Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Look up keyword
Like this
6Activity
0 of .
Results for:
No results containing your search query
P. 1
Bio Stability Analysis for Drinking Water

Bio Stability Analysis for Drinking Water

Ratings: (0)|Views: 110|Likes:
Published by bryesanggalang

More info:

Published by: bryesanggalang on Jan 19, 2010
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

08/29/2010

pdf

text

original

 
Available at www.sciencedirect.com
 journal homepage: www.elsevier.com/locate/watres
Biostability analysis for drinking waterdistribution systems
Soumya Srinivasan
Ã
, Gregory W. Harrington
Department of Civil and Environmental Engineering, University of Wisconsin-Madison, 1415 Engineering Drive, Madison, WI53705, USA
a r t i c l e i n f o
Article history:
Received 20 April 2006Received in revised form30 January 2007Accepted 6 February 2007Available online 3 April 2007
Keywords:
AocBiostabilityChlorinated systemsChloraminated systemsDrinking waterRegrowth
a b s t r a c t
The ability to limit regrowth in drinking water is referred to as biological stability anddepends on the concentration of disinfectant residual and on the concentration of substrate required for the growth of microorganisms. The biostability curve, based on thisfundamental concept of biological stability, is a graphical approach to study the twocompeting effects that determine bacterial regrowth in a distribution system: inactivationdue to the presence of a disinfectant, and growth due to the presence of a substrate.Biostability curves are a practical, system specific approach for addressing the problem of bacterial regrowth in distribution systems. This paper presents a standardized algorithmfor generating biostability curves and this will enable water utilities to incorporate thisapproach for their site-specific needs. Using data from pilot scale studies, it was found thatthis algorithm was applicable to control regrowth of HPC in chlorinated systems whereAOC is the growth limiting substrate, and growth of AOB in chloraminated systems, whereammonia is the growth limiting substrate.
&
2007 Elsevier Ltd. All rights reserved.
1. Introduction
Water quality deterioration due to regrowth of bacteria in thedistribution system is a practical problem that continues topose a challenge towater utilities (van der Kooij, 2003;Momba et al., 2000;LeChevallier et al., 1987;Brazos and O’Connor, 1996). Regrowth occurs when large numbers of bacteria areobserved in a distribution system that was supplied withwater containing only a few bacteria (van der Kooij, 2003;Hu et al., 1999). The ability to limit regrowth in drinking water isreferred to as biological stability (Rittmann and Snoeyink,1984) and mainly depends on the concentration of disin-fectant residual and on the concentration of substrate or foodrequired for the growth of microorganisms (Liu et al., 2002;Escobar et al., 2001;Skadsen, 1993). In Europe, biological treatment has been recognized as an important operationsince the early 1900s. It has been used for producinbiologically stable drinking water to prevent bacterial re-growth in water distribution systems and thus reduce thedose of disinfectant required to suppress regrowth (Kruithof,2001).Several researchers have found that the disinfectant canoxidize the natural organic matter (NOM) dissolved in water toproducemoresubstrateforheterotrophicbacteriaandnitrifying bacteria (Liu et al., 2002;Harrington et al., 2003). This makes the distribution system more vulnerable to bacterial regrowthbecause bacteria have substrate available from both sourcewater and from reactions between NOM and disinfectants.These studies imply that there is a trade off betweenmaintaining a high disinfectant residual and achieving a lowsubstrate concentration for controlling regrowth. Based onthis implication, bacterial regrowth can be controlled byunderstanding the biological stability in the distributionsystem, which is fundamentally governed by a complexinteraction between bacteria, their energy or food source,and the disinfectant. This paper presents a graphical
ARTICLE IN PRESS
0043-1354/$-see front matter
&
2007 Elsevier Ltd. All rights reserved.doi:10.1016/j.watres.2007.02.014
Ã
Corresponding author.
Tel.: +16082620848; fax: +16082625199.E-mail address:soumyas@cae.wisc.edu (S. Srinivasan).
WAT ER RESEARCH
41 (2007) 21272138
 
approach to study the two competing effects that determinebacterial regrowth in a distribution system: inactivation dueto the presence of a disinfectant, and growth due to thepresence of a food or energy source.
1.1. Biostability concepts for each species of bacteriawithin a distribution system
Assuming that bacterial growth can be described using Monod kinetics and that bacterial inactivation can bedescribed using Chick–Watson kinetics, Eq. (1) can be derivedfor each species of bacteria that is present in the distributionsystem (Harrington et al., 2002;Woolschlager et al., 2002; Fleming et al., 2005). Eq. (1) states that regrowth of abacterium can be prevented if the inactivation rate equalsor exceeds the growth rate of the bacterium at each locationwithin the distribution system. This equation can be simpli-fied to give Eq. (2) by replacing 
m
m
 /
k
d
with a single constant
R
.Fig. 1shows a hypothetical curve that was derived from Eq. (2)and this curve is called the biostability curve for thebacterium of interest. A system operated at any point abovethe curve will have a sufficient amount of disinfectantpresent to prevent regrowth of the bacterium of interest:
C
d
X
m
m
k
d
C
s
K
S
þ
C
s
, (1)
C
d
X
RC
s
K
S
þ
C
s
. (2)The significance of the curveparameters
R
and
K
S
is shown inFig. 1, where
R
is the asymptotic limit of 
C
d
when
C
s
b
K
S
, and
C
d
¼
0.5
R
when
C
s
¼
K
S
. An important observation about thebiostability curve is that systems operating at
C
d
X
R
can achievebiostability at all substrate concentrations. However, if thesystem is operated at lower substrate concentrations such as
C
s
p
K
S
, then a smaller disinfectant residual can be used. Morespecifically,
C
d
p
0.5
R
when
C
s
p
K
S
. At very low substrateconcentrations where
C
s
5
K
S
, the required disinfectant concen-tration becomesa linearfunctionofthesubstrateconcentrationand the ratio of substrate concentration to biocide concentra-tion is the controlling operational parameter (Harrington et al.,2002). Eq. (2) has been used to estimate values of 
R
and
K
S
thatachieve control of nitrification in chloraminated distributionsystems (Fleming et al., 2005).It is important to note that the values of 
R
and
K
S
arespecific to the conditions present in the distribution system.For example, values of 
R
and
K
S
will depend on temperature,pH, the disinfectant being used, pipe material, thepresence of multiple substrates, and the species of bacteria present in thesystem. As a result, values of 
R
and
K
S
are specific to a utility’sdistribution system.
1.2. Biostability concepts for two species of bacteria
As noted in the previous section, each bacterium will have acorresponding biostability curve defined by Eq. (2). Therefore,there will be multiple biostability curves for a distributionsystem in which multiple species of bacteria are present. Forsimplicity, assume that a given distribution system containstwo species of bacteria, namely
A
and
B
. Let
R
A
and
K
S,A
be thebiostability curve parameters for species
A
, and
R
B
and
K
S,B
bethe biostability curve parameters for species
B
. With twospeciesofbacteria,ninepossiblescenariosexistandthesecasesare shown inTable 1. One of these cases, when
R
A
¼
R
B
and
ARTICLE IN PRESS
Nomenclature
C
d
residual disinfectant concentration at a specificlocation in the system (ML
À
3
)
C
s
limiting substrate concentration at a specificlocation in the system (ML
À
3
)
m
m
maximum specific growth rate of bacteria (T
À
1
)
k
d
inactivation rate constant (L
3
T
À
1
M
À
1
)
R
ratio of maximum specific growth rate of bacteriato the inactivation rate (
m
m
 /
k
d
 )
(ML
À
3
)
K
S
half-maximum substrate concentration (ML
À
3
)
R
A
ratio of maximum specific growth rate to theinactivation rate of bacteria species
A
(ML
À
3
)
K
S,A
half-maximum substrate concentration of bacter-ia species
A
(ML
À
3
)
R
B
ratio of maximum specific growth rate to theinactivation rate of bacteria species
B
(ML
À
3
)
K
S,B
half-maximum substrate concentration of bacter-ia species
B
(ML
À
3
)
C
I
concentration of a regrowth indicator at a specificlocation in the system (ML
À
3
)
Regrowth
a boolean variable used to indicate whether aregrowth event of concern was present or absentbased on the
C
I
value (
Yes
 /
No
)
C
s
      C
      d
0.5 RR
 Asymptotic limit of biostability curve 
C
S
 
=
S
C
S
 
>>
S
Fig. 1 – Hypothetical biostability curve for a single species of  bacteria. ( 
C
d
is the residual disinfectant concentration at aspecific location in the system,
C
s
is the growth limitingsubstrate concentration,
R
is the ratio of maximum specificgrowth rate of bacteria to the inactivation rate ( 
l
m
 / 
k
d
 )
, and
K
S
is the half-maximum substrate concentration.)
WATER R ESEARCH
41 (2007) 2127– 2138
2128
 
K
S,A
o
K
S,B
, is illustrated inFig. 2. As shown in this figure, for0
o
C
s
o
N
, thevalue of 
C
d
needed to prevent growth of species
A
is greater than or equal to the value of 
C
d
needed to preventgrowth of species
B
. Therefore, the graph shows that species
A
is the most resistant of the two species to inactivation. If autility bases its distribution system operation on the biostabilitycurve for species
A
, then it has also adequately controlledspecies
B
and Eq. (3) is the control equation for both species of bacteria when
R
A
¼
R
B
and
K
S,A
o
K
S,B
.A similar analysis can be performed for the case when
R
A
4
R
B
and
K
S,A
o
K
S,B
(seeFig. 3). When
C
s
4
0, the value of 
C
d
needed to prevent growth of species
A
is greater than thevalue of 
C
d
needed to prevent growth of species
B
. As with theabove case,Fig. 3shows that species
A
is the most resistant of the two species to inactivation and Eq. (3) becomes thecontrol equation for this case. Similar results are obtainedwhen
R
A
4
R
B
and
K
S,A
¼
K
S,B
. Eq. (3) illustrates that
C
d
X
R
A
when
C
s
b
K
S,A
and
C
d
X
R
A
C
s
 /
K
S,A
when
C
s
5
K
S,A
:
C
d
X
R
A
C
s
K
S
;
A
þ
C
s
. (3)In other cases, species
B
becomes the controlling bacteriumfor biostability. As shown inTable 1, this occurs when
R
A
¼
R
B
and
K
S,A
4
K
S,B
, when
R
A
o
R
B
and
K
S,A
4
K
S,B
, and when
R
A
o
R
B
and
K
S,A
¼
K
S,B
. For these cases, basing system operation onEq. (4) will achieve biostability for both species
A
and
B
. Forthe case where both species of bacteria have the samebiostability curve (
R
A
¼
R
B
and
K
S,A
¼
K
S,B
), biostability canbe achieved by either Eqs. (3) or (4). Eq. (4) illustrates that
C
d
X
R
B
when
C
s
b
K
S,B
and
C
d
X
R
B
C
s
 /
K
S,B
when
C
s
5
K
S,B
:
C
d
X
R
B
C
s
K
S
;
B
þ
C
s
. (4)Biostability curvesfor species
A
and
B
areshown inFig. 4forthe casewhen
R
A
o
R
B
and
K
S,A
o
K
S,B
. For this case, species
A
ismore difficult to control at low
C
s
values while species
B
ismore difficult to control at high
C
s
values. From a utilityoperations perspective, it would be simpler to have a singlecurve that is always greater than or equal to either of the twocurves shown inFig. 4. Eq. (5) is an equation that meets thiscriterion. For this equation,
C
d
X
R
B
when
C
s
b
(
R
B
 /
R
A
)
K
S,A
and
C
d
X
R
A
C
s
 /
K
S,A
when
C
s
5
(
R
B
 /
R
A
)
K
S,A
:
C
d
X
R
B
C
s
ð
R
B
=
R
A
Þ
K
S
;
A
þ
C
s
. (5)
ARTICLE IN PRESS
Table 1 – The nine possible biostability scenarios for systems with two species of bacteria
a
Properties of bacteria Implication
R
A
4
R
B
and
K
S,A
4
K
S,B
Bacterium B requires higher
C
d
than bacterium Afor 0
o
C
s
o
R
B
K
S
;
A
À
R
A
K
S
;
B
R
A
À
R
B
:
Bacterium A requires higher
C
d
than bacterium Bfor
o
R
B
K
S
;
A
À
R
A
K
S
;
B
R
A
À
R
B
o
C
s
p
1
:
R
A
4
R
B
and
K
S,A
¼
K
S,B
Bacterium A requires higher
C
d
than bacterium B for 0
o
C
s
p
N
.
R
A
4
R
B
and
K
S,A
o
K
S,B
Bacterium A requires higher
C
d
than bacterium B for 0
o
C
s
p
N
.
R
A
¼
R
B
and
K
S,A
4
K
S,B
Bacterium B requires higher
C
d
than bacterium A for 0
o
C
s
o
N
.
R
A
¼
R
B
and
K
S,A
¼
K
S,B
Bacteria A and B require the same
C
d
for 0
p
C
s
p
N
.
R
A
¼
R
B
and
K
S,A
o
K
S,B
Bacterium A requires higher
C
d
than bacterium B for 0
o
C
s
o
N
.
R
A
o
R
B
and
K
S,A
4
K
S,B
Bacterium B requires higher
C
d
than bacterium A for 0
o
C
s
p
N
.
R
A
o
R
B
and
K
S,A
¼
K
S,B
Bacterium B requires higher
C
d
than bacterium A for 0
o
C
s
p
N
.
R
A
o
R
B
and
K
S,A
o
K
S,B
Bacterium A requires higher
C
d
than bacterium Bfor 0
o
C
s
o
R
B
K
S
;
A
À
R
A
K
S
;
B
R
A
À
R
B
.Bacterium B requires higher
C
d
than bacterium Afor
R
B
K
S
;
A
À
R
A
K
S
;
B
R
A
À
R
B
o
C
s
p
1
:
a
C
d
is the residual disinfectant concentration at a specific location in the system,
C
s
is the growth limiting substrate concentration,
R
A
and
R
B
is the ratio of maximum specific growth rate to the inactivation rate of bacteria species A and B,respectively,
K
S,A
and
K
S,B
is the half-maximumsubstrate concentration of bacteria species A and B, respectively.
Cs
      C      d
Biostability curve for Species ABiostability curve for Species B
R
 A
 
=
R
B
 
=
R
S
,
 A
S
,
B
Fig. 2 – Biostability curves for a system with two species of  bacteria when
R
A
¼
R
B
and
K
S,A
o
K
S,B
.
C
d
is the residualdisinfectant concentration at a specific location in thesystem,
C
s
is the growth limiting substrate concentration,
R
A
and
R
B
istheratioofmaximumspecificgrowthratetotheinactivation rate of bacteria species A and B, respectively,
K
S,A
and
K
S,B
is the half-maximum substrate concentrationof bacteria species A and B respectively.)
WATER RESEARCH
41 (2007) 2127– 2138
2129

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->