EVALUATION OF A HOUSEHOLD CONTACT DISINFECTION DEVICE FOR
INACTIVATION OF BACTERIOPHAGE MS2 AND MURINE NOROVIRUS
By
Emaly Leak

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Microbiology and Molecular Genetics- Master of Science
2013

ABSTRACT
EVALUATION OF A HOUSEHOLD CONTACT DISINFECTION DEVICE FOR
INACTIVATION OF BACTERIOPHAGE MS2 AND MURINE NOROVIRUS
By
Emaly Leak
Obtaining safe drinking water is problematic in many developing nations
around the world. The HaloPure disinfection canister designed by HaloSource
was created to provide household water treatment for middle-class families in
India. The Waterbird device was tested for its effectiveness in reducing viruses.
The main objective of this study was to determine the effectiveness of the
Waterbird device in removing or inactivating bacteriophage MS2 and murine
norovirus as surrogates for human pathogens. Secondary objectives were to
determine the potential impacts, if any, that added organic contaminants (in the
form of raw sewage) and pH adjustment have on the effectiveness of the device.
The Waterbird device was tested by adding MS2 and murine norovirus stock to
five liters of well water (with or without sewage); the pH of the water was adjusted
to 7.5 or 9 before treatment. The murine norovirus samples were all reduced to
the detection limit, achieving a minimum of 2.0 to 4.0 log10 reductions after
treatment, but the performance of the device could not be accurately examined
using these results. For the MS2, the Waterbird device inactivated or removed
an average of 5.4 log10. The MS2 log10 reduction was affected by pH (p=0.006)
and sampling times (p<0.001). Overall, the Waterbird device met the U.S. EPA’s
guidelines for water purifiers (which requires at least 4 log10 removal of viruses).

ACKNOWLEDGEMENTS

I would like to thank my advisor Dr. Joan B. Rose for her help and support in the
development of this project, and throughout my time at MSU. I am grateful to my
committee members, Dr. Susan Masten and Dr. Shannon Manning, for their
guidance during my Master’s research. I would like to thank the Rose Lab
members, especially Dr. Kyle Enger and Dr. Tiong Gim Aw, for their support and
help in carrying out this project. And I would like to thank HaloSource and Dr.
Jeff Williams for funding this research.

!

iii!

TABLE OF CONTENTS
LIST OF TABLES……………………………………………………………….…..v
LIST OF FIGURES…………………………………………………………………vii
CHAPTER 1
INTRODUCTION…………………………………………………………..….…….1
1.1 SAFE WATER FOR THE DEVELOPING WORLD…………...……….......1
1.2 NOROVIRUS…………….…….………………….……………………………4
1.3 WATER TREATMENT & DISINFECTION…………..………………….….10
1.4 OBJECTIVES………………………………………..………………………..21
CHAPTER 2
MATERIALS & METHODS……………….………………………………………22
2.1 EXPERIMENTAL DESIGN………………..………………………….……..22
2.2 BROMINE DEMAND OF THE ADDED SEWAGE…………………..……25
2.3 CHEMICAL OXYGEN DEMAND OF THE ADDED SEWAGE…………..26
2.4 VIRUS PROPAGATION………….……………………………………….…27
2.5 VIRUS QUANTIFICATION……………………..……………………………30
2.6 STATISTICAL ANALYSIS…………………………………….……………..31
CHAPTER 3
RESULTS…………………..………………………………………………………33
3.1 INTRODUCTION………………………..……………………….…………...33
3.2 MS2 VIRUS………………………………………………………….………..35
3.2.1 INFLUENT CHARACTERIZATION……………………………..……35
3.2.2 EFFLUENT CHARACTERIZATION……………………………….....39
3.2.2.1 BROMINE RESIDUAL……………………………………....40
3.2.2.2 FLOW RATE……………………………………………….....45
3.2.2.3 DEVICE PERFORMANCE……………………………….....47
3.2.2.3.1 EFFECT OF PH AND SEWAGE ADDITION..….53
3.2.2.3.2 DEVICE VARIABILITY……………………….……57
3.3 MURINE NOROVIRUS……………………………………….………….…..61
3.3.1 INFLUENT CHARACTERIZATION……………………………….….61
3.3.2 EFFLUENT CHARACTERIZATION……………………………….....66
3.3.3 DEVICE PERFORMANCE…………………………………....66
3.4 CONCLUSIONS……………………………………………………………...73
CHAPTER 4
DISCUSSION………………………………………………………………………75
APPENDIX…..................................................................................................85
REFERENCES…………………………………………………………..……….104
!

iv!

LIST OF TABLES
Table 1: Viruses common in water that cause disease in humans .......................3
Table 2: Disinfection of murine norovirus and MS2 with different forms of
chlorine..................................................................................................................8
Table 3: Advantages and disadvantages of various methods of drinking water
disinfection…………….........................................................................................13
Table 4: Different methods of drinking water disinfection and their efficacies on
viruses……………………………………………………...…………………………...16
Table 5: Chemical oxygen demand measurements (in mg/L) and percent raw
sewage added for each raw sewage experiment………………….……………….27
Table 6: Influent concentrations for MS2 for the pH, sewage and well water
variables. Shown are the mean values (+/- standard deviation) for virus influent
concentrations for each combination of pH and water treatment variables..……35
Table 7: Average bromine residual (+/- standard deviation) and cartridges used
for each type of influent (pH 7.5 or 9, well water or sewage)……………………..42
Table 8: Physical characteristics (flow rate and bromine residual), and MS2
influent and effluent concentrations and log10 reductions. The mean values (+/standard deviation) are given for each measurement for each combination of pH
and water treatment variables………………………………………………………..48
Table 9: Average MS2 concentrations and device characteristics in the effluent
post treatment for well water samples at pH 7.5…………………………………..49
Table 10: Average MS2 concentrations and device characteristics in the effluent
post treatment for sewage addition samples at pH 7.5…………………………...50
Table 11: Average MS2 concentrations and device characteristics in the effluent
post treatment for well water samples at pH 9……………………………………..51
Table 12: Average MS2 concentrations and device characteristics in the effluent
post treatment for sewage addition samples at pH 9………………………………52
Table 13: Influent concentrations for murine norovirus for the pH, sewage and
well water variables. Shown are the mean values (+/- standard deviation) for
virus influent concentrations for each combination of pH and water treatment
variables………………………………………………………………………………...63

!

v!

Table 14: Murine norovirus influent and effluent concentrations and log10
reductions. The mean values (+/- standard deviation) are given for each
measurement for each combination of pH and water treatment variables………68
Table 15: Average murine norovirus concentrations and device characteristics in
the effluent post treatment for well water samples at pH 7.5……………………..70
Table 16: Average murine norovirus concentrations and device characteristics in
the effluent post treatment for sewage addition samples at pH 7.5……………...71
Table 17: Average murine norovirus concentrations and device characteristics in
the effluent post treatment for well water samples at pH 9………………………..72
Table 18: Average murine norovirus concentrations and device characteristics in
the effluent post treatment for sewage addition samples at pH 9………………..73
Table A-1: Results of all challenge experiments…………….……………………..86
Table A-2: Murine norovirus influent concentrations……………………………..103

!

vi!

LIST OF FIGURES
Figure 1: The HaloSource Waterbird device. The white part at the top is where
water is poured in; it flows through the ceramic pre-filter and the bromine
cartridge before being deposited in the bottom (clear) reservoir…………...........23
Figure 2: Diagram of a device similar to the HaloSource Waterbird. Arrows show
the direction of water flow; water flows from the upper reservoir (U), though the
bromine cartridge (H) containing packed N-bromine beads (N), to the lower
reservoir (L), and finally is deposited in the lower reservoir (L) where it can flow
out of the tap (McLennan et al., 2009).................................................................24
Figure 3: Average MS2 influent concentrations (in log10) for the pH 7.5 samples.
N=9 for well water and sewage (N=18 total)…………………...............................36
Figure 4: Average MS2 influent concentrations (in log10) for the pH 9 samples.
N=9 for well water and sewage (N=18 total)…………………….………………….37
Figure 5: Average MS2 influent concentrations (in log10) for all samples. N=18
for pH 7.5 and n=18 for pH 9 (N=36 total)…………………..………………………39
Figure 6: Bromine residual (mg/L) between cartridges used (for the pH 9 sewage
samples). Values were averaged for all time points. The bromine residual was
not significantly different between cartridges (p=0.201). N=4 for cartridge 1,2,3,
N=8 for cartridge 4, N=7 for cartridge 5, N=6 for cartridge 6, N=33 total. The *
symbol indicates an extreme outlier………………..………………………………..41
Figure 7: Bromine residual (mg/L) between cartridges used (for the pH 7.5
samples). N=71……………….………….………………………………………..….43
Figure 8: Average bromine residual (mg/L) for well water and sewage addition
samples for pH 7.5 and 9 over sampling time. N=135...………………………….44
Figure 9: Bromine residual (mg/L) between the two pH variables. N=48 for
cartridges 1, 2 and 3 only……………………………….....…………………………45
Figure 10: The flow rate (in mL/min) for all samples, separated by sampling time.
There was a significant difference in flow rates between samples (p<0.001); the
first flush sample was different than all other samples (p<0.001), the 15 minute
sample was different than the 45 minute sample (p=0.044) and the 120 minute
sample (p<0.001), and the 45 minute sample was also significantly different from
the 120 minute sample (p=0.004). N=135……………….………………..……….46
Figure 11: Log10 reduction of MS2 between well water and sewage addition
samples (pH 7.5). The sewage samples did have significantly different log10
!

vii!

reduction compared to the well water samples (p=0.005). N= 9 for well water and
9 for sewage (N=18 total)……….…………………………………………………….54
Figure 12: Log10 reduction of MS2 between well water and sewage addition
samples (pH 7.5 and 9 samples together). N= 18 for well water and 18 for
sewage (N=36 total)………………………………………………………….…….....55
Figure 13: Log10 reduction of MS2 between pH 7.5 and pH 9 samples. There
was a significant difference in the log10 reduction of MS2 between the samples
(p=0.006). N= 9 for pH 7.5 and 3 for pH 9 (N=12 total)…………………………..56
Figure 14: Log10 reduction of MS2 for the pH 7.5 samples. There was also a
significant difference between samples (p<0.001); the first flush (2-38 minutes)
sample differed significantly from both the 15 minute sample (p<0.001) and the
45 minute sample (p<0.001). The 15 minute sample also differed significantly
from the 120 minute sample (p=0.005). N=18……………………………………..58
Figure 15: Log10 reduction of MS2 for the pH 9 samples. There was a significant
effect of sampling time (p=0.002), but only the first flush sample and 45 minute
sample differed significantly (p=0.004). N=18……………………………………..59
Figure 16: MS2 log10 reduction at each sampling time. There was a significant
effect of sample on the log10 reduction of MS2 (p<0.001); the first flush and the
15 minute samples (p<0.001), the first flush and 45 minute samples (p=0.001),
the 15 minute and 120 minute samples (p=0.001), and the 45 minute and 120
minute samples (p=0.006) were all significantly different. N=18 for well water
and 18 for sewage (N=36 total)………………………………………………………61
Figure 17: Average murine norovirus influent concentrations (in log10) for pH 7.5
samples. N=9 for well water and sewage (N=18 total)…………………..............64
Figure 18: Average murine norovirus influent concentrations (in log10) for pH 9
samples. N=3 for well water and sewage (N=6 total)……………………………..65
Figure 19: Average murine norovirus influent concentrations (in log10) for all
samples. N=12 for well water and sewage (N=24 total)………………………….66
Figure 20: The log10 concentrations of murine norovirus in untreated influent
(time=0, n= 24) and treated samples. The treated samples were all reduced to
the detection limit (2 PFU/mL)………………………………………………………..67

!

viii!

CHAPTER 1
INTRODUCTION
1.1 SAFE WATER FOR THE DEVELOPING WORLD
Obtaining clean, safe drinking water is something most people in
developed countries take for granted, but in many parts of the world it is not so
easy to come by. In 2011, the WHO estimated that approximately 11% of the
world’s population was without adequate access to safe drinking water (World
Health Organization, 2011b). Roughly 2.5 million people die every year from
diarrheal diseases, with 88% of those cases due to the combination of unsafe
drinking water, inadequate sanitation, and poor hygiene (Kosek et al., 2003;
World Health Organization, 2011b; Schwarzenbach et al., 2010). About two
million of those deaths are children under the age of five, for which unsafe water
is the number one killer (Boschi-Pinto et al., 2008; Elimelech, 2006). Diarrheal
diseases caused by unsafe water and inadequate sanitation are responsible for
6.1% of all health-related deaths worldwide (Schwarzenbach et al., 2010).
Unsafe water alone is thought to be responsible for 15-30% of all gastrointestinal
illnesses (Schwarzenbach et al., 2010; Sobsey et al., 2003).
There has been an emergence of waterborne pathogens in recent years,
due to an increase in the number of people sensitive to the pathogens, an
increase in the importation of food from developing countries (where the water
quality is poor), and the natural evolution of increased virulence in microbial
pathogens (Reynolds et al., 2008). Environmental change, including climate
change, has even been shown to play a role in the increase of infectious

!

1!

diseases (Eisenberg et al., 2007; Lloyd et al., 2007). Environmental changes can
also alter transmission cycles of infectious pathogens and modify human
exposure to contaminated sources such as water or infected animals or other
humans (Eisenberg et al., 2007). Drought has been shown to increase diarrheal
disease, as people may be forced to use lower quality water sources; low rainfall
has been shown to be a determinant of the prevalence of childhood diarrheal
disease in developing countries (Lloyd et al., 2007). About 1.7 billion people
currently live in a state of water scarcity; that number is expected to increase to 5
billion people by the year 2025 (Lloyd et al., 2007).
There are many bacterial, viral, and protozoan pathogens that can be
spread through water. Common pathogenic bacteria include E. coli 0157:H7,
Salmonella spp., Klebsiella spp., Staphylococcus aureus, Vibrio spp., and
Campylobacter jejuni; protozoa include Cryptosporidium spp. , Naegleria fowleri,
Toxoplasma gondii, and Giardia spp. and viruses include adenovirus,
enteroviruses, Hepatitis A and E, norovirus, and rotavirus (World Health
Organization, 2011a; World Health Organization, 2011c). Bacteria have been
shown to be the easiest to remove by chemical disinfection, and protozoa are
larger and easier to remove by physical methods. However, the extremely small
size and unique outer coating (capsid) of viruses makes them harder to
physically remove or chemically disinfect. Table 1 lists some of the common
viruses that are often found in water and that cause disease in humans.

!

2!

Table 1. Viruses common in water that cause disease in humans.
Virus Name

Genome

Year
Discovered

Reference

Adenoviruses

1
DS DNA Respiratory
distress,
gastroenteritis

1950

Goncalves
and de Vries,
2006

Poliovirus

+ RNA

Gastroenteritis,
paralysis,
respiratory
distress

1909

Skern, 2010

Hepatitis A

+ RNA

Liver
inflammation

1973

WHO, 2000

Rotavirus

DS RNA

Gastroenteritis

1973

Flewett and
Woode, 1978

Norovirus

+ RNA

Gastroenteritis

1972

Patel et al.,
2009

2

Diseases

Noroviruses in particular are becoming a worldwide concern. Human
noroviruses are one of the major causes of non-bacterial gastroenteritis; they are
responsible for more than 50% of general gastroenteritis and more than 90% of
viral gastroenteritis worldwide (Wobus et al., 2006; Seitz et al., 2011; Patel et al.,
2009; World Health Organization, 2011a, Seo et al., 2011). It is estimated that
there are 218,000 deaths and 1.1 million hospitalizations among children in
developing countries caused by norovirus infections each year (Seitz et al.,
2011). Norovirus is responsible for about 23% of reported waterborne diseases;

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
1
Double-stranded DNA.
2

Positive single-stranded RNA. This genome enables the virus to proceed
directly to translation following infection, without the need for genome replication
or conversion.
!

3!

there are approximately 23 million cases of norovirus each year in the United
States alone (Keswick et al., 1985; Mead et al., 1999).
1.2 NOROVIRUS
Noroviruses (previously known as Norwalk-like viruses) are singlestranded RNA viruses that are part of the Caliciviridae family (World Health
Organization, 2011a, Seo et al., 2011). They have a non-enveloped icosahedral
capsid and are usually 35-40nm in diameter (World Health Organization, 2011a).
Human noroviruses are small (27nm) enteric viruses (Bae and Schwab, 2008).
There are currently seven genotypes of norovirus, five of which are found in
humans (Lysén et al., 2009; Shirasaki et al., 2010). The noroviruses as a group
have a genome that contains 7,500 nucleotides organized into three open
reading frames (ORFs); ORF 1 encodes the RNA-dependent RNA polymerase
and other proteins, ORF 2 encodes the major capsid protein (VP1), and ORF 3
encodes the small capsid protein (VP2) (Lysén et al., 2009). ORFs 1 and 2 are
used for PCR and genotyping (Lysén et al., 2009). The virus was discovered in
1972 (Patel et al., 2009).
Gastroenteritis caused by human norovirus has a very rapid onset and
resolution (about two days), but the virus can be shed by an infected individual
for up to three weeks after infection, which leads to efficient transmission (Wobus
et al., 2006; Lopman et al., 2004). Symptoms include severe vomiting and nonbloody diarrhea (Lopman et al., 2004; Pang et al., 2000). Children and the
elderly usually have more severe symptoms, and individuals with compromised
immune systems can experience long-term infections (Centers for Disease

!

4!

Control and Prevention, 2012; Wobus et al., 2006). Norovirus is extremely
infectious; a single infectious virus particle has a 49% chance of causing an
infection (Teunis et al., 2008). Because of this, environmental contamination can
prolong outbreaks (Cannon et al., 2006). As with many enteric viruses, there is
no vaccine or drug available for prevention or treatment (Wobus et al., 2006).
Norovirus is mostly spread by the fecal-oral route, usually through
consumption of contaminated water and food but also through exposure to
vomitus (Seitz et al., 2011; Bae and Schwab, 2008; Cannon et al., 2006;
Borchardt et al., 2010; Kim et al., 2005). It affects people of all ages and typically
occurs in crowded areas; transmission through infectious vomit (from contact
with contaminated surfaces or by inhaling aerosolized particles) accounts for its
rapid spread in close quarters (Bae and Schwab, 2008; Wobus et al., 2006; Patel
et al., 2009). Norovirus is a common cause of many foodborne epidemics; it
could be responsible for 50% of all foodborne outbreaks in the US (Cannon et al.,
2006). Norovirus is also frequently associated with outbreaks of gastroenteritis
caused by contaminated drinking water (World Health Organization, 2011a; Chan
et al., 2006; Kim et al., 2005). Lysén et al. (2009) estimated that 18% of
norovirus outbreaks in Sweden were due to contaminated water. Norovirus has
even been found in fecally-polluted drinking water that had been terminally
disinfected; however these studies used PCR to detect the norovirus particles so
infectivity cannot be determined (Keller et al., 2010; Maunula et al., 2005; Gallay
et al., 2006; O’Reilly et al., 2007). Waterborne outbreaks of norovirus often occur
after heavy rains, due to wastewater entering the drinking water system (Lysén et

!

5!

al., 2009; Lloyd et al., 2007; Halonen et al., 2012) and during hot weather when
people are more frequently visiting recreational water sites (Lysén et al., 2009).
Viruses are the primary concern in groundwater contamination, due to their small
size and related ease of transport (Reynolds et al., 2008).
Human noroviruses may be resistant to environmental degradation and
chemical inactivation (Bae and Schwab, 2008; Seitz et al., 2011; Wu et al., 2005;
D’Souza et al., 2006; Reynolds et al., 2008). A study by Seitz et al. (2011) used
RT-PCR for detection of norovirus and human volunteers to determine infectivity;
norovirus samples were spiked into groundwater and allowed to sit for varying
amounts of time before being ingested by the volunteers or measured with RTPCR. The authors found that norovirus can remain detectable in groundwater for
over three years and remain infectious for at least 61 days (Seitz et al., 2011).
Borchardt et al. (2010) found that in a conventional septic system (in Wisconsin),
it would take 200 days to achieve the recommended four log10 reduction in
norovirus. Keller et al. (2010) found that murine norovirus did not degrade in raw
river water at 4 or 25˚C over 24 hours. Resistance to degradation may be due to
aggregation of the virus, which is how the virus would be found in natural (fecal)
contamination scenarios (Keswick et al., 1985).
Human noroviruses currently cannot be grown in cell culture or in an
animal model, so different forms of polymerase chain reaction (PCR) are used to
detect and quantify norovirus in samples. For studies where infectivity of the
virus needs to be measured, surrogate viruses are used in place of human
noroviruses. Feline calicivirus was historically the most common human
!

6!

norovirus surrogate, but recent studies are turning to murine norovirus as a more
appropriate surrogate. Murine (mouse) norovirus (MNV-1) is currently the only
norovirus that replicates in cell culture and in a small animal model (Wobus et al.,
2006; Duizer et al., 2004). There are many biochemical and genetic similarities
between human noroviruses and murine norovirus, including pH stability (both
are stable at high pH) (Wobus et al., 2006; Bae and Schwab, 2008; Cannon et
al., 2006). Murine norovirus is also an enteric virus, its size is similar to human
norovirus (27-35 nm), and its genome has the three open reading frames (ORFs)
that are characteristic of noroviruses (Wobus et al., 2004; Shirasaki et al., 2010).
Murine norovirus also seems to have a similar persistence against free chlorine
disinfection (when comparing viral RNA reduction rates) (Kitajima et al., 2010).
MS2 bacteriophage is a common indicator and model for human enteric
RNA viruses (Charles et al., 2009). Lim et al. (2010) found that murine norovirus
and MS2 were inactivated similarly by chlorine and chlorine dioxide. Table 2
shows how murine norovirus and MS2 bacteriophage respond to different forms
of chlorine disinfection. Both murine norovirus and MS2 have been shown to be
sensitive to many forms of chlorine disinfection; murine norovirus may be slightly
less sensitive than MS2. Free chlorine and chlorine dioxide appear to be more
effective disinfectants than monochloramine, and low temperature may have a
protective effect on the viruses.

!

7!

Table 2. Disinfection of murine norovirus and MS2 with different forms of
chlorine.
Virus

Disinfect
ant

Disinfectant Log10
Dose (mg/L) Reduction

Temperatu
re

Reference

Murine
norovirus

Free
chlorine

0.1

4.00

20-25˚C

Kitajima et
al., 2010

Monochlo
ramine

1.89

<1

4˚ C

2.5

25˚C

>4

4˚ C

>4

25˚C

0.184-0.193

3

5˚C & 25˚C

0.172-0.174

4

5˚C & 25˚C

0.255-0.288

3

5˚C & 25˚C

0.174-0.288

4

5˚C & 25˚C

Murine
norovirus
Free
chlorine
Murine
norovirus

Chlorine

MS2
Murine
norovirus
MS2

Chlorine
dioxide

0.5

Keller et
al., 2010

Lim et al.,
2010

Past studies suggested that human noroviruses were resistant to chemical
inactivation, but this may have been due to the methods used. Keswick et al.
(1985) used human volunteers to determine the infectivity of norovirus after
treatment with chlorine; they found that norovirus was more resistant to chlorine
disinfection than poliovirus, rotavirus, or f2 bacteriophage. However, more recent
research supports a different conclusion. Shin and Sobsey (2008) found that
norovirus might not be very resistant to free chlorine disinfection. Their
experiment, which used RT-PCR, found that norovirus is disinfected similar to
MS2 bacteriophage, and faster than poliovirus (Shin and Sobsey, 2008). A study

!

8!

by Cromeans et al. (2010) could achieve at least three log10 reduction of murine
norovirus within five seconds using 0.2 mg/L of free chlorine. The experiments
were run in flasks using buffered reagent-grade water of pH 7 and 8 at 5˚C, and
infectivity was determined using plaque assay (Cromeans et al., 2010). Kitajima
et al. (2010) found that free chlorine disinfection could result in approximately
four log10 reduction of murine norovirus. Aggregation of virions and the
presence of culture media have been found to hinder disinfection (Shin and
Sobsey, 2008; Floyd et al., 1976).
Many characteristics of the water being used for testing can influence the
disinfection capacity, especially temperature and pH. Both murine norovirus and
MS2 bacteriophage were found to be more sensitive to inactivation at
temperatures over 60˚C (Seo et al., 2011). Murine norovirus also appears to be
somewhat resistant to strong acids (pH 2) but tolerant of slightly acidic or neutral
conditions (pH 4 or 7) (Seo et al., 2011). Water temperature can influence the
disinfection capacity of monochloramine on murine norovirus; murine norovirus
appears to be resistant to disinfection at cooler temperatures (4˚C) (Keller et al.,
2010). Lim et al. (2010) found that less chlorine is needed at higher
temperatures to achieve the same amount of disinfection of murine norovirus and
MS2 bacteriophage but it was unclear if this was synergistic or just additive effect
of the heat and the disinfectant.

!

9!

1.3 WATER TREATMENT & DISINFECTION
There is need for improving the methods to effectively disinfect drinking
water, especially in developing countries. There are many reasons why water
may be unsafe to drink: there may not be a public water supply available, the
water supply may be unreliable and people are forced to frequently use unsafe
sources of water, or the public water supply may actually distribute unsafe water
(Wegelin et al., 1994; Mintz et al., 1995). Most of the world’s population
consumes untreated drinking water that they collect in small volumes and store in
their home; most of this water is untreated and unprotected from further
contamination (Sobsey et al., 2003). It is estimated that over 90% of diarrheal
cases could be prevented through changes to the environment, which includes
interventions to provide more clean water (McGuigan et al., 2012). Treating
drinking water at the point of use or household level has been shown to be about
twice as effective in reducing the incidence of diarrheal disease, compared to
distributed water systems that treat the water and provide points of distribution
(eg. community taps) (Clasen et al., 2006; McGuigan et al., 2012). Possible
reasons for this include the potential for recontamination in the distribution
system (due to faulty pipes, connections, etc.) and the potential for
recontamination in storage containers in the home (Wegelin et al., 1994, Mintz et
al., 1995). Treating drinking water at the point of use (in the household) has
been shown to reduce diarrheal disease by up to 30-40%, even in the absence of
improved sanitation or hygiene (Sobsey et al., 2008; Sobsey, 2002). Quick et al.
(1999) found that the combination of a chlorine disinfectant and a safe water

!

10!

storage container reduced the incidence of diarrhea by 44% within five months in
Bolivia. A similar intervention studied by Sobsey et al. (2003) found a 20.8%
reduction in diarrhea during an eight-month study in Bangladesh and a 43%
reduction during a six-month intervention in Bolivia. It is also the most costeffective method of preventing diarrheal disease (World Health Organization,
2002).
Household water treatment devices or point of use devices are now being
developed for use in areas without consistent access to clean drinking water.
Household water treatment devices that are designed to be used in developing
countries must be inexpensive and easy to use to be successful (Sobsey et al.,
2008; Clasen et al., 2006; McGuigan et al., 2012). Educational efforts also need
to be considered when implementing a point of use water treatment device in a
particular community, as the understanding that treating one’s water will prevent
diarrhea has been shown to contribute to the success of devices (Elimelech,
2006). Point of use devices are even being considered for people living in the
United States that get their drinking water from shallow groundwater sources, or
for small or rural communities (Abbaszadegan et al., 1997).
Devices utilize technology such as liquid halogen solution (chlorine,
bromine, or iodine), ultraviolet (UV) light (in the form of lamps or sunlight
(SODIS)), or filtration (through materials such as sand, gravel, ceramic, etc.).
Liquid halogen solutions are simply added to a volume of water (at a specific
concentration) and allowed to treat the water for a specific amount of time.
Liquid chlorine is normally used to disinfect drinking water on both a community

!

11!

and household scale, while bromine has historically only been used to treat water
on navy ships (Dunk, 2007). Solar water disinfection (SODIS) uses clear plastic
bottles that are filled with water and then set in the sun for several hours (more
than six) to disinfect (Wegelin et al., 1994; McGuigan et al., 2012). It aims to
treat two liters of water per person per day, or 10-15 liters for a family (Wegelin et
al., 1994). Mäusezahl et al. (2009) found no significant effect of SODIS
treatment of drinking water on the prevalence of diarrheal disease in rural Bolivia.
Slow sand filters consist of 1-1.25m of medium sand over a layer of gravel
through which the water flows (Sobsey, 2002). Particulate and microbial removal
occurs in the slime layer that forms in the top few centimeters of the sand
(Sobsey, 2002).
Table 3 shows some of the most common methods of disinfecting drinking
water, as well as the advantages, disadvantages, and costs of each. SODIS and
free chlorine or bromine treatment are by far the least expensive methods
available, and along with ceramic filters are the ones most recommended for
household use.

!

12!

Table 3. Advantages and disadvantages of various methods of drinking water disinfection.
Disinfection
Advantages
Disadvantages
Recommended
Method
for Household
Use
Boiling
Easy to use, effective
Large energy
No
against most pathogens, no requirement, no residual
disinfection by-products
Free
Safe, easy to distribute and Bad taste and smell
Yes
chlorine/bromine
use, effective against most
(chlorine), often not used
pathogens, provides
correctly, inhibited by
residual
turbidity, can form
disinfection by-products
Ceramic filter

Fairly sustainable, no
disinfection by-products,
easy to use, can be
effective against most
pathogens

Expensive, clog easily,
often crack, less
effective against viruses,
no residual, requires
regular cleaning

Yes

SODIS (solar)

Low cost, sustainable, easy
to use, mostly effective
against pathogens, no
disinfection by-products

Geographic variation; no
residual, only treats
small volumes, high
interference (turbidity),
long treatment time
Not sustainable, harder
to use, expensive

Yes

Floculation/disinfec Effective against most
tion
pathogens

!

13!

No

Reference
Wegelin et al.,
1994; Sobsey,
2002
Wegelin et al.,
1994; World Health
Organization,
2011a; McGuigan
et al., 2012; Mintz
et al., 1995;
Sobsey, 2002
Wegelin et al.,
1994; Clasen et al.,
2006; McGuigan et
al., 2012; Sobsey et
al., 2008; Sobsey,
2002
Wegelin et al.,
1994; McGuigan et
al., 2012; Sobsey,
2002
McGuigan et al.,
2012; Sobsey,
2002

Table 3 (cont’d).
Biosand filters

UV irradiation

!

Very sustainable, no
disinfection by-products,
fairly inexpensive, can be
effective against most
pathogens
Easy to use, effective
against most pathogens, no
disinfection by-products

No residual, harder to
maintain properly

No

Sobsey et al., 2008;
Sobsey, 2002

High cost, high
interference (turbitidy),
requires electricity

No

Sobsey, 2002

14!

Table 4 shows various forms of drinking water disinfectants and their
efficacy on different viruses and bacteriophages. All methods of drinking water
disinfection shown here appear to have good efficacy on various human and
model viruses. Model viruses may actually be less sensitive to disinfection than
the pathogenic viruses. Bromine also appears to be a more effective disinfectant
than chlorine.

!

15!

Table 4. Different methods of drinking water disinfection and their efficacies on viruses.
Disinfectant
Log10
Disinfectant Type of Method
Virus(es) Tested
Dose
Reduction
128
Ultraviolet
Point of use
Hepatitis A, Simian
4
2
light
(POU)
rotavirus, Poliovirus, MS2
mWs/cm
Continuous flow
apparatus, liquid 0.13 mg/L
Poliovirus
3.5
bromine (lab)
Beakers, liquid
0.07 mg/L
Reovirus
4
bromine (lab)
Bromine
MS2
4
POU
Unknown
Human adenovirus
5
POU
Unknown
MS2
5
POU
Unknown
Coliphage
1.8
Norwalk virus (norovirus) <1
Poliovirus
4
Beakers, liquid
3.75 mg/L
Simian rotavirus
4
chlorine (lab)
Human rotavirus
4
F2 bacteriophage
2
Glass test tubes,
Norovirus
3
Chlorine
1 mg/L
liquid chlorine
Poliovirus & MS2
4
(lab)
Flasks, liquid
Norovirus
3.5
0.5 mg/L
chlorine (lab)
Poliovirus
4
POU
Unknown
MS2
3
POU
Unknown
Coliphage
1
POU
6 mg/L
MS2
7
Ceramic filter POU
n/a
MS2
1-2

!

16!

Reference
Abbaszadegan et al., 1997
Floyd et al., 1976
Sharp et al., 1975
Enger et al., Draft
Manuscript
Coulliette et al., 2010
McLennan et al., 2009
Keswick et al., 1985

Shin & Sobsey, 2008
Kitajima et al., 2010
Coulliette et al., 2010
McLennan et al., 2009
Clasen et al., 2006
Brown & Sobsey, 2010

Table 4 (cont’d).
Glass tubes (lab
& field)
SODIS

!

Glass tubes (lab)

n/a

POU

n/a

f2, encephalomyocarditis
virus, bovine rotavirus

3

Somatic phage, f2,
rotavirus

3

Poliovirus

4.4

MS2

POU
Simulation (lab)

Coagulationrapid sand
filtration
Ultrafiltration
(LifeStraw)

199,800
2
mWs/cm
1,209,600
2
mWs/cm
1,836,000
2
mWs/cm

>3

rNV-VLPs

~3

MS2

4.7

17!

Wegelin et al., 1994

McGuigan et al., 2012

Shirasaki et al., 2010
Clasen et al., 2009

Another consideration for safe water is the need for safe storage
containers. Drinking water can be easily (re)contaminated when hands or
utensils come in contact with water (Mintz et al., 1995). Mintz et al. (1995) found
that mean coliform levels in water containers were significantly higher than in the
water sources.
One common method of testing the efficacy of a point of use device is to
challenge the device with bacteria, viruses, and protists, and compare the
reduction in biological organisms to the US EPA’s Guide Standard and Protocol
for Testing Microbiological Water Purifiers. The Guide Standard states that a
microbiological water purifier must be capable of reducing bacteria, virus, and
protozoan pathogens by 99.9999% (six log10 reduction), 99.99% (four log10),
and 99.9% (three log10), respectively (United States Environmental Protection
Agency, 1987). Testing is done with general and worst-case water conditions
over the life of the device (United States Environmental Protection Agency,
1987). The World Health Organization also has its own requirements for
treatment capacity of household water treatment devices. The WHO
recommends that a device be capable of four, five, and four log10 reductions of
bacteria, viruses, and protozoa, respectively, to be labeled as highly protective;
only two, three, and two log10 reductions of bacteria, viruses, and protozoa,
respectively, are required for the device to be labeled as protective (World Health
Organization, 2011c).

!

18!

Contact disinfectants are relatively new and have been incorporated into
household water treatment devices. One of these devices, the HaloSource
Waterbird, was chosen to be tested as part of this research. The device uses
physical removal and chemical disinfection to treat water to produce drinking
water, through the use of a ceramic filter and brominated N-halamine polystyrene
beads. N-halamines contain one or more nitrogen-halogen covalent bonds
(Chen & Sun, 2006; Timofeeva & Klescheva, 2011). The polystyrene beads
used in the device are 0.5mm diameter spherical polymer beads, with a large
surface area to bind to halogen ions (Enger et al., Draft Manuscript). The Nhalamine media has advantages over other methods of microbial disinfection:
better overall performance, pH stability, rechargeability, less toxicity, and a lower
price (Chen et al., 2003; Kenawy et al., 2007; Padmanabhuni et al., 2012). The
halogenated polystyrene beads do not leach decomposition products (Chen et
al., 2003). The suspected mechanism of action is that when a cell or virion
makes contact with a halogenated bead, the halogen ion is transferred to the
biological particle where it oxidizes proteins on or in the biological particle (Chen
et al., 2003; Ahmed et al., 2008; Chen & Sun, 2006; Timofeeva & Klescheva,
2011). The N-halogen bond does not break to form free halogen in the water
(Chen et al., 2003).
The device contains both a reservoir for untreated water (at the top), and a
reservoir for treated water (at the bottom), so it can act as a safe water storage
device as well as a system that disinfectants the water. As a disinfectant, the
device meets many of the features of a sustainable household water treatment

!

19!

device specified by Sobsey et al. (2008): it can consistently produce enough safe
water for an entire household for one day (9L), it has been shown to be effective
against many groups of microorganisms (Dr. Jeff Williams, Personal
Communication), it should effectively treat many different types and qualities of
water, it requires very little user time to treat the water, and it is relatively
inexpensive. As a safe water storage container, it also meets many of the criteria
proposed by the Centers for Disease Control & Prevention (CDC) and the Pan
American Health Organization (PAHO): it is made of translucent plastic that is
durable, lightweight, nonoxidizing, and easy to clean, it has a stable base, it
holds all the water that the device can disinfect at a single time, and it has a
durable and nonrusting spigot to remove water (Mintz et al., 1995).
Previous experiments with an older model of the HaloSource device
(same brominated polystyrene beads, but without the ceramic pre-filter) achieved
four to six log10 reductions of Salmonella and four to seven log10 reductions of
Vibrio cholerae (Enger et al., Draft Manuscript; Coulliette et al., 2012). There
were four log10 reductions of MS2 bacteriophage, and six log10 reduction of
human adenovirus type two (Enger et al., Draft Manuscript). Later time points
had higher inactivation, accompanied by slower flow rate and higher total
bromine residual (Enger et al., Draft Manuscript; Coulliette et al., 2010). The flow
rate was found to decrease over time, but there was no significant impact on the
halogen residual (Coulliette et al., 2010).

!

20!

Traditionally, chlorine has been the halogen of choice for use in drinking
water disinfection. Bromine has been used as a water disinfectant in hot tubs
and swimming pools, but has really only been used to treat drinking water on
navy ships (Dunk, 2007). However, recent studies suggest that certain forms of
bromine may be a good option for drinking water disinfection. Studies comparing
the chlorine and bromine forms of HaloPure N-halamine media have shown that
the bromine disinfectant is actually more effective than its chlorine counterpart.
Coulliette et al. (2010) found that the chlorine version of the media could remove
an average of 2.98 +/- 0.26 log10 of MS2, and that the bromine version could
remove an average of 5.02 +/- 0.19 log10 of MS2.
1.4 OBJECTIVES
The main objective of this study was to determine the effectiveness of the
Waterbird device in removing or inactivating MS2 bacteriophage and murine
norovirus with sub-objectives to determine if pH or added organic materials
impact the device’s effectiveness. The hypotheses are that the device will be
able to reduce both MS2 bacteriophage and murine norovirus to levels required
by the EPA standard (four log10 reduction), and that the addition of organic
materials will not impact the efficiency of the device. I hypothesize that an
increase in pH will decrease the effectiveness of the device.

!

21!

CHAPTER 2
MATERIALS & METHODS
2.1 EXPERIMENTAL DESIGN
The main objective for this study was to determine the effectiveness of the
Waterbird device in removing or inactivating MS2 bacteriophage and murine
norovirus; sub-objectives were to determine if pH or added organic materials
impact effectiveness of the device. To test these objectives, the study was
designed to test three replicates of the device with both well water and well water
seeded with approximately five percent raw sewage. The water was adjusted to
either pH 7.5 or 9. Pure culture stock of MS2 and murine norovirus were added
to each mixture of water before treatment. Each device was challenged in
triplicate for each combination of pH (7.5 or 9) and sample types (well water only
or well water with five percent raw sewage), giving a total of 36 experiments. For
pH 7.5, 18 influent samples were collected and analyzed for MS2 and murine
norovirus. Seventy-two effluent samples were collected at four time points for
each experiment for a total of 90 samples for MS2 and murine norovirus. For pH
9, nine influent samples from the sewage treatment and six influent samples from
the well water treatment were collected and analyzed for MS2; three influent
samples from each treatment were collected and analyzed for murine norovirus.
Effluent samples were collected at the same four time points (first flush, 15
minutes, 45 minutes, and 120 minutes) for each experiment, giving a total of 45
samples for the sewage treatment and 30 samples for the well water treatment
for MS2, and 15 samples for each treatment for murine norovirus.

!

22!

Figure 1 shows the device tested, where the upper reservoir was the chamber
seeded with the viruses representing the influent and the clear chamber contains
the effluent or treated water. Samples were collected from the tap. Figure 2
shows a schematic of the chlorinated system that is similar to the model of the
device tested previously (McLennan et al., 2009).

Figure 1. The HaloSource Waterbird device. The white part at the top is where
water is poured in; it flows through the ceramic pre-filter and the bromine
cartridge before being deposited in the bottom (clear) reservoir. For
interpretation of the references to color in this and all other figures, the reader is
referred to the electronic version of this thesis.

!

23!

Figure 2. Diagram of a device similar to the HaloSource Waterbird. Arrows
show the direction of water flow; water flows from the upper reservoir (U), though
the bromine cartridge (H) containing packed N-bromine beads (N), to the lower
reservoir (L), and finally is deposited in the lower reservoir (L) where it can flow
out of the tap (McLennan et al., 2009).
To test the effectiveness of the Waterbird device, diluted mixtures of MS2
bacteriophage and murine norovirus were treated through the device and the
concentrations of virus before and after treatment were measured and compared.
A control of well water (from Williamston, MI) was used, as well as well water
with a low volume of raw sewage added (from the East Lansing wastewater
treatment plant). Two pH values were used (pH 7.5 and 9). Three Waterbird
devices were analyzed in triplicate, giving nine replicates for each treatment and
pH combination (36 total samples).
The Waterbird devices were pre-flushed with five liters of NanoPure water
the night before each set of experiments. The influent mixture used to test the
devices was made with 5L of well water. For the runs with raw sewage,
!

24!

approximately 2-4% raw sewage was added. The exact amount of raw sewage
added was adjusted based on the COD measured each day (the COD was
adjusted to 8.1 mg/L (final concentration)). The influent was adjusted for pH (to
either 7.5 or 9) using hydrochloric acid or sodium hydroxide, and temperature
was measured using a digital thermometer. MS2 suspension in TSB (2mL) and
purified murine norovirus in PBS (3mL) were added to the influent, and the
mixture was stirred for 15 minutes. The mixed influent was poured into the top
reservoir of the device (with the bromine cartridge and ceramic pre-filter already
in place), and allowed to flow through the device by keeping the tap in the bottom
reservoir open and flowing. Samples were taken of the untreated influent and of
the treated water at first flush (0 minutes), 15 minutes, 45 minutes, and 2 hours.
Flow rate (mL per minute) was also measured at each time point. The time from
the start of the experiment to the first flow at the tap was recorded.
Bromine residual was measured using the Hach colorimetric DPD method
was used, with an adjustment for bromine (total chlorine * 2.25) (Coulliette et al.,
2010) at each time point.
2.2 BROMINE DEMAND OF THE ADDED SEWAGE
To determine the optimum concentration of raw sewage to use in the
challenges, a bromine consumption experiment was run. Three concentrations
(1%, 5%, and 10%) of raw sewage (from the East Lansing wastewater treatment
plant) were analyzed to see how quickly they depleted the available bromine. An
ideal concentration would likely show about 80% reduction in bromine residual
over two hours (Eaton et al., 1992). The colorimetric DPD method was used and

!

25!

adjusted for bromine (total chlorine * 2.25). A solution of hypobromous acid was
prepared from 1.725 mL hydrogen bromide, 10.25 mL sodium hypochlorite, and
448.3 mL NanoPure water. Half a milliliter of the solution was added to 200mL of
diluted raw sewage (diluted with NanoPure water), and allowed to react while
continuously stirred for five hours. The 1% solution showed 54% reduction in
bromine, the 5% solution showed 87% reduction, and the 10% solution showed
89% reduction; 5% raw sewage was determined to be the optimum sewage
concentration for the challenge experiments.
2.3 CHEMICAL OXYGEN DEMAND OF THE ADDED SEWAGE
Since raw sewage varies by day in concentration of organic matter,
chemical oxygen demand (COD) was chosen to standardize each experiment
where raw sewage was added to the well water. The Hach COD kit was used
(having a range of 0-1500 mg/L). Initial studies were undertaken to set up the
experimental design based on COD measurements and then used to set up each
experiment. Raw sewage samples were taken in the morning from the East
Lansing wastewater treatment plant and immediately analyzed. Samples were
added to the pre-made reagent tubes, and then digested on a heating block for
two hours. Samples were then analyzed with a spectrophotometer to determine
the COD. The absorbance reading from the spectrophotometer was converted to
parts per million (PPM). The volume added each day was determined by dividing
the day’s COD PPM value by the initial COD value (162.5 PPM) to get a dilution
factor; the initial volume (250 mL) was divided by the dilution factor to get the
day’s required volume of raw sewage (Table 5). Table 5 shows the chemical

!

26!

oxygen demand values and percent raw sewage added for each raw sewage
experiment. The overall COD concentration for the 5L influent was adjusted to
be 8.1 mg/L.
Table 5. Chemical oxygen demand values (in mg/L) and percent raw sewage
added for each raw sewage experiment.
Experiments

COD (mg/L) of Raw
Sewage

Percent Raw Sewage (in
5L)

3,4,5

233

3.5

9,10,11

401

2.0

14,15,16

353

2.3

17,18,19

348

2.1

32,33,34

263

3.1

38,39,40

208

3.9

2.4 VIRUS PROPAGATION
Murine norovirus was propagated in RAW 264.7 host cells (ATCC# TIB71); these cells are mouse macrophage cells.

Frozen RAW 264.7 cells were

2

thawed and started in 25cm flasks with 8 mL growth medium (high-glucose
DMEM (Hyclone, Logan, UT, #SH30243.02) with 10% low-endotoxin fetal bovine
serum (FBS) (Hyclone, Logan, UT, #SH30070.03), 1% HEPES free acid
(Amaresco, Solon, OH, #J848) 1% MEM NEAA (Lonza, Walkersville, MD, #13114E), 1% penicillin/streptomycin (Hyclone, Logan, UT, SV30010), and 1% Lglutamine (Hyclone, Logan, UT, SH30034.02)) and incubated at 37˚C and 5%
CO2. Once the cells showed >80% confluence in the flasks they were removed
by scraping and split at a 1:8 ratio to create new flasks. Subsequent passages of
!

27!

the cells were grown in maintenance media (high-glucose DMEM with 2% lowendotoxin fetal bovine serum (FBS), 1% HEPES free acid, 1% MEM NEAA, 1%
penicillin/streptomycin, and 1% L-glutamine).
Murine norovirus stock was obtained from Dr. Kellogg Schwab at Johns
Hopkins University. To propagate the murine norovirus, >80% confluent RAW
2

264.7 cells in 150 cm flasks were infected with 6 ml of diluted virus stock
(diluted in maintenance media); the virus stock was diluted so that there are not
more virus particles than cells in the flasks. Infected flasks were incubated for
one hour at 37˚C and 5% CO2, and were rocked every 15 minutes. After one
hour of incubation, the innoculum was removed from the flasks and complete
DMEM without FBS was added. Flasks were incubated until they showed
cytopathic effect (CPE) for two to seven days. The flasks were then frozen at 80˚C overnight and subsequently thawed at room temperature; this was
repeated three times. The resulting cell/virus mixture was filtered through a 0.45
and 0.22 micron filter (Millipore) and then frozen at -80˚C. Enger et al. (Draft
Manuscript) found that 1X Eagle’s MEM exerted a bromine demand, as observed
by significantly lower bromine residuals, and protected MS2 from inactivation, so
the filtered virus stock was ultrapurified to remove all media. Frozen virus stock
was thawed and ultrapurified using Amicon ultrafilters (Millipore, Billerica, MA,
#UFC910024) to remove the MEM and concentrate the virus. Resulting stock (at
7

a concentration of approximately 1 x 10 PFU/mL as determined by plaque
assay) was frozen at -80˚C. Throughout the course of the study, approximately
!

28!

200 flasks were prepared and 120 ml of a concentrated purified virus stock was
prepared.
MS2 (ATCC# 15597-B1) was propagated in E. coli Famp host bacterial
cells (ATCC# 700891). Frozen bacterial cells (frozen at -80˚C) were thawed and
grown in tryptic soy broth (TSB) overnight at 37˚C. Cells were transferred to
fresh TSB and incubated for four to six hours. Two milliliters of diluted MS2 stock
(diluted in phosphate buffered water (PBW)) were added to 0.5 mL of the E. coli
host cells and 2.5 mL of melted 1.5% trypicase soy agar (TSA). The melted TSA
was boiled and then equilibrated to 50˚C before use. The resulting mixture was
2

poured onto a thin layer of solidified TSA in a 75 cm flask, thus described as the
double agar overlay method. Plates were incubated for 16-24 hours at 37˚C,
after which plaques or lysis in the bacterial monolayer could be visualized and
counted. If sufficient concentrations of MS2 bacteriophage were present in the
flasks (the flasks should have a lacey pattern, where the plaques are growing
into one another), 30 ml of sterile TSB was added to each flask and the flasks
were rocked at 4˚C for an hour to elute the viruses from the bacterial monolayer.
The resulting cell/bacteriophage suspension was pipetted from the flask and
filtered through 0.45 and 0.22 filters. The purified suspension was transferred to
a sterile centrifuge tube, covered in aluminum foil, and stored at 4˚C.

!

29!

2.5 VIRUS QUANTIFICATION
For the quantification of murine norovirus, the stock solution, influent and
treated samples were all assayed using the plaque assay protocol. The treated
water samples contained sodium thiosulfate (final concentration of 1% w/v) to
neutralize the disinfectant. One-half milliliter of a 2% sterile sodium thiosulfate
solution was added to the 15 ml tubes. Samples were frozen at -80˚C. Richards
et al. (2012) have shown that norovirus can be frozen and thawed up to 14 times
without decreasing infectivity. When ready to be processed, frozen samples
were thawed and diluted with maintenance media. Samples were analyzed on
RAW 264.7 host cells using the plaque assay method described below.
Confluent flasks of RAW 264.7 host cells were split to make 6-well 9.6
2

6

cm cell culture plates (Corning, Corning, NY, #3516); approximately 3 x 10

cells were added to each well. The well plates were incubated for 24 hours in a
37˚C 5% CO2 incubator. After 24 hours, the growth media was removed from
each well and wells were infected with 1:10 dilutions of the samples (diluted in
maintenance media) (0.5 mL innoculum was used per well). Duplicates were
processed of each sample, as well as positive and negative controls. Well plates
were inoculated and then incubated and rocked for one hour at room
temperature. Overlay media was prepared using one half 1.5% agarose (1.5 g
agarose (Cambrex, Rockland, ME, #50111) dissolved in 50 mL NanoPure water,
autoclaved for 15 minutes, and brought to 48˚C) and one half 2xMEM (2xMEM
(Sigma, St. Louis, MO, #M3024) with 10% low-endotoxin FBS, 2% L-glutamine,
!

30!

2% penicillin/streptomycin, and 1% HEPES) brought to 37˚C. Immediately
before use the agarose and 2xMEM were combined and thoroughly mixed. Two
milliliters of the overlay solution were added to each well, and the plates were left
to solidify at room temperature for 30 minutes. After solidifying, the plates were
incubated at 37˚C and 5% CO2 for 24 hours. After 24 hours, two milliliters of
fresh overlay solution (with 2% neutral red (Sigma, St. Louis, MO, #N2889)) were
added on top of the other overlay in each well to stain the living cells and
visualize the plaques. The plates were allowed to solidify at room temperature
and then incubated for 24 hours at 37˚C and 5% CO2. After 24 and 48 hours,
plaques were counted to obtain a measurement of PFU per milliliter of sample.
For quantification of MS2 samples, a double agar overlay method was
used. Sodium thiosulfate (final concentration of 1% w/v) was added to the treated
water samples as described above. A 1/10 dilution series were prepared with
PBS and then plated with E. coli Famp using the double agar overlay method as
described above. Plates were incubated at 37˚C for 16-24 hours, and then read
to determine the plaque forming units (PFU) per milliliter sample.
2.6 STATISTICAL ANALYSIS
For the MS2 and murine norovirus results, the concentrations of virus
before and after treatment were measured and compared to determine a log10
reduction for each time point (the log10 of the effluent was subtracted from the
log10 of the influent for each sample). The data were also converted to ratios
!

31!

(the effluent virus concentration divided by the influent virus concentration) and
then arcsine transformed. For the MS2 results, a repeated measures ANOVA
was used to determine whether or not the water type (sewage or pH) or other
variables (sampling time, cartridge, or ceramic) had a significant effect on the
efficacy of the device in removing the virus. A Tukey’s post-hoc test was used to
distinguish which groups were significantly different. For the murine norovirus
results, an ANOVA was used to look for significant effects. A Tukey’s post-hoc
test was used to distinguish which groups were significantly different. Linear
regression analysis was used to look for correlations between the log10
reductions of the two viruses and the physical parameters (flow rate and bromine
concentration). P-values equal to or less than 0.05 were considered significant.

!

32!

CHAPTER 3
RESULTS
3.1 INTRODUCTION
This study set out to determine the effectiveness of the bromine-based
disinfectant HaloSource Waterbird device in reducing or inactivating MS2
bacteriophage and murine norovirus as model enteric viruses. Sub-objectives
were to determine the effect (if any) that the addition of raw sewage or the
manipulation of pH had on the overall effectiveness of the device.
To test these objectives, the study was designed to test three replicates of
the device with both well water and well water seeded with approximately five
percent raw sewage. The water was adjusted to either pH 7.5 or 9. Each device
was challenged in triplicate for each combination of pH (7.5 or 9) and sample
types (well water only or well water with five percent raw sewage), giving a total
of 36 experiments. For pH 7.5, 18 influent samples (nine for well water and nine
for sewage) were collected and analyzed for MS2 and murine norovirus.
Seventy-one effluent samples were collected at four time points (first flush, 15
minutes, 45 minutes, and 120 minutes). For each experiment a total of 89
samples for MS2 and murine norovirus were analyzed. For pH 9, 18 influent
samples (nine for well water and nine for sewage) were collected and analyzed
for MS2; three influent samples from each treatment were collected and analyzed
for murine norovirus. Effluent samples were collected at the same four time
points for each experiment, however if the first flush was beyond 15 minutes this
was considered first flush sample with no 15 minute sample collected, giving a

!

33!

total of 44 samples for the sewage treatment and 40 samples for the well water
treatment for MS2, and 14 samples for the sewage treatment and 13 samples for
the well water treatment were analyzed for murine norovirus.
There were several issues with testing this device as a whole, mostly
concerning the variability inherit in the device itself and in running samples
across a long time period. The variability in the device comes from differences
between cartridges and ceramic pre-filters, and variability within the cartridge
itself (this shows in the variability in the bromine that is released or the flow rate
through the cartridge). Variability due to the experimental design comes from the
raw sewage addition. Variation due to the long time period during which
experiments were run comes from differences in batches of virus stock produced
and variation in sewage used (which could impact the MS2 concentration if
additional natural bacteriophage is introduced). The study was designed to
attempt to control the variation from these factors, but it is never possible to
control all variation in an experiment. The variability due to the device itself could
not be controlled. The same three ceramic pre-filters were used for all of the
experiments, but the cartridges had to be replaced partway through the study
due to inefficient performance and apparent exhaustion of the bromine. The raw
sewage was measured for the chemical oxygen demand (COD) and adjusted so
that the same amount of organic material was added to each experiment,
however while this controlled for demand, this did not control for a variation in
naturally-occurring coliphage that may be observed with the host used.

!

34!

The

same procedures were used to propagate and process virus stock, in attempt to
reduce variability.
3.2 MS2 VIRUS
3.2.1 INFLUENT CHARACTERIZATION
Table 6 shows the average influent concentrations for the experimental
challenges run using MS2. The goal was to keep the influent concentrations for
each virus as consistent as possible between replicates. The MS2 stock used
gave an average influent concentration of 6.9, 6.7, 7.3, and 7.7 log10 for the
experiments at pH 7.5 for well water and sewage and pH 9 for well water and
sewage, respectively. The pH 9 experiments had slightly higher influent
concentrations of MS2. The same MS2 stock was used for all experiments, so
the variation might have been due to clumping of the virus particles (and not
enough mixing). If the virus particles were clumped, the concentration would be
underestimated in the assay. Differences between well water and sewage
samples could also be due to the occasional higher level of naturally occurring
coliphage added from the sewage.
Table 6. Influent concentrations for MS2 for the pH, sewage and well water
variables. Shown are the average values for virus influent concentrations for
each combination of pH and water treatment variables.
pH 7.5

pH 9

Well Water
MS2 Influent
Concentration
(log10
PFU/mL)

!

Sewage

Well Water

Sewage

6.9 (n=9)

6.7 (n=9)

7.3 (n=9)

7.7 (n=9)

35!

Figure 3 shows the average MS2 influent concentrations for the pH 7.5
samples. There was an average MS2 influent concentration of 6.9 log10 for the
well water samples and 6.7 log10 for the sewage samples (Figure 3).

Figure 3. Average MS2 influent concentrations (in log10) for the pH 7.5 samples.
N=9 for well water and sewage (N=18 total).
Figure 4 shows the average MS2 influent concentrations for the pH 9
samples. There was an average MS2 influent concentration of 7.3 log10 for the
well water samples and 7.7 log10 for the sewage samples (Figure 4).

!

36!

Figure 4. Average MS2 influent concentrations (in log10) for the pH 9 samples.
N=9 for well water and sewage (N=18 total).
Overall, there was an average of 7.2 log10 of MS2 for all experiments
when measuring the untreated influent samples. There was found to be no
significant difference in the MS2 influent concentrations between the well water
and sewage samples (p=0.754), but there was a significant difference between
pH treatments (p=0.011) (Figure 5). Figure 5 shows the average MS2
concentrations between the pH 7.5 samples and the pH 9 samples. The pH 7.5
samples had a mean of 6.8 log10, while the pH 9 samples had a higher mean of
7.5 log10. Looking closer at the data, it appears that two sets of experiments

!

37!

(one in the pH 7.5 sewage experiments and one in the pH 9 sewage
experiments) may be responsible for the difference between pH variables. One
set of pH 7.5 sewage experiments had an average MS2 influent concentration of
4.9 log10, while the rest of those experiments had an average of 6.9 log10; that
set of experiments would have lowered the overall average. One set of pH 9
sewage experiments had an average MS2 influent concentration of 8.3 log10,
while the rest of those experiments had an average of 7.6 log10; that set of
experiments would have raised the overall average. Due to these two
experimental results there was a statistically significant difference in the influent
concentrations of the two pH experiments for MS2. When those two influent
values are removed from the analysis, there is no significant effect of pH
(p=0.293) on the influent concentrations. The role of the influent concentration in
relationship to reduction of MS2 is discussed later.

!

38!

Figure 5. Average MS2 influent concentrations (in log10) for all samples. N=18
for pH 7.5 and n=18 for pH 9 (N=36 total).
3.2.2 EFFLUENT CHARACTERIZATION
The time of the first sample (t0, or first flush) differed with each
experiment. This is probably due to variation between cartridges and ceramic
pre-filters used in each unit. For each combination of pH and water condition,
three devices were tested in triplicate, giving a total of nine experiments. The
results for bromine residuals, flows, device performance and effect of sewage
and pH are presented below.

!
!

39!

3.2.2.1 BROMINE RESIDUAL
Table 6 shows the average bromine residual values (+/- standard
deviation) and which cartridges were used for each combination of influent
experimental conditions (pH 7.5 or 9, well water or sewage). The bromine
residual was not significantly different between cartridges (p=0.127) (Figure 6).
Figure 6 shows the bromine residual for each cartridge used. The cartridges 4, 5
and 6 had bromine levels that were one third of the levels of the cartridges 1, 2
and 3. Only the values from the pH 9 sewage experiments were compared,
since that was the only group of similar samples that used all six cartridges.

!

40!

Figure 6. Bromine residual (mg/L) between cartridges used (for the pH 9 sewage
samples). Values were averaged for all time points. The bromine residual was
not significantly different between cartridges (p=0.201). N=4 for cartridge 1,2,3,
N=8 for cartridge 4, N=7 for cartridge 5, N=6 for cartridge 6, N=33 total. The *
symbol indicates an extreme outlier.
While the addition of sewage to the well water had no statistically
significant effect on bromine residual (p=0.409) (Table 7), there was a slight
increase in the bromine residuals when sewage was added.

!

41!

Table 7. Average bromine residual (+/- standard deviation) and cartridges used
for each type of influent (pH 7.5 or 9, well water or sewage).
pH 7.5

pH 9

Well Water

Sewage

Well Water

Sewage

N=35

N=36

N=31

N=34
1, 2, 3*

Cartridges
Used

Bromine
Residual
(mg/L)

1, 2, 3

1, 2, 3

4, 5, 6

4, 5, 6**
1-6***

0.70 +/- 0.45*
3

0.50 +/- 0.42

0.54 +/- 0.48

0.19 +/- 0.29

(N=12)

0.22 +/- 0.62**
(N=21)
Avg. 0.39 +/0.61***

Figure 7 shows the bromine residual for each cartridge used in the pH 7.5
samples. There was no statistical difference between cartridges (Figure 7),
however the lower levels could still have had an effect on disinfection. This will
be addressed later.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
3
N=12 is for cartridges 1,2,3 combined (*), and N=21 is for cartridges 4,5,6
combined (**). The average bromine residual shown is for all six cartridges
combined (***).
!

42!

Figure 7. Bromine residual (mg/L) between cartridges used (for the pH 7.5
samples). N=71.
The bromine residual showed a statistically significant difference between
timed samples (p<0.001) with increasing residuals; the first flush sample differed
from the 45 minute sample (p=0.001) and the 120 minute sample (p<0.001), and
the 15 minute sample differed from both the 45 minute sample (p=0.044) and the
120 minute sample (p=0.010) (Figure 8). Figure 8 shows the bromine residual
over sampling time, and compares the two pH variables. The first flush samples
had an average bromine residual of 0.16 +/- 0.31 mg/L, the 15 minute sample
had an average of 0.27 +/- 0.22 mg/L, the 45 minute sample had an average of

!

43!

0.57 +/- 0.61 mg/L, and the 120 minute sample had an average of 0.62+/- 0.48
mg/L.

Figure 8. Average bromine residual (mg/L) for well water and sewage addition
samples for pH 7.5 and 9 over sampling time. N=135. The º indicates an outlier,
and the * indicates an extreme outlier.
There was no significant effect of pH (p=0.402) on bromine residual
(Figure 9). Figure 9 shows the bromine residual between the two pH variables.
Only the experiments that used cartridges 1, 2, or 3 were considered (only the
sewage samples could be analyzed). For this subset of data, the pH 7.5

!

44!

samples had an average bromine residual of 0.54 +/- 0.48 mg/L; the pH 9
samples had an average bromine residual of 0.70 +/- 0.45 mg/L.

Figure 9. Bromine residual (mg/L) between the two pH variables. N=48 for
cartridges 1, 2 and 3 only. The º indicates an outlier.
3.2.2.2 FLOW RATE
Overall, the flow rate did not differ significantly between treatments
(p=0.626), pH (p=0.869), or cartridges (p=0.433) (Table 13). There was a
significant difference in flow rates between timed samples (p<0.001) as expected
with the decrease in the influent volume and head; the first flush sample was
different than all other samples (p<0.001), the 15 minute sample was different
than the 45 minute sample (p=0.044) and the 120 minute sample (p<0.001), and
!

45!

the 45 minute sample was also significantly different from the 120 minute sample
(p=0.004) (Figure 10). Figure 10 shows the flow rate between sampling times.
The first flush samples had an average flow rate of 61 +/- 25 mL/min, the 15
minute samples had an average of 38 +/- 14 mL/min, the 45 minute samples had
an average of 27 +/- 14 ml/min, and the 120 minute samples had an average of
13 +/- 5.3 mL/min.

Figure 10. The flow rate (in mL/min) for all samples, separated by sampling time.
There was a significant difference in flow rates between samples (p<0.001); the
first flush sample was different than all other samples (p<0.001), the 15 minute
sample was different than the 45 minute sample (p=0.044) and the 120 minute
sample (p<0.001), and the 45 minute sample was also significantly different from
the 120 minute sample (p=0.004). N=135.

!

46!

3.2.2.3 DEVICE PERFORMANCE
For each of the influent and effluent samples, MS2 concentration was
determined by plaque assay. The log10 reduction values were calculated by
taking the log10 concentration of the influent minus the log10 concentration of the
effluent. The same influent log10 concentration was used for four corresponding
effluent log10 concentrations (in four separate calculations). Each effluent
sample (from each of the four time points when effluent samples were taken) had
a separate log10 reduction value calculated; thus, each experiment has one
influent concentration and four effluent concentrations or log10 reduction values.
For MS2, the detection limit was one PFU/mL; non-detect samples were
considered to have one PFU/mL for statistical analysis.
The average influent concentration of MS2 was 7.16 log10 (n=36), and the
average effluent concentration was 1.79 log10 (n=135 from all timed samples
from all experiments). The overall reduction of MS2 was 5.4 log10. The device
was capable of 4.7, 5.7, 5.3, and 5.9 log10 reduction of MS2 virus for the
experiments at pH 7.5 for well water and sewage and pH 9 for well water and
sewage, respectively (Table 8). Table 8 shows the average (+/- standard
deviation) flow rate, bromine residual, MS2 influent and effluent concentrations,
and log10 reduction. There was no correlation between influent concentration of
2

MS2 and the log10 reduction (R =0.000).
!

47!

Table 8. Physical characteristics (flow rate and bromine residual), and MS2
influent and effluent concentrations and log10 reductions. The average values
(+/- standard deviation) are given for each measurement for each combination of
pH and water treatment variables.
pH 7.5

pH 9

Well
Water

Sewage

Well
Water

n=9

n=9

n=9

n=3

n=6

Cartridge

1, 2, 3

1, 2, 3

4, 5, 6

1, 2, 3

4, 5, 6

Flow Rate
(mL/min)

37 +/- 23

32 +/- 20

32 +/30

42 +/- 20

33 +/- 28

Bromine
Residual
(mg/L)

0.50 +/0.42

0.54 +/0.48

0.19 +/0.29

0.70 +/- 0.45

0.22 +/- 0.62

MS2 Influent
Concentration
(log10)

6.9

6.7

7.3

7.6

7.7

MS2 Effluent
Concentration
(log10)

2.3

1.0

2.0

3.1

1.3

MS2 Log10
Reduction

4.7

5.7

5.3

4.6

6.7

Sewage

For the pH 7.5 treatment, there was an average influent MS2
concentration of 6.8 log10, an average effluent concentration of 1.6 log10, and an
average log10 reduction of MS2 of 5.2 log10 (Table 9 and 10). Table 9 shows the
average MS2 concentrations and device characteristics (average flow rate and
bromine residual) in the effluent post treatment for well water samples at pH 7.5.
Table 10 shows the average MS2 concentrations and device characteristics

!

48!

(average flow rate and bromine residual) in the effluent post treatment for
sewage addition samples at pH 7.5.
Table 9. Average MS2 concentrations and device characteristics in the effluent
post treatment for well water samples at pH 7.5.
Experim Cartr Time
# of
Average MS2 Average Average
ent #
idge (min) post Samples Concentration Flow
Bromine
addition of
Rate
Residual
(log10)
influent to
(mL/min (mg/L)
upper
)
reservoir
0

0

3

0.0

46

0.41

3

3.1

41

0.19

3

2.9

39

0.49

120
2

4.1

45
2, 8, 12,
25

3

15;15;17

1

4

2;4;10

6, 13, 23

3

1.7

15

0.83

0

4

0

0

0

4.0
3;3;5;24

53

0.11

3

4.4

39

0.34

45

4

3.5

27

0.62

120

4

1.4

14

0.87

0

2

3.8

0

0

4;7

2

0.0

84

0.11

15

2

4.7

48

0.28

45

2

3.9

40

0.62

120

3

0.3

15;15;17

7, 24

4

2

2.7

18

1.18

39

0.50

Total
n=44

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
4
Time of 0 (zero) is the untreated influent sample.
!

49!

Table 10. Average MS2 concentrations and device characteristics in the effluent
post treatment for sewage addition samples at pH 7.5.
Experim Cartr Time
# of
Average MS2 Average Average
ent #
idge (min) post Samples Concentration Flow
Bromine
addition of
Rate
Residual
(log10)
influent to
(mL/min (mg/L)
upper
)
reservoir

3

0.0

43

0.19

3

1.2

30

0.26

3

1.3

28

0.64

3

0.8

12

0.56

0

3

6.8

0

0

3

0.0

36

0.11

15

3

2.3

31

0.30

45;45;46

3

0.6

28

0.68

120

3

0.7

15

0.75

0

3

6.6

0

0

3;3;9

3

0.8

64

0.64

15

3

2.0

43

0.43

45

3

1.4

31

0.99

120

3

0

2;3;5

3, 9, 16

0

120
2

6.7

45
4, 10, 14

3

15

1

5

3;11;13

5, 11, 15

3

0.8

18

0.94

32

0.54

0

Total
n=45

For the pH 9 treatment, there was an average influent MS2 concentration
of 7.5 log10, an average effluent concentration of 2.0 log10, and average log10
reduction of MS2 was 5.6 log10 (Table 11 and 12). Table 11 shows the average
MS2 concentrations and device characteristics (average flow rate and bromine
residual) in the effluent post treatment for well water samples at pH 9. Table 12
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
5
Time of 0 (zero) is the untreated influent sample.
!

50!

shows the average MS2 concentrations and device characteristics (average flow
rate and bromine residual) in the effluent post treatment for sewage addition
samples at pH 9.
Table 11. Average MS2 concentrations and device characteristics in the effluent
post treatment for well water samples at pH 9.
Experim Cartr Time
# of
Average MS2 Average Average
ent #
idge (min) post Samples Concentration Flow
Bromine
addition of
Rate
Residual
(log10)
influent to
(mL/min (mg/L)
upper
)
reservoir
29, 36,
44

4

0

6

3

0

0

7.3

9;10;13

62

0.05

3

3.4

31

0.07

45

3

3.3

19

0.17

120
5

1.7

15

30, 37,
45

3

3

2.8

11

0.35

0

3

0

0

7.4
20;22;37

61

0.06

3

1.9

11

0.11

120
6

0.9

45
31, 35,
46

3
3

0.0

8

0.35

0

3

0

0

7.1
12;20;26

3

1.7

95

0.07

14

1

5.1

26

0.11

45

3

2.4

17

0.26

120

3

1.2

8

0.44

32

0.19

Total
n=40

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
6
Time of 0 (zero) is the untreated influent sample.
!

51!

Table 12. Average MS2 concentrations and device characteristics in the effluent
post treatment for sewage addition samples at pH 9.
Experi
Cartr Time
# of
Average MS2 Average Average
ment #
idge (min) post Samples Concentration Flow
Bromine
addition of
Rate
Residual
(log10)
influent to
(mL/min (mg/L)
upper
)
reservoir

1

5.1

59

0.56

1

5.0

40

1.13

1

3.0

19

1.01

0

1

7.4

0

0

1

0.0

64

0.11

1

4.2

56

0.23

1

4.6

46

0.90

1

3.0

15

1.13

0

1

8.3

0

0

1

0.0

80

0.11

15

1

4.6

48

0.79

45

1

3.9

38

1.24

120

1

3.3

20

1.13

0

2

7.7

0

0

6

2

0.0

42

0.02

15;24

2

3.7

31

0.10

45

2

5.7

29

1.54

120

2

4.0

11

0.17

0

2

7.7

0

0

10;18

2

0.0

83

0.02

17

1

0.0

28

0.07

45

5

0.11

3

33, 38

24

120

4

0.0

45

32, 40

1

15

3

0

4

19

0

120
2

7.3

45
17

1

15

1

7

2

18

2

0.0

19

0.09

0

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
7
Time of 0 (zero) is the untreated influent sample.
!

52!

Table 12 (cont’d).
120

10

0.09

0

2

7.8

0

0

2

0.0

84

0.02

45

2

0.0

16

0.09

120

6

0.0

25

34, 39

2

2

0.0

10

0.09

38

0.47

Total
n=42
3.2.2.3.1 EFFECT OF pH AND SEWAGE ADDITION

For the pH 7.5 samples, the sewage samples showed higher log10
reduction of MS2 than the well water samples (p=0.005) (Figure 11). Figure 11
shows the log10 reduction of MS2 between well water and sewage addition
samples, for pH 7.5. The well water samples had an average influent
concentration of MS2 of 6.9 log10 and an effluent concentration of 2.3 log10,
giving an average log10 reduction of 4.7. The sewage samples had an average
influent concentration of MS2 of 6.7 log10 and an effluent concentration of 1.0
log10, giving an average log10 reduction of 5.7.

!

53!

Figure 11. Log10 reduction of MS2 between well water and sewage addition
samples (pH 7.5). The sewage samples did have significantly different log10
reduction compared to the well water samples (p=0.005). N= 9 for well water and
9 for sewage (N=18 total).
For the pH 9 samples, the well water samples had an average influent
concentration of MS2 of 7.3 log10 and an effluent concentration of 2.0 log10,
giving an average log10 reduction of 5.3. The sewage samples had an average
influent concentration of MS2 of 7.7 log10 and an effluent concentration of 1.9
log10, giving an average log10 reduction of 5.9 There was no significant effect of
sewage addition on the log10 reduction of MS2 (p=0.389), however the data did

!

54!

show an increased removal overall with the addition of sewage. Only the
samples that used cartridges 4, 5, and 6 were used for this comparison, which
may have influenced the statistical significance.
Overall, the sewage samples showed an average log10 reduction of MS2
of 5.8 and the well water samples had an average log10 reduction of 5.0 (Figure
12). Figure 12 shows the log10 reduction of MS2 between well water and
sewage addition samples.

Figure 12. Log10 reduction of MS2 between well water and sewage addition
samples (pH 7.5 and 9 samples together). N= 18 for well water and 18 for
sewage (N=36 total).

!

55!

When pH effects were compared with the sewage only samples that used
cartridges 1, 2, or 3 it was found that there was a statistically significant effect of
pH on the log10 reduction of MS2 (p=0.006) (Figure 13). Figure 13 shows the
log10 reduction of MS2 between pH 7.5 and pH 9 samples for this comparison.
Otherwise pH could not be shown to impact reductions.

Figure 13. Log10 reduction of MS2 between pH 7.5 and pH 9 sewage samples.
There was a significant difference in the log10 reduction of MS2 between the
samples (p=0.006). N= 9 for pH 7.5 and 3 for pH 9 (N=12 total).

!

56!

3.2.2.3.2 DEVICE VARIABILITY
For the pH 7.5 samples, the log10 reduction of MS2 starts high at the first
flush sample (2-38 minutes) (about 7 log10), drops to around 4 log10 when the
second sample was collected at 15 minutes, and then goes back up to about 5
log10 for the 45 and 120 minute time points. There was a significant effect of
sampling time (p<0.001); the first flush sample differed significantly from both the
15 minute sample (p<0.001) and the 45 minute sample (p<0.001). The 15
minute sample also differed significantly from the 120 minute sample (p=0.005)
(Figure 14). Figure 14 shows the log10 reduction of MS2 for the pH 7.5 samples.
There was no significant effect of cartridge (p=0.979) or ceramic (p=0.979).
Cartridge 1 had an average log10 reduction of MS2 of 5.4, cartridge 2 had an
average of 5.3, and cartridge 3 had an average of 4.8. Ceramic 1 had an
average log10 reduction of MS2 of 5.2, ceramic 2 had an average of 5.3, and
ceramic 3 had an average of 5.1.

!

57!

Figure 14. Log10 reduction of MS2 for the pH 7.5 samples. There was also a
significant difference between samples (p<0.001); the first flush (2-38 minutes)
sample differed significantly from both the 15 minute sample (p<0.001) and the
45 minute sample (p<0.001). The 15 minute sample also differed significantly
from the 120 minute sample (p=0.005). N=18. The º indicates an outlier, and
the * indicates an extreme outlier.
For the pH 9 samples, the log10 reduction of MS2 started around 7 log10
for the first flush sample, dropped to around 4 log10 for the 15 minute sample,
and then increased back to about 7 log10 for the 45 and 120 minute samples.
There was a significant effect of sample (p=0.002), but only the first flush sample
and 45 minute sample differed significantly (p=0.004) (Figure 15). Figure 15

!

58!

shows the log10 reduction of MS2 for the pH 9 samples. There was no
significant effect of ceramic (p=0.270) or cartridge (p=0.542). Cartridge 4 had an
average log10 reduction of MS2 of 4.7, cartridge 5 has an average of 7.1, and
cartridge 6 has an average of 6.1. Ceramic 1 had an average log10 reduction of
MS2 of 5.9, ceramic 2 had an average of 5.3, and ceramic 3 had an average of
5.6.

Figure 15. Log10 reduction of MS2 for the pH 9 samples. There was a
significant effect of sampling time (p=0.002), but only the first flush sample and
45 minute sample differed significantly (p=0.004). N=18. The * indicates an
extreme outlier.

!

59!

Overall, there was a significant effect of sampling time on the log10
reduction of MS2 (p<0.001); the first flush and the 15 minute samples (p<0.001),
the first flush and 45 minute samples (p=0.001), the 15 minute and 120 minute
samples (p=0.001), and the 45 minute and 120 minute samples (p=0.006) were
all significantly different (Figure 16). Figure 16 shows the MS2 log10 reduction at
each sampling time. The first flush samples showed an average log10 reduction
of MS2 of 6.7, the 15 minute samples had an average of 3.9, the 45 minute
samples had an average of 4.8, and the 120 minute samples had an average of
5.7. There was no significant effect of ceramic (p=0.517). No correlation was
2

found between MS2 log10 reduction and flow rate (R =0.005, p=0.421) or with
2

bromine residual (R =0.025, p=0.069). There was also no significant effect of
cartridge (p=0.479). Only the pH 9 sewage results were used for this analysis,
since that was the only group of the same type of samples that used all six
cartridges.

!

60!

Figure 16. MS2 log10 reduction at each sampling time. There was a significant
effect of sample on the log10 reduction of MS2 (p<0.001); the first flush and the
15 minute samples (p<0.001), the first flush and 45 minute samples (p=0.001),
the 15 minute and 120 minute samples (p=0.001), and the 45 minute and 120
minute samples (p=0.006) were all significantly different. N=18 for well water
and 18 for sewage (N=36 total). The º indicates an outlier, and the * indicates
an extreme outlier.
3.3 MURINE NOROVIRUS
3.3.1 INFLUENT CHARACTERIZATION
Table 13 shows the average influent concentrations for the experimental
challenges using murine norovirus. The murine norovirus stock used gave an
average influent concentration of 4.0, 4.3, 2.8, and 2.3 log10 PFU/mL for the
experiments at pH 7.5 for well water and sewage and pH 9 for well water and
!

61!

sewage, respectively. The pH 7.5 experiments had higher influent
concentrations of murine norovirus. All of the murine norovirus stock was
prepared using the same procedure and materials; the differences seen in the
influent concentrations cannot be explained at this time. Murine norovirus stocks
were purified on 9 different dates (2-15-12, 2-29-12, 3-19-12, 3-21-12, 4-12-12,
5-10-12, 10-23-12, 11-5-12, and 12-14-12). The stocks were stored at -80ºC
until use. Due to issues with a working plaque assay, only the stock that was
purified on 5-10-12 was assayed to determine the concentration; the pure stock
6

had a concentration of 10 PFU/mL (assay run 10-15-12). The purified stocks
produced 2-15-12 to 5-10-12 were used for the pH 7.5 experiments, and the
purified stocks produced 5-10-12 to 12-14-12 was used for the pH 9 experiments.
The purified stocks were a maximum of five months old when used, but most of
the stocks were less than a month old when used. The differences seen in the
influent concentrations of murine norovirus are probably due to slight differences
in the starting concentrations of the unpurified norovirus stock. Even though the
same conditions and protocols were used to grow up each batch of virus, slight
differences in initial cell and virus concentrations could result in large differences
in the amount of new viruses produced in each batch.

!

62!

Table 13. Influent concentrations for murine norovirus for the pH, sewage and
well water variables. Shown are the mean values (+/- standard deviation) for
virus influent concentrations for each combination of pH and water treatment
variables.
pH 7.5

pH 9

Well Water
Murine
Norovirus
Influent
Concentration
(log10)

Sewage

Well Water

Sewage

4.0 (n=9)

4.3 (n=9)

2.8 (n=3)

2.3 (n=3)

Figure 17 shows the average murine norovirus influent concentrations for
the pH 7.5 samples. There was an average murine norovirus influent
concentration of 4.0 log10 for the well water samples and 4.3 log10 for the
sewage samples (Figure 17).

!

63!

Figure 17. Average murine norovirus influent concentrations (in log10) for pH 7.5
samples. N=9 for well water and sewage (N=18 total). The º indicates an
outlier.
Figure 18 shows the average murine norovirus influent concentrations for
the pH 9 samples. There was an average murine norovirus influent
concentration of 2.8 log10 for the well water samples and 2.3 log10 for the
sewage samples (Figure 18).

!

64!

Figure 18. Average murine norovirus influent concentrations (in log10) for pH 9
samples. N=3 for well water and sewage (N=6 total).
For the murine norovirus samples, there was an average of 3.8 log10 in
the untreated influent samples (Figure 19). Figure 19 shows the average murine
norovirus influent concentrations for all samples. There was found to be no
significant difference in the murine norovirus influent concentrations between the
water treatments (p=0.768), but there was a significant difference between pH
treatments (p<0.001). The pH 7.5 samples had a mean of 4.1 log10, while the
pH 9 samples had a mean of 2.6 log10. All of the norovirus samples have not
been analyzed due to time constraints.
!

65!

Figure 19. Average murine norovirus influent concentrations (in log10) for all
samples. N=12 for well water and sewage (N=24 total).
3.3.2 EFFLUENT CHARACTERIZATION
3.3.2.1 DEVICE PERFORMANCE
For the murine norovirus samples, all of the treated samples were reduced
to the detection limit (2 PFU/mL), so no sample variation could be analyzed
(Figure 20). Log10 reductions are presented as “greater than or equal to” the
values given.

!

66!

Figure 20. The log10 concentrations of murine norovirus in untreated influent
(time=0, n= 24) and treated samples. The treated samples were all reduced to
the detection limit (2 PFU/mL).
For each of the influent and effluent samples, murine norovirus
concentration was determined by plaque assay. The log10 reduction values were
calculated by taking the log10 concentration of the influent minus the log10
concentration of the effluent.
Overall, the average influent concentration of murine norovirus was 3.8
log10, the average effluent concentration was <0.3 log10, and the average log10
reduction was >3.8 log10 (Table 14). Table 14 shows the mean norovirus
!

67!

influent and effluent concentrations, as well as the log10 reductions for each
combination of pH and water treatment variables. Although a true statistical
comparison is not possible due to the effluent samples being reduced to the
detection limit, it is important to note that the well water and sewage samples did
perform similarly in regards to log10 reduction of murine norovirus. All of the
samples (both well water and sewage) were reduced to the detection limit, so it is
possible to presume that the device is capable of the same performance even
with contaminated water.
Table 14. Murine norovirus influent and effluent concentrations and log10
reductions. The mean values (+/- standard deviation) are given for each
measurement for each combination of pH and water treatment variables.
pH 7.5

pH 9

Well Water

Sewage

Well Water

Sewage

N=9

N=9

N=3

N=3

Cartridge

1, 2, 3

1, 2, 3

1, 2, 3

4, 5, 6

Murine
Norovirus
Influent
Concentration
(log10)

4.0

4.3

2.8

2.3

Murine
Norovirus
Effluent
Concentration
(log10)

<0.3

<0.3

<0.3

<0.3

Murine
Norovirus Log
Reduction
(log10)

>3.7

>4.0

>2.5

>2.0

!

68!

For the pH 7.5 samples, the average influent concentration of murine
norovirus was 4.1 log10, the average effluent concentration was <0.3 log10, and
the average log10 reduction was >3.9 log10 (Table 15 and 16). Table 15 shows
the average murine norovirus concentrations and device characteristics (average
flow rate and bromine residual) in the effluent post treatment for well water
samples at pH 7.5. Table 16 shows the average murine norovirus concentrations
and device characteristics (average flow rate and bromine residual) in the
effluent post treatment for sewage addition samples at pH 7.5.

!

69!

Table 15. Average murine norovirus concentrations and device characteristics in
the effluent post treatment for well water samples at pH 7.5.
Experi
Cartr Time
# of
Average
Average Average
ment #
idge (min) post Samples Murine
Flow
Bromine
addition of
Concentration Rate
Residual
influent to
(mL/min (mg/L)
(log10)
upper
)
reservoir
6, 13,
23

1

0

8

3

0

0

4.4

2;4;10

0.41

3

0.3

41

0.19

3

0.3

39

0.49

120

3

0.3

15

0.83

0

4

3.9

0

0

3;3;5;24

4

0.3

53

0.11

15;15;17

3

0.3

39

0.34

45

4

0.3

27

0.62

120

4

0.3

14

0.87

0

2

3.8

0

0

4;7

2

0.3

84

0.11

15

2

0.3

48

0.28

45

2

0.3

40

0.62

120

3

46

45

7, 24

0.3

15;15;17

2, 8, 12, 2
25

3

2

0.3

18

1.18

39

0.50

Total
n=44

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
8
Time of 0 (zero) is the untreated influent sample.
!

70!

Table 16. Average murine norovirus concentrations and device characteristics in
the effluent post treatment for sewage addition samples at pH 7.5.
Experi
Cartr Time
# of
Average
Average Average
ment #
idge (min) post Samples Murine
Flow
Bromine
addition of
Concentration Rate
Residual
influent to
(mL/min (mg/L)
(log10)
upper
)
reservoir

3

0.3

43

0.19

3

0.3

30

0.26

3

0.3

28

0.64

3

0.3

12

0.56

0

3

4.3

0

0

3

0.3

36

0.11

15

3

0.3

31

0.30

45;45;46

3

0.3

28

0.68

120

3

0.3

15

0.75

0

3

4.3

0

0

3;3;9

3

0.3

64

0.64

15

3

0.3

43

0.43

45

3

0.3

31

0.99

120

3

0

2;3;5

3, 9, 16

0

120
2

4.2

45
4, 10,
14

3

15

1

9

3;11;13

5, 11,
15

3

0.3

18

0.94

32

0.54

0

Total
n=45

For the pH 9 samples, the average influent concentration of murine
norovirus was 2.6 log10, the average effluent concentration was <0.3 log10, and
the average log10 reduction was 2.2 log10 (Table 17 and 18). Table 17 shows
the average murine norovirus concentrations and device characteristics (average
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
9
Time of 0 (zero) is the untreated influent sample.
!

71!

flow rate and bromine residual) in the effluent post treatment for well water
samples at pH 9. Table 18 shows the average murine norovirus concentrations
and device characteristics (average flow rate and bromine residual) in the
effluent post treatment for sewage addition samples at pH 9.
Table 17. Average murine norovirus concentrations and device characteristics in
the effluent post treatment for well water samples at pH 9.
Experi
Cartr Time
# of
Average
Average Average
ment #
idge (min) post Samples Murine
Flow
Bromine
addition of
Concentration Rate
Residual
influent to
(mL/min (mg/L)
(log10)
upper
)
reservoir

1

0.3

55

0.02

1

0.3

16

0.05

1

0.3

15

0.05

1

0.3

9

0.07

0

1

2.8

0

0

1

0.3

56

0.05

45

1

0.3

11

0.07

120

1

0.3

7

0.09

0

1

3.3

0

0

20

1

0.3

100

0.07

45

1

0.3

16

0.09

120

6

0

20

31

0

120
5

2.3

45
30

1

15

4

10

10

29

1

0.3

7

0.09

30

0.06

0

Total
n=14

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
10
Time of 0 (zero) is the untreated influent sample.
!

72!

Table 18. Average murine norovirus concentrations and device characteristics in
the effluent post treatment for sewage addition samples at pH 9.
Experi
Cartr Time
# of
Average
Average Average
ment #
idge (min) post Samples Murine
Flow
Bromine
addition of
Concentration Rate
Residual
influent to
(mL/min (mg/L)
(log10)
upper
)
reservoir

1

0.3

46

0.02

1

0.3

28

0.11

1

0.3

20

2.92

1

0.3

10

0.18

0

1

2.2

0

0

1

0.3

75

0.02

17

1

0.3

28

0.07

45

1

0.3

20

0.11

120

1

0.3

12

0.09

0

1

2.8

0

0

25

1

0.3

70

0.02

45

1

0.3

18

0.09

120

6

0

10

34

0

120
5

2.1

45
33

1

24

4

11

6

32

1

0.3

7

0.09

30

0.34

0

Total n=
14
3.4 CONCLUSIONS

Overall, the device was shown to be capable of removing an average of
5.4 log10 PFU/mL of MS2. Due to the low influent concentrations for the murine
norovirus samples, the effectiveness of the device against that virus cannot be
fully determined, but the device did remove an average of 3.5 log10 PFU/mL of
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
11
Time of 0 (zero) is the untreated influent sample.
!

73!

murine norovirus. The addition of raw sewage did not have a significant effect on
the performance of the device overall. However, for the pH 7.5 samples, the
sewage addition samples did have significantly higher log10 reduction of MS2
compared to the well water samples. The increased pH did have a negative
effect on the performance of the device but was only shown for those samples
with sewage.

!

74!

CHAPTER 4
DISCUSSION
The HaloSource Waterbird is a novel type of point of use device. Rather
than using a liquid bromine solution, the bromine is attached to polystyrene
beads (HaloPure media). As water rushes over the beads, bromine comes in
contact with the biological particles and inactivates them. The unique delivery
system for bromine makes the device difficult to evaluate as a disinfectant
because the active dose of bromine cannot be measured, only the residual that is
left after treatment. The exact contact time can also not be evaluated, and the
mechanism is still not understood. The device utilizes both ceramic filtration and
halogen disinfection (bromine) to remove and/or inactivate bacteria and viruses.
The HaloPure media has been tested and approved by the EPA to be capable of
six log10 reduction of bacteria and four log10 reduction of viruses (poliovirus and
rotavirus) (Dr. Jeff Williams, Personal Communication). This study aimed to test
the Waterbird device as a whole and particularly bromine media disinfection,
similar to how a buyer would use it to treat water in their home. The main
objective for the study was to determine the effectiveness of the Waterbird device
for inactivating and or removing MS2 bacteriophage and murine norovirus. Subobjectives were to determine if pH or added organic materials (in the form of raw
sewage) had an impact on the device’s performance.
Past studies suggest that MS2 is easier to inactivate with chlorine than
murine norovirus; higher log10 reductions can be achieved for MS2 with the
same contact time and chlorine concentration (Lim et al., 2010). The study by
!

75!

Lim et al. (2010) used demand-free water at pH 7.2; experiments were run at
5ºC and 20ºC. The experiments were set up in batch reactors, with 30 mL
volume per experiment. Four log10 reduction of MS2 was possible after four
minutes of contact time with 0.17 mg/L chlorine; only two minutes of contact time
with 0.19 mg/L chlorine resulted in three log10 reduction of murine norovirus.
While the present study showed a lower titer of the murine norovirus in the
influent than anticipated, it was possible to show at least two to four log10
reductions, which seems to be somewhat comparable result. !
The same disinfection delivery system (HaloPure bromine media) has
been used previously in three studies (Enger et al., Draft Manuscript; Coulliette et
al., 2010; McLennan et al., 2009). However, all of these studies simply tested
the effectiveness of the HaloPure media (without the carbon pre-filter), while the
present study also included the ceramic pre-filter during the device testing. For
MS2 bacteriophage, Coulliette et al. (2010) found that the device was capable of
an average log10 reduction of 5.0, which was not different overall from the
present study (5.4). Their system had an average flow rate of 82 mL/min, and an
average bromine residual (for samples taken at 60 and 90 minutes) of 1.2 mg/L
(Coulliette et al., 2010). The average flow rate for the present study was half this
(34 mL/min), and the average bromine residual (for the samples taken at 45 and
120 minutes) was lower (0.60 mg/L). The slightly higher log10 reduction seen in
the present study could be due to the presence of the ceramic pre-filter, which

!

76!

might be able to remove some viruses. The ceramic pre-filter also lowered the
flow rate. The second study by Enger et al. (Draft Manuscript) showed an
average of 4.7 log10 reduction of MS2; the average flow rate was 160 mL/min,
and the average bromine residual was 1.0 mg/L. As with the Coulliette et al.
(2010) study, the absence of the ceramic pre-filter could explain the lower log10
reduction values for MS2 and the higher flow rate (Enger et al., Draft
Manuscript). The final study by McLennan et al. (2009) used natural coliphage
from raw sewage (10%) instead of adding pure stock of MS2. They observed an
average log10 reduction of 1.8, but their starting influent concentration of
coliphage was much lower than in the present study. The chlorine system
produced residuals ranging from zero to 0.60 mg/L and bromine system had
residuals of 0.68 to 1.8 mg/L. All of their coliphage samples were reduced to the
detection limit for the bromine but only one log10 reduction for the chlorine, so the
bromine device could be capable of more than a 1.8 log10 reduction of coliphage
and was more effective than the chlorine.
The study by Enger et al. (Draft Manuscript) used human adenovirus
(Adenovirus Type 2), where by a log10 reduction of 4.9 was observed. As
mentioned above, this study used a different type of cartridge (though the
HaloPure media inside was about the same) and no ceramic pre-filter.
Adenoviruses are double-stranded DNA viruses, so they may be more resistant
to bromine. The Waterbird device has been tested with the EPA protocol, and

!

77!

has been shown to be capable of reducing four log10 of both poliovirus and
rotavirus (Dr. Jeff Williams, Personal Communication).
Two other studies tested liquid bromine against human viruses (Floyd et
al., 1976; Sharp et al., 1975). Floyd et al. (1976) found a 3.5 log10 reduction of
poliovirus, and Sharp et al. (1975) found a four log10 reduction of reovirus. Floyd
et al. (1976) used 0.13 mg/L bromine, and it took 12 seconds of contact time to
get 3.5 log10 reduction of poliovirus. The experiment was done at 10ºC and pH
7. Sharp et al. (1975) used 0.07 mg/L bromine, and it took 1.5 minutes of contact
time to get a four log10 reduction of reovirus. The experiment was done at 2ºC
and pH 7. Reoviruses are double-stranded RNA viruses, and poliovirus is a
positive-strand RNA virus like norovirus. These studies demonstrate that
bromine residual similar to what is delivered with the HaloPure media can
inactivate both DNA and RNA viruses similar to what was found for the murine
norovirus. It is important to note that these two studies used a low temperature
which also affects disinfection efficacy (lowering it) and buffered water (no raw
sewage added) to do their testing (which should increase the efficacy) compared
to the present study of the WaterBird device.
In the present study, for the MS2 samples, there was a significant effect of
pH, as would be expected (but only the sewage experiments could be
compared). Previous studies suggest that halogen disinfection works better at a
lower pH. One study using a chlorine solution found that it took more chlorine
and longer contact time to get the same level of poliovirus inactivation at high pH
!

78!

(10.0) than at low pH (6.0) (Kott et al., 1975). Amiri et al. (2010) found that it was
more difficult to inactivate E. coli at pH 8.0 than at pH 6.0 or 6.9 when using
organic N-chloramines as a disinfectant. Another study with chlorine dioxide
(and six different viruses) found that the disinfectant loses its effectiveness at pH
9.0 (Junli et al., 1997). In accordance with those results, this study found that the
device works better at pH 7.5 than at pH 9. At pH 7.5, bromine is mostly in the
-

form of hypobromous acid (HOBr); at pH 9 it is mostly OBr (Song et al., 1996).
At a low pH, the amino acids in the virus capsid are likely to be neutral, while at
high pH the amino acids are likely to be negatively charged. If both the bromine
and virus capsid are neutral the two won’t repel one another, but if they are both
negative they will repel one another and disinfection will be less successful.
For the MS2 samples overall, there was no significant effect of sewage
addition. Additionally, neither bromine residual nor flow rate differed between the
well water and sewage addition samples. It might be expected that the increased
organic materials in the raw sewage would exert an additional bromine demand
on the device, resulting in a higher bromine residual, and that the increased
turbidity in the raw sewage treatment could result in the ceramic pre-filter
clogging more rapidly and lowering the flow rate. However, neither observation
occurred often enough to be statistically significant. Enger et al. (Draft
Manuscript) also found that the addition of raw sewage had no effect on the
performance of the HaloPure bromine media. It is unusual that the addition of
raw sewage did not negatively affect the performance of the device, as that is the
common result with many other types of disinfection systems. It appears that the
!

79!

HaloPure media cartridges are less susceptible to variation in the organic content
of the water to be treated. In addition to increasing the disinfectant demand, raw
sewage can physically block contact and light-based disinfectants (if the virus
particles are inside the organic material) (McGuigan et al., 2012). For the
Waterbird device, the ceramic pre-filter is instrumental in removing the solid
organic particulates present in the raw sewage, thus decreasing the bromine
demand and allowing the bromine contact disinfectant to work better.
When considering only the samples taken at the 15 minute time point, the
trends for the effect of sewage addition and pH manipulation are the same as
when all the data are considered; there was no significant effect of sewage
addition or pH, but the pH 9 samples did have significantly lower log10 reduction
of MS2 than the pH 7.5 samples (when only the sewage samples were
analyzed). The average log10 reduction of MS2 was 4.7, 5.7, 5.3, and 5.9 for the
pH 7.5 well water and sewage samples, and the pH 9 well water and sewage
samples, respectively. The 15 minute samples had the lowest log10 reduction of
MS2 compared to almost all of the samples overall. The average flow rate for
the 15 minute samples was 38 mL/min, compared to an overall average of 34
mL/min. The average bromine residual was 0.27 mg/L, compared to an overall
average of 0.41 mg/L. The low bromine residual may be responsible for the
lower log10 reduction of MS2 seen in the 15 minute samples.
The rechargeability of the HaloPure beads may be responsible for the
differences seen in MS2 log10 reduction between sampling times. In the first
!

80!

flush samples, the bromine on the outside of the polystyrene beads is released to
disinfect the water that flows through the cartridge early on; there is more
bromine available, so there is more reduction of viruses. In the short amount of
time between the first flush sample and the 15 minute sample, there may not be
enough time for the outside of the polystyrene beads to become covered with
bromine atoms again (atoms from the inside move to the outside of the bead);
not enough bromine is available to treat the water, so there is less reduction of
viruses. By the 45 and 120 minute samples, there has been ample time for the
displaced bromine atoms from the outside of the beads to be replaced by new
bromine atoms from the inside, so there is once again enough bromine to treat
the water flowing through the cartridge. However, it is impossible to measure the
actual disinfecting dose of bromine released from the cartridges, so it may not be
possible to verify this theory.
There was also a significant difference in bromine residual between
sampling times. The bromine residual typically would start very low and increase
over time. This result might explain the variability seen in the log10 reductions of
MS2 between sampling times. However, no statistical correlation was found
between MS2 log10 reduction and bromine residual. The bromine residual was
lowest for some of the samples when the MS2 log10 reduction was highest (the
first flush samples). Flow rate also differed significantly between sampling times,
changing the contact times (less contact at the beginning when flow rate was
faster). For almost every device, the flow rate decreased with increased

!

81!

sampling time. This is due to the decreasing volume in the top reservoir of the
device; since the device is gravity fed, as the top volume decreases, there is less
pressure to force water through the device, so the flow rate decreases.
One area of concern with the Waterbird device is the short lifespan that
the HaloPure media cartridges seem to have, compared to the manufacturer’s
stated lifespan. The first three HaloPure bromine cartridges used for the study
only treated approximately 100 L of water before losing effectiveness (reducing
less than one log10 of MS2) when challenged with sewage contaminated well
water. HaloSource claims that the cartridges have a lifespan of 1,500 L. The
cartridges have a built-in end-of-life shut-off device that doesn’t allow water to
flow through the cartridge after approximately 1,500 L, but the three cartridges
did not have the shut-off device engaged.
Bromine has typically been avoided for use in drinking water treatment
systems because of the potential to form more hazardous disinfection byproducts such as bromate, bromoform, and dibromoacetic acid (Guo and Chen,
2009; Song et al., 1996). Disinfection by-products have been shown to cause
cancer and reproductive anomalies in lab animals (Krasner, 2009). Brominated
organic disinfection by-products can actually be more dangerous than the
chlorinated forms, and hypobromous acid has been shown to react better and
faster with organic matter than hypochlorous acid (Singer, 1999; Guo and Chen,
2009). The HaloPure delivery system for bromine has been tested for the
potential to form disinfection by-products (bromate and bromide), and has
received the National Sanitation Foundation (NSF) certification (standard 042)
!

82!

(National Sanitation Foundation, 2012). The NSF testing determined that up to
94 g of HaloPure bromine media could be placed in a cartridge and still release
acceptable levels of bromate and bromide; a normal cartridge for the Waterbird
device holds an average of 15 g of media (National Sanitation Foundation, 2012).
There is still more work that could be done with the Waterbird device to
better understand how well it works. The virus challenges should be repeated,
but rather than taking samples during the treatment process, all of the water
should be allowed to pass through the device before taking samples. This would
more closely represent how consumers use the device in their own homes. It
would also be interesting to do an observation study on how the device actually
is used in the field. No household water treatment device is going to be used
with 100% compliance by consumers, so understanding how consumers actually
use the water is always helpful.
Overall, I do feel that the Waterbird device is a worthy drinking water
disinfection device. It has been shown to be capable of a 5.4 log10 reduction of
MS2, and at least a 3.5 log10 reduction of murine norovirus. I would recommend
that the device is used as the manufacturer suggests, adding water at night and
letting it treat for at least 8 hours before consumption. That would provide the
bromine residual in the treated water additional time to inactivate some of the
remaining microorganisms that may still be present in the treated water.
The device is a bit expensive for use in third world countries, but it seems
to be a good option for the intended market (middle-class India). The device
reportedly has a good lifespan (1,500 L), but after that point a new HaloPure
!

83!

cartridge must be purchased. Using the purchase price for the entire device, this
technology costs $0.02/L of water. The replacement cost for new cartridges
would be less than the original purchase price, so the cost would go down over
time. However, the Waterbird device still ends up being a more expensive way
to disinfect drinking water compared to traditional chlorination (Sobsey, 2002).

!

84!

APPENDIX

!

85!

APPENDIX
Table A-1. Results of all challenge experiments.

Experi
ment

2

Date
22Mar
22Mar
22Mar
22Mar
22Mar

3

10Apr

3

10Apr

3

10Apr

3

10Apr

2
2
2
2

!

Mat Cartri
rix dge

Cera
mic

Tempe
rature

pH

Sam
ple

Total
Chlori
ne
(mg/L)

MS2
Concentr
ation
(PFU/mL)

Murine
Norovirus
Concentrat
ion
(PFU/mL)

0.00

Flow
Rate
(mL/min)

Bromi
ne
Resid
ual
(mg/L)
0.00

2.80E+07

2.00E+03

ww

2

3

10.5

7.40 inf

ww

2

3

10.5

7.40 t0

45

0.05

0.11

1.00E+00

2.00E+00

ww

2

3

10.5

7.40 t1

25

0.10

0.23

7.00E+04

2.00E+00

ww

2

3

10.5

7.40 t2

22

0.25

0.56

1.80E+03

2.00E+00

ww
se
wa
ge
se
wa
ge
se
wa
ge
se
wa
ge

2

3

10.5

7.40 t3

13

0.30

0.68

1.00E+00

2.00E+00

3

3

14.6

7.44 inf

0.00

0.00

8.00E+04

1.00E+04

3

3

14.6

7.44 t0

35

0.75

1.69

1.00E+00

2.00E+00

3

3

14.6

7.44 t1

40

0.12

0.27

1.00E+00

2.00E+00

3

3

14.6

7.44 t2

30

0.27

0.61

1.00E+00

2.00E+00

86!

Table A-1 (cont’d).
se
10- wa
3
Apr ge
se
10- wa
4
Apr ge
se
10- wa
4
Apr ge
se
10- wa
4
Apr ge
se
10- wa
4
Apr ge
se
10- wa
4
Apr ge
se
10- wa
5
Apr ge
se
10- wa
5
Apr ge
se
10- wa
5
Apr ge

!

3

3

14.6

7.44 t3

22

0.45

1.01

1.00E+00

2.00E+00

2

2

14.5

7.65 inf

0.00

0.00

3.30E+05

2.00E+04

2

2

14.5

7.65 t0

27

0.05

0.11

1.00E+00

2.00E+00

2

2

14.5

7.65 t1

20

0.05

0.11

4.00E+01

2.00E+00

2

2

14.5

7.65 t2

18

0.20

0.45

1.00E+00

2.00E+00

2

2

14.5

7.65 t3

10

0.30

0.68

1.00E+00

2.00E+00

1

1

14.7

7.51 inf

0.00

0.00

4.90E+05

2.00E+04

1

1

14.7

7.51 t0

35

0.15

0.34

1.00E+00

2.00E+00

1

1

14.7

7.51 t1

15

0.05

0.11

1.00E+00

2.00E+00

87!

Table A-1 (cont’d).
se
10- wa
5
Apr ge
se
10- wa
5
Apr ge
14Apr ww
6
14Apr ww
6
14Apr ww
6
14Apr ww
6
14Apr ww
6
14Apr ww
7
147
Apr ww
147
Apr ww
147
Apr ww
147
Apr ww
148
Apr ww

!

1

1

14.7

7.51 t2

15

0.20

0.45

1.00E+00

2.00E+00

1

1

14.7

7.51 t3

9

0.30

0.68

1.00E+00

2.00E+00

1

2

11.3

7.54 inf

0.00

0.00

2.40E+05

1.10E+04

1

2

11.3

7.54 t0

72

0.45

1.01

1.00E+00

2.00E+00

1

2

11.3

7.54 t1

50

0.10

0.23

4.90E+02

2.00E+00

1

2

11.3

7.54 t2

68

0.25

0.56

9.00E+01

2.00E+00

1

2

11.3

7.54 t3

20

0.50

1.13

1.40E+01

2.00E+00

3

1

11.4

7.48 inf

0.00

0.00

3.10E+05

2.00E+04

3

1

11.4

7.48 t0

115

0.05

0.11

1.00E+00

2.00E+00

3

1

11.4

7.48 t1

42

0.10

0.23

1.10E+03

2.00E+00

3

1

11.4

7.48 t2

40

0.25

0.56

5.70E+01

2.00E+00

3

1

11.4

7.48 t3

15

0.45

1.01

8.00E+00

2.00E+00

2

3

11.6

7.45 inf

0.00

0.00

5.50E+05

1.00E+05

88!

Table A-1 (cont’d).
148
Apr ww
14Apr ww
8
14Apr ww
8
14Apr ww
8
se
17- wa
9
Apr ge
se
17- wa
9
Apr ge
se
17- wa
9
Apr ge
se
17- wa
9
Apr ge
se
17- wa
9
Apr ge
se
17- wa
10
Apr ge

!

2

3

11.6

7.45 t0

60

0.05

0.11

1.00E+00

2.00E+00

2

3

11.6

7.45 t1

46

0.15

0.34

3.30E+02

2.00E+00

2

3

11.6

7.45 t2

38

0.30

0.68

4.50E+02

2.00E+00

2

3

11.6

7.45 t3

21

0.50

1.13

7.00E+00

2.00E+00

3

1

16.6

7.6 inf

0.00

0.00

2.10E+07

1.10E+04

3

1

16.6

7.6 t0

75

0.05

0.11

1.80E+02

2.00E+00

3

1

16.6

7.6 t1

38

0.10

0.23

6.00E+02

2.00E+00

3

1

16.6

7.6 t2

18

0.15

0.34

4.20E+01

2.00E+00

3

1

16.6

7.6 t3

7

0.20

0.45

1.00E+00

2.00E+00

2

3

16.1

7.58 inf

0.00

0.00

3.10E+07

2.80E+04

89!

Table A-1 (cont’d).
se
17- wa
10
Apr ge
se
17- wa
10
Apr ge
se
17- wa
10
Apr ge
se
17- wa
10
Apr ge
se
17- wa
11
Apr ge
se
17- wa
11
Apr ge
se
17- wa
11
Apr ge
se
17- wa
11
Apr ge
se
17- wa
11
Apr ge

!

2

3

16.1

7.58 t0

15

0.05

0.11

1.00E+00

2.00E+00

2

3

16.1

7.58 t1

16

0.05

0.11

2.50E+02

2.00E+00

2

3

16.1

7.58 t2

22

0.20

0.45

1.00E+00

2.00E+00

2

3

16.1

7.58 t3

21

0.20

0.45

1.00E+00

2.00E+00

1

2

16.4

7.56 inf

0.00

0.00

1.50E+07

1.80E+04

1

2

16.4

7.56 t0

30

0.05

0.11

1.00E+00

2.00E+00

1

2

16.4

7.56 t1

24

0.05

0.11

1.00E+00

2.00E+00

1

2

16.4

7.56 t2

22

0.15

0.34

3.00E+00

2.00E+00

1

2

16.4

7.56 t3

10

0.15

0.34

1.00E+00

2.00E+00

90!

Table A-1 (cont’d).
1912
Apr ww
19Apr ww
12
19Apr ww
12
19Apr ww
12
19Apr ww
13
19Apr ww
13
19Apr ww
13
19Apr ww
13
1913
Apr ww
se
24- wa
14
Apr ge
se
24- wa
14
Apr ge
se
24- wa
14
Apr ge

!

2

1

11.6

7.57 inf

0.00

0.00

3.40E+06

4.00E+03

2

1

11.6

7.57 t0

68

0.05

0.11

1.00E+01

2.00E+00

2

1

11.6

7.57 t2

12

0.05

0.11

1.70E+02

2.00E+00

2

1

11.6

7.57 t3

7

0.10

0.23

1.00E+00

2.00E+00

1

2

13

7.53 inf

0.00

0.00

2.90E+07

2.60E+04

1

2

13

7.53 t0

22

0.05

0.11

1.00E+00

2.00E+00

1

2

13

7.53 t1

22

0.05

0.11

1.00E+00

2.00E+00

1

2

13

7.53 t2

6

0.10

0.23

6.00E+00

2.00E+00

1

2

13

7.53 t3

9

0.10

0.23

1.00E+00

2.00E+00

2

1

16.7

7.51 inf

0.00

0.00

2.20E+07

2.00E+04

2

1

16.7

7.51 t0

65

0.05

0.11

1.00E+00

2.00E+00

2

1

16.7

7.51 t1

57

0.30

0.68

1.06E+03

2.00E+00

91!

Table A-1 (cont’d).
se
24- wa
14
Apr ge
se
24- wa
14
Apr ge
se
24- wa
15
Apr ge
se
24- wa
15
Apr ge
se
24- wa
15
Apr ge
se
24- wa
15
Apr ge
se
24- wa
15
Apr ge
se
24- wa
16
Apr ge
se
24- wa
16
Apr ge

!

2

1

16.7

7.51 t2

44

0.50

1.13

7.00E+01

2.00E+00

2

1

16.7

7.51 t3

15

0.50

1.13

1.10E+02

2.00E+00

1

2

16.9

7.54 inf

0.00

0.00

2.20E+07

1.00E+04

1

2

16.9

7.54 t0

64

0.05

0.11

1.00E+00

2.00E+00

1

2

16.9

7.54 t1

50

0.25

0.56

3.00E+03

2.00E+00

1

2

16.9

7.54 t2

46

0.50

1.13

3.00E+03

2.00E+00

1

2

16.9

7.54 t3

18

0.30

0.68

2.90E+02

2.00E+00

3

3

18

7.51 inf

0.00

0.00

2.90E+07

9.00E+04

3

3

18

7.51 t0

0.05

0.11

1.00E+00

2.00E+00

82

92!

Table A-1 (cont’d).
se
24- wa
16
Apr ge
se
24- wa
16
Apr ge
se
24- wa
16
Apr ge
se
27- wa
17
Apr ge
se
27- wa
17
Apr ge
se
27- wa
17
Apr ge
se
27- wa
17
Apr ge
se
27- wa
17
Apr ge
se
27- wa
18
Apr ge

!

3

3

18

7.51 t1

52

0.35

0.79

2.10E+03

2.00E+00

3

3

18

7.51 t2

46

0.90

2.03

4.40E+02

2.00E+00

3

3

18

7.51 t3

24

0.60

1.35

2.30E+02

2.00E+00

2

2

9.7

8.97 inf

0.00

0.00

2.40E+07

2

2

9.7

8.97 t0

64

0.05

0.11

1.00E+00

2

2

9.7

8.97 t1

56

0.10

0.23

1.70E+04

2

2

9.7

8.97 t2

46

0.40

0.90

4.00E+04

2

2

9.7

8.97 t3

15

0.50

1.13

1.00E+03

1

1

11

9.02 inf

0.00

0.00

2.00E+07

93!

Table A-1 (cont’d).
se
27- wa
18
Apr ge
se
27- wa
18
Apr ge
se
27- wa
18
Apr ge
se
27- wa
18
Apr ge
se
27- wa
19
Apr ge
se
27- wa
19
Apr ge
se
27- wa
19
Apr ge
se
27- wa
19
Apr ge
se
27- wa
19
Apr ge

!

1

1

11

9.02 t0

24

0.05

0.11

1.00E+00

1

1

11

9.02 t1

59

0.25

0.56

1.40E+05

1

1

11

9.02 t2

40

0.50

1.13

1.00E+05

1

1

11

9.02 t3

19

0.45

1.01

1.00E+03

3

3

12.3

8.96 inf

0.00

0.00

2.00E+08

3

3

12.3

8.96 t0

80

0.05

0.11

1.00E+00

3

3

12.3

8.96 t1

48

0.35

0.79

4.00E+04

3

3

12.3

8.96 t2

38

0.55

1.24

8.00E+03

3

3

12.3

8.96 t3

20

0.50

1.13

2.00E+03

94!

Table A-1 (cont’d).
1123 May ww
1123 May ww
1123 May ww
1123 May ww
1123 May ww
1124 May ww
1124 May ww
1124 May ww
1124 May ww
1124 May ww
1125 May ww
1125 May ww
1125 May ww
1125 May ww

!

1

2

9.1

7.48 inf

0.00

0.00

9.60E+07

1.00E+04

1

2

9.1

7.48 t0

45

0.05

0.11

1.00E+00

2.00E+00

1

2

9.1

7.48 t1

50

0.10

0.23

3.00E+06

2.00E+00

1

2

9.1

7.48 t2

42

0.30

0.68

1.09E+06

2.00E+00

1

2

9.1

7.48 t3

17

0.50

1.13

6.90E+03

2.00E+00

3

1

10.1

7.53 inf

0.00

0.00

1.50E+08

2.00E+03

3

1

10.1

7.53 t0

52

0.05

0.11

1.00E+00

2.00E+00

3

1

10.1

7.53 t1

54

0.15

0.34

2.20E+06

2.00E+00

3

1

10.1

7.53 t2

40

0.30

0.68

1.08E+06

2.00E+00

3

1

10.1

7.53 t3

20

0.60

1.35

2.50E+04

2.00E+00

2

3

14

7.58 inf

0.00

0.00

1.10E+08

1.00E+04

2

3

14

7.58 t0

38

0.05

0.11

1.00E+00

2.00E+00

2

3

14

7.58 t1

45

0.20

0.45

7.00E+05

2.00E+00

2

3

14

7.58 t2

37

0.50

1.13

1.00E+06

2.00E+00

95!

Table A-1 (cont’d).
1125 May ww
27Oct ww
29
27Oct ww
29
27Oct ww
29
27Oct ww
29
27Oct ww
29
27Oct ww
30
27Oct ww
30
2730
Oct ww
2730
Oct ww
2731
Oct ww
2731
Oct ww
2731
Oct ww
2731
Oct ww

!

2

3

14

7.58 t3

16

0.65

1.46

5.30E+04

2.00E+00

4

3

9

9.03 inf

0.00

0.00

4.00E+07

2.00E+02

4

3

9

9.03 t0

55

0.01

0.02

3.00E+01

2.00E+00

4

3

9

9.03 t1

16

0.02

0.05

1.10E+03

2.00E+00

4

3

9

9.03 t2

15

0.02

0.05

1.00E+00

2.00E+00

4

3

9

9.03 t3

9

0.03

0.07

7.00E+00

2.00E+00

5

1

9.1

9.02 inf

0.00

0.00

3.00E+07

7.00E+02

5

1

9.1

9.02 t0

56

0.02

0.05

4.00E+00

2.00E+00

5

1

9.1

9.02 t2

11

0.03

0.07

7.00E+02

2.00E+00

5

1

9.1

9.02 t3

7

0.04

0.09

1.00E+00

2.00E+00

6

2

8.6

9.03 inf

0.00

0.00

1.00E+07

2.00E+03

6

2

8.6

9.03 t0

104

0.03

0.07

3.00E+01

2.00E+00

6

2

8.6

9.03 t2

16

0.04

0.09

8.00E+01

2.00E+00

6

2

8.6

9.03 t3

7

0.04

0.09

5.00E+00

2.00E+00

96!

Table A-1 (cont’d).
se
31- wa
32
Oct ge
se
31- wa
32
Oct ge
se
31- wa
32
Oct ge
se
31- wa
32
Oct ge
se
31- wa
32
Oct ge
se
31- wa
33
Oct ge
se
31- wa
33
Oct ge
se
31- wa
33
Oct ge
se
31- wa
33
Oct ge

!

4

3

11.5

8.97 inf

0.00

0.00

1.00E+08

1.20E+02

4

3

11.5

8.97 t0

46

0.01

0.02

1.00E+00

2.00E+00

4

3

11.5

8.97 t1

28

0.05

0.11

3.00E+04

2.00E+00

4

3

11.5

8.97 t2

20

1.30

2.92

5.00E+05

2.00E+00

4

3

11.5

8.97 t3

10

0.08

0.18

1.00E+04

2.00E+00

5

2

11.1

9.00 inf

0.00

0.00

1.00E+08

1.50E+02

5

2

11.1

9.00 t0

75

0.01

0.02

1.00E+00

1.00E+00

5

2

11.1

9.00 t1

28

0.03

0.07

1.00E+00

1.00E+00

5

2

11.1

9.00 t2

20

0.05

0.11

1.00E+00

1.00E+00

97!

Table A-1 (cont’d).
Se
31- wa
33
Oct ge
se
31- wa
34
Oct ge
se
31- wa
34
Oct ge
se
31- wa
34
Oct ge
se
31- wa
34
Oct ge
635
Nov ww
635
Nov ww
635
Nov ww
6Nov ww
35
6Nov ww
36
6Nov ww
36

!

5

2

11.1

9.00 t3

12

0.04

0.09

1.00E+00

1.00E+00

6

1

11.9

8.99 inf

0.00

0.00

1.00E+08

6.00E+02

6

1

11.9

8.99 t0

70

0.01

0.02

1.00E+00

1.00E+00

6

1

11.9

8.99 t2

18

0.04

0.09

1.00E+00

1.00E+00

6

1

11.9

8.99 t3

7

0.04

0.09

1.00E+00

1.00E+00

6

1

13.4

9.00 inf

0.00

0.00

1.00E+07

6

1

13.4

9.00 t0

110

0.01

0.02

1.00E+00

6

1

13.4

9.00 t2

10

0.01

0.02

1.00E+00

6

1

13.4

9.00 t3

7

0.04

0.09

1.00E+00

4

2

13.7

8.95 inf

0.00

0.00

1.00E+07

4

2

13.7

8.95 t0

0.01

0.02

4.00E+01

86

98!

Table A-1 (cont’d).
636
Nov ww
6Nov ww
36
6Nov ww
36
6Nov ww
37
6Nov ww
37
6Nov ww
37
6Nov ww
37
se
9- wa
38
Nov ge
se
9- wa
38
Nov ge
se
9- wa
38
Nov ge
se
9- wa
38
Nov ge

!

4

2

13.7

8.95 t1

25

0.02

0.05

1.00E+03

4

2

13.7

8.95 t2

23

0.01

0.02

1.00E+05

4

2

13.7

8.95 t3

13

0.04

0.09

3.00E+03

5

3

13.5

8.99 inf

0.00

0.00

3.00E+07

5

3

13.5

8.99 t0

64

0.01

0.02

1.00E+00

5

3

13.5

8.99 t2

7

0.01

0.02

1.00E+00

5

3

13.5

8.99 t3

6

0.03

0.07

1.00E+00

5

3

10.7

8.99 inf

0.00

0.00

3.00E+07

5

3

10.7

8.99 t0

90

0.01

0.02

1.00E+00

5

3

10.7

8.99 t2

18

0.03

0.07

1.00E+00

5

3

10.7

8.99 t3

8

0.04

0.09

1.00E+00

99!

Table A-1 (cont’d).
se
9- wa
39
Nov ge
se
9- wa
39
Nov ge
Se
9- wa
39
Nov ge
se
9- wa
39
Nov ge
se
9- wa
40
Nov ge
se
9- wa
40
Nov ge
se
9- wa
40
Nov ge
se
9- wa
40
Nov ge
se
9- wa
40
Nov ge

!

6

2

11.3

8.96 inf

0.00

0.00

5.00E+07

6

2

11.3

8.96 t0

98

0.01

0.02

1.00E+00

6

2

11.3

8.96 t2

14

0.04

0.09

1.00E+00

6

2

11.3

8.96 t3

12

0.04

0.09

1.00E+00

4

1

12.5

8.97 inf

0.00

0.00

2.00E+07

4

1

12.5

8.97 t0

38

0.01

0.02

1.00E+00

4

1

12.5

8.97 t1

34

0.04

0.09

1.00E+03

4

1

12.5

8.97 t2

38

0.07

0.16

4.00E+05

4

1

12.5

8.97 t3

12

0.07

0.16

1.00E+04

100!

Table A-1 (cont’d).
2044
Dec ww
20Dec ww
44
20Dec ww
44
20Dec ww
44
20Dec ww
44
20Dec ww
45
20Dec ww
45
20Dec ww
45
2045
Dec ww
2046
Dec ww
2046
Dec ww
2046
Dec ww
2046
Dec ww

!

4

3

8.5

9.04 inf

0.00

0.00

3.10E+07

4

3

8.5

9.04 t0

46

0.05

0.11

1.40E+02

4

3

8.5

9.04 t1

52

0.05

0.11

1.40E+04

4

3

8.5

9.04 t2

18

0.20

0.45

6.70E+04

4

3

8.5

9.04 t3

10

0.40

0.90

1.50E+04

5

1

8.1

9.00 inf

0.00

0.00

2.40E+07

5

1

8.1

9.00 t0

63

0.05

0.11

1.30E+02

5

1

8.1

9.00 t2

14

0.10

0.23

6.60E+02

5

1

8.1

9.00 t3

10

0.40

0.90

1.00E+00

6

2

7.8

8.98 inf

0.00

0.00

2.40E+07

6

2

7.8

8.98 t0

72

0.05

0.11

3.60E+03

6

2

7.8

8.98 t1

26

0.05

0.11

1.20E+05

6

2

7.8

8.98 t2

24

0.30

0.68

2.10E+05

101!

Table A-1 (cont’d).
2046
Dec ww

!

6

2

7.8

8.98 t3

102!

11

0.50

1.13

6.00E+02

Table A-2. Murine norovirus influent concentrations.
Date Stock
Water
Experiment
Date
Purified
Type
well
2
22-Mar
2/15/12 water
3
10-Apr
2/29/12 sewage
4
10-Apr
2/29/12 sewage
5
10-Apr
2/29/12 sewage
well
6
14-Apr
3/19/12 water
well
7
14-Apr
3/19/12 water
well
8
14-Apr
3/19/12 water
9
17-Apr
3/19/12 sewage
10
17-Apr
3/19/12 sewage
11
17-Apr
3/19/12 sewage
well
12
19-Apr
3/21/12 water
well
13
19-Apr
3/21/12 water
14
24-Apr
3/21/12 sewage
15
24-Apr
3/21/12 sewage
16
24-Apr
3/21/12 sewage
well
23
11-May
5/10/12 water
well
24
11-May
5/10/12 water
well
25
11-May
5/10/12 water
well
29
27-Oct
5/10/12 water
well
30
27-Oct
5/10/12 water
well
31
27-Oct
5/10/12 water
32
31-Oct
10/23/12 sewage
33
31-Oct
10/23/12 sewage
34
31-Oct
10/23/12 sewage

!

103!

pH

Norovirus Influent
Concentration

7.5
7.5
7.5
7.5

2.00E+03
1.00E+04
2.00E+04
2.00E+04

7.5

1.10E+04

7.5

2.00E+04

7.5
7.5
7.5
7.5

1.00E+05
1.10E+04
2.80E+04
1.80E+04

7.5

4.00E+03

7.5
7.5
7.5
7.5

2.60E+04
2.00E+04
1.00E+04
9.00E+04

7.5

1.00E+04

7.5

2.00E+03

7.5

1.00E+04

9.0

2.00E+02

9.0

7.00E+02

9.0
9.0
9.0
9.0

2.00E+03
1.20E+02
1.50E+02
6.00E+02

REFERENCES

!

104!

REFERENCES
Abbaszadegan, M., Hasan, M.N., Gerba, C.P., Roessler, P.F., Wilson, B.R.,
Kuennen, R., van Dellen, E. 1997. The disinfection efficacy of a point-of-use
water treatment system against bacterial, viral and protozoan waterborne
pathogens. Water Research 31:574-582.
Ahmed, A.E.I., Hay, J.N., Bushell, M.E., Wardell, J.N., Cavalli, G. 2008. Biocidal
polymers (II): Determination of biological activity of novel N-halamnine
biocidal polymers and evaluation for use in water. Reactive & Functional
Polymers 68:1448-1458.
Amiri, F., Mesquita, M.M.F., Andrews, S.A. 2010. Disinfection effectiveness of
organic chloramines, investigating the effect of pH. Water Research 44:845853.
Bae, J., Schwab, K.J. 2008. Evaluation of murine norovirus, feline calicivirus,
poliovirus, and MS2 as surrogates for human norovirus in a model of viral
persistence in surface water and groundwater. Applied and Environmental
Microbiology 74:477-484.
Borchardt, M.A., Bertz, P.D., Spencer, S.K., Battigelli, D.A. 2003. Incidence of
enteric viruses in groundwater from household wells in Wisconsin. Applied
and Environmental Microbiology 69:1172-1180.
Borchardt, M.A., Bradbury, K.R., Alexander, E.C., Kolberg, R.J., Alexander, S.C.,
Archer, J.R., Braatz, LA., Forest, B.M., Green, J.A., Spencer, S.K. 2010.
Norovirus outbreak caused by a new septic system in a dolomite aquifer.
Ground Water 49:85-97.
Boschi-Pinto, C., Velebit, L., Shibuya, K. 2008. Estimating child mortality due to
diarrhoea in developing countries. Bulletin of the World Health Organization
86:710-717.
Brown, J., Sobsey, M.D. 2010. Microbiological effectiveness of locally produced
ceramic filters for drinking water treatment in Cambodia. Journal of Water
and Health 8:1-10.
Cannon, J.L., Papafragkou, E., Park, G.W., Osborne, J., Jaykus, L., Vinje, J.
2006. Surrogates for the study of norovirus stability and inactivation in the
environment: a comparison of murine norovirus and feline calicivirus. Journal
of Food Protection 69:2761-2765.
Centers for Disease Control and Prevention. 2012. Norovirus: Trends and
outbreaks. Available from http://www.cdc.gov/norovirus/trendsoutbreaks.html. Accessed June 11, 2012.
!

105!

Charles, K.J., Shore, J., Sellwood, J., Laverick, M., Hart, A., Pedley, S. 2009.
Assessment of the stability of human viruses and coliphage in groundwater by
PCR and infectivity methods. Journal of Applied Microbiology 106:18271837.
Chan, M.C.W., Sung, J.J.Y., Lam, R.K.Y.,m Chan, P.K.S., Lee, N.L.S., Lai,
R.W.M., Leung, W.K. 2006. Fecal viral load and norovirus-associated
gastroenteritis. Emerging Infectious Diseases 12:1278-1280.
Chen, Y., Worley, S.D., Kim, J., Wei, C., Chen, T., Santiago, J.I., Williams, J.F.,
Sun, G. 2003. Biocidal poly(strenehydantoin) beads for disinfection of water.
Industrial and Engineering Chemical Research 42:280-284.
Chen, Z., Sun, Y. 2006. N-halamine-based antimicrobial additives for polymers:
preparation, characterization, and antimicrobial activity. Industrial &
Engineering Chemistry Research 45:2634-2640.
Clasen, T., Nadakatti, S., Menon, S. 2006. Microbiological performance of a
water treatment unit designed for household use in developing countries.
Tropical Medicine and International Health 11:1399-1405.
Clasen, T., Naranjo, J., Frauchiger, D., Gerba, C. 2009. Laboratory assessment
of a gravity-fed ultrafiltration water treatment device designed for household
use in low-income settings. American Journal of Tropical Medical Hygiene
80:819-823.
Coulliette, A.D., Enger, K.S., Peterson, L.A., Weir, M.H., Rose, J.B. 2012. Risk
reduction assessment of waterborne Salmonella and Vibrio by a chlorine
contact disinfectant point-of-use device. International Journal of Hygiene and
Environmental Health In Press.
Coulliette, A.D., Peterson, L.A., Mosberg, J.A.W., Rose, J.B. 2010. Evaluation
of a new disinfection approach: efficacy of chlorine and bromine halogenated
contact disinfection for reduction of viruses and microcystin toxin. American
Journal of Tropical Medicine and Hygiene 82:279-288.
Cromeans, T.L., Kahler, A.M., Hill, V.R. 2010. Inactivation of adenoviruses,
enteroviruses, and murine norovirus in water by free chlorine and
monochloramine. Applied and Environmental Microbiology 76:1028-1033.
D’Souza, D.H., Sair, A., Williams, K., Papafragkou, E., Jean, J., Moore, C.,
Jaykus, L. 2006. Persistence of caliciviruses on environmental surfaces and
their transfer to food. International Journal of Food Microbiology 108:84-91.

!

106!

Duizer, E., Schwab, K.J., Neill, F.H., Atmar, R.L., Koopmans, M.P., Estes, M.K.
2004. Laboratory efforts to cultivate noroviruses. Journal of General
Virolology 85: 79–87.
Eaton, A.D., Greenberg, A.E., Clesceri, L.S., Franson, M.H. 1992. Standard
Methods for the Examination of Water and Wastewater, eighteenth ed.
American Public Health Association Publications, DC, pp. 309-347.
Eisenberg, J.N.S., Desai, M.A., Levy, K., Bates, S.J., Liang, S., Naumoff, K.,
Scott, J.C. 2007. Environmental determinants of infectious disease: a
framework for tracking causal links and guiding public health research.
Environmental Health Perspectives 115:1216-1223.
Elimelech, M. 2006. The global challenge for adequate and safe water. Journal
of Water Supply: Research and Technology 55:3-10.
Engelbrecht, R.S., Weber, M.J., Salter, B.L., Schmidt, C.A. 1980. Comparative
inactivation of viruses by chlorine. Applied and Environmental Microbiology
40:249-256.
Enger, K., Coulliette, A., Field, M., Rose, J. Draft Manuscript. Antimicrobial
effectiveness of a household water treatment device that delivers bromine.
Fankhauser, R.L., Monroe, S.S., Noel, J.S., Humphrey, C.D., Bresee, J.S.,
Parashar, U.D., Ando, T., Glass, R.I. 2002. Epidemiologic and molecular
trends of ‘Norwalk-like viruses’ associated with outbreaks of gastroenteritis in
the United States. Journal of Infectious Diseases 186:1–7.
Flewett, T.H., Wooe, G.N. 1978. The rotaviruses. Archives of Virology 51:1-23.
Floyd, R., Johnson, J.D., Sharp, D.G. 1976. Inactivation by bromine of single
poliovirus particles in water. Applied and Environmental Microbiology 31:298303.
Gallay, A., De Valk, H., Cournot, M., Ladeuil, B., Hemery, C., Castor, C., Bon, F.,
Mégraud, F., Le Cann, P., Desenclos, J.C. 2006. A large multi-pathogen
waterborne community outbreak linked to faecal contamination of a
groundwater system. Clinical Microbiology and Infection 12:561-570.
Goncalves, M.A.F.V., de Vries, A.A.F. 2006. Adenovirus: from foe to friend.
Reviews in Medical Virology 16:167-186.
Guo, G., Chen, X. 2009. Halogenating reaction activity of aromatic organic
compounds during disinfection of drinking water. Journal of Hazardous
Materials 163:1207-1212.

!

107!

Hale, A., Mattick, K., Lewis, D., Estes, M., Jiang, X., Green, J., Eglin, R., Brown,
D. 2000. Distinct epidemiological patterns of Norwalk-like virus infection.
Journal of Medical Virology 62:99–103.
Halonen, J.I., Kivimäki, M., Oksanen, T., Virtanen, P., Virtanen, M.J., Pentti, J.,
Vahtera, J. 2012. Waterborne outbreak of gastroenteritis: effects on sick
leaves and cost of lost workdays. PLoS ONE 7:e33307.
Junli, H., Li, W., Nenqi, R., Li, L.X., Fun, S.R., Guanle, Y. 1997. Disinfection
effect of chlorine dioxide on viruses, algae, and animal planktons in water.
Water Research 31:455-460.
Kaplan, J.E., Feldman, R., Campbell, D.S., Lookabaugh, C., Gary, G.W. 1982.
The frequency of Norwalk-like pattern of illness in outbreaks of acute
gastroenteritis. American Journal of Public Health 72:1329-1332.
Keller, R., Tetro, J.A., Springthorpe, V.S., Sattar, S.A. 2010. The influence of
temperature on norovirus inactivation by monochloramine in potable waters:
testing with murine norovirus as a surrogate for human norovirus. Food and
Environmental Microbiology 2:97-100.
Kenawy, E., Worley, S.D., Broughton, R. 2007. The chemistry and applications
of antimicrobial polymers: a state-of-the-art review. Biomacromolecules
8:1359-1384.
Keswick, B.H., Satterwhite, T.K., Johnson, P.C., DuPont, H.L., Secor, S.L.,
Bitsura, J., Gary, G.W., Hoff, J.C. 1985. Inactivation of Norwalk virus in
drinking water by chlorine. Applied and Environmental Microbiology 50:261264.
Kim, S., Cheon, D., Kim, J., Lee, D., Jheong, W., Heo, Y., Chung, H., Jee, Y.,
Lee, J. 2005. Outbreaks of gastroenteritis that occurred during school
excursions in Korea were associated with several waterborne strains of
norovirus. Journal of Clinical Microbiology 43:4836-4839.
Kitajima, M., Tohya, Y., Matsubara, K., Haramoto, E., Utagawa, E., Katayama, H.
2020. Chlorine inactivation of human norovirus, murine norovirus and
poliovirus in drinking water. Letters in Applied Microbiology 51:119-121.
Kosek, M., Bern, C., Guerrant, R.L. 2003. The global burden of diarrhoeal
disease, as estimated from studies published between 1992 and 2000.
Bulletin of the World Health Organization 81:197–198.
Kott, Y., Nupen, E.M., Ross, W.R. 1975. The effect of pH on the efficiency of
chlorine disinfection and virus enumeration. Water Research 9:869-872.

!

108!

Krasner, W.S. 2009. The formation and control of emerging disinfection byproducts of health concern. Philosophical Transactions of the Royal Society
367:4077-4095.
Lim, M.Y., Kim, J., Ko, G. 2010. Disinfection kinetics of murine norovirus using
chlorine and chlorine dioxide. Water Research 44:3243-3251.
Lloyd, S.J., Kovats, R.S., Armstrong, B.G. 2007. Global diarrhoea morbidity,
weather and climate. Climate Research 34:119-127.
Lopman, B.A., Reacher, M.H., Vipond, I.B., Sarangi, J., Brown, D.W.G. 2004.
Clinical manifestation of Norovirus gastroenteritis in health care settings.
Clinical Infectious Diseases 39:318-324.
Lysén, M., Thorhagen, M., Brytting, M., Hjertqvust, M., Andersson, Y., Hedlund,
K. 2009. Genetic diversity among food-borne and waterborne norovirus
strains causing outbreaks in Sweden. Journal of Clinical Microbiology
47:2411-2418.
Maunula, L., Miettinen, I.T., von Bonsdorff, C. 2005. Norovirus outbreaks from
drinking water. Emerging Infectious Diseases 11:1716-1721.
Mäusezahl, D., Christen, A., Pacheco, G.D., Tellez, F.A., Iriarte, M., Zapata,
M.E., Cevallos, M., Hattendorf, J., Cattaneo, M.D., Arnold, B., Smith, T.A.,
Colford, J.M. 2009. Solar drinking water disinfection (SODIS) to reduce
childhood diarrhoea in rural Bolivia: a cluster-randomized, controlled trial.
PLoS Medicine 6:e1000125.
McGuigan, K.G., Conroy, R.M., Mosler, H., du Preez, M., Ubomba-Jaswa, E.,
Fernandez-Ibañez, P. 2012. Solar water disinfection (SODIS): a review from
bench-top to roof-top. Journal of Hazardous Materials 235-236:29-46.
McLennan, S.D., Peterson, L.A., Rose, J.B. 2009. Comparison of point-of-use
technologies for emergency disinfection of sewage-contaminated drinking
water. Applied and Environmental Microbiology 75:7283-7286.
Mead, P.S., Slutsker, L., Dietz, V., McCaig, L.F., Bresee, J.S., Shapiro, C.,
Griffin, P.M., Tauxe, R.V. 1999. Food-related illness and death in the United
States. Emerging Infectious Diseases 5:607-625.
Mintz E, Reiff F, Tauxe R. 1995. Safe water treatment and storage in the home:
a practical new strategy to prevent waterborne disease. Journal of the
American Medical Association 273: 948-953.

!

109!

Monroe, S.S., Ando, T., Glass, R.I. 2000. Introduction: Human enteric
Caliciviruses-An emerging pathogen whose time has come. Journal of
Infectious Diseases 181:S249-251.
National Sanitation Foundation. 2012. NSF Product and Service Listings.
Available from
http://www.nsf.org/Certified/DWTU/Listings.asp?Company=C0041807&Stand
ard=042. Accessed January 9, 2013.
O’Reilly, C.E., Bowen, A.B., Perez, N.W., Sarisky, J.P., Shepherd, C.A., Miller,
M.D., Hubbard, B.C., Herring, M., Buchanan, S.D., Fitzgerald, C.C., Hill, V.,
Arrowood, M.J., Xiao, L.X., Hoekstra, R.M., Mintz, E.D., Lynch, M.F. 2007. A
waterborne outbreak of gastroenteritis with multiple etiologies among resort
island visitors and residents: Ohio, 2004. Clinical Infectious Diseases 44:506512.
Padmanabhuni, R.V., Luo, J., Cao, Z., Sun, Y. 2012. Preparation and
characterization of N-halamine-based antimicrobial fillers. Industrial and
Engineering Chemistry Research 51:5148-5156.
Pang, X., Honma, S., Nakata, S., Vesikari, T. 2000. Human Caliciviruses in
acute gastroenteritis of young children in the community. Journal of
Infectious Diseases 181:S288-294.
Patel, M.M., Hall, A.J., Vinjé, J., Parashar, U.D. 2009. Noroviruses: A
comprehensive review. Journal of Clinical Virology 44:1-8.
Payment, P., Richardson, L., Siemiatycki, J., Dewar, R., Edwardes, M., Franco,
E., 1991. A randomized trial to evaluate the risk of gastrointestinal disease
due to consumption of drinking water meeting current microbiological
standards. American Journal of Public Health 81:703–707.
Podewils, L.J., Zanardi Blevins, L., Hagenbuch, M., Itani, D., Burns, A., Otto, C.,
Blanton, L., Adams, S., Monroe, S.S., Beach, M.J., Widdowson, M. 2007.
Outbreak of norovirus illness associated with a swimming pool. Epidemiology
and Infection 135:827-833.
Quick, R.E., Venczel, L.V., Mintz, E.D., Soleto, L., Aparicio, J., Gironaz, M.,
Hutwagner, L., Greene, K., Bopp, C., Maloney, K., Chavez, D., Sobsey, M.,
Tauxe, R.V. 1999. Diarrhoea prevention in Bolivia through point-of-use water
treatment and safe storage: a promising new strategy. Epidemiology and
Infection 122:83-90.
Reynolds, K.A., Mena, K.D., Gerba, C.P. 2008. Risk of waterborne illness via
drinking water in the United States. Reviews of Environmental Contamination
and Toxicology 192:117-158.
!

110!

Richards, G.P., Watson, M.A., Meade, G.K., Hovan, G.L., Kingsley, D.H. 2012.
Resilience of norovirus GII.4 to freezing and thawing: Implications for virus
infectivity. Food and Environmental Virology 4:192-197.
Schwarzenbach, R.P., Egli, T., Hofstetter, T.B., von Gunten, U., Wehrli, B. 2010.
Global water pollution and human health. Annual Review of Environment and
Resources 35:109-136.
Seitz, S.R., Leon, J.S., Schwab, K.J., Lyon, G.M., Dowd, M., McDaniels, M.,
Abdulhafid, G., Fernandez, M.L., Lindesmith, L.C., Baric, R.S., Moe, C.L.
2011. Norovirus infectivity in humans and persistence in water. Applied and
Environmental Microbiology 77:6884-6888.
Seo, K., Lee, J.E., Lim, M.I., Ko, G. 2011. Effect of temperature, pH, and NaCl
on the inactivation kinetics of murine norovirus. Journal of Food Protection
75:533-540.
Shin, G., Sobsey, M.D., 1998. Reduction of Norwalk Virus, poliovirus 1 and
coliphage MS2 by monochloramine disinfection of water. Water Science and
Technology 38:151–154.
Shin, G., Sobsey, M.D. 2008. Inactivation of norovirus by chlorine disinfection of
water. Water Research 42:4562-4568.
Shirasaki, N., Matsushita, T., Matsui, Y., Oshiba, A., Ohno, K. 2010. Estimation
of norovirus removal performance in a coagulation-rapid sand filtration
process by using recombinant norovirus VLPs. Water Research 44:13071316.
Singer, P.C. 1993. Formation and characterization of disinfectant by-products.
In: Safety of Water Disinfection: Balancing Chemical and Microbial Risks
(Craun, G.F., ed.) ILSI Press, DC, pp. 201.
Singer, P.C. 1999. Formation and Control of Disinfection By-Products in
Drinking Water, American Water Works Association, DC, pp. 27-40.
Skern, T. 2010. 100 years poliovirus: from discovery to eradication. A meeting
report. Archives of Virology 155:1371-1381.
Sobsey, M.D. 2002. Managing water in the home: accelerated health gains from
improved water supply. Available from
http://www.who.int/water_sanitation_health/dwq/wsh0207/en/index.html.
Accessed October 29, 2012.

!

111!

Sobsey, M.D., Handzel, T., Venczel, L. 2003. Chlorination and safe storage of
household drinking water in developing countries to reduce waterborne
disease. Water Science and Technology 47:221-228.
Sobsey, M.D., Stauber, C.E., Casanova, L.M., Brown, J.M., Elliot, M.A., 2008.
Point of use household drinking water filtration: a practical, effective solution
for providing sustained access to safe drinking water in the developing world.
Environmental Science and Technology 42:4261–4267.
Song, R., Westerhoff, P., Minear, R.A., Amy, G.L. 1996. Interactions between
bromine and natural organic matter. In: Water Disinfection and Natural
Organic Matter: Characterization and Control (Minear, R.A., Amy, G.L., eds.).
American Chemical Society, DC, pp. 298-319.
Teunis, P.F.M., Moe, C.L., Liu, P., Miller, S.E., Lindesmith, L., Baric, R.S., Le
Pendu, J., Calderon, R.L. 2008. Norwalk virus: how infectious is it? Journal
of Medical Virology 80:1468-1476.
Timofeeva, L., Klescheva, N. 2011. Antimicrobial polymers: mechanism of
action, factors of activity, and applications. Applied Microbial Biotechnology
89:475-492.
United States Environmental Protection Agency. 1987. Guide standard and
protocol for testing microbiological water purifiers. Washington, DC: United
States Environmental Protection Agency, Registration Division, Office of
Pesticide Programs and Criteria and Standards Division, Office of Drinking
Water.
Vega, E., Barclay, L., Gregoricus, N., Williams, K., Lee, D., Vinjé, J. 2011. Novel
surveillance network for Norovirus gastroenteritis outbreaks, United States.
Emerging Infectious Diseases 17:1398-1395.
Wegelin, M., Canonica, S., Mechsner, K., Fleischmann, T., Pesaro, F., Metzler,
A. 1994. Solar water disinfection: scope of the process and analysis of
radiation experiments. Journal of Water Supply: Research and Technology Aqua 43:154-169.
Westerhoff, P., Chao, P., Mash, H. 2004. Reactivity of natural organic matter
with aqueous chlorine and bromine. Water Research 38:1502-1513.
Williams, J., Bridges, M. 2010. Drinking water: New disinfecting medium boosts
water treatment. Filtration and Separation 47:16-19.
Wobus, C.E., Karst, S.M., Thackray, L.B., Chang, K.O., Sosnovtsev, S.V., Belliot,
G., Krug, A., Mackenzie, J.M., Green, K.Y., Virgin, H.W. 2004. Replication of

!

112!

norovirus in cell culture reveals a tropism for dendritic cells and macrophages.
PLoS Biology 2:2076-2084.
Wobus, C.E., Thackray, L.B., Virgin, H.W. 2006. Murine norovirus: a model
system to study norovirus biology and pathogenesis. Journal of Virology
80:5104-5112.
World Health Organization. 2000. Hepatitis A. Available from
http://www.who.int/csr/disease/hepatitis/whocdscsredc2007/en/. Accessed
October 18, 2012.
World Health Organization. 2002. The World Health Report 2002. Available
from http://www.who.int/whr/2002/en. Accessed October 29, 2012.
World Health Organization. 2004. Water, Sanitation, and Hygiene Links to
Health. Available from
http://www.who.int/water_sanitation_health/publications/facts2004/en/.
Accessed June 11, 2012.
World Health Organization. 2011a. Guidelines for Drinking-Water Quality.
Available from
http://who.int/water_sanitation_health/publications/2011/dwq_chapters/en/ind
ex.html. Accessed June 11, 2012.
World Health Organization. 2011b. UN-Water Global Analysis and Assessment
of Sanitation and Drinking-Water. Available from
http://whqlibdoc.who.int/publications/2012/9789241503365_eng.pdf.
Accessed September 20, 2012.
World Health Organization. 2011c. Evaluating Household Water Treatment
Options: Health-based targets and microbial performance specifications.
Available from
http://who.int/water_sanitation_health/publications/2011/evaluating_water_tre
atment.pdf. Accessed September 20, 2012.
Wu, H.M., Fornek, M., Schwab, K.J., Chapin, A.R., Gibson, K., Schwab, E.,
Spencer, C., Henning, K. 2005. A norovirus outbreak at a long-term-care
facility: the role of environmental surface contamination. Infection Control and
Hospital Epidemiology 26:802-810.

!

113!