 the online turbidity and particle count data acquisition software crashed.
Since this occurred over a holiday weekend, it caused an approximately 3 day period where no
online particle and turbidity data were collected.
Test Period 2
Ionics UF Membrane System
The membrane system experienced 3 incidents of broken fibers over the course of Test Period 2.
In all cases, the fiber breakage was identified by visual observation of online filtrate particle count
data.
In the first incident on March 13, 2000, 2 bubble points were identified and repaired. In the
second fiber breakage incident on March 20, 2000, one bubble point was identified and repaired.
After the final incident on March 26, 2000, which identified one bubble point, repairs were not
successful in restoring particle counts to previous levels because of damage to the potting material
around the broken fiber end. A second attempt at repairing this fiber end on March 31, 2000 was
also unsuccessful. Repairs were easily accomplished by inserting stainless steel pins into the
29
-------
leaking fiber ends at the top of the module. The most time-consuming aspect of the repair
procedure was removing the approximately two dozen bolts connecting the pressure vessel cover
to the base of the pressure vessel.
A chronological listing of all problems experienced during ETV testing of the Ionics UF system,
along with their associated corrective actions, is provided in Appendix A of this report.
30
-------
Chapter 5
References
Adham, S.S., J.G. Jacangelo, and J-M. Laine (1995). Low pressure membranes: assessing
integrity, JournalAWWA, 87(3)62-75.
APHA, AWWA and WPCF (1992). Standard Methods for Examination of Water and
Wastewater. 18th ed. Washington, D.C. APHA.
Jacangelo, J.G., S.S. Adham, and J-M. Laine (1995). Mechanism of Cryptosporidium, Giardia,
and MS2 virus removal by MF and UF, Journal AWWA, 87(9)107-121.
Montgomery Watson (1997), Membrane Prequalification Pilot Study. Final Report prepared for
the City of San Diego, October 1997.
Sobsey, M.D., Schwab, K.J., and Handzel, T.R. (1982) A simple membrane filter method to
concentrate and enumerate male-specific RNA coliphages. Jour AWWA, (9): 52-59.
31
-------
Tables and Figures
32
-------
Table 2-1. Characteristics of the Ionics UF 1-7T ultrafiltration membrane.
Units
Value
Membrane Manufacturer
Membrane Model
Commercial Designation
Available Operating Modes
Approximate Size of Membrane Module
Active Membrane Area
Number of Fibers per Module
Number of Modules (Operational)
Inside Diameter of Fiber
Outside Diameter of Fiber
Approximate Length of Fiber
Flow Direction
Nominal Membrane Pore Size
Absolute Membrane Pore Size
Membrane Material/Construction
Membrane Surface Characteristics
Membrane Charge
Design Operating Pressure
Design Flux at Design Pressure
Maximum Transmembrane Pressure
Standard Testing pH
Standard Testing Temperarture
Acceptable Range of Operating pH Values
Maximum Permissible Turbidity
Chlorine/Oxidant Tolerance
ft (m)
ft2 (m2)
inch (mm)
inch (mm)
ft (m)
micron
micron
psi (bar)
gfd (l/hr-m2)
psi (bar)
degF (degC)
NTU
TO RAY
TP-TE07-S
IONICS UF 1-7T
dead-end
3.3 (1.0) length x 0.29 (0.089) diam
130(12)
3,600 (approx.)
7
0.016 (0.40)
0.027 (0.68)
5.2 (1.6)
outside-in
0.01
0.04
PolyAcryloNitrile
Hydrophilic
Neutral
7.3 (0.50)
69 (117)
44 (3.0)
7
77 (25)
2-10 (operating), 1-12 (cleaning)
300 (experience up to)
Chlorine tolerant
33
-------
Table 3-1. Water quality analytical methods.
Parameter
Facility
Standard Method
General Water Quality
PH
Alkalinity
Total Hardness
Calcium Hardness
Temperature
Total Suspended Solids
Total Dissolved Solids
On-Site
Laboratory
Laboratory
Laboratory
On-Site
Laboratory
Laboratory
4500H+
2320 B
2340 C
3500Ca D
2550 B
2540 D
2540 C
Particle Characterization
Turbidity (Bench-Top)
Turbidity (On-Line)
Particle Counts (On-Line)
On-Site
On-Site
On-Site
2130 B
Manufacturer
Manufacturer
Organic Material Characterization
TOC and DOC
UV Absorbance at 254 nm
Laboratory
Laboratory
5310 B
5910 B
Microbiological Analyses
Total Coliform
HPC Bacteria
MS2 Virus
Laboratory
Laboratory
Laboratory
9221 B
9215 B
EPA ICR Method for Coliphage Assay
34
-------
Table 4-1. Ionics UF membrane system operating conditions.
Parameter Unit
Test Period
Run
Start Date & Time
End Date & Time
Run Length
Run Terminating Condition
Filter Cycle Length
Feed Flow
Filtrate Flow
Operating Flux
"B" Backwash Settings
Backwash Cycle Length
Backwash Filtrate Consumed
Fast Flush Feedwater Consumed
Number of "B" backwash before a "C"
"C" Backwash Settings
Backwash Cycle Length
Backwash Filtrate Consumed
Fast Flush Feedwater Consumed
Backwash Chlorine Dose (during reverse flow)
Feed Water Recovery
1 2
1-1 2-1
12/7/99 17:00 3/6/00 15:16
1/11/00 10:32 4/6/00 14:23
days - hrs 34days-18hrs 30days-23hrs
Time Time
min 30 30
gpm (Ipm) 21 (79) 21 (79)
gpm (Ipm) 21 (79) 21 (79)
gfd (L/hr-m2) 33 (57) 33 (57)
sec 60 65
gal (liter) 15(57) 30(113)
gal (liter) 30(114) 7(26)
9 9
sec 130 155
gal (liter) 10(38) 30(113)
gal (liter) 45(170) 45(170)
mg/L 5-10 5-10
% 93% 92%
35
-------
Table 4-2. Evaluation of cleaning efficiency for Ionics UF membrane.
Clean Clean
Number Date
Specific Flux
@20degC
Before Clean
Jsf
gfd/psi
(l/hr-m2-bar)
Specific Flux
@20degC
After Clean
Jsi
gfd/psi
(l/hr-m2-bar)
Recovery of
Specific Flux
Loss of Original
Specific Flux
100(1 - Jsf I Jsi) 100(1-(Jsi I Jsio))
%
%
Start
—
5.1 (126)
—
—
1-1 1/11/00
1.1 (27)
4.8 (120)
78
4.8
2-1 4/7/00
3.2 (80)
4.2 (104)
23
17
Table 4-3. Onsite lab water quality analyses for Ionics UF membrane system.
Standard
Parameter Unit Count Median Range Average Deviation
95 Percent
Confidence
Interval
TEST PERIOD 1
Raw Water
PH
Desktop Turbidity
Temperature
NTU
degC
23
46
46
8.3
1.2
17
8.1 -8.5
1.0-1.5
7.0-21
8.3
1.2
15
0.071
0.12
3.5
8.3-8.3
1.2-1.2
14-16
Filtrate
Desktop Turbidity
NTU
22
0.05
0.05-0.05
0.05
0.00
0.05-0.05
Backwash Waste
Desktop Turbidity
NTU
44
15
9.8-46
17
8.0
15-19
TEST PERIOD 2
Raw Water
PH
Desktop Turbidity
Temperature
NTU
degC
23
46
46
8.3
1.3
19
7.9-8.4
1.0-2.1
11-29
8.2
1.3
19
0.16
0.24
4.0
8.1 -8.3
1.2-1.4
18-20
Filtrate
Desktop Turbidity
NTU
23
0.05
0.05-0.05
0.05
0.00
0.05-0.05
Backwash Waste
Turbidity
NTU
47
9.3
4.8-35
12
7.7
9.8-14
36
-------
Table 4-4. Summary of online particle and turbidity data for Ionics UF membrane system.
95 Percent
Standard Confidence
Parameter Unit Count Median Range Average Deviation Interval
TEST PERIOD 1
Raw Water
Turbidity
ntu
191
1.2
0.95-1.5
1.2
0.12
1.2-1.2
> 2 um Particles
#/mL
186
5200
4000-7200
5300
680
5200 - 5400
2-3 um Particles
#/mL
186
2700
2300 - 3500
2700
300
2700 - 2700
3-5 um Particles
#/mL
186
1600
1200-2300
1700
230
1700-1700
5-15 um Particles
#/mL
186
880
530-1400
900
170
880 - 920
5-7 um Particles
#/mL
186
550
360 - 850
570
100
560 - 580
7-10 um Particles
#/mL
186
240
130 -390
250
49
240 - 260
10-15 um Particles
#/mL
186
85
40-200
85
20
82-88
>15 um Particles
#/mL
186
20
8.7-120
21
8.9
20-22
Filtrate
Turbidity
ntu
186
0.05
0.05-0.10
0.05
0.0037
0.05-0.05
> 2 um Particles
#/mL
186
0.14
0.038 -5.7
0.60
1.2
0.43-0.77
2-3 um Particles
#/mL
186
0.085
0.038 -3.0
0.31
0.61
0.22-0.40
3-5 um Particles
#/mL
186
0.062
0.038 - 1.7
0.19
0.35
0.14-0.24
5-15 um Particles
#/mL
186
0.047
0.038 - 1.9
0.16
0.26
0.12-0.20
5-7 um Particles
#/mL
186
0.043
0.038 -0.63
0.090
0.12
0.073 -0.11
7-10 um Particles
#/mL
186
0.040
0.038 -0.67
0.065
0.077
0.054 -0.076
10-15 um Particles
#/mL
186
0.038
0.038 -0.72
0.051
0.059
0.043 -0.059
>15 um Particles
#/mL
186
0.038
0.038 -0.33
0.043
0.026
0.039 -0.047
Log Removal 2-3 um Particles
33
4.3
3.0-4.8
4.2
0.45
4.0-4.4
Log Removal 3-5 um Particles
33
4.3
3.0-4.6
4.2
0.40
4.1 -4.3
Log Removal 5-15 um Particles
33
4.0
3.0-4.4
3.9
0.40
3.8-4.0
Log Removal 5-7 um Particles
33
4.0
3.0-4.2
3.9
0.30
3.8-4.0
Log Removal 7-10 um Particles
33
3.7
3.0-3.9
3.6
0.24
3.5-3.7
Log Removal 10-15 um Particles
33
3.3
2.8-3.4
3.2
0.18
3.1 -3.3
Log Removal >15 um Particles
33
2.7
2.3-2.9
2.7
0.13
2.7-2.7
37
-------
Table 4-4. Continued.
Standard
95 Percent
Confidence
Parameter
Unit
Count
Median
Range
Average
Deviation
Interval
TEST PERIOD 2
Raw Water
Turbidity
ntu
187
1.2
0.95-2.5
1.3
0.21
1.3-1.3
> 2 um Particles
#/mL
187
4500
2400-7500
4500
920
4400 - 4600
2-3 um Particles
#/mL
187
2400
1400-3500
2400
410
2300 -2500
3-5 um Particles
#/mL
187
1400
690-2400
1400
310
1400-1400
5-15 um Particles
#/mL
187
680
270-1500
680
200
650-710
5-7 um Particles
#/mL
187
440
180 -900
430
120
410-450
7-10 um Particles
#/mL
187
180
66-410
180
59
170-190
10-15 um Particles
#/mL
187
63
32-170
67
22
64-70
>15 um Particles
#/mL
187
17
8.6-60
19
7.8
18-20
Filtrate
Turbidity
ntu
187
0.05
0.05-0.05
0.05
0.00
0.05-0.05
> 2 um Particles
#/mL
187
1.4
0.50-11
2.1
1.6
1.9-2.3
2-3 um Particles
#/mL
187
0.69
0.28-5.4
1.2
0.86
1.1 -1.3
3-5 um Particles
#/mL
187
0.39
0.15-3.4
0.63
0.46
0.56-0.70
5-15 um Particles
#/mL
187
0.28
0.11 -2.3
0.37
0.24
0.34-0.40
5-7 um Particles
#/mL
187
0.14
0.063 - 1.3
0.20
0.14
0.18-0.22
7-10 um Particles
#/mL
187
0.081
0.044-0.62
0.10
0.063
0.091 -0.11
10-15 um Particles
#/mL
187
0.052
0.040 -0.38
0.062
0.031
0.058 -0.066
>15 um Particles
#/mL
187
0.044
0.038 -0.33
0.048
0.024
0.045 -0.051
Log Removal 2-3 um Particles
32
3.5
2.9-3.9
3.4
0.30
3.3-3.5
Log Removal 3-5 um Particles
32
3.5
2.9-3.9
3.4
0.30
3.3-3.5
Log Removal 5-15 um Particles
32
3.3
2.8-3.7
3.3
0.24
3.2-3.4
Log Removal 5-7 um Particles
32
3.4
2.9-3.9
3.4
0.27
3.3-3.5
Log Removal 7-10 um Particles
32
3.3
2.8-3.7
3.3
0.23
3.2-3.4
Log Removal 10-15 um Particles
32
3.0
2.6-3.4
3.0
0.16
2.9-3.1
Log Removal >15 um Particles
32
2.6
2.2-2.9
2.6
0.15
2.5-2.7
38
-------
Table 4-5. Summary of microbial water quality analyses for the Ionics UF membrane system.
Parameter
Unit
95 Percent
Standard Confidence
Count Median Range Average Deviation Interval
TEST PERIOD 1
Raw Water
Total Coliforms
HPC
MPN/100mL 6 18 <2-80 <25
cfu/mL 6 95 2 - 120 83
28
42
2.6-47
49-120
Filtrate
Total Coliforms MPN/100mL 6 <2 <2 - <2 <2 0.00
HPC cfu/mL 6 105 11 - 140 100 46
<2-<2
63-140
Backwash Waste
Total Coliforms
MPN/100mL
40 <2-130 <50
48
12-88
TEST PERIOD 2
Raw Water
Total Coliforms
HPC
MPN/100mL 4 6 <2-17 <7.7 6.7
cfu/mL 5 48 2 - 1400 310 610
1.1 - 14
0-840
Filtrate
Total Coliforms MPN/100mL 4 <2 <2 - <2 <2 0.00
HPC cfu/mL 5 200 48 - 580 200 210
<2-<2
16-380
Backwash Waste
Total Coliforms
MPN/100mL
2-30 <10
12
0-21
Note: All calculations with below detection limit values used the detection limit value in the calculation
as a conservative estimate.
39
-------
Table 4-6. Summary of general water quality analyses for the Ionics UF membrane system.
95 Percent
Standard Confidence
Parameter
Unit
Count
Median
Range
Average
Deviation
Interval
TEST PERIOD 1
Raw Water
Alkalinity
mg/L as CaC03
5
120
120 -130
120
1.1
120 -120
Total Hardness
mg/L as CaC03
5
240
220 - 240
240
11
230 - 250
Calcium Hardness
mg/L as CaC03
4
150
140 -160
150
6.6
140 -160
Total Suspended Solids
mg/L
5
10
1.8-15
8.0
5.5
3.2 -13
Total Dissolved Solids
mg/L
5
500
490 - 500
500
6.4
490-510
TOC
mg/L
5
3.1
2.5 - 4.1
3.2
0.7
2.6 - 3.8
UV254 Unfiltered
/cm
5
0.07
0.06-0.1
0.07
0.02
0.05 - 0.09
UV254 Filtered
/cm
5
0.06
0.05 - 0.06
0.06
0.003
0.06 - 0.06
Filtrate
Alkalinity
mg/L as CaC03
5
120
120 -120
120
0.00
120 -120
Total Hardness
mg/L as CaC03
5
230
220 - 240
230
8.5
220 - 240
Calcium Hardness
mg/L as CaC03
5
150
140 -150
150
5.1
150 -150
Total Suspended Solids
mg/L
5
<1.0
<1.0 - <1.0
<1.0
0.00
<1.0 - <1.0
Total Dissolved Solids
mg/L
5
500
490-510
500
8.7
490-510
TOC
mg/L
5
2.5
2.4 - 2.6
2.5
0.09
2.4 - 2.6
UV254 Unfiltered
/cm
5
0.06
0.05 - 0.06
0.06
0.003
0.06 - 0.06
Backwash Waste
Total Suspended Solids
mg/L
5
25
16-68
32
21
14-50
TEST PERIOD 2
Raw Water
Alkalinity
mg/L as CaC03
6
120
120 -120
120
1.9
120 -120
Total Hardness
mg/L as CaC03
6
220
210-220
220
6.4
210-230
Calcium Hardness
mg/L as CaC03
6
140
130 -140
140
4.7
140 -140
Total Suspended Solids
mg/L
5
5.9
3.9 - 20
9.4
7.0
3.3 -16
Total Dissolved Solids
mg/L
6
470
460 - 480
470
7.4
460 - 480
TOC
mg/L
6
3.6
2.9 - 4.0
3.6
0.4
CO
CO
1
CO
CO
UV254 Unfiltered
/cm
6
0.08
0.07 - 0.09
0.08
0.007
0.07 - 0.09
UV254 Filtered
/cm
6
0.07
0.06 - 0.08
0.07
0.005
0.07 - 0.07
Filtrate
Alkalinity
mg/L as CaC03
5
120
120 -120
120
1.3
120 -120
Total Hardness
mg/L as CaC03
5
220
210-230
220
6.8
210-230
Calcium Hardness
mg/L as CaC03
5
130
130 -140
140
6.4
130 -150
Total Suspended Solids
mg/L
3
<1.0
<1.0 - <1.0
<1.0
0.00
<1.0 - <1.0
Total Dissolved Solids
mg/L
5
470
470 - 480
470
3.9
470 - 470
TOC
mg/L
5
3.8
CO
i
to
CO
3.9
0.3
3.6 - 4.2
UV254 Unfiltered
/cm
5
0.07
0.06 - 0.08
0.07
0.007
0.06 - 0.08
Backwash Waste
Total Suspended Solids
mg/L
4
9.8
1.0-13
8.5
5.4
3.2 -14
Note: All calculations with below detection limit values used the detection limit value in the calculation as a conservative estimate.
All results rounded to two significant digits with the exception of UV254, which is rounded to one significant digit.
40
-------
Table 4-7. Comparison of calculated and measured total suspended solids for Ionics UF membrane
system.
Filtration "B" "B" Measured Measured Calculated Calculated
Date Filtrate Cycle Volume Fast Reverse Raw Backwash Backwash Backwash
Flow Length Filtered Flush Flow TSS TSS TSS TSS
Volume Volume 1st Cycle 9th Cycle
(gpm) (min) (gal) (gal) (gal) (mg/L) (mg/L) (mg/L) (mg/L)
TEST PERIOD 1
12/8/99
21
30
630
30
15
2.8
20.2
14
20
12/14/99
21
30
630
30
15
14.6
25
71
106
12/20/99
21
30
630
30
15
10.4
15.8
51
76
12/27/99
21
30
630
30
15
1.8
67.6
9
13
1/10/00
21
30
630
30
15
10.3
32
50
75
TEST PERIOD 2
3/8/00
21
30
630
6.5
45
12.7
13.4
88
172
3/15/00
21
30
630
6.5
45
3.9
1
27
53
3/22/00
21
30
630
6.5
45
4.2
8.45
29
57
4/5/00
21
30
630
6.5
45
5.9
11.2
41
80
Note: For both Test Periods, 9 "B" backwashes were performed, followed by a "C" backwash.
41
-------
Table 4-8. Feed and filtrate concentrations of MS2 virus for the Ionics UF membrane system.
Seeding #1
Seeding Date: 1/12/00
Specific Flux at 20 degC =
4.9 gfd/psi (119 L/hr-m2-bar)
Feed
Cone.
(pfu/100mL)
Filtrate
Cone.
(pfu/100mL)
Log
Removal
Backwash Waste
Cone.
(pfu/100mL)
7.6E+6
7.4E+6
2.8E+7
1.9E+7
2.8E+7
7.6E+6
7.1 E+2
2.0E+2
2.1 E+2
3.4E+1
5.7E+1
1.7E+2
4.0
4.6
5.1
5.7
5.7
4.7
4.7E+7
5.2E+7
Seeding #2
Seeding Date: 3/6/00
Specific Flux at 20 degC
6.2 gfd/psi (152 L/hr-m2-bar)
Feed
Cone.
(pfu/100mL)
Filtrate
Cone.
(pfu/100mL)
Log
Removal
Backwash Waste
Cone.
(pfu/100mL)
3.6E+7
3.9E+7
6.0E+7
4.0E+7
3.5E+7
4.2E+7
3.1 E+4
3.5E+3
2.9E+3
4.9E+4
4.4E+3
3.1 E+3
3.1
4.0
4.3
2.9
3.9
4.1
2.7E+8
1.2E+8
42
-------
Table 4-9. Review of manufacturer's operations and maintenance manual for the Ionics UF
membrane system.
O & M Manual Grade Comment
Overall Organization + • The O&M manual is very well organized. The
table of contents includes the following main
sections: Introduction, Safety Procedures,
Equipment Description, Unit Installation, Start-up
and Shut Down Procedures, Operation
Instructions, Maintenance and Repair and
Troubleshooting.
• The manual also includes the following
appendices: Definition of terms and
abbreviations, calculations, consumable material
information, system drawings, PLC program, bill
of materials and complete manufacturers
literature for every purchased component on the
system.
Operations Sections + • Includes start up procedures section describing
position of all manual valves during system
operation. Includes detailed step by step
instruction for both initial start up and normal start
up after brief shutdown. Initial startup includes
section detailing preliminary checks which should
be made before start up.
• Shut down procedures sections include normal
shutdown for events such as maintenance or long
term storage, and emergency shutdown
procedures. Also includes section on long term
shutdown of unit.
• Operations section includes operations
constraints section listing feed requirements,
operating limits including operational feed
pressure, TMP, pH range, and air scrub pressure.
Another section describes both "B" and "C"
backwash sequences, and finally a section
describes the integrated integrity test which can
be performed automatically as part of the "C"
backwash sequence.
• The operations section also includes sections
on alarms, control logic with tables showing
position of all automatic valves during each phase
of filtration and backwash modes, operator
interface section with detailed descriptions of all
screens on the Allen Bradley PLC.
• The operations sections are extremely well
organized and make excellent use of tables and
graphics.
Maintenance Section + • Includes sections detailing daily, weekly,
monthly and quarterly, yearly and 5-year
maintenance checks.
• Maintenance sections discussed include UF
Clean-ln-Place and loading UF elements.
43
-------
Table 4-9. Continued.
O & M Manual
Grade Comment
Troubleshooting
Ancillary Equipment Information +
Drawings and Schematics +
Use of Tables +
OVERALL COMMENT +
• Manual does not include a troubleshooting
section with description of all alarm conditions
and tables for each alarm condition detailing
possible causes and solutions.
• Equipment manufacturers' literature included
as an appendix for all system components.
• Overall makes good use of drawings and
schematics.
• Should include process schematics showing
water flow during filtration and backwash.
• Includes schematics of the Allen Bradley
PanelView display and all associated screens.
• Manual makes very good use of tables to
organize and present information.
• An excellent O&M manual. It is very well
organized, well written, clear and complete. An
excellent table of contents makes locating
information in the manual a simple process.
• The manual includes a good use of graphics to
assist the reader's understanding.
• The manual includes as an appendix a list of
components used on the Ionics UF unit, such as
pumps, flow meters, valves and pressure gauges
including manufacturer and model number.
44
-------
Table 4-10. Efficiency of the Ionics UF membrane system.
Parameter Unit
Value
ELECTRICAL USE
Voltage Volt - three phase 460
Feed Pump Current Amp 0.5
Feed Pump Power Watt 420
WATER PRODUCTION
Transmembrane Pressure psi 9.3
pascal 6.4E+04
Flow Rate gpm 21
m3/s 1.3E-03
Power Watt 85
EFFICIENCY % 20%
Table 4-11. Chemical consumption for the Ionics UF membrane system.
Unit Value
Backwash Chlorine [1]
Average Chlorine Dose mg/L 8.7
Stock Chlorine Concentration % 10
Average Backwash Volume gal (L) 29(110)
Chlorine Stock Volume per Backwash mL 9.6
Backpulse Per Day # 48
Stock Chlorine Use Per Day gal (L) 0.12(0.46)
Cleaning Chemicals [2]
Citric Acid 2% lb (kg) 17(7.7)
RO Clean L211 gal (L) 1.8(7.0)
Acid Rinse pH 2 (Hydrochloric Acid) gal (L) 0.053(0.20)
111 Based on average chlorine dose and average backwash volume
121 Chemical use per cleaning
45
-------
Project Manager
SamerAdham, Ph.D.
Montgomery Watson
Figure 1-1. Organizational chart showing lines of communication.
Figure 2-1. Photographs of ETV test unit.
#:::::::
46
-------
PLAN VIEW
-28.0 in.-
-22.0 in.-
AIR
COMPRESSOR
FEED/
CHEMICAL
FILTRATE
CLEAN-IN-
STORAGE
PLACE
TANK
TANK
T
28
.0
in.
1
Back-
wash
Pump
480V Electrical
Feed Pump
Chlorine
Tank/
Pump
110V Electrical
-105.0 in.-
SIDE VIEW
AIR
COMPRESSOR
Differential
Pressure
O ,
Feed 54
"l|emper-.0
ature jp
Feed
Pump
480V Electrical
FEED/
CHEMICAL
CLEAN-IN-
PLACE
TANK
Filtrate
Rotameter
-28.0 in.-
FILTRATE
STORAGE
TANK
Feed
Pressure
Backwash
Pump
110V Electrical
O
Hour
Meter
All^n.BradLey. ¦
ParielView Display
® o
System Manual
Control BW
o
Emergency
Stop
o o
Integrity Alarm
Test
Filtrate
Flow
Magmeter
o
Filtrate
Pressure
"\
7 MODULE
MEMBRANE
PRESSURE
VESSEL
5b
.0
in.
-18.0 in.-
Jt
Chlorine
Tank
and
Pump
-105.0 in.
Figure 2-2. Spatial requirements for the Ionics UF unit.
47
-------
~
Waste
To
Drain
Air Scour
Figure 2-3. Schematic diagram of the Ionics UF membrane process.
Conference Room & Office
Trailer
Tertiary Filter
Blower
Building
Figure not drawn to scale.
Figure 3-1. Schematic of Aqua 2000 Research Center test site.
48
-------
o
o
5 <3
.& in
e
= _l
J2 o)
I!
in o
m _i
| E
£ E
350
300
250
200
150
100
50
0
Hardness, Alkalinity and Calcium
Hardness
Alkalinity
-A 1 *
Calcium
Nov-97
Jan-98
Mar-98
May-98
Month
Jul-98
Sep-98
Nov-98
700-
600-
5" 500-
u>
E. 400-
Q 300-
I-
200-
100-
o-
Nov-97 Jan-98 Mar-98 May-98 Jul-98 Sep-98 Nov-98
Month
Total Dissolved Solids
PH
Nov-97 Jan-98 Mar-98 May-98 Jul-98 Sep-98 Nov-98
Month
Figure 3-2. Lake Skinner raw water quality.
49
-------
¦g
!q
Nov-97
Turbidity
Jan-98
Mar-98
May-98
Month
J u 1-98
Sep-98
Nov-98
Nov-97
Temperature
Jan-98
Mar-98
May-98
Month
J u 1-98
Sep-98
Nov-98
TOC
3.50
3.00
2.50
U)
b
2.00
o
o
1.50
1-
1.00
0.50
0.00
Nov-97
Jan-98
Mar-98
May-98
Month
J u 1-98
Sep-98
Nov-98
Figure 3-3. Lake Skinner raw water quality.
50
-------
• Raw
- Filtrate
(n
o
ro
a.
10000
1000
100
10
1
KHM-D-O-O-O-n
O I I I I I I I I
2-3 um Particles
4/13/00 4/13/00 4/13/00 4/13/00
13:48 13:55 14:02 Time 14:09
4/13/00 4/13/00 4/13/00
14:16 14:24 14:31
-Raw
H— Filtrate
(n
o
ro
o.
10000
1000
100
10
1
o-o-o-o-o-o-oo-o
3-5 um Particle?
o I I I I I I I I
4/13/00 4/13/00 4/13/00 4/13/00
13:48 13:55 14:02 Time 14:09
4/13/00 4/13/00 4/13/00
14:16 14:24 14:31
-Raw
H— Filtrate
(n
o
cz
a.
10000
1000
100
10
1
5-15 um Particles
O t I 4 I I I I I
4/13/00 4/13/00 4/13/00 4/13/00
13:48 13:55 14:02 Timei4:09
4/13/00 4/13/00 4/13/00
14:16 14:24 14:31
(n
o
o.
10000
1000
100
10
1
• Raw
H— Filtrate
>15 um Particles
4/13/00
13:48
¦•-MMFOif-cPQ-o-n"
4/13/00 4/13/00 4/13/00
13:55 14:02 Time 14:09
4/13/00 4/13/00 4/13/00
14:16 14:24 14:31
Figure 3-4. Response of online particle counters to Duke Monosphere solution.
51
-------
Year
1999
2000
Month
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Task 1:
Membrane Flux &
Recovery
_
_
_
_
Task 2:
Cleaning Efficiency
1
_
_
Task3:
Finished Water Quality
1
Task 4:
Reporting of Membrane
Pore Size
,
]
Task5:
Membrane Integrity
1
Task 6:
Data Management
1
Task 7:
QA/QC
1
in
1
Task 8:
Microbial Removal
~
~
Figure 3-5. Membrane verification testing schedule.
52
-------
-Transmembrane Pressure
-Temperature
40
35
30
to
Q.
3 25
ai
c
ns
5
E
at
E
20
15
2 10
0
12/06/99
0 hrs
40
35
30
25
20
O
15
o
of
3
10
TO
aJ
5
a
E
-------
- Flux@20C
-Specific Flux@20C
50
40
1) 30
x
= 20
10
Target Flux = 33 gfd
to
5 5-
2
U)
6
4 °
x
3 =
2 a>
^ a.
-------
A Clean Membrane: Start of Test Period 1
100 -
90 -
80 -
70 -
jo"
n 60 -
O
k 50-
©
| 40 -
u_
30 -
20 -
10 -
0 -
0 2 4 6 8 10 12 14 16 18 20
Transmembrane Pressure (psi)
A - Clean membrane: start of Test Period 1.
~ Before Cleaning I After Chemical 1 A After Chemical 2
100
80 -
70 -
jo"
S
p
k 50 -
©
3 40 -
20 -
10 -
0 10 15 25 30 40
Transmembrane Pressure (psi)
B - Test Period 1: cleaning 1-1 (1/11/00).
Figure 4-3. Clean water flux profile during membrane chemical cleanings - Test Period 1.
Clean membrane:
Start of Test Period 1
RO Clean L211
y = 4.8x + 2.9
/
Before Cleaning
/
y = 1.1x + 7.7
/
Citric Acid
y = 1.1x + 5.8
55
-------
A Clean Membrane: Start of Test Period 2
~ Before Cleaning O After Chemical 1 A
90 -
70 -
jo"
3 60 -
o
o
O
CN|
©
| 40 -
u_
30 -
10 -
0 -
5 10 20 25 35 40
B - Test Period 2: cleaning 2-1 (4/7/00).
Figure 4-4. Clean water flux profile during membrane chemical cleanings - Test Period 2.
After Chemical 1
Citric Acid
56
-------
Raw Online Turbidity
A Filtrate Desktop Turbidity
O Raw Desktop Turbidity
10
£ 1
~ ~ D
dPdd
-6-6-6 6-6-66-6 iWHW 66—6-6 6-i
Note: - Online values averaged over 4 hour period
- Turbidity values rounded to nearest 0.05 NTU
0.001
12/4/99
0 hrs
12/11/99
168 hrs
12/18/99
336 hrs
12/25/99
Time
1/1/00 1/8/00
672 hrs 840 hrs
1/15/00
1008 hrs
Figure 4-5. Turbidity profile for raw water and Ionics UF membrane system - Test Period 1.
1000
Raw Online Turbidity
Ionics Filtrate Online Turbidity
o Raw Desktop Turbidity
A Filtrate Desktop Turbidity
~ Backwash Waste Desktop Turbidity
100
~ ~
tng-^-o crn-
0.1
0.01
-6-6-6-6 6-6-6-6-6 6-6-6-6-6 6-6-6-6-6 6-6-6-6
Note: - Online values averaged over 4 hour period
- Turbidity values rounded to nearest 0.05 NTU
0.001 H r-
3/5/00
0 hrs
3/12/00
168 hrs
3/19/00 3/26/00
336 hrs Time 504 hrs
4/2/00
672 hrs
4/9/00
840 hrs
Figure 4-6. Turbidity profile for raw water and Ionics UF membrane system - Test Period 2.
57
-------
12/5/99 12/12/99 12/26/99 1/2/00
168hrs 336 hrs Time
840 hrs
100000
¦jj 1000
r 100
c
3
o
o
0)
o
t
ro
Q.
0.001
12/5/99
0 hrs
- Raw Water Particles
- Ionics Filtrate Particles
3-5 um Particles
168 hrs
12/19/99
336 hrs
12/26/99
Time
672 hrs
1/9/00
840 hrs
Note:
100000
10000
E
8
*—l
c
3
o
o
a;
o
t
ro
Q.
0.001
12/5/99
0 hrs
12/12/99
168 hrs
12/19/99
336 hrs
12/26/99
Time
1/2/00
672 hrs
1/9/00
840 hrs
Gap in data at 12/25/99 due to electrical outage ot particle counters.
Gap in data at 1/1/00 due to Y2K software failure.
Figure 4-7. Particle counts for raw water and Ionics filtrate - Test Period 1.
58
-------
100000
10000
E
s
+¦»
c
3
o
o
CJ
o
TO
a.
E
c
3
o
o
a;
o
t
ro
Q.
0 hrs
100000
10000
100
10
0.1
0.01
12/12/99
168 hrs
12/19/99
1/2/00
672 hrs
1/9/00
840 hrs
- Raw Water Particles
- Ionics Filtrate Particles
0 hrs
12/12/99
168 hrs
12/19/99
1/2/00
672 hrs
1/9/00
840 hrs
Note:
E
¦+->
C
3
o
o
CJ
o
TO
a.
100000
10000
100
10
0.1
0.01
- Raw Water Particles
i.
0 hrs
12/12/99
168 hrs
12/19/99
- Ionics Filtrate Particles
1/2/00 1/9/00
672 hrs 840 hrs
Gap in data at 12/25/99 due to electrical outage ot particle counters.
Gap in data at 1/1/00 due to Y2K software failure.
Figure 4-7. Continued.
59
-------
E
c
3
o
o
o
o
t
ro
Q.
100000
10000
100
10
- Raw Water Particles
- Ionics Filtrate Particles
0.1
0.01
0 hrs
2-3 um Particles
3/12/00
168 hrs
3/19/00
336 hrs Time
504 hrs
4/2/00
672 hrs
E
8
c
3
o
o
a;
o
t
ro
Q.
100000
10000
1000
100
10
1
0.1
0.01
0.001
3/5/00
0 hrs
- Raw Water Particles
• Ionics Filtrate Particles
3-5 um Particles
3/12/00
168 hrs
3/19/00
336 hrs Time
3/26/00
504 hrs
4/2/00
672 hrs
100000
10000
^ 1000
* 100
£
3
o
o
a;
o
t
ro
Q.
10
0.1
0.01
0.001
- Raw Water Particles
- Ionics Filtrate Particles
5-7 um Particles
3/5/00 3/12/00 3/19/00 3/26/00 4/2/00
0 hrs 168 hrs 336 hrs Time 504 hrs 672 hrs
Figure 4-8. Particle counts for raw water and Ionics filtrate - Test Period 2.
60
-------
E
+¦»
c
3
o
o
CJ
o
TO
a.
E
+¦»
c
3
o
o
CJ
o
TO
a.
E
tt
*—i
c
3
o
o
a;
o
t
ro
0.
100000
10000
1000
0.001
3/5/00 3/12/00 3/19/00 3/26/00 4/2/00
0 hrs 168hrs 336hrsjime 504 hrs 672 hrs
100000
10000
0.001
3/5/00
0 hrs
100000
10000
1000
100
10
1
0.1
0.01
0.001
3/5/00
0 hrs
• Raw Water Particles
- Ionics Filtrate Particles
3/12/00 3/19/00 3/26/00 4/2/00
168 hrs 336 hrs Time 504 hrs 672 hrs
- Raw Water Particles
- Ionics Filtrate Particles
>15 um Particles
3/12/00
168 hrs
3/19/00
336 hrs Time
3/26/00
504 hrs
4/2/00
672 hrs
Figure 4-8. Continued.
61
-------
ro
>
o
E
at
a.
O)
o
- 2
0)
o
t
ro
Q.
12/4/99
0 hrs
2-3 um Particles
~-C-CKjXI
12/11/99
168 hrs
12/18/99
336 hrs
12/25/99
Time
1/1/00
672 hrs
1/8/00
840 hrs
1/15/00
1008 hrs
a;
o
t
ro
0.
3-5 um Particles
12/4/99 12/11/99 12/18/99 12/25/99 1/1/00 1/8/00 1/15/00
0 hrs 168 hrs 336 hrs Time 672 hrs 840 hrs 1008 hrs
Note:
ra
>
o
E
at
a.
O)
o
— 9
o ^
12/4/99
0 hrs
5-7 um Particles
n-D-o-u-n
12/11/99
168 hrs
12/18/99
336 hrs
12/25/99
Time
1/1/00
672 hrs
1/8/00
840 hrs
Gap in data at 12/25/99 due to electrical outage ot particle counters.
Gap in data at 1/1/00 due to Y2K software failure.
1/15/00
1008 hrs
Figure 4-9. Particle removal for Ionics UF membrane system - Test Period 1.
62
-------
7-10 um Particles
>
o
E
0)
a:
O)
o
- 2
t\ ^
o
o
TO
a.
o-n-ckQj:
y~y
12/4/99
0 hrs
12/11/99
168 hrs
12/18/99
336 hrs
12/25/99
Time
1/1/00
672 hrs
1/8/00
840 hrs
1/15/00
1008 hrs
10-15 um Particles
ra
o 4
E
at
a.
a) 3
o
— 9
o ^
~-n-ckQ^i
12/4/99
0 hrs
12/11/99
168 hrs
12/18/99
336 hrs
12/25/99
Time
1/1/00
672 hrs
1/8/00
840 hrs
1/15/00
1008 hrs
>15 um Particles
ra
>
o
E
at
a.
O)
o
— 9
o ^
Note:
12/4/99
0 hrs
12/11/99
168 hrs
12/18/99
336 hrs
12/25/99
Time
1/1/00
672 hrs
1/8/00
840 hrs
Gap in data at 12/25/99 due to electrical outage ot particle counters.
Gap in data at 1/1/00 due to Y2K software failure.
1/15/00
1008 hrs
Figure 4-9. Continued.
63
-------
3/4/00 3/11/00 3/18/00 3/25/00 4/1/00 4/8/00
0 hrs 168 hrs 336 hrs Time 504 hrs 672 hrs 840 hrs
3/4/00 3/11/00 3/18/00 3/25/00 4/1/00 4/8/00
0 hrs 168 hrs 336 hrs Time 504 hrs 672 hrs 840 hrs
ra
>
o
E
at
a.
O)
o
— 9
o ^
0
3/4/00
0 hrs
3/11/00
168 hrs
3/18/00
336 hrs
Time
3/25/00
504 hrs
5-7 um Particles
4/1/00
672 hrs
4/8/00
840 hrs
Figure 4-10. Particle removal for Ionics UF membrane system - Test Period 2.
64
-------
ro
>
o
E
at
a.
O)
o
0)
o
t
ro
Q.
- 2
1
7-10 um Particles
0
3/4/00
0 hrs
3/11/00
168 hrs
3/18/00
336 hrs
Time
3/25/00
504 hrs
4/1/00
672 hrs
4/8/00
840 hrs
ra
>
o
E
at
a.
O)
o
— 9
o ^
1
10-15 um Particles
0
3/4/00
0 hrs
3/11/00
168 hrs
3/18/00
336 hrs
Time
3/25/00
504 hrs
4/1/00
672 hrs
4/8/00
840 hrs
ra
>
o
E
at
a.
ut
o
— 9
o ^
>15 um Particles
0
3/4/00
0 hrs
3/11/00
168 hrs
3/18/00
336 hrs
Time
3/25/00
504 hrs
4/1/00
672 hrs
4/8/00
840 hrs
Figure 4-10. Continued.
65
-------
.01
J I l_l I l_l I l_
5 10 20 30 50 70 80 90 95
99
_l
99.9 99.99
Removal of 3-5um Particles
(0
o
o
t
(5
a.
w
¦
CO
15
>
o
E
-------
¦0—Start of Test Period 1 —0—1 Broken Fiber (1/4/00) —A—End of Test Period 1
10 11 12
¦H— Start of Test Period 2 (no bubbles)
Time, minutes
Figure 4-12. Air pressure hold test results for the Ionics UF membrane system.
67
-------
¦H—2 bubble points (3/15/00) A After repair (3/15/00)
Time, minutes
¦a— 1 bubble point (3/22/00) A After repair (3/22/00)
Time, minutes
Figure 4-12. Continued.
68
-------
¦a—1 bubble point (3/28/00)
-A—After repair (3/28/00)
Time, minutes
¦0—1 bubble point (3/31/00) A After repair (3/31/00)
Time, minutes
Figure 4-12. Continued.
69
-------
o
0: 0.6
o.
0.4
0.2
-A—End of Test Period 2 (1 bubble point)
Test Period 2
0.0
0123456789
Time, minutes
10 11 12
Figure 4-12. Continued.
Seeding:
Date:
Seeding 1
1/12/00
Specific Flux: 4.9 gfd/psi @ 20°C (119 L/hr m bar)
_ 4
rc
>
o
h
O)
o
Seeding 2
3/6/00
6.2 gfd/psi @ 20°C (152 L/hr m2 bar)
3 4
Sample
Seeding 2
17
i n
5 6
Figure 4-13. Log removal of seeded MS2 virus by Ionics UF membrane system.
70
-------
Appendix A
-------
Appendix B
Raw Data Sheets
-------
Appendix C
Hardcopy Electronic Data
------- |