&EPA
EPA/600/R-12/644 | December 2012 | www.epa.gov/gateway/science
   United States
   Environmental Protection
   Agency
   Evaluation of Radial  Flow Fluidized Filter (R3F)
   Followed  by Microfiltration and Ultrafiltration
   Systems in Calimesa, California
                            <.

                            '#

    •'
                                   •»
     Office of Research and Development

-------
  EVALUATION OF RADIAL FLOW FLUIDIZED FILTER (R3F)
FOLLOWED BY MICROFILTRATION AND ULTRAFILTRATON
              SYSTEMS IN CALIMESA, CALIFORNIA
                            Craig L. Patterson, P.E.
                   Water Supply and Water Resources Division
                  National Risk Management Research Laboratory
                            Cincinnati, Ohio 45268

                              Rajib Sinha, P.E.
                  Shaw Environmental & Infrastructure (E&I), Inc.
                            Cincinnati, Ohio 45212
This report was compiled in cooperation with Shaw Environmental & Infrastructure (E&I), Inc.
         Under EPA Contract EP-C-09-041, Work Assignment No. 0-03, Task 2

                                Submitted to

                  National Risk Management Research Laboratory
                      Office of Research and Development
                     U.S. Environmental Protection Agency
                            Cincinnati, OH 45268

                                    and

                          South Mesa Water Company
                              391 W. Avenue L
                             Calimesa, CA 92320
                             September 18, 2012

-------
                                        Notice

 The U.S. Environmental Protection Agency, through its Office of Research and Development,
funded and managed, or partially funded and collaborated in, the research described herein. It
 has been subjected to the Agency's peer and administrative review and has been approved for
publication. Any opinions expressed in this report are those of the  author  (s) and do not
 necessarily reflect the views of the Agency, therefore, no official endorsement should be inferred.
Any mention of trade names or  commercial products does not constitute endorsement or
 recommendation for use.

-------
                                Acknowledgements

EPA acknowledges the contributions from George Jorritsma of the South Mesa Water Company.
His efforts included coordinating and managing the research studies in Calimesa, California.
EPA acknowledges the in-kind support and contributions of John Martin with R3f Filtration
(Email: martinr3f@yahoo.com) and Kenneth W. Clark with NOK Corporation (Contact: Mr. Y.
Inaba, Email: yosinaba@nok.co.jp, Tel: +81-3-3432-8415). Mr. Martin provided the R3f pilot-
unit and Mr. Clark provided the MF and UF housings and membranes and participated in the
research studies in Calimesa, CA.

                                EPA Contributors

Dr. John Ireland served as Project Officer on EPA Contract No. EP-C-04-034 and Mr. Craig L.
Patterson P.E., served as the Work Assignment Manager for this research project.  Mr. Jeffrey Q.
Adams and Dr. Joel Allen performed technical reviews of the document.  Mr. John Olszewski
was the EPA Quality Assurance Managers, and was responsible for the quality assurance review
of the document.  Dr. Bruce Macler with U.S. EPA Region 9  in  San Francisco, California
reviewed the experimental plan for this research study.

-------
                                       Abstract

U.S. EPA  coordinated a  field study with South Mesa Water Utility to look for treatment
alternatives for California  State Project Water in the small  community of Calimesa, California.
EPA evaluated the performance of a system comprised of Radial Flow Fluidized Filtration (R3f)
followed by  microfiltration (MF)  and ultrafiltration (UF) through a series of turbidity and
microbial challenges. The  R3f and MF-UF system was challenged to remove turbidity, particles,
and bacteria  and viral surrogates (B.  Subtilis, MS-2)  from State Project Water.   Turbidity,
particle counts and Total  Organic  Carbon (TOC) were  also  monitored  to  obtain  additional
information on the  performance and maintenance requirements of the systems.   The report
documents  the results of these tests and also summarizes the  critical  observations on  the
maintenance of the R3f-MF-UF system and the use of turbidity and particle count as a surrogate
for measuring the performance of the treatment plant.

-------
                                       Contents

Notice	ii
Acknowledgements	iii
Abstract	iv
Contents	v
List of Figures	vi
List of Tables	vii
Acronyms and Abbreviation	viii

1.0  Introduction	1

2.0  System Description, Operation and Testing Procedures	3
     2.1  Field Test Site	3
     2.2  Radial Flow Fluidized Filter (R3f) with MF and UF Membrane Polishing Filters	3
     2.3  System Operation and Test Conditions	4
          2.3.1  Microbial and Turbidity Challenges	4
          2.3.2  Sampling and Analytical Procedures	5

3.0  Test Results	6
     3.1  MS2 bacteriophage Test Results	6
     3.2  B. subtilis Test Results	8
     3.3  Turbidity, Particle Count and TOC Results	10
          3.3.1  Turbidity Results	10
          3.3.2  Particle Count Results	11
          3.3.3  Pressure Buildup during the Challenge Tests	16
          3.3.4  TOC Results	18

4.0  Conclusions	20

5.0  References	21

-------
                                    List of Figures


Figure 1. Conceptual Flow Diagram of the R3f System	3
Figure 2. Typical R3f Filter Setup	3
Figure 3. MF Filters Placed in a Shelco Multi-Cartridge Housing Unit	4
Figure 4. Correlation between Turbidity and 1.2 jim Size Particle	15
Figure 5. Correlation between Turbidity and 2  - 5 jim Size Particles	16

-------
                                   List of Tables

Table 1. Summary of the Sampling Strategy	5
Table 2. Analytical Methods For Physical and Microbial Parameters in this Study	5
Table 3. Summary of Challenge Tests Conducted on R3f-MF-UF System	6
Table 4. Results of MS2 bacteriophage Challenges on the R3f-MF-UF System	2
Table 5. Summary of R3f-MF-UF System Performance in Removing MS2 bacteriophage	8
Table 6. Results of B.  subtilis Challenges on the R3f-MF-UF System	9
Table 7. Summary of the R3f-MF-UF System Performance in Removing B. subtilis	10
TableS. Turbidity Results for the R3f-MF-UF System	11
Table 9. Summary of the R3f-MF-UF System Performance in Removing Turbidity	11
Table 10. Results of 1.2 |im Size Particle Count Monitoring for the R3f-MF-UF System	13
Table 11. Results of 2  - 5 |im Size Particle Count Monitoring for the R3f-MF-UF System	14
Table 12. Summary of the R3f-MF-UF System Performance in Removing Particles	14
Table 13. Pressure Buildup in R3f System during the Challenge Tests	17
Table 14. Pressure Buildup in the MF and UF Units during the Challenge Tests	18
Table 15. Results of TOC Monitoring for the R3f-MF-UF System	19

-------
                           Acronyms and Abbreviations
B. subtilis    Bacillus subtilis
E. coli       Escherichia coli
gpm         gallons per minute
L            liter
LRV         log removal value
LT2ESWTR  Long Term 2 Enhanced Surface Water Treatment Rule
mg          milligram
mL          milliliter
MF          microfiltration (microfilter)
MS2         MS2 bacteriophage
NTU         Nephelometric Turbidity Unit
psi          pounds per square inch
PSL         polystyrene latex
QA          quality assurance
QAPP       Quality Assurance Proj ect Plan
QC          quality control
R3f         Radial Flow Fluidized Filter
Shaw        Shaw Environmental & Infrastructure (E&I), Inc.
SMWC      South Mesa Water Company
Std. Dev.     standard deviation
T&E         Test and Evaluation
TOC         total organic carbon
UF          ultrafiltration (ultrafilter)
ug          microgram
jam          microns

-------
1.0    Introduction
The  U.S.  Environmental Protection Agency  (EPA) evaluated the performance of a system
comprised of Radial Flow Fluidized Filtration (R3f)  followed by Microfiltration  (MF) and
Ultrafiltration (UF) through a series of turbidity and microbial  challenges.   The tests were
performed at the EPA Test and Evaluation (T & E) Facility in Cincinnati, Ohio and at the  South
Mesa Water Company (SMWC) site in Calimesa, CA. These tests were performed by EPA and
Shaw Environmental & Infrastructure (E&I), Inc.  (Shaw) under Contract No. EP-C-09-041,
Work Assignment No. 0-03 with the EPA and under a separate contract with SMWC.

The R3f and MF-UF system was challenged with the following contaminants:

   •   Escherichia coli (E. coli) as a representative organism for bacterium
   •   Bacillus subtilis (B. subtilis) as a representative organism for aerobic spores
   •   MS2 bactedophage, a surrogate for enteric virus
   •   Polystyrene Latex (PSL) beads as a surrogate of Cryptosporidium
   •   Cryptosporidium. One test was conducted on the R3f system only using
       Cryptosporidium to confirm the surrogacy equivalence of PSL beads.

SMWC had previously evaluated the use of the  R3f system as a treatment device for a surface
water source. The R3f system had successfully produced water that met the required turbidity
standard of 0.30 Nephelometric Turbidity Unit (NTU). However, California required SMWC to
achieve at least a  1-log removal of viruses from the filtration step.  Since the media size  in the
R3f  system (33 jim) is too large for removal of bacteria and viruses, a MF  and UF system
downstream of the R3f system was considered for inclusion for microbial removal.

EPA had evaluated the R3f system at the T&E  facility and  then had evaluated an R3f system
followed by a MF and UF system at a field site in Ely, Minnesota and the results showed nearly
complete removal of bacteria and viruses.  The  same treatment system was shipped to SMWC
and tested in this study to confirm the removal of bacteria and viruses in this water matrix as well
as to evaluate the likely cartridge backwash and changeout frequency.  Turbidity, particle counts
and Total  Organic Carbon (TOC) were also monitored to obtain additional information on the
performance and maintenance requirements of the systems. This report documents the results of
these tests and also summarizes the critical observations on the maintenance of the R3f-MF-UF
system and the use of turbidity and particle count as surrogate for measuring the performance of
the treatment plant.

-------
2.0    System Description, Operation and Testing Procedures	
This section provides a summary  description  of the systems evaluated and the associated
operation and testing procedures employed at the field site and at the T&E Facility in Cincinnati,
OH.   The test procedures employed are presented in the following  EPA-endorsed  Quality
Assurance Project Plans (QAPPs):
   •   EPA QA ID No. 627-Q-10-0 (EPA and Shaw, 2007a)
   •   Amendment 1 to EPA QA ID No. 627-Q-10-0 (EPA and Shaw, 2008)

       2.1 Field Test Site
The field test site was located at SMWC in Calimesa, California. The tested units were installed
in the test trailer owned and operated by SMWC.

       2.2 Radial Flow Fluidized Filter (R3f) with MF and UF Membrane Polishing Filters
The radial flow fluidized filtration (R3f) system utilizes radial flow (similar to the flow pattern of
a cartridge filter) through use  of non-bonded garnet media that can be fluidized and backwashed.
The technology uses fine garnet media (33 microns)  for depth filtration.  The  filter system
housing consists of an upper chamber and a lower chamber.  The upper chamber provides the
necessary head for flow through the media and for backwashing. Water to be treated enters the
system through the annular section at the top of the inner core.  The lower chamber contains the
media and water travels radially from the annular section through the media to the bottom inner
core as a plug flow. The treated water then flows out of the system through the pipe connected at
the bottom of the inner core. The plug in the inner core prevents any reverse flow. A typical R3f
system is 150 mm (6 inches) in  diameter and 1.8 m (6 feet) high as shown in Figure  1.  The
media  can be fluidized  and backwashed  very  quickly  resulting in  backwash  volumes
significantly lower than other technologies.  The R3f system is suitable for  small community
potable water systems. It can also serve as pre-filter to membranes or disposable final filtration
system. The R3f system is comprised of one lead and one lag unit in series.  A typical R3f
system setup is shown in Figure 2.

-------
                                      Filtration
                                        Cycle
v-


t

•


'

i


\
&
^
J^^-™
•
*
*
*
I
S~TL~
"***^__ _
ii :
1
• !
i
i
1 1

^^
C- ' -3
> *

# «,

* «
j;


H.






^^zl.
:
*
t
*
1
""jL^
__ __*-!**'
m
i i
i y
1 1

«

^^
^
-.>

•

-:>


;

>


1
^
^-
                     Figure 1. Conceptual Flow Diagram of the R3f System
Lag Filter
                                                                                 Inlet
                                                                                 Lead Filter
                                                                                Control Panel
 Effluent
                              Figure 2. Typical R3f Filter Setup
  The UF and MF polishing filters were tested by placing them in a 7-cartridge Shelco Multifilter
  Cartridge Housing (Figure 3).  Two Shelco housings were placed in series - the first housing
  contained MF  cartridges while the second housing contained UF  cartridges.  The system was
  configured so that both Shelco housing units could be backwashed at the same frequency as the
  R3f system.

-------
            Figure 3. MF Filters Placed in a Shelco Multi-Cartridge Housing Unit

       2.3 System Operation and Test Conditions

          2.3.1 Microbial and Turbidity Challenges
Tests were carried out to evaluate the performance of the R3f-MF-UF systems in removing
bacteria and virus using the following test matrix, organisms or surrogates:

   •   For bacteria removal, the systems were challenged with B. subtilis endospores,

   •   For evaluation  of virus  removal  potential,  the systems  were  challenged  with  MS2
       bacteriophage, a surrogate for enteric viruses.

   •   For evaluation of turbidity removal performance, the systems were operated using surface
       water from the lake as the feed source with as-received turbidity levels.  Samples were
       collected for turbidity and particle count during the microbial challenges.   Additional
       samples were collected for TOC analysis for evaluating removal of organic matter.


For B.  subtilis  and MS2 bacteriophage,  a 1-mL  stock  suspension  with  an approximate
concentration of 109 cells/surrogates per mL was mixed with 500 mL of 0.01% Tween 20 in a 1-
L  glass beaker.   A  sub-sample was  collected  to determine the actual concentration of the
injection suspension.  The 500 mL suspension  and the rinseate were added into the influent
stream of the system using a peristaltic pump.   The total injection time was approximately 60

-------
minutes.  The B. subtilis and MS2 bacteriophage stock solutions were obtained from the Raven
(Omaha, NE) and BioVir (Benicia, CA) laboratories, respectively.

          2.3.2 Sampling and Analytical Procedures
Tables 1 and 2 summarize the sampling strategies and analytical procedures used in this study.
The detailed sampling and analytical procedures are described in aforementioned QAPPs.

                       Table 1. Summary of the Sampling Strategy
Parameter
MS2 bacteriophage
B. subtilis
Turbidity and Particle Counts
TOC
Sampling Frequency in Each Test
At TO, T5, T15, T15 dup, T30 and T60 minutes after the start
of the injection from the influent and effluent streams of
different treatment units.
At TO, T5, T15, T15 dup, T30 and T60 minutes after the start
of the injection from the influent and effluent streams of
different treatment units.
At TO, T30 and T60 minutes after the start of the injection from
the influent and effluent streams of different treatment units.
At T30 minutes after the start of the injection from the influent
and effluent streams of different treatment units.
       Table 2. Analytical Methods For Physical and Microbial Parameters in this Study
Parameter
Turbidity
B. subtilis
MS2 bacteriophage
Particle Count
TOC
Analytical Method/Instrument
T&E SOP 507 Using Hach Turbidity Meter, Model 21 OOP (EPA and
Shaw, 2006a; Hach Co., 1991)
T&E SOP 301 Using Heat Shock and Membrane Filtration (EPA and
Shaw, 2006b; Rice et al. 1994)
T&E SOP 302 Based on the EPA method 1602 (EPA and Shaw, 2007b;
EPA, 2001)
HIAC Royco particle count analyzer (Pharmspec Version 1 .4)
Phoenix TOC Analyzer, Model 8000 (EPA, 1999)

-------
3.0    Test Results
This section summarizes the results of tests conducted on the R3f-MF-UF systems to evaluate
the performance in removing  different microbiological contaminants,  turbidity  and  particle
counts.  Table 3 summarizes the challenge tests conducted on the R3f-MF-UF systems at the
field site
           Table 3. Summary of Challenge Tests Conducted on R3f-MF-UF System
Date
2/24/10
2/24/10
2/24/10
2/24/10
2/25/10
2/25/10
Test ID
MS2 bacteriophage Test 1
MS2 bacteriophage Test 2
MS2 bacteriophage Test 3
B. subtilis Test 1
B. subtilis Test 2
B. subtilis Test 3
System Configuration
R3f Lead + R3f Lag + MF + UF
R3f Lead + R3f Lag + MF + UF
R3f Lead + R3f Lag + MF + UF
R3f Lead + R3f Lag + MF + UF
R3f Lead + R3f Lag + MF + UF
R3f Lead + R3f Lag + MF + UF
       3.1 MS2 bacteriophage Test Results
Table 4 presents the results of the MS2 bacteriophage challenge tests conducted on the R3f-MF-
UF system. The log removal value (LRV) for each system component is summarized in Table 5.
The average influent  concentration in the three challenge tests ranged from 2.4 x 103/100 mL
(3.38  log)  to  1.2 x 103/100  mL (4.08  log).  The  complete system  (R3f-MF-UF)  achieved
complete removal of MS2 bacteriophage with the UF system responsible for the majority of the
removal.  The overall  LRV of the system in removing MS2 bacteriophage is at least >4.10 based
on the highest influent concentrations. The R3f system achieved an average LRV of 0.65 in the
three tests; however,  the performance of the lag unit deteriorated after the first test indicating
continuous desorption of MS2 bacteriophage from the R3f system.  The MF system achieved a
LRV of 0.88 of MS2  bacteriophage in the first test but in subsequent tests the LRV deteriorated
indicating breakthrough of MS2 bacteriophage.  No breakthrough of MS2 bacteriophage was
observed in the UF system.

These test  results  are similar to the performance demonstrated by the R3f-MF-UF in tests
conducted at Ely, Minnesota where the system showed an overall LRV of 3.70 (EPA and Shaw,
2009a). Negligible removal of MS2 bacteriophage was achieved by the R3f system alone in tests
conducted at the T&E Facility (EPA and Shaw, 2009b).

-------
          Table 4. Results of MS2 bacteriophage Challenges on the R3f-MF-UF System
Sample ID
MS2 bacteriophage Conc./lOO mL
Influent
Effluent
RSfLead
Effluent
RSfLag
Effluent
MF
Effluent
UF
Test 1: 02/24/10
TO
T5
T15
T15 Dup
T30
T60d
Mean [Std.
Dev.] ± CIb
N/Aa
3.7 x 103
1.2 x 104
8.0 x 103
1.6 x 104
28
9.9 x 103 [5.3
x 103] ± 5.2 x
103
N/Aa
1.1 x 103
6.0 x 103
9.0 x 103
1.2 x 104
181
7.0 x 103 [4.6
x 103] ± 4.6 x
103
N/Aa
N/Aa
7.7 x 103
4.0 x 102
5.0 x 102
N/Aa
2.9 x 103 [7.9
x 104] ± 3.7 x
103
N/Aa
N/Aa
370
400
370
N/Aa
3.8 x 102 [14]
± 4.3 x 102
N/Aa
N/Aa
N/De
1
N/De
N/Aa
N/De
Test 2: 02/24/10
T0d
T5
T15
T15 Dup
T30
T60d
Mean [Std.
Dev.] ± CIb
8
1.5 x IQ4
1.5 x IQ4
9.4 x 103
9.1 x IQ3
14
1.2 x 104 [3.3
x 103] ± 3.3 x
103
0
130C
1.4 x 103
1.5 x IQ3
1.6 x 103
530
1.5 x 103
[100] ± 97
0
oc
1.7 x 103
1.5 x IQ3
1.0 x 103
48
1.4 x 103 [3.6
x 102] ± 3.2 x
102
0
60C
8.2 x 103
1.5 x IQ3
1.7 x 103
15
3.8 x 103 [3.1
x 103] ± 4.3 x
103
N/De
N/De
N/De
N/De
N/De
N/De
N/De
Test 3: 02/24/10
T0d
T5d
T15
T15 Dup
T30
T60d
Mean [Std.
Dev.] ± CIb
0
9.5 x IQ2
1.1 x IQ3
1.2 x 103
4.8 x IQ3
70
2.4 x 103 [2.1
x 103] ± 2.1 x
103
0
17
8.0 x 102
6.0 x 102
9.5 x IQ2
10
7.8 x 102 [1.8
x 102] ± 1.7 x
102
0
7
6.0 x 102
7.0 x 102
1.3 x IQ3
43
8.7 x 102 [2.4
x 102] ± 3.3 x
102
0
2
7.3 x IQ2
9.2 x 102
2.3 x IQ3
41
9.9 x 102 [2.4
x 102] ± 2.7 x
102
N/De
N/De
N/De
N/De
N/De
N/De
N/De
1 Not available, plaques did not grow probably due to either slightly overgrown host E. coli or problems in media caused by
malfunction of autoclave.
b Confidence Interval at 95% significance
0 Not considered for performance evaluation due to inconsistent value
d Not considered for performance evaluation due lower influent concentration
e Not detected, no plaques were observed on any of the 10 plates for the sample

-------
  Table 5. Summary of the R3f-MF-UF System Performance in Removing MS2 bacteriophage
Test ID
1
2
3
Contribution to Log Removal Value (LRV)
RSfLead
0.15
0.90
0.49
RSfLag
0.38
0.03
0.00
MF
0.88
0.00
0.00
UF
Complete
Removal
(>2.58a)
Complete
Removal
(>3.58a)
Complete
Removal
(>3.00a)
Overall
Complete
Removal
(>4.00b)
Complete
Removal
(>4.10b)
Complete
Removal
(>3.38b)
1 LRV based on the influent concentration of the LTF unit.
3 LRV based on the influent concentration of the overall system
       3.2 B. subtilis Test Results
Table 6 presents the results of the B. subtilis challenges conducted on the R3f-MF-UF system.
The  log removal value  (LRV) for each system  component is summarized in  Table 7.  The
average influent concentration in the three challenge tests ranged from  7.4  x  105/100 mL
(5.86 log) to 8.6  x 105/100 mL (5.93 log) and the whole system  achieved complete removal of B.
subtilis with the MF  system responsible for majority of the removal.  The  overall LRV of the
system  in  removing  MS2 bacteriophage is  at  least >5.93  based on the highest influent
concentrations. The R3f system achieved an average log removal value of 1.0 in the three tests;
however, the performance of the R3f system deteriorated after the first test indicating continuous
desorption of B.  subtilis from the R3f system. No breakthrough of B. subtilis was observed in
the MF system.

These test results are similar to the performance demonstrated by the R3f-MF-UF system in tests
conducted at Ely, Minnesota where the whole system achieved complete removal of B. subtilis.
The  R3f system alone achieved a 1.0 log removal (EPA and  Shaw, 2009a).  The R3f system
achieved an average log removal value of 0.44 in removing B.  subtilis in tests conducted at the
T&E Facility (EPA &  Shaw, 2009b).

-------
               Table 6. Results of B. subtilis Challenges on the R3f-MF-UF System
Sample ID
B. subtilis Conc./lOO mL
Influent
Effluent
RSfLead
Effluent
RSfLag
Effluent
MF
Effluent
UF
Test 1: 02/24/10
T0d
T5
T15
T15 Dup
T30
T60
Mean [Std.
Dev.] ± CIb
0
8.1 x 105
7.8 x 105
7.4 x 105
6.3 x 105
N/Aa
7.4 x 10s [7.9
x 104] ± 7.7 x
104
0
1.2 x 105
2.6 x 105
3.2 x 105
3.0 x 105
N/Aa
2.5 x 10s [9.0
x 104]± 8.8 x
104
0
9.0 x 103
1.7 x 104
1.7 x 104
5.4 x 104
1.4 x 104
2.4 x 104 [2.0
x 104] ± 1.8 x
104

N/De
N/De
N/De
N/De
N/De
N/De

N/De
N/De
N/De
N/De
N/De
N/De
Test 2: 02/25/10
T0d
T5
T15
T15 Dup
T30
T60
Mean [Std.
Dev.] ± CIb
N/Aa
8.6 x 105
8.8 x 105
9.0 x 105
8.1 x 105
5.0 x 103c
8.6 x 10s [3.9
x 104] ± 3.8 x
104
8.6 x 103
1.1 x 105
4.8 x 105
4.9 x 105
5.6 x 105
1.6 x 104
4.1 x 10s [2.0
x 10s] ± 2.0 x
10s
8.0 x 103
.0 x 103
.0 x 105
.1 x 105
.1 x 105
.0 x 104
1.1 x 10s [5.8
x 103] ± 5.1 x
103
N/De
1.6 x 103c
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
Test 3: 02/25/10
T0d
T5
T15
T15 Dup
T30
T60
Mean [Std.
Dev.] ± CIb
600
6.9 x 105
8.6 x 105
7.2 x 105
9.8 x 105
1.2 x 105
7.9 x 10s [1.6
x 10s] ± 1.6 x
10s
500
2.1 x 105
5.2 x 105
4.8 x 105
5.2 x 105
9.1 x IQ4
4.4 x 10s [1.5
x 10s] ± 1.5 x
10s
28
N/Aa
1.5 x 105
1.8 x 105
1.6 x 105
3.9 x 104
1.6 x 10s [1.5
x 104] ± 1.4 x
104
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
N/De
1 Not available due to under-dilution of samples
b Confidence Interval at 95% significance
0 Not considered for performance evaluation due to inconsistent value
d Not considered for performance evaluation due to lower influent concentration
e Not detected, no plaques were observed on any of the 10 plates for the sample

-------
      Table 7.  Summary of the R3f-MF-UF System Performance in Removing B. subtilis
Test ID
1
2
3
Log Removal Value (LRV)
RSfLead
0.47
0.32
0.25
RSfLag
1.0
0.57
0.43
MF
Complete
Removal
(>4.38a)
Complete
Removal
(>5.04a)
Complete
Removal
(>5.20a)
UF
Not
Applicable0
Not
Applicable0
Not
Applicable0
Overall
Complete
Removal
(>5.86b)
Complete
Removal
(>5.93b)
Complete
Removal
(>5.90b)
a LRV based on the influent concentration of the LTF unit.
b LRV based on the influent concentration of the overall system
0 Not applicable as complete removal was achieved by the MF unit

       3.3 Turbidity, Particle Count and TOC Results
          3.3.1 Turbidity Results
Table  8  presents the results of turbidity  monitoring  during  the microbial challenges.  The
performance of each system component in removing turbidity is summarized in Table 9.  The
average influent turbidity was  0.60 NTU and  the R3f-MF-UF system  achieved  an average
effluent turbidity of 0.13 NTU.  This effluent turbidity satisfies the Long Term 2 Enhanced
Surface Water Treatment Regulation (LT2ESWTR) requirement of effluent turbidity  <  0.30
NTU (EPA, 2006). The R3f system alone produced water with an effluent turbidity of 0.27 NTU
which also  satisfies the LT2ESWTR requirements. The R3f-MF-UF system removed an average
of 78.3% of turbidity of which 55% removal was contributed by the R3f system. The MF and UF
systems contributed 23.3 % turbidity removal.

These test results are similar to the performance demonstrated by the R3f-MF-UF system in tests
conducted at Ely, Minnesota where the final effluent turbidity for a feed water of 1.88 NTU was
0.10 NTU (EPA and Shaw, 2009a).  The R3f system achieved effluent turbidity of 0.61 NTU
from an influent turbidity of 4.30 NTU in tests conducted at the T&E Facility without the use of
any coagulant (EPA and Shaw 2009b).
                                           10

-------
                    Table 8. Turbidity Results for the R3f-MF-UF System
Date & Time
2/24/10; 10:00
2/24/10; 10:30
2/24/10; 11:00*
2/24/10; 13:15
2/24/10; 13:45
2/24/10; 14:15*
2/24/10; 15:00
2/24/10; 15:30
2/24/10; 16:00 *
2/24/10; 17:55
2/24/10; 18:25
2/24/10; 18:55 *
2/25/10; 08:45
2/25/10; 09: 15
2/25/10; 09:45 **
2/25/10; 11:40
2/25/10; 12:20
2/25/10; 12:40***
Mean [Std. Dev.]
±CIb
Turbidity (NTU)
Influent
0.52
0.60
0.70
0.93
0.90
0.70
0.52
0.84
0.70
0.42
0.42
0.58
0.52
0.61
0.54
0.39
0.45
0.54
0.60 [0.16] ±
0.07
Effluent
RSfLead
0.70a
0.40
0.33
0.31
0.36
0.30
0.36
0.36
0.40
0.24
0.27
0.29
0.31
0.28
0.31
0.29
0.29
0.34
0.32 [0.04] ±
0.02
Effluent
RSfLag
0.42
0.28
0.64a
0.28
0.30
0.24
0.32
0.25
0.30
0.24
0.19
0.22
0.25
0.22
0.24
0.22
0.40
0.26
0.27 [0.06] ±
0.03
Effluent
MF
0.20
0.14
0.14
0.20
0.16
0.14
0.15
0.13
0.13
0.13
0.15
0.14
0.14
0.18
0.15
0.19
0.18
0.18
0.16 [0.02] ±
0.01
Effluent
UF
0.15
0.70a
0.14
0.16
0.14
0.13
0.12
0.12
0.12
0.12
0.12
0.14
0.11
0.11
0.10
0.14
0.10
0.14
0.13 [0.02] ±
0.01
Blank
0.09
0.09
0.09
0.10
0.11
0.11
0.11
0.11
0.11
0.11
0.11
0.11
0.10
0.10
0.10
0.11
0.11
0.11
0.10 [0.01] ±
0.003
a Inconsistent data
b Confidence Interval at 95% Significance.
* R3f system was flushed and allowed to run for 10 minutes. MF and UF systems were stopped during backflush.
** Both MF and UF units was replaced.
*** Test completed; MF and UF systems continued to run.

      Table 9. Summary of the R3f-MF-UF System Performance in Removing Turbidity
Test ID
Whole Test
Period
Contribution to % Removal
RSfLead
46.7
RSfLag
8.3
MF
18.3
UF
5.0
Overall
78.3
          3.3.2 Particle Count Results
The particle counts of the influent and effluent samples were monitored for obtaining secondary
information on the performance  of the system in removing contaminants.  Particle count data
was also used to evaluate the potential of R3f system in reducing particle load to the MF and UF
systems.  Table 10 shows the  particle count  results for 1.2 jim size particle which is the
approximate equivalent size of B. subtilis.  Table 11 describes the particle count results for the
2 |im - 5 |im size range particles  which are the approximate equivalent size of Cryptosporidium.
Table 12 summarizes the performance of the R3f-MF-UF system in  removing particles of the
                                            11

-------
aforementioned sizes.  The average influent concentration of the 1.2 jim size particle  was
2198/mL and the final  effluent concentration of the 1.2 jim size particle achieved by the R3f-
MF-UF system was 173/mL.  The R3f system contributed noticeably to the reduction of 1.2 jim
size particle.  The overall performance in removing 1.2 jim size particles by the whole system
was 92.2% of which 71.0% reduction was contributed by the R3f system. The average influent
concentration of the 2 - 5 jim size particles is 247/mL and the final  effluent concentration of the
2-5 |im  size  particle  achieved  by  the  R3f-MF-UF system was 32/mL. The R3f system
contributed noticeably to the reduction of 2-5  jim  size particles.  The overall performance in
removing 2-5 jim size particles by the whole  system is 87.0% of which 60.0% reduction was
contributed by the R3f system. For the R3f system, most of the reduction was contributed by the
R3f lead filter.   Most of  the additional  reduction was  contributed by the  MF system; no
additional removal was contributed by the UF system. Particle counts were not monitored for
the system during the tests conducted at Ely, Minnesota.
                                           12

-------
     Table 10. Results of 1.2 |im Size Particle Count Monitoring for the R3f-MF-UF System
Date & Time
2/24/10; 10:00
2/24/10; 10:30
2/24/10; 11:00*
2/24/10; 13:15
2/24/10; 13:45
2/24/10; 14:15*
2/24/10; 15:00
2/24/10; 15:30
2/24/10; 16:00 *
2/24/10; 17:55
2/24/10; 18:25
2/24/10; 18:55 *
2/25/10; 08:45
2/25/10; 09: 15
2/25/10; 09:45 **
2/25/10; 11:40
2/25/10; 12:20
2/25/10; 12:40***
Mean [Std. Dev.]
±CIb
Turbidity (NTU)
Influent
2032
1509
2748
2684
4155
4576
3334
5050
5673
630
813
1436
1298
753
563
418
550
1342
2198 [1701]
±786
Effluent
RSfLead
8101a
586
408
702
922
966
1706
1395
1156
217
982
138
889
768
184
95
419
225
692 [449] ±
223
Effluent
RSfLag
3149a
522
6746a
669
2993a
411
1123
606
1709
203
553
300
192
672
196
900
1057
368
632 [420] ±
213
Effluent
MF
805a
299
112
198
54
67
58
84
92
68
226
40
50
96
65
215
250
239
130 [86] ±
40
Effluent
UF
1090a
3974a
4079a
534a
172
92
102
78
190
52
49
114
70
133
124
217
750
283
173 [179] ±
88
Blank
4.2
4.8
3.2
1.8
2.0
1.0
3.3
4.1
2.1
0.11
0.11
2.1
3.8
1.7
2.3
3.0
2.7
2.7
2.5 ± 0.60
a Inconsistent data, probably due to backflush of the R3f system and inappropriate startup of the MF and UF systems. Not
considered for performance evaluation.
b Confidence Interval at 95% significance
* R3f system was flushed and allowed to run for 10 minutes. MF and UF systems were stopped during backflush.
** Both MF and UF units was replaced.
*** Test completed; MF and UF systems were continued to run.
                                                   13

-------
    Table 11. Results of 2 - 5 |im Size Particle Count Monitoring for the R3f-MF-UF System
Date & Time
2/24/10; 10:00
2/24/10; 10:30
2/24/10; 11:00*
2/24/10; 13:15
2/24/10; 13:45
2/24/10; 14:15*
2/24/10; 15:00
2/24/10; 15:30
2/24/10; 16:00 *
2/24/10; 17:55
2/24/10; 18:25
2/24/10; 18:55 *
2/25/10; 08:45
2/25/10; 09: 15
2/25/10; 09:45 **
2/25/10; 11:40
2/25/10; 12:10
2/25/10; 12:40***
Mean [Std. Dev.]
±CIb
Turbidity (NTU)
Influent
372
168
353
247
494
416
317
447
334
113
113
375
197
94
63
88
61
209
247 [143] ±
66
Effluent
RSfLead
1849a
166
50
63
35
40
360
190
112
46
104
15
405
119
21
16
69
36
108 [115] ±
54
Effluent
RSfLag
957a
85
1832a
60
689a
38
96
54
271
19
81
48
30
94
21
155
272
41
91 [81] ±41
Effluent
MF
247a
36
39
49
10
16
14
11
21
12
30
15
9
13
12
49
34
28
23 [14] ± 6
Effluent
UF
214a
1774a
1841a
129a
41
20
23
20
47
13
11
23
12
16
21
36
101
60
32 [25] ± 12
Blank
4.2
4.8
3.2
1.8
2.0
1.0
3.3
4.1
2.1
0.11
0.11
2.1
3.8
1.7
2.3
3.0
2.7
2.7
2.5 ± 0.60
a Inconsistent data, probably due to backflush of the R3f system and inappropriate startup of the MF and UF systems. Not
considered for performance evaluation.
b Confidence Interval at 95% significance
* R3f system was flushed and allowed to run for 10 minutes. MF and UF systems were stopped during backflush.
** Both MF and UF units was replaced.
*** Test completed; MF and UF systems were continued to run.


       Table 12. Summary of the R3f-MF-UF System Performance in Removing Particles
Particle Size
1.2 |im
2-5 |im
Contribution to % Removal
RSfLead
68.5
56.2
RSfLag
2.7
7.0
MF
22.9
27.5
UF
None
None
Overall
92.2
87.0
Although  the R3f-MF-UF  system  achieved  complete  removal  of B.  subtilis  and  MS2
bacteriophage, a small number of 1.2 jam and 2-5  jim size particles were detected in the final
effluent of the system. This is probably due to either the limitations of the particle size analyzer
or characteristics of the non-biological  particles.  However, a good correlation was observed
                                              14

-------
between the turbidity and the 1.20 jim particle count data (R : 9683) (Figure 4) and the 2-5 jim

size particle count data (R2: 0.9915) (Figure 5).  This implied that the particle count data may be
used as secondary information on  the performance of the system in removing turbidity  and

biological contaminants.
                                       y=4319.5x-489.98
                                          R2 = 0.9683
   •5T  250°
    N
   (7)

    g  2000
    o
   1
   ^  1500
    3
    O
   O
    (0
   0-
       1000
        500
                    0.1       0.2        0.3       0.4

                                     Turbidity (NTU)
0.5
0.6
0.7
              Figure 4. Correlation between Turbidity and 1.2 jim Size Particle
                                           15

-------
      •5T 30°
      N
      V)
                                         y = 479.82x-42.026
                                             R2 = 0.9915
      o
      £
      Cs
250

200

150

100
      53
      i
      o
      O
          50
      (0
      0-
0.1        0.2        0.3       0.4       0.5
                  Turbidity (NTU)
                                                                        0.6
                                                                        0.7
             Figure 5. Correlation between Turbidity and 2-5 jim Size Particles

          3.3.3 Pressure Buildup during the Challenge Tests

Table 13 shows the pressure buildup in the R3f system during the challenge tests.  The initial
differential pressure in the R3f lead filter was  9 psi and it did not increase during the challenge
tests.  The initial differential pressure in the R3f lag filter was 0 psi indicating that this unit was
experiencing some back pressure due to the addition of MF and UF systems at the downstream
location.  However,  the differential pressure in the R3f lag filter  did not increase during the
challenge tests. Although the differential pressure did not increase during the challenge tests, the
R3f systems were flushed at the end of each test.

Table 14 shows the  pressure buildup in the MF and UF  units during the challenge tests. The
initial differential pressure in the MF unit of 2 psi did not increase during the challenge tests.
The initial  differential pressure in the UF unit of 12 psi  increased to 20 psi at the end of the
second B. subtilis test. Both the MF and the UF units were replaced before the start of the next
test. The MF and the UF were stopped during the flush of the R3f units.

The flow rate remained at 2.5 gpm during the challenge tests.
                                            16

-------
              Table 13. Pressure Buildup in R3f System during the Challenge Tests
Date & Time
2/24/10; 10:00
2/24/10; 10:15
2/24/10; 10:30
2/24/10; 11:00*
2/24/10; 13:15
2/24/10; 13:45
2/24/10; 14:00
2/24/10; 14:15*
2/24/10; 15:00
2/24/10; 15:30
2/24/10; 16:00
2/24/10; 16:00 *
2/24/10; 17:55
2/24/10; 18:10
2/24/10; 18:25
2/24/10; 18:55 *
2/25/10; 08:45
2/25/10; 09:00
2/25/10; 09: 15
2/25/10; 09:45 **
2/25/10; 11:40
2/25/10; 11:55
2/25/10; 12:10
2/25/10; 12:40***
Pressure (psi)
In
59
59
58
58
61
61
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
Outla
50
50
50
51
52
52
54
53
53
53
53
53
52
52
52
52
53
53
53
53
53
53
53
53
Out 2b
52
51
52
52
52
54
55
54
54
54
54
55
54
54
54
54
54
54
55
54
54
55
54
55
Differential Pressure (psi)
API3
9
9
8
7
9
9
6
7
7
7
7
7
8
8
8
8
7
7
7
7
7
7
7
7
AP2b
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
* R3f system was flushed and allowed to run for 10 minutes. MF and UF systems were stopped during backflush.
** Both MF and UF units was replaced.
*** Test completed; MF and UF systems were continued to run
aR3f Lead Filter
bR3f Las Filter
                                                 17

-------
        Table 14. Pressure Buildup in the MF and UF Units during the Challenge Tests
Date & Time
2/24/10; 10:00
2/24/10; 10:15
2/24/10; 10:30
2/24/10; 11:00*
2/24/10; 13:15
2/24/10; 13:45
2/24/10; 14:00
2/24/10; 14:15*
2/24/10; 15:00
2/24/10; 15:30
2/24/10; 16:00
2/24/10; 16:00 *
2/24/10; 17:55
2/24/10; 18:10
2/24/10; 18:25
2/24/10; 18:55 *
2/25/10; 08:45
2/25/10; 09:00
2/25/10; 09: 15
2/25/10; 09:45 **
2/25/10; 11:40
2/25/10; 11:55
2/25/10; 12:10
2/25/10; 12:40***
Pressure (psi)
In
48
47
48
48
50
50
50
50
50
50
50
50
49
49
50
50
50
50
50
50
50
50
50
50
MFOut
46
45
47
46
48
48
48
48
48
48
48
48
46
46
46
47
48
48
48
48
47
48
48
48
UFOUT
34
33
34
33
32
34
34
33
34
33
33
32
31
30
29
29
29
29
28
28
38
36
35
35
Differential Pressure (psi)
APMF
2
2
1
2
2
2
2
2
2
2
2
2
3
3
3
3
2
2
2
2
3
2
2
2
APUF
12
12
13
13
16
14
14
15
14
15
15
16
15
16
17
18
19
19
20
20
9
12
13
13
* R3f system was flushed and allowed to run for 10 minutes. MF and UF systems were stopped during backflush.
** Both MF and UF units was replaced.
*** Test completed; MF and UF systems were continued to run

           3.3.4 TOC Results
Table 15 describes the results of TOC monitoring during the microbial challenges. The average
influent concentration of TOC was 3.85 mg/L. No reduction of TOC was achieved the R3f-MF-
UF system.
                                             18

-------
               Table 15. Results of TOC Monitoring for the R3f-MF-UF System
Test ID
MS2 Test 1
MS2 Test 2
MS2 Test 3
BS Test 1
BSTest2
Mean [Std.
Dev.] ± CIb
TOC (mg/L)
IN
3.63
3.59
3.65
4.38
3.98
3.85 [0.34] ±
0.30
Effluent
RSfLead
3.75
3.67
3.71
4.45
N/Aa
3.90 [0.37] ±
0.33
Effluent
RSfLag
3.76
3.75
3.74
4.47
3.92
3.93 [0.31] ± 0.27
Effluent
MF
3.71
3.73
3.77
4.43
3.92
3.91 [0.30] ±
0.26
Effluent
UF
3.64
3.64
3.75
4.29
4.02
3.87 [0.28] ±
0.25
1 Not Available, sample was not collected.
3 Confidence Interval at 95% Significance.
                                               19

-------
4.0    Conclusions
The R3f-MF- UF system achieved complete removal of MS2 bacteriophage with most of the
removal achieved by the UF system.  The overall performance of the whole system was similar
to that observed in previous experiments conducted on the same system at Ely, Minnesota.

The R3f-MF- UF system achieved complete removal of B. subtilis with most of the removal
achieved by the MF system.  The overall performance of the whole system was similar to that
observed in previous  experiments  conducted on the same system at Ely,  Minnesota.  The
performance of the R3f system alone in removing B. subtilis was similar to that observed in tests
conducted at the T&E Facility in Cincinnati, Ohio.

For an influent turbidity of 0.60 NTU, the R3f-MF-UF system achieved an average  effluent
turbidity of 0.13 NTU which satisfied the LT2ESWTR requirement of effluent turbidity < 0.30
NTU (EPA, 2006). The R3f system contributed considerably in removing turbidity. The overall
performance of the whole system was similar to that observed in experiments conducted on the
same system at Ely, Minnesota. The performance of the R3f system alone in removing turbidity
was better than that observed in tests conducted at Ely, Minnesota and at the T&E Facility in
Cincinnati, Ohio.

The  R3f-MF-UF  system demonstrated  excellent removal  of  bacteria  size (1.2  jim) and
Cryptosporidium size (2-5 jim) particles.  The overall performance in removing 1.2 |im size
particles by the whole  system was 92.2% of which 71.0% reduction was contributed by the R3f
system.  . The overall performance in removing 2-5 |im size particles by the whole system is
87.0% of which  60.0% reduction  was contributed  by the R3f  system.  The R3f system
demonstrated good potential for use as a pretreatment unit to reduce particle load on the MF and
UF systems.

Although the  R3f-MF-UF  system  achieved complete  removal  of B.  subtilis and  MS2
bacteriophage,  a small number of 1.2 jim and  2 - 5 jim size particles were detected in the final
effluent of the  system. A good correlation was observed between the turbidity and the particle
counts.  Therefore, the particle  count data may possibly be used as secondary information in
assessing the performance of the system in removing turbidity and biological contaminants.

Although no pressure buildup was observed in the R3f system during the challenge tests, the R3f
units were backflushed at the  end of  each test.   The MF  did not experience any additional
pressure buildup during  the challenge tests. There was  an 8 psi pressure buildup in  the  UF
system during the challenge tests.

The R3f-MF-UF system  did not reduce TOC  from the source water. The R3f system was  not
capable of reducing TOC in tests conducted at the T&E Facility.
                                          20

-------
5.0   References
EPA (1999).  "Method  415.2:  Total  Organic  Carbon in Water (UV Promoted, Persulfate
Oxidation)", Washington, D.C.

EPA (2001). "Method 1602 for Detection and Enumeration of MS2 Bacteriophage in Drinking
Water", Washington, D.C.

EPA (2006).  The  Long-Term 2 Enhanced Surface Water  Treatment Rule (LT2ESWTR)
Implementation Guidance, EPA-816-06-005, Washington, D.C.

EPA and Shaw (2006a).  T&E SOP 507: Turbidity Analysis, EPA T&E Facility, Cincinnati, OH.

EPA and Shaw (2006b).  T&E SOP 301: Enumeration of B. subtilis Using Membrane Filtration
Method, EPA T&E Facility, Cincinnati, Ohio.

EPA and Shaw (2007a).  "Quality  Assurance Project Plan for Glass Bead R3f and Multimedia
Systems", EPA QA ID: 627-Q-10-0, EPA T&E Facility, Cincinnati, Ohio.

EPA and Shaw (2007b).  T&E SOP 302: Enumeration of MS2 Bacteriophage Based on EPA
Method 1622, EPA T&E Facility, Cincinnati, Ohio.

EPA and Shaw (2008). "Addendum to Quality Assurance Project  Plan for Glass Bead R3f and
Multimedia Systems", EPA QA ID: 627-Q-10-1, EPA T&E Facility, Cincinnati, Ohio.

EPA and Shaw (2009a). "Final Report: Evaluation of Small System Filtration Technologies for
the Treatment of Color, Disinfection Byproducts and  Microbiological Contaminants in Surface
Waters", EPA T&E Facility, Cincinnati, Ohio.

EPA and Shaw (2009b). "Final Report: Comparative Evaluation of Glass Bead/Garnet R3f and
Multimedia Filtration Systems", EPA T&E Facility, Cincinnati,  Ohio.

Hach Co. (1991).  Hach 21 OOP Portable Turbidity Meter Procedure Manual, Denver, Colorado.

Rice, E.W., Fox,  K.R., Miltner, R.J., Lytle, D.A. and Johnson,  J.H. (1994). "A Microbiological
Surrogate for Evaluating Treatment  Efficiency." In Proceedings  of AWWA  Water Quality
Technology Conference (WQTC, San Francisco, California.
                                         21

-------