Innovative biofilm control strategies
by Alex Martin Bargmeyer
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in
Environmental Engineering
Montana State University
© Copyright by Alex Martin Bargmeyer (2003)
Abstract:
The drinking water industry is continually interested in finding new methods for preventing growth
and/or eliminating biofilms in the water delivery system. Concern for reducing biofilms in public water
distribution systems is due to problems such as undesirable taste and odor, corrosion of the
infrastructure, and the possibility for harboring pathogens. These drinking water biofilms may be
defined as a modular community of largely benign microorganisms attached to a surface within a
microbe-derived hydrated matrix. The biofilm’s basic structure and mode of growth enables resistance
to antimicrobial-based removal strategies. The objectives of this research were to screen novel
technologies or strategies in the laboratory using annular reactors and mixed population biofilms of
drinking water origin and to test the best technology for application in a realistic setting.
This research investigated three strategies or technologies to enhance the removal or the prevention of
growth of biofilm under drinking water conditions. The first was the use of specific chemical signalling
compounds implicated in previous research, to have the ability to cause detachment of an established
biofilm. The second, was the use of the bioelectric effect phenomenon in which the efficacy of
antibiotics, has been shown to be enhanced through the application of weak electric fields. The final
strategy in this investigation is a contact biocide which can be continually replenished by bulk fluid
chlorine in the system.
The results of the cell signaling compounds did not show a significant effect on an established biofilm.
The bioelectric effect proved to be corrosive to metal components in the reactor system and actually
provided liberated metal ions that were more conducive for bacterial growth. The final strategy proved
to be the best candidate for a drinking water system application. The contact biocide was able to
facilitate faster and greater removal of biofilm given typical drinking water chlorine disinfectant
concentrations. Experimental data suggests however, that low concentrations of residual chlorine in the
system would not be adequate to render the surface biocidal. INNOVATIVE BIOFILM CONTROL STRATEGIES
by
Alex Martin Bargmeyer
A thesis submitted in partial fulfillment
of the requirements for the degree
of
Master of Science
in
Environmental Engineering
MONTANA STATE UNIVERSITY
Bozeman, Montana
August 2003
ii
APPROVAL
of a thesis submitted by
Alex Martin Bargmeyer
This thesis has been read by each member of the thesis committee and has been
found to be satisfactory regarding content, English usage, format, citations, biblio­
graphic style, and consistency, and is ready for submission to the College of Graduate
Studies.
Dr. Anne K. Camper
fyTLyyu? /
(Signature)
Date
Approved for the Department of Environmental Engineering
Dr. Brett Gunnink
(Signature)
Date
Approved for the College of Graduate Studies
Dr. Bruce R. McLeodv^ A - ^ /z ^ /^
(Signature)
Date
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of the requirements for a master’s
degree at Montana State University, I agree th at the Library shall make it available
to borrowers under rules of the Library..
If I have indicated my intention to copyright this thesis by including a Copyright
notice page, copying is allowable only for scholarly purposes, consistent with “fair
use” as prescribed in the U. S. Copyright Law. Requests for permission for extended
quotation from or reproduction of this thesis in whole or in parts may be granted
only by the copyright holder.
Signature
Date
iv
TABLE OF CONTENTS
LIST OF TA B LE S.......................................................................................................
vi
LIST OF FIGU RES........'............................................................. ............................... viii
1. IN TRO D U CTIO N .................................................................................................
I
G o a l s .....................................................................................................................
2
2. LITERATURE REVIEW ....................................................................................
3
D rinking W a ter B io film s ..............................................................................
C ell S ignaling C om po unds ...........................................................................
B io electric E f f e c t .............................................
C ontact B io c id e s ...................................................................
D isinfectants.................................................-...................... ■...........................
Free Chlorine C hem istry.........................................................................
Mono chloramine chem istry....................................
S ummary ...............................
3
6
9
13
17
18
18
19
MATERIALS AND M ETH O D S.......................................
21
A nnular R e a c t o r s ............................... .: .........................................................
S imulation of D rinking W ater C o n d it io n s ..........................................
Dilution W ate r.................................................................................................
Humic Substances Solution P rep aratio n .....................................................
Nitrogen/Phosphorus Preparation..................................................... j......
C hemical T reatments P repa ra tio n ..........................................................
Cell Signaling Com pounds.............................................................................
C hlorine.................................................. ■....................i ..................................
Monochloramine ..............................................................................................
E xperim ental C o n d itio n s ..............................................
Signaling Compounds ........................................................ '...........................
Bioelectric................................................................................. •......................
Contact Biocide Surface - Laboratory..........................................................
N-halamine Polyurethane A ctivation....................................................
Biofilm R eduction....................................................................................
Biofilm Prevention......... ■.........................................................................
Chlorine Uptake By Coating ..................................................................
Contact Biocide Surface - Field.....................................................................
Southern Nevada Water A uthority........................................................
Seattle Public U tilities..........................................................■..................
21
23
23
24
26
26
26
27
28
29
29
31
36
36
36
38
39
41
41
44
3.
V
Sewerage & Water Board of New O rleans....................................'■......
Utility Reactor Operation........................................................................
B iological A nalysis ......................................................
Sampling Laboratory Annular Reactors.......................................................
Sampling Field Annular Reactors ............................... ' ...............................
■Heterotrophic Plate C ounts__ : ......................................... ..............•...........
Total Direct Counts ................. ......................................................................
D ata A nalysis ...................................................
45
46
47
47
48
49
49
51
4. RESULTS.................................................................................................................
53
S ignaling C o m po u n d s ..........................................................•...........................
B io electric E f f e c t .........................................................................................
C ontact B iocide S urface - L a bo ra to ry ................................................
Biofilm Reduction................................................................
Biofilm Prevention............ :............................................................................
Chlorine Uptake By Coating.........................................................................
C ontact B iocide S urface - F ield .............................................................
Southern Nevada Water A uthority...............................................................
Seattle Public Utilities ...................................................................................
Sewerage & Water Board of New Orleans ..................................................
Statistical Analysis - Field D a ta ................................
53
58
61
62
73
75
78
78
78
79
80
5. DISCUSSION ......................................
83
C ell S ignaling C om pounds ...............................
B io electric E f f e c t ........................................................................ . : .............
C ontact B iocide S u rfa ce ..............................................................................
Chlorine Uptake By Coating.........................................................................
Field S tudy...................................................................................: ..................
83
86
87
91
92
CONCLUSIONS ....................................................................................................
94
REFERENCES C IT E D ............................ ....................................................... : .........
97
6.
APPENDICES.......................................................................................... •................... 103
A P P E N D IX A - B O ZEM A N W A TER Q U A LITY D A T A ......... ■........ 104
A P P E N D IX B - C H L O R IN E U P T A K E D A T A ..................................... 108
A P P E N D IX C - N -H A LA M IN E PO L Y U R E T H A N E M S D S ............ 112
LIST OF TABLES
Table
Page
1. Problematic Microorganisms in Distribution Systems ..............................
5
2. Signaling molecule reactor configuration.....................................................
32
■3. Bioelectric reactor configurations..................................................................■ 35
4.
Contact biocide reactor configuration with different polycarbonate
slide compliments, I mg/L chlorine treatm ent............................................
37
Contact biocide reactor configuration with different polycarbonate
slide compliments, 4 mg/L chlorine treatm ent............................................
38
Contact biocide reactor configuration with different polycarbonate
slide compliments, low residual chlorine and monochloramine treatment
40
7. Average finished water characteristics at participating u tilities..............
42
5.
6.
8.
Average chlorine concentrations during the 1st biofilm reduction treat­
ment period, I m g/L target residual. Mean (Standard Deviation) ........
65
Average chlorine concentrations during the 2nd N-halamine biofilm
reduction treatm ent period, I mg/L target residual. Mean (Standard
D eviation).........................................................................................................
67
10. Two tailed t-test for surfaces comparing treated polycarbonate control
to the activated N-halamine surface biofilm HPC density (CFU/cm2) ..
70
9.
11. Average chlorine concentrations during the 5 day treatm ent period experiments I and 2, 4 mg/L target residual. Mean (Standard Deviation) 71
12. Average chlorine concentrations during biofilm prevention treatment
period. Mean (Standard D eviation).............................................................
73
13. Average monochloramine concentrations during biofilm prevention
treatm ent period. Mean (Standard D eviation)..........................................
75
14. N-halamine polyurethane surface uptake model sum m ary......................
77
vii
15. Two-tailed t-test for surfaces comparing polycarbonate control to the
activated N-halamine surface biofilm HPC density (CPU/cm2) ..............
82
16. Two-tailed t-test for the average reactor chlorine residual comparing
the polycarbonate control to the activated N-halamine surface.............
82
17. Bozeman water quality data for the year 2000 ......................................... 105
18. Bozeman water quality data for the year 2001 ......................................... 106 '
19. Bozeman water quality data for the year 2002 ......................................... 107
viii
LIST OF FIGURES
Figure
•
Page
1. Structures of cell signaling molecules...........................................................
8
2.
Structure of cyclic N-halamine polymerp o ly lC l..................
15
3.
Preparation of biocidal polyurethane coating.........................;..................
16
• 4.
Activation of biocidal polyurethane........... ................................................
17
5. Laboratory annular reactor model 1120
L S .............................................
22
6. Laboratory BAG treatm ent ap p aratu s........................................................
25
7.
30
General annular reactor configuration for all biotic experiments............
8. Modified annular reactor with DC electrodes for evaluation of the
bioelectric effect..........................
9.
33
Control and test reactor setup in Wet Training Room of. the Las Vegas
Valley Water District, Las Vegas, N e v ad a ..................................................
43
10. Control and test reactor setup at the Carrollton Water Plant, Sewer­
age & Water Board of New Orleans, Louisiana..........................................
46
11. Modified HPC drop streak m ethod..............................................................
50
12. Comparative planktonic HPC data for control and signaling molecule
FOdHSL tre a tm e n t............................
54
13. Comparative biofilm HPC data for control and signaling molecule
FOdHSL tre a tm e n t........................
55
14. Comparative planktonic HPC data for control and signaling molecule
wrs-I-51 tre a tm e n t...........................................................................................
55
15. Comparative biofilm HPC data for control and signaling molecule
wrs-I-51 tre a tm e n t...........................................................................................
56
16. Comparative planktonic HPC data for control and signaling molecule
BHL treatm ent.................................................................................................
56
17. Comparative biofilm HPC data for control and signaling molecule
BHL treatm ent.................................................................................................
57
18. Comparative planktonic HPC data for the indicated chlorine and cur­
rent treatments .....................................................................
59
19. Comparative biofilm HPC data for the indicated chlorine and current
treatm ents.........................................................................................................
60
20. 0.5 m g/L chlorine treatment and/or bioelectric effluent chlorine con­
centrations.........................................................................................................
61
21. Electrolytic generation of chlorine from reactor input water and feed ...
62
22. Biofilm reduction N-halamine experiment # 1 reactor influent and ef­
fluent free chlorine concentrations over the 5 week treatm ent p erio d __
63
23. Biofilm reduction N-halamine experiment # 1 reactor influent and ef­
fluent total chlorine concentrations over the 5 week treatm ent period...
64
24. Biofilm reduction experiment # 1 reactor planktonic HPC concentra­
tions over the 10 week experiment, I mg/L chlorine tre a tm e n t..............
65
25. Biofilm reduction experiment # 1 surface biofilm HPC densities over
the 10 week experiment, I mg/L chlorine treatm en t.................................
66
26. Biofilm reduction N-halamine experiment # 2 reactor planktonic HPC
CFU concentrations over the 10 week experiment, I mg/L target residual 68
27. Biofilm reduction N-halamine experiment # 2 biofilm HPC CPU den­
sities over the 10 week experiment, I m g/L target resid u al..... ...............
68
28. Picture of annular reactor slides comparing surface abnormalities be­
tween biofilm reduction experiment I and 2. (A) Polycarbonate, (B)
N-halamine coated polycarbonate (exp. I), (C) N-halamine coated
polycarbonate (exp.2) ...................................................................................
69
29. 4 m g/L biofilm reduction experiment reactors 1-3 biofilm HPC CFU
densities over the 10 day treatment p e rio d ................................................
72
30. 4 m g/L biofilm reduction experiment reactors 4-6 biofilm HPC CFU .
densities over the 10 day treatment p e rio d .................... ............. ______;
72
31. 0.2 m g/L (0-14 weeks) and 0.5 mg/L (14-24 weeks) chlorine treatment
biofilm prevention experiment, reactors 1-3 biofilm HPC densities over
the 24 week treatm ent period............................................. .........., ........... ■.
74
32. 0.2 mg/L (0-14 weeks) and 0.5 m g/L (14-24 weeks) monOchloramine
treatm ent biofilm prevention experiment, reactors 4-6 biofilm HPC
densities over the 24. week treatment p erio d ...............................................
76
33. Comparative biofilm density HPC data for uncoated polycarbonate
slides and N-Halamine coated slides at SNW A....................... .'.........: ___
79
34. Comparative biofilm density HPC data for uncoated polycarbonate
slides and N-Halamine coated slides at SP U ...............................................
80
35. Comparative biofilm density HPC data for Uncoated polycarbonate
slides and N-Halamine coated slides at SP U ...............................................
81
36. Theorized N-halamine activation of previously colonized surface biofilm
90
37. I mg/L Chlorine degradation data and regression curve......................... 109
38. 2 mg/L Chlorine degradation data and regression curve. .•...................... HO
39. 4 mg/L Chlorine degradation data and regression curve.......................... • 111
ABSTRACT
The drinking water industry is continually interested in finding new methods for
preventing growth and/or eliminating biofilms in the water delivery system. Concern
for reducing biofilms in public water distribution systems is due to problems such
as undesirable taste and odor, corrosion of the infrastructure, and the possibility for
harboring pathogens. These drinking water biofilms may be defined as a modular
community of largely benign microorganisms attached to a surface within a microbederived hydrated matrix. The biofilm’s basic structure and mode of growth enables
resistance to antimicrobial-based removal strategies. The objectives of this research
were to screen novel technologies or strategies in the laboratory using annular reactors
and mixed population biofilms of drinking water origin and to test the best technology
for application in a realistic setting.
This research investigated three strategies or technologies to enhance the removal
or the prevention of growth of biofilm under drinking water conditions. The first
was the use of specific chemical signalling compounds implicated in previous research,
to have the ability to cause detachment of an established biofilm. The second, was
the use of the bioelectric effect phenomenon in which the efficacy of antibiotics, has
been shown to be enhanced through the application of weak electric fields. The final
strategy in this investigation is a contact biocide which can be continually replenished
by bulk fluid chlorine in the system.
The results of the cell signaling compounds did not show a significant effect on an
established biofilm. The bioelectric effect proved to be corrosive to metal components
in the reactor system and actually provided liberated metal ions th a t were more con­
ducive for bacterial growth. The final strategy proved to be the best candidate for a
drinking water system application. The contact biocide was able to facilitate faster
and greater removal of biofilm given typical drinking water chlorine disinfectant con­
centrations. Experimental data suggests however, th at low concentrations of residual
chlorine in the system would not be adequate to render the surface biocidal.
I
CHAPTER I
INTRODUCTION
Drinking water distribution system biofilms are an inevitable fact of our infras­
tructure. These biofilms 1are for the most part benign and pose ho direct health
threat. However, there are a number of problems associated with these biofilms that
can arise if their growth is left unchecked. The problems th at are associated with
drinking water microorganisms and their biofilms are taste and odor, corrosion of
metal structures and the potential for harboring pathogens.
Taste and odor primarily relates to the general aesthetics of the water and can
be an indication of maintenance and operational problems of the treatment and/or
distribution system. Corrosion of the infrastructure can reduce the systems efficiency
and invoke costly repairs. Recently, it has been shown th at distribution systems
th a t have predominantly iron pipe have the greatest potential for biofilm regrowth. [1]
The persistence of pathogens in these biofilms presents the most important concern.
This is because biofilms are inherently resistant to disinfection by conventional means.
Harbored pathogens within these biofilms may then be released into the system posing
a possible public health threat. [2]
Current methods employed to combat nuisance biofilms include the use of disin­
fectants, reduction of organic material in the water, pipe materials th at reduce the
2
amount of biofilm accumulation and modification of plant operations to discourage
biofilm growth. In addition to these practices, other strategies or technology applied
to the management of distribution systems may result in greater biofilm reduction
and prevention or stalling regrowth events. Continuing research into the nature of
biofilm growth/development and emerging technologies to enhance biofilm control
may present additional strategies for this problem.
Goals
The goal of this research is to investigate and provide the drinking water industry
with novel biofilm control strategies. Recommendations for application of promising
technologies to distribution systems is not the intent of this study but rather to
provide an exploratory look at possible complementary solutions to current biofilm
control measures.
3
CHAPTER 2
LITERATURE REVIEW
Drinking Water Biofilms
Drinking water biofilms may be defined as a modular community of microorgan­
isms embedded within a microbe-derived hydrated matrix of extracellular polymers,
th at exists at an interface. This interface may be between the aqueous phase and
a solid support such as polycarbonate, polyvinyl chloride (PVC), ductile iron and
coated ductile iron pipe. It can also be between two different fluid phases such as air
and water, or between two liquids of differing densities like oil and water. [3]
Biofilms in drinking water are typically in environments th at do not provide op­
timum growth conditions, yet their persistence may lead to the problems mentioned
previously. A distribution system’s pipe network is the common interface for biofilm
growth. [4] Conditions within these systems have variations in temperature, disinfec­
tant residual, flow and chemical conditions such as pH, organic carbon, and nutrient
levels. These variations may produce biofilm sloughing events' which can increase the
number of planktonic bacteria. [1] Therefore distribution system biofilm maintenance
can be an arduous task. •
Distribution systems can be colonized by a number of microorganisms that cause
4
a variety of problems. Organisms such as Actinomycetes or fungi can result in taste'
and odor problems. [5, 6, 7] Bacteria may grow on ferrous metal surfaces [8] and
result in the presence of iron particulate in finished water. [9]' Corrosion of distribution
system pipe materials may be enhanced due to the presence of bacterial biofilms. [10]
Coliform bacteria present in treated water may be an indicator of the prior growth and
release of these organisms within distribution system biofilms. [11, 12, 13] Coliform
regrowth events are generally proclaimed to be a regulatory nuisance rather than a
hazard to public health. However, it has been noted that opportunistic pathogens,
including Aeromonas spp., Mycobacterium spp., and Legionella spp. can and do
grow in drinking water distribution system biofilms. These and other problematic
microorganisms associated with distribution system biofilms are summarized in Table
1[14],
The biofilm mode of growth enables resistance to antimicrobial-based removal
strategies. Previous research indicates that in order for disinfectants and antibiotics
to be as effective against biofilm bacteria, concentrations of 500 to 5000 times greater
than those required for killing planktonic strains of the same bacterial species are
required. [15] One of the mechanisms by which inherent resistance to antimicrobial
factors is mediated in biofilms is through very low metabolic levels and dramati­
cally downregulated rates of cell division of the deeply embedded microbes. [16] Thus
the strategies, for disinfecting agents th at depend upon robust and actively divid­
ing microbes are often ineffective. [17] Another is th at these structures demonstrate a
5
Table I. Problematic Microorganisms in Distribution Systems.
Type of Microorganism_________ ■Infrastructure or Water Quality Problem
Coliforms
Positive samples may be a violation of the To­
tal Coliform Rule.
Actinomycetes, Molds and Fungi
Produce earthy-musty-moldy taste and odor
compounds and are commonly found in sur­
face waters.
Iron Bacteria
Oxidize soluble iron to precipitate forms in­
creasing the mass of corrosion products on
pipe walls and pump casings. Excessive iron
deposits cause increased pipe friction and
lower pump efficiency.
Sulfate Reducing Bacteria
Reduces sulfate to hydrogen sulfide gas (rotten
egg odor) and increases corrosion rates.
Nitrifying Bacteria
Oxidize ammonia to nitrate and consume al­
kalinity, which may result in pH reduction.
Protozoans
May reside in biofilms posing a health risk
(Abernathy, 1998)
physical barrier to penetration of the antimicrobial agents. [18] It has been previously
shown th a t the polymeric matrix that encompasses the majority of biofilms retards the
inward diffusion of a number of antimicrobial agents. [19, 18, 20] Also, biocides com­
posed of common chlorine compounds used in water treatment such as hypochlorite,
chlorine dioxide,, mono chloramine as well as other antimicrobial / antifouling agents
may be deactivated in the outer layers of the biofilm faster than they can diffuse into
the lower layers. [21, 22] Lastly, a number of studies have shown th a t the gene ex­
pression within biofilms is altered due to the physical action of attachment. [23] This
6
change in gene expression is a biologically programmed response to attachment and
not due to nutrient deprivation. However, the link between antimicrobial resistance
and altered gene expression is presently being studied. [24]
Cell Signaling Compounds
Research in bacterial biofilms has just begun to yield information about the com­
plex nature of its survival strategies. One. of these strategies increasingly studied
since the 1960’s, is the mechanism of cell-to-cell signaling or quorum sensing. This
phenomenon was initially believed to be unique to bioluminescent marine bacteria
such as Vibrio fischeri and Vibrio harveyi. In these organisms the exhibition of bi­
oluminescence corresponds to high cell densities within high nutrient conditions in
the light emitting organs of their aquatic hosts. The specific mechanism responsible
for activation of bioluminescence was discovered to be the production and accumu­
lation of a signaling molecule or ” autoinducer” which is accumulated in response to
high bacterial cell densities. [25] Since then there have been numerous other bacte­
rial species identified to possess a quorum sensing system. [26] This mechanism for
cellular communication has been implicated as instrumental in the cycle of bacterial
attachment, biofilm formation/ maturation and detachment. [27, 28]
Quorum sensing compounds associated with specific detachment signals for biofilm
bacteria may be of value to the drinking water industry. If cells are chemically stimu­
lated to detach from the biofilm mode of growth, it could potentially make them more
7
susceptible to disinfection by residual disinfectants. Furthermore, if cellular response
to attachment and the formation of biofilm could be interrupted through the use of
cell signaling compounds, biofilm could be stalled or prevented.
A comprehensive review by Shirliff et al. [24] covers the current understanding
of molecular interactions within bacterial biofilms and the impact it has on biofilm
development and phenotype.- From the review there were three specific molecules
th a t displayed properties th at could applied to drinking water biofilms. These three
potential cell-to-cell signaling molecules were evaluated (Figure I. for structures)
for their ability to promote biofilm detachment from mixed species drinking water
biofilms.
The first of these molecules was 1,2 fluorodecal acyl homoserine lactone (FOdHSL),
which had previously demonstrated efficacy in the prevention of Pseudomonas aerug­
inosa biofilm development. [29] This is a synthetic molecule and acts as a’non-specific
antagonist to bacterial signaling molecules th a t have previously been implicated in the
formation of the fully mature biofilm phenotype. Therefore, by applying the bacterial
signal antagonist FOdHSL to the biofilm, the hope is to disrupt this mature phenotype
and promote biofilm detachment. [30, 31] Ideally, biofilm detachment would facilitate
flow-mediated removal, and microbial death in the presence of residual disinfectants
without the protective structure of the biofilm.
Another signaling molecule antagonist th a t was tested, wrs-I-51, was designed by
8
FOdHSL (1,2 fluorodecal acyl homoserine lactone)
wrs-l- 51
BHL (N-butyryl homoserine lactone)
Figure I. Structures of cell signaling molecules.
organic chemists at the Montana State University Chemistry Department as an irre­
versible inhibitor of the quorum sensing systems of microbes. These antagonists were
used at greater than 100 times the concentration of the bacterial signaling molecules
in biofilm environments. This molar excess of antagonist has been previously shown
9
to effectively inhibit quorum sensing in biofilms. [29]
The last signaling molecule that was evaluated was not an antagonist, but a na­
tive autinducing signaling molecule from the P. aeruginosa quorum sensing system,
butyryl homoserine lactone (BHL) [32, 33, 34]. BHL has also been implicated as a
more freely diffusible compound within a biofilm compared to the other HSL signaling
compounds. The physiological BHL concentration at which quorum sensing-activated
genes are upregulated is approximately 40 - 60 nM[35]. This signaling molecule has
been implicated as a possible'biofilm size limiting or detachment signal for P. 'aerugm osa[24]. Therefore, we tested the ability of BHL at 60 nM to promote detachment
from mixed population biofilms in annular reactors. In addition, we evaluated the
role of BHL to promote biofilm detachment at a concentration above (6000 nM) the
physiological BHL concentration due to the effectiveness of the first BHL experiment.
Bioelectric Effect
The problem of biofilm-mediated resistance to disinfectants may be circumvented
through the application of weak direct current fields. This “bioelectric effect” Was
initially discovered by Blenkinsopp et al.[36] and also by Costerton, et al.[37, 38] in
which the efficacy of antibiotics was shown to be increased through the application
Of weak electric fields. In one particular study, biofilms were grown on dialysis tub­
ing and immersed within a. minimal media filled chamber. These membranes Were
either left untreated, treated with a weak electric field alone, treated with antibiotic
•
10
alone, or treated with a combination of antibiotic and electric field. In order to avoid
the electrochemical generation of toxic products through electrolysis (discussed later),
biofilms were formed in minimal salts medium th at excluded chloride-containing com­
pounds. The biofilms th at were untreated, treated with a weak electric field alone,
or treated with antibiotics alone were not affected. However, when an electric field
was applied to the minimal salt medium containing antibiotics, the researchers found
a dramatic effect on biofilm structure and bacterial concentration. It has also been
found th a t the concentrations of antibiotics needed to be effective against biofilm bac­
teria in a weak electric field fell to only 1.5 - 4.0 times those necessary for planktonic
bacteria. [39]
Another study conducted by Stewart et al. [40] found similar results of enhanced
efficacy of applied antibiotics when coupled with a weak DC electric current. These
experiments utilized P. aeruginosa biofilms grown on a polycarbonate substratum.
When the biofilm was exposed to the antibiotic tobramycin the effective reduction
in biofilm was 2.88 logs which was increased to 5.58 logs in the presence of the weak
electric field. Because these experiments require an electric potential (not indicated)
and current flow (2 mA) they realized th at the potential electrolysis of water can
create species such -as oxygen and oxygen intermediates. These electrolysis reactions
are shown in equations 2.1 through 2.3. Other electrolytically-generated oxygen
compounds possible are super oxide anion, peroxide and hydroxyl radicals.
11
2H20
O2 + AH+ + 4e™
'
AH2O + 4e“ =» 2H2 + 4 0 f T
(2.1)
(2.2)
The overall net reaction becomes:
2H20 =v- O2 + H 2
(2.3)
Oxygen did enhance the ability of tobramycin to reduce the biofilm by 4.68 logs.
However, biofilm reduction was enhanced further by the addition of the electric cur­
rent (5.58 logs), which implies the claimed bioelectric effect did have a synergistic
effect on antibiotic effectiveness in this study.
Electrolysis or iontophoresis of other constituents when applying an electric po­
tential to an electrolyte are also possible. It was found th at the application of small
electric currents in solutions that contain chloride ions can create chlorine gas shown
in equation 2.4 .[41]
2NaCl + 2H20 => 2 N a O H .+ Cl2 + H 2
(2.4)
Chlorine gas quickly dissolves in water and forms hypochlorous acid and hypochlorite.
These observations lead to the conclusion th at in situations where there is a significant
number of chloride ions present, antimicrobial efficacy is due to the iontophoretically
produced chlorine based substances and not a synergistic bioelectric effect. [41]
Electric fields in direct contact with biofilm have been observed to dramatically
change the structure of the biofilm. Biofilms grown on an electrode wire where the
12
electrical current could pass directly through the bacterial structure displayed syn­
chronous expansion and contraction of the biofilm thickness with alternating current.
This has been attributed possibly to the localized change in pH due to the change in
current flow and charge reversal and also charged groups within the biofilm and the
S
charge of the electrode. It was also thought th a t this induced structural “effect” may
present a possible mechanism for the increasing the efficacy of biocides when com­
bined with an electric field. Prom this research conjecture for the bioelectric effect’s
usefulness to reduce biofilms in industry has been made; but only if growing on a
conductive surface in an aqueous environment and a counter electrode. [42]
Continued research by McLeod et al. in the bioelectric effect created a dose
response curve for the effectiveness of applied electric, current and antibiotic. [43] A
current density within their system of 360 /jlA / cm2 yielded consistent effectiveness for
5 times the minimum inhibitory concentration or MIC of tobramycin at 5 mg/L. These
findings served as a starting point for drinking water biofilm experiments utilizing this
strategy.
In review of the bioelectric effect, a number of possible explanations for the en­
hancement of antimicrobial efficacy have been proposed. These include electrophoreticallymediated augmentation of antimicrobial transport, cell membrane permeabilization,
electrolytic generation of oxygen and potentiating oxidants, increased convective
transport of antimicrobials due to contraction and expansion of the biofilm, and
pH alterations in solutions to which current is applied. This portion of the study
13
evaluated the phenomenon of the bioelectric effect as a strategy to enhance the an­
timicrobial properties of chlorine against mixed population biofilms of drinking water
origin.
Contact Biocides
The most promising novel strategy or technology for the reduction of biofilm
within distribution systems is the contact biocide. Traditionally, reduction of bacte­
rial growth in a water system is achieved through maintaining a residual concentration
of disinfectant in the system. These disinfectants, mainly chlorine or monochloramine
must then diffuse through the biofilm to effectively kill the cells. As mentioned pre­
viously, biofilms are resistant to disinfection, which is often due to the resulting mass
transfer limitations of the disinfectant.[44] The effectiveness of typical chlorinated
drinking water may be enhanced if the surface were to be inhospitable to the pro­
liferation of microbial growth. A contact biocide which has been made available to
this study comes from Vanson-HaloSource Inc. and is based on a group of organic
compounds called N-halamines. These N-halamines and their derivatives' have the
ability to bind oxidative/ reactive chlorine, thereby rendering a biocidal property.
A study of the comparative antimicrobial activities of these N-halamine deriva­
tives was first published in 1976 by Kaminski et al.[45] The comparison evaluated
a number of water soluble N-halamine compounds for their bacterial inactivation
efficiencies. It was determined that these compounds could provide an alternative
14
stabilization method of the active Cl+ ion and their subsequent use as a disinfectant.
More recently the application of this technology for the disinfection of water
became apparent. [46] The compounds used in this study are similar to the N-halamine
compounds described above but contain a heterocyclic carbon nitrogen or oxygen ring
structure which has limited solubility in water. Again these compounds have a unique
property th at stabilizes the active chlorine or bromine species th a t makes it useful as
a soluble disinfectant. In this study some of the chlorinated compounds tested were
able to remain stable as an effective disinfectant over a period of 15 weeks.
Later development of insoluble polymeric N-halamine compounds suggested an
application as a biocidal water filter. [47] This polymer has the same heterocyclic
ring structure as previously described and is a derivative of a polystyrene hydantoin
(polylCl, Figure 2). The chlorinated structure exists in solid form as pale yellow
granules. In these filter experiments the solid granules were loaded into columns and
challenged with water containing four bacterial species with concentrations greater
than IO6 colony forming units (CPUs) per mL. The filter’s capacity for complete dis­
infection was about 160 mL contaminated water per gram of polylCl for E. coli at
contact times of 1-2 s/mL.[48] The disinfectant properties of this filter could be re­
newed when flushed with a strong bleach solution to clean and reactivate the polylCl.
Other work in this area included coating various substrates with a poly-N-halamine
compound.[49] Substrates used in these experiments were glass, plastic, cotton and
a cotton polyester blend. The hard surfaces coated with the polymer achieved 6-log
15
(
H2C ------ CH
)
X'
O
C-------- N
X1X1= H 1CI1Br
Figure 2. Structure of cyclic N-halamine polymer polylCl.
inactivation of S. aureus in 5-10 minutes. The textiles were tested for their zone of
inhibition which was 0.5 mm for the cotton and 0.1mm for the cotton blend. Once
again the unique properties of the poly N-halamine compounds is the ability to retain
a stabile active N-Cl moiety until there is a collision or hit by a reactant i.e. bacterial
cells, organic compounds, etc. The active chlorine is then used or donated from the
structure. The antimicrobial properties can be regenerated by repeated exposure to
Cl2 or NaOCl.
A recently developed coating that incorporates an N-halamine hydantoin monomer
into the backbone of a polyurethane polymer has promise for testing in drinking wa­
ter surface applications. This particular coating is prepared by copolymerizing a
16
X
CH2CH2OH
CH1— N
\
CH2CH2OH
+
OCN ....................NCO
Hydantoin modified
Diol Monomer
Polyisocyanate
Figure 3. Preparation of biocidal polyurethane coating.
previously disinfection-proven, water-soluble, N-halamine derivative with commer­
cially available polyurethane. The N-halamine used to create a diol monomer used in
copolymerization is a l,3-dihalo-5,5-dimethylhydantoin combined with diethanolamine
and a methanol formaldehyde solution. The modified diol monomer is then combined
with polyisocyanate (Figure 3) to create a surface active polyurethane . The coating
is cured at room temperature and activated by exposure to a strong bleach solution
(10% by volume) for 3-12 hours (Figure 4).[50]
The chemical structure of the N-halamine hydantoin is the key to the superior
stability of the active nitrogen-chlorine bond. The electron donating alkyl groups
adjacent to the nitrogen-chlorine moieties prevent significant release of chlorine into
an aqueous solution. It has been shown that the coating is able to retain its biocidal
17
Polyurethane
o
Activated Polyurethane
O
O
Il
Il
o
Il
Il
.....HNCO
.OCNH....
HNCO
OCNH-
N
+
Bleach
Figure 4. Activation of biocidal polyurethane.
properties for up to 6 months. [50]
Disinfectants
While traditional water treatment may employ a variety of disinfectants to re­
duce or eliminate the growth of microorganisms in drinking water, only secondary
disinfectants th at are used to produce a residual will be be considered here. Once
biofilm has detached from a pipe surface it becomes more susceptible to disinfection
by these residual disinfectants in the system. Free chlorine and monochloramine are
the only practical types for this application. Chlorine dioxide is also an active chlo­
rine species that is used in drinking water applications. However this disinfectant was
18
not considered due to it’s potentially explosive properties and time allowed for the
experiments.
Free Chlorine Chemistry.
Chlorine is the most common secondary disinfectant used in distribution sys­
tems today. It is added in sufficient quantities to create a residual in the system to
discourage microbial growth and prevent recontamination from system maintenance,
backflow and break events. It can be added as a gas (CI2 ), liquid sodium hypochlo­
rite (NaOCl), or calcium hypochlorite (Ca(OCl)Z) in solid form. Once in water, the
chemistry is basically similar for all, creating hypochlorous acid and hypochlorite.
The concentration of dissociated species is dependant upon the pH of the Water (see
equation 2.5) where hypochlorous acid is the more active disinfectant.
# #+ + ocz-
(par* = 7.6)
-
(2.5)
Typical concentrations of available free chlorine are maintained between 0.20 4.0 mg/1 depending on existing organic carbon concentration in the finished water.
Evaluation of the efficacy of free chlorine coupled with the bioelectric and coating
strategies presented here used chlorine doses within this range.
Monochloramine chemistry.
Monochloramine is becoming more widely used as a secondary disinfectant. This
is mainly because it does not create as many disinfectant byproducts as free chlorine.
19
It does require higher concentrations and longer contact time to obtain the same
disinfection efficacy. However it does have the capacity to maintain a residual in
the system longer than chlorine. Monochloramine also does not leave a detectable
residual taste and odor which is common with chlorine. In these experiments ammonia
is mixed with chlorine to create monochloramine shown in equation 2.6.
N H t + HO Cl => N H 2Cl + H 2O + H +
(2.6)
Monochloramine was used in the later experiments testing the contact biocide. It was
not known by previous research of the N-halamine compounds whether they could be
effectively charged by chloramines.
Summary
The use of signalling compounds to prevent the attachment of cells to form biofilm
or promote detachment could be a useful maintenance tool for utilities. Providing the
necessary information about the toxicity or health concerns of adding these chemicals
to drinking water was not considered in this study. The Environmental Protection
Agencies approval would be the final step for the application of such treatments.
Using the bioelectric effect as a control measure for biofilm growth has a host of
engineering problems associated with it. Delivering a uniform current flux throughout
a pipe network presents a very significant challenge. This strategy may be of use
in this type of application but only if the system is free of metals susceptible to
electrochemical corrosion. Anodic protection of ductile iron pipes is one possible
20avenue of the application of the bioelectric effect, but may not produce the current
needed to be effective. The use of alternating current has had little attention in this
area but may present a mechanical means of enhanced biocide delivery or diffusion
into biofilm [42] and should not be overlooked for future experimentation in this area.
Based on work in the laboratory with the bioelectric effect, this strategy did not
demonstrate significant efficacy for application to distribution systems, but may be
useful in other areas. '
The coating of drinking water distribution system pipe networks is used as a strat­
egy to reduce the amount of corrosion and prolong system integrity. If the N-halamine
technology can be combined with these coatings it could provide added protection to
the system. From a practical standpoint, implementation of this technology to cur­
rent coatings would be relatively simple. Adding the N-halamine compounds could
easily become part of the mixing process of polymer coatings.
Furthermore, the
cost of adding this technology to current coatings would be low given that these Nhalamines are inexpensive to create. The N-halamine contact biocide has shown very
good efficacy in the application of vinyl pool liners against bacteria and algae growth
(Jeff Williams, personal communication). While pools generally operate with higher
chlorine concentrations than drinking water, it is evident that the technology needs
attention to enhancing the rate of activation of the contact biocide. A high rate of
chlorine uptake at the surface in the presence of a low residual chlorine concentration
would present a viable solution to the problem herein.
21
CHAPTER 3
MATERIALS AND METHODS
Annular Reactors
To model drinking water conditions annular reactors manufactured by Biosurface
Technology Corporation in Bozeman, Montana, were used for laboratory biotic,, ki­
netic abiotic and field study experiments (see Figure 5). The annular reactors used in
these experiments consisted of a glass outer cylinder and an inner rotating cylinder
driven by a variable speed motor. The inner cylinder holds 20 flush-mounted slides
th at can be removed for sampling biofilm. Each reactor contains inlet ports for di­
lution water, substrate/nutrients and addition of disinfectants. Outflow leaves the
bottom of the reactor via a standpipe and vent system to maintain the correct water
level in the reactor. The liquid space within the reactor has been found by others
to be completely mixed, and therefore the system model for growth kinetics in the
reactor is a continuous flow stirred tank reactor (CFSTR). The fluid in the annular
space inside the reactor is I liter and for these experiments a total inflow of 8.33 mL/s
was used to achieve a 120 minute hydraulic residence time.
The rotational speed of the inner cylinder was set at 60 revolutions per minute.
This correlates to a hydraulic shear stress at the sample slides’ surface equivalent
22
Continuous Flow (Once Through)
Laboratory Regrowth Monitor
System Schematic
Pertstaltic Pum p
Dilution
Vteter
R ow
Pertstaltic P u m p
Breaks O)
A it Vent
-fr *
icfc
Slide Removal
Port
Knurled Nuts (4)
Nutrient
Solution
Motor
R eactor
Slides (20)
Threaded Rod (4)
Inner Cylinder
Draft Mixing
Tubes (4)
Figure 5. Laboratory annular reactor model 1120 LS.
Air Vent
23
to I foot per second flow velocity in a 4-inch diameter pipe. The inner cylinder
has 4 vertical offset draft tubes to assist in mixing of the reactor fluid. Substrate,
nutrients and dilution water were pumped to each reactor using peristaltic pumps
and Norprene®/silicone tubing (Cole-Parmer, Masterflex, Vernon Hills, IL).
Simulation of Drinking Water Conditions
To simulate drinking water for the laboratory experiments, Bozeman tap water
was treated and used for the bulk dilution, water. The dilution water also served
to provide a continuous source of indigenous, mixed species population of organisms
acclimated to Bozeman tap water. Soil-derived humic substances served as the pri­
mary organic carbon supplement or substrate for the mixed species drinking water
biofilms.
Nitrogen, from potassium nitrate, and phosphorus, from dibasic potas­
sium phosphate, were added to the humic substances feed solution in stoichiometric
amounts to ensure th at carbon would be the limiting growth factor in the reactors.
The carbon:nitrogen:phosphorus concentration ratio used in these experiments was
100:10:1. All laboratory experiments received feed concentrations th at resulted in a
final reactor inflow concentration of 1.5 mg/1 total organic carbon (TOC), 0.15 mg/L
nitrogen and 0.015 m g/L phosphorus.
Dilution Water
-
Bozeman tap water was pumped through two parallel upflow granular activated
24
carbon (GAC) columns to remove residual chlorine. The dechlorinated water then
passed though another column which contained a biologically activated carbon (BAG)
media originally "from a water treatment facility in Laval, Quebec. This media has
acclimated to Bozeman tab water for a period greater than 6 years. The BAG col­
umn served to remove easily degradable organic carbon from the water and provide
the inoculation of drinking water microorganisms to the reactors. A similar BAG
filter system operated at the CBE measured 1014±394/ug dissolved organic carbon
(DOC) and 32±38)Ug biodegradable dissolved organic carbon (BDOC) in a previous
study[51]. The BDOC level of the dilution water from these systems is very low,
so th at any carbon added to this water will dominate. The laboratory scale filter
treatm ent system is illustrated in figure 6. Water quality data for the Bozeman’s tap
water from the start of the experiments is shown in Appendix A. The dilution water
was periodically checked to assure there was no chlorine in the effluent of the two
GAC columns. The temperature of the dilution water system was passively warmed
to laboratory temperature (21-24°C) with a pH range of 6.8-9.1 throughout the ex­
perimental period.. Continuous feed of disinfectant solutions were added directly to
the dilution water 0.6 m upstream of the reactors to allow for mixing and dilution. All
flowrates were adjusted accordingly to maintain a hydraulic residence time of «120
min.
25
Untreated Bozeman
tap water reservoir
Treated dilution water
reservoir for
distribution to reactors
Peristaltic pump
Figure 6. Laboratory BAC treatment apparatus.
Humic Substances Solution Preparation
Humic substances are high molecular weight, largely unidentifiable, polymeric,
organic compounds commonly found in soil and were used as the substrate for biofilm
growth in these experiments.
Bulk Elliot Silt Loam Soil for extraction of humic
substances was obtained from the International Humic Substances Society in St.
Paul, MN. This material has been utilized in previous studies at the Center for Biofilm
Engineering in similar drinking water studies.
A stock solution of humic substances was prepared by adding 120 g of soil to I
L of sterile 0.1 N NaOH, and mixing vigorously for at least 48 hours. This mixture
26
was then centrifuged at 10,OOOxg at 4°C for 20 minutes to separate the Soil particles.
TOC analysis of the humic stock solution was performed on a Dohrman DC-80 TOC
analyzer with an average concentration of approximately 1300 mg/L. The stock so­
lution was stored in the dark at 4°C. To prepare the feed to the reactors, the stock
solution was diluted in sterile reverse osmosis (RO) water to a final concentration
th at resulted in the TOC level indicated previously (1.5 mg/L). The pH of the feed
solution is adjusted from «10 to 7.204=0.05 pH units using pure hydrochloric acid
before connection to the system. •
Nitrogen/Phosphorus Preparation
The nitrogen/phosphorus stock solution was prepared by dissolving potassium
nitrate and dibasic potassium phosphate into NANOpure Diamond Ultrapure Water
(Barnstead/Thermolyne, Dubuque, Iowa). The stock solution concentrate was deter­
mined so th at 0.1 mL of stock nitrogen/ phosphorus solution added, to I L of humic
feed solution would result in the desired ratio of 100C:10N:1P stated previously. The
solution was then filter sterilized with a 0.2 fim syringe filter into a sterile bottle and
stored in the dark at 4°C.
Chemical Treatments Preparation .
Cell Signaling Compounds
The previously described signaling molecules FOdHSL and wrs-I-51 which are
non-specific antagonist to biofilm formation, and BHL, an analog signaling molecule
27
to promote detachment were individually diluted in 95% ethanol. The final annular re­
actor concentrations of both the antagonist signaling molecules FOdHSL and wrs-I-51
(10 yuM) was 100 times the concentration of signaling molecule levels found in biofilm
environments. A BHL concentration of OOnM is equal to the upper limit in which
quorum sensing-activated genes are upregulated. Therefore, OOnM and 0000nM(100x
the physiological level) concentrations were prepared in 95% ethanol. Dilutions of the
concentrates were chosen so the addition of 200 pL of the ethanol-diluted solution
produced a final concentration of lO/zM, lO/rM, GOnM and GOOOnM respectively in the
annular reactor. Controls for these treatm ent experiments consisted of the addition
of 95% ethanol (200 /rL) alone. Signaling molecule treatment solutions were prepared
immediately before treatm ent and sampling of the reactors.
Chlorine
All chlorine solutions used for cleaning and treatments were derived from house­
hold bleach. The bleach concentrate ranged from 5.25-7.20% by weight as sodium
hypochlorite (NaOCl). For each batch of bleach solution the concentration was
checked and then diluted accordingly to the appropriate level to obtain.the desired
concentration for treatm ent within the reactors. All treatment, feed containers were
chlorine demand free Pyrex glassware and were mixed using chlorine demand free
NANOpure water prior to use. Chlorine levels were checked daily for the treatment
28
feed container, reactor influent and reactor effluent using a DPD colorimetric chlo­
rine method and digital chlorine colorimeter (Model DC1100, LaMotte Company,Chestertown, Maryland). Minor adjustments to the flowrate and feed concentrations
were made in order to maintain constant residuals in each reactor. The chlorine feed
container, tubing and reactors were covered to prevent UV light degradation.
Monochloramine
Mono chloramine treatm ent solutions were prepared using two separate techniques.
The first protocol used 5.5 mg/L of dibasic potassium phosphate in NANOpure water
as a buffer. This solution was adjusted to a pH of 8.9-9.2 using sodium hydroxide.
Lastly, enough ammonium chloride was added to achieve a final Cl2:N mass ratio
of 4:1 in the concentrate. This amount of ammonium chloride was back-calculated
assuming a 100% conversion from chlorine to monochloramine and a desired concen­
tration in the feed of 8 mg/L. Diluted sodium hypochlorite was dripped slowly into
this solution at a rate greater than 20/iL per 6 seconds. This technique had vary­
ing success at producing consistent concentrations of monochloramine and another
method was proposed based upon similar practice in the drinking water industry.
Instead of slowly adding sodium hypochlorite to a solution of ammonium chloride,
separate streams of buffered ammonium chloride and sodium hypochlorite solution
were combined simultaneously. Each solution was buffered with 5.5 mg/L dibasic
potassium phosphate and the pH adjusted to 8.25-8.35. The concentration ratios of
29
ammonium chloride to sodium hypochlorite were.the same as in the first method. The
two solutions were pumped at the same flow rate using peristaltic pumps into a 1.5
m length mixing tube allowing sufficient time (%1 minute) for the chlorine-ammonia
reaction to occur. According to the EPA, 99% of this conversion occurs in just 0.069
seconds at a pH of 8.3 [52]. This method produced a much more stable and reliable
treatm ent solution and was used for the duration of the monochloramine treatment
experiment. Monochloramine concentrations were measured using the same method
as for free chlorine where the difference between total chlorine and free chlorine mea­
surements were assumed to be the monochloramine concentration.
Experimental Conditions
The general reactor configuration remained the same throughout the study (Fig­
ure 7). Specific treatments or arrangements of reactors is described in the following.
Signaling Compounds
These experimental conditions were chosen to reflect field conditions in which a
treatm ent solution would be applied to a target water system as a slug dose. This dose
of signaling molecules would be applied to a segment of the distribution system that
was contaminated and allowed to diffuse and interact with the biofilm. The treated
segment would then be flushed to remove the detached microbes and returned to
operation. Simulation of this scenario was conducted in the laboratory.
TreatmentFeed
Humic substances
& nutrient feed
Dilution
W ater —
I
Control
Treatm ent I
Figure 7. General annular reactor configuration for all biotic experiments.
Treatment 2
Treatm ents
I
31
The reactor configuration for testing the signaling compounds is outlined in Table
2. Reactors were allowed to reach steady state prior to addition of the compounds.
Steady state was the point in annular reactor biofilm growth where planktonic con­
centrations stabilized for five consecutive days at approximately 1.0 x IO5 — 1.0 x IO6
colony forming units (CFU)/m L and biofilm concentrations reached approximately
1.0 x IO7 — 1.0 x IO8 (C F U )/cm 2. This usually occurred within 5 weeks of operation.
One of the three signaling molecules was then added to the mixed population biofilm
annular reactor. Following the addition of ethanol alone (Control) or ethanol diluted
signaling molecule (FOdHSL, wrsT-51, or BEL), the reactor was allowed to rotate
for 2 minutes (60 rpm) to promote mixing, then infeed and rotation in the reactor
was paused for two hours. Infeed was then reconnected and reactors were allowed
to rotate for 24 hours. Four biofilm samples for each time point were taken (via
removable polycarbonate slides (15 cm2)) at 0, 2, 3, 6, and 26 hours after addition
of the signaling molecule .and planktonic samples were obtained in triplicate at 0, 2,
2.5, 2.75, 3, 6, and 26 hours post-addition.
Bioelectric
In previous bioelectric effect experiments reviewed in Chapter 2, the electrodes
used were typically composed of platinum'. However, the extremely high expense of
platinum makes the use of these electrodes financially prohibitive if they are to be
considered for general water system treatment. Therefore, stainless steel was chosen
32
Table 2. Signaling molecule reactor configuration.
Reactor #
Description of conditions
1
Control-no treatment (200 /rL 95% ethanol only)
2
Treated with FOdHSL (200 /rL 95% ethanol 10 /rM final concen­
tration)
3
Treated with wrs-I-51 (200 jjL 95% ethanol 10 yuM final concen­
tration)
4
.
Treated with BHL (200 /iL 95% ethanol 60 nM final concentration)'
as the electrode material. The annular reactor system was modified for the delivery
of a clamped, direct current and allow for periodic removal of electrodes for corrosion
monitoring. The completed reactor modifications in which electrodes (+ and -) were
placed against the walls of the reactor at opposite sides of the inner rotating cylinder
as shown in Figure 8. Electrodes were connected to a power supply th at provided
a constant current of 15.7 mA (70 - 80 V) current flux of 3.7 fj,A/m m 2 through the
reactor as determined by a multi-test ammeter connected in series. This current flux
was previously shown to be effective in producing a significant bioelectric effect as
noted in the Chapter 2.
While previous experiments evaluating the bioelectric effect have utilized chloridefree media, this system has measurable levels of chloride ions like standard water de­
livery systems (also contained in Appendix A). The chloride ions in this system are
derived from the tap water th a t is combined with the pH-adjusted humic concentrate.
33
Figure 8. Modified annular reactor with DC electrodes for evaluation of the bioelectric
effect.
34
Also, the base-extracted humic substance concentrate (0.1M NaOH) is pH-adjusted
with concentrated hydrochloric acid to a final pH of 7.2. Therefore, the passage of cur­
rent through a solution containing these ions has enough potential to create reactive
chlorine based substances such as free chlorine (e.g. Cl2, OCl- , and HOCl), chlorine
dioxide (ClO2), chlorite (C lO j), monochloramine (NH2Cl), dichloramine (NHCl2),
and trichloramine (NCl2) due to electrolysis.
Preliminary experiments were per­
formed in which significant concentrations of these reactive chlorine products were
measured in annular reactors.
Other initial experiments also noted corrosion of the stainless steel electrodes in
the annular reactor systems. After three days of constant current-application, moder­
ate corrosion was noted on the electrodes which warranted replacement after five days.
Significant iron corrosion products were also noted after seven days of current applica­
tions resulting in elevated iron concentrations (approximately 2-3 mg/L) as measured
by Hach Ferrozine method (Hach, Loveland, Colorado) • on a Spectronic Genesys 5
spectrophotometer (Spectronic Genesys, Rochester, New York). The source of the
excess corrosion was from the annular reactor’s internal stainless steel components
in contact with the fluid of electrically-treated reactors. The reactors that had no
current applied to the fluid space did not display these corrosion products or iron con­
centrations. Therefore, the metal surfaces of all reactors were sealed from the fluid
component with silicon sealant. Once these surfaces were appropriately sealed, the
corrosion products and iron concentrations were reduced to negligible levels. These
35
sealed reactors were used for the remainder of the studies. Evaluation then began of
the bioelectric effect to enhance the antimicrobial properties of free chlorine against
mixed population biofilms of drinking water origin.
Table 3. Bioelectric reactor configurations.
Reactor #
Description of conditions
I
Untreated control-No current or free chlorine was applied to the
reactor.
2
Treated by current application (15.7 mA) proportional to a cur­
rent flux of 3.7 /JlA fm m 2.
3
Treated with a constant free chlorine supply of 0.5 mg/L.
4
Treated with free chlorine (0.5 mg/L - administered at the same
rate as Reactor 3) and by current application (15.7 mA) propor­
tional to a current flux of 3.7 /j A fm m 2.
5
Treated with a constant free chlorine supply of approximately 5.0
•mg/L.
6
Treated with free chlorine (administered at the same rate as Re­
actor 5) and by current application (15.7 mA) proportional to a
current flux of 3.7 /iA fm m 2.
Biofilms were grown in the modified annular reactors until they had attained
steady state and then evaluated with the control and experimental groups as shown
in Table 3. Free and total chlorine concentrations were determined at 0, I, 2, 3,
4, 24, 48 and 168 hours following initiation of current application in the annular
reactors. Free and total chlorine concentrations were determined using a DPD col­
orimetric method. The two infeed free chlorine concentrations, 0.5 mg/L and 5.0
mg/L, provided residual chlorine concentrations of approximately 0.06 mg/L and 1.5
36
mg/L, respectively, within the annular reactors. Biofilm and planktonic samples were
obtained at 0, 6, 12; 24, 48, and 168 hours after initiation of the experiment. The
stainless steel electrodes were monitored daily for visible corrosion.
Contact Biocide Surface - Laboratory
N-halamine Polyurethane Activation.
The N-halamine polyurethane coating was provided and applied by VansonHaloSource to clean polycarbonate annular reactor slides. The N-halamine coating
used in the following experiments was received in an uncharged state without biocidal
properties. In all of the following experiments (except one test reactor in the first I
mg/L chlorine test series) the coating was activated or charged with 100% household
chlorine bleach (5-7% NaOCl by weight). The coated annular reactor slides were
immersed in the bleach for 2 hours then removed and rinsed with chlorine demand
free NANOpure water. The slides were then allowed to dry in th,e dark overnight
before inserting into the reactors. This was the activation protocol recommended by
the manufacturer, Vanson-HaloSource Inc.
■ Biofilm Reduction.
Much research has been done previously by the manufacturer and other re­
searchers in the area of challenging the contact biocide coating with bacteria once
fully charged. The focus of the enclosed study was to see if this coating could be
effectively charged in the presence of typical drinking water chlorine residuals. Mixed
37
species drinking water biofilms were grown in the annular reactors until they reached
steady state and then chlorine treatments were administered. Table 4 outlines the
experimental conditions for the first experiment.
Table 4. Contact biocide reactor configuration with different polycarbonate slide
compliments, I mg/L chlorine treatment.
Reactor #
Description of conditions
1
Polycarbonate slides, 1.0 m g/L residual chlorine treatment.
2
Polycarbonate slides coated with non-chlorinated N-halamine
polyurethane-based coating, 1.0 m g/L residual chlorine treatment.
3'
polycarbonate slides coated with chlorinated N-halamine
polyurethane-based coating, 1.0 mg/L residual chlorine treatment.
4
Polycarbonate slides, untreated control.
Annular reactor 3 contained the chlorine activated N-halamine polyurethane coated
slides at the start of the experiment while reactor 2 contained inactivated N-halamine
slides. Reactor inputs, flow, and shear conditions were set up as described previously,
and the reactors were allowed to run for a period of 10 weeks. Once per week, two
slides from each reactor and triplicate planktonic samples (10 ml) were obtained.
Once the biofilm attained steady state, chlorine was administered' to reactors 1-3
at an influent concentration of 2.5 mg/1 (+ /- 0.08). This influent concentration of
free chlorine results in approximately 1.0-1.2 mg/1 free chlorine in the effluent of the
annular reactors. Reactor 4 received no treatm ent of chlorine and served as an un­
treated control. Planktonic and biofilm samples were taken at 3 days, I, 2, 3, and 5
38
weeks after the start of chlorine treatment to the reactors. Chlorine concentrations
were monitored daily and chlorine delivery adjusted if needed to maintain the de­
sired residual of I mg/L. A replicate of this experiment was performed without the
non-chlorinated slide reactor (#2).
Table 5. Contact biocide reactor configuration with different polycarbonate slide
compliments, 4 mg/L chlorine treatment.
Reactor #
Description of conditions '
1
Polycarbonate slides, 4.0 m g/L residual chlorine treatment.
2
3
Polycarbonate slides coated with chlorinated N-halamine
polyurethane-based coating, 4.0 m g/L residual chlorine treatment.
Polycarbonate slides, untreated control.
4
Polycarbonate slides, 4.0 mg/L residual chlorine treatment.
5
Polycarbonate slides coated with chlorinated N-halamine
polyurethane-based coating, 4.0 mg/L residual chlorine treatment.
Polycarbonate slides, untreated control.
6
Another group of these reduction experiments was performed using the maxi­
mum allowable residual free chlorine for drinking water distribution systems of 4.0
mg/L (disinfectant byproduct rule, EPA). All experimental conditions except the
chlorine treatm ent and reactors containing previously chlorinated (uncharged) coat­
ings remained the same. The configuration of this experiment is outlined in Table 5.
Reactors 4-6 were an identical replicate group for this experiment.
Biofilm Prevention.
39
In addition to testing the coating’s ability to reduce biofilm it was also evalu­
ated for its ability to prevent biofilm development in the presence of a low chlorine
concentration. This experiment started with clean reactors and a low chlorine and
monochloramine treatm ent of 0.20 m g/I/ effluent concentration. There were three
reactors in each group; a treated reactor with polycarbonate slides, another treated
reactor with the contact biocide coated slides as in the reduction experiment, and an
untreated control with polycarbonate slides. The contact biocide coated slides were
charged or activated with bleach prior to insertion into the reactor. Planktonic and
biofilm samples were taken every two weeks over a 24 week period. After week 14,
the chlorine and monochloramine effluent concentration was adjusted to 0.5 mg/L for
the remainder of the experiment. Chlorine and mono chloramine concentrations were
monitored routinely. The reactor configuration is shown in Table 6.
Chlorine Uptake By Coating.
Throughout collection of the data with the N-halamine contact biocide surface
it was apparent th at exploration of the coating’s ability to bind free chlorine needed
attention. It was hypothesized that the uptake rate of the coating is quite slow and
is competing with other chlorine demand factors in the water. Chlorine uptake ex­
periments were performed to determine the rate and a kinetic model for the coating’s
chlorine uptake.
The uptake experiments used a clean and virtually chlorine demand free annular
40
Table 6. Contact biocide reactor configuration with different polycarbonate slide
compliments, low residual chlorine and monochloramine treatment.
Reactor #
Description of conditions
1
Polycarbonate slides, 0.2(14 weeks)-0.5(remaining 6 weeks) mg/L
residual chlorine treatment.
2
Polycarbonate slides coated with chlorinated N-halamine
polyurethane-based coating, 0.2(14 weeks)-0.5(remaining 6 weeks)
m g/L residual chlorine treatment.
Polycarbonate slides, untreated control.
3
4■
Polycarbonate slides, 0.2(14 weeks)-0.5(remaining 6 weeks) mg/L
residual monochloramine treatment.
5
Polycarbonate slides coated with chlorinated N-halamine
polyurethane-based coating, 0.2(14 weeks)-Q.5(remaining 6 weeks)
mg/L residual monochloramine treatment.
Polycarbonate slides, untreated control.
6
reactor to test the coating. The chlorine solution was prepared with diluted bleach
in chlorine demand free water and chlorine measurements were made using the same
DPD colorimetric method described previously. The temperature of all experiments
ranged from 22-24°C. A minimum of three experiments for each concentration were
completed for a polycarbonate control and the contact biocide coating.. The con­
tact biocide was prepared by activating with chlorine using the protocol described
previously. The slides were then deactivated for 2 hours with 0.02 N sodium thiosul­
fate solution to quench the adsorbed active chlorine, then rinsed and dried overnight
before beginning isotherm experiments.
The reactors were operated in batch mode with the same mixing/shear conditions
41
as all other experiments. Chlorine measurements were taken at various time points
throughout each experiment and the disappearance of chlorine was assumed to be
the adsorbed quantity. The polycarbonate control for each concentration was used to
determine the chlorine demand of the annular reactor system.
Contact Biocide Surface - Field
Long-term testing was conducted at three participating utilities to determine
the ability of surfaces containing N-halamines to resist and/or control the growth
of biofilm in distribution systems using free chlorine and chloramine residuals to
recharge the surfaces. Two annular reactors were installed at each utility - one re­
actor contained polycarbonate slides as the control and the test reactor contained
polycarbonate slides coated with the N-halamine polyurethane. All of the physical
parameters for the reactors are as previously described for the laboratory experiments
with the exception of the substrate/ nutrient feed and disinfectant treatments. Two of
the utilities operate with free chlorine residuals (Southern Nevada Water Authority,
Las Vegas, Nevada and Seattle Public Utilities, Seattle, Washington) and one util­
ity uses chloramine residuals (Sewerage & Water Board of New Orleans, Louisiana).
The annular reactors were exposed to finished water at each site and operated by the
utilities during the test period from August 2002 through May 2003. Each utilities
finished water characteristics for this study are summarized in Table 7.
:
Southern Nevada Water Authority.
42
Table 7. Average finished water characteristics at participating utilities.
U tility
pH
T em p
A lk
(°C)
(m g /L
TOC
as
D i s i n f e c t a n t R e s id u a l
(m g /L )
CaCOg)
S o u th e r n
W a te r
N evada
7 .9
18
132
2 -3
1 .0 - 1 .5 m g / L C L (free)
7.82
1 4 .5
1 6 -2 1
0 .6 8 -1 .1
0 .9 2 m g / L C l2 (free)
8 .8 1
22
92
3 .1
0 .1 0
A u t h o r it y ,
L as V egas, N ev a d a
S e a ttle
U tilitie s ,
P u b li c
S e a t t le ,
W a s h in g to n
S e w e r a g e & W a te r
B oard o f N ew
O r­
m g /L
C l2 (fr e e );
2 .9 0 m g / L C l2 ( t o t a l)
le a n s , N e w O r le a n s ,
L o u is ia n a
The Southern Nevada Water Authority (SNWA) serves a population of approx­
imately 1.5 million and purifies water diverted from Lake Mead, which is supplied
by the Colorado River as its primary source. SNWA operates two water treatment
plants with total system capacity of 750 million gallons per day (mgd). At the Alfred
M erritt Smith Water Treatment Facility, treatment includes coagulation (1.6 mg/L
ferric chloride), flocculation, sedimentation, and disinfection with chlorine (2.5 mg/L
for final concentration of 1.5 m g/L). Additional treatment at the plant includes zinc
orthophosphate (2.0 mg/L) and fluoride (final concentration of 0.8 mg/L) additions
prior to filtration, clear well storage, and pumping into the distribution system.
The two annular test reactors were placed side by side on a table in the Operations
Building at Las Vegas Valley Water District (Figure 9). The reactors were connected
by plastic tubing and brass fittings to a line th at services the wet training room with
normal distribution system water. The test reactors were indoors, protected from the
43
Figure 9. Control and test reactor setup in Wet Training Room of the Las Vegas
Valley Water District, Las Vegas, Nevada.
elements, wrapped with aluminum foil to protect from direct and indirect sunlight.
The ambient temperature of the room was maintained between 16-25 °C. For the
duration of this field-testing program, the Treatment Facility operated normally, but
the water feeding into the Operations Building was either a blend of groundwater
from LVVWD Wells and treated surface water from AMSWTP (8/21/02 - 10/3/02
and from 5/5/03 to the end of the project) or surface water (10/4/03 - 5/4/03).
There was no ozone treated water during these periods. The test reactors were kept
in the Operations Building Wet training room from August 21, 2002 until February
24, 2003 and were allowed to function undisturbed. Due to painting that had to be
44
done in the Wet Training room the reactors had to be moved (February 24, 2003)
and were carefully relocated to the Operations Building Water Quality Laboratory
where conditions were very similar to the Wet training room
Seattle Public Utilities.
Seattle Public Utilities (SPU) serves a population of approximately 1.3 million
and purifies water diverted from the Tolt and Cedar Rivers. SPU operates two water
treatm ent facilities with a combined capacity of approximately 300 mgd. Currently,
at the Cedar River Facility water treatment includes screening, corrosion control and
a final disinfection with chlorine (target dosage 1.3 mg/L). Additional treatment at
the Cedar River Facility includes the addition of lime (1-3 mg CaO/L) and fluoride
(1.0 mg/L) prior to gravity feed into the distribution system.
The two annular reactors were placed side by side on a laboratory bench near
a sink in the Water Quality Laboratory. The reactors were connected by plastic
tubing and brass fittings to a line connected to a sink providing distribution system
water. They were protected from direct/indirect sunlight within the laboratory and
exposed to ambient temperature conditions. For the first 13 weeks of operation, the
test reactors were allowed to function undisturbed. During the weekend of December
21-22, 2002, the feed tubing from the sink began to leak, potentially impacting the
flow to the reactors. The problem was discovered immediately on Monday morning,
and the reactors remained full despite the water leak. Thus, the feed line was replaced
45
and sampling continued according to schedule.
Sewerage k. Water Board of New Orleans.
The Sewerage & Water Board of New Orleans (S&WBNO) serves a population
of approximately 497,000 and purifies water diverted from the Mississippi River.
S&WBNO operates two water treatment plants with total system capacity of 280
mgd. At the Carrollton Water Plant, conventional treatment includes coagulation,
flocculation, sedimentation, disinfection and filtration. The coagulants used are ferric
sulfate (30-60 mg/L) and a cationic polymer (2-5 mg/L). Calcium oxide (25-50 mg/L)
is added near the point of coagulant addition for corrosion control, and this also aids
in the coagulation process. Disinfection is accomplished by the addition of chlorine
followed by ammonia to produce chloramine at a target level of 3.0 mg/L. Additional
I
treatm ent at the Carrollton Plant includes sodium hexametaphosphate (0.6 mg/L)
and fluoride (1.0 mg/L) additions prior to filtration, clear well storage, and pumping
into the distribution system.
The two annular test reactors were placed side by side on a table in a corridor
at the Carrollton Plant outside of the operator’s office (Figure 10). The reactors
were connected by plastic tubing and brass fittings to a line th at normally provides
Continuous sampling from the clear well at the outlet of the dual media filters.. The
test reactors were protected from rain, but they were not covered and they were
exposed to indirect sunlight and ambient temperature.conditions. For the duration
46
Figure 10. Control and test reactor setup at the Carrollton Water Plant, Sewerage &
Water Board of New Orleans, Louisiana.
of this field-testing program, the Carrollton Plant operated normally and the test
reactors were kept in the same location and allowed to function undisturbed.
Utility Reactor Operation.
1. Daily (or near-daily) inspection for leaks, spin rate, and bulk-water feed rate
with adjustments as necessary, plus associated record keeping on a Reactor
Operations Log sheet.
2. Measurement of chlorine residual, pH, and temperature of the bulk water influ­
ent to the reactors. Frequency varied from daily to once-a-week depending on
operator’s availability and was mandatory for each sampling event. Tasks also
included associated record keeping on a Water Quality Log sheet.
47
3. Sampling of slides for analysis were accomplished at the Center for Biofilm En­
gineering (CBE), Montana State University at Bozeman. Sampling frequency
was once every two weeks except for a single 3-week period during late Decem­
ber. Sampling involved transfer of one coupon from each reactor to a test tube
filled with 0.02M sodium thiosulfate solution and placement of a new “blank”
coupon into the vacant coupon seat within the reactor (to avoid introducing
turbulence due-to rotor slide void). Conclusion of sampling included record
keeping on a Sample Identification Log sheet.
4. Shipping of coupons was in cushioned and insulated test-tube containers. Three
freshly prepared tubes were provided to each utility just before each scheduled
sampling date. The third tube was provided as a back up' in the event one of
the three tubes was broken during shipment to the utilities. The two couponcontaining tubes and the Sample Identification Log were placed in the shipping
container and shipped overnight back to the CBE. ■
Biological Analysis
Sampling Laboratory Annular Reactors
Triplicate 10 mL planktonic and two biofilm samples were taken- for each sample
period. Planktonic samples from chlorine/monchloramine treated reactors received
0.1 mL of 2 M sodium thiosulfate to remove residual disinfectant. Biofilm samples
were harvested by removing a slide from the reactor and scraping the growth from
48
the surface into 10 mL of sterile reverse osmosis water in a petri dish. The suspended
biofilm was then transferred to a sterile test tube. The water for the removed biofilm
samples from treated reactors contained 0.1 mL of sodium thiosulfate.
A sterile
rubber policeman was used to scrape the biofilm from the slides in order to minimize
surface damage of the contact biocide coating. All samples were then homogenized
with a Janke & Kunkel tissuemiser (model T 25 SI, IKA Labortechnik, Germany)
for 50 seconds at 20,500 rpm to break up biofilm and suspended cell conglomerates
before serial dilution. The homogenizer probe was autoclaved before use and rinsed
with sterile RO water between samples. Samples were then serially diluted and plated
for heterotrophic plate counts (HPC) as described below.
Sampling Field Annular Reactors
Field study reactor samples were transported via overnight delivery to the CBE
lab for analysis (see above, Biofilm Prevention - Field). Only biofilm was sampled
from the field reactors. Upon arrival the slide was removed from the test tube and
placed into a sterile 100 mL beaker where the suspension water from the test tube was
emptied. Accurate volume of the sample was accounted for by weight, see equation
3.1.
.
Sample Vol. = (Beaker & Sample W t. — D ry Beaker W t.)
Ig H 2 O
(3.1)
The biofilm was then harvested and homogenized as described above. The homoge­
nized sample was then transferred to a sterile 30 mL test tube before serial dilution
49
and plating.
Heterotrophic Plate Counts
Serially diluted samples were plated using a modified drop streak method (see
Figure 11). This method allows for triplicate plating of the sample for each nutrient
agar plate (R2A nutrient agar). The plated samples were then placed upside down at
room temperature for 7 days, after which colony forming units (CFUs) were counted
on each plate. The average number of CFUs were calculated for each sample period (3
samples for planktonic and 2 samples for biofilm). Heterotrophic planktonic bacteria
were calculated using equation 3.2.
C FU s _
Average Plate Count
mL
Plated Volume x 10(Dilution Series
(3.2)
biofilm bacteria were calculated using equation 3.3.
C FU s _
Average Plate Count
Sample Volume
cm2
Plated V olume x 10(DlluUon Senes #1
S lid eA rea
(3.3)
Selected original samples were then fixed with 0.2 mL of filter sterilized 37 %
formaldehyde and refrigerated at 4°C for total direct Counts.
Total Direct Counts
Only biofilm samples were used for total direct counts in these experiments. The
fixed samples' for total direct counts were sonicated using a probe sonicatOr at 50%
power for 30 seconds to disaggregate the cells. The sample was then vortexed briefly
50
Three 20
drops at one end of
the plate (from same sample)
Petri dish with
R2A nutrient
agar.
The plate is then tilted in the
indicated direction to allow the
drops to streak down the plate.
When the streak nears the
other end of the plate the
plated is then tilted in the
opposite direction
momentarily to allow the
streak to travel about half
Figure 11. Modified HPC drop streak method.
51
and a selected volume filtered through 0.22 /um polycarbonate filter (Osmonics, Inc.,
material #1215609). The filtered cells were then covered with ssO.5 mL of 100 mg/L
DAPI (4’,6-diamidino-2-phenylindole, Sigma) intercalating stain for 20 minutes. The
filter was washed with filter sterilized (0.2 /rm pore size) RO water, then mounted
on a microscope slide with oil and a coverslip and observed under UV light at 100x
objective magnification (Nikon elcipse, model e800). Ten randomly selected fields were
selected and counted. The cell counts from these fields were averaged. The number
of cells for biofilm samples were calculated using equation 3.4.
Cells
cm*
A vera g eF ield C o u n t
F ilteredV olum e
F iltefed A re a
' Sample Volume,
(3.4)
M icroscopeF ieldA rea
S lid eA rea
Total direct count data for biofilm samples are not included in the results due
to the extreme variability which makes the data useless for interpretation. Addition­
ally, the mature biofilm samples contained very large amounts of debris which made
accurate bacterial counting extremely difficult, only the HPC data was used to show
trends and for the data analysis.
Data Analysis
All chlorine data for biotic experiments were combined for each reactor once the
effluent concentration stabilized. The average and standard deviation was computed
for each reactor throughout the treatm ent period. This analysis was adequate to
indicate any differences between reactors of the same treatment concentration.
HPC data for surface biqfilm comparison was graphed over the respective time
52
period of each -experiment to show trends or whether there was significant differences
between surface types for the given treatment. Each data point for the given sample
time was calculated from duplicate biofilm Samples from each reactor and triplicate
HPC for each sample (6 total). Each point on the graphs represent the average count
with y-axis error bars th at indicate =Lone standard deviation. D ata or trends that
indicate possible significant difference between treatments or strategies were tested
using a Student’s t-test with Microsoft Excel’s Data Analysis Tools. The paired ttest compared the means of the control surface (polycarbonate) to the test surface
(N-halamine polyurethane coating) for the given treatment regiment. A t-test value
of zero indicates th a t there is no difference between the two calculated means, t-test
values th a t lie within the 95% confidence interval with 2 tails indicate there is no
significant difference between the means where a t-test value outside of this interval
does. A two-tailed distribution was used over a one-tailed for increased stringency.
r
53
CHAPTER 4
RESULTS
This chapter encompasses the results of the three biofilm control strategies exper­
iments. It continues in the same order as outlined in methods and materials starting
with results from the signaling compounds, the bioelectric effect and finally the con­
tact biocide surface experiments. More attention was directed toward the contact
biocide surface experiments due to promising results that may have application to
the drinking water industry and the possible reduction of biofilms. Initial laboratory
experiments using the contact biocide surface tested the ability of the coating to re­
duce an established biofilm compared to a non-chlorine reactive surface using typical
drinking water chlorine residuals. Other laboratory experiments tested the ability of
the N-halamine coating to resist biofilm growth in the presence of a very low chlorine
concentration. The last biotic experiments included field study results of the coat­
ing technology followed by kinetic experiments of the surfaces chlorine adsorption
characteristics.
Signaling Compounds
It was found th a t the antagonist signaling molecules, FOdHSL and wrs-I-51,
did not demonstrate a statistically significant reduction of biofilm bacteria or an
54
1E -06
—
Control - Planktonic
—
F Od H S L - Planktonic
1E+04
Time (hrs)
Figure 12. Comparative planktonic HPC data for control and signaling molecule
FOdHSL treatment.
increase in the number of detached bacteria when compared to controls (see Figures
12, 13, 14 and 15). While effective in promoting detachment in Pseudomonas aerug­
inosa biofilms [29], these antagonist molecules were not effective in mixed population
biofilms of drinking water origin.
However, BHL at a final annular reactor concentration of 60 nM was able (in a
transient manner) to significantly increase the concentration of planktonic bacteria
and decrease biofilm concentrations (see Figures 16 and 17). This signaling molecule
(at 60nM) promoted an order of magnitude increase and decrease in planktonic and
biofilm bacterial concentrations, respectively. However, this change in biofilm and
planktonic bacterial concentrations was transient and once the BHL was subsequently
55
1 E+07
-----
I
E
5
1 E + 06
Control - Biofilm
FO dH SL - Biofilm
I E+05
O
5
10
15
20
25
30
T im e (hrs)
Figure 13. Comparative biofilm HPC data for control and signaling molecule FOdHSL
treatment.
1 E+06
Control - Planktonic
*
wrs-l-51 - Planktonic
I E+04
Time (hrs)
Figure 14. Comparative planktonic HPC data for control and signaling molecule
wrs-I-51 treatment.
56
1E+07
I F--—
11'
----
g
I 5—
—g
m
T
E
=J 1E +06
Li-
O
— Control - Biofilm
« - wrs-I-51 - Biofilm
1E +05
0
5
10
15
20
25
3
Time (hrs)
Figure 15. Comparative biofilm HPC data for control and signaling molecule wrs-I-51
treatment.
— Control - Planktonic
» BHL - 60 nM - Planktonic
* BHL - 6000 nM - Planktonic
1E+04
Tlme(hrs)
Figure 16. Comparative planktonic HPC data for control and signaling molecule BHL
treatment.
57
I E »07
...... .......... ............................ .................... *
“s
5
1E+06
U-
O
—• —Control - Biofilm
* B H L - 60 nM - Biofilm
— B H L - 6000 n M - B iofilm
I E+05
T-------------------------------------------- 1-------------------------------------------- 1------------------------------------------ 1— -
0
6
12
18
24
30
Time (hrs)
Figure 17. Comparative biofilm HPC data for control and signaling molecule BHL
treatment.
washed out of the reactor (120 minute residence time), the biofilm recovered within
25 hours. These results incited further testing of the ability of BHL to promote de­
tachment from mixed population biofilms in annular reactors at concentrations above
(60 nM) the physiological BHL concentration to 6000 nM . The resulting increased
concentration of BHL, did not affect biofilm levels and did not significantly affect
annular reactor planktonic bacterial concentrations.
The antagonists and the one analog signaling molecule did not show any apparent
increased efficacy on biofilm detachment from mixed species biofilms of drinking water
origin. While some increased detachment was noted using BHL, the transient effect
would not be effective for biofilm control in a long term application. Therefore, the
58
utilization of signaling compounds to promote biofilm detachment was not effective
in this experimental system.
Bioelectric Effect
These experiments evaluated the application of current and chlorine in annular
reactors with an established biofilm as a novel treatment method. All chlorine treat­
ments had a control reactor that did not receive the application of current for testing
the bioelectric effect. Figure 18 shows planktonic bacterial concentrations in the ex­
perimental and control reactors during the seven day study. Planktonic bacteria were
detected at concentrations between IO5 - IO6 CFU/m l for all groups studied with the
exception of reactors treated with 5.0 mg/L free chlorine influent with and without
current application. When compared to one another, these last two groups did not
display a significant difference in planktonic bacterial concentrations. Therefore, the
application of current had no effect on planktonic bacterial kill rates when applied
alone to the reactor system or in combination with free chlorine administration.
These trends were also noted in the biofilm bacterial numbers as demonstrated
in Figure 19. As one can see, the addition of current to the reactor systems did
not result in a reduction in the numbers of biofilm bacteria when applied alone and
did not have an additive or synergistic bactericidal action when coupled with free
chlorine administration. In fact the application of current increased the number of
biofilm bacteria when applied alone and reduced the bactericidal efficacy of the infused
59
I E +05
*■ " C o n t r o l b i o e l e c t r i c
I E +00
“ Control
—* — b i o e l e c t r i c + 0 . 5 mg/I Cl
“ 0 . 5 mg/I Cl
— * — B i o e l e c t r i c + 5.0 m g /I Cl
“ 5 . 0 mg/I Cl
T im e (d a y s )
Figure 18. Comparative planktonic HPC data for the indicated chlorine and current
treatments.
free chlorine. This may have been due to the generation of nutrients such as soluble
iron beneficial to the growth of some organism’s, such as iron oxidizing bacteria.
The residual chlorine concentrations in the annular reactors were determined for
each group (Figure 20). Residual chlorine concentrations for reactors that received
an influent of 5.0 mg/L free chlorine (with or without bioelectric application) are not
provided in this figure. However, the concentrations in these reactors ranged between
1.0 - 2.0 m g/L and the bioelectric-treated and non-treated reactors did not show
any notable difference in the detected total chlorine concentrations when compared.
All chlorine residuals in the four experimental and control groups where the dose
60
I E+09
" C o n tro l b io e le c tric
*
" C o n tr o l
B io e le c tric + 0 .5 m g/l Cl
™ 0.5 m g/l Cl
— * — B io e le c tric + 5 .0 m g/l Cl
~ 5 .0 mg/1 Cl
I E +08
I E +07
I E +06
I E +05
T im e (d a y s )
Figure 19. Comparative biofilm HPC data for the indicated chlorine and current
treatments.
was 0.5 m g/L (Figure 20), are less than 0.06 mg/L. Specifically, 90% of the 0.5mg/L chlorine influent was used to satisfy the chlorine demand of the reactors as
shown in the detection of only 0.06 mg/L of residual chlorine. There was also a
demonstrable lack of measurable current-generated reactive chlorine products in. the
reactor systems. In fact when current was applied to the reactor receiving 0.5 mg/L
chlorine influent, the observed residual chlorine concentrations were less than those
reactors that received 0.5 mg/L influent alone.
To determine if the presence of the humic substances was responsible for low pro­
duction of chlorine, an additional set of experiments was performed. Using cleaned
61
Bioelectric & 0 5 mg/I Cl Treatm ent
—
0. 5 mg/I Q Treatment
"Bioelectric & No Cl Treatment
"No CITreatment
Time (days)
Figure 20. 0.5 mg/L chlorine treatment and/or bioelectric effluent chlorine concen­
trations..
reactors (i.e. no biofilm or carbon adsorbed to the internal surfaces), the same cur­
rent flux was applied to reactors with BAC treated water supplemented with the
humic/nutrient solution and residence time as previous experiments. Free and total
chlorine were then measured (see Figure 21). The applied current was readily able
to generate reactive chlorine products to levels of almost 2.5 mg/L within minutes.
Therefore, the low residual chlorine levels in the biofilm-containing reactors (Figure
20) may be due to the reaction between chlorine products and the biofilm or reactor
surface adsorbed organic carbon.
62
a Total
Time(min)
Figure 21. Electrolytic generation of chlorine from reactor input water and feed.
Contact Biocide Surface - Laboratory
Biofilm Reduction
These first experiments were designed to evaluate the biofilm reduction capability
of the HaloSource N-halamine coating. Chlorine treatment began after verifying a
steady-state biofilm in all reactors. Influent and effluent chlorine measurements were
made on a daily basis following initiation of treatment. Figure 22 shows the free
chlorine concentrations and Figure 23 the total chlorine concentrations for both the
influent and effluent of all reactors during the I O-week period. Initially the chlorine
demand of the mature biofilms was high resulting in low effluent concentrations, but
63
these residuals stabilized after about 7 days.
— + - • Influent
—■ — P o ly c a r b o n a te
A U n c h a r g e d N -h alam in e
— e — C h a r g e d N -h alam in e
T im e ( d a y s )
Figure 22. Biofilm reduction N-halamine experiment #1 reactor influent and effluent
free chlorine concentrations over the 5 week treatment period.
A summary of the chlorine data after reactor effluent residual stabilization is
shown in Table 8 indicating the mean concentration with the standard deviation in
parenthesis. There were minor fluctuations in the effluent concentrations which were
most likely due to disturbance of the biofilm during sampling, increasing the chlorine
demand of the system. The total effluent chlorine concentrations were slightly higher
than free chlorine concentrations in the system due to reaction of the free chlorine
with humic substances, nitrates, and other organic complexes in the dilution water.
Figure 24 shows the planktonic bacterial concentrations throughout the I O-week
study. The data indicates that the planktonic bacteria in all reactors remained >105
64
— + - • Influent
E. 1 .5
®
P o ly c a r b o n a te
A
U n c h a r g e d N -h alam in e
— C h a r g e d N -h alam in e
T im e ( d a y s )
Figure 23. Biofilm reduction N-halamine experiment # 1 reactor influent and effluent
total chlorine concentrations over the 5 week treatment period.
CFU/mL during the first 5-week period. While the planktonic concentrations re­
mained high in the untreated control reactor, the reactors that received chlorine after
week 5 demonstrated a 2-3 log reduction in bacterial numbers. However, there was
no significant difference in bacterial concentrations between the different treated re­
actors.
Figure 25 shows the HPC data from the biofilm samples taken over the I O-week
period. The results from the first five weeks of the study demonstrate that that there
was no apparent difference between the polycarbonate and the N-halamine surfaces to
grow and maintain a biofilm without chlorine treatment residuals. Even the reactor
containing slides coated with the N-halamine polyurethane that was initially charged
65
Table 8. Average chlorine concentrations during the 1st biofilm reduction treatment
period, I mg/L target residual. Mean (Standard Deviation).
Influent (mg/L)
R eactor#
1
2
3
Free
2.50 (0.08)
2.50 (0.08)
2.50 (0.08)
Effluent (mg/L)
Total
Free
2.51 (0.07) 1.04 (0.15)
2.51 (0.07) 1.01 (0.18)
2.51 (0.07) 1.07 (0.13)
Total
1.20 (0.14)
1.17 (0.17)
1.23 (0.12)
1 .E + 0 7
1 .E + 0 6
1 .E + 0 5
I
,
r
/
i
‘
1 .E + 0 4
&
"Y
.
gL 1 E + 0 3
I
I .E + 0 2
I T
— ♦— P o ly c a r b o n a te
T
—■ — U n c h a r g e d N -h a la m in e
1.E +01
a
— x-
C h a r g e d N -h a la m in e
P o ly c a r b o n a te C o n tro l (N o C I2)
I E+00
0
1
2
3
4
5
6
7
8
9
10
11
T im e ( w e e k s )
Figure 24. Biofilm reduction experiment #1 reactor planktonic HPC concentrations
over the 10 week experiment, I mg/L chlorine treatment.
with chlorine did not show appreciable resistance to biofilm formation in the begin­
ning of the experiment. There was a significant reduction in the numbers of HPC
bacteria. However, the subsequent sample periods after starting chlorine treatment
shows that the N-halamine surface seemed to have an increased effect in reducing
66
N o C l2 -4-------- 1---------► C l2
I .E+OS
I E+07
„
1E+06
-y
^
1.E+05
U
o.
x
,-------------------^
I
I -E+OA
— • — P o ly c a r b o n a te
,------------------- — '
—■— U n c h a r g e d N -h a la m in e
I .E+03
—* — C h a r g e d N -h a la m in e
— X r-
P o ly c a r b o n a te C o n tro l (N o C I2)
1.E+02
0
I
2
3
4
5
6
7
8
9
10
11
T im e ( w e e k s )
Figure 25. Biofilm reduction experiment # 1 surface biofilm HPC densities over the
10 week experiment, I mg/L chlorine treatment.
the numbers of HPC bacteria when compared to the chlorine treated polycarbonate
surface (see Figure 25). By the 8th week there was almost a 2 log difference between
the polycarbonate and the initially charged N-halamine polyurethane slides and a I
log difference between the polycarbonate and uncharged slides. By the end of the
10-week time period the N-halamine surface samples did not show an appreciable
change in biofilm. There was still about a I log difference between the polycarbonate
and charged N-halamine surfaces by the end of the experiment.
A replicate of this first experiment with I mg/L chlorine residual treatment was
performed in order to verify these initial findings. Chlorine measurements were made
in the same manner as in the previous experiment and a summary of the chlorine
67
effluent concentrations once stabilized (in %6 days) are shown in Table 9. The efflu­
ent concentrations throughout the 5 week sample period indicate th at all treatment
reactors were operating within similar effluent limits when compared to the first ex­
periment (Table 8). Sampling of the reactors was completed using the same time
intervals (as in the first experiment) for planktonic and biofilm HPC analysis. Re­
sults of this experiment, are shown in Figures 26 and 27.
Table 9. Average chlorine concentrations during the 2nd N-halamine biofilm reduction
treatm ent period, I mg/L target residual. Mean (Standard Deviation).
Influent (mg/L)
R eactor#
1
2
3
Free
2.47 (0.08)
2.47 (0.08)
2.47 (0.08)
Total
2.48 (0.11)
2.48 (0.11)
2.48 (0.11)
Effluent (mg/L)
Free
1.05 (0.05)
1.02 (0.05)
0.98 (0.07)
Total
1.17 (0.05)
1.15 (0.04)
1.11 (0.07)
The same trends are noted in this experiment except the N-halamine coating’s
performance is not as dramatic as the first experiment. The coating displays increased
reduction of biofilm densities but only about a I log difference between the polycar­
bonate surface once chlorine treatment is administered. The difference in the results
from this replicate experiment may be attributed to the poor condition of the coating.
The slides displayed visible surface blistering of the coating from the polycarbonate
possibly due to inadequate surface preparation or application of the polyurethane (see
Figure 28). The data from this experiment also shows that there are slightly lower
68
1 .E + 0 7
N o C l2
1 .E + 0 6
1 .E + 0 5
I
1 .E + 0 4
1 .E + 0 3
1 .E + 0 2 -
— • — P o ly c a rb o n a te
—» — U n c h a r g e d N -h a la m in e
1 .E + 0 1
A
C h a r g e d N -h a la m in e
— x - • P o l y c a r b o n a t e C o n tro l (N o C L 2 )
1 .E + 0 0
T im e ( w e e k s )
Figure 26. Biofilm reduction N-halamine experiment # 2 reactor planktonic HPC
CFU concentrations over the 10 week experiment, I mg/L target residual.
1 E+09
1 E+08
—
---------------
1 .E + 0 7
« 1 .E + 0 6
O 1 .E + 0 5
— ♦— P o ly c a r b o n a te
1 .E + 0 4 - —■ — U n c h a r g e d N -h a la m in e
1 .E + 0 3 - -
A
C h a r g e d N -h a la m in e
— x - • P o ly c a r b o n a te C o n tro l (N o C I2)
1 .E + 0 2
T im e ( w e e k s )
Figure 27. Biofilm reduction N-halamine experiment # 2 biofilm HPC CFU densities
over the 10 week experiment, I mg/L target residual.
69
Figure 28. Picture of annular reactor slides comparing surface abnormalities between
biofilm reduction experiment I and 2. (A) Polycarbonate, (B) N-halamine coated
polycarbonate (exp. I), (C) N-halamine coated polycarbonate (exp. 2).
planktonic bacteria in the reactor that contained the activated biocidal coating.
To determine the significance of the N-halamine coating’s ability to reduce biofilm
when compared to the polycarbonate surface, a paired t-test was performed on the
data collected in both of these biofilm reduction experiments. The comparison began
after chlorine treatm ent commenced and the samples reached quasi-steady state (1-2
weeks). Table 10 shows the results of the t-test comparing the treated polycarbonate
70
Table 10. Two tailed t-test for surfaces comparing treated polycarbonate control to
the activated N-halamine surface biofilm HPC density (CPU/cm2).
Experiment I
Log Mean
R eactorl
5.41
Variance
# Obs
df
t Stat
P (T < t)
t Critical
Experiment 2
ReactorS
4.18
0.44
Reactorl
4.39
0.20
Reactor2
4.71
0.21
0.19
Reactor2
4.22
0.23
22
22
22
21
24
24
21
5.15
. 4 .1 7 x l0 -5
Reactor3
3.56
0.13
24
23
8.01
8.02x10-8
23
1.54
7.33
0.14
L 8 4 x l0 -7
2.08
2.07
2.07
2.07
R e a c t o r I , p o ly c a r b o n a t e s u r f a c e
R e a c t o r 2 , u n c h a r g e d N - h a la m in e s u r fa c e
R e a c t o r 3 , c h a r g e d N - h a la m in e s u r fa c e
control reactor (I) to the uncharged and charged treated N-halamine surface reactors
(2 and 3). Comparisons of the t-statistic show that there is a significant difference
between the polycarbonate and the previously charged N-halamine surfaces (boldface
numbers, p< IO-5). This is also true for the previously uncharged surface in the first
experiment, but was not in the second which had a p-value equal to 0.14 and the t
statistic within the critical t value for the distribution.
The final surface mediated reduction experiment used a treatm ent dose of 4 mg/L
residual after the biofilm had formed to see if the N-halamine surface was enhanced by
a greater chlorine residual. The data was combined for each reactor after the chlorine
effluent concentrations stabilized, after about I day of treatment. Each reactor’s
average and standard deviation was then calculated for the treatm ent period. A
71
summary of the chlorine data over the 5 day treatm ent period is shown in Table 11.
Reactors 4 and 5 are replicates of I and 2 with the same respective surfaces and
treatment.
Table 11. Average chlorine concentrations during the 5 day treatm ent period - ex­
periments I and 2, 4 mg/L target residual. Mean (Standard Deviation).
■ Influent (mg/L)
R eactor#
I
2
Effluent (mg/L)
Total
5.39 (0.34)
Free
3.99 (0.14)
5.58 (0.48)
3.88 (0.15)
Total ■
4.12 (0.14)
4.06 (0.12)
4
Free
5.34 (0.36)
5.54 (0.48)
5.15 (0.43)
5.18 (0.42)
3.70 (0.20)
3.85 (0,18)
5
5.21 (0.32)
5.25 (0.33)
3.70 (0.24)
3.83 (0.23)
Planktonic HPC data (not shown) did not display a notable difference between
the treated reactors. The HPC biofilm data from these experiments are shown in
Figures 29 and 30. After steady state growth was achieved, a chlorine treatment of
~ 4 m g/L was administered. The treatment period lasted for 5 days, where planktonic
and biofilm data were collected. After 5 days the chlorine treatm ent was stopped and
the biofilm allowed to regrow. This regrowth period was also sampled. Initially after
chlorination there was a sharp drop in biofilm and some slight divergence between the
polycarbonate surface biofilm and the N-halamine surface biofilm. However the differ­
ence was not great enough to tell in the shoft period of time whether the N-halamine
surface was more effective. There was also a noticeable inconsistency between biofilm
results from the two identical experiments which prevents coming to any conclusion.
72
1 .E + 0 6
1 E +04
1 .E + 0 1
T im e ( d a y s )
P o ly c a r b o n a te t r e a t e d (# 1 )
— — N -h a la m in e t r e a t e d (# 2 )
—▲— P o ly c a r b o n a te co n tro l (#3)
Figure 29. 4 mg/L biofilm reduction experiment reactors 1-3 biofilm HPC CFU
densities over the 10 day treatment period.
1 E +08
1 E +07
1 E +06
a
1 .E + 0 5
O
1 E +04
1 .E + 0 3
1 .E + 0 2
1 .E + 01
T im e ( d a y s )
—
- P o ly c a r b o n a te t r e a t e d (# 4 )
—m — N -h a la m in e tr e a t e d (# 5 )
—A — P o ly c a r b o n a te co n tro l (#6)
Figure 30. 4 mg/L biofilm reduction experiment reactors 4-6 biofilm HPC CFU
densities over the 10 day treatment period.
73
Biofilm Prevention
. These experiments were designed to represent dead end line conditions that might
be encountered in a distribution system. Low residual chlorine concentrations (resid­
ual goal of 0.2 and 0.5 mg/L) were administered to evaluate the ability of the Nhalamine surface to prevent the growth of biofilm. Monochloramine was also used
(also at 0.2 and 0.5 m g/L target residuals) to see if it would be effective in the same
setting. The sampling period for this part of the study took place over a 24 week
period.
-
Reactors I and 2 were treated with chlorine with reactor 3 serving as a control.
Again, the difference in effluent and influent concentrations indicate the chlorine
demand of the system. After 14 weeks the concentration was increased to ~0.5 mg/L
for the duration of the experiment.
Table 12. Average chlorine concentrations during biofilm prevention treatment pe­
riod. Mean (Standard Deviation).
Influent (mg/L)
Treatment Goal
0.2 mg/L
(0-14 weeks)
R eactor#
I
2
0.5 mg/L
(14-24 weeks)
I
2
Free
0.83 (0.07)
Total
0.85 (0.07)
0.89 (0.07)
1.34 (0.09)
1.38 (0.10)
Effluent (mg/L)
0.91 (0.07)
Free
0.19 (0.05)
0.18 (0.04)
Total
0.28 (0.04)
0.30 (0.10)
1.37 (0.09)
0.46 (0.09)
0.58 (0.08)
1.41 (0.09)
0.47 (0.08)
0.59 (0.08)
The planktonic data is not shown for this experiment because it did not display
a notable difference between the treated reactors. Biofilm data shown in Figure 31
74
also indicates th at there was no marked difference between the surfaces when treated
with a low residual chlorine concentration. Even after increasing the concentration to
0.5 mg/L for the remaining 10 weeks there was no change in biofilm density on either
of the treatm ent reactor’s surfaces. The steep rise in biofilm density by the 24t/l week
is due to a chlorine feed malfunction which happened about 24 hours before the last
sample was taken.
i E+oe I E +08
—
—
J ir .
—
—
1 E +07
I E +06
1 .E + 0 5
1 .E + 0 4
0 .2 m g /L C l2
0 .6 m g /L Cl2
1 E +03
T im e ( w e e k s )
♦ — P o ly c a r b o n a te tr e a t e d (# 1 )
—■ — N -h alam in e t r e a t e d (# 2 )
—A — P o ly c a r b o n a te co n tro l(# 3 )
Figure 31. 0.2 m g/L (0-14 weeks) and 0.5 mg/L (14-24 weeks) chlorine treatment
biofilm prevention experiment, reactors 1-3 biofilm HPC densities over the 24 week
treatm ent period.
Reactors 4 and 5 were treated with monochloramine with the same concentration
residuals as the chlorine treated reactors where reactor 6 served as a control. The
monochloramine concentrations in each of the treated reactors are summarized in
75
Table 13.
Table 13. Average monochloramine concentrations during biofilm prevention treat­
ment period. Mean (Standard Deviation).
Treatment Goal
R eactor#
Influent (mg/L)
Effluent (mg/L)
0.2 mg/L
4
0.27 (0.05)
0.18 (0.04)
5
4 ,
0.26 (0.05)
0.19 (0.03)
0.53 (0.07)
0.55 (0.07)
(0-14 weeks)
0.5 mg/L
(14-24 weeks)
5
0.63 (0.10)
0.65 (0.09)
Planktonic data from reactors 4-6 (not shown) did not display a notable difference
in bacterial concentrations between the treated reactors. Biofilm surface densities in
reactors 4-6 over the 24 week treatment period are shown in Figure 32. Again the
low residual treatm ent of monochloramine represents findings similar to those of the
chlorine treated reactors. The presence of the N-halamine polyurethane coating did
not show enhanced prevention of biofilm growth when compared to the polycarbonate
surface exposed to low residual mono chloramine treatment.
Chlorine Uptake By Coating
Representative drinking water concentrations of chlorine were used in the uptake
experiments to determine a constant th at describes a 1st order decay rate at which
free chlorine reacts or binds to the N-halamine polyurethane surface. The initial
concentration groups consisted of I, 2 arid 4 mg/L free chlorine. A minimum of
3 timed batch runs were performed for each initial concentration group. The time
76
1 .E + 0 9 -I
1 .E + 0 8
"fe 1 .E + 0 7
I E +06
1 .E + 0 5
0 .2 m g /L C l2
0 .5 m g /L C l2
1 E+04
T im e ( w e e k s )
— P o ly c a r b o n a te tr e a t e d (# 4 )
—■ — N -h a la m in e t r e a t e d (# 5 )
— P o ly c a r b o n a te co n tro l (#6)
Figure 32. 0.2 mg/L (0-14 weeks) and 0.5 mg/L (14-24 weeks) monochloramine
treatment biofilm prevention experiment, reactors 4-6 biofilm HPC densities over the
24 week treatment period.
span for sampling was between 0 and 600 minutes. Only the degradation of bulk
fluid chlorine in the system was measured with the assumption that most of the
disappearing chlorine was binding to the N-halamine moieties in the polyurethane.
The proposed model to describe these observations was an equilibrium concentration
degradation model (1st order), equation 4.1:
C = (Cinit —Cequu) • e k 1+ Cequn
(4 .1)
where the concentration at a given point in time is governed by the initial concentra­
tion, Cinit, the system’s equilibrium concentration with the bulk fluid, CequiI and the
rate constant of degradation or disappearance of chlorine, k. The chlorine demand of
the system alone was negligible.
77
The data and regression curves for each initial concentration group are contained
in Appendix A. Nonlinear regression analysis was accomplished using TableCurve™
software (SPSS Inc.) for each concentration. The regression curve is indicated by the
central curve with progressive curves radiating outward indicating the prediction in­
terval (inside) and 90% confidence interval (outside). A summary of the experimental
conditions and regression parameters are contained in Table 14.
Table 14. N-halamine polyurethane surface uptake model summary.
________ Model P aram eter_________
Cmit (mg/L) Gequil (mg/L) k (min -
Initial Gone. (mg/L)
Ave. Temp. (0C)
I
2
4
22.43
20.40
0.99
1.99
0.26
22.35
3.79
0.69
0.28 .
0.0086
0.0033
0.0052
The initial chlorine concentration parameters correlate well with the observed
data. There was no definitive way to measure the equilibrium concentration in this
system. While degradation in this system was considered negligible, the time needed
to measure the equilibrium concentration would affect the outcome of this measure­
ment. However, the increasing trend of this equilibrium parameter correlating to
an increase in initial concentration lends credibility to this prediction as the surface
approaches chlorine saturation. The variability in the rate constant predictions may
78
be due to the small fluctuations in the temperature when the experiments were per­
formed; the lower temperatures correspond to a lower rate constant which may be
kinetically valid. This is not entirely conclusive due to the difference between the I
and 4 m g/L experiments. Although there is a notable amount of variability in the
rate constant the resulting numbers indicate th at the surface is not “charged” at a
rapid rate under these low chlorine concentrations. In fact these results infer that
the coating would take >10 hours to reach an equilibrium state with the bulk fluid
concentration, providing th at there were no other competing reactions taking place.
Contact Biocide Surface - Field
Southern Nevada Water Authority
SNWA began operating the reactors on August 21, 2002, initiated sampling on
September 23, 2002, and continued sampling every two weeks through May 28, 2003.
HPC data are shown in Figure 33. In general, HPC data ranged from IO4 to IO6
(CFU)'/cm2 and chlorine residuals ranged from 0.2 to 0.8 mg/L. Results indicate
no preferential control of biofilm growth on polycarbonate surfaces coated with Nhalamine materials compared to non-coated surfaces during this study period. The
statistical evaluation of all field study date is presented later in this section.
Seattle Public Utilities
.
SPU began operating the reactors on September 19, 2002, initiated sampling on
October 7, 2002, and continued sampling every two weeks through^ May 28, 2003.
79
1 .E + 0 7
1 E+06
U
1 .E + 0 4
>—P o ly c a r b o n a te s u r f a c e
i—N h a la m in e s u r f a c e
1 .E + 0 2
S a m p l e D a te
Figure 33. Comparative biofilm density HPC data for uncoated polycarbonate slides
and N-Halamine coated slides at SNWA.
HPC data are shown in Figure 34. Similar to SNWA, HPC data generally ranged
from IO4 to IO6 (CFU)/cm2 and indicate no preferential control of biofilm growth on
surfaces coated with N-halamine materials compared to control samples.
Sewerage & Water Board of New Orleans
S&fWBNO began running their reactors on September 3, 2002, initiated sampling
on September 23, 2002, and continued sampling every two weeks through May 28,
2003. HPC data are shown in Figure 35. S&WBNO began running their reactors on
September 3, 2002, initiated sampling on September 23, 2002, and continued sam­
pling every two weeks through May 28, 2003. HPC data exhibited greater variability
compared to SNWA and SPU, but the average counts for all three utilities were
80
1E+07
1 E +06
£
I E +05
1 E +04
P o ly c a r b o n a te s u r f a c e
N -h alam in e s u r f a c e
1 E 4-03
S a m p l e D a te
Figure 34. Comparative biofilm density HPC data for uncoated polycarbonate slides
and N-Halamine coated slides at SPU.
comparable within an order of magnitude.
Statistical Analysis - Field Data
A statistical comparison of the two HPC data sets was performed for each of
the three participating utilities using a paired t-test. Statistical comparisons were
performed at the 95th percentile level (a = 0.05).
A two-tailed t-test was used
to determine if the two reactors produced data sets with mean HPC levels which
differed significantly (test hypothesis). This analysis is the same method used for the
laboratory biofilm reduction experiments # I and 2. The results of the statistical
analysis shown in Table 15, determined that for all utilities, t Stat (in bold) is less
than t Critical and the null hypothesis (the two data sets do not vary significantly)
81
1E+08 n
1E+07
1E+06
I E+04
1 E+03
1E+02
P o ly c a r b o n a te s u r f a c e
N -h alam in e S u rf a c e
I E +01
S a m p le D a te
Figure 35. Comparative biofilm density HPC data for uncoated polycarbonate slides
and N-Halamine coated slides at SPU.
should be retained.
A statistical comparison of the reactor disinfectant residual data sets was per­
formed for each of the three participating utilities using a paired t-test. The statis­
tical comparisons here were also performed at the 95th percentile level (a = 0.05).
The t-test was again used to determine if the two reactors produced data sets with
mean disinfectant levels which differed significantly (test hypothesis). The results of
the statistical analysis are shown in Table 16, and indicate that for all utilities the t
Stat is less than t Critical; therefore the two data sets do not vary significantly and
the null hypothesis is retained.
82
Table 15. Two-tailed t-test for surfaces comparing polycarbonate control to the acti­
vated N-halamine surface biofilm HPC density (CPU/cm2).
Log'Mean
Variance
# Obs
df
t Stat
P(T<t)
t Critical
Las Vegas_____ ______ Seattle______ ____New Orleans^
Control
Test
Control
Test
Control
Test
7.72 x IOb 9.19xlOb 6.55x10* 1.22x10* 1.61x10*
1.55x10*
■•2.20xl012 3.79xl0i2 1.18 x IO12 2.26xl012 7.16xl012 1.44xl012
18
18
16
16
18
18
17 '
15
17.
-0.81
-1.23
0.06
0.22
0.12
0.48
1.74
1.75
1.74
Table 16. Two-tailed t-test for the average reactor chlorine residual comparing the
polycarbonate control to the activated N-halamine surface.
__Las Vegas__ ____ Seattle___ .________ New Orleans ________
Free CL
FYee CL
FYee CL
Total CL
Control Test Control Test Control Test Control Test
Mean (mg/L)
0.17
0.25
0.18
0.26
0.13
0.13
2.25
2.87
Variance (mg/L)
0.014
0.012 ■ 0.108
0.102
0.001
0.001
0.264 0.060
61
61
17
17
4
4
4
# Obs
4
df
60
60
16
17
3
3
3
3
t Stat -0.78
-0.43
-0.27
-1.95
0.22
0.34
0.07
0.40
P(T<t)
2.12
t Critical
2.00
3.18 ■
3.18
83
CHAPTER 5
DISCUSSION
The prevention of drinking water biofilms is virtually impossible where the prac­
ticed control of these biofilms is adequate. None of the strategies investigated pre­
sented a clear solution to the problem through promotion of detachment using sig­
naling compounds or the enhancement of disinfectants through a bioelectric effect.
The moderately enhanced control of biofilm growth found with the N-halamine coat­
ing presents a possible solution, but only if the coating is exposed to high chlorine
residuals. Low chlorine residuals appear to be ineffective in maintaining the coatings
biocidal properties.
Cell Signaling Compounds
Drinking water can contain many different species of bacteria th a t either inhabit
or comprise a biofilm in a distribution system. When compared to pure culture
biofilms grown in the laboratory there may be some correlative properties between the
two given similar environments. Apart from the similar environment of which these
biofilms persist, the quorum sensing system within a mixed species community may
have an inherent complexity that is beyond existing understanding of cell signaling.
The effective use of signaling molecules in a pure culture biofilm setting is apparently
84
not reciprocated in a mixed species biofilm of drinking water origin possibly due to
this undefined complexity.
It has been suggested th at the development and maintenance of a fully mature
P. aeruginosa biofilm phenotype was under the direction of it’s intrinsic quorum
sensing system. [53] A m utant used in this particular study formed biofilm with cell
clusters th at were only 20% of the wild type P. aeruginosa biofilm thickness and were
sensitive to detergent removal. These m utant biofilms were homogenous sheets of cells
without any indication of the highly developed and differentiated structure observed
in the wild type strain. The mutant deficient in the ability to produce the signalling
molecules associated with the biofilm phenotype was re-exposed to these signaling
compounds, it restored the ability to form complex biofilms. The signalling antagonist
molecules FOdHSL and wrs-I-51 were applied to this reactor system in anticipation
th at they would disrupt the mature biofilm phenotype and promote detachment.
However, both antagonists were unable to promote detachment in the mixed species
drinking water biofilms.
In another study the properties of a BHL-deficient P. aeruginosa mutant were
evaluated. [54] They found th at the biofilms were much thicker, contained 10 x as many
cells, detached as large conglomerates, and were very sensitive to shear. From these
findings it was thought th a t this smaller homoserine lactone may be responsible for
either limiting the size of the biofilm or upregulating detachment. The experiments
using this molecule to mediate detachment in drinking water biofilms at a 60 nM
85
concentration have shown th at some moderate detachment was occuring. The results,
however, indicated only a transient reduction of biofilm which was most likely due to
the dilution of initial BHL concentrations once flow was restored to the system. The
increased concentration of BHL to IOOx physiological concentrations did not have an
effect on these biofilms. A possible explanation for this result may be th at these high
concentrations of signalling analogs actually abrogate their effect on biofilm (Dave
Davies - personal communication).
The physical structure of biofilms also present a barrier th at m ay slow or pre-
\
vent significant mass transfer of the signaling molecules in this reactor system. Mass
transfer limitations based on molecular size However have not presented a notable dif- •
ference in diffusivity in water between the larger (FOdHSL and wrs-I-51) and smaller
(BHL) molecules (Steve Hunt - personal communication). However, the permeabil­
ity of cell membranes by these-molecules has been shown to be different. The BHL
molecule is freely permeable where FOdHSL and wrs-L51 which are antagonistic to
the autoinducer molecule (similar size and structure), S-oxo-Cig-HSL, implicated in
biofilm formation and maintenance is not. [24] The permeation of this larger autoinducing molecule is largely due to the activation of efflux pumps in the cell membrane.
•Antagonists and analog signaling molecules have not shown applicable efficacy
on biofilm detachment from mixed species biofihns of drinking water origin, with the
arguable exception of BHL at 60 nM. I t’s use is not likely at this time because of
the limited effect and high post of this compound. Consideration for the molecules
86
safe use in this setting is another more daunting task to overcome. Only pending all
of the necessary evaluation and approval of the chemicals safety for use in drinking
water by the EPA and FDA could this become a possible non regulatory strategy for
biofilm control.
Bioelectric Effect
Results from the experiments testing the efficacy of the bioelectric effect run
contrary to previous studies th at describe the synergistic effect between weak current
fields and antimicrobial agents. [37, 38, 43, 40, 39] However, these investigators used
pure culture biofilms under well-defined media conditions and reactor systems that
do not have metals which corrode under the influence of an electrical current. The
bioelectric experiments described in this thesis more accurately reflect the results
expected given the complex; nature of drinking-water delivery systems.
These studies may also call into question the validity of the synergistic biocidal
nature of the bioelectric effect when current was combined with antibiotics. As re­
viewed in these earlier studies when current was passed through a chlorine-free buffer,
no biocidal effect was noted. [37, 38, 43] However, when antibiotics were added to the
buffer, the efficacy of the antibiotic to reduce bacterial populations was enhanced.
The bioelectric effect may have well been due to unintended chloride contamina­
tion of solutions upon the addition of the saline diluted (NaCl solution) antibiotics
to the “chloride-free” solutions, resulting in the electrolytic generation of reactive
87
chlorine species. While the reactor system used in this study had the potential for
generation of these reactive chlorine species, it was not observed in the biotic exper­
iments. This could be th a t oxygen generated by electrolysis of water reacted with
electrolytically-generated metal ions to create metal oxides which then readily react
with any electrically generated chlorine.
Barring any further conjecture about the reasons why the bioelectric effect did not
work in this system, a more relevant question exists as to how to apply this technology
to a distribution system. The implications of applying an electrical current to a water
system which is composed largely of ferrous metals (e.g. ductile iron, steel, or stainless
steel) could be catastrophic. Although anodic protection (corrosion control) of some
systems does imply th at there is an electrical current involved it does not apply the
needed current flux required for the bioelectric effect as described in previous studies.
This may be an area of exploration to see if there is an added benefit of reduced
biofilm on a system th a t uses this type of corrosion control. This may be a better
approach to the application of this strategy to distribution systems. Regardless,, the
bioelectric effect did -not demonstrate a synergistic effect on drinking water biofilm
reduction when coupled with typical chlorine disinfectant concentrations found in
distribution systems.
Contact Biocide Surface
The last innovative strategy or technology investigated was believed to be the
88
most promising for application to the drinking water industry. A surface coating
th at can be continually replenished by bulk water chlorine residuals to maintain a
biocidal property is of obvious value. The N-halamine technology has demonstrated
its effectiveness as a contact biocide in a number of studies. [46, 47, 48, 49, 50], In all of
these studies the contact biocide was activated with a very strong bleach solution and
then challenged with a potent bacterial culture (> IO6 cells/mL). The result is nearly
always total kill when the culture contacts the biocidal surface. The efficacy of this
technology is dependant upon how much reactive chlorine is bound to the surface,
but is finite and in these studies was always theoretically at saturation. The primary
goal of the experiments presented in this part of the enclosed thesis was to test the
ability of the" N-halamine coating to effectively remain biocidal in the presence of low
chlorine concentrations typically found in drinking water distribution systems.
The first experiments were designed to see if the N-halamine coating would grow
a typical drinking water biofilm compared to a non-reactive surface when no residual
chlorine was present. Then chlorine treatm ent was started to determine its effects
on biofilm development. Polycarbonate surfaces continued to be used throughout all
of the experiments as controls because a polyurethane coating without N-halamine
was not available for the study. These experiments demonstrated th at there was no
apparent difference in the N-halamine’s surface ability to support 'culturable biofilm
when compared to a plain, non-reactive polycarbonate surface. Once the reactors
89
were treated with a I m g/L residual concentration of chlorine, there was a statisti­
cally significant difference in the biofilm between the N-halamine and polycarbonate
surfaces. This biofilm reduction experiment was reproduced to verify these initial
findings. The same trends were observed in the N-halamine coatings ability to reduce
the biofilm more readily and to a larger degree after 5 weeks of treatment. The sur.
faces ability to facilitate reduction of biofilm even though the coating began with the
same cellular density from the initiation of chlorine treatment may have been due to
its ability to increase the diffusion of chlorine to the surface through a chemical affinity
or concentration gradient from the active binding sites' of the N-halamine. Another
may be that the coating may be charged through eroded “holes” within the biofilm
th at allow uninhibited passage of free chlorine to the surface. These hypothesized
mechanisms may be further described by Figure 36.
The observed enhanced biofilm removal by the N-halamine surface in the first
reduction experiments was not as pronounced when a higher chlorine residual was
used in a subsequent experiment. Both experiments with the 4.0 m g/L treatment
showed similar trends in the first hours of sampling however, subsequent samples
displayed inconclusive results. A possible explanation for this observation is that the
chlorine at the higher concentration (4.0 mg/L) simply works so fast on the biofilm
(e.g. through killing and oxidation reactions) th at any notable difference between the
two surfaces could not be determined. The HPC method for analysis used in this
study was approaching the limit of detection for these organisms and conditions (4
90
Bulk Fluid
Bulk Fluid
Blofllm
Polycarbonate substratum
coated substratum
Chlorine diffusion vector
Chlorine charged N-halamine
Polycarbonate substratum
Uncharged N-halamine
Figure 36. Theorized N-halamine activation of previously colonized surface biofilm.
mg/L free chlorine). Contamination of the sample when scraping the slides is also
a notable concern. The N-halamine coating was only applied to the exposed surface
of the slide rendering the bevelled edges un-coated and susceptible biofilm growth
(Figure 28). Also, if sampling were conducted similar to the first experiments (I
mg/L treatments) with a longer treatment period, more conclusive evidence for the
coating performance at the higher residual may have been presented.
In most distribution systems, the location of the lowest residual in the system
is also the location with the highest microbial growth. Experiments planned to test
a minimum chlorine as well as monochloramine concentration (0.2-0.5 mg/L, goal)
to represent a worst case scenario. If the N-halamine can resist growth in a low
91
residual environment compared to other surfaces this would truly be a viable strat­
egy. The data however, from the laboratory experiments did not show any sign of
increased resistance to biofilm growth at 0.2 mg/L, nor were differences seen when the
concentration was increased to 0.5 mg/L. The ability of mono chloramine to charge
the N-halamine surface is not well defined at this time. Difficulties in measuring
mono chloramine accurately prevented studies of this kind. It may be th at the similar
nitrogen-chlorine bond between activated N-halamine moieties and the chloramine
species may prevent significant reactions with the surface.
Chlorine Uptake By Coating
Prior research with the N-halamine technology has been with high chlorine con­
centrations to load or activate the surface. The experimental data obtained from
this study implies th a t the coating is not adequately activated when exposed to low
chlorine concentrations. The first question to be answered was to measure a rate of
activation of the N-halamine moieties within the coating. From the measured data
a proposed model th a t describes a mechanism associated with this rate could also
be hypothesized. Second, surface chlorine/ solution chlorine isotherm experiments
would be appropriate to describe the affinity of the surface. Surface chlorine density
measurements were not feasible because the method (iodometric titration) lacked the
sensitivity to directly measure bound chlorine at low concentrations. The approach
of this part of the study looked at the disappearance of chlorine in the bulk fluid with
92
the assumption th at all of the missing chlorine mass was reacting with the surface
active sites of the N-halamine polyurethane. This was a crude but effective method
in this case because the chlorine demand of the system was negligible.
The results from these chlorine uptake experiments, have shown that the rate
of charging the N-halamine surface is quite slow, with more than 10 hours required
to reach surface/liquid equilibrium. From these observations it can be said that the
rate of charging of the surface is slow compared with other competing reactions in the
water, such as bulk fluid reactions with cells, humic substances and other constituents
of the dilution water. All of these elements in the bulk fluid are in random motion
within the liquid where each impact with the N-halamine surface could relinquish
the stored reactivity of each N-halamine moiety impacted. At the same time the
surface is constantly being recharged by bulk fluid chlorine but at a rate that is not
.able to overcome the other surface interactions (e.g. adsorption of organic carbon
and attachment of planktonic cells). Therefore the use of the N-halamine coatings is
probably not an effective maintenance strategy for biofilm control in a distribution
system at this time. The rate of activation of the N-halamine surface requires a more
rapid rate in order to maintain sufficient biocidal properties.
Field Study
Annular reactors were used to perform long-term testing at the three participating
utilities to determine the ability of surfaces containing N-halamines to resist and/or
93
control the growth of biofilms in distribution systems. Reactors were installed at
each utility and used to test finished waters for a period of approximately 8 months.
Two of the utilities use free chlorine and one utility uses chloramines as the residual
disinfectant in the distribution system. For all three utilities, results indicate no
preferential control of biofilm growth on surfaces coated with N-halamine materials >
compared to control samples.
It is possible th at the N-halamine materials failed to perform as expected because
adequate free chlorine residuals could not be maintained within the test or control
reactors for both Las Vegas and Seattle. Thus, adequate chlorine was not present
to maintain a. charged state in the N-halamine material. If surface charging were
occurring, there should have been some indication from the HPC analysis. Results
from New Orleans indicate th at chloramines were apparently not able to recharge
the N-halamine material, even at reactor effluent total chlorine concentrations of 2.3
- 2.9 mg/L there should have been some indication (HPC data) of biofilm reduc­
tion/prevention if the N-halamine surface was activated by chloramines. The field
study findings complement the laboratory results in this case and also conclude th at
the use of the N-halamine coatings is probably not an effective maintenance strategy
for biofilm control in a distribution system at this time. Continued research and
development of the N-halamine technology may present a possible solution in the
future.
94
CHAPTER 6
CONCLUSIONS
The following conclusions can be drawn from the results of this research:
1. Introduction of cell signaling compounds to the model drinking water systems
did not present an applicable solution to management of biofilm.
2. The bioelectric effect as employed in this study did not reduce the amount of
biofilm in the model drinking water systems.
3. In laboratory experiments the N-halamine contact biocidal coating did show
some indication th a t it could be used as a biofilm control strategy at chlorine
residual concentrations of I mg/L, but the results at low and high chlorine
residuals were inconclusive. A positive effect was not clearly demonstrated in
the field studies. The rate of activation of the N-halamine surface at chlorine
concentrations used in drinking water does not appear to be great enough to
maintain the biocidal properties of the coating. Any application of this control
strategy would require additional exploration and development.
While the cell signalling and bioelectric strategies evaluated in this research did
not present results th at incite further exploration into the validity of the previous re­
search claims, the particular experimental system in which each was tested should be
95
considered. Organisms grown in a pure culture- with defined media behave differently
than a mixed population within undefined media. These experiments Were designed
to model drinking water which is inherently complex and considered a, natural mi­
crobial system. Natural systems are difficult to predict or control. However, these
experimental conditions were representative of an actual drinking water distribution
system and this form of testing is the most fitting for these prospective strategies.
The N-halamine technology demonstrated the most viable solution to the problem
presented here. Although results have shown that, the coating does not work as effi­
ciently in this type of system, future work with the coating should not be overlooked.
The N-halamine compounds are cost effective and can easily be manufactured for the
application of new installations or retrofitting an existing distribution system’s infras­
tructure. The N-halamine compounds do not leach significantly into the surrounding
bulk fluid and are not toxic. These coatings may have more application within plant
processes where chlorine residuals are higher, or are exposed to shock loads of disin­
fectant chlorine. This could also be true for the portions of the distribution system
which are in close proximity of the treatm ent plant where the chlorine residuals are
higher, or in the installation of new lines following a main break, when the new lines
are super chlorinated prior to use.
N-halamine coatings however, may have more application in a setting where they
are not continually challenged by their surrounding environment. Food preparation
surfaces and hospital surfaces may be more fitting applications. These settings have
96
the ability to be continually re-expOsed to much higher concentrations of chlorine
to restore the biocidal properties where drinking water rarely has this opportunity.
Similar coatings may have application to surgical implants as well, where the implant ■
would retain sterility throughout pre-op and operative installation. Maybe future
work in the coating technology will present a solution for the application to drinking
water distribution systems.
97
REFERENCES CITED
98
REFERENCES CITED
[1] E. Vanderwende, W. G. Charaoklis, and J. Grochowski, “Bacterial-growth in
water distribution-systems,” Water Science and Technology 20(11-12), pp. 521' 524,1988.
[2] M. Warnecke and A. Camper, “Pathogens in model distribution system biofilms,”
tech, rep., American Water Works Association-Research Foundation, 1998.
[3] J. Wimpenny, “An overview of biofilms as functional communities,” in Commu­
nity Structure and Co-operation in Biofilms, D. Allison, P. Gilbert, H. LappinScott, and M. Wilson, eds., Symposia of the Society for General Microbiology,
pp. 1-24, Cambridge University Press, Cambridge, UK, 2000.
[4] A. Camper, M. Burr, B. Ellis, P. Butterfield, and C. Abernathy, “Develop­
ment and structure of drinking water biofilms and techniques for their study.,”
J.Appl.Microbiol. 85, pp. 1S-12S, 1999.
[5] B. H. Olson, “Assessment and implications of bacterial regrowth in water distri­
bution systems.,” tech, rep., U.S. Environmental Protection Agency, 1982.
[6] N. P. Burman, “Taste and odor due to stagnation and local warming in long
lenghts of piping,” in Proceedings of the Society for Water Treatment and Ex­
amination, vol. 14, pp. 125-131, 1965.
,[7] N. P. Burman, “The occuraiice and significance of actinomycetes in water sup­
ply,” in Actinomycetales; characteristics and practical importance, G. S. Skinner
and F. A., eds., pp. 219-230, Academic Press, London and New York, 1973.
[8] H. F. Ridgway, E. G. Means, and B. H. Olson, “Iron bacteria in drinkingwater distribution-systems - elemental analysis of gallionella stalks, using xray energy- dispersive microanalysis,” Applied and Environmental Microbiology'
41(1), pp. 288-297, 1981.
[9] H. T. Victoreen, “Control of water-quality in transmission and distribution
mains,” Journal American Water Works Association
, pp. 369-370, 1974.
[10] S. H. Lee, S. O ’Conner, and B. K. Banerji, “Biologically mediated corrosion
and its effects on water quality in distribution systems,” J. Amer. Water Works
Assoc. 72 , pp. 636-645, 1980.
[11] D. B. Smith, H. A.F., and S. Hubbs, “Survey of distribution system coliform oc­
currences in the united states.,” in Proc. Water Qual. Technol. Conf., American
Water Works Association Research Foundation, Denver, CO., (San Diego, CA),
1990.
'
99
[12] W. G. Characklis, “Bacterial-growth in distribution-systems,” Journal American
Water Works Association 79(9), pp. 58-58, 1987.
[13] K. Haudidier, J. L. Paquin, T. Prancais, P. Hartemann, G. Grapin, F. Colin,
M. J. Jourdain, J. C. Block, J. Cheron, 0 . Pascal, Y. Levi, and J. Miazga,
“Biofilm growth in 'a drinking-water network - a preliminary industrial pilotplant experiment,” Water Science and Technology 20(11-12), pp. 109-115, 1988.
[14] C. G. Abernathy, The effect of corrosion control treatments and biofilm disinfec­
tion on unlined ferrous'pipes. Doctor of philosophy, Montana State University,
. 1998.
[15] J. Nickel, I. Ruseska, J. Wright, and J. Costerton, “Tobramycin resistance of
Pseudomonas aeruginosa cells growing as a biofilm on urinary catheter material,”
Antimicrob.Agents Chemother. 27(4), pp. 619-624, 1985.
[16] M. Brown, D. Allison, and P. Gilbert, “Resistance of bacterial biofilms to antibi­
otics: a growth-rate related effect?,” J.Antimicrob.Chemother. 22(6), pp. 777780,1988.
[17] E. Tuomanen, R. Cozens, W. Tosch, 0 . Zak, and A. Tomasz, “The rate of killing
of Escherichia coli by beta-lactam antibiotics is strictly proportional to the rate
of bacterial growth,” J.Gen.Microbiol. 132(Pt 5), pp. 1297-1304, 1986.
[18] P. Stewart, “Theoretical aspects of antibiotic diffusion into microbial biofilms,”
Antimicrob.Agents Chemother. 40(11), pp. 2517-2522, 1996.
[19] J. Anderl, M. Franklin, and P. Stewart, “Role of antibiotic penetration limita­
tion in Klebsiella pneumoniae biofilm resistance to ampicillin and ciprofloxacin,”
• Antimicrob.Agents Chemother. 44(7), pp. 1818-1824, 2000.
[20] K. Xu, G. McFeters, and P. Stewart, “Biofilm resistance to antimicrobial agents,”
Microbiology 146(Pt 3), pp. 547-549, 2000.
[21] . D . De Beer, R. Srinivasan, and P. Stewart, “Direct measurement of chlorine
penetration into biofilms during disinfection,” A p p lEnviron.Microbiol. 60(12),
pp. 4339-4344, 1994. •
[22] P. Stewart, “Biofilm accumulation model that predicts antibiotic resistance
of Pseudomonas aeruginosa biofilms,” Antimicrob.Agents Chemother. 38(5),
pp. 1052-1058, 1994.
[23] D. Davies, A. Chakrabarty, and G. Geesey, “Exopolysaccharide production in
biofilms: substratum activation of alginate gene expression by Pseudomonas
aeruginosa,” A p p lEnviron.Microbiol. 59(4), pp. 1181-1186, 1993.
100
[24] M. Shirtliff, J. Mader, and A. Camper, “Molecular interactions in biofilms,”
Chem.Biol. 9(8), pp. 859-871,-2002.
[25] K. Nealson, “Early observations defining quorum-dependent gene expression,”
in Cell-Cell Signaling in Bacteria, G. Dunny and S. W inans, eds., pp..277-289,
ASM Press, Washington, D.C., 1999.
[26] Cell-Cell Signaling in-Bacteria, ASM Press, Washington, D.C., 1999.
[27] P. Singh, A. Schaefer, M. Parsek, T. Moninger, M. Welsh, and E. Greenberg,
“Quorum-sensing signals indicate th at cystic fibrosis lungs are infected with bac­
terial biofilms,” Nature 407(6805), pp. 762-764, 2000.
[28] R. McLean, M. Whiteley, D. Stickler, and W. Fuqua, “Evidence of autoinducer
activity in naturally occurring biofilms,” FEMS Microbiol.Lett. 154(2), pp. 259263, 1997.
[29] T. Kline, J. Bowman, B. Iglewski, T. de Kievit, Y. Kakai, and
L. Passador, “Novel synthetic analogs of the Pseudomonas autoinducer,”
Bioorg.Med.Chem.Lett. 9(24), pp. 3447-3452, 1999.
[30] M. Manefield, R. de Nys, N. Kumar, R. Read, M. Givskov, P. Steinberg, and
S. Kjelleberg, “Evidence th a t halogenated furanones from Delisea pulchra inhibit
acylated homoserine lactone (AHL)-mediated gene expression by displacing the
AHL signal from its receptor protein,” Microbiology 145(Pt 2), pp. 283-291,
1999.
[31] S. Rice, M. Givskov, P. Steinberg, and S. Kjelleberg, “Bacterial signals
and antagonists; the interaction between bacteria and higher organisms,”
J.Mol.Microbiol.Biotechnol. 1(1), pp. 23-31, 1999.
[32] J. Pearson, L. Passador, B. Iglewski, and E. Greenberg, “A second
N-acylhomoserine lactone signal produced by Pseudomonas aeruginosa,”
Proc.Natl.Acad.Sci.U.S.A. 92(5), pp. 1490-1494, 1995.
[33] E. Pesci, J. Pearson, P. Seed, and B. Iglewski, “Regulation of las and rhl quorum
sensing in Pseudomonas aeruginosa,” J.Bacterial. 179(10), pp. 3127-3132, 1997.
[34] P. Seed, L. Passador, and B. Iglewski, “Activation of the Pseudomonas aerugi­
nosa IasI gene by IasR and the Pseudomonas autoinducer PAL an autoinduction
regulatory hierarchy,” J.Bacterial. 177(3), pp. 654-659, 1995.
[35] J. Pearson, C. Van Delden, and B. Iglewski, “Active efflux and diffusion are in­
volved in transport of Pseudomonas aeruginosa cell-to-cell signals,” J.Bacterial.
181(4), pp. 1203-1210, 1999.
101
[36] S. A. Blenkinsopp, A. E. Khoury, and; J. W. Costerton, “Electrical enhance­
ment of biocide efficacy against Pseudomonas aeruginosa biofilms,” Applied and
Environmental Microbiology 58(11), pp. 3770-3773, 1992.
[37] J. Costerton, B. Ellis, K. Lam, F. Johnson, and A. Khoury, “Mechanism of
electrical enhancement of efficacy of antibiotics in killing biofilm bacteria,” Antimicrob.Agents Chemother. 38(12), pp. 2803-2809, 1994.
[38] J. J ass, J. Costerton, and H. Lappin-Scott, “The effect of electrical currents
and tobramycin on Pseudomonas aeruginosa biofilms,” J.Ind.Microbiol 15(3),
pp. 234-242, 1995.
[39] N. Wellman, S. Fortun, and B. McLeod, “Bacterial biofilms and the bioelectric
effect,” Antimicrob.Agents Chemother. 40(9), pp. 2012-2014, 1996.
[40] P. Stewart, W. Wattanakaroon, L. Goodrum, S. Fortun, and B. McLeod, “Elec­
trolytic generation of oxygen partially explains electrical enhancement of to­
bramycin efficacy against Pseudomonas aeruginosa taofilm,” Antimicrob.Agents
Chemother. 43(2), pp. 292-296, 1999.
[41] C. Davis, M. Shirtliff, N. Trieff, S. Hoskins, and M. Warren, “Quantification,,
qualification, and microbial killing efficiencies of antimicrobial chlorine-based
substances produced by iontophoresis,” Antimicrob.Agents Chemother. 38(12),
pp. 2768-2774, 1994.
[42] P. Stoodley, D. deBeer, and H. Lappin-Scott, “Influence of electric fields and
ph on biofilm structure as related to the bioelectric effect,” Antimicrob.Agents
Chemother. 41(9), pp. 1876-1879, 1997.
[43] B. McLeod, S. Fortun, J. Costerton, and P. Stewart, “Enhanced bacterial biofilm
control using electromagnetic fields in combination with antibiotics,” Methods
Enzyrhol. 310, pp. 656-670, 1999.
[44] P. S. Stewart, G. A. McFeters, and C.-T. Huang, “Biofilm control by antimicro­
bial agents,” in Biofilms II Wiley Series in Ecological and Applied Microbiology
. : Process Analysis and Applications, J. D. Bryers, ed., Wiley, John and Sons,
Incorporated, 2000.
[45] J. Kaminski, M. Huycke, S. Selk, N. Bodor, and T. Higuchi, “N-halo derivatives
v: comparative antimicrobial activity of soft n-chloramine systems,” Journal of
Pharmaceutical Sciences 65(12), pp. 1737-1742, 1976.
[46] D. E. Williams, S. D. Worley, S. B. Barnela, and L. J. Swango, “Bactericidal
activities of selected organic N-halamines,” Applied and Environmental MicrobioZo#/ 53(9), pp. 2082-2089, 1987.
102
[47] G. Sun, W. B. Wheatley, and S. D. Worley, “A new cyclic N-halamine biocidal
polymer,” Industrial and Engineering Chemistry Research 33(1), pp. 168-170,
1994.
[48] G. Sun, L. C. Allen, E. P. Luckie, W. B. Wheatley, and S. D. Worley, “Disin­
fection of water by N-halamine biocidal polymers,” Industrial and Engineering
" Chemistry Research 34(11),, pp. 4106-4109, 1995.
[49] M. W. Eknoian, J. H. Putm an, and S. D. Worley, “Monomeric and polymeric Nhalamine disinfectants,” Industrial and Engineering Chemistry Research 37(7),
pp. 2873-2877,1998.
[50] S. D. Worley, F. Li, E. Wu, J. Kim, C.-I. Wei, J. Williams, J. Owens, J. Wander,
A. M. Bargmeyer, and M. E. Shirtliff, “A novel N-halamine monomer for pre­
pared biocidal polyurethane coatings,” in Surface Coatings International, 2002.
[51] P. Butterfield, A. Camper, B. Ellis, and W. Jones, “Chlorination of model drink­
ing water biofilm: implications for growth and organic carbon removal,” Water
Res. 36(17), pp. 4391-4405, 2002.
[52] USEPA, “Alternative disinfectants and oxidants,” April 1999.
[53] D. Davies, M. Parsek, J. Pearson, B. Iglewski, J. Costerton, and E. Greenberg,
“The involvement of cell-to-cell signals in the development of a bacterial biofilm,”
Science 280(5361), pp. 295-298, 1998.
[54] D. Allison, S. Heys, L. Willcock, J. Holah, and P. Gilbert, “Cellular detachment
and dispersal from bacterial biofilms: A role for quorum sensing?,” in Biofilms:
The Good, The Bad and The Ugly, J. Wimpenny, P. Gilbert, J. Walker, M. Brading, and R. Bayston, eds., pp. 279-286, Bioline, Cardiff, UK, 1999.
APPENDICES
104
APPENDIX A
BOZEMAN WATER QUALITY DATA
Table 17. Bozeman water quality data for the year 2000.
B O Z E M A N W A T E R Q U A L IT Y IN F O R M A T IO N
MCL
Alkalinity
Chlorine residual, free (>0.2)
Fluoride (add to = 1.0)
Hardness, calcium
Calcium
Harness, magnesium
Magnesium
Harness, total
Harness, total
pH
Sodium
Sulfate
Iron
Total dissolved soils
Turbidity (daily average)
Total coliforms
Units
2000 Range
2000 Average
ppm .
N /A
83.73
5 0 -1 0 3 .8
4
ppm
1.17-1.92
1.54
4
ppm
0.00 -1.38
1.01
ppm
N /A
38-73
59.29
ppm
N /A
15. - 29.2
23.71
ppm
17.5 - 42.8 ■
N /A
29.45
ppm
N /A
4.27 -10.45
7.35
ppm
N /A
56.8 -110.4
8 8.74'
N /A
grain/gal
3.32 - 6.46
5.19
6.5 - 9.3
Units
.7.41 - 8.97
8.45
20
ppm
1.32 - 7.94
3.5
'500
ppm
0.00-9,30
0.95
ppm
0.3
0.01 - 0.21
0.04
ppm
56.2 -111.4
500
92.44
0.5
NTU
0.03 - 0.18
0.06
There were no positive samples in Bozeman’s drinking
water after treatment.
(Source: The C ity of Bozeman’s annual water quality report.)
Table 18. Bozeman water quality data for the year 2001.
B O Z E M A N W A T E R Q U A L IT Y IN F O R M A T IO N
MCL
Alkalinity
Chlorine residual, free (>0.2)
Fluoride (add to = 1.0)
Hardness, calcium
Calcium
Harness, magnesium
Magnesium
Harness, to ta l
Harness, total
pH
Sodium
Sulfate
Iron '
Total dissolved soils
Turbidity (daily average)
Total coliforms
Units
2001 Range
2001 Average
ppm
63.20 - 100.00
N /A
80.28
ppm
4
1.28 -1.93
1.56
4
ppm
0.00 -1.29
1.03
ppm
N /A
42.80 - 74.00
56.95
ppm
17.12 - 29.60
N /A
22.78
ppm
N /A
12.00 - 44.00
27.37
ppm
N /A
2.93 -10.74
6.68
ppm
N /A
58.80 -116.00
84.33
grain/gal
3.43 - 6.77
N /A
4.93
units
6.5 - 9.3
7 .9 6 -8 .9 4
8.63
ppm
0.37 -16.44
20
4.24
ppm
500
0.00 -14.20
2.84
0.3
ppm
■ -0.01 - 0.08
0.03
500
ppm
64.60 - 119.10
87.92
0.5
NTU
0.02 - 0.11
0.05
There were no positive samples in Bozeman’s drinking
water after treatment.
(Source: The C ity of Bozem an’s annual water quality report.)
Table 19. Bozeman water quality data for the year 2002.
B O Z E M A N W A T E R Q U A L IT Y IN F O R M A T IO N
MCL
Alkalinity
Chlorine residual, free (>0.2)
Fluoride (add to = 1.0)
Hardness, calcium
Calcium
Harness, magnesium
Magnesium
Harness, total
Harness, total
pH
Sodium
Sulfate
Iron
Total dissolved soils
Turbidity (daily average)
Total coliforms
Units
2002 Range
2002 Average
N /A
ppm.
55.60 - 116.80
89.01
4.
ppm
1.35 - 2.30 ■
1.65
4
ppm
0.64 - 3.36
1.07
N /A
- ppm
33.20 - 75.00
57.59
ppm
N /A
13.30 - 30.00
23.04
ppm
N /A
18.40 - 53.60
33.6
ppm
4.49 -13.09
N /A
,
8.06
N /A
ppm
91.19
52.80 -120.40
grain/ gal
N /A
3.09 - 7.04
5.33
units
7.76 - 8.99
6.5 - 9.3
8.69
20
ppm
2.51 - 17.86
5.88
ppm
500
0.00 - 10.40
3.33
0.3
ppm
0.00 - 0.09
0.02
ppm '
500
62.40 - 120.20
93.33
0.5
NTU
0.03 - 0.12
0.05
■There were no positive samples in Bozeman’s drinking
water after treatment.
(Source: The C ity of Bozeman’s annual water quality report.)
108
APPENDIX B
CHLORINE UPTAKE DATA
C hlonne D egradation
Rank 1 Eqn 8 1 4 3 [Equii ConcJ y=(a-b)exp(-cx) +b
r = 0 .9 6 7 4 3 7d D F A djf2=O 949S692S FitStdErr=0.043929575 Fstat=137.1271
0. 7~ .....
0.5
0 .5
- .......................
200
Time (min)
Figure 37. I mg/L Chlorine degradation data and regression curve.
Concentiation (mg/L)
Concentration (mg/L)
a = 0.98834753 b=0 25642022
c=0.0086439396
Chlorine Degradation
Rank 3 Eqn 8143 [Equii Cone]y=(a-b)exp(-cx)+b
Concentration (mg/L)
^=O 97257997 DF AdjI^=O 97000934 FitSidErr=0.078572125 Fstaf=585.24982
a= 1.985938 7 5=0.28153458
c=0.0032872319
Time (min)
Figure 38. 2 mg/L Chlorine degradation data and regression curve.
C hfotine D egradation
R a n k l Eqn 8 1 4 3 [Equil Cone] y= (a-b) exp(-cx)+b
r = 0 .d 3 d 1 1 4 9 3 D F A d jk = 0 .9 3 2 3 4 9 9 2 FitStdErr=O.2 8448313 Fstat=215.94144
3 .5 - * ^
\
Concentration (mg/L)
Concentration (mg/L)
a=3. 7886719 5=0 69457229
c=0.0051511507
400
Time (min)
Figure 39. 4 mg/L Chlorine degradation data and regression curve.
APPENDIX C
N-HALAMINE POLYURETHANE MSDS
113
010151706-00
PAGE
FOR COATINGS, RESINS, AND RELATED MATERIALS
(APPROVED B Y THE U.S. DEPARTMENT OF LABOR AS
'ESSENTIALLLY SIMILAR' TO FORM OSHA-20)
(MEETS REQUIREMENTS OF GFR 29 PART 1910.1200,
OSHA1S HAZARD COMMUNICATION STANDARD)
3
NPCA 1-84
SECTION I - MANUFACTURER AND PRODUCT INFORMATION
CHEMICAL PRODUCT IDENTIFICATION:
PRODUCT I D ......................... .. F475-0475B QT
PRODUCT C L A S S ......................... : WB ISOCYANATE
TRADE NAME . . . : TRITON CONVERTER CONVERTS
FORMULA VERSION N U M B E R ............ :
5
MSDS PREPARATION D A T E .................. : 07/12/2002
MANUFACTURER IDENTIFICATION:
N A M E ................................. : TNEMEC COMPANY,
INC.
A D D R E S S ............................. .. 123 WEST 23RD AVENUE
T E L E P H O N E ........................... .. ^
7
^
"U6-3064
TRANSPORTATION EMERGENCY.............. : 800-535-5053 (INFOTRAC)
SECTION 2 - HAZARDOUS INGREDIENTS
CAS# 28182-81-2
HEXAMETHYLENE DIISOCYANATE (HDD POLYMER
PCT BY MT:
>90
EXPOSURE LIMIT:
ACGIH TLV/TWA:
0000.500 MG/M3 MFG REC
ACGIH TLV/STEL:
0001.000 MG/M3 MFG REC
This product contains no reported carcinogens or suspected carcinogens.
Contains.isocyanate monomer. If subject to spray application,
engineering and administrative controls must be instituted to
maintain an exposure level below .OOSppm. If these controls are
not adequate, the use of an air-supplied respirator is mandatory.
SECTION 3 - HEALTH HAZARD INFORMATION
EMERGENCY OVERVIEW:
POTENTIAL HEALTH EFFECTS:
EYE:
Redness, tearing, blurred vision.
SKIN:
Allergic skin responses.
INHALATION - OVEREXPOSURE TO SOLVENT VAPORS OR SPRAY MIST:
INHALATION - OVEREXPOSURE TO FREE PIGMENT DUST:
INGESTION:
Gastrointestinal irritation.
CHRONIC EFFECTS:
Prolonged contact or repeated exposure to isocyanate concentrations
greater than the recommended TLV may result in permanent respiratory
and skin sensitization.
Once diagnosed as being sensitized to
isocyanates, no further exposure can be permitted.
TARGET ORGANS:
Can cause eye damage.
Can cause respiratory tract irritation.
Can cause lung damage.
Can be corrosive to gastrointestinal tract.
Can cause skin sensitization.
Can cause respiratory tract sensitization.
OTHER:
This product when mixed with other components acquires the hazards
of all components.
PRIMARY ROUTES OF ENTRY:
Dermal and Inhalation.
PROPOSITION 65:
Pigments and/or other raw materials present in this product contain
trace amounts of a chemical or chemicals known to the State of
California to cause cancer, birth defects or other reproductive harm.
SECTION 4 - FIRST AID MEASURES
EYE CONTACT:
Flush immediately with large amounts of clean water under low
I kinsCONTACT•at least 15 nunuteS• Consult a physician.
Wash affected area with soap and water. Remove contaminated
clotbinq. Dispose of or launder aoorvrdi nrrl v , r o n a u i <- * , , W o i
^■i an
114
010151706-00
F475-0475B
PAGE
QT
4
TMEMEC COMPANY, INC.
MATERIAL SAFETY DATA SHEET
TRITON CONVERTER CONVERTS
if skin irritation persists.
INHALATION:
Remove affected individual to fresh air. Treat
eat symptom,
tomatically. If
breathing is difficult, administer oxygen,
__
If breathing has stopped
give artificial respiration. Consult a physician.
phy
INGESTION:
Drink I or 2 glasses of water to dilute,
te. Do not induce vomiting.
Consult
a physician or poison control center IMMEDIATELY.
Treat symptomatically.
NOTE TO PHYSICIAN:
Exposure to isocyanate products may aggravate persons with
asthmatic type conditions, chronic bronchitis, other chronic
respiratory diseases, s k i n eczema,
----—
--or skin sensitization.
SECTION 5 - FIRE AND EXPLOSION HAZARD DATA
FIRE AND EXPLOSIVE PROPERTIES OF THE CHEMICAL:
Flammabilitv Classification ........ :
F l a s h p o i n t ........................... : -N/A
Explosion L e v e l ...................... : Low
- -N/A
High
- -N/A
Flammability Limits ................ : Lower
- -N/A
Higher - -N/A
EXTINGUISHING MEDIA:
Foam, carbon dioxide, and dry chemical.
FIRE-FIGHTING PROCEDURES AND EQUIPMENTS:
For closed containers, pressure build-up followed by possible
rupture or explosion might occur due to extreme heat exposure.
Product will not burn, but may splatter if temperature exceeds
boiling point. Once water has evaporated, combustion can occur.
Small traces of HCN may be evolved under fire conditions.
Water may be used to cool unruptured containers. Wear self-contain­
ed breathing apparatus with a full facepiece operated in pressuredemand or other positive pressure mode to prevent inhalation of
hazardous decomposition products. Use appropriate extinguishing
media to control fire. Water may cause violent frothing if sprayed
directly into containers of burning liquid.
SECTION 6 - SPILL CR LEAK PROCEDURES
CLEAN-UP:
Remove all sources of ignition.
Spills may be collected with inert,
absorbent material for proper disposal. Use non-sparking tools,
protective gloves, goggles and clothing, adequate ventilation, avoid
the breathing of vapors and use respiratory protective devices.
Transfer absorbent material to suitable containers for proper
disposal.
Remove containers to a safe place and cover loosely until
carbon dioxide has finished evolving.
SECTION 7 - SPECIAL PRECAUTIONS
HANDLING AND STORAGE:
Store in dry area. Keep closures tight and upright to prevent
leakage. Do not store in high temperature areas or near fire or
open flame. Refer to product data sheet for recommended storage
temperatures.
SPECIAL COMMENTS:
Prevent prolonged breathing of airborne contaminants such as vapor, spray
mists, or dusts. Prevent contact with skin and eyes. Do not take
internally.
Keep out of reach of children. Do not reuse or alter
containers without proper industrial cleaning. Do not weld or flame cut
empty, uncleaned containers due to potential fire and explosion hazard.
Consult product data sheet for proper application instructions.
SECTION 8 - SAFE HANDLING AND USE INFORMATION
HYGIENIC PRACTICES:
Wash hands and other contaminated skin areas with warm soap and
water before eating.
EYE PROTECTION:
Use chemical resistant splash type goggles.
115
010151706-00
F475-0475B
PAGE
5
TNEMEC COMPANY, INC.
MATERIAL SAFETY DATA SHEET
QT
TRITON CONVERTER CONVERTS
RESPIRATORY PROTECTION:
Respiratory,protective devices must be used when engineering and
administration^contro^-s ^re_not_adequate_tg maintain Threshold Limit
Values,(TLV) and Permissible Exposure Limits (PEL) of airborne
contaminants below,the listed values for those hazardous ingredients
identified in Section II of this MSDS. Observe OSHA regulations for
respirator use (CFR 29, 1910.134) whenever a respirator is used.
Particulate, chemical cartridge, air purifying half-mask respirators
can be used within certain limitations; consult the respirator
manufacturer for specific uses and limitations. Where airborne
contaminant concentrations are unknown, the use of a NIOSH/MSHA
O^Er^pTECTIONt"
respirator is mandatory.
Use Chemical resistant gloves.
Use chemical resistant coveralls or apron to protect against skin
and clothing contamination.
Use protective cream where skin contact is likely.
VENTILATION:
Sufficient ventilation, in volume and pattern, should be provided through .
both local and general exhaust to keep the air contaminant concentration
below current applicable OSHA Permissible Exposure Limits (PEL) and ACGIH1s
Threshold Limit Values (TLV). Appropriate ventilation should be employed
to remove hazardous decomposition products formed during welding or flame
cutting operations of surfaces coated with this product.
SECTION S - PHYSICAL AND CHEMICAL PROPERTIES
Vapor Pressure
........
Vapor Density ..........
Boiling Range ..........
Formula Weight per Volume
VOC IN LBS PER GALLON . .
Evaporation Rate
. . . .
% Volatile by Weight
. .
.
.
.
..
.
: -N /A
.. . : -N/A
.. . : Lower - - N / A
°F
Higher - -N/A
°F
. . . . :
9.6039 LB/C L
.
.. . :
.000
.
:
.000 (Ether = I )
. . . . :
. 00 0
SECTION 10
STABILITY AND REACTIVITY
INCOMPATIBILITIES:
Water, alcohols, amines, strong bases, metal components, surface
active materials.
DECOMPOSITION:
Carbon monoxide, carbon dioxide, hydrocarbon fragments
Nitrogen monoxide, nitrogen dioxide
Trace amounts of HC N .
CONDITIONS TO AVOID:
POLYMERIZATION3 under uncontrolled conditions.
Will not occur.
STABILITY:
Stable.
SECTION 11 - TOXICOLOGICAL INFORMATION
OTHER:
Consult various toxicology references such as NIOSH's "Registry of
!rasm samwssm: tsiss StsessLtmmu
regarding hazardous ingredients.
SECTION 12 - ECOLOGICAL INFORMATION
ECOTOXICOLOGICAL INFORMATION:
SECTION 13 - DISPOSAL CONSIDERATIONS
WASTE DISPOSAL:
2i?:?2!?_Sf_i?1?5??Ed'an::e
with Federal, state, and local regulations
116
010151706-00
F475-0475B
PAGE
QT
6
TNEMSC COMPANY, INC.
MATERIAL SAFETY DATA SHEET
TRITON CONVERTER CONVERTS
SECTION 14 - TRANSPORT INFORMATION
DOT HAZARD CLASS ..................... ;
TRANSPORTATION ASSISTANCE:
Contact Tnemec1a Traffic department S (816) 474-3400.
SECTION 15 - REGULATORY INFORMATION
FEDERAL REGULATIONS:
There are no SARA reportable materials in this product.
.This product contains NONE of the substances subject to the reporting
requirements of Section 313 ot Title III of the Superfund Amendments and
Reauthorization Act of 1986 and 40 CFR Part 372.
STATE REGULATIONS:
SECTION 16 - OTHER INFORMATION
Prepared by . . .
Date of issue . .
Last Revision Date
NONE
MSDS Prepared for
VANSON-HALOSOURCE, INC
ATTN: FRANK LI
14716 NE 87TH STREET
REDMOND
WA
98052
MSDS Last P r e p a r e d .................: 02/18/2002
HMIS Information:
HealthReactivity-
3*
I
Flammability-
I
For specific information regarding occupational safety and health
standards, please refer to the Code of Federal Regulations, Title 29,
Part 1910.
To the best of our knowledge, the information contained herein is accurate.
However, neither the Tnemec Company or any of its subsidiaries assume
any liability whatsoever for the accuracy of completeness of the
information contained herein. Final determination of suitability of any
material is the sole responsibility of the user. All materials may present
unknown health hazards and should be used with caution. Although certain
hazards are described herein, we cannot guarantee that these are the only
hazards which exist.
117
010151706-00
PAGE
FOR COATINGS, RESINS, AND RELATED MATERIALS
(APPROVED BY THE U.S. DEPARTMENT OF LABOR AS
'ESSENTIALLLY SIMILAR' TO FORM OSHA-20)
(MEETS REQUIREMENTS OF CFR 29 PART 1910.1200,
OSH A 1S HAZARD COMMUNICATION STANDARD)
3
NPCA 1-84
SECTION I - MANUFACTURER AND PRODUCT INFORMATION
CHEMICAL PRODUCT IDENTIFICATION:
PRODUCT I D ......................... : 0475-WHITE IG
PRODUCT CLASS ....................... : ACRYLIC POLYOL
TRADE NAME . . .
TRITON WHITE BASE
FORMULA VERSION N U M B E R ............ :
6
MSDS PREPARATION D A T E .................: 07/12/2002
MANUFACTURER IDENTIFICATION:
N A M E ................................: TNEMEC COMPANY, INC.
A D D R E S S ..............................: 123 WEST 23RD AVENUE
NORTH KANSAS CITY, MO. 64116-3064
TELEPHONE ...........................
: 816-474-3400
TRANSPORTATION EMERGENCY............ : 800-535-5053 (INFOTRAC)
SECTION 2 - HAZARDOUS INGREDIENTS
1
CAS# 13463-67-7
TITANIUM DIOXIDE (TOTAL DUST)
PCT BY WT': 21-30
EXPOSURE LIMIT:
ACGIH TLV/TWA:
0010.000 MG/M3
OSHA PEL/TWA:
0010.000 MG /M3
2 CLAY
CAS# 8031-18-3
FULLER'S EARTH (TOTAL DUST)
PCT BY W T :
1-5
EXPOSURE LIMIT:
ACGIH TLV/TWA:
0010.000 MG /M3
OSHA PEL/TWA:
0015.000 MG/M3
3 ACETIC ACID BUTYL ESTER
CAS# 123-86-4
N-BUTYL ACETATE
PCT BY W T : 1.9470 VAPOR PRESSURE:
EXPOSURE LIMIT:
ACGIH TLV/TWA:
0150.000 PPM
ACGIH TLV/STEL:
0200.000 PPM
OSHA PEL/TWA:
0150.000 PPM
OSHA STEL:
0200.000 PPM
10.000 MMHG ® 68F
4
CAS# 64742-95-6
AROMATIC PETROLEUM DISTILLATE
PCT BY W T :
1-5
5 2 -BUTOXYETHANOL
CAS# 111-76-2
ETHYLENE GLYCOL MONOBUTYL ETHER (SKIN)
PCT BY W T : 2.3170 VAPOR PRESSURE:
EXPOSURE LIMIT:
ACGIH TLV/TWA:
0025.000 PPM
OSHA PEL/TWA:
0025.000 PPM
.600 MMHG S 68F
This suDstance contains a material classified as a hazardous air pollutant.
This product contains no reported carcinogens or suspected carcinogens.
This product contains pigment dusts which may be releases when subjected to
abrasive blasting, sanding, or arinding.
SECTION 3 - HEALTH HAZARD INFORMATION
EMERGENCY OVERVIEW:
POTENTIAL HEALTH EFFECTS:
EYE:
Redness, tearinc, blurred vision.
Severe irritation.
SKIN:
Moderate irritation, drying of skin, defatting and possible
dermatitis.
INHALATION - OVEREXPOSURE TO SOLVENT VAPORS OS SPRAY MIST:
Nasal
and 1respiratory
irritation,
anesthetic
effects,
dizziness,
mkDc Cs i r"vI o
1c*
a z**i
s o v\ o c? a o
» r*V
--V
-*-t r - v n
-t
w>> /-%
—
I^
——
118
010151706-00
PAGE
4
TNEMEC COMPANY, INC.
MATERIAL SAFETY DATA SHEET
0475-WHITE
IG
TRITON WHITE BASE
fatigue, nausea, and headache.
INHALATION - OVEREXPOSURE TO FREE PIGMENT DUST:
Coughing, wheezing, shortness of breath, restricted nasal passages,
lung injury.
INGESTION:
Gastrointestinal irritation, nausea, vomiting, diarrhea, death,
aspiration into the lungs which can be fatal.
CHRONIC EFFECTS:
NOTICE:
Reports have associated repeated and prolonged occupational
overexposure to solvents with permanent brain and nervous system
damage. Intentional misuse by deliberately concentrating and
inhaling the vapors may be harmful or fatal.
Ethylene Glycol Monobutyl Ether may cause blood damage based on animal
data.
TARGET ORGANS:
Can cause eye irritation.
Can cause respiratory tract irritation.
Can cause skin irritation.
Can cause gastrointestinal tract irritation.
Can cause nervous system effects.
Can cause liver damage.
Can cause kidney damage.
Can cause blood disorders.
This product when mixed with other components acquires the hazards
of all components.
PRIMARY ROUTES OF ENTRY:
Dermal and Inhalation.
PROPOSITION 65:
. ,
,
Pigments and/or other raw materials present in this product contain
trace amounts of a chemical or chemicals known to the State of
California to cause cancer, birth defects or other reproductive harm.
SECTION 4 - FIRST AID MEASURES
EYE CONTACT:
Flush immediately with large amounts of clean water under low
pressure for at least 15 minutes. Consult a physician.
SKIN CONTACT:
Wash affected area with soap and water. Remove contaminated
clothing. Dispose of or launder accordingly. Consult a physician
if skin irritation persists.
Remove affected individual to fresh air. Treat symptomatically.
If
breathing is difficult, administer oxygen.
If breathing has stopped
give artificial respiration. Consult a physician.
INGESTION
'ink I or 2 glasses of water to dilute. Do not induce vomiting. Consult
Drink
a physician or poison control center IMMEDIATELY. Treat symptomatically.
NO'!TE TO PHYSICIAN:
SECTION 5 - FIRE AND EXPLOSION HAZARD DATA
FIRE AND EXPLOSIVE PROPERTIES OF THE CHEMICAL:
Flammability Classification.......... : „
„
Flashpoint
. ......................... : 125.0
Explosion L e v e l .......................:
Low
-1.1
High
- -N/A
Flammability Limits ................. : Lower - -NZA
Higher - -N/A
EXTINGUISHING MEDIA:
, ,
Foam, carbon dioxide, and dry chemical.
FIRE-FIGHTING PROCEDURES AND EQUIPMENTS:
and areas where static
Euild-up and possible explosion might occur due to extreme heat
exposure.
Solvent vapors are heavier than air and may travel
considerable distance to a source of ignition and flash back.
' ■
--- *---- -=---*------Wear self-contain__ _ pressureinhalation of
119
010151706-00
0475-WHITE
PAGE
IG
5
TNEMEC COMPANY, INC.
MATERIAL SAFETY DATA SHEET
TRITON WHITE BASE
hazardous decomposition products. Use appropriate extinguishing
media to control fire. Water may cause violent frothing if sprayed
directly into containers of burning liquid.
SECTION 6 - SPILL OR LEAK PROCEDURES
CLEAN-UP:
Remove all sources of ignition.
Spills may be collected with inert,
absorbent material for oroper disposal. Use non-sparking tools,
protective gloves, goggles and clothing, adequate ventilation, avoid
the breathing of vapors and use respiratory protective devices.
Transfer absorbent material to suitable containers for proper
disposal.
Remove containers to a safe place and cover loosely until
carbon dioxide has finished evolving.
SECTION 7 - SPECIAL PRECAUTIONS
HANDLING AND STORAGE:
Store in dry area.
Keep closures tight and upright to prevent
leakage. Do not store in high temperature areas or near fire or
open flame. Refer to product data sheet for recommended storage
temperatures.
SPECIAL COMMENTS:
Prevent prolonged breathing of airborne contaminants such as vapor, spray
mists, or dusts.
Prevent contact with skin and eyes. Do not take
internally.
Keep out of reach of children.
Do not reuse or alter
containers without proper industrial cleaning. Do not weld or flame cut
empty, uncleaned containers due to potential fire and explosion hazard.
Consult product data sheet for proper application instructions.
SECTION S - SAFE HANDLING AND USE INFORMATION
HYGIENIC PRACTICES:
Wash hands and other contaminated skin areas with warm soap and
water before eating.
EYE PROTECTION:
Use chemical resistant splash type goggles.
RESPIRATORY PROTECTION:
Respiratory protective devices must be used when engineering and
administration controls are not adequate to maintain Threshold Limit
Values,(TLV) and Permissible Exposure Limits (PEL) of airborne
contaminants below the listed values for those hazardous ingredients
identified in Section II of this MSDS. Observe OSHA regulations for
respirator use (CFR 29, 1910.134) whenever a respirator is used.
Particulate, chemical cartridge, air purifying half-mask respirators
can be used within certain limitations; consult the respirator
manufacturer for specific uses and limitations. Where airborne
contaminant concentrations are unknown, the use of a NIOSH/MSHA
approved fresh-air supplied respirator is mandatory.
OTHER PROTECTION:
Use Chemical resistant gloves.
Use chemical resistant coveralls or apron to protect against skin
and clothing contamination.
Use protective cream where skin contact is likely.
VENTILATION:
Sufficient ventilation, in volume and pattern, should be provided throuch
both local and general exhaust to keep the air contaminant concentration
below current applicable OSHA Permissible Exposure Limits (PEL) and ACGIH':
Threshold Limit Values (TLV). Appropriate ventilation should be employed
to remove hazardous decomposition products formed during welding or"flame
cutting operations of surfaces coated with this product.
Heavier than air solvent vapors should be removed from lower levels
of work area due to potential explosion hazard and all ignition
sources (non-explosion proof equipment) should be eliminated if
flammable mixtures will be encountered.
SECTION 9 - PHYSICAL AND CHEMICAL PROPERTIES
Veper Pressure
10, CD
120
010151706-00
0475-WHITE
PAGE
TKEMEC COMPANY, INC.
MATERIAL SAFETY DATA SHEET
IG
TRITON WHITE BASE
Vapor Density ..........
Boiling Range ..........
Formula Weight per Volume
VOC IN LBS PER GALLON . .
Evaporation Rate
....
% Volatile by Weight
. .
: -N/A
:
Lower - 244.0
°F
Higher - 343.0
cF
:
11.6570 LB/GL
:
1.581
:
7.800 (Ether = I)
:
7.408
SECTION 10 - STABILITY AND REACTIVITY
INCOMPATIBILITIES:
Strong oxidizing agents.
Caustics.
DECOMPOSITION:
Carbon monoxide, carbon dioxide, hydrocarbon fragments
CONDITIONS TO AVOID:
Heat, sparks, open flames.
POLYMERIZATION:
Will not occur.
STABILITY:
Stable.
SECTION 11 - TOXICOLOGICAL INFORMATION
OTHER:
Consult various toxicology references such as NIOSH's "Registry of
Toxic Effects of Chemical Substances" or Sax's "Dangerous Properties
of Industrial Chemicals" for specific toxicity information
regarding hazardous ingredients.
SECTION 12 - ECOLOGICAL INFORMATION
ECOTOXICOLOGICAL INFORMATION:
SECTION 13 - DISPOSAL CONSIDERATIONS
WASTE DISPOSAL:
Dispose of in accordance with Federal, state , and local regulations
regarding pollution.
SECTION' 14 - TRANSPORT INFORMATION
DOT HAZARD CLASS ..................... :
TRANSPORTATION ASSISTANCE:
Contact Tnemec1s Traffic department SI (816) 474-3400.
SECTION 15 - REGULATORY INFORMATION
FEDERAL REGULATIONS:
^ ^ „,
.
.
v
. .
,. _ _
This product contains the following toxic chemicals subject to the
reporting requirements of Section 313 of the Emergency Planning and
Community Right -To-Know Act of 1986 and of 40 CFR 372:
ETHYLENE GLYCOL MONOBUTYL ETHER (SKIN)
CAS# 111-76-2
PCT BY W T :
2.3170
STATE REGULATIONS:
SECTION 16 - OTHER INFORMATION
Prepared by . . .
Date of issue . .
Last Revision Date
Kevin Settles
07/12/2002
NONE
MSDS Prepared for
VANSON -HALOSOURCE, INC
ATTN: FRANK LI
14716 NE 87TH STREET
6
121
PAGE
010151706-00
7
TNEMEC COMPANY, INC.
MATERIAL SAFETY DATA SHEET
0475-WHITE
IG
TRITON WHITE BASE
REDMOND
WA
98052
MSDS Last P r e p a r e d ................. : NONE
HMIS Information:
HealthReactivity-
2
I
Flammability-
I
For specific information regarding occupational safety and health
standards, please refer to the Code of Federal Regulations, Title 29,
To the best of our knowledge, the information contained herein is accurate.
However, neither the Tnemec Company or any of its subsidiaries assume
any liability whatsoever for the accuracy of completeness of the
information contained herein. Final determination of suitability of any
material is the sole responsibility of the use r . All materials may present
unknown health hazards and should be used with caution. Although certain
hazards are described herein, we cannot guarantee that these are the only
hazards which exist.
MONTANA STATE UNIVERSITY - BOZEMAN
3 1762 10388402 7