Comparison of Three Household Water TreatmentTechnologies in.docx

Comparison of Three Household Water Treatment
Technologies in San Mateo Ixtatán, Guatemala
Jonathan E. Mellor1; Erin Kallman2; Vinka Oyanedel-Craver,
A.M.ASCE3; and James A. Smith, F.ASCE4
Abstract: Silver-impregnated ceramic water filters (CWFs) are a
simple and sustainable low-cost technology that has shown
promise in
improving household drinking water quality and reducing
incidences of early childhood diarrhea in a variety of settings.
Despite this promise,
lower reservoir contamination is thought to be a contributing
factor to the decline in the effectiveness being seen over time.
A novel silver-
impregnated ceramic torus that can be placed in the lower
reservoir was designed to minimize this contamination. This
study uses a one-year
randomized trial to compare the relative effectiveness of the
CWF þ torus design with a standard CWF and point-of-use
chlorination. The
effectiveness of each technology was measured at project
inception and subsequently after six and 12 months. Results
indicate that the
toruses, as designed, are not able to consistently maintain
lower-reservoir silver concentrations above those of the simple
CWF design
and are hence unable to prevent contamination. Furthermore,
after six months, only 65% of households that used point-of-use
chlorination
maintained sufficient chlorine levels above the 0.2 mg=L
needed to be effective. All three technologies showed

statistically equivalent log
removal efficiencies for total coliform bacteria and all three
declined in effectiveness over the first six months. Combined
average log removal
efficiencies for all three technologies ranged from 2.22 � 0.21
initially but declined to 1.45 � 0.35 after six months and to
1.42 � 0.29 after
one year. DOI: 10.1061/(ASCE)EE.1943-7870.0000914. © 2014
American Society of Civil Engineers.
Introduction
Worldwide, there are an estimated 780 million people who lack
access to improved water sources and an additional 2.2 billion
without consistent access to microbiologically safe water (Onda
et al. 2012). This poor access is a leading cause of death for
842,000 children a year who suffer from poor access to water,
san-
itation, and hygiene (WASH) services (Pruss-Ustun et al. 2014).
Even when a community has a clean water source, water is fre-
quently contaminated after collection or treatment (Wright et al.
2004) by a myriad of contamination sources as well as
biological
regrowth (Mellor et al. 2013). This is a particular problem for
res-
idents who must travel long distances to collect water (Mellor et
al.
2012b) and thus store their water for extended periods.
One point-of-use (POU) technology that has shown promise
at improving household drinking water quality is a ceramic
water
filter (CWF). CWFs consist of a porous ceramic pot-shaped
filter
that purifies water using size exclusion and an antimicrobial
silver

nano-particle coating. Contaminated water can be poured into
the
top from where it gradually percolates down through the
ceramic
and is collected in a plastic lower reservoir. CWFs have been
shown
to be highly effective at removing bacteria in both laboratory
(97.8–100% removal) (Oyanedel-Craver and Smith 2008) and
field
environments (∼90% average removal) (Kallman et al. 2011)
and
reducing diarrhea incidences both generally (Hunter 2009) as
well
as in human immunodeficiency virus-positive adult populations
(Abebe et al. 2014). CWFs are also likely able to remove
viruses
and protozoa, but with lower and higher respective effectiveness
(Bielefeldt et al. 2010). Indeed, recent work has shown Crypto-
sporidium parvum removal to be 99.2% through ceramic disks
(Abebe 2013).
Although these are positive results, some researchers have
recently questioned the ability of ceramic filters and other
point-
of-use interventions to reduce diarrhea incidence especially
over
the long term. Hunter (2009) found that the effectiveness of
point-of-use interventions as a means of diarrhea reduction
declines with follow-up duration and that blinded studies
indicate
lower effectiveness. Others have suggested that household water
treatment technologies should not be promoted widely until
further
research is conducted (Schmidt and Cairncross 2009). One
possible

cause of the long-term decline in effectiveness is biological
buildup
that has been shown to be present in household water storage
con-
tainers (Jagals et al. 2003) (which are similar in form and
function
to the lower reservoir of CWFs). This, along with biological re-
growth, may be a significant contributing factor to early-
childhood
diarrhea incidence in South Africa (Mellor et al. 2012a). In fact,
previous results obtained by the authors have suggested that this
biological buildup coupled with poor cleaning regimes might be
significant factors in CWFs’ declining effectiveness (Mellor et
al.
2014).
CWFs effectively remove bacteria both through size exclusion
and the antimicrobial action of colloidal silver or silver nitrate
that
is painted on or infused into the ceramic. The bactericidal
proper-
ties of the applied silver are dependent on the applied mass of
colloidal silver (Oyanedel-Craver and Smith 2008) and the
reten-
tion of silver in the filter (Ren et al. 2013). The bacterial
growth
inhibition by silver is dependent on the silver dose applied to
the
CWF (Rayner et al. 2013) and the number of bacteria present
(Sondi and Salopek-Sondi 2004).
1Postdoctoral Research Associate, Dept. of Chemical and
Environmen-
tal Engineering, Yale Climate and Energy Institute, Yale Univ.,
P.O. Box
208286, New Haven, CT 06520. E-mail: [email protected]

2Engineer, Dept. of Civil and Environmental Engineering, Univ.
of
Virginia, P.O. Box 40072, Charlottesville, VA 22904-4742.
3Assistant Professor, Dept. of Civil and Environmental
Engineering,
Univ. of Rhode Island, Bliss Hall 213, Kingston, RI 22881. E-
mail:
[email protected]
4Professor, Dept. of Civil and Environmental Engineering,
Univ. of
Virginia, P.O. Box 40072, Charlottesville, VA 22904-4742
(corresponding
author). E-mail: [email protected]
Note. This manuscript was submitted on November 11, 2013;
approved
on October 16, 2014; published online on November 18, 2014.
Discussion
period open until April 18, 2015; separate discussions must be
submitted
for individual papers. This paper is part of the Journal of
Environmental
Engineering, © ASCE, ISSN 0733-9372/04014085(7)/$25.00.
© ASCE 04014085-1 J. Environ. Eng.
J. Environ. Eng. 2015.141.
D
ow
nl
oa
de
d

fr
om
a
sc
el
ib
ra
ry
.o
rg
b
y
N
or
th
er
n
A
ri
zo
na
U
ni
v

on
0
9/
09
/1
5.
C
op
yr
ig
ht
A
S
C
E
. F
or
p
er
so
na
l
us
e

on
ly
;
al
l
ri
gh
ts
r
es
er
ve
d.
http://dx.doi.org/10.1061/(ASCE)EE.1943-7870.0000914
The safe water system (SWS) is another common point-of-use
water treatment system developed by the U.S. Centers for
Disease
Control and Prevention (Mintz et al. 1995). It combines house-
hold water treatment and safe water storage. Treatment is
typically
done using sodium hypochlorite (NaOCl) added to household
drinking water. Free chlorine residual should be between 0.2
and

2.0 mg=L (Lantagne 2008). The safe water storage containers
have
small openings to prevent contamination and a spigot to allow
easy
access.
Attempting to address the problem of lower-reservoir reconta-
mination, members of the nonprofit organization Potters for
Peace
proposed placing a ceramic torus painted with colloidal silver
(Fig. 1) in the lower reservoir of the filter system (Fig. 2). It
was
hypothesized that the release of silver ions from the colloidal
silver-impregnated torus in the lower reservoir would help to
pre-
vent microbial regrowth and biofilm formation. To test this
claim,
the microbial effectiveness of three technologies was compared
concurrently in the same community over the course of a year:
a CWF; a CWF with the torus placed in the lower reservoir; and
the safe water system. Therefore, the overarching goal of this
study
was to compare the longitudinal microbial effectiveness of the
three technologies over a year-long period while testing the
novel
silver-impregnated torus’s ability to slowly release silver thus
im-
proving the water quality in the lower reservoir. This goal was
achieved through the recruitment of 107 participants from San
Mateo Ixtatán, Guatemala, who were randomly divided into
three
study groups. Each group received one intervention.
Methods

Community Setting and Cohort
The study was undertaken in the community of San Mateo
Ixtatán
in the Guatemalan highlands. Access to suitable WASH infra-
structure is severely limited and diarrhea is common in this
region
making it a leading cause of death among children in Guatemala
(Guatemala 2002). San Mateo Ixtatán is the poorest community
in
the poorest department of Huehuetenango and has a population
of
approximately 30,000. Although the community has an
extensive
spring-fed water distribution system, the water is not treated
and
is of poor quality (Kallman et al. 2011). This system is the only
source of drinking water and is used by all members of the com-
munity with the exception of the occasional use of bottled
water.
In June 2009, 107 participants were recruited to participate
in the study and were randomly assigned to one of three groups
of approximately equal size as shown in Fig. 3. Participants
were
invited through flyers and radio announcements. Technology as-
signments were done via a rotation where the first person in line
got a CWF, the second a CWF þ torus, the third the SWS, etc.
The
SWS group received their first chlorine bottles free of charge
and
could purchase dosed chlorine bottles from a local distributor
for
approximately 25 quetzales (≈US$3.14) for a six-month supply.
Instructions were given to the participants about operation and
maintenance of their technologies at project inception.

However,
there was no systematic attempt to assesses actual compliance
or
maintenance practices. Study attrition was mostly due to partici-
pants not being present during the Period 2 and 3 sampling
rounds.
Two attempts were made at each household to find participants.
Institutional review board approval was obtained for this study
from the University of Virginia.
Torus Fabrication
The toruses were fabricated in San Mateo Ixtatán using a
method
similar to the one described by Kallman et al. (2011) to
fabricate
CWFs, however they were hand-molded instead of pressed. In
brief, approximately 60 lb (27.2 kg) of locally collected clay is
combined with 8–10 lb (3.6–4.5 kg) of sieved sawdust. Once
mixed, 10 L of water is added and the toruses are molded by
hand.
They are then allowed to air-dry for 8 days, after which time
they
are fired at a temperature of 800°C. The temperature was slowly
increased from ambient by 75°C=h for 4 h and then by 150°C=h
until the maximum temperature was reached. They are then
hand-painted with approximately 23 mL of 200 ppm silver
nano-particle solution and allowed to dry. The finished torus
mean
mass was 175 g with a range of 158–213 g.
Fig. 1. Silver-impregnated torus investigated in this study
(image by
James Smith)

Torus
Ceramic Water Filter
Water Level
Lower Reservoir
Fig. 2. Ceramic water filter plus silver-infused ceramic torus
(CWF þ
torus) configuration; the ceramic water filter is placed inside
the lower
reservoir as shown; the torus is then placed underwater at the
bottom of
the reservoir where it can inhibit biological growth
D
ow
nl
oa
de
d
fr
om
a
sc
el

ib
ra
ry
.o
rg
b
y
N
or
th
er
n
A
ri
zo
na
U
ni
v
on
0
9/
09
/1

5.
C
op
yr
ig
ht
A
S
C
E
. F
or
p
er
so
na
l
us
e
on
ly
;
al
l

ri
gh
ts
r
es
er
ve
d.
Analytical Methods
For each household sampling event, water samples were
collected
from the household tap (which was the source water) and from
the
spigot of the CWF or SWS. Sampling took place in June 2009,
January 2010, and June 2010, which will hereafter be referred
to as Periods 1, 2, and 3, respectively. Period 1 sampling was
con-
ducted within approximately two days of a household receiving
their filter.
Silver concentration was measured in the field each time using a
Hach DR/4000 spectrophotometer and the Hach 8120 silver
colori-
metric method (Hach 2003) (Hach Company, Loveland,
Colorado).
The detection limits for that method are 0.02–0.70 mg=L. Free
chlorine levels were measured during Periods 1 and 2 using
Hach

Chlorine Color Disk Test Kits Model CN-66 (Hach Company,
Loveland, Colorado). The effective range is 0.1–3.4 mg=l Cl2.
The samples taken from each household at each period were
tested for total coliform bacteria during each of the three visits
and
E. coli bacteria during the last two visits using standard
membrane
filtration methods. All samples were collected in sterile, 250
mL
plastic bottles and stored in a cooler with ice during transport to
the field laboratory. All samples were tested within eight hours
of
collection. The selective media m-ColiBlue24 (Millipore,
Billerica,
Massachusetts) was not available during the first sampling
period
necessitating the use of m-Endo Total Coliform Broth
(Millipore,
Billerica, Massachusetts), which does not differentiate between
E. coli and total coliforms. m-ColiBlue24 was used for the
subse-
quent periods.
The membrane filtration protocol is as follows. A hand pump
was first used to pass 100 mL of each undiluted sample through
a sterilized 0.45 μm membrane filter (Fisher Brand, Pittsburgh).
The filter was then placed in a petri dish containing m-
ColiBlue24
or m-Endo total coliform broth and incubated at 35°C for 24 h
in
a portable incubator (Millipore, Billerica, Massachusetts).
Colonies
were then counted and reported as CFU=100 mL. Plates with
too

many colonies to count were recorded as having 2,000 CFU=
100 mL. Daily boiled water samples all had zero colonies.
Statistical Methods
Mean differences in silver concentrations over time (within-
subjects effects) and between technologies (between-subjects
ef-
fects) were assessed using repeated-measures analysis of
variance
(ANOVA). Shapiro-Wilk’s tests of normality, Levene’s test for
equality of error variances, Box’s test of equality of covariance
and
Mauchly’s test for sphericity were all conducted on the data to
test
if ANOVA assumptions were met.
A paired t-test was conducted to assess the chlorine concentra-
tion changes.
Log reduction values were calculated from the influent and
effluent water samples and used in the subsequent analyses of
the microbial concentrations. One-way ANOVA analyses were
conducted at each time period to assess the equity of the
influent
water for the three technologies and two bacteria types. The
micro-
bial effectiveness of the three technologies was assessed using a
repeated-measures ANOVA in an identical manner to the silver
concentrations. Pearson correlation coefficients were calculated
to
see if effluent water silver concentrations were correlated to the
log
removal efficiency of the filters.
Finally, Student’s t-test were conducted to assess the difference

in log removal rates between households with and without suffi-
cient chlorine during Period 2. F-tests were used to assess
variance
for t-test analyses.
IBS SPSS statistical analysis software version 21.0 (IBM SPSS,
Chicago, 2011) as well as Microsoft Excel were used for all
analy-
ses. All tests were conducted with a significance level of 0.05.
Results
Fig. 4 presents mean silver concentrations (with 95%
confidence
intervals) in treated water for the CWF and CWF þ torus inter-
ventions for each of the three sampling periods. All samples
were
107 Initial Participants
37 CWF 36 CWF + Torus 34 SWS
P
e
ri
o
d
1
(
Ju
n
e
2

0
0
9
)
P
e
ri
o
d
3
(
Ju
n
e
2
0
1
0
)
P
e
ri
o
d
2
(

Ja
n
2
0
1
0
)
Fig. 3. Three-product study design; 107 initial participants were
re-
cruited in June 2009 and were approximately evenly divided
randomly
into three study arms; there was significant dropout during the
subse-
quent follow-up visits
Sampling Period
[A
g
]
m
g
/l
1 2 3
CWF
CWF + Torus
0
.0

2
0
.0
3
0
.0
4
0
.0
5
Fig. 4. Silver concentration for the three time periods sampled;
repeated-measures ANOVA analyses indicated there was no
variation
with either time or between technologies; error bars indicate
95% CI
D
ow
nl
oa
de
d
fr
om

a
sc
el
ib
ra
ry
.o
rg
b
y
N
or
th
er
n
A
ri
zo
na
U
ni
v
on
0

9/
09
/1
5.
C
op
yr
ig
ht
A
S
C
E
. F
or
p
er
so
na
l
us
e
on
ly

;
al
l
ri
gh
ts
r
es
er
ve
d.
below the World Health Organization (WHO) recommendation
for
silver concentration, which is 0.1 mg=L (WHO 2011). Although
apparent mean differences in effluent water silver concentration
were seen over time (Fig. 4), these differences were not
statistically
significant according to the repeated measures ANOVA analysis
[Fð2; 34Þ ¼ 0.812, p ¼ 0.412]. Likewise, there were no
significant
differences between technologies [Fð1; 17Þ ¼ 0.160, p ¼
0.694].
Chlorine concentrations fell from Period 1 to Period 2 for
house-
holds that received the safe water system, however, this result
was not statistically significant according to a paired t-test (p ¼

0.125). During the first sampling round, chlorine concentration
was 1.43 mg=L on average with 100% (n ¼ 34) having more
than
0.2 mg=L. However, during the Period 2 sampling only 65% (n
¼
20) had levels of 0.2 mg=L or higher although the average rose
to 2.66 mg=L. This is due to the fact that 20% of households
had
concentrations in excess of 5 mg=L, indicating a minority of
house-
holds were possibly overchlorinating.
Table 1 summarizes mean tap (influent) water quality for the
three technologies, two bacteria types, and three sampling
periods.
According to one-way ANOVA analyses, mean tap water
samples
for both E. coli and total coliform bacteria were statistically
equiv-
alent for all three technologies for Periods 2 and 3. However,
they
were not equivalent for Period 1 (p ¼ 0.037) as is shown in
Table 1.
Mean log reduction values for all three technologies over the
sam-
pling periods for the two bacteria types are shown in Figs. 5 and
6,
while the same data is displayed as boxplots in Fig. 7.
Combined
average log removal efficiencies for all three technologies for
total
coliform bacteria ranged from 2.22 � 0.21 initially but declined
to
1.45 � 0.35 after six months and to 1.42 � 0.29 after 1 year.
Total
coliform removal declined between Period 1 and Period 3 from

2.20 to 1.18 for CWF, from 2.10 to 1.48 for the CWF þ torus
design, and from 2.37 to 1.60 for the SWS. E. coli removal
changed
from the Period 2 to Period 3 sampling rounds from 1.45 to 1.37
for
the CWFs, from 1.51 to 1.87 for the CWF þ torus design and
from
1.26 to 1.05 for the SWS.
A repeated-measures ANOVA of all three technologies indi-
cated that there were mean differences over time [Fð2; 92Þ ¼
12.410, p < 0.001], but not between the technologies [Fð2; 46Þ
¼
0.417, p ¼ 0.661] for total coliform bacteria. Pairwise
comparisons
identified mean differences between Period 1 and Period 2 (p <
0.001), but not between Period 2 and Period 3 (p ¼ 1.000).
How-
ever, there was no similar temporal decline for E. coli [Fð1; 32Þ
¼
0.008, p ¼ 0.930] nor was there a difference between
technologies
for E. coli [Fð2; 32Þ ¼ 1.409, p ¼ 0.259]. The lack of temporal
decline for E. coli is likely due to the lack of data for Period 1.
There was also no correlation between silver concentration in
the effluent water and log reduction values for total coliform (R
¼
0.08) or E. coli (R ¼ 0.087).
Results are also displayed in terms of the WHO risk categories
of <1, 1–10, 10–100, 100–1,000 and >1,000 CFU=100 mL in
Figs. 8 and 9. There plots are largely consistent with the
previous
results and indicate that the effectiveness of all three
technologies

declines between Periods 1 and 2, but remains relatively
constant
between Periods 2 and 3. There are few differences between the
three technologies.
Finally, the log removal rates for households with and without
sufficient residual chlorine during the Period 2 sampling was
com-
pared. Households with chlorine concentrations of 0.2 mg=L
had
Table 1. Mean Tap/Influent Water Quality for the Three
Sampling Periods
for Each of the Three Technologies and Two Bacteria Types; P-
Values are
the Result of One-Way ANOVA Analyses to Compare Means;
Results
Indicate that Means are Statistically Equivalent Except for the
First
Sampling Period
Sampling
period
Bacteria
type
Mean tap (influent) water quality
(CFU=100 mL)
p-ValueCWF CWF þ torus SWS
1 TC 465 499 914 0.037
2 TC 525 773 651 0.746
2 EC 352 328 255 0.886
3 TC 798 632 328 0.333
3 EC 269 221 66 0.405

Note: EC = E. Coli; TC = total coliform.
Total Coliform
Sampling Period
L
o
g
R
e
d
u
ct
io
n
0
.5
1
.0
1
.5
2
.0
2
.5
1 2 3

CWF
CWF + Torus
SWS
Fig. 5. Mean log reduction over the three sampling periods for
total
coliform bacteria; repeated-measure ANOVA test indicates
temporal
decline for total coliform bacteria between Periods 1 and 2; no
signif-
icant differences were found between technologies; plot is
consistent
with a temporal decline in effectiveness, but limited differences
between technologies; error bars indicate 95% CI
E. Coli
Sampling Period
L
o
g
R
e
d
u
ct
io
n
0
.5

1
.0
1
.5
2
.0
2
.5
2 3
CWF
CWF + Torus
SWS
Fig. 6. Mean log reduction over the two sampling periods for E.
coli
bacteria; there was no temporal decline for E. coli and no
significant
difference between technologies; the lack of temporal decline is
likely
due to the fact that no E. coli data was taken for Period 1; error
bars
indicate 95% CI
D
ow
nl

oa
de
d
fr
om
a
sc
el
ib
ra
ry
.o
rg
b
y
N
or
th
er
n
A
ri
zo
na

U
ni
v
on
0
9/
09
/1
5.
C
op
yr
ig
ht
A
S
C
E
. F
or
p
er
so
na

l
us
e
on
ly
;
al
l
ri
gh
ts
r
es
er
ve
d.
significantly higher log removal rates for total coliform bacteria
than those that did not (1.79 versus 0.41, p ¼ 0.022) (Student’s
t-test). However, log removal rates were equivalent for E. coli
bacteria (1.09 versus 1.09, p ¼ 0.936) (Student’s t-test).
Discussion
This paper reports on a randomized trial of the longitudinal
field

effectiveness of three point-of-use water treatment systems. To
the
best of the authors’ knowledge, this is the first concurrent,
compar-
ative study of these three technologies. Results indicate that all
three technologies decline in effectiveness over the first six
months
and that the toruses, as designed, are not sufficient to improve
performance over the simple CWF. However, it is also
noteworthy
that the microbial effectiveness decline appears to level off
after
six months for all three technologies. Furthermore, it is evident
that chlorination adherence falls precipitously during the first
six months, which is something that can affect its suitability as
a
sustainable point-of-use intervention.
The longitudinal declines in CWF effectiveness are consistent
with those reported previously from a study conducted in South
Africa (Mellor et al. 2014) and are likewise consistent with that
of
Hunter (2009) who found that longer duration studies of POU
devices showed decreased effectiveness at reducing diarrhea.
The
high variability and negative log reduction values have likewise
been seen by Brown (2007) who found that 17% of CWF
effluent
samples had higher E. coli concentrations in the treated water
com-
pared to the influent water.
One reason for the apparent effectiveness declines seen in the
CWF and the CWF þ torus designs may be the depletion of the
silver in the filters. As noted by Ren et al. (2013), silver

nanopar-
ticles are relatively mobile through a ceramic porous media, and
colloidal silver solutions that are painted onto porous ceramic
filters
(as they were in this case) result in silver nanoparticle release
into
the treated water at a rate much faster than filters fabricated by
firing the nanosilver into the filter.
The relative ineffectiveness of the torus design is surprising.
If designed properly, such a technology should reduce the
biofilm
buildup quantified by others (Jagals et al. 2003) and help to
mit-
igate regrowth in such settings (Mellor et al. 2013). One
possibility
is that the reservoir silver concentrations, which ranged from
∼15 to
45 ppb (Fig. 4), were insufficient to deactivate high
concentrations
of bacteria. However, prior research indicates that ∼2 log
reduction
occurs in about 30 min at concentrations down to at least 50 ppb
(Jung et al. 2008). In the filter design, the microbes pass
through the
pores of the silver-impregnated ceramic, which forces them into
very close contact with the silver-impregnated pore surfaces.
This
is apparently not occurring as efficiently in the toruses because
the
water is not forced through the torus. It is possible that the
toruses
could be more effective if they were painted with higher
concen-
trations of colloidal silver, had different pore sizes, or had a
differ-

ent geometry. However, the Pearson’s test indicates that there is
no
L
o
g
R
e
d
u
ct
io
n
-2
-1
0
1
2
3
4
5
TC Period 1 TC Period 2 TC Period 3 EC Period 2 EC Period 3
SWS
Technology
CWF + Torus CWF

Fig. 7. Boxplots of log reductions for the three sampling periods
for
each of the three technologies and two bacteria types [EC (E.
coli) and
TC (total coliform)]
CWF CWF+Torus SWS CWF CWF+Torus SWS CWF
CWF+Torus SWS
P
e
rc
e
n
t
(%
)
0
2
0
4
0
6
0
8
0
1

0
0 Period 1 Period 2 Period 3
Risk Category (CFU/100ml)
<1 1-10 10-100 100-1000 >1000
Fig. 8. Bar plots showing the percent of samples with total
coliform concentrations in WHO risk categories at the three
periods; results indicate that
water quality declines between Period 1 and Period 2 for all
three technologies but remains relatively constant between
Period 2 and Period 3; there is
minimal difference between the three technologies
D
ow
nl
oa
de
d
fr
om
a
sc
el
ib

ra
ry
.o
rg
b
y
N
or
th
er
n
A
ri
zo
na
U
ni
v
on
0
9/
09
/1
5.

C
op
yr
ig
ht
A
S
C
E
. F
or
p
er
so
na
l
us
e
on
ly
;
al
l
ri

gh
ts
r
es
er
ve
d.
relationship between silver concentrations and log reduction
values.
It is also notable that silver concentrations showed little varia-
tion with time or between the two CWF configurations. The one
exception to this was during the second sampling period when
the
torus design appears to have had higher levels of silver. This
could
have led to the statistically insignificant mean increase in log
reduction for that period for both total coliform (1.42 versus
1.59,
p ¼ 0.636) and E. coli (1.41 versus 1.53, p ¼ 0.738) for the
CWF
versus CWF þ torus designs, respectively. Another possibility is
that reservoir contamination was not due to contamination from
the reservoir itself, but rather from the ceramic filter walls or
pores as was found by others in a controlled laboratory
experiment
(Bielefeldt et al. 2009). Also, the detection range of the chosen
method (Hach 8120) is 0.02–0.70 mg=L. Since the silver
concen-

trations were near the lower detection limit, there might be
uncer-
tainty in those measurements (Hach 2003). It is important to
note
that this does not mean that lower-reservoir biofilm buildup is
not
occurring, or that silver-impregnated toruses are infective
gener-
ally, but it does call for an improved torus design and further
research to pinpoint the source of the recontamination.
The mean log reduction values seen for the CWFs in this study
were 2.20 � 0.30 initially and declined to 1.34 � 0.65 and to
1.18 � 0.49. This makes these filters comparable to those
reported
previously by Brown (2007) who found log reduction values of
∼2.
A recent study that compared the effectiveness of chlorination
with a silver-coated porous ceramic candle element found a
mean
log reduction value of 1.21 for households provided
WaterGuard
(a dilute hypochlorite solution) while the ceramic candles
provided
a log reduction of only 0.91 (Albert et al. 2010). Likewise, the
73%
reduction in the number of households with detectable levels of
E. coli before and after chlorination for Period 2 was consistent
with a major meta-analysis that found an 80% reduction in the
pro-
portion of stored water samples with detectable E. coli (Arnold
and
Colford 2007) after chlorination interventions. Although the
per-
centage in this study declined to 58% during Period 3, the
Arnold

and Colford meta-analysis relied on studies that had a median
length of only 30 weeks. Finally, it is worth noting that if the
log
reduction values measured in the current study were 3 or better,
it could lead to improved outcomes (Mellor et al. 2012a; Enger
et al.
2012). However, given that the influent water frequently had
less
than 1,000 CFU=100 mL of bacterial contamination, the
reported
log reduction values might be higher if the influent water were
more
highly contaminated.
The use of silver in water treatment technologies is not without
risk, however, this risk is minimal and the WHO has not
established
a firm limit due to inadequate data (WHO 2011). Indeed, the
WHO
suggests that the only known risk for silver ingestion is argyria,
which is a condition that discolors the skin and hair. To prevent
this, the WHO recommends a lifetime limit of 10 g of silver.
Based
on this limit, the WHO recommends that silver concentrations
of
0.1 mg=L can be tolerated for 70 years without any health risk
(WHO 2011). The WHO limit for chlorine is 5 mg=L and there
are no specific adverse health effects that have been observed
(WHO 2011).
The current study had a number of limitations that warrant
discussion. First of all, the nonuniform influent water supplies
may
have biased some of the results. Secondly, the torus design did
not

generally increase the silver concentration in the lower
reservoir,
which is likely why it proved to be equally as effective at
removing
bacteria as the filter only design. This does not mean that the
torus
cannot be effective, or that biofilm layer buildup is not
occurring,
but it does mean that the torus needs to be redesigned to
increase
efficiency. That improved design should then be tested in future
laboratory and field trials. Also, systematic baseline
demographic
data about socioeconomic status or other covariates was not
con-
ducted. Such factors might have affected the use and
maintenance
of the three technologies. The fact that the participants using
chlo-
rine had to purchase chlorine bottles every six months, while
the
filter users did not have any additional costs, might have biased
the
results. However, the periodic purchase of chlorine is realistic
for
that method. Finally, the high drop-out rate might have biased
the
results and lowered their statistical power.
Conclusions
The first randomized trial has been conducted to study the rela-
tive microbial effectiveness of three different point-of-use water
CWF CWF+Torus SWS CWF CWF+Torus SWS

P
e
rc
e
n
t
(%
)
0
2
0
4
0
6
0
8
0
1
0
0 Period 2 Period 3
Risk Category (CFU/100ml)
<1 1-10 10-100 100-1000 >1000
Fig. 9. Bar plots showing the percent of samples with E. coli
concentrations in WHO risk categories at the two periods

D
ow
nl
oa
de
d
fr
om
a
sc
el
ib
ra
ry
.o
rg
b
y
N
or
th

er
n
A
ri
zo
na
U
ni
v
on
0
9/
09
/1
5.
C
op
yr
ig
ht
A
S
C
E

. F
or
p
er
so
na
l
us
e
on
ly
;
al
l
ri
gh
ts
r
es
er
ve
d.

treatment technologies in the developing-world community of
San Mateo Ixtatán, Guatemala. The POU technologies studied
were
chlorination (i.e., the safe water system advocated by the
Centers
for Disease Control and Prevention), CWFs, and CWFs with a
ceramic torus impregnated with silver placed in the lower filter
res-
ervoir. Surprisingly, the CWF þ torus design did not
significantly
increase silver concentrations in lower reservoirs as expected.
Furthermore, the percent of households in the chlorination
group
with adequate residual chlorination dropped from 100 to 65%
be-
tween Periods 1 and 2 of the study. Log removal efficiency was
highly variable and declined over the first six months with all
three
technologies, and there were no statistically significant
differences
seen between the three technologies in terms of microbial
removal
efficiency. These results highlight the need for further study
into
the causes of lower-reservoir contamination in CWFs and ways
to
remedy this problem. Improved silver-impregnated ceramic
toruses
might be an effective technology, but further research is needed
to
improve their design.
Acknowledgments
The authors thank Beth Neville Evans and the Ixtatán
Foundation

for assistance in participant enrollment and logistical field
support.
We would also like to thank Dr. Relana Pinkerton for her assis-
tance with some of our methods. This work was supported by
the National Science Foundation (CBET 651996). It was also
developed under STAR Fellowship Assistance Agreement no.
FP91728601 awarded by the U.S. Environmental Protection
Agency (USEPA). It has not been formally reviewed by USEPA.
The views expressed in this publication are solely those of the
authors, and USEPA does not endorse any products or
commercial
services mentioned in this publication.
References
Abebe, L. (2013). “Silver-impregnated ceramic-water filters to
im-
prove water quality and health.” Ph.D. thesis, Univ. of Virginia,
Charlottesville, VA.
Abebe, L., et al. (2014). “Ceramic water filters impregnated
with silver
nanoparticles as a point-of-use water-treatment intervention for
HIV-
positive individuals in Limpopo Province, South Africa: A pilot
study
of technological performance and human health benefits.” J.
Water
Health, 12(2), 288–300.
Albert, J., Luoto, J., and Levine, D. (2010). “End-user
preferences for and
performance of competing POU water treatment technologies
among
the rural poor of Kenya.” Environ. Sci. Technol., 44(12), 4426–
4432.

Arnold, B. F., and Colford, J. M. (2007). “Treating water with
chlorine
at point-of-use to improve water quality and reduce child
diarrhea in
developing countries: a systematic review and meta-analysis.”
Am. J.
Trop. Med. Hyg., 76(2), 354–364.
Bielefeldt, A. R., Kowalski, K., Schilling, C., Schreier, S.,
Kohler, A.,
and Scott Summers, R. (2010). “Removal of virus to protozoan
sized
particles in point-of-use ceramic water filters.” Water Res.,
44(5),
1482–1488.
Bielefeldt, A. R., Kowalski, K., and Summers, R. S. (2009).
“Bacterial
treatment effectiveness of point-of-use ceramic water filters.”
Water
Res., 43(14), 3559–3565.
Brown, J. M. (2007). “Effectiveness of ceramic filtration for
drinking
water treatment in Cambodia.” M.S. thesis, Univ. of North
Carolina,
Chapel Hill, NC.
Enger, K., Nelson, K., Clasen, T., Rose, J., and Eisenberg, J.
(2012).
“Linking quantitative microbial risk assessment and
epidemiological
data: Informing safe drinking water trials in developing
countries.”
Environ. Sci. Technol., 46(9), 5160–5167.

Guatemala. (2002). Health in the Americas, Vol. 2, Pan
American Health
Organization, Washington, DC, 306–322.
Hach. (2003). Silver colorimetric method: Method 8120
DR/4000 pro-
cedure, 11th Ed., Hach Company, Loveland, CO.
Hunter, P. (2009). “Household water treatment in developing
countries:
Comparing different intervention types using meta-regression.”
Envi-
ron. Sci. Technol., 43(23), 8991–8997.
Jagals, P., Jagals, C., and Bokako, T. (2003). “The effect of
container-
biofilm on the microbiological quality of water used from
plastic house-
hold containers.” J. Water Health, 1(3), 101–108.
Jung, W. K., Koo, H. C., Kim, K. W., Shin, S., Kim, S. H., and
Park, Y. H.
(2008). “Antibacterial activity and mechanism of action of the
silver
ion in Staphylococcus aureus and Escherichia coli.” Appl.
Environ.
Microbiol., 74(7), 2171–2178.
Kallman, E., Oyanedel-Craver, V., and Smith, J. (2011).
“Ceramic filters
impregnated with silver nanoparticles for point-of-use water
treatment
in rural Guatemala.” J. Environ. Eng., 10.1061/(ASCE)EE.1943-
7870
.0000330, 407–415.

Lantagne, D. S. (2008). “Sodium hypochlorite dosage for
household and
emergency water treatment (PDF).” J. Am. Water Works Assoc.,
100(8),
106–119.
Mellor, J., Abebe, L., Ehdaie, B., Dillingham, R., and Smith, J.
(2014).
“Modeling the sustainability of a ceramic water filter
intervention.”
Water Res., 49, 286–299.
Mellor, J. E., Smith, J. A., Learmonth, G. P., Netshandama, V.
O., and
Dillingham, R. A. (2012a). “Modeling the complexities of
water,
hygiene, and health in Limpopo Province, South Africa.”
Environ. Sci.
Technol., 46(24), 13512–13520.
Mellor, J. E., Smith, J. A., Samie, A., and Dillingham, R. A.
(2013). “Coli-
form sources and mechanisms for regrowth in household
drinking water
in Limpopo, South Africa.” J. Environ. Eng.,
10.1061/(ASCE)EE
.1943-7870.0000722, 1152–1161.
Mellor, J. E., Watkins, D., and Mihelcic, J. (2012b).“Rural
water usage
in east Africa: Does collection effort really impact basic
access?”
Waterlines, 31(3), 215–225.
Mintz, E. D., Reiff, F. M., and Tauxe, R. V. (1995). “Safe water

treatment
and storage in the home: A practical new strategy to prevent
waterborne
disease.” JAMA, 273(12), 948–953.
Onda, K., LoBuglio, J., and Bartram, J. (2012). “Global access
to safe
water: Accounting for water quality and the resulting impact on
MDG
progress.” Int. J. Environ. Res. Public Health, 9(3), 880–894.
Oyanedel-Craver, V. A., and Smith, J. A. (2008). “Sustainable
colloidal-
silver-impregnated ceramic filter for point-of-use water
treatment.”
Prüss-Ustün, A., et al. (2014). “Burden of disease from
inadequate water,
sanitation and hygiene in low-and middle-income settings: A
retrospec-
tive analysis of data from 145 countries.” Trop. Med. Int.
Health., 19(8),
894–905.
Rayner, J., Zhang, H., Schubert, J., Lennon, P. D. L., and
Oyanedel-Craver, V. (2013). “Laboratory investigation into the
effect
of silver application on the bacterial removal efficacy of filter
material
for use on locally produced ceramic water filters for household
drinking
water treatment.” ACS Sustainable Chem. Eng., 1(7), 738–745.
Ren, D., Colosi, L. M., and Smith, J. A. (2013). “Evaluating the
sus-

tainability of ceramic filters for point-of-use drinking water
treatment.”
Schmidt, W., and Cairncross, S. (2009). “Household water
treatment
in poor populations: Is there enough evidence for scaling up
now?”
Sondi, I., and Salopek-Sondi, B. (2004). “Silver nanoparticles
as antimi-
crobial agent: A case study on E. coli as a model for gram-
negative
bacteria.” J. Colloid Interface Sci., 275(1), 177–182.
WHO (World Health Organization). (2011). Guidelines for
drinking-water
quality, 4th Ed, Geneva.
Wright, J., Gundry, S., and Conroy, R. (2004). “Household
drinking water
in developing countries: A systematic review of microbiological
con-
tamination between source and point-of-use.” Trop. Med. Int.
Health,
9(1), 106–117.
D
ow
nl

l
us
e
on
ly
;
al
l
ri
gh
ts
r
es
er
ve
d.
http://dx.doi.org/10.1021/es1000566
http://dx.doi.org/10.1016/j.watres.2009.10.043
http://dx.doi.org/10.1021/es204381e
http://dx.doi.org/10.1128/AEM.02001-07
http://dx.doi.org/10.1128/AEM.02001-07

http://dx.doi.org/10.3362/1756-3488.2012.022
http://dx.doi.org/10.1001/jama.1995.03520360062040
http://dx.doi.org/10.3390/ijerph9030880
http://dx.doi.org/10.1021/es071268u
http://dx.doi.org/10.1021/es802232w
http://dx.doi.org/10.1016/j.jcis.2004.02.012
http://dx.doi.org/10.1046/j.1365-3156.2003.01160.x
http://dx.doi.org/10.1046/j.1365-3156.2003.01160.x
For this assignment, you will need to answer the questions
below. Type your responses where indicated -- and delete the
words in the square brackets: [ ]. In order to successfully
answer these questions, you'll need to:
· Start by navigating to http://azhin.org/nau/EGR-186-
Homework-Fall-2015. You'll need to read everything on that
page, and then follow the prompts for answering specific
questions below.
· Download the research article: Comparison of Three
Household Water Treatment Technologies in San Mateo Ixtatán,
Guatemala (provided on Bb Learn).
· Note: research articles are difficult to read unless you are an
expert, but as you look at this article, you can probably
understand enough to get a general idea of what it is about.

Many of the questions below relate to that article.
1. Notice that the article you are reading is divided into these
sections: Title, Abstract, Introduction, Methods, Results,
Discussion, Conclusion, and References. Most research articles
are divided into these same sections. The Introduction section
of an article will describe the research problem that the authors
investigated. Look over the Introduction and summarize the
research problem that the author's investigated:
[type your response here]
2. Notice that the authors cite a lot of other research articles in
the introduction, before they start describing the research that
they investigated. It is typical for authors to do this in the
Introduction section. Why do you think authors do this? What
purpose does it serve?
3. Which section of the article provides a short summary of the
entire article?
4. To find out what the authors learned from their research
project, you can skip to the Discussion and Conclusions
sections of the article. Take a look at those sections. Were the
authors successful in solving their research problem? Explain
your answer.
5. Why do you think the authors of this article took to the time
to write it and get it published?
6. Imagine that you are the owner of a private engineering
company and you want to manufacture and sell high-quality,
long-lasting, ceramic water filters to people at risk from using
unsafe water. Would reading this article be useful to you?
Explain why or why not.
7. Do you think this article is a reliable source of information?
Think about reasons why or why not. List at least two reasons in

any mix of categories below.
Reasons I think this article is likely to be reliable:
Reasons why it is hard for me to tell if this article is reliable:
Reasons why I think this article might not be reliable:
·
·
·
·
·
·
·
·
·
8. What type of publication is the citation below? (Pick from:
journal article, conference paper, book, or website.)
X. Zhu and X. Wu, “Class noise vs. attribute noise: A
quantitative study of their impacts,” Artificial Intelligence
Review, vol. 22, no. 3/4, pp. 177–210, Nov. 2004. doi:
10.1007/s10462-004-0751-8
9. Name the parts of this citation:
· X. Zhu and X. Wu: [type your response here]
· Class noise vs. attribute noise: A quantitative study of their
impacts: [type your response here]
· Artificial Intelligence Review: [type your response here]
· 22: [type your response here]
· 3/4: [type your response here]
· 177–210: [type your response here]
· Nov. 2004: [type your response here]
· 10.1007/s10462-004-0751-8: [type your response here]
10. Are the references listed at the end of the article you've
been reading (Comparison of Three Household Water Treatment
Technologies in San Mateo Ixtatán, Guatemala ) formatted in
the IEEE citation style? What clues helped you determine if
they are or are not?

11. What is the most comprehensive engineering database for
finding articles and conference papers?
12. Let's say you were just getting familiar with the topic of
ceramic water filters. In what order would you do the following:
Random Order:
Your Order:
· Look for books covering the topic
· Look for research articles on the topic
· Search the web for information on this topic
1.
2.
3.
Explain the order you chose:
Page 1

Comparison of Three Household Water TreatmentTechnologies in.docx

More Related Content

Similar to Comparison of Three Household Water TreatmentTechnologies in.docx (20)

More from maxinesmith73660 (20)

Recently uploaded (20)

Comparison of Three Household Water TreatmentTechnologies in.docx