United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600 2-79-091
August 1979
Research and Development
Acid-Fast
Bacteria and
Yeasts as
Indicators of
Disinfection
Efficiency
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1 Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6 Scientific and Technical Assessment Reports (STAR)
7 Interagency Energy-Environment Research and Development
8. "Special" Reports
9, Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-79-091
August 1979
ACID-FAST BACTERIA AND YEASTS AS
INDICATORS OF DISINFECTION EFFICIENCY
by
Richard S. Engelbrecht
Charles N. Haas
Jeffrey A. Shular
David L. Dunn
Dipak Roy
Ajit Lalchandani
Blaine F. Severin
Shaukat Farooq
University of Illinois
Urbana, Illinois 61801
EPA-IAG-D6-0432
Project Officer
Raymond H. Taylor
Drinking Water Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
This study was conducted
in cooperation with
U.S. Army Medical Research
and Development Command
Washington, D.C. 20314
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health
and welfare of the American people. Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environment.
The complexity of that environment and the interplay between its components
require a concentrated and integrated attack on the problem.
Research and development is that necessary step in problem solution
and it involves defining the problem, measuring its impact, and searching
for solutions. The Municipal Environmental Research Laboratory develops
new and improved technology and systems for the prevention, treatment,
and management of wastewater and solid and hazardous waste pollutant
discharges from municipal and community sources, for the preservation and
treatment of public drinking water supplies, and to minimize the adverse
economic, social, health, and aesthetic effects of pollution. This publi-
cation is one of the products of that research; a most vital communications
link between the researcher and the user community.
This paper describes enumeration techniques for alternative bacterial
indicators of pollution and evaluates their resistance to chlorine and
ozone disinfection. This provides a much more meaningful measure of the
effectiveness of disinfection of wastewater and water supplies, thereby
limiting the introduction of contaminants into our waters and helping to
protect the quality of our water supplies.
Francis T. Mayo
Director
Municipal Environmental Research Laboratory
m
-------
ABSTRACT
Since the coliform group of organisms is considered to be less resistant
to chlorine than many pathogens, including viruses, the utility of both yeast
and acid-fast organisms as potential indicators of disinfection efficiency
was evaluated. In most laboratory studies, these two groups of organisms
were represented by Candida paAap^-iio^-Lf, and Myc.obaatnsu.um ^ohJjjJMm.,
respectively.
Yeast and acid-fast organisms were found to be consistently present in
raw municipal wastewater at a density of approximately lOVlOO m£'> fecal and
total coliforms were present at a density of about KP/lOO mi and lO^/lOO mi,
respectively. Wastewater chlorination, following secondary treatment, reduced
the density of these organisms by 1-3.5 logs. Laboratory studies involving
lime/alum coagulation were observed to remove yeasts and acid-fast organisms
by 90-99 percent. Removal of acid-fast organisms by sand filtration was from
40-70 percent, and yeasts and coliforms were removed by at least 90 percent.
The relative resistance of the test organisms to free chlorine under
continuous flow conditions was: acid-fast > yeast > coliforms, using both
mixed pure-cultures and diluted clarified activated sludge effluent.
M. ^oJitUsitum occasionally survived breakpoint chlorination conditions.
Relative resistance to ozonation was: M. ^ofctu/jtum > poliovirus > C. pcmap^-i-
Loi«it> > EAck&u.c.kia coti > SalmoneJLia typkimu&ium. Variations in pH between
5 and 10 did not significantly affect percent organism survival of either
yeasts or acid-fast organisms under constant ozone residual while increasing
the temperature from 9° to 40°C increased the inactivation of both organisms.
Large volume sampling techniques for the enumeration of yeasts and
acid-fast organisms were developed for membrane filtration of 1 £ water
samples. Yeasts and acid-fast organisms were enumerated in finished drinking
water at densities of 1.5 and 2.2/i, respectively. Increased resistance of
M. ^ofitiMtum to free available chlorine was judged to be the result of the
impermeability of its cell wall; the increased resistance of C. paAap^^io^^u,
appeared to be the result of the thickness and rigidity of its cell wall.
It was concluded that the primary mode of action of chlorine in disinfection
was disruption of the cell membrane and with a resultant change in cell
permeability, and physical damage to cellular DNA.
This report was submitted in fulfillment of Contract No. DADA 17-72-C-
2125 (U.S. Army Medical Research and Development Command) under the partial
sponsorship of the U.S. Environmental Protection Agency through an inter-
agency agreement (EPA-IAG-D6-0432). This report covers the period from
May 1, 1975 through December 31, 1977.
-------
CONTENTS
Foreword iii
Abstract iv
Figures vi
Tables x
Acknowledgments xii
1. Introduction 1
Overall Objectives of the Study 1
Summary of Previous Work 2
Background and Objectives of the Report 4
2. Conclusions 6
3. Recommendations 9
4. Materials and Methods 10
Introduction 10
Source of Test Organisms 10
Organism Enumeration Techniques 10
Preparation of Pure Cultures (Laboratory Studies) 12
Physical and Chemical Methods 16
Field Methods 17
Unit Operations/Unit Processes Laboratory Methods 27
5. Results and Discussion 38
St. Joseph/Oakwood Field Studies 38
Decatur Field Studies 53
Continuous Flow Chlorination Experiments 61
Removal of Indicator Organisms by Chemical Coagulation
with Alum, Ferric Chloride and Lime 80
Removal of Indicator Organisms by Sand Filtration 87
Continuous Ozonation Studies 89
Mechanism of Inactivation by Chlorine 106
References 127
Appendix 131
A. The Relative Resistance of Acid-Fast and Other Organisms
to Chlorination: A Review 131
-------
FIGURES
Number Page
1 Revised Enumeration Technique for Acid-Fast Organisms 14
2 St. Joseph, Illinois, Wastewater Treatment Plant 18
3 Oakwood, Illinois, Water Treatment Plant 20
4 Salt Fork of the Vermillion River and Sampling Stations 22
5 Water Treatment Plant and Distribution System with Sampling
Stations, Decatur, Illinois 26
6 Continuous Flow Reactor for Chlorine Inactivation Studies .... 28
7 Step-Input Dye Tracer Study Performed on Continuous Flow
Inactivation Reactor 30
8 Pulse-Input Dye Tracer Study Performed on Continuous Flow
Inactivation Reactor 31
9 Schematic Diagram of Reactor Arrangement Nos. 1, 2 and 3 used
in Ozone Inactivation Studies 35
10 Densities of Yeasts at the St. Joseph Wastewater Treatment Plant
Over a 24 Hr Period 39
11 Densities of Acid-Fast Organisms at the St. Joseph Wastewater
Treatment Plant Over a 24 Hr Period 40
12 Densities of Total Coliforms at the St. Joseph Wastewater
Treatment Plant Over a 24 Hr Period 41
13 Densities of Fecal Coliforms at the St. Joseph Wastewater
Treatment Plant Over a 24 Hr Period 42
14 Mean Normalized Organism Densities at the St. Joseph Wastewater
Treatment Plant Over a 24 Hr Period 43
15 Densities of Yeasts at the St. Joseph Wastewater Treatment Plant. 45
16 Densities of Acid-Fast Organisms at the St. Joseph Wastewater
Treatment Plant 46
-------
17 Densities of Fecal Coliforms at the St. Joseph Wastewater
Treatment Plant ........................ 47
18 Densities of Total Coliforms at the St. Joseph Wastewater
Treatment Plant ........................ 48
19 Densities of Yeasts in the Salt Fork River System ........ 49
20 Densities of Acid-Fast Organisms in the Salt Fork River System . 50
21 Densities of Fecal Coliforms in the Salt Fork River System. ... 51
22 Densities of Total Coliforms in the Salt Fork River System. ... 52
23 Continuous Flow Inactivation Studies with 0.53 mg/£ Free Available
Chlorine Residual Using Mixed Pure Cultures in Chlorine Demand
Free Buffer at pH 7 and 22.5°C ................ 63
24 Continuous Flow Inactivation Studies with 0.96 mg/£ Free Available
Chlorine Residual Using Mixed Pure Cultures in Chlorine Demand
Free Buffer at pH 10 and 24.5°C ................ 64
25 Continuous Flow Inactivation Studies with Various Free Available
Chlorine Residuals and Temperatures Using a Pure Culture of
C. paAapA-ULoA-it, in Chlorine Demand Free Buffer at pH 7 ..... 65
26 Continuous Flow Inactivation Study with Various Free Available
Chlorine Residuals and Temperatures Using a Pure Culture of
M. ^ofituyitwm in Chlorine Demand Free Buffer at pH 7 ...... 66
27 Continuous Flow Inactivation Study with Various Free Available
Chlorine Residuals and Temperatures Using a Pure Culture of
C. pasiapA-LlLoAiA in Chlorine Demand Free Buffer at pH 10 .... 67
28 Continuous Flow Inactivation Study with Various Free Available
Chlorine Residuals and Temperatures Using M. ^ofvtaA-tum in
Chlorine Demand Free Buffer at pH 10 .............. 68
29 Non-Breakpoint Chlorination Study with 0.63 mg/£ Total Combined
Residual Using Clarified Activated Sludge Effluent ...... 71
30 Non-Breakpoint Chlorination Study with 0.73 mg/£ Total Combined
Residual Using Clarified Activated Sludge Effluent ...... 72
31 Non-Breakpoint Chlorination Study with 0.84 mg/£ Total Combined
Residual Using Clarified Activated Sludge Effluent ...... 73
32 Non-Breakpoint Chlorination Study with 1.73 mg/£ Total Combined
Residual Using Clarified Activated Sludge Effluent ...... 74
VII
-------
33 Breakpoint Chlorination Study for Acid-Fast Organisms Present
in Clarified Activated Sludge Effluent ............ 77
34 Removal of Indicator Organisms and Turbidity with Alum (Tap
Water, Kaolinite Clay I and 1.0 Percent Raw Wastewater;
Temperature = 10-15°C) .................... 82
35 Removal of Indicator Organisms by Sand Filtration Using Tap
Water Plus 1.0 Percent Raw Wastewater Inoculum ......... 88
36 Survival of C. paA.a.p^^io&^, in Reactor Arrangement Nos. 1 and 3 . 91
37 Survival of M. fiovtu^tum in Reactor Arrangement Nos. 1 and 3 . . 92
38 Survival of M. fioitLUtum in the Presence of Ozone Bubbles with
no Ozone Residual, Reactor Arrangement No. 2 ......... 94
39 Survival of C. poAap^^o-i-ci in the Presence of Ozone Bubbles
with no Ozone Residual, Reactor Arrangement No. 2 ....... 95
40 Effect of pH on the Survival of M. ^ontu/jtum at a Constant
Rate of Applied Ozone ..................... 96
41 Effect of pH on the Survival of C. paJicLp&-Uioi>it> at a Constant
Rate of Applied Ozone ..................... 97
42 Effect of pH on the Survival of M. fioKtiutum for a Constant
Ozone Residual at a Given DT ................. 98
43 Effect of Temperature on the Survival of M. ^ontuJJLim at a
Constant Rate of Applied Ozone ................ 100
44 Effect of Temperature on the Survival of M. ^ontu^tum for a
Constant Ozone Residual at a Given DT ............. 101
45 Effect of Ultraviolet Light on the Survival of M.
at a Constant Rate of Applied Ozone .............. 103
46 Effect of Initial Density of C. paAapAltoA-U, on Degree of
Inactivation for a Constant Rate of Applied Ozone ....... 104
47 Response of Five Test Organisms to Ozone in a DI Phosphate
Buffer Solution ........................ 105
48 Chlorine Uptake Kinetics of E. aoti ............... 107
49 Uptake of Chlorine at Constant Contact Time ........... 109
50 Uptake of Chlorine as HOC1 at Constant Contact Time ....... HO
v i i i
-------
51 Growth of E. c.oti after Chlorination at pH 7 112
52 Growth of C. pasia.p£>4JLoi>JJ> after Chlorination at pH 7 113
53 Growth of M. fastta-itum after Chlorination at pH 7 114
54 Effect of Chlorine on Cellular Potassium: E. c.oti, pH 7 .... 118
55 Effect of Chlorine on Cellular Potassium: C. p4J>, pH 7. 119
56 Effect of Chlorine on Cellular Potassium: M. fioKtiutum, pH 7 . . 120
57 Effect of Chlorination at pH 7 on Protein Synthesis 122
58 Effect of Chlorination at pH 7 on DNA Synthesis 123
-------
TABLES
Number Page
1 Acid-Fast Pretreatment Methodology, Sample: M. faontuAJum Pure
Culture in Buffered Deionized-Distilled Water ......... 12
2 Acid-Fast Pretreatment Methodology, Sample: 0.5% Raw Wastewater. 13
3 Acid-Fast Pretreatment Methodology Using Different Concentrations
of Oxalic Acid, Sample: 0.5% Raw Wastewater ......... 15
4 Operating Characteristics of the St. Joseph, Illinois Wastewater
Treatment Plant (March 1975 January 1976) .......... 19
5 Operating Characteristics of the Oakwood, Illinois Water Treat-
ment Plant (March 1975 - January 1976) ............ 21
6 Detailed Description of the Salt Fork of the Vermillion River
Sampling Stations ....................... 23
7 Operating Data For the South Side Water Treatment Plant,
Decatur, Illinois ....................... 25
8 Location of Sampling Stations, Decatur, Illinois ........ 25
9 Density of Organisms at the Oakwood Water Treatment Plant .... 54
10 Density of Total Col i form Organisms During Decatur Field
Sampling (No./lOO mi by the MF Enumeration Technique) ..... 55
11 Density of Total Coliform Organisms During Decatur Field
Sampling (No./lOO ml by the MPN Enumeration Technique) .... 56
12 Density of Yeast Organisms During Decatur Field Sampling
(No./£) ............................ 57
13 Density of Acid-Fast Organisms During Decatur Field
Sampling (Ho. /I) ....................... 58
14 Estimated Standard Plate Count During Decatur Field
Sampling (Ho. /ml) ....................... 59
-------
TABLES (Continued)
15 Geometric Mean Values of Organism Density and Chlorine
Residual During Decatur Field Sampling Between 4 April 1977
and 11 October 1977 ...................... 60
16 Statistical Correlation Analysis of Raw Water Parameters
for Decatur Field Studies Between 4 April 1977 and
11 October 1977; Number of Samples - 14 ............ 61
17 Inactivation Data for Breakpoint Chlorination Experiments .... 76
18 Physical and Chemical Characteristics of Activated Carbon
Column Influent and Effluent ................. 78
19 Results of Continuous Flow Chlorination Inactivation of
Organisms Present in Clarified Activated Sludge Effluent
After Treatment by Activated Carbon .............. 79
20 Comparison of Chlorine Inactivation Results Using Clarified
Activated Sludge Effluent and Activated Carbon Treated
Activated Sludge Effluent Following 20 Minutes Contact Time . . 80
21 Removal of Indicator Organisms at Optimum Conditions for
Turbidity Removal ....................... 83
22 Optimum Physical-Chemical Conditions for Optimum Turbidity
Removal ............................ 84
23 Initial Conditions for the Experimental Coagulation ....... 85
24 Removal of Indicator Organisms by Sand Filtration ........ 89
25 Effect of Mixing Rate on the Survival of M. ^of^tiiLtwm and
C. paAapAA-loA-ib in Experimental Arrangement No. 3 for a
DT of 24 Seconds ....................... 99
26 Release of UV Absorbing Material After Chlorination, pH 7 . . . . 115
27 Release of Cellular TOC After Chlorination, pH 7 ........ 115
-------
ACKNOWLEDGMENTS
This study was performed in the Environmental Engineering Laboratories
of the Department of Civil Engineering, University of Illinois at Urbana-
Champaign, and was funded jointly by the United States Army Medical Research
and Development Command and the U.S. Environmental Protection Agency. The
cooperation of personnel of the Urbana-Champaign Sanitary District; the City
of Decatur, Illinois; the villages of Oakwood and St. Joseph, Illinois; the
Illinois State Water Survey; and the United States Geological Survey,
Champaign, Illinois, is appreciated. Much of the data and analysis of the
results of removal of indicator organisms by coagulation-flocculation were
provided by Mr. Michael Price.
The critical comments and worthwhile suggestions, as well as the
encouragement, provided by Mr. Raymond Taylor, EPA Project Officer, and
Mr. William Dennis of the U.S. Army Medical Research and Development Command,
are greatly appreciated.
XII
-------
SECTION 1
INTRODUCTION
OVERALL OBJECTIVES OF THE STUDY
The practice of disinfection as applied to water treatment and to waste-
water effluents has generated considerable interest in recent years. At the
present time, the major disinfectant in use in the United States is chlorine.
The major objective of chlorination, or of disinfection in general, is to
produce a finished product that is acceptable from a public health standpoint.
In order to assess the efficiency of disinfection in terms of destruction
of pathogens, the usual approach is to utilize a bioindicator believed to be
at least as resistant to disinfection as the most resistant pathogens. To be
a satisfactory bioindicator, an organism or group of organisms must meet cer-
tain requirements (1,2). For example, a bioindicator should be rapidly and
unambiguously quantifiable in water and wastewater samples by simple and
easily applied techniques. The utility of the bioindicator would be enhanced
if it also could serve as an indicator of fecal contamination.
Since it appears that total and fecal coliforms, currently the most com-
monly used groups of bioindicators, do not meet the criterion of being as
resistant to chlorination as the most resistant pathogens, the suitability of
coliforms for evaluating chlorine disinfection efficiency may be seriously
questioned. This is particularly true where wastewater reuse schemes are
being considered; in such situations protection of the public health is
paramount.
The coliform organisms have been useful in the past in providing infor-
mation on the potential presence of bacterial pathogens in waters and waste-
waters. However, in light of their inability to satisfy the intended purpose
in certain situations (3), it would seem appropriate to reevaluate the appli-
cation of coliform organisms as bioindicators on a case-by-case basis. In
this approach, the most appropriate indicator organism could be selected on
the basis of its purpose and the information required, e.g., the inactivation
of resistant viral pathogens by chlorination. It was with this purpose in
mind that this study was undertaken.
Representatives of two groups of organisms, yeasts and acid-fast bacteria,
were isolated from chlorinated wastewater secondary effluent. Early studies
indicated that these organisms were substantially more resistant to chlorine
and somewhat more resistant to ozone than were coliforms and, as a result,
could be considered as possible new indicators of disinfection efficiency.
-------
It was the purpose of this research to critically analyze these groups of
organisms as to their utility as indicator organisms.
SUMMARY OF PREVIOUS WORK
In a previous report to the Office of Research and Development of the
U.S. Environmental Protection Agency, dated February 1974, a detailed litera-
ture review concerning the applicability of the currently used fecal coliform
index for determining disinfection efficiency was presented (4). The basic
conclusions drawn from the literature review and other studies from this
initial phase of the project are briefly restated here.
The literature indicates that the currently used coliform group of organ-
isms is too sensitive to chlorine to be a totally reliable indicator of the
potential presence of chlorine resistant pathogenic agents. For example,
existing data suggest that many enteric viruses are more resistant to chlorine
than coliforms. While the resistance of vegetative bacteria to chlorine is
varied, the majority exhibit similar or less resistance to chlorine than coli-
forms; spore-forming bacteria, however, are found to be considerably more
resistant to chlorine than vegetative cells. The resistance of cyst-forming
pathogenic protozoans to chlorine has also been reported to be greater than
the resistance of coliforms in laboratory studies; however, since a strong
dependency of cysticidal chlorine dose on cyst density was noted, comparisons
with field conditions may be inappropriate.
In cases where the presence of waterborne acid-fast bacilli are of con-
cern from a public health standpoint, such as with Mycobac.t&iiLjm tubeAc.uJlo&
-------
standard, the isolates displaying the highest degree of chlorine resistance
were yeasts and acid-fast bacilli. Of these isolates, one yeast, previously
reported as No. 30, and two acid-fast bacilli, previously reported as Nos.
132 and 134, were selected for further study; these organisms were identified
subsequently as Candida pJJio&'U> , MycobacteAJ-um pkLoJi, and
respectively.
Pure culture inactivation experiments were performed with these organ-
isms using free chlorine at residuals of 0.1-2.0 mg/£ in pH 7 chlorine demand
free phosphate buffer (CDFB) at 20°C. A comparison of these data with those
published in the literature for several enteric viruses led to the conclusion
that yeasts and acid-fast organisms show sufficient resistance to chlorine to
warrant further study.
Substantial effort was directed towards the development of enumeration
techniques for acid-fast organisms and yeasts. After examination of several
media and selective agents, it was determined that acidified yeast extract-
malt extract-dextrose agar plus several selective and enrichment agents were
adequate for the enumeration of yeast organisms (5,6). The development of
an acid-fast enumeration technique was approached in three ways, i.e., the
evaluation of a pre-treatment step, selective media, and a final staining
procedure. The results of these three approaches when combined produced a
satisfactory method for the enumeration of acid-fast organisms (5,6). More
recent modifications are discussed in Section 4, Materials and Methods.
The results of the second phase of the overall study were covered in a
report to the U.S. Army Medical Research and Development Command and to the
U.S. Environmental Protection Agency, prepared in July 1975 and published in
1977 (5). The basic conclusions from this report are summarized below.
To ensure the consistent presence of the acid-fast and yeast organisms
in wastewater, a detailed study of acid-fast organisms, yeasts, and fecal
col i forms in raw domestic wastewater and treated effluents was performed.
In addition, a study of the occurrence of acid-fast and yeast organisms in
fecal matter was performed so as to identify their source when these organ-
isms are isolated from wastewater. From these studies, it was found that
yeasts and acid-fast organisms are consistently present in raw and treated
domestic wastewater. However, while fecal matter was a fairly consistent
and reliable source of yeasts, it could not account for the high densities
of acid-fast organisms observed in raw wastewater. To further define the
source of these organisms in wastewater, as well as in surface water, several
of the more common acid-fast and yeast organisms isolated from wastewater
were identified as to genus and species.
To avoid the dangers associated with comparing the response of indicator
organisms and pathogenic organisms from separate chlorine inactivation exper-
iments performed under possibly differing experimental conditions, mixed
culture experiments were undertaken using two different yeasts, four differ-
ent acid-fast organisms, Et>ck&u.chsia c.oLi, So£mona££a typh-imuAA-wm , and
poliovirus type 1 (Mahoney). In these studies, varying conditions of pH and
temperature were evaluated in an attempt to further identify the relative
-------
resistance of these organisms to inactivation by free chlorine. These experi-
ments confirmed the superior chlorine resistance of the proposed indicator
organisms.
Studies with inorganic chloramine were performed using pure cultures of
one yeast, two acid-fast organisms, and E. c.oLi in CDFB at pH 7. These
experiments indicated that the resistance of the yeast and the acid-fast
organisms to chloramine was superior to that of E. coti. An investigation
to test the proposed indicator organisms under conditions of ozone disinfec-
tion was initiated, also.
BACKGROUND AND OBJECTIVES OF THE REPORT
This study was initiated 1 December 1969 under a contract (17060 EYZ)
from the U.S. Environmental Protection Agency. A report covering the period
1 December 1969 through 30 April 1972 was submitted in September 1972 and
subsequently published by the U.S. Environmental Protection Agency in
February 1974 (4). During the period of 1 May 1972 through 31 December 1977,
the study was funded equally by the U.S. Environmental Protection Agency and
the U.S. Army Medical Research and Development Command under interagency
agreement EPA-IA6-D6-0432. A report covering the period 1 May 1972 through
31 April 1975 was submitted to the funding agencies in July 1975, and was
published as a report of the U.S. Environmental Protection Agency in August
1977 (5). The present report covers the period 1 May 1975 through 31 Decem-
ber 1977-
Work covered in the previous reports (4,5) finalized the first two
phases of the study which involved a search for a valid indicator organism
of disinfection efficiency, development of enumeration methodology, verifica-
tion of increased resistance of the organism to disinfection, confirmation of
the presence of the organism in wastewater influents and effluents, and other
laboratory studies exploring the utility of the proposed new organisms. This
has been reviewed briefly under "Summary of Previous Work" in this report.
The final phase of the study, as discussed in this report, includes
additional work on the development of enumeration techniques for use in
treated water samples of larger volumes. Using methods as developed, a
detailed study of acid-fast bacteria, yeasts, and fecal and total coliforms
was undertaken in a wastewater treatment plant, the associated receiving
stream, and a downstream water treatment plant using the same stream as a
raw water source. From these studies, over a period of 9 months, earlier
results were confirmed noting the consistent presence of the proposed indi-
cator organisms in water and wastewater samples. Analysis of the variation
between the organisms confirmed the increased resistance of the acid-fast and
yeast group of organisms to wastewater treatment and disinfection when com-
pared with the coliform group.
To further determine the occurrence of the proposed indicator organisms,
a separate study, using large-volume samples, was performed in a water treat-
ment plant and its associated distribution system. The results of this
-------
investigation confirmed the indicator organism resistance relative to coli-
forms and, indirectly, showed that a well operated water treatment plant is
capable of achieving a moderate degree of removal of acid-fast and yeast
organisms.
Since the densities of the proposed indicator organisms observed in
field studies were generally low, it seemed desirable to confirm that removal
of the organisms in water treatment unit operations, prior to disinfection,
was not so great as to produce negligible indicator populations at the influ-
ent to the disinfection process. Laboratory coagulation-flocculation studies
using lime, alum, or ferric chloride as coagulants, and sand filtration
experiments were performed.
To verify the resistance of the proposed new indicator organisms to free
and combined chlorine, continuous flow chlorination studies were performed
using mixed pure-cultures in CDFB; diluted secondary wastewater effluent was
also used. Results confirmed the data from batch studies in which increased
resistance of the proposed indicator organisms to free and combined chlorine
was observed, as compared to E. coti.
Because ozone is frequently used as a disinfectant for water and waste-
water in European countries, and may be a substitute for chlorine in the U.S.
under certain conditions, studies were performed to determine the ozone sensi-
tivity of the proposed indicator organisms, E. coti, poliovirus type 1
(Mahoney), and S. typhMnm>vium. While the acid-fast organism exhibited maxi-
mum resistance to ozone, the yeast organism appeared to be less resistant
than poliovirus.
Experiments to determine the mechanism of inactivation of E. call and
the proposed new indicator organisms by free chlorine were also undertaken.
-------
SECTION 2
CONCLUSIONS
Additional information has been presented regarding the potential use
of acid-fast organisms and/or yeasts as indicators of chlorination efficiency.
In a study of a wastewater treatment plant (WWTP) at St. Joseph, Illinois,
its associated receiving stream, and a downstream water treatment plant (WTP)
at Oakwood, Illinois, the proposed indicator organisms were enumerated. Com-
pared with the coliform group, the yeast and acid-fast organisms were found
to be more resistant to secondary wastewater treatment and chlorination. A
24-hr study to determine the diurnal variation based upon organism density
in the WWTP revealed that yeast and acid-fast organisms were consistently
present at a density of approximately 104/100 mi in raw wastewater. Fecal
and total coliform densities in the raw wastewater were consistently higher
than the proposed indicators by an order of 2-2.5 and 3 logs, respectively.
Reduction of organism density as a result of chlorinating the secondary waste-
water effluent was observed to be in the range of 1-3.5 logs. The relative
resistance of the four organism groups to chlorination was found to be:
acid-fast > yeast > total coliforms > fecal coliforms.
The occurrence of these four groups in the receiving stream for the WWTP
effluent was also monitored. It was found that the wastewater effluent did
not significantly change the quality of the water in the receiving stream.
Physical-chemical parameters monitored above and below the effluent outfall
supported this conclusion. Actual organism densities in the stream exhibited
seasonal variations, being generally lower during colder periods than during
warmer periods.
Recovery of all organisms from a holding reservoir, serving as the raw
water supply of the WTP, was much lower than their respective densities in
the river system supplying the reservoir. The majority of organisms in the
raw water were removed by coagulation/flocculation/prechlorination within the
WTP- Further reductions in organism density were observed after sand filtra-
tion. Yeasts were not recovered from any finished water samples in this WTP;
acid-fast organisms were recovered from only one out of seven finished water
samples.
One-liter volume sampling for enumeration of the proposed indicator
organisms, yeasts and acid-fast bacteria, was found to be feasible. Tech-
niques were developed to allow membrane-filtration of 1 £ samples of both
raw and treated drinking water at a water treatment plant. Utilizing these
techniques, the water treatment plant at Decatur, Illinois, and its associ-
ated distribution system was studied. Yeasts and acid-fast organisms were
-------
detected in all raw water samples, as were total coliform organisms. The
density of yeasts averaged 5.7/1 in raw water samples and 1.5/£ in finished
(treated) water. Organism densities of acid-fast bacteria averaged 37.0/£ in
raw water samples and 2.2/1 in finished water. The proposed indicators were
recovered from finished water samples with a frequency of 7 percent for
yeasts and 20 percent for acid-fast organisms. Estimated standard plate
counts were also performed and were found to be approximately 2 x 104/m£ 1n
raw water samples and approximately 7 x 103/m£ in finished water samples.
No significant correlation was found between the results of the standard
plate count and the density of the other organisms determined, i.e., coli-
forms and the proposed indicator organisms.
Continuous flow chlorination, using mixed pure-cultures and diluted
secondary wastewater effluent, confirmed the increased resistance of the
proposed indicator organisms to free and combined chlorine, as compared to
E. coti or coliforms. Experiments performed at pH 7 and 10 confirmed that
greater yeast and acid-fast inactivation rates were achieved at the lower pH,
thus supporting the claim that hypochlorous acid is a more effective disin-
fectant than hypochlorite ion. The relative resistance of these organisms
to chlorination was found to be: acid-fast > yeast > coliforms.
Similar experiments were performed under breakpoint chlorination condi-
tions. It was again observed that acid-fast organisms were the most resis-
tant of the three organism groups studied, as evidenced by their ability to
occasionally survive the severe conditions of breakpoint chlorination.
A step-wise regression analysis of the data from these studies showed
that the independent variable of chlorine residual could explain greater
than 85 percent of the variance in the inactivation of each of the test
organisms.
Experiments were performed to study the reductions in the density of the
proposed indicator organisms using clarified activated sludge effluent which
had been passed through a granular activated carbon column. Yeast organisms
were reduced approximately 8 percent, while coliforms and acid-fast organisms
were each reduced by approximately 4 percent.
Laboratory coagulation/flocculation studies using alum or ferric chlor-
ide as the coagulant showed that removal of acid-fast and yeast organisms
was similar to that for coliforms. In using lime, removal of yeasts was less
than that observed for total coliforms. Studies on sand filtration indicated
that acid-fast organisms were removed to the same extent as the coliforms,
whereas yeasts showed a higher degree of removal than the coliforms. By
optimizing coagulation conditions for turbidity removal, it was found that
total coliforms, yeasts, and acid-fast organisms could be removed from inocu-
lated water in the range of 90-99 percent. The percent removal of these
organisms by sand filtration ranged from 40-70 percent for acid-fast and
total coliform organisms, and was greater than 90 percent for yeasts.
Studies of organism inactivation by ozone were performed with acid-fast
and yeast organisms, E. coti, S. typkimusu.wm, and poliovirus type 1 (Mahoney)
7
-------
Various parameters were analyzed to determine the extent of their influence
on ozonation efficiency. Experiments performed with C. paSLapA-LtoA+A and M.
fiositiiitum showed that the presence of ozone bubbles in addition to ozone
residual was more effective in inactivating the test organisms than ozone
residual alone. Ozone bubbles alone, without any ozone residual, were also
found to cause a slight amount of inactivation. In a study on the effect of
mixing rate in the ozone contact chamber, it was found that as the rate of
agitation increased there was a higher decomposition rate of ozone and, con-
sequently, a lower organism inactivation. Variations in pH did not signifi-
cantly affect percent organism survival under constant ozone residuals.
Experiments to study the effect of temperature indicated that the degree of
inactivation of M. ^o^tu^Ltum increased significantly with an increase in tem-
perature, at constant ozone residual. The relative resistance of test organ-
isms to disinfection with ozone was found to be: M. fiofctu-Ltum > poliovirus >
C. pasiap&JJLo&4A > E. c.otL > 5.
Studies performed to determine the mechanism of organism inactivation
due to the action of free available chlorine demonstrated that the primary
mode of action was cellular penetration at the level of the outer cell wall.
This was additionally followed by disruption of cell permeability, secondary
metabolic disturbances, and lethal lesions to the deoxyribonucleic acid (DNA).
It was concluded that the increased resistance of C. paJvpaA and
M. fiofctu-itm to free available chlorine was a result of their cellular physi-
ology, principally the thickness and rigidity, and impermeability, respec-
tively, of their cell walls.
Information to date concerning the resistance of the proposed new indi-
cator organisms, yeasts and acid-fast bacteria, to various forms of chlorine
and the distribution of these organisms in water and wastewater indicates
their potential use as indicators of chlorination efficiency.
-------
SECTION 3
RECOMMENDATIONS
1. A large scale field sampling program to determine the densities of
acid-fast and yeast organisms, as well as coliforms, and chlorine resis
tant pathogens, e.g., amoebic cysts, enteric viruses and bacteria, is
required in order to develop criteria of disinfection efficiency based
upon the proposed new indicator organisms with respect to the desired
degree of microbial purity of a finished water supply or a treated
wastewater effluent.
2. Providing the proposed indicator organisms continue to appear useful as
a means of determining disinfection efficiency, the selective growth
medium for the enumeration of acid-fast organisms and yeasts should be
reexamined with the objective of reducing the incubation time required,
or alternatively, rapid enumeration techniques should be investigated
which enable the determination of the indicator organism densities
within 24 hr.
3. If either the yeast or acid-fast group of organisms is accepted as a
new indicator of disinfection efficiency, studies to determine the
species of organisms most commonly found in water and wastewater should
be performed to better define the nature of the indicator group.
-------
SECTION 4
MATERIALS AND METHODS
INTRODUCTION
This section covers the biological, physical, and chemical techniques
used in both the field and laboratory experiments related to the presence of
the proposed indicator organisms and their role as indicators of disinfection
efficiency by chlorine and ozone. Also included in this section are the
methods of preparing the pure cultures of the yeasts and acid-fast organisms
used in the laboratory experimental studies. The enumeration techniques for
acid-fast organisms, yeasts, fecal coliforms, E. coti, S. typkunusu.wm, and
poliovirus have been reported in detail in earlier publications (5,6). In
this section, modifications in the selective methods made since last reported
shall be discussed.
SOURCE OF TEST ORGANISMS
The M. fiotLttM-twm, M. phJLtLi, and C. poA.ap^-c£.o4^4 test organisms used in
this study were isolated from wastewater by Engelbrecht oJL o£.(4). Polio-
virus type 1 (Mahoney) and the cell line used (African Green Monkey Kidney
[BGM] cells) were obtained from Dr. Gerald Berg, U.S. EPA, Cincinnati, Ohio.
The pure cultures of 5. typhimuJvium and E. coti were from the collection of
the Department of Microbiology, University of Illinois at Urbana-Champaign.
ORGANISM ENUMERATION TECHNIQUES
Total Plate Count, Total Col i forms, Fecal Col i forms,
E. co Li, and S. typhJjnusu.wm
The methodology used for the enumeration of total coliforms, fecal colv
forms, E. coti, and S. typkimusiium has been previously described (5). The
determination of total plate count bacteria was performed according to
Standard Methods (7).
Poliovirus
The enumeration method used for poliovirus has been described in
detail (5).
10
-------
Yeast
Depending on whether field or laboratory samples, yeasts were enumerated
according to the procedures described earlier (5).
Acid-Fast Organisms
The methodology for enumerating acid-fast organisms in pure culture
studies (laboratory studies) has been described (5,6). The standard enumera-
tion technique for acid-fast organisms, used with mixed cultures or field
samples as reported earlier, consisted of pretreating samples with 2.5 per-
cent oxalic acid in a 1:1 volume ratio for 10 min and neutralizing with
2 percent sodium hydroxide (5,6). The samples were then membrane-filtered,
with membranes being placed on Middlebrook 7H9 broth base (BBL, Cockeysville,
MD) enriched with oleic acid-albumin-dextrose-catalase (OADC; Difco Labora-
tories, Detroit, MI), propionate, and antibiotics in a culture dish and
incubated at 37°C for 72 hr. The filters were then removed from the agar
surface, heat dried, and stained by means of the Brook's acid-fast stain.
Colonies which appeared pink to red when viewed under a dissecting microscope
were counted as acid-fast organisms.
Studies were performed in an attempt to simplify the pretreatment steps
required in the enumeration of acid-fast organisms as described above. Two
modifications were evaluated using both a pure culture of M. ^o^itaitum sus-
pended in pH 7 buffered deionized-distilled water and the natural population
of acid-fast organisms occurring in raw wastewater, diluted with pH 7
buffered deionized-distilled water.
Modification 1. The untreated sample was filtered and an equal volume
of 2.5 percent oxalic acid was added and allowed to
remain on the filter by disconnecting the vacuum.
After 10 min, the oxalic acid was filtered and the
membrane filter rinsed with pH 7 buffer.
Modification 2. The sample was treated with 2.5 percent oxalic acid
(1:1 volume ratio) in a flask and then filtered after
10 min contact time, without sodium hydroxide neutra-
lization.
Results of these modified procedures are presented in Tables 1 and 2.
As indicated, no statistically significant difference among the results
obtained by use of the three treatments was observed for either the waste-
water or the pure culture suspended in buffered deionized-distilled water.
Figure 1 is a diagram of the revised enumeration procedure for acid-fast
organisms.
An investigation was also performed in an attempt to more easily pro-
cess large volume samples containing low organism densities by increasing
the concentration of oxalic acid stock solution, with a parallel decrease in
the volume of the filtrate produced. Results of these experiments are pre-
sented in Table 3 and indicate a statistically significant difference in
11
-------
TABLE 1. ACID-FAST PRETREATMENT METHODOLOGY, SAMPLE: M.
PURE CULTURE IN BUFFERED DEIONIZED-DISTILLED WATER
RAW DATA
Replicates
Mean
Standard Deviation
Counts per 100 mi
Pretreated on Pretreatment in
filter - NaOH flask - NaOH
omitted omitted
78, 85, 62, 78, 108, 77, 65
59, 91, 71 62, 84
74.3 79.0
12.7 16.5
sample
Standard
Pretreatment
76, 83, 78,
69, 93, 56
75.8
12.6
ANALYSIS OF VARIANCE
Source of
variation
Among Treatments
Within Treatments
Degrees of Sum of
freedom squares
2 68.11
15 2950.19
Mean
squares F
34.06 0.17*
196.68
Total 17
3018.3
Not significant (p > 0.05)
counts/100 ml with concentrations of oxalic acid greater than 2.5 percent.
It may also be noted that the average organism recovery decreased with an'
increasing concentration of oxalic acid.
The standard pretreatment technique, described earlier, with neutrali-
zation by sodium hydroxide was used to study the frequency of occurrence of
acid-fast organisms at the St. Joseph wastewater and Oakwood water treatment
plants and in the Salt Fork of the Vermillion River as well as in the coaqu-
lation-flocculation and sand filtration studies. Pretreatment Modification 2
was used in the distribution study at Decatur in the continuous flow chlori-
nation study. The pretreatment step and the addition of antibiotics and
selective agents to the medium was omitted in performing the studies related
to mechanism of inactivation of acid-fast organisms by chlorine.
PREPARATION OF PURE CULTURES (Laboratory Studies)
Acid-Fast Organisms
Acid-fast organisms (M. ^oituAtum and M. phlex.) were grown in
12
-------
TABLE 2. ACID-FAST PRETREATMENT METHODOLOGY,
SAMPLE: 0.5% RAW WASTEWATER
RAW DATA
Counts per 100 mi sample
Replicates
Mean
Standard Deviation
Source of
variation
Among Treatments
Within Treatments
Total
Pretreated on
filter - NaOH
omitted
163, 144, 154,
128, 154
148.6
13.3
ANALYSIS
Degrees
freedom
2
14
16
Pretreatment in
flask - NaOH
omitted
190, 188, 165,
176, 158, 158
172.5
14.4
OF VARIANCE
of Sum of
squares
1559.42
3906.7
5466.12
Standard
pretreatment
138, 157, 166,
199, 150, 156
161.0
20.8
Mean
squares F
779.71 2.79*
279.05
Not significant (p > 0.05)
Middlebrook 7H9 medium for 36 hr at 37°C in a water shaker bath and then
centrifuged in a Sorval GLC-2 (DuPont Instruments, Newton, CT) general labor-
atory centrifuge at 1800 rpm for 15 min. In the ozone studies, the incuba-
tion time was 72 hr. The pellets were then washed twice with a total of at
least 150 ml of pH 7 phosphate buffer. Cells prepared in this manner were
resuspended in phosphate buffer and were kept at 4°C until needed. Before
the organisms were used, the optical density was measured at 660 nm using
cuvetts with a 1 cm light path in a Bausch-Lomb Spectronic 20 colorimeter
(Bausch-Lomb, Inc., Rochester, NY). The optical density was correlated with
a previously developed extinction coefficient. The concentration of cells
was calculated as the ratio of the optical density to the extinction coeffi-
cient, times the length of the light path.
Yeast
The yeast culture (C. paJiapA-iloA-U} was grown in modified yeast extract-
malt extract medium for 24 hr at room temperature on a shaker. In the case
of the ozone studies, incubation was for 48 hr. Centrifugation, washing,
and subsequent determination of organism density was performed by the same
procedure as described for acid-fast organisms.
13
-------
Sample
Homogenization
Waring Blender (30 sec)
or agitation
Pretreatment*
1.25% Oxalic Acid (10 min)
4-
Dilution
Phosphate Buffer, pH 7
Membrane Filtration
Cultivation
7H9 Middlebrook
(plus OADC, propionate, and antibiotics)
4-
Incubation
72 hr - 37°C
Staining*
Heat fix colonies on membrane
Brook's carbol fuchsin (primary stain)
10% oxalic acid rinse
Brook's malachite green (counterstain)
Distilled water rinse
4-
Enumeration
Dark pink to red colonies
*
These steps were omitted in the laboratory studies using pure acid-fast
cultures.
Figure 1. Revised Enumeration Technique for Acid-Fast Organisms.
14
-------
TABLE 3. ACID-FAST PRETREATMENT METHODOLOGY USING DIFFERENT
CONCENTRATIONS OF OXALIC ACID, SAMPLE: 0.5% RAW WASTEWATER
RAW DATA
Counts per 100 mi
Replicates
Mean
Standard Deviation
Source of
variation
Among Treatments
Within Treatments
Total
Pretreated with
10% oxalic acid
129, 108, 64,
71, 94, 106, 95,
97, 83, 59
90.6
21.69
ANALYSIS OF
Degrees of
freedom
2
27
29
sample
Standard
Pretreated with pretreatment
5% oxalic acid 2.5% oxalic acid
102, 48, 45,
122, 96, 56,
137, 149, 49,
108
109.2
26.03
VARIANCE
Sum of
squares
4,956.47
17,052.90
22,009.37
175, 125, 159,
97, 118, 89,
127, 117, 95,
117
121.9
27.33
Mean
squares F
2,478.24 3.924*
631.59
Significant (p > 0.05)
Coti
Preparation of E. coti suspensions was performed by inoculating a young
culture of the organism into a flask containing nutrient broth and incubating
overnight at 37°C in a shaker water bath. Centrifugation, washing, and deter-
mination of cell density was carried out by the same procedure as that for
acid-fast organisms.
The method followed for preparation of S. typlrimu^ium was the same as
that for E. c.oti. However, the incubation period was 48 hr.
Poliovirus
Poliovirus type 1 (Mahoney) was grown on monolayers of BGM cells.
Infected cells, along with the suspending medium, were frozen at -70°C and
thawed twice, then centrifuged for 1 hr at 10,000 rpm in a Beckman L2-65B
15
-------
ultracentrifuge (Beckman Instruments, Inc., Palo Alto, CA). The supernate
was decanted and recentrifuged at 50,000 rpm for 2 hr. The pellet from the
second centrifugation, containing the virus, was resuspended in 10 to 20 ml
of buffered deionized water and kept overnight at 4°C, after which the virus
suspension was again centrifuged at 50,000 rpm for 2 hr. The virus was
finally resuspended in 150 ml of phosphate buffer distributed in small an-
quots, and then frozen and stored at -70°C for use in subsequent experiments
The virus was thawed and diluted before use.
PHYSICAL AND CHEMICAL METHODS
Temperature
Temperature was recorded by mercury-filled centri grade thermometers in
laboratory experiments and by a temperature measuring probe attached to a
YSI Model 54 DO analyzer (Yellow Springs Instrument Co., Yellow Springs, OH)
in the field studies.
The electrometric method, using glass and reference electrodes with a
Beckman Electromate pH meter (Beckman Instruments, Inc., Palo Alto, CA), was
used to measure pH.
Dissolved Oxygen
Dissolved oxygen was measured by means of a YSI Model 54 DO meter
(Yellow Springs Instrument Co., Yellow Springs, OH).
Total Suspended Solids
Total suspended solids were determined according to the procedure stated
in Standard Methods using glass fiber filter discs (7).
Total Organic Carbon and Chemical Oxygen Demand
Total organic carbon (TOC) measurements were made with a Beckman Model
915 TOC analyzer (Beckman Instruments, Inc., Palo Alto, CA) using the two-
channel method, as given in Standard Methods (7). Chemical oxygen demand
(COD) was measured according to the procedure given in Standard Methods (7).
Turbidity
Turbidity was measured in a Hach Turbidimeter Model 2100A (Hach Chemical
Co., Ames, IA) and reported in nephelometric turbidity units (NTU).
Chlorine Residuals
Chlorine residuals were measured by the N,N-diethyl-p-phenylene-diamine
(DPD) method in accordance with Standard Methods (7) for the batch
16
-------
chlorination experiments and field studies. The measurement of chlorine
residuals in the continuous chlorination and in the mechanism of inactivation
studies was determined by the amperometric method as described in Standard
Methods (7) using an amperometric titrator (Series A-790, Wallace and Tiernan,
Belleville, NJ).
Ozone Residual
Ozone residual in the aqueous phase was determined by using the UV spec-
trophotometric method of Shechter (8).
Chlorine-Demand Free Buffer
Chlorine-demand free buffer (CDFB) was prepared as described in a pre-
vious report (5). Deionized-distilled water was chlorinated by adding sodium
hypochlorite to achieve a free chlorine residual of approximately 3 mg/l, and
stored at room temperature for one week. The water was then divided into 5 £
portions and the appropriate quantity of phosphate or borate salt was added
to each portion to achieve the desired pH. Each portion was subsequently
boiled for sterilization and then placed under UV light for at least 72 hr
for final dechlorination. The CDFB was then ready for use.
FIELD METHODS
Occurrence of Indicator Organisms in Water and Wastewater
A detailed field survey of the densities of acid-fast organisms, yeasts,
fecal coliforms, and total coliforms was undertaken. The location selected
for this survey was the St. Joseph, Illinois, wastewater treatment plant,
its receiving stream, the Salt Fork of the Vermillion River, and the Oakwood,
Illinois, water treatment plant, which uses the Salt Fork as a water supply.
The wastewater treatment plant at St. Joseph, Illinois, serves a popula-
tion of 1850. A combined sewer system carries domestic wastewater and storm
runoff to the treatment plant. A schematic of the plant is given in Figure 2.
Raw wastewater flows into an inlet chamber for removal of grit and is then
fed into two Model R Oxigest Tanks (Smith and Loveless Co., Lenexa, KS) which
are operated on the contact stabilization principle. Waste sludge is aerob-
ically digested and then lagooned. The secondary effluent is chlorinated
prior to discharge into the Salt Fork of the Vermillion River. The operating
characteristics of the treatment plant for the period March 1975 to January
1976 are summarized in Table 4.
As noted in Figure 2, there were three sampling locations within the
treatment plant. SJ1 was the de-gritted wastewater immediately upstream from
the aerator influent. SJ2 was the activated sludge effluent prior to chlori-
nation, and SJ3 was the chlorinated effluent immediately prior to discharge
to the receiving stream. Grab samples were used for analysis and samples
were taken without compensating for hydraulic detention time.
17
-------
Digested Sludge
Disposal to
Lagoon
Raw Wastewater
SJI
Inlet- Chamber
with Comminutor
Mixing
Aeration
Zone
00
Chlorine
_ J
SJ3 Treated
Effluent
Chlorine Contact
Chamber
Wastewater
— Sludge
Sampling Location
Figure 2. St. Joseph, Illinois, Wastewater Treatment Plant
-------
TABLE 4. OPERATING CHARACTERISTICS OF THE ST. JOSEPH, ILLINOIS WASTEWATER
TREATMENT PLANT (March 1975 - January 1976)
Parameter
3
Flow, m /day
Design capacity
Actual
Range
-
Mean
1135
662
BODr, mg/£
Influent
Effluent
Suspended Solids, mg/l
Influent
Effluent
Chlorine, mg/£
Dosage
Total Residual
310-460
3-9
330-560
2-6
3.5-4
0.6-1.1
380
5
400
4
0.8
The Oakwood, Illinois, water treatment plant serves a population of
1600. A schematic diagram of the plant is shown in Figure 3. Raw water is
pumped from the Salt Fork of the Vermin ion River, at a point approximately
59.5 km downstream from the St. Joseph wastewater treatment plant discharge,
into a 5677.5 m3 reservoir adjacent to the treatment plant as it is needed.
the reservoir is two weeks; however, short cir-
summer months, copper sulfate is added to the
at a dose sufficient to produce a final copper
1 mg/£. Operating characteristics of the
A typical turnover time in
cuiting may occur. During
reservoir to control algae,
concentration of approximately
plant are shown in Table 5.
As noted in Figure 3, four sampling locations within the treatment plant
were used. OW1 was located at the head end of the raw water reservoir. The
OW2 sample was taken from the overflow of the clarifiers. OW3 was located
immediately after sand filtration, and the OW4 sample, the finished water,
was taken at a service tap in the treatment plant.
The Salt Fork of the Vermillion River was sampled at locations between
the U.S. Highway 150 bridge, upstream of the St. Joseph wastewater plant,
and the Oakwood water treatment plant intake. The approximate locations of
the sampling stations are noted in Figure 4. Table 6 gives an exact descrip-
tion of the sampling sites used. It should be noted in Table 6 that sample
station RSO is located above the outfall of the St. Joseph wastewater treat-
ment plant, whereas the remaining five sample stations are all located below
the outfall.
19
-------
Copper Sulfate in Summer
ro
o
Water from
Salt - Fork
OWI
•
L
Alum
Lime
Chlorine
Holding
Reservoir
Sand
Filters
LLP
Chlorine
OW3
4-*-
Clear Wei
OW4
—•—
To
HLP Distribution
System
Filter Back Wash Pump
LLP - Low Lift Pump
Legend( HLP -High Lift Pump
• -Sampling Location
Figure 3. Oakwood, Illinois, Water Treatment Plant
-------
TABLE 5. OPERATING CHARACTERISTICS OF THE OAKWOOD, ILLINOIS, WATER
TREATMENT PLANT (March 1975 - January 1976)
Parameter Range Mean
3
Design Flow Capacity, m /day 545.2
Actual Flow, m3/day 321.8
Alum Dosage, kg/day 13.6
Lime Dosage, kg/day 36.3
Filtration Rate, £/min/m2 101.8
Length of Filter Run, days 2
Terminal Chlorination Dosage, mg/£ 7.0
Finished Water
Total Hardness, mg/l 180-270 220
Total Alkalinity, mg/£ 100-180 150
pH 7.0-8.6 7.8
Turbidity, NTU 1-9 2.0
Total Residual Chlorine, mg/l 0.6-1.5 1.0
The Salt Fork of the Vermin ion arises in east Champaign County,
Illinois, at an altitude of 243.8 m above sea level. It flows toward the
southeast and joins the Middle Fork of the Vermillion River southwest of
Danville, Vermillion County, Illinois. The Vermillion River joins the
Wabash River near the Illinois-Indiana state line (9).
The discharge from the city of Rantoul municipal wastewater treatment
plant enters the Upper Salt Fork drainage ditch. Wastewater discharge from
the Chanute Air Force Base enters the same drainage ditch approximately
3.2 km below the Rantoul outfall. Effluent from the East Side Treatment
Plant of the Urbana-Champaign Sanitary District, and urban runoff from the
cities of Urbana and Champaign enter the Saline Branch on the northeast edge
of the city of Urbana. The Saline Branch subsequently joins the Upper Salt
Fork approximately 1.6 km west of the village of St. Joseph to form the Salt
Fork of the Verrnillion River. Wastewater effluent from the St. Joseph waste-
water treatment plant is discharged into the Salt Fork approximately 1.6 km
21
-------
ro
ro
Upper Salt Fork
Drainage Ditch
Champaign
County
Conservation
Area
4 8
Scale , kilometers
Figure 4. Salt Fork of the Vermillion River and Sampling Stations
-------
TABLE 6. DETAILED DESCRIPTION OF THE SALT FORK OF THE
VERMILLION RIVER SAMPLING STATIONS
Site designation
Description
RSO U.S. Highway 150 bridge over the Salt Fork, west of
St. Joseph
RSI Bridge over the Salt Fork, 2.4 km south of St. Joseph,
approximately 88° 03' 00" and 40° 05' 30"
RS2 Bridge over the Salt Fork, 1.6 km north of Sidney, Illinois,
approximately 88° 04' 30" and 40° 02' 20"
RS3 Bridge over the Salt Fork, 4.8 km west of Homer, Illinois,
approximately 88° 00' 30" and 40° 02' 50"
RS4 Bridge over the Salt Fork, 4.8 km northeast of Homer,
approximately 87° 54' 30" and 40° 04' 20"
RS5 Bridge over the Salt Fork, 3.2 km south of Oakwood, on
Federally Aided Secondary Highway 331
south of the confluence of the Saline Branch and the Upper Salt Fork. Other
than field tile drainage nets, which are common in the east central Illinois
agricultural area, there are no other known continuous discharges into the
Salt Fork between St. Joseph and Oakwood. However, diffuse sources of waste-
water, largely untreated, may enter this stretch of the stream from other
communities located within the drainage basin.
The sampling program was initiated in March, 1975. Two sets of samples,
collected in March and April, 1975, were used to establish the sampling pro-
cedures and protocol. River stations RSI and RS5 were not included in these
two sets of samples. Data for the first two sets have not been included in
the statistical analyses.
Initially, a set of samples was collected and analyzed twice monthly.
After reviewing data of the first four sets, it was decided to reduce sam-
pling frequency to monthly. No sampling could be performed during December
1975 due to the extremely poor weather conditions; however, two sets of sam-
ples were collected in January 1976. This report considers the data obtained
during the period May 1975 through January 1976. Eleven sets of samples were
collected and analyzed during this nine-month period.
All stream water samples for microbiological analysis were collected
5-10 cm below the water surface in sterile 1 I plastic bottles; usually the
sample was collected by lowering the freshly opened bottle from a bridge.
23
-------
Samples were collected from the Oakwood and St. Joseph treatment plants from
the points previously described. All samples were dechlorinated by the addi-
tion of 1 mt of 0.05 N sodium thiosulfate per liter of sample immediately
after collection.
Sample collection was initiated at RSO between 0900 and 1000 hr; samples
were then collected at the St. Joseph wastewater treatment plant, i.e.,
RS1-RS5, and finally at the Oakwood water treatment plant. Immediately
after collection, the dechlorinated samples were stored in an ice chest.
Samples were transported back to the laboratory by 1200 hr, and it was gen-
erally possible to complete the analyses by 1700 hr.
During the 24-hr field study conducted at the St. Joseph wastewater
treatment plant, samples were collected at 1200, 1500, 1700, 1900, 2200,
0100, 0400, 0700, and 1200 hr. Predetermined dilutions of the samples were
membrane-filtered on site, and placed on the appropriate agar medium and
refrigerated; the prepared samples were then transported back to the labora-
tory for incubation and enumeration.
Occurrence of Indicator Organisms in Distribution Systems
Another field survey was performed at Decatur, Illinois, South Side
Water Treatment Plant between February 8 and September 11, 1977. This survey
involved the enumeration of acid-fast, yeast, and total coliform organisms
as well as the standard plate count in the raw water, in the finished water
at the plant following treatment, and in treated water at various locations
in the distribution system. A total of 18 sets of samples were collected
during the study period; however, this report summarizes data from only the
last 14 sets of samples. The first four sets were used to refine and modify
the laboratory procedures for processing of the samples.
The South Side Plant is one of two water treatment plants serving a
population of 95,000 in the city of Decatur. Water is withdrawn .from Lake
Decatur, a man-made impoundment on the Sangamon River. Treatment consists
of alum coagulation-flocculation, sand filtration, and both pre- and post-
chlorination. The water is softened by only a small degree. Plant operating
data for the South Side Plant for the period April 1977 through October 1977
are summarized in Table 7.
A major distirubtion line extends for approximately 6.4 km to the west
of the South Side Water Treatment Plant. This water main was selected for
sampling purposes because it is supplied by water treated almost exclusively
by the South Side Plant, and because it terminates in a deadend. The North
Side Water Treatment Plant distribution network does interconnect with the
previously described water main in the vicinity of sample station 4, and
this may have resulted in a partial mixing of water from each treatment plant
in samples taken from stations 4 and 5. A schematic map showing the location
of sampling points is given in Figure 5, and a more detailed description of
each sample station is given in Table 8.
24
-------
TABLE 7. OPERATING DATA FOR THE SOUTH SIDE WATER TREATMENT
PLANT, DECATUR, ILLINOIS
Parameter _ Average value _
3
Flow, m /day
Design capacity 113,550
Actual 56,775
Turbidity, NTU
Raw water 21
Pre-filtration 6.3
At plant tap (finished water) 0.6
Temperature, °C
Raw 69
At plant tap (finished water) 76
Raw 8.0
At plant tap (finished water) 8.5
TABLE 8. LOCATION OF SAMPLING STATIONS, DECATUR, ILLINOIS
Sampling Station Location
1 Raw water sampled at the South Side Plant
2 Finished water sampled at a service tap of the South Side
Plant
3 Fire station near Main and Franklin Streets
4 Bowling alley at Fairview Avenue and Eldorado Streets
5 Holiday Inn on Eldorado Street
25
-------
Eldorado
No. 4
t
h-
Street
v
No. 5 , Holiday Inn
Sangamon
-n
Q
•- Distribution Main
Sampling Station
'No. 3
Nos. I and 2 ,
South Treatment
Plant
Figure 5. Water Treatment Plant and Distribution System with Sampling Stations, Decatur, Illinois
-------
A sample of approximately 8 I was taken at each station by personnel of
the city of Decatur and immediately dechlorinated with sodium thiosulfate.
At the same time, a field determination of free available and total chlorine
residual was made using the modified DPD method (LaMotte Chemical, Chester-
town, MD). The samples were then transported to the laboratory for further
analysis. All analyses were completed within 8 hr of the time of sample
collection.
Total coliform organisms were enumerated by both membrane filtration
(MF) and most probable number (MPN) techniques (7). With the MF technique,
raw water samples were analyzed using triplicate filtrations of three dif-
ferent dilutions, selected so as to yield between 20 and 80 colonies per
plate. Treated water was uniformly filtered in duplicate 100 mi volumes.
Colony counts were geometrically averaged and reported as "number/100 mi."
Analysis of the water for coliforms by the MPN method consisted of five tubes
for each of three different dilutions. Treated water was analyzed using a
5-tube, single-dilution design. Coliform densities have been reported as
"MPN index/100 ml," using the MPN tables in Standard Methods (3).
Acid-fast organisms and yeasts were enumerated using the techniques pre-
viously described (5). Duplicate 1-1 sample volumes from each station were
membrane filtered; colony counts at each station were geometrically averaged
and reported as "number/^."
Standard plate count enumeration was performed on every sample using
duplicate plates for each of three different dilutions. Colonies were
counted on a Darkfield Quebec Colony Counter (Fischer Scientific, Pittsburgh,
PA), and the "estimated standard plate count/m£" was calculated as described
in Standard Methods (7).
UNIT OPERATIONS/UNIT PROCESSES LABORATORY METHODS
Continuous Flow Chlorination Experiments
Continuous flow experiments were designed to determine the chlorine
resistance of the proposed indicator organisms and coliform organisms under
various idealized water and wastewater chlorination conditions. To simulate
the conditions used in actual water and wastewater chlorination practice, a
tubular reactor was fabricated using a plexiglass tube of approximately 3.9 cm
internal diameter and 183.5 cm length. The overall theoretical detention
time of the reactor was approximately 20 min for a flow rate of 100 m£/min.
Additional sampling ports corresponding to detention times of 4, 8, 12, and
16 min were located along the longitudinal axis of the reactor. A schematic
diagram of this reactor is shown in Figure 6. The experimental arrangement
for all continuous flow experiments was such that the influent flow to the
reactor was a combination of both the solution containing the test organisms
and the free chlorine feed solution, with mixing of these two flows occurring
just prior to their entry into the reactor. The total combined influent flow
rate was controlled at approximately 100 m£/min, thus resulting in the reac-
tor effluent detention time of approximately 20 min. To verify the plug flow
27
-------
00
CL Soln.
Organisms
-Sampling
Ports \i
Figure 6. Continuous Flow Reactor for Chlorine Inactivation Studies
-------
nature of the reactor, dye tracer studies were made. Figures 7 and 8 show
typical results of these tracer studies. These figures show that plug-flow
conditions prevailed in the reactor.
The continuous flow experiments performed using the above described
reactor consisted of two basic types: 1) mixed pure culture inactivation
studies, and 2) natural population inactivation studies. In the mixed pure
culture chlorination studies, experimental conditions were established to
assure the observation of several logs of organism inactivation, as well as
a low chlorine demand with a reasonable free chlorine concentration. These
studies were made with a laboratory-prepared "mixed culture" of the yeast,
acid-fast, and coliform organisms. This mixed culture of test organisms was
suspended in CDFB and fed to the reactor at a flow rate of approximately
25 m£/min, along with the free chlorine solution at a flow rate of approxi-
mately 75 m£/min. The reactor was allowed to flow full with the combined
test solutions for a minimum of two reactor detention times before test sam-
ples were collected. Experimental test samples from the reactor, at all but
the reactor effluent port, were collected by means of syringes inserted
through the membrane-covered ports located along the reactor length so as to
give the desired retention times of approximately 4, 8, 12, and 16 min.
Samples having a retention time of 20 min were collected in sterile flasks
by means of plastic tubing connected to the end of the reactor.
In all test samples collected, 0.05 N sodium thiosulfate was added
immediately to destroy the chlorine residual. Culturing and enumeration
techniques for the acid-fast, yeast and coliform organisms were performed
as previously described for the laboratory studies (5). the amperometric
method was used to measure the free chlorine residual of both the chlorine
feed solution and the reactor effluent.
The natural population studies were made using the clarified activated
sludge effluent of the East Side Wastewater Treatment Plant of the Urbana-
Champaign Sanitary District as the organism feed solution. The East Side
Plant consists of primary treatment followed by parallel secondary treatment,
i.e., activated sludge and trickling filter units, clarification and chlori-
nation. It had previously been shown that acid-fast, yeast, and coliform
organisms were routinely present in both the plant's raw influent and in the
activated sludge effluent (5).
The sample of activated sludge effluent for each natural population
experiment was collected at the point of overflow from the activated sludge
clarifier to the chlorination contact tanks. A sufficient quantity of eff-
luent, approximately 12 I, was collected to allow each experiment to be com-
pleted using a single sample and to allow a physical and chemical analysis
to be made. Each sample of effluent was analyzed for pH, temperature, TOC,
suspended solids, turbidity, and occasionally COD.
These chlorination experiments were performed in a manner similar to
the "mixed culture" studies with free chlorine, but several notable proced-
ural changes were made. The influent flow to the reaction chamber was again
approximately 100 m£/min with a resulting effluent detention time of
29
-------
1.00
0.75
Q>
U
O
U
c
O
O
±f 0.50
c
U
O
U
O
O
0.25
Tirne/Theo. Dentention Time
Figure 7. Step-Input Dye Tracer Study Performed on Continuous Flow
Inactivation Reactor
30
-------
0.5
0.4 —
O»
o
o
c
o
o
0)
o
o
0.3
0.2
20
40
Time (min)
60
80
Figure 8.
Pulse-Input Dye Tracer Study Performed on Continuous
Flow Inactivation Reactor
31
-------
approximately 20 min, but the reactor inflow was a combination of 90 m£/min
of the sample feed solution (clarified activated sludge) and 10 m£/min of the
free chlorine feed solution. This flow ratio of 9:1 resulted in an undesir-
able dilution of the sample to be disinfected, but was unavoidable due to the
limits of the feed pumps used in the experiments. Due to the chlorine
demand of the clarified activated sludge, it was found that whereas the
chlorine feed solution was free available chlorine, the residual chlorine,
as measured in the reactor effluent, was always present as combined available
chlorine, except in the case of the breakpoint chlorination experiments. For
the non-breakpoint experiments, a 20 min contact time was a sufficient
reaction time to convert the free chlorine in solution to combined chlorine,
and the chlorine residual was thus measured as total combined available
chlorine. The breakpoint chlorination experiments involved measurement of
free residual chlorine only. The natural population chlorination studies
were performed without any adjustment of the pH or temperature of the acti-
vated sludge effluent feed solution; the chlorine feed solution was buffered
at pH 7 and at room temperature.
In an attempt to investigate the inactivation behavior of the test
organisms under conditions resembling advanced wastewater treatment and/or
water reuse, chlorination experiments were performed utilizing an activated
carbon treated secondary wastewater. All experiments were performed using
the same activated carbon treated clarified activated sludge effluent. The
activated carbon treatment consisted of passing the secondary wastewater
effluent by upflow through a semi-expanded column of granular activated
carbon (approximately 340 g, Grade 10 x 30, Darco Activated Carbon, ICI
Americus Inc., Wilmington, DE). The theoretical column detention time was
8.5 min at a hydraulic loading of approximately 114 £/min/nr for the 3.8 cm
inside diameter, tubular column reactor. The effluent from the activated
carbon column was collected and used as the influent organism test solution
to the chlorination reactor. The clarified activated sludge solution was
passed through the carbon column for at least 8 hr at the described applica-
tion rate before a sample was collected for the actual chlorination experi-
ment; this permitted the carbon column to reach a steady state condition.
The influent and effluent of the carbon column were characterized as to pH,
turbidity, suspended solids, temperature and total organic carbon. The
chlorination experiments followed the same experimental procedures as out-
lined for the other natural population studies.
Coagulation-Flocculation and Filtration Experiments
To determine the removal of yeasts and acid-fast organisms by various
treatment processes, a series of laboratory studies were performed using
coagulation-flocculation and sand filtration.
Coagulation-flocculation performance with respect to removing the test
organisms was studied using typical jar-test procedures. Each series of
experiments was designed to achieve the optimum removal of turbidity in the
water tested. This was accomplished by optimizing the coagulant dosage,
taking into account pH, flocculation time, and mixing speed. The jar tests
were performed using six-bladed paddles and a Phipps and Bird multiple
32
-------
position flocculator (Phipps and Bird, Inc., Richmond, VA). Water samples
were placed in 1 L beakers and stirred for 2 min at 100 rpm after which suf-
ficient 1.0 N HC1 or 1.0 N NaOH was added for pH adjustment, when necessary,
along with the appropriate dosage of coagulant. Following this, flash mixing
at 100 rpm for 2 min was accomplished. The stirring rate was then reduced
to the desired flocculation speed for a predetermined time. After 30 min of
settling, the supernate was decanted and the turbidity measured. Samples
were also removed for the enumeration of yeasts, acid-fast, and total coli-
form organisms.
Seven series of coagulation experiments were performed. Three of these
used alum as a coagulant, and two each used ferric chloride and lime. One
series with each coagulant was conducted using river water collected from
the Salt Fork of the Vermillion River at Station RSI. The remaining series
of experiments for each coagulant were performed using dechlorinated tap
water to which a caolinite clay suspension and one percent raw municipal
wastewater were added. The experiments with river water were made at 20°-
25°C, while the dechlorinated tap water experiments were performed at 10°-
15°C. The tap water had an alkalinity of 180 mg/£ as CaC03 and a pH of
7.9-8.2, while the river water typically had an alkalinity of 350 mg/£ and
a pH of 7.1-7.2.
The stock suspension of kaolinite was prepared by adding 50 g kaolinite
clay (Kaolinite #7, Ward's Natural Science Establishment, Rochester, NY) to
20 L of tap water. After shaking, the suspension was allowed to settle for
1.5 hr. The suspension was decanted to remove large, settleable, clay par-
ticles, and then used to attain the desired turbidity in the test water prior
to each experiment. Stock solutions of coagulants were prepared using deion-
ized water.
For the filtration experiments, a column was constructed from an acrylic
plastic cylinder. It was 1.83 m in length and had an inside diameter of
3.81 cm. The filter media consisted of washed sand having a D&Q of 0.53 mm
and a DIQ of 0.38 mm, with a uniformity coefficient of 1.4. The depth of
sand in the column was 0.91 m and was supported by well-ground gravel, graded
into three sizes. The base support was a 16.5 cm layer of gravel with a
diameter in excess of 4.7 mm. The middle layer was 3.8 cm of gravel with a
diameter between 3.36 and 4.7 mm, while the upper layer of the underdrain was
gravel having a diameter between 2.36 and 3.36 cm and was 2.54 cm deep.
The sand column was backwashed immediately prior to each experiment.
Dechlorinated tap water was used as the suspending medium. As will be noted
later, one percent raw municipal wastewater was used as the organism inoculum
in several experiments. In all other experiments, washed cell suspensions of
E. co&t, C. paSiapA-iloA-Ui, M. fiofutuLtum, and M. pkleA., prepared according to
the aforementioned procedures, were used as the source of organisms and were
added at the same time to the influent to the filter. The inoculated water,
having either the pure cultures or wastewater as the source of organisms,
was prepared in a large volume and was continuously mixed and, at the same
time, pumped to the head of the filter to maintain a standing head of between
33
-------
15 and 35 cm of water- At periodic intervals during the course of each
experiment, samples from the influent tank and from the filtrate were taken
for microbial analysis.
Continuous Flow Ozonation Experiments
To study the relative effects of ozone bubbles and ozone residual on
the inactivation of yeasts and acid-fast organisms, three different continu-
ous flow reactor arrangements were devised. In arrangement No. 1, ozone
demand free water, inoculated with the test organisms, was fed into the top
of the reactor and removed from the bottom. A gaseous mixture of ozone and
air was supplied at the bottom of the reactor through a fritted glass dif-
fuser and exhausted through the top. The reactor consisted of a 55 mm Pyrex
column 270 mm long, with a 500 mi volume. This arrangement provided informa-
tion on the combined effects of ozone residual and ozone bubbles on the
inactivation of the test organisms.
In arrangement No. 2, using the same physical reactor as above, inocu-
lated water and sodium thiosulfate solution were fed into the top of the
reactor through separate streams. The ozone/air mixture was supplied at the
bottom of the reactor and exhausted through the top. The use of sodium thio-
sulfate reduced all residual dissolved ozone, permitting study of only the
effect of ozone bubbles on organism inactivation. In arrangement No. 3, an
aqueous solution of ozone was added at the top of the reactor along with a
stream of organism-inoculated water. The reactor used in this arrangement
was a Pyrex glass bottle having a diameter of 80 mm, a height of 120 mm, and
a volume of 500 ml. A Teflon-coated magnetic stirring bar, driven by an
external magnet, was used for mixing. In this arrangement the effect of
ozone bubbles was eliminated, thus permitting study of the effect of ozone
residual on organism inactivation.
A preliminary study using tracers confirmed that all three of the reac-
tor arrangements approximated an ideal continuously-stirred tank reactor
(CSTR). The three experimental arrangements are shown schematically in
Figure 9.
The ozone demand free water used in these experiments was prepared by
ozonating deionized water for 15 min, followed by boiling. To further ensure
dissipation of any ozone residual, the water was exposed overnight to ultra-
violet light.
The air/ozone mixture used in these studies was produced in a Welsbach
T-408 Laboratory Ozonator (Welsbach, Philadelphia, PA) under a feed pressure
of 0.703 kg/cm2 gauge. The ozonator was operated at a pressure of 0.562 kg/
cm2 gauge and a voltage setting of 80-115 v. Feed air was supplied from the
laboratory service line, dried with calcium chloride, and filtered through a
30 cm column of activated silica gel and fiberglass wool.
Initially, two dissolved ozone analyzers were employed in this study.
However, due to poor instrument performance, analysis of dissolved ozone was
performed using the method of Shechter (8). All experiments were performed
34
-------
OJ
en
Inoculated
Water
To Hood To Hood
Inoculated f N_a2 S2 03
Reactor
A ~~
Water
uy
Reactor
A
Drain
Ozonated
~A7r
Air
Arrangement
No. I
Solution
Inoculated
Water
Reactor
B
^To Drain
Ozonated
Air
Arrangement
No. 2
Dissolved
Ozone
•To Drain
Magnetic
Stirrer
Arrangement
No. 3
Figure 9. Schematic Diagram of Reactor Arrangement Nos. 1, 2 and 3
Used in Ozone Inactivation Studies
-------
using deionized water buffered at pH 7. The temperature of the water was
maintained at 24°C throughout all experiments. The entire experimental set-
up was sterilized before use by passing a stream of ozonated air or ozonated
water through the system for 15 min.
Mechanism of Inactivation by Chlorine
Hypochlorous acid used in these experiments was prepared according to
the method of Moeller (10), yielding material low in chlorides. Radioactive
hypochlorite was prepared by an isotope exchange between hydrochloric acid
(3t>Cl) and unlabeled hypochlorous acid (11,12).
Radioactivity analysis was performed in a Beckman LS-100 Liquid Scintil-
lation Spectrometer (Beckman Instruments, Inc., Palo Alto, CA) using a 2:1
v/v mixture of phase-combining scintillant (Amersham Searle, Arlington
Heights, IL) and scintillation grade p-xylene (Amersham Searle). Soluble
radioactivity was measured by pipetting the sample directly into the scin-
tillation cocktail; particulate radioactivity was measured by removing the
particulate matter by either glass fiber (Schleicher and Schuell #25) or
membrane filters (Mi 11ipore Type HAWP) and then placing them into the cock-
tail. Corrections were made, where necessary, for background and non-specific
binding of radiotracer, but not for quench.
In the chlorine binding experiments, known aliquots of radio-labelled
hypochlorite were added to erlenmeyer flasks containing CDFB. At appropriate
intervals, the flasks were manually agitated and a sample withdrawn for
particulate and total radioactivity analysis. Corrections were made for the
binding of radio-labelled chloride. In determining the adsorption isotherms,
equilibration times used were 30 min for C. p&napA'iloAiA and M. ^o^tu^tam,
and 6 min for E. coti. The use of cell and chlorine controls was identical
to that used in earlier studies (5,13).
The ability of microorganisms to grow after exposure to chlorine was
assessed using a chlorine-exposed suspension of log phase cells and, after
dechlorination with thiosulfate, resuspending the cells in fresh concen-
trated growth medium. The cells were incubated at the appropriate tempera-
ture, and growth was assessed by optical density measurements at periodic
intervals.
Release of either ultraviolet-absorbing material or TOC by cells after
chlorination was determined by exposing the test organisms to chlorine and
dechlorinating aliquots withdrawn after periodic intervals. Cells were
removed by centrifugation, and the supernate analyzed. The UV-absorbing
material released was measured at 260 and 280 nm with a Beckman Acta III
recording spectrophotometer (Beckman Instruments, Inc., Palo Alto, CA),
using quartz cuvettes of 1 cm pathlength in double beam mode referenced
against a solution containing buffer and identical concentrations of thio-
sulfate and chlorine.
The effect of chlorine on cellular respiration was assessed by equili-
brating cells with an appropriate substrate, which was found to have no
36
-------
chlorine demand. After a stable respiration rate was reached, as measured
by a DO electrode, an aliquot of chlorine was added, and the resulting res-
piration rate was monitored. After 10 min of contact, thiosulfate was added
and the final respiration rate assessed. Throughout each experiment, stir-
ring was achieved by magnetic bar so that the incubation vessel could be
closed to the atmosphere.
The effect of chlorine on cellular potassium uptake was determined
using radiotracers. 42« Was prepared by irradiation of K2C03 in the
University of Illinois' TRIGA nuclear reactor. The cell suspension plus
1.5 g/£ dextrose was equilibrated for 5 min, after which 42|< was added.
At appropriate intervals, samples were withdrawn for particulate radioactivity
determination. At 45 min, chlorine was added and, after various time inter-
vals, samples were withdrawn for radioactivity determination.
The effect of chlorination on protein synthesis, and DNA synthesis was
determined using radiotracers. Log phase cells grown in the presence of
5 g/£ lactose were prepared. Chlorine demand free preparations of organisms
were exposed to chlorine for 10 min in the case of E. coti, and 30 min for
C. paJtap^-Uio^-U, and M. fiofctiiLtum. After exposure and dechlorination, cells
were split into two portions. To one portion of cells, 0.8 yCi/m£ of
(2,3,4,5-3H)-L-Proline (Amersham Searle, Arlington Heights, IL) was added.
At periodic intervals, samples were withdrawn and added to 1 volume of 10
percent (w/v) hot (70°C) trichloro-acetic acid (TCA). After contact at 70°C
for 30 to 45 min, the treated cells were analyzed for particulate radioactiv-
ity subsequent to washing with cold TCA; radioactivity was interpreted as
protein synthesis. To the second portion, 1.2 yCI/m£ of (methyl-3H)-thymine
was added. At periodic intervals, samples were withdrawn and each was added
to an equal volume of cold (5°C) TCA. After contact for 30 to 45 min, the
treated cells were analyzed for particulate radioactivity after washing with
cold TCA; radioactivity was interpreted as DNA synthesis.
37
-------
SECTION 5
RESULTS AND DISCUSSION
ST. JOSEPH/OAKWOOD FIELD STUDIES
This investigation was carried out to ascertain the presence of the
proposed indicator organisms in a typical wastewater treatment effluent, in
the receiving stream above and below the outfall, and at a water treatment
plant downstream. A related objective was to determine whether the proposed
indicator organisms were ever absent in the presence of either fecal or total
coliforms in such waters. It was also proposed to evaluate the diurnal and
seasonal variability of these indicator organisms as related to the waste-
water treatment plant, and the seasonal variability both in the receiving
stream and at the downstream water treatment plant. The St. Joseph Waste-
water Treatment Plant and its receiving stream, the Salt Fork of the
Vermillion River, and the Oakwood Water Treatment Plant made up the system
of study.
Analysis of yeast, acid-fast, and total and fecal coliform organisms
was performed at the St. Joseph Wastewater Treatment Plant on samples of raw
wastewater (SJ1), secondary effluent (SJ2), and the chlorinated effluent
(SJ3) over a period of 24 hr on July 22-23, 1975 (Figure 2). Graphical vari-
ations of organism density as a function of time are presented in Figures 10
to 13. The results indicate the consistent occurrence of these organisms in
the wastewater treatment plant. The density of yeast and acid-fast organisms
in the raw wastewater was found to be approximately 10^/100 m£. The density
of fecal coliforms was consistently higher than the proposed indicator organ-
isms by an order of 2-2.5 logs, while total coliforms were usually higher
than fecal coliforms by a magnitude of one log.
The mean density of the four groups of organisms over the 24 hr period
at the three sampling stations is presented in Figure 14. In general, a
constant removal pattern of 1-1.5 log reduction for all four organism groups
was observed through secondary treatment, contact stabilization process of
aeration, and settling. It may be concluded that the four groups of organ-
isms showed similar behavior with respect to the contact stabilization pro-
cess. Reduction of organism density as a result of chlorination was observed
to be in the range of 1-3.5 logs. The relative resistance of these four
groups of organisms to chlorination may be determined from Figure 14 by com-
paring the slopes of the plot between points SJ2 and SJ3. It can be seen
from this comparison that yeasts and total coliforms showed similar resis-
tance, while acid-fast organisms were most resistant and fecal coliforms were
38
-------
!0C
E
o
o
-------
SJI - Raw Wastewater
SJ2 - Secondary Effluent
SJ 3 - Chlorinated Effluent
12
Midnight
Figure 11. Densities of Acid-Fast Organisms at the St. Joseph
Wastewater Treatment Plant Over a 24 Hr Period
40
-------
E
O
O
v>
E
o
o
SJI - Raw Wostewater
SJ2- Secondary Effluent
SJ3- Chlorinated Effluent
Figure 12. Densities of Total Coliforms at the St. Joseph Wastewater
Treatment Plant Over a 24 Hr Period
41
-------
E
i_
o
o
O
"o
o
i I i i i i ' '
i i i I ii i
SJ I - Row Wastewater
SJ2 -Secondary Effluent
SJ3 -Chlorinated Effluent
Figure 13. Densities of Fecal Coliforms at the St. Joseph Wastewater
Treatment Plant over a 24 Hr Period
42
-------
100
10
o»
c
o
E
«>
QC
in
E
in
c
o
o>
c
0)
o
Acid-Fast
Organisms
O.I
Fecal Coliforms
0,01
SJI - Raw Wastewater
SJ2 - Secondary Effluent
SJ3 -Chlorinated Effluent
0.001
SJI SJ2 SJ3
Sampling Station
Figure 14. Mean Normalized Organism Densities at the St. Joseph
Wastewater Treatment Plant Over a 24 Hr Period
43
-------
the least resistant to disinfection by chlorine. A typical chlorine dosage
ranged from 3.5-4.0 mg/l and the plant effluent contained an average total
chlorine residual of 0.8 mg/l.
Seasonal variation of yeast, acid-fast, fecal, and total coliform organ-
ism density at the St. Joseph Wastewater Treatment Plant was evaluated by
sampling the wastewater at the three sampling stations previously described
during the period 2 May 1975 to 29 January 1976. The results are presented
as three dimensional bar diagrams in Figures 15 to 18. The density of yeast,
acid-fast, and fecal coliform organisms in the raw wastewater was generally
low during the winter months (November to January) as compared with their
density during warmer periods (May to October). Reductions in organism den-
sity through the contact stabilization process and as a result of chlorina-
tion were very similar to the results of the 24 hr study discussed above.
The density of yeasts in the raw wastewater at the St. Joseph Wastewater
Treatment Plant varied from 2.73 x 103-3.94 x 104/100 ml during the period
of study. A similar variation in the yeast density, 4.86 x 10^-3.13 x 104/
100 ml, was observed during the 24 hr study. Acid-fast organism densities
during the same period of study varied from 1.43 x 103-1.43 x 104/100 ml in
the raw wastewater. Less variation in the acid-fast density was observed in
the raw wastewater during the 24 hr study, where it ranged from 6.50 x 1Q3-
1.26 x 10V100 ml- It is significant to note that yeasts and acid-fast
organisms were always detected in the raw wastewater over the entire period
of study.
The density of fecal coliforms in the raw wastewater varied from
6.67 x 104-9.45 x 106/100 ml during the period of study. This represents a
two-log variation as compared with a one-log variation in the raw wastewater
density of both yeast and acid-fast organisms. The variation in fecal coli-
form density during the 24 hr study was much less than that observed in the
seasonal study, ranging from 2.96 x 106-9.26 x 106/100 ml. Enumeration of
total coliforms in the raw wastewater was difficult due to their extreme vari-
ation in density. In almost all cases, the density of total coliforms was
0.5-2 logs greater than the fecal coliform density.
The Salt Fork River system was also examined to ascertain whether yeasts
and acid-fast organisms were present whenever the commonly accepted indicator
organisms (fecal and total coliforms) were present in such waters. Data were
also collected and analyzed to establish the variability, both temporal and
spatial, of yeasts and acid-fast organisms in the river system and to compare
their variability to that of the coliform group of organisms. The location
and description of the sampling stations for the Salt Fork River system are
given in Figure 4 and Table 6.
The density of yeasts, acid-fast organisms, fecal coliforms, and total
coliforms in the Salt Fork River system are presented in Figures 19 to 22.
It should be noted that the chlorinated wastewater effluent from the St.
Joseph Wastewater Treatment Plant is introduced into the river system between
stations RSO and RSI. Although somewhat difficult to determine from Figure
21, the density of fecal coliforms decreased continuously between RSI and RS5
44
-------
-p.
en
SJ3
'2,75
May 19,75
June 4,75
June 23,75
July 14,75
Aug28,75
Sept 18,75
Oct 16,75
Nov 18,75
Jan 8,76
Jan 29,76
Figure 15. Densities of Yeasts at the St. Joseph Wastewater Treatment Plant
-------
en
SJ3
Moy2,75
Moyl9
June 4
June 23
Julyl4
Aug28,75
Sept 18,75
Oct 16,75
Nov 18,75
Jon 8,76
Jon 29,76
SJ I
Figure 16. Densities of Acid-Fast Organisms at the St. Joseph Wastewater Treatment Plant
-------
-pi
-vj
SJ3
May 2,75
May 19,75
June 4,75
June 23,75
July 14,75
Aug28,75
Sept 18,75
Oct 16,75
I8,75_
Jon 8,76
Jan 29,76
Figure 17. Densities of Fecal Coliforms at the St. Joseph Wastewater Treatment Plant
-------
OO
SJ3
May 2,75
May 19,75
June 4,75
June 23,75
July 14,75
Aug28,75
Sept 18,75
Oct 16,75
Nov 18,75
Jan 8,76
Jan 29,76
Figure 18. Densities of Total Coliforms at the St. Joseph Wastewater Treatment Plant
-------
RS5
S May 2,75
May 19,75
June 4,75
June 23,75
July 14,75
Aug28,75
Sept 18,75
Octl6,75
Nov 18,75
Jan 8,76
Jan 29,76
RSI
RSO
Figure 19. Densities of Yeasts in the Salt Fork River System
-------
on
O
RS5
May 19,75
June
June
23
July 14
Aug 28
Sepl 18
Octl6
Nov 18
Jan 8,76
Jan 29,76
RSO
Figure 20. Densities of Acid-Fast Organisms in the Salt Fork River System
-------
RS5
May 2,75
May 19,75
June 4,75
June 23,75
July 14,75
Aug28,75
Sept 18,75
Octl6.75
Nov 18,75
Jan 8,76
Jan 29,76
RSO
Figure 21. Densities of Fecal Coliforms in the Salt Fork River System
-------
Moyl9,75
June 4,75
June 23,75
July 14,75
Aug28,75
Sept 18,75
Octl6,75
Nov 18,75
Jon 8,76
Jon 29,76
RS5
RSO
Figure 22. Densities of Total Coliforms in the Salt Fork River System
-------
in the field samples collected on June 23, September 18, October 16, and
November 18, 1975, and in those on January 8 and 29, 1976. A maximum decrease
in fecal coliform density of 2.5 logs was observed from RSI to RS5 based upon
the results of samples obtained on October 16. The samples collected on the
other dates did not show a definite trend with respect to a change in density
of fecal coliforms. Figure 19 indicates a continuous increase in yeast den-
sity on two occasions between RSI and RS5 (May 2, 1975 and July 14, 1975).
A continuous decrease in yeast density was observed on September 18 and
November 18, 1975. The density of yeasts in the river system remained rela-
tively constant between RSI and RS5 on the other sampling dates. The density
of acid-fast organisms in the river system fluctuated from station to station
on most of the sampling dates (Figure 20). Exceptions to this observation
occurred on October 16, 1975, when there was a minor decrease in density, and
on August 23, 1975 and November 18, 1975, when there was a continuously
decreasing trend in the density of acid-fast organisms. The enumeration of
total coliform organisms was difficult in the river samples due to their
extreme variability (Figure 22). It was observed, however, that the density
of total coliforms was less during the cold periods than during the warmer
months.
Differences in the density of each of the four groups of organisms at
sampling stations located above and below the point of discharge of chlori-
nated secondary effluent by the St. Joseph Wastewater Treatment Plant were
not found to be significant, i.e., the wastewater effluent did not appear
to cause significant changes in the biological quality of the Salt Fork
River water in terms of the organism groups studied. Physical-chemical
parameters monitored above and below the outfall of the wastewater treatment
plant also showed that the effluent had limited impact on the quality of
water in the stream.
The density of the four organism groups was also determined at four
stations within the Oakwood Water Treatment Plant system (Figure 3). The
results are presented in Table 9. During the sampling period covered by this
investigation, the density of all four groups of organisms in the raw water
reservoir (OW1) was much lower than their respective density in the Salt Fork
River system. This may have been due to the settling and dieoff of the organ-
isms in the holding reservoir; the addition of copper sulfate to control
algal growth in the reservoir during the summer months may also have had an
effect. The majority of the organisms present at OW1 were removed from the
water by the process of coagulation-flocculation, prechlorination, and fil-
tration. In most of the samples analyzed, none of the test organisms could
be detected in the filtered water (OW3), and in almost all cases, any organ-
isms present at OW3 were inactivated by chlorine during the disinfection
process, resulting in their absence in the finished water at station OW4.
DECATUR FIELD STUDIES
A field sampling program was performed at Decatur, Illinois, over a
seven-month period to analyze the occurrence of the proposed indicator
organisms through a water treatment plant and in its associated distribution
53
-------
TABLE 9. DENSITY OF ORGANISMS AT THE OAKWOOD WATER TREATMENT PLANT
en
Sampling
Date
3/4/75
4/4/75
5/2/75
5/19/75
6/4/75
'6/23/75
7/14/75
8/28/75
9/18/75
10/16/75
11/18/75
1/8/76
1/29/76
Yeasts
no./100 mi
OW1
0
9.00
0.33
0
125.00
5.50
9.00
0.67
14.00
3.50
0.34
78.34
106.00
OW2
0
0
0.33
0.67
0
0
0
0
0
22.70
1.00
OW3
0
0
0
0.34
0
0
0
0
0
0.67
0
OW4
0
0
0
0
0
0
0
0
0
0
0
0
0
Acid-fast organisms
no./lOO mJL
OW1 OW2 OW3 OW4
10.00 000
65.00 53.00 0.67 0
6.67 000
12.00 7.67 0 0
20.00
13.34 0 1.00 0:67
110.00 3.00 2.30 0
17.00 3.67 0.34 0
Fecal col i forms
no./lOO mi
OW1 OW2
0
5.50
0 0
0 0
3.30 0
0 0
0.50 0
0 0
0
0.67
0.34 0
8.67 2.34
0.34 0
OW3
0
0
0
0
0
0
0
0
0
0
0
OW4
0
0
0
0
0
0
0
0
0
0.67
0
0
0
Total col i forms
no./lOO m£
OW1 OW2 OW3
0
0 0
0-0 0
29.68 0.51 0.26
0 0
0 0
7.34 1.50
1.67 0 0
8.75 0.34
1 1 . 34 0 0
OW4
0
0
0
0
0
0
1.00
0
0
0
-------
system. Previously described enumeration techniques were used to analyze
large-volume samples for yeasts, acid-fast organisms, total coliforms, and
standard plate count bacteria. A complete description of the treatment plant,
distribution system, and sample stations has been given (Figure 5 and Table 8)
Total coliform analysis by MF and MPN techniques revealed that coliform
organisms were always present in the influent raw water (station No. 1). The
density of total coliform organisms in the influent raw water as determined
by the MF technique ranged from 190-29,800/100 m£ over the entire sampling
period (Table 10). Total coliform densities as determined by the MPN tech-
nique were less than by MF in all cases, with a range of 31- >2400/100 mi.
The mean log difference between the MF and MPN techniques in determining
total coliforms in the raw water was 0.825, i.e., MF > MPN. Table 10 shows
that coliforms were found in the treated water on two occasions, at station
No. 2 on 23 August 1977, and at station Nos. 4 and 5 on 11 October 1977. The
presence of coliforms at two of these three sample stations was confirmed by
the MPN technique (Table 11). The occurrence of coliforms at station No. 4
on 11 October 1977 was not confirmed by the MPN technique.
TABLE 10. DENSITY OF TOTAL COLIFORM ORGANISMS DURING DECATUR FIELD
SAMPLING (No./lOO ma by the MF Enumeration Technique)
Sampling Station
Date
4/05/77
4/19/77
5/24/77
6/01/77
6/07/77
6/14/77
6/28/77
7/06/77
7/19/77
8/23/77
9/07/77
9/20/77
9/27/77
10/11/77
1
190
290
1680
1480
1200
190
464
4900
673
1990
29800
7400
5800
5530
2
0
0
0
0
0
0
0
0
0
132
0
0
0
0
3
0
0
0
0
0
0
-
0
0
0
0
0
0
0
4
0
0
0
0
0
0
0
0
0
0
0
0
0
0.5
5
0
0
0
0
0
0
0
0
0
0
0
0
0
30
55
-------
TABLE 11. DENSITY OF TOTAL COLIFORM ORGANISMS DURING DECATUR FIELD
SAMPLING (No./lOO nU by the MPN Enumeration Technique)
Sampling Station
Date
4/05/77
4/19/77
5/24/77
6/01/77
6/07/77
6/14/77
6/28/77
7/06/77
7/19/77
8/23/77
9/07/77
9/20/77
9/27/77
10/11/77
1
70
31
240
170
no
170
170
33
70
920
>2400
1600
350
>2400
2
*
*
*
*
*
*
*
*
*
>16
*
*
*
*
3
*
*
*
*
*
*
-
*
*
*
*
*
*
*
4
*
*
*
*
*
*
*
*
*
*
*
*
*
*
5
*
*
*
*
*
*
*
*
*
*
*
*
*
5.1
Results expressed as organism density <2.2/100 m£
Yeasts were detected with 100 percent frequency in the influent raw water
during the sampling period; the geometric mean value was 5.7/£. Yeasts were
enumerated in the finished tap water at the treatment plant (station No. 2)
with 15 percent frequency; the geometric mean density was 5.3/a. No yeasts
were recovered from stations Nos. 3 and 4 in the distribution system. The
most terminal sampling point in the distribution system (station No. 5)
showed the occurrence of yeasts in two samples, or in 15 percent of the
samples examined. These results are summarized in Table 12.
Acid-fast organisms were recovered from all of the influent raw water
samples examined (Table 13). The frequency of recovery from a potable water
tap at the treatment plant (station No. 2) was 36 percent, and the incidence
of acid-fast organisms in the distribution system ranged from 7-23 percent.
The density of acid-fast organisms in the influent raw water ranged from
6-169/£, with a geometric mean of 37.0/1. Table 13 summarizes the results of
the occurrence of acid-fast organisms during this study; it should be noted
that these organisms were detectable at all sampling stations at one time or
another. Chlorination and other treatment processes caused a reduction in
56
-------
TABLE 12. DENSITY OF YEAST ORGANISMS DURING DECATUR FIELD SAMPLING (Ho./a)
Sampling Stati
Date
4/05/77
4/19/77
5/24/77
6/01/77
6/07/77
6/14/77
6/28/77
7/06/77
7/19/77
8/23/77
9/07/77
9/20/77
9/27/77
10/11/77
1
16
20
6
5
3
7
9
3
3
12
1
2
16
—
2
0
0
0
0
0
0
0
0
0
9.5
0
3
0
~
3
0
0
0
0
0
0
-
0
0
0
0
0
0
"
on
4
0
0
0
0
0
0
0
0
0
0
0
0
0
"
5
0
0
0
0
0
0
0
0
0
0.5
1
0
0
'
the density of acid-fast organisms within the treatment plant of between 27
and 100 percent.
Bacterial densities as measured by the estimated standard plate count
(SPC) were found to be very high. At sample station No. 1 (raw water influ-
ent), the SPC values ranged from 4850-48,840/m£, with a geometric mean dens-
ity of 19,879/m£. Of particular interest was the fact that between 14 and
43 percent of the total number of samples obtained from the distribution
system showed a SPC which exceeded that of the SPC in the raw water influent
(station No. 1, Table 14). It should be noted that the SPC values at station
No. 4 on 7 June 1977 and 19 July 1977 were on the order of 106 and 105/m£,
respectively. These values appear to be unreasonably high relative to the
other data for station No. 4 and, therefore, may not be an accurate indica-
tion of the existing bacterial quality in the distribution system at this
sampling station. Alternatively, the high bacterial counts observed at
station No. 4 on these dates might have been due to the dislodgement of bio-
logical slimes by high flow conditions in the main. As previously mentioned,
the fact that the North Treatment Plant may contribute some flow at this
location in the distribution network could also account for the observed high
densities.
57
-------
TABLE 13 DENSITY OF ACID-FAST ORGANISMS DURING DECATUR
FIELD SAMPLING (No.
Sampling Station
Date
4/05/77
4/19/77
5/24/77
6/01/77
6/07/77
6/14/77
6/28/77
7/06/77
7/19/77
8/23/77
9/07/77
9/20/77
9/27/77
10/11/77
1
61
6
108
51
33
48
100
8
93
96
48
6
8
169
2
0
0
0
0
24
4
0
0
0
5
1
0
0.5
0
3
0
0
0
0
1
2
-
0
0
2.5
0
0
0
0
4
0
0
0
0
1
3
0
0
0
6
0
0
0
0
5
0
0
0
0
2
0
0
0
0
0
0
0
0
0
It may be calculated that the treatment processes, including chlorina-
tion, lead to a 13-98 percent reduction in the SPC between sample stations
Nos. 1 and 2 (Table 14). On 28 June 1977, however, the number of bacteria
as determined by the SPC actually increased between the raw and the treated
water. Table 15 gives the geometric mean density for each of the organism
groups at each of the sample stations. The increase in SPC as seen at station
No. 4 was due to the inclusion of the data from 7 June 1977 to 19 July 1977
as previously mentioned. If these two values are omitted due to their ques-
tionable accuracy, the geometric mean SPC at station No. 4 becomes 5729/m£.
A statistical analysis was performed on the organism density data to
determine if any correlation might exist between the occurrence of any two
organism groups in the influent raw water, particularly between coliforms and
either the yeasts or acid-fast organisms. Product-moment correlation coeffi-
cients were also computed for raw water turbidity vs. log organism density
and for raw water temperature vs. log organism density (14). Results of this
analysis are given in Table 16. It can be seen that the only significant
correlation among the parameters was an inverse correlation between yeast and
total coliform organism density, at the 5 percent confidence level, i.e.,
58
-------
TABLE 14. ESTIMATED STANDARD PLATE COUNT DURING DECATUR
FIELD SAMPLING (No./m£)
Date
4/05/77
4/19/77
5/24/77
6/01/77
6/07/77
6/14/77
6/28/77
7/06/77
7/19/77
8/23/77
9/07/77
9/20/77
9/27/77
10/11/77
Sampling Station
1
15,940
19,080
20,135
4,850
34,260
10,880
20,880
47,840
37,780
44,660
14,280
16,880
30,700
10,900
2
290
6,490
2,000
2,280
6,290
1,080
31 ,080
41,800
14,880
18,630
11,630
12,390
19,760
140
3
490
18,270
23,870
1,040
6,200
1,785
-
25,460
9,660
15,690
30,780
14,710
24,270
15
4
3,050
910
4,310
7,440
>106
7,370
2,780
28,420
190,820
22,650
32,720
1,620
3,510
0
5
9,270
4,440
28,960
8,935
39,030
1,880
26,260
590
8,650
10,830
9,530
20,830
41,940
103
there was a negative association between yeast and total coliform density in
the raw water samples.
Water leaving the treatment plant for distribution contained a free
available chlorine residual ranging from 0.6-1.2 mg/£, with a mean value of
0.88 mg/£. In only 3 of the 14 series of samples was there no free chlorine
residual detected at the most terminal sampling point of the distribution
main under study (station No. 5).
59
-------
en
o
TABLE 15. GEOMETRIC MEAN VALUES OF ORGANISM DENSITY AND CHLORINE RESIDUAL
DURING DECATUR FIELD SAMPLING BETWEEN
4 APRIL 1977 and 11 OCTOBER 1977
Sampling Station
Parameter
Free C^ residual, mg/£
Total do residual, mg/£
Total Coliform
MPN/100 m£
MF/100 m£
Yeast/£
Acid-fast/£
Est. Standard Plate
Count/m£
1
-
-
238.5
(14/14)*
1592.2
(14/14)
5.7
(13/13)
37.0
(14/14)
19,879
(14/14)
2
0.88
1.09
**
132
(1/14)
5.3
(2/13)
3.0
(5/14)
5114
(14/14)
3
0.70
1.06
**
0.0
(0/13)
0.0
(0.12)
1.7
(3/13)
5096
(13/13)
4
0.52
0.82
**
0.5
(1/14)
0.0
(0/13)
2.6
(3/14)
11,160
(13/14)
5
0.36
0.40
**
30
(1/14)
0.7
(2/13)
1.4
(1/14)
7178
(14/14)
Numbers in parentheses indicate frequency of samples which were positive, e.g., 14/14, 14 positive
out of 14 samples examined.
**
MPN data for stations Nos. 2 through 5 were always <2.2/100 m£, except on two occasions:
23 August 1977 (station No. 2): count was >16/100 m£
11 October 1977 (station No. 5): count was 5.1/100 m£
-------
TABLE 16. STATISTICAL CORRELATION ANALYSIS OF RAW WATER PARAMETERS
FOR DECATUR FIELD STUDIES BETWEEN 4 APRIL 1977 and
11 OCTOBER 1977; NUMBER OF SAMPLES = 14
Significance at
Variables Correlation coefficient the 5% level
TC and yeast densities
TC and AF densities
Yeast and AF densities
Water Temp, and TC density
Water temp, and AF density
Water temp, and yeast density
Water temp, and SPC density
Turbidity and TC density
Turbidity and AF density
Turbidity and yeast density
Turbidity and SPC density
-0.593
-0.174
-0.231
0.377
-0.070
-0.544
0.132
-0.198
0.245
0.020
0.127
Significant
Not significant
Not significant
Not significant
Not significant
Not significant
Not significant
Not significant
Not significant
Not significant
Not significant
AF = acid-fast organisms
TC = total coliform organisms
SPC = estimated standard plate count
CONTINUOUS FLOW CHLORINATION EXPERIMENTS
Continuous flow chlorine-inactivation experiments were performed to verify
the resistance of the acid-fast, yeast, and coliform organisms under dynamic
conditions and to investigate the inactivation behavior of the organisms under
conditions more closely resembling that of water and wastewater chlorination.
In an attempt to reproduce these conditions, a plug flow tubular reactor was
used as the disinfection contact chamber and both pure culture and natural
population inactivation studies were conducted. The pure culture chlorination
studies were performed as a "mixed culture" of the test organisms in all cases;
this permitted a direct comparison of the relative resistance of the organisms
to chlorine under the same, simultaneous conditions. The natural population
studies consisted of chlorinating a clarified activated sludge effluent and,
since this effluent was also a "mixed culture" of the test organisms, all
experiments established in a comparable way the relative resistance of the
yeast, acid-fast, and coliform organisms to chlorine under constant experi-
mental conditions.
61
-------
Mixed Culture Studies
In the mixed pure culture studies the yeast, C. paAap*U.oAiA , the acid-
fast, M. £o/itu,tt.iim, and the E. c.oli organisms were added together in CDFB and
subjected to a free available residual chlorine concentration of between 0.515
and 1.620 mg/l for contact times up to 20 min. The initial organism density
used in these studies was approximately 7 x 105 ml for E. coti; the density
ranged from 4.19 x 105-1.65 x 107/100 ml for M. fiotitiUtw, and from 6.12 x 10b-
6.76 x 106/100 ml for C. poJta.pt>JUiobit,. Experiments were performed at pH 7 and
10, and at room temperature which varied from 22°-24.5°C. Using these experi-
mental conditions, it was possible to study the response of the test organisms
to a range of free available chlorine residuals and ratios of hypochlorous
acid to hypochlorite ion concentrations. It was also possible under these
conditions to compare, to a certain degree, the results obtained in these
experiments with those obtained in the batch inactivation studies.
The response of the "mixed culture" of organisms to free available resid-
ual chlorine is shown in Figures 23 and 24. The results shown in these two
figures are typical in that only one concentration of free available chlorine
is given for the two pH values studied; the response of the three organisms
was also determined using other concentrations of chlorine at both pH 7 and 10
(Figures 25 through 28). In all the experiments, at least 4 logs of inactiva-
tion of E. coli occurred in the shortest contact time studied, 4 min. Because
of the nature of the experimental reactor plus the high rate of inactivation,
it was impossible to successfully enumerate E. coti during the first 4 min of
contact. Acid-fast organisms were observed to have the highest degree of
resistance to free residual chlorine, with yeast and coliform organisms show-
ing progressively less resistance, respectively. In all experiments, the
response of the mixed culture of yeast and acid-fast organisms to free resid-
ual chlorine clearly demonstrates the superior chlorine resistance of both of
these groups of organisms over that of the coliforms. Figures 25 and 26 show
the response of the yeast and acid-fast organisms, respectively, to various
levels of free residual chlorine at pH 7 while Figures 27 and 28 summarize the
response of the test organisms at pH 10.
At pH 7 and 20°C, approximately 80 percent of the free available residual
chlorine is in the form of hypochlorous acid, whereas at pH 10, only approxi-
mately 0.4 percent is hypochlorous acid with the rest being hypochlorite ion.
In general, hypochlorous acid is accepted as being a more potent disinfectant
than hypochlorite ion in the inactivation of microorganisms. The results of
this study support this observation in that it was shown that as the pH of
the system increased, from pH 7 to pH 10, the rate of inactivation of the
yeast and acid-fast organisms decreased. The inactivation of the yeast
(C. paAapi>Lf>} decreased significantly at the higher pH, i.e., with a lower
concentration of hypochlorous acid; the acid-fast organism (M. ^ontu^Ltwm) was
not affected to the same degree. However, it may be concluded that yeasts and
acid-fast organisms are more effectively inactivated by hypochlorous acid than
hypochlorite ion. No conclusions can be drawn concerning E. cotL since it
could not be successfully enumerated at either pH due to their high degree of
inactivation under all experimental conditions.
62
-------
o
>
3
(O
"c
0)
u
w.
0)
Q.
8
12 16
Contact Time (min)
20
24
28
Figure 23.
Continuous Flow Inactivation Studies with 0.53 mg/£ Free
Available Chlorine Residual Using Mixed Pure Cultures in
Chlorine Demand Free Buffer at pH 7 and 22.5°C
63
-------
100
C. pgrapsilosis
20
E. coli
I
I
I
8
12
16
20
24
28
Contact Time (min)
Continuous Flow Inactivation Studies with 0.96 mg/£ Free
Available Chlorine Residual Using Mixed Pure Cultures in
Chlorine Demand Free Buffer at pH 10 and 24.5°C
64
-------
o
>
en
c
0)
o
w
4)
Q.
Free
Available
Chlorine (mg/1)
O 0.58
• 0.72
A 0.96
D 1.15
12 16 20
Contact Time (min)
Figure 25. Continuous Flow Inactivation Studies with Various Free Avail-
able Chlorine Residuals and Temperatures Using a Pure Culture
of C. po^apix^o-i-u in Chlorine Demand Free Buffer at pH 7
65
-------
o
>
>
k.
3
0)
u
w
-------
100
80
50
D
>
3
CO
0)
Q-
Free Available
Chlorine (mg/i)
O 0.93
O 1.00
A 1.48
• 1.52
Temp (°C)
22.0
21.5
23.0
24.5
20
10'
Figure 27.
8
20
24
28
12 16
Contact Time (min)
Continuous Flow Inactivation Study with Various Free Available
Chlorine Residuals and Temperatures Using a Pure Culture of
C. panapt>-LtoA-i* in Chlorine Demand Free Buffer at pH 10
67
-------
100
80
50
o
3
CO
0)
u
w
0)
Q.
Free Available
Chlorine (mg/l)
O 0.93
0 1.48
A 1.62
Temp (°C)
22
23
22
20
10
I
8
16
20
24
28
Figure 28.
12
Contact Time (min)
Continuous Flow Inactivation Study with Various Free Available
Chlorine Residuals and Temperatures Using M.
Chlorine Demand Free Buffer at pH 10
68
in
-------
Comparison of these observations with those of the batch reactor
studies with free chlorine show general agreement (5). In both studies,
acid-fast organisms were found to be more resistant than yeasts, which in
turn were more resistant than the coliform organisms to free available
residual chlorine under similar experimental conditions. This was true for
experiments at both pH 7 and pH 10, with the rate of inactivation of the
yeast and acid-fast organisms always being less at the higher pH value. The
response of the yeast and acid-fast organisms to the higher pH value was
similar in these studies as in the batch reactor studies. In the batch
reactor studies it was calculated that in the inactivation of the yeast,
hypochlorous acid was approximately 5-20 times more effective than the hypo-
chlorite ion. In the case of the acid-fast organism, it was found that
hypochlorous acid was only 1.1-2.5 times more effective than the hypochlorite
ion (5). Considering the results of the continuous flow experiments, it was
observed that in the inactivation of the yeast and acid-fast organisms, the
hypochlorous acid concentration at pH 7 was approximately 4-10.5 and 1.5-3
times more effective, respectively, than the hypochlorite ion which predomi-
nates at pH 10. Therefore, both the continuous flow and batch reactor
studies support the conclusion that hypochlorous acid is a more effective
disinfectant for inactivation of yeasts and acid-fast organisms than hypo-
chlorite ion.
A comparison of the E. coti data from the two studies is of question-
able significance based on the fact that in the batch studies the E. coli
suspensions were prepared from nutrient agar slants and were enumerated as
fecal coliforms, whereas in these continuous flow studies, E. coti suspen-
sions were prepared from nutrient broth cultures and were enumerated as
total coliforms. Although these differences exist, it may be concluded that
there was general agreement in the inactivation of E. coti. between the two
studies in comparing the experiments performed at pH 7. In neither study
was it possible to reliably demonstrate the presence of E. c.oLi, even with
the shortest contact time, due to their high degree of inactivation. With
the batch reactor studies, more than 3 logs reduction of E. coti occurred in
4 min, while these continuous flow studies showed greater than 4 logs inacti-
vation with the same contact time. In the batch reactor studies at pH 10,
up to 1.8 percent of E. coLi survived 1.0 mg/£ free available residual chlor-
ine after a contact time of 10 min; inactivation of E. colti. in the continu-
ous flow system amounted to more than 4 logs in 4 min. The reason for this
discrepancy is not apparent.
Natural Population Studies
The natural population studies were performed as either breakpoint or
non-breakpoint chlorination experiments. In all cases, the test solution in
each inactivation experiment was a grab sample of the clarified activated
sludge effluent from the East Side Wastewater Treatment Plant, Sanitary
District of Urbana-Champaign. A review of the operating and performance
characteristics for the activated sludge unit of the treatment plant, as
supplied by the Sanitary District, together with additional data (TOC and
COD), indicated that the plant was achieving satisfactory results with
69
-------
respect to removal of organic matter and suspended solids. Considering all
the grab samples of activated sludge effluent analyzed and used in the
natural population study, the average BOD5, TOC, COD, and suspended solids
in the effluent was found to be 14.3, 21.5, 77.5, and 9.0 mg/£, respectively.
The pH varied from 7.30 to 7.76 while the ammonia concentration in the efflu-
ent averaged 5.35 mg/t as N. A statistical analysis was performed to deter-
mine if there was any correlation between inactivation of the test organisms
and the physical/chemical characteristics of the test solution. Using a
step-wise regression, it was determined that the independent variable of_
chlorine residual accounted for greater than 85 percent of the variance in
the inactivation of each of the three test organisms.
In the non-breakpoint chlorination experiments, the dosage of chlorine
approximated that used in practice when chlorinating a wastewater effluent.
The chlorine dose varied from 1.5-4.5 mg/£ and the corresponding reactor
effluent residual chlorine concentration, measured as total combined chlo-
rine, varied from 0.48-2.6 mg/£. In these experiments, the reactor effluent
pH and temperature varied between 6.9 and 7.3 and 22° and 24°C, respectively.
Results of the non-breakpoint natural population chlorination studies,
based upon the inactivation of acid-fast, yeast, and fecal coliform organisms
by various levels of combined residual chlorine, are shown in Figures 29
through 32. Under all conditions studied in these experiments, the acid-fast
organisms consistently showed the lowest inactivation rate, followed by that
of the yeasts; the fecal coliform organisms always exhibited the highest
rate of inactivation. In general, for a contact time of 20 min, the overall
reduction in acid-fast organisms was slightly less than 1 log, while the
yeast showed a reduction greater than 1 log; approximately 3.5 log reduction
was observed with the fecal coliform organisms. Some variation in inactiva-
tion occurred for all three groups of organisms, depending on the experi-
mental conditions. As seen from these figures showing the results of chlori-
nating a natural population of organisms such as might be found in an acti-
vated sludge effluent, the acid-fast and yeast organisms were consistently
more resistant to combined available chlorine than the coliform organisms.
These plots also show that as the chlorine dosage and resulting residual
increased, the inactivation of the test organisms also increased. This
expected increase in inactivation with increased combined chlorine residual
was not as pronounced in the current study as in the study with chloramines
reported earlier (5). In the current natural population study, chlorine was
added as free chlorine which then reacted with the ammonia to form combined
chlorine, while the chloramine study was done with preformed chloramines.
It may be hypothesized that organism inactivation under the former conditions
should be greater than that observed in the chloramine experiments, because
of the initial presence of free available chlorine.
The results of the current study also agree with the data presented
earlier in this report for the inactivation of the test organisms at the
St. Joseph Wastewater Treatment Plant. Under similar chlorination condi-
tions, the St. Joseph study showed a mean log reduction, due to chlorination
of secondary effluent, of 0.40, 1.3, and 3.2 logs for the acid-fast, yeast,
70
-------
>
'>
w
3
0)
u
w
0>
Q.
Fecal Coliform (MPN
Time (min)
Figure 29. Non-Breakpoint Chlorination Study with 0.63 mg/£ Total
Combined Residual Using Clarified Activated Sludge Effluent
71
-------
10
o
>
0>
O
w.
O>
Q.
Fecal Coliform (MPN)
Fecal Coliform (MF)
8 12 16 20
Contact Time (min)
Figure 30. Non-Breakpoint Chlorination Study with 0.73 mg/l Total
Combined Residual Using Clarified Activated Sludge Effluent
72
-------
o
>
>
k.
3
c
4)
O
w
0)
Q.
Fecal Coliform MPN
Fecal Coliform (MF)
8 12 16 20
Contact Time (min)
24
28
Figure 31
Non-Breakpoint Chlorination Study with 0.84 mg/l Total
Combined Residual Using Clarified Activated Sludge Effluent
73
-------
10
a
>
C
-------
and fecal coliform organisms, respectively, which compares favorably with
the inactivation of the same organisms in the current study.
Experimental results for studies in which fecal coliforms were enumer-
ated by both the MPN and MF techniques are presented in Figures 30, 31, and
32. In these experiments, the MPN technique always gave a higher recovery
of fecal coliforms than the MF procedure. The densities of fecal coliforms,
as determined by the MF procedure, agree with the range of fecal coliform
densities found in activated sludge effluents in past studies (5).
The breakpoint chlorination experiments were performed so as to obtain
a free available residual of approximately 0.5-1.0 mg/£ in the reactor efflu-
ent. A sufficient volume of activated sludge effluent was collected so that
experiments at the two different levels of free available residual chlorine
could be made with the same grab sample. For the breakpoint studies, the
chlorine dosage varied from approximately 63-87 mg/£, whereas the free
available residual chlorine concentration varied from 0.45-1.10 mg/£. Traces
of combined chlorine residual were detected in the reactor effluent on
occasion, but it could not be measured with any accuracy or consistency and,
thus, was not considered valid. The reactor effluent pH and temperature
for the breakpoint experiments varied from 7.15-7.35 and 18°-21°C, respec-
tively.
Results from the breakpoint chlorination experiments were limited due
to the high degree of inactivation that occurred for each group of organisms
studied. With the experimental arrangement used, neither the yeasts nor the
coliform organisms could be accurately enumerated under any of the experi-
mental conditions studied. The acid-fast organisms, on the other hand, were
successfully enumerated in several of the breakpoint experiments, but only
with the shorter contact times of 4 and 8 min, and in the reactor effluent
(contact time, 20 min) when a large sample volume could be collected and
filtered.
Table 17 presents the chlorine dosage and resultant residual of each
breakpoint experiment, and the log reduction for each organism group
measured after a 20 min contact time. Of the three organism groups studied,
the acid-fast organisms were the most resistant to chlorination. The inac-
tivation results for the yeast and coliform organisms do not permit a
definite conclusion to be reached regarding their relative resistance to
breakpoint chlorination.
The survival curves in Figure 33 for the acid-fast organisms are given
for the breakpoint experiments in which the acid-fast organisms were success-
fully enumerated. Based upon these data, it would appear that initially,
within approximately 4 min contact time, there was a very rapid inactivation
of the acid-fast organisms; this was then followed by a much slower rate of
inactivation. This observation is consistent with the characteristics
associated with members of the acid-fast group of organisms and their vari-
able resistance to chlorination. The natural population of acid-fast organ-
isms in wastewater, i.e., activated sludge effluent, probably includes species
with varying degrees of sensitivity to chlorination, e.g., M. pkt&i is more
75
-------
TABLE 17. INACTIVATION DATA FOR BREAKPOINT CHLORINATION EXPERIMENTS
Log reduction*
Chlorine
dosage,
Date mg/£
11/28 63.0
65.0
12/06 86.0
87.0
12/17 78.0
81.0
Free Chlorine
residual ,
mq/£
0.59
1.04
0.45
1.10
0.46
0.95
(20
Col i forms
>4.70
>4.56
>3.63
>3.84
>4.03
>3.88
min contact
Yeasts
>2.24
>2.40
>1.86
>1.98
>2.17
>2.25
time)
Acid-Fast
Organisms
2.16
2.26
>1.96
>1.77
2.13
>2.29
Because of enumeration difficulties, much of the data must be expressed as
greater than (>).
sensitive to chlorination than M. fiosituAtum (5). Under the severe conditions
of breakpoint chlorination, it might be expected that the more sensitive or
a percentage of the total population of acid-fast organisms would be ini-
tially inactivated at an apparent high rate while the more resistant species
would be inactivated at a slower rate.
In the previous chloramine study (5) using a contact time of 20 min,
M. fiotittutum showed approximately 0.3 log reduction when subjected to 3.25
mg/£ of chloramine, while C. poMupk-ULob-ik exhibited approximately 0.52 log
reduction with 2.83 mg/£ of chloramine. In the current breakpoint study,
the initial density of M. fioKtuLtum was reduced by 0.7 log and C. pasiapt>t-
loA-lb showed greater than 4 logs reduction when both were exposed to 1.15
mg/l of free available residual chlorine for 20 min. Considering the non-
breakpoint chlorination study, a chlorine dosage of 4.0 mg/£ which produced
a residual of 1.73 mg/£ of total combined chlorine resulted in approximately
0.83 and 1.46 log reduction of the acid-fast and yeast organisms, respectively.
Care must be taken in making any direct comparison of the results of
the natural population study with those of the other studies mentioned. The
activated sludge used in the natural population study probably contained
several members of each test organism group, members which are known to show
a variable response to chlorine disinfection. In contrast, the free chlorine
and chloramine studies used a known chlorine resistant member of each organ-
ism group. The presence of other organisms of these groups in the secondary
effluent used in the natural population study, which may have been less
resistant to chlorination, may be the reason for the observed discrepancy
between the expected and observed inactivation of the test organisms among
the three studies.
76
-------
10s
10'
o
>
'>
u.
3
0)
o
Free Chlorine Residual
O 0.46 mg/f
O 0.59 mg/Jt
A 0.95 mg/f
Based Upon Minimum Reduction
Determined
8
20
24
28
Figure 33.
12 16
Contact Time (min)
Breakpoint Chlorination Study for Acid-Fast Organisms Present
in Clarified Activated Sludge Effluent
77
-------
Another series of chlorine inactivation experiments was performed with
a natural population of organisms using activated sludge effluent which was
subsequently treated by activated carbon. The measured physical and chemical
characteristics of the activated carbon column influent, i.e., clarified
activated sludge, and effluent are given in Table 18. Comparison of the
influent and effluent values presented in Table 18 shows that only TOC was
significantly affected by treatment with activated carbon, i.e., 55 to 75
percent reduction in TOC.
TABLE 18. PHYSICAL AND CHEMICAL CHARACTERISTICS OF ACTIVATED CARBON
COLUMN INFLUENT* AND EFFLUENT
Exper.
No.
pH Temp,
Inf. Eff. Inf.
°C
Eff.
Suspended
Turbidity Solids
NTU mg/£
Inf. Eff. Inf. Eff.
TOC, mg/£
Inf. Eff.
1 7.40 7.20 20.5 20.5 12.5 10.5 6.5 3.0 21.0 5.5
2 7.35 7.30 21.0 21.0 7.5 6.0 11.0 9.0 29.0 13.0
3 7.50 7.38 20.0 20.0 9.0 8.0 13.5 12.5 17.5 6.6
*
Influent - clarified activated sludge effluent, East Side Wastewater Treat-
ment Plant, Sanitary District of Urbana-Champaign
Experimental data and inactivation results of the chlorination experi-
ments performed using the activated carbon treated secondary effluent are
shown in Table 19. A slight reduction in the densities of all three test
organisms occurred as a result of activated carbon treatment. In no case,
however, was the reduction in organism density through the activated carbon
column significant.
The actual chlorine inactivation results shown in Table 19 agree in
general with the results of the previous natural population chlorination
study. Acid-fast organisms were the most resistant of the three test
organisms to chlorination, with the yeast being less sensitive to chlorina-
tion than the fecal coliform organisms. Comparison of the disinfection
results as given in Table 19 for activated carbon treated secondary waste-
water with those of the non-breakpoint natural population study shows that
the acid-fast and fecal coliform organisms were inactivated to about the
same degree in the two studies. On the other hand, the yeast organisms
appear to be less chlorine resistant in the study utilizing activated carbon
treated effluent than in the non-breakpoint experiments. Table 20 compares
the log reductions of each of the three test organisms in the two studies
78
-------
TABLE 19. RESULTS OF CONTINUOUS FLOW CHLORINATION INACTIVATION OF
ORGANISMS PRESENT IN CLARIFIED ACTIVATED SLUDGE EFFLUENT
AFTER TREATMENT BY ACTIVATED CARBON
Exper
No.
1
2
3
Exper
No.
1
2
3
1
2
3
1
2
3
Chlorine dosage
mg/£
1.5
3.3
2.5
Organism density in
clarified activated
sludge effluent
No./lOO mi
Acid-Fast
1.09 x 103
6.23 x 103
2.7 x 103
Yeasts
4.8 x 103
9.4 x 103
6.85 x 103
Fecal Col i forms
8.6 x 105
5.53 x 105
6.9 x 105
Total combined
Organism density in
activated carbon
column effluent*
No./lOO mi
1.036 x 103
6.105 x 103
2.6 x 103
4.5 x 103
8.366 x 103
6.23 x 103
7.998 x 105
5.31 x 105
6.66 x 105
chlorine residual
mg/£
0.66
1.69
1.31
Organism density
in chlori nation
reaction chamber
effluent**
No./lOO mi
2.07 x 102
9.16 x 102
2.7 x 102
2.4 x 102
1.924 x 102
2 x 102
5.6 x 102
1.274 x 102
1.465 x 102
**
Effluent from the activated carbon column was collected and used as the
influent organism test solution to the chlorination reactor chamber.
Contact time 20 min.
79
-------
TABLE 20. COMPARISON OF CHLORINE INACTIVATION RESULTS USING CLARIFIED
ACTIVATED SLUDGE EFFLUENT AND ACTIVATED CARBON TREATED ACTIVATED
SLUDGE EFFLUENT FOLLOWING 20 MINUTES CONTACT TIME
Total Total
combined Activated carbon treatment, combined
chlorine log reduction chlorine
Non-breakpoint chlorination
log reduction
residual
mq/l
0.66
1.31
1.69
acid-
fast
0.699
0.984
0.824
yeast
1.274
1.490
1.638
fecal
col iform
3.155
3.658
3.620
residual
0.63
1.20
1.73
acid-
fast
0.634
0.894
0.901
yeast
1.11
1.32
1.50
fecal
col iform
3.20
3.59
3.68
under similar conditions of chlorine residual. In general, it might be
thought that, due to the lower TOC level in the chlorination experiments
utilizing activated carbon treated effluent, a higher degree of inactivation
of the test organisms would be achieved than in the non-breakpoint experi-
ments because of the lower chlorine demand in the former. However, this
does not appear to be the case, and in fact the opposite is true for the
yeast organisms.
REMOVAL OF INDICATOR ORGANISMS BY CHEMICAL
COAGULATION WITH ALUM, FERRIC CHLORIDE AND LIME
The major emphasis of these laboratory experiments was to evaluate the
removal of a natural population of acid-fast organisms, yeasts and total
coliforms by chemical coagulation when the conditions for turbidity removal
were optimum. In order to define the conditions for optimum turbidity
removal, each test water was subjected to a series of four experiments:
1) optimization of coagulant dose; 2) optimization of pH at the optimum
coagulant dose; 3) optimization of flocculation time at a constant floccula-
tion speed of 25 rpm; and 4) optimization of flocculation speed at a constant
flocculation time of 20 min.
Each coagulant, alum, ferric chloride, and lime, was evaluated with two
different test waters. The first water used consisted of dechlorinated tap
water to which a stock suspension of kaolinite clay was added as a source of
turbidity and a 1 percent volume of raw municipal wastewater was added as an
organism inoculum. The second water was obtained from the Salt Fork of the
Vermillion River in Vermillion County, Illinois. In this water, no organism
inoculation was necessary because a natural population of the indicator
organisms to be studied was present. Seven series of coagulation experiments
80
-------
were performed in all. Three of these used alum as a coagulant, and two
each used ferric chloride and lime.
Figure 34 shows the results of a typical series of experiments. This
particular series (experimental series 1) was performed with alum and dechlor-
inated tap water with kaolinite clay and raw wastewater added. As may be
seen in Figures 34(a) and 34(b), the optimum removal of turbidity occurred
with a dosage of 10 mg/£ alum as A1203 at a pH of approximately 7. Effective
flocculation times and flocculation speeds showed a wide range of values.
It may be observed in Figure 34(c) that with a flocculation speed of 25 rpm,
stable turbidity removals of 96-97 percent were achieved with mixing times
of between 20 and 50 min. Figure 34(d) shows that with a flocculation time
of 20 min, all mixing rates between 25 and 50 rpm gave turbidity removals of
96-97 percent. The wide range of flocculation times and mixing speeds at
which it was possible to produce a constant removal of turbidity indicate
the stability of the floe particle to mixing stresses.
Since emphasis was given to the removal of the indicator organisms when
the conditions for turbidity removal were optimum, more consideration will
be given to the results in Figures 34(c) and 34(d) than those in Figures
34(a) and 34(b). Organism removals are reported as the range of experimental
values observed throughout the range of optimum turbidity removal seen in the
experiments performed to determine the optimum flocculation time and mixing
speed. For example, Figure 34(c) shows that the least removal of yeasts
occurred at a flocculation time of 5 min with a flocculation mixing rate of
25 rpm. The removal of yeasts was approximately 40 percent. Figure 34(d)
shows that the greatest removal of yeast (98-99 percent) occurred with a
mixing speed of 25 rpm with a flocculation time of 20 min. Therefore, a
realistic estimate of yeast removal at the optimum conditions of turbidity
removal would be a range of between 80 and 99 percent. A similar analysis
of these figures indicates the range of acid-fast organism removal to be
approximately 91-99.1 percent and the range of total coliform removal to be
approximately 97-98 percent. The original turbidity in this series of
experiments was 49 NTU. The original density of organisms, reported as
number per 100 mi, was 2.6 x 104, 150, and 140 for total coliforms, acid-
fast organisms, and yeasts, respectively.
Data obtained for the other six series of experiments were analyzed in
a similar manner to that described for the results presented in Figure 34.
The range of removal efficiency for each of the indicator group of organisms
and turbidity, at optimum conditions for turbidity removal, are given in
Table 21. Table 22 provides the physical-chemical conditions for optimum
turbidity removal in all seven experimental series. Control densities of
the indicator organisms (No/100 ml) and the initial turbidity in all seven
series are shown in Table 23.
Experimental series 2 and 3 were also performed with alum. Series 2
was essentially the same as series 1 with the exception that the initial
turbidity was 19 NTU rather than 49. The range of removal of the indicator
organisms was essentially the same as in the first series, being approxi-
mately 90-99 percent for all three groups of indicator organisms when the
81
-------
o>
o
E
4»
or
*-
u
I
0
10 20 30 40 50
Dosage AI203 (mg/l)
Dosage Optimization! pH 7, flocc.
time = 20min, flocc. rate =30 rpm
a Y = yeast organisms
• T = turbidity
60
o»
_c
'E
'5
E
c
4)
O
4)
a.
10 20 30 40
Time (min)
50
60
Flocculation Time Optimization! pH 7,
IOmg/1 Al 0 , flocc. rate =25 rpm
10
(b)
I
I
5 6 7 8 9 10
PH
pH Optimization! IOmg/1 AI2 03
flocc. time = 20min, flocc.
rate = 30 rpm
o AF = acid-fast organisms
A TC = total coliform organisms
Flocculation Speed Optimization!
pH 7, 10 mg/l AI203 , flocc.
time = 20 min
Figure 34. Removal of Indicator Organisms and Turbidity with Alum (Tap Water,
Kaolinite Clay, and 1.0 Percent Raw Wastewater; Temperature =
10-15°C)
82
-------
TABLE 21. REMOVAL OF INDICATOR ORGANISMS AT OPTIMUM
CONDITIONS FOR TURBIDITY REMOVAL
Experimental
series
Turbidity,
%
Total
coliforms,
Acid-fast
organisms.
Yeasts,
1) Alum, kaolinite 96-97
clay and 1%
raw wastewater
2) Alum, kaolinite 80-98
clay and 1%
raw wastewater
3) Alum, 85-95
river water
4) Ferric chloride, 95-99
kaolinite clay
and 1% raw
wastewater
5) Ferric chloride, 70-98
river water
6) Lime, 85-96
kaolinite clay
and 1% raw
wastewater
7) Lime, 75-80
river water
97-98
93-99
93-98
98-99.5
60-97
99.97-99.98
99.7-99.97
91-99.1
89-93
>93
>95
>90
>99.8
80-99
89-98
92-98
97-99
78-94
98-99
97-99.8
removal of turbidity was optimum. The range of optimum turbidity removal
was found to be slightly higher (80-98 percent) than in the first series.
This was due to the method of recording optimum results. The third experi-
ment of the second series (optimization of flocculation time) showed a
constant removal of turbidity from 97-98 percent with flocculation times of
20-50 min. The fourth experiment of the series (optimization of mixing speed)
indicated a removal of turbidity of 80-85 percent with mixing speeds of 25-50
rpm. Hence, the range of turbidity removal recorded for the second series in
Table 21 is the combined result of the two experiments and is given as the
range between 80 and 98 percent.
Experimental series 3 was performed utilizing river water. In these
83
-------
TABLE 22. OPTIMUM PHYSICAL-CHEMICAL CONDITIONS FOR
OPTIMUM TURBIDITY REMOVAL
Experimental
series
Coagulant dose
mq/£ pH
Flocculation
time (min)
at 25 rpm
Flocculation
speed (rpm)
for 20 min
1) Alum,
kaolinite clay
and 1% raw
wastewater
2) Alum,
kaolinite clay
and 1% raw
wastewater
3) Alum,
river water
4) Ferric chloride,
kaolinite clay
and 1% raw
wastewater
5) Ferric chloride,
river water
6) Lime,
kaolinite clay
and 1% raw
wastewater
7) Lime,
river water
10.0 as A1203
10.0 as A1203
5.0 as A1203
30.0 as
30.0 as Fe0.
300 as CaCO-
500 as CaCO-
7.0
7.0
7.0
9.6
7.9
10.9
10.7
20-50
20-50
30-50
30-50
30-50
40-50
40-50
25-50
25-50
30-50
30-50
30-50
40-50
35-50
experiments, the optimum removal of turbidity was found when 5.0 mg/£ alum
as A1203 was used. Other considerations, such as pH, flocculation time and
mixing speed, as well as organism and turbidity removal ranges were very
similar to those recorded in series 1 and 2. In considering the first three
series of experiments as a whole, it may be summarized that under optimum
conditions for turbidity removal, approximately 90-99 percent removal of the
turbidity and indicator organisms may be expected when alum was used as the
coagulant.
Special consideration should be given to the data for the removal of
acid-fast organisms reported for experimental series 3-7 (Table 21). In
84
-------
TABLE 23. INITIAL CONDITIONS FOR THE EXPERIMENTAL COAGULATION
Total Acid-fast
Experimental Turbidity coliforms organisms Yeasts
series NTU No.7100 ml No./TOO ml No./IOQ ml
1) Alum, 49 2.6 x 104 150 140
kaolinite clay
and 1% raw
wastewater
2) Alum, 19 2.0 x 106 170 120
kaolinite clay
and 1% raw
wastewater
3) Alum, 19 1.8 x 103 230 95
river water
4) Ferric chloride, 23 1.5 x 105 20 125
kaolinite clay
and 1% raw
wastewater
5) Ferric chloride, 12 8.3 x 102 58 340
river water
6) Lime, 37 1.2 x 105 130 230
kaolinite clay
and 1% raw
wastewater
7) Lime, 19 4.7 x 103 -- 170
river water
series 3-6, removal ranges are given as "greater than" an indicated level of
removal. No removal range is cited for the acid-fast organisms in series 7.
A problem was encountered with a contaminating mold growth which inhibited
the recovery of acid-fast organisms in some of the experiments. In series
3-6, only one of the two final optimum experiments in each series, i.e.,
optimization of flocculation time and optimization of mixing speed, yielded
reliable results. The results of the reliable experiments are the recorded
levels in Table 21. Data from the other experiments, i.e., those in which
the acid-fast plates showed signs of contamination, tended to always show
a higher removal of these organisms than the level recorded in Table 21.
The range of removal efficiencies is therefore presented as "greater than"
85
-------
the reliable values obtained. Reliable results for acid-fast organism
removal were not obtained in experimental series 7-
The results of the coagulation experiments using ferric chloride with
tap water plus kaolinite clay and raw wastewater (experimental series 4) were
very similar to those obtained with alum. Removal of turbidity and the indi-
cator organisms was generally 95-99 percent. The removal of total coliforms
reached as high as 99.5 percent in some experiments. Optimum conditions of
turbidity removal in this series were a coagulant dose of 30.0 mg/£ as Fe203
and a pH of 9.6. The ranges of flocculation times and mixing speeds which
produced optimum turbidity removal were somewhat less than with alum, being
from 30-50 min and 30-50 rpm, respectively.
A much wider range of results was obtained with river water and ferric
chloride than was reported for the series utilizing ferric chloride and the
tap water suspension. With the river water, turbidity removal was 70-98
percent, total coliform removal was 60-97 percent, yeast removal was 78-94
percent, and the removal of acid-fast organisms was estimated to be greater
than 90 percent (Table 21). Optimum chemical conditions for turbidity
removal were a coagulant dose of 30.0 mg/£ as Fe203 and a pH of 7.9 (Table
22). Although flocculation time and mixing ranges were similar to those
cited for turbidity removal in the tap water suspension, i.e., 30-50 min and
30-50 rpm, no river water turbidity removal was noted if the flocculation
time was less than 30 min in the third test of the series, i.e., optimization
of flocculation time with a mixing speed of 25 rpm.
Turbidity removal by lime precipitation showed the greatest dependence
upon flocculation time and mixing speed compared to the other two coagulants.
Ranges of flocculation time and mixing speeds producing optimum turbidity
removal conditions were limited to 40-50 min and 40-50 rpm in the experiments
with tap water plus kaolinite clay and raw wastewater, and 40-50 min and
35-50 rpm in the river water experiments (Table 22). Dosage requirements in
these two series of tests were very critical. Failure to add sufficient lime
caused an increase in turbidity. Optimum turbidity removal with the tap
water suspension was accomplished with 300 mg/£ as CaC03 at pH 10.9 while
500 mg/£ lime as CaC03 at pH 10.7 was required with the river water.
These two series of experiments gave the best results for using yeasts
and/or acid-fast organisms as indicator organisms instead of the total coli-
forms. While the removal of the yeasts and acid-fast organisms generally
paralleled that of the total coliforms with alum and ferric chloride as the
chemical coagulants, yeast removals were 100 times less than that recorded
for the total coliforms in the experiments with tap water plus kaolinite
clay and raw wastewater, and between 10-100 times less than that for the
total coliforms in the river water experiments with lime. In these two
series of experiments (series 6 and 7), total coliform removal ranged from
99.7 to 99.98 percent, yeast removal ranged from 97-99.8 percent, and the
removal of acid-fast organisms was estimated to be approximately 99.8 percent
(Table 21). A high pH can inactivate microorganisms and, admittedly, the
observed results indicating a higher removal of total coliforms than yeasts
may be an effect of the high pH in these experiments and a superior resis-
tance of the yeasts to these conditions.
86
-------
REMOVAL OF INDICATOR ORGANISMS BY SAND FILTRATION
Figure 35 shows the results of a typical sand filtration experiment
performed with dechlorinated tap water plus a 1 percent wastewater inoculum
as a source of seed organisms. In this experiment, the filter was operated
for 20 hr with an applied flow rate of between 167-190 m£/min and a mean rate
of 180 m£/min. This rate gave a loading of 159 £/min/m2. Yeast removal
ranged from 86-97 percent. Removal of acid-fast organisms was initially 65
percent but increased to 76 percent after 4 hr of filtration. After this
time, a continuous decrease in removal was noted until the end of the filter
run when the removal was only approximately 40 percent. Data for the removal
of total coliforms were very similar to that of the acid-fast organisms.
Initial removals of total coliforms were nearly 70 percent and showed a
gradual reduction to 42 percent removal at the end of 20 hr. It should be
noted that the test waters were applied directly to the filter without any
prior treatment, i.e., coagulation-sedimentation.
Results of the above experiment, another filter run (12 hr) with
dechlorinated tap water plus 1 percent wastewater, and two 36 hr filter runs
with pure culture organisms added to dechlorinated tap water are summarized
in Table 24. In this table, the removal efficiencies are reported at vari-
ous times during the filtration period. Data are presented as combined
results of the two experiments with wastewater inoculum and the two experi-
ments with pure culture organisms. Removal percentages are recorded as an
average for a given length of filter run if the results of the two experi-
ments were within 5 percent of the average, and as a range of values if the
results differ by more than 5 percent of the average. For example, the
initial removal (0 hr) of total coliforms in the two wastewater inoculum
experiments was 55 percent in one experiment and 70 percent in the other.
This result is presented as the range 55-70 percent. After 6 hr of filtra-
tion, total coliofrm removal in one experiment was 68 percent and 72 percent
in the second test. These results are reported in Table 24 as an average
70 percent removal.
It should be pointed out that the rate of filtration in all four exper-
iments was not the same. Loading rates in the two wastewater inoculum
experiments were 119 £/min/m2 and 159 £/min/m2 while in the two pure culture
experiments the rates averaged 106 and 143 £/min/m2, respectively. Removal
results with each test water compared well between the two experiments for
each water and are therefore reported as discussed above. Results reported
for 18 hr of filtration in the wastewater inoculum experiments represent
data from only one experiment since one experiment was terminated after
12 hr. Data presented in Table 24 for the wastewater inoculum experiments
indicate a greater removal of yeasts as compared to the removal of acid-fast
organisms and total coliforms.
In the two wastewater inoculum tests, the control densities of total
coliforms and yeasts were 1 x 105 and 2.2 x 105/100 mi, respectively. The
control densities of acid-fast organisms ranged from 2.3 x 103/100 mi in
one experiment to 2.1 x 102/100 mi in the other experiment. These two dif-
ferent values with the acid-fast organisms apparently did not affect the
removal percentage in the two experiments.
87
-------
o
E
0>
tr
c
0)
o
t_
Q>
Q.
too
90 —
80 —
70
I 1 1 1
10 15
Hours of Filter Operation
20
Figure 35. Removal of Indicator Organisms by Sand Filtration
Using Tap Water Plus 1.0 Percent Raw Wastewater
Inoculum
88
-------
TABLE 24. REMOVAL OF INDICATOR ORGANISMS BY SAND FILTRATION
Percent removal at various times during filter run, hr
0 6 12 18 24 32
1% Wastewater Inoculum
Total coliforms
Acid-fast organisms
Yeasts
Pure Culture Organisms
55-70
67
87
70
66
96
55
43-54
97
45*
40*
97*
E.
M.
C.
COli
fioituLtum and M. pkl&i
paAa.p-f> u,, the control
density in the feed water in the two experiments was 1.2 x 103 and 4.2 x 103/
100 ml in the low rate and high rate experiments, respectively. That
C. po^ap4^£o4-c6 was removed to a higher degree than E. coli and the two acid-
fast organisms is consistent with the results of the wastewater inoculum
experiments. It is conceivable that the larger cell size of the yeast organ-
ism, as compared to that of coliforms and the acid-fast organisms, may have
contributed to the greater removal of the yeasts.
CONTINUOUS OZONATION STUDIES
Ozonation studies were performed in the three reactors previously
89
-------
described, using pure cultures of C. poMip^-ULo^i^, M. fioKtu-itwm, E. coti,
5. typhA.mLifiiu.m, and poliovirus type 1 (Mahoney). Mixed culture studies using
these organisms were also carried out to determine the relative resistance of
these organisms to ozone. The following parameters which influence the
efficacy of ozone disinfection were studied:
a) Ozone residual and ozone bubbles
b) Ozone residual only
c) Ozone bubbles only
d) Effect of mixing in the reactor
e) Initial density of organisms
f) Effect of pH
g) Effect of temperature
h) Presence of UV light
Pure Culture Studies
Effect of Ozone Residual and Ozone Bubbles --
Reactor arrangement No. 1 was employed to study the effect of ozone
residual plus ozone bubbles. Mixing was provided by a gaseous mixture of
ozone and air, supplied at the bottom of the reactor, and the ozone residual
was monitored at each detention time (DT). Figures 36 and 37 show the sur-
vival of C. panaptxil.ot>-Li> and M. fiovtuAtLLm, respectively, for a detention time
of 24 sec.
Effect of Ozone Residual --
Using experimental arrangement No. 3, the effect of ozone residual with-
out bubbles on the inactivation of C. paJia.pt>-ULotxit, and M. ^ovta-itum was
determined. Figures 36 and 37 show the percent survival of these organisms
vs. ozone residual for a 24 sec detention time.
Comparison of Figures 36 and 37 clearly show that inactivation of both
test organisms was greater with reactor arrangement No. 1 than with arrange-
ment No. 3 for a given ozone residual. This indicates that the presence of
ozone bubbles, along with ozone residual, was more effective in inactivating
the test organisms than ozone residual alone.
A conceptual model was prepared to explain this phenomenon. Based on
the film theory of gas transfer, when an organism comes in contact with ozone
in aqueous solution, a liquid film develops around the aqueous layer associ-
ated with the cell wall of the organism. This adsorbed aqueous layer may be
considered an integral part of the cell wall of the organism. Also, accord-
ing to the film theory, a concentration gradient similar to the gas-liquid
interface develops in the liquid film surrounding the aqueous layer. Gen-
erally, the thickness of the liquid film at the gas-liquid interface is much
larger than the diameter of the organism. Further, microorganisms tend to
concentrate at the gas-liquid interface. Because of their presence in the
interface, microorganisms are surrounded by ozone concentrations which are
higher than the bulk liquid ozone concentration. Therefore, with the same
ozone residual in the bulk liquid the tendency for inactivation would be
greater for a cell which is either in direct contact with an ozone bubble or
90
-------
icf
10
10
10
o
u
l_
(U
Q.
-2
10
ic?
-4
10
-5
10
Arrangement No. I
Arrangement No. 3
DT= 24 sec.
C. parapsilosis =1.83-5.3xlO/m)
Figure 36. Survival of C.
0.001 0.01 O.I
Ozone Residual, mg/Jt
in Reactor Arrangement Nos. 1 and 3
91
-------
lOO
10
a
>
CO
c
a>
u
O.I
0.01
Arrangement No. 3
Arrangement No.
DT=24sec. 4 5
M. fortuitum =4.4x10 -2.48xlO/mA
\
I I I I I I
0.01
10
0.1 I
Ozone Residual, mg/^
Figure 37. Survival of M. {,0/itu-ittum in Reactor Arrangement Nos. 1 and 3
92
-------
within the liquid film than for a cell in the bulk liquid. This concept of
concentration effect is in agreement with the observed inactivation data as
shown in Figures 36 and 37, i.e., greater inactivation in a system having
both ozone bubbles and a residual.
Effect of Ozone Bubbles --
Studies involving inactivation by ozone bubbles in the absence of any
residual were performed in reactor arrangement No. 2, where the ozone resid-
ual was destroyed by the addition of an appropriate quantity of sodium thio-
sulfate. Survival curves for M. ^ofvtoJMm and C. poAap4x^o4^4 are presented
in Figures 38 and 39, respectively; the data indicate a slight amount of
inactivation even in the absence of an ozone residual. This may be due to
the fact that an ozone residual may have existed for a short period of time
before being reduced by sodium thiosulfate. No information is available
about the reaction kinetics of ozone and thiosulfate except that it is
believed to be an extremely rapid reaction.
Effect of Mixing --
Ozone decomposition rate has been reported by Hewes
-------
100
o
>
C
o>
o
l_
ft)
Q.
10
10
A Control, Air Flow I S/min.
0 Ozone Gas Flow I S/min.
Gaseous Ozone Cone. = I6.8mg.//f
Thiosulphate = 0.005 M
pH = 7
Temperature =24°C fl
M. fortuitum = 1.22 - 1.6 x 10 /m*
I
I
I
20
30 40
DT, sec
50
60
70
Figure 38. Survival of M. ^ositu-ltm in the Presence of Ozone Bubbles
with no Ozone Residual, Reactor Arrangement No. 2
94
-------
100
o
>
c
4>
O
£.
10
10
Ozone Gas Flow I ^/min.
Gaseous Ozone Cone. = I6.8mg/Jf
Thiosulphate =0.005 M
pH =7
Temperature = 24°C
C. parapsilosis =1.09 x 10 /m A
20
30 40
DT, sec.
50
70
Figure 39.
80
Survival of C. patia.p&4JLo&-i!> in the Presence of Ozone
Bubbles with no Ozone Residual, Reactor Arrangement No. 2
95
-------
o
>
c
tt)
o
10
itf
10
5x ID*
Figure 40.
(1.13)
(0.57)
M.fortuitum = 1.28-2.04 x 10/ml
( ) = Ozone Residual, mg/I
O pH , 10.1
D pH,8.5
A p H , 7.0
• pH , 5.7
,(1.27)"
20
40
60
DT, sec
80
Effect of pH on the Survival of M.
Constant Rate of Applied Ozone
100
at a
120
96
-------
o
>
c
0>
u
Control pH =5.6
Control pH=9.8
Temperature = 24 °C
C. paropsilosis =1.95-2.45 x
6
pH = 9.8
(0.26-0.04mg/J?)
(0.026)
( ) = Ozone Residual, mg/2.
pH=5.6
(0.74- 0.79 mg/50
(0.79)
0 20 40 60 80 100 120
DT , sec
Figure 41. Effect of pH on the Survival of C.
at a Constant Rate of Applied Ozone
97
-------
10
.(0.54)
pH=5.7
10
o
'>
k.
3
C
-------
TABLE 25. EFFECT OF MIXING RATE ON THE SURVIVAL OF M.
AND C. poA^p^cM^i IN EXPERIMENTAL ARRANGEMENT NO. 3
FOR A DT OF 24 SECONDS
1.27
1.27
1.24
63
210
345
Applied
ozone
mg/£
Mixing
rpm
Ozone
residual
mg/£
Number cells/
m£
Percent
survival
0.58
0.55
0.52
M.
Control, 2.32 x
,4
2.19 x 10
2.19 x 10
4
3.23 x 10
C.
100.00
9.44
12.54
13.91
Control, 3.1 x 10°
1.12
1.15
1.15
63
210
345
0.257
0.245
0.233
4.65
24.20
957.00
100.00
0.00015
0.00078
0.0308
In the first set of experiments, the applied ozone level was maintained
constant at 15.8 mg/£, with an ozone/air gas flow rate of 0.5 £/min. The
reactor temperatures studied were 9°, 20°, 30°, and 40°C. Figure 43 shows
the results of the experiments using M. fiofctu^tiw as the test organism. An
increase in temperature resulted in a higher degree of inactivation of
M. ^ont(jJUt(m even though the ozone residual was considerably lower at the
higher temperatures. A control experiment was also performed at 30° and 37°C
in the absence of ozone but with an air flow rate of 0.5 £/min. Results in
Figure 43 indicate that a small degree of inactivation occurred presumably
due to the direct effect of a temperature increase on inactivation.
Since the ozone residual was observed to be one of the most important
parameters affecting ozone disinfection, the ozone residual must be held
constant in order to study the sole effect of temperature on inactivation.
Therefore, another set of experiments was performed at a constant ozone
residual at a given detention time for four different temperatures, i.e., 9°,
20°, 30°, and 37°C. The survival data for M. fioKtuAttm in Figure 44 clearly
indicate the true effect of temperature on inactivation, i.e., the degree of
inactivation increases significantly with an increase in temperature and at
a constant ozone residual.
99
-------
. Control, 30°C-i
ILST^---"•-•&:::::-
M.fortuitum = 3.1-4.14 x 10 /ml
( ) = Ozone Residual, mg/l
DT , sec
Figure 43. Effect of Temperature on the Survival of M.
at a Constant Rate of Applied Ozone
100
-------
w
"c
o
s.
M.fortuitum =3.1-4.02 x 10/ml
( ) = Ozone Residual, mg/l
DT, sec
Figure 44. Effect of Temperature on the Survival of M.
for a Constant Ozone Residual at a Given DT
101
-------
Effect of Ultraviolet (UV) Light —
A study was undertaken to investigate whether the presence of UV light
augments inactivation of M. fioKtuitum by ozone. This study was conducted in
a fermentor (New Brunswick Scientific Co., New Brunswick, NJ) consisting of
a 15 cm dia and 30 cm long Pyrex glass reactor having an effective volume of
4 £. This reactor was equipped with a 15 w low-pressure mercury germicidal
lamp with the major ultraviolet energy being emitted at a wavelength of
253.7 nm. The gaseous ozone concentration was 16.8 mg/£. Survival curves
from this study are presented in Figure 45. It may be noted that comparable
inactivation of M. fiofitusttum occurred with the three different conditions,
i.e., UV light, UV light plus ozone, and ozone alone. However, the impor-
tance of ozone residual in the presence of UV light becomes less significant
because the degree of inactivation of M. ^ofutusLtwm is increased in spite of
a decrease in the ozone residual. It can also be seen in Figure 45 that UV
light alone is a strong disinfecting agent. Therefore, the increase in
inactivation for UV light plus ozone may be due to the application of UV
light rather than to the catalytic effect of UV light.
Effect of Initial Density of Microorganisms --
Preliminary experiments indicated that the degree of inactivation by
ozone was profoundly affected by the initial organism density. Therefore,
three different initial densities of C. p were studied with
respect to inactivation in reactor arrangement No. 3. An applied dissolved
ozone concentration was maintained at 0.15 mg/£ for the three initial yeast
densities. Survival data for C. pafiap^-lto&i^ are given in Figure 46, which
indicate that the percent survival is greater for higher initial organism
density. It may be noted that a 4 log reduction occurred when the initial
density of yeast was 1.35 x 105 cells/m£, while little observable inactiva-
tion took place when the initial density was 1.55 x 10? cells/m£. The
limited degree of inactivation seen in Figure 46 for the two highest organ-
ism densities can be attributed to the substantial ozone demand exerted by
the large amount of organic cell matter present. The measured TOC for the
system having an initial organism density of 1.55 x 10? cells/m£ was about
120 mg/£, compared to a TOC of approximately 4 mg/£ when the initial density
of C. pcvia.p& was 1.35 x 1(P cells/m£. These experiments indicate that
0.15 mg/£ of applied dissolved ozone was not sufficient to inactivate yeasts
when the initial density of cells was high, such as in the case with
1.55 x 107 cells/me.
Mixed Culture Study
Relative Resistance of Microorganisms to Ozonation --
A mixed culture study, including E. dotl, M. ^ofutuJMm, C. paAap^^U.o^>-li>,
S. typkimunium and poliovirus type 1 (Mahoney), was performed in reactor
arrangement No. 1. All of these organisms were grown separately and then
added together in the feed reservoir about 15 min before the initiation of
the experiment. The reservoir was mixed continuously to provide a homo-
geneous suspension. The ozone/air gas flow rate was maintained at a rate
of 0.5 £/min. The pH and temperature were 7 and 24°C, respectively. It can
be seen from Figure 47 that M. fiovtuA-tum was the most resistant organism of
the five, while S. tgphJjmuJvium was the least resistant. E. coti,
102
-------
>
'>
c
V
u
M.fortuitum = 9.94 x 10 /ml
( ) = Ozone Residual , mg/l
-Ozone
(0.387)
(0.52)
Ozone*Ultraviolet Light
(0.13)
0(0.155)
Ultraviolet Light
6'!
Figure 45.
40
80
120
160
200
DT , sec
Effect of Ultraviolet Light on the Survival of M.
at a Constant Rate of Applied Ozone
103
-------
C.parapsilosis =1.55 x 10 /ml
C. parapsilosis = 1.25 x 10/ml
Applied Dissolved Ozone
= 0.14 -0.15 mg/l
C.parapsilosis =1.35 x 10 /ml
DT, min
Figure 46. Effect of Initial Density of C. paAapt>-ito&iA on Degree
of Inactivation for a Constant Rate of Applied Ozone
104
-------
< I I
Ozone Residual = 0.23-0.26 mg/I
M.fortuitum = 8.1 xlOVml
Poliovirus = 2.32 x lOVml
C. parapsilosis =l.55x 10/ml
E.coli =2.09x 10/ml
S.typhimurim=6.56 x lOVml
Figure 47.
DT, sec
Response of Five Test Organisms to Ozone in a DI
Phosphate Buffer Solution
105
-------
C. pcuta.pA-iio£>4A , and poliovirus showed an increasing order of resistance.
The resistance of S. typkimu>vium was comparable to that of E. coti; however,
poliovirus was more resistant than E. coli and C. paA.apA-itoA'i}, but less
resistant than M.
MECHANISM OF INACTIVATION BY CHLORINE
If either acid-fast or yeast organisms are to be accepted as a new indi-
cator of disinfection efficiency, it is necessary that they be more resistant
to disinfection under all circumstances likely to be encountered. This
ability can be ascertained either by extensive field testing or by showing
that the increased resistance they exhibit is a logical result of known
aspects of their microbial physiology. Since chlorine is currently the most
common disinfectant of water and wastewater in the United States, it was
decided to investigate the mechanism of resistance of the acid-fast and
yeast organisms to free available chlorine, using C. paJia.pt> and
as model organisms, respectively
The question of mechanisms of inactivation by free available chlorine
may be considered in two parts: 1) What is the basis for the increased
efficiency of HOC! vs. OCl'?, 2) What is(are) the site(s) of inactivation
of free available chlorine within the cell?
The most prevalent hypothesis by way of answering the first question
is that the uncharged HOC1 molecule is better able to penetrate the cell
wall and be taken up by the microorganism (18). Although this theory has
been directly confirmed in the case of amoebic cysts (18), bacterial
spores (19), and bacteriophage (20), no direct verification has been
reported using vegetative bacteria or yeasts, although the hypothesis has
been supported, in the case of E. c.oti, by work using organic chloramines
(21). In an apparent contradiction to the permeability hypothesis, it has
been reported that bacteria take up more chlorine after exposure to free
available chlorine at high pH than at low pH (22,23).
In regard to the second question, the classical hypothesis has been one
of damage to respiratory enzymes (24,25); however, this hypothesis was
developed without benefit of the current level of knowledge of cellular
biochemistry (26). More recent work has suggested that damage to the cell
permeability barriers may be a cause of inactivation (27,28). In addition,
a variety of studies have suggested that chlorine and chloramines can damage
cellular deoxyribonucleic acid (DNA) (29-36). Thus, damage to DNA may be at
least partially responsible for cell inactivation (29,31).
Chlorine Dynamics
Initial experiments were performed with radioactive chlorine to deter-
mine the kinetics and extent of chlorine binding. These studies showed that
the association of chlorine with the cells was fairly rapid (Figure 48). It
was determined that a contact time of 6 min for E. coti, and 30 min for
either C. panapALtoA'U, or M. fioKtiutum was sufficient to ensure completion
of chlorine uptake.
106
-------
160
0>
0)
c
o
0)
o
'o
o
tf>
tf>
o
128 —
pH 7
2.10-1.57 mg/l Free Chlorine
128
pH 9.14
2.13-1.75 mg/I Free Chlorine
96
64
32
I
T
I
3 6 9 12 15
Contact Time (min)
Figure 48. Chlorine Uptake Kinetics of E. coti
107
18
21
-------
Using these fixed contact times, further studies on the extent of
chlorine interactions with the cells were performed. In these studies, the
amount of chlorine associated with the cells was determined using radio-
activity measurements, and the initial and final free available chlorine
concentrations in solution were measured amperometrically; from a mass bal-
ance, it was determined that a significant fraction of the initial free
available chlorine was not present at the termination of the experiment as
either cell associated chlorine or free residual. This fraction was desig-
nated as "reacted chlorine."
It was found that cell-associated chlorine could be adequately described
by a Freundlich isotherm (Figure 49). In most cases the amount of chlorine
present in association with the cells was much less than the reacted chlo-
rine; in some cases this deviation was more than an order of magnitude. The
ability of C. paruLpt>Ato6 and M. fiosLtuitum to take up more chlorine at low
pH than at high pH was verification of the permeability theory. Further,
the amount of cell-associated chlorine was plotted against the theoretical
concentration of HOC1 (Figure 50). Isotherms for the yeast and acid-fast
organisms at pH 7 and 9.14, obtained by linear regressional analysis, were
found to be not significantly different from each other. In the case of
E. coLL, a substantial difference in cell-associated chlorine was noted
between cells contacted with a constant concentration of hypochlorous acid
at pH 7 vs. pH 9.14.
An additional observation from these studies was that there appeared
to be a correlation between the amount of chlorine taken up by each organism
and its inherent sensitivity to chlorine. On the basis of the continuous
flow chlorination studies and similar work performed earlier (2), the order
of sensitivity of these organisms towards free chlorine was observed to be
E. coti > C. poJuL^-LLo^i^ > M. fiovtuAtum. Considering cell size, C. pan.a.p&^-
lot>jj> has about 20-100 times the surface area and volume of E. co&c, while
M. ^ofvtidtim is of a similar size to E. c.oti. Based on Figures 49 and 50,
after correction for cell size, the order of ease of uptake of chlorine was
E. c-oLL > C. pcuicipt>'itot> > M. fiosituAtum. The only exception to this rela-
tionship was E. c.oti at pH 7 which, it will be shown, may have been due to
underestimation of cell-associated chlorine and excessive leakage after
chlorination.
On the basis of these experiments, it was determined that the basis for
increased efficiency of HOC! as a disinfectant compared to OC1~ was its
enhanced uptake by microorganisms due to its increased ability to penetrate
the cell wall.
Biochemical Effects
To determine the nature of the lethal event by which chlorine inacti-
vates microorganisms, it was necessary to perform experiments monitoring
various aspects of cellular activity presumably influenced by chlorination.
In the first set of experiments, cells suspended in CDFB were dosed with
chlorine, held for a given contact time, and dechlorinated with thiosulfate.
Following this, the test organism was inoculated into an appropriate liquid
108
-------
io13
io14
Q>
O
_4>
O
£
-------
O
^
in
o>
o
E
•o
0)
o
'o
o
in
3
-13
10
-14
10
-18
10
-16
10
-17
0
-18
0
-19
10
EC— E. coli
Cp — C. poropsilosis
Mf — M. fortuitum
10
-7
10
-6
10
-5
10
-4
10
-3
10
Final Hypochlorous Acid Concentration (moles/1)
Figure 50. Uptake of Chlorine as HOC! at Constant Contact Time
110
-------
culture medium and incubated; subsequent growth was monitored using optical
density at 660 nm as a criterion.
Results from the growth experiments (Figures 51-53) indicated that
microorganisms inactivated by chlorine were rapidly prevented from growing;
no initial increase was seen with the treated cells. Only with M. ^ontvuLtmrn,
where the higher levels of survival were found, was an increase in optical
density of chlorine treated cells observed. Even in the face of this result,
in no case was any organism able to grow immediately after chlorination-
dechlorination. This may be interpreted as showing the lethal event caused
by chlorine did not require the participation of growth - death was imme-
diate rather than after a preparatory phase of reproduction. This obser-
vation rules out the possibility of a lethal point mutation, such as a base
mispair or frameshift, as the principal lethal event. If such mutation
were the cause of cell death, it would require at least one generation of
replication before the genotype could be expressed. On this basis, it was
decided to investigate cell properties most likely to be immediately
affected by chlorine.
From the viewpoint of the disinfectant molecule, the first structural
feature of the cell contacted, and therefore the first potential lethal
site, is the cell wall and cell membrane. Damage to these structures may
result in severe damage to cell integrity and/or metabolism, due to a
permeability change, or because of a disorganization of respiratory enzymes
and transport systems localized near the cell surface. If such damage
occurred, the cell may be expected to release ultraviolet-absorbing mater-
ial, e.g., proteins, nucleic acids or precursors, into solution as well as
lower molecular weight organics and inorganic compounds.
Experiments to determine the release of ultraviolet-absorbing mater-
ial showed that release can occur in certain circumstances. The super-
nate optical densities for the controls and chlorine treated cells are
shown in Table 26. The reported values for percent of total extractables
released were calculated by subtracting control optical densities from
that of the treated cells and dividing by total extractable material.
The latter quantity was measured from the supernatant of an equivalent
density of untreated cells boiled for 30 min in pH 7 CDFB. The material
released showed a broad absorption band from 250-300 nm; the wavelengths
used for calculation were 260 nm, characteristic of nucleic acids, and
280 nm, characteristic of proteins (37). It is of interest that the most
resistant test organism, M. fiovtuitum, did not exhibit release of UV
material, nor did C. paMip&U.o&AJ>, at survival of 2-3 percent, whereas
E. c.oJU did demonstrate release at 23 percent survival. In the case of
conditions where detectable release was observed, confirmation was obtained
in at least two replicates of each experiment.
Corroborative evidence that cytoplasmic leakage may occur, following
exposure of cells to chlorine, was obtained by measuring supernatant TOC.
Results are summarized in Table 27- The primary conclusion from these
111
-------
o
tO
(0
in
c
0)
o
o
u
•^
Q.
O
O.I
0.01
A Control
O 0.25 mg/l Chlorine Dose, 0.0158 *70 Survival
0 0.50 mg/l Chlorine Dose, 0.0021 % Survival
Contact Time 6 min.
I
I
40
240
80 120 160 200
Incubation Time (min)
Figure 51. Growth of E. c.oLL after Chlorination at pH 7
280
112
-------
6
c
O
ID
<0
at
c
0)
O
— O.I
O
O
*-
0.
O
A Control
O 1.2 mg/l Chlorine Dose, 0.012 70 Survival
D 2.4 mg/ I Chlorine Dose, 0.005 % Survival
Contact Time 30 min.
O.OlJ
Figure 52.
12
24 6 8 10
Incubation Time (hr)
Growth of C. paAapA-tlo-6-cA after Chiorination at pH 7
113
-------
£
c
O
CO
(O
w
c
O
O
4~
ex
O
O.I
0.01
A Control
O 3.08 mg/l Chlorine Dose, 18.8 % Survivol
D 6.16 mg/ I Chlorine Dose, 0.06 % Survival
Contact Time 30 min.
I
I
Figure 53. Growth of M.
46 8 10 12
Incubation Time (hr)
after Chlorination at pH 7
14
114
-------
TABLE 26. RELEASE OF UV ABSORBING MATERIAL AFTER CHLORINATION, pH 7
Organism
C. paA.a.p&4Jtot><(
E. coLi
M. faofctuttum
C. paJia.pA 1.97
3.70
0.57
3.19
1.93
LA 1.97
3.70
0.57
3.19
1.93
measured
27. RELEASE OF
Chlorine
dose
'J, 2.1
2.1
6.1
280 nm
2.3 0.0225
* 0.013
23.0 0.058
* 0.115
3.1 0.0515
260 nm
2.3 0.024
* 0.011
23.0 0.0566
* 0.127
3.1 0.0175
0.011
0.017 4.
0.539 5.
0.128 4.
0.044
0.011
0.025 9.
0.593 3.
0.146 4.
0.006
7
0
7
-
5
1
9
-
CELLULAR TOC AFTER CHLORINATION, pH 7
TOC, mg/£
Control Treated
0.7 2.9
5.7 10.3
4.7 6.3
Release as
percent
of total TOC
extractable
35
24
16
115
-------
results was that, upon chlorination all three organisms, C. patia.p&U.o&,
M. fiovtuAtwm, and E. co&t, exhibited an increased release of TOC. The
concentration of TOC from the control cells may represent a low degree
of organic matter in the CDFB, as well as a possible release of material
from the metabolizing cells, or from the cell surface. Whether the
enhanced TOC released from the chlorine treated cells was from the cell
interior or from the cell surface currently is unknown.
The reason the release of UV-absorbing material was not noted with
either the acid-fast organism or the yeast at low survival while E. coti
did release material may relate to the relative impermeability of the
cell wall of these organisms. This impermeability may have prevented
the cytoplasmic contents from leaking into the medium following chlori-
nation. With the yeast, at high chlorine dose where release was observed,
further damage may have occurred to the wall to permit such release.
By comparing the release of percent extractable UV-absorbing mater-
ial to percent extractable TOC, it was noted that larger percentages of
TOC contributing material were released, upon chlorination, than the UV-
absorbing material. The measurement of UV-absorbing material probably
included large amounts of high molecular weight components, such as pro-
teins and nucleic acids. However, the TOC analyses included these, as
well as a large number of low molecular weight substances. The observa-
tion that larger percentages of TOC than UV material were released may
have indicated that smaller molecular weight organic compounds were
preferentially released after chlorination. This observation was in
accord with expectations, where the permeability barrier is only par-
tially damaged, and also agreed with the results of Venkobachar &£ at.
(28) who observed a sequential release of protein, followed by RNA and
DNA, with higher chlorine dosages.
Results of the above studies implicated the cell membrane as the
site of action of chlorine. To verify this hypothesis, it was necessary
to show that membrane-dependent functions, such as respiration, trans-
port, and synthetic processes dependent on transport were also affected
by chlorination.
Studies were performed to ascertain the effect of chlorination on
respiration using E. co&c, C. pawpAJJ-oA-iA, and M. fiosituAttw. Although
some inhibition of oxygen uptake was noted for all three organisms, the
results were inconsistent with the hypothesis that enzymatic inhibition
at a site within the catabolic pathway is the sole lethal lesion (25).
The effects noted could be explained on the basis of damage to cell
transport systems during the lethal event itself.
As previously discussed, the TOC and UV release experiments suggested
damage to the cell permeability barrier as being of significance in the
chlorination process. To ascertain the effects of this disturbance on trans'
port processes associated with the cell membrane, potassium transport was
116
-------
studied. Potassium uptake was selected for investigation since current
evidence suggests that nearly all intracellular potassium is present as the
free monovalent cation, which is accumulated by an active osmoregulatory
mechanism (38). Additionally, potassium transport is readily subject to
measurement using radiotracers. It would be expected that the uptake or
leakage of potassium by the cell would be highly sensitive to the effects
of chlorine at the cell membrane and cell wall, due to the small size of the
K+ ion.
Figures 54 through 56 present the results of the potassium uptake and
retention experiments with E. coti, C. pivia.pi>JJLot>u>, and M. fiosituAtum,
respectively, at pH 7. Upon treatment with chlorine, E. coti showed a defi-
nite decrease in the amount of potassium associated with the cell. This
decrease, about 33 percent following 40 min exposure to chlorine, was rela-
tively greater than that noted for the release of TOC (Table 27) or UV-
absorbing material (Table 26). The loss of cellular potassium may be
attributed to the action of chlorine promoting leakage through the cell
membrane, with a preferential release of lower molecular weight compounds.
With C. pojw.pt>jJioku>, the chlorine treated cells showed an initial
increase in 42K uptake when compared to the control cells (Figure 55). With
prolonged contact, however, the treated cells lost cellular potassium and
potassium levels at the end of the experiment were about 40 percent of the
control cells. This behavior may have been due to a progressive deteriora-
tion in the cell wall leading, at first, to increased permeability of extra-
cellular material; following this, further cell wall or membrane damage may
have occurred leading to leakage of the potassium. It should be noted that
potassium levels in the experimental cells did not decrease below those of
control cells until about 30 min after chlorine addition.
With M. ^o^taitLm, the only effect of chlorine upon 42K transport
appeared to be a substantial enhancement of uptake (Figure 56). This
increase may be due to a partial degradation of the impermeable cell wall
of the mycobacteria, leading to greater transport rates. While there
appeared to be a decline in the rate of uptake after about 30 min of chlorine
contact, no absolute decrease was noted, as in the case of the other organisms.
The differential response of E. zoti from the other two organisms in
the potassium uptake experiments may be explained on the basis of permeabil-
ity differences. The effect of chlorine on all three organisms may be simi-
lar, i.e., change in permeability of the outer layers of the cell. Since
E. colA- has a more permeable cell wall than either C. paM.p&U,o&*A of
M. hoKtuitum (39,40), the effect resulted in leakage of material in the
former organism, whereas an increase in permeability was seen in the other
two organisms.
The release of potassium in E. coti and C. pU,o& may be accom-
panied by the release of anionic compounds, as well, to maintain internal
electrical neutrality. One possibility is that anionic organic compounds,
such as amino acids, or low molecular weight phosphates, are released simul-
taneously with the loss of potassium. This phenomenon, similar to that
117
-------
E
Q.
U
>l
'>
'*-
U
o
o
•o
o
tr
O
Q.
t-
O
U
c
A Control
0 3.0 mg/I Chlorine Dose
Incubation Time (min)
Figure 54. Effect of Chlorine on Cellular Potassium: E. coli, pH 7
118
-------
A Control
O 3.0mg/l Chlorine Dose
20
40 60 80
Incubation Time (min)
Figure 55.
Effect of Chlorine on Cellular Potassium:
pH 7
100
C.
120
119
-------
6
Q.
U
O
O
O
•5
o
01
TJ
0)
O
Q.
b.
O
U
Chlorine
Added
A
O
Control
5.0mg/l Chlorine Dose
I
20
40 60 80
Incubation Time (min)
100
120
Figure 56.
Effect of Chlorine on Cellular Potassium:
pH 7
120
-------
observed in osmotic downshock of E. c.oti (38), would also account for the
observed release of organic materials in this study and for the release of
organic compounds (28,41) and phosphate compounds (23) reported in the
literature.
Within the overall objectives of this study, the potassium uptake experi
ments provide primary evidence for the involvement of the cell membrane as a
site of action of chlorine. Since the potassium transport system appears to
be cellularly controlled (38), and since such control may be exercised at
the level of the cell membrane, these data, in addition to those previously
described, support cell membrane involvement.
After chlori nation, the ability of the three test organisms to synthe-
size protein appeared to be greatly reduced (Figure 57). This reduction in
activity was confirmed, in several cases, by replicate experiments. The
observed reduction is identical to that made by Bernarde at at. (42) who
noted that protein synthesis ceased immediately after treatment of E. c.oti
with chlorine dioxide. The apparent blockage of protein synthesis noted in
Figure 57 does not, per se, mean that intracellular protein synthesis was
stopped. Rather, it may mean that only the ability of the cells to trans-
port the radioactive proline was affected.
As previously noted, the cell permeability barrier is affected as a
result of chlorination. The response in protein synthesis may have been due
to an impaired transport ability. Alternatively, chlorine may have acted
directly on protein synthesis, or the alteration in cellular potassium
levels may have interfered with optimum ionic environments for ribosome
activity.
The observation that protein synthesis was immediately affected after
chlorination was supportive of the findings of the growth experiments
(Figures 51-53) in that no growth of the treated cells occurred after chlori-
nation. It apparently was not necessary for any cell growth or synthesis to
occur before microbial inactivation.
The effect on protein synthesis provides evidence for an effect of
chlorine at a point prior to completion of translation. Using arguments
similar to Benarde at aJt. (42), this damage may be at the level of amino
acid transport, activation, transcription, or translation. Since an effect
on the cell membrane would cause damage to many, if not all, transport pro-
cesses, this damage, which has already been demonstrated, is sufficient to
explain the observed effect.
In the case of C. pafiapAiZoA-Lb , E. co&t, and M. £oKtu
-------
ro
ro
28
E 24
*v
e
CL
o
°2 20
O 16
O
•5
o
(T
J£ 12
_
O
10
c
O
E. coli
A Control
O 2.2 mg/1 Dose
Contact Time lOmin.
84
72
60
48
36
24
12
C. paropsilosis
A Control
O 2.68 mg/ I Dose
Contact Time 30 min.
1 J J
28
24
20
16
12
M. fortuitum
A Control
O 5 mg/I Dose
Contact Time 30 min.
oepo^^Fi ii i ii o
0 10 20 30 40 50 60 0 10 20 30 40 50 60 0 10 20 30 40 50 60
Incubation Time (min)
Figure 57. Effect of Chiorination at pH 7 on Protein Synthesis
-------
2.8
2.4
E
o.
o
10.
O 2.0
>»
">
U
O
o
•o
o
o:
CO
1.6
1.2
O
(A
.£ 0.8
o
0.4
E. coll
A Control
0 2.2 mg/l Dose
Contact Time 10 min.
3.5
3.0
2.5
2.0
1.5
1.0
0.5
C. parapsilosis
A Control
O 2.68 mg/l Dose
Contact Time 30 min.
3.5
3.0
2.5
2.0
I .5
1.0
0.5
I *
M. fortuitum
A Control
O 5 mg/l Dose
Contact Time 30 min.
-A
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Incubation Time (min)
0 10 20 30 40 50 60
Figure 58. Effect of Chlorination at pH 7 on DNA Synthesis
-------
synthesis. Alternatively, chlorine may have directly interfered with the
synthetic process itself, or with the integrity of the DNA template. Obser-
vations by Shih and Lederberg (30) point to the formation of single and
double strand DNA breaks upon chloramine treatment. These breaks might
result in a reduced rate of replication.
The results of the DNA synthesis experiments provide evidence for the
involvement of the cell membrane and possibly the DNA, as targets for chlo-
rine damage. As in the case of the effect on protein synthesis, membrane
damage leading to loss of transport activity would produce the observed
effect.
Lethal Lesions
As noted earlier, discussion of the mode of action of chlorine can be
reduced to two questions. First, is the toxic action dependent on penetra-
tion of the disinfectant molecule into the cell, and, if so, is this the
basis for the increased efficiency of hypochlorous acid, vis-a-vis hypo-
chlorite ion? Second, what are the lethal lesions caused by chlorine?
On the basis of the results, plus a literature review, chlorine appears
to be capable of producing lethal events at or near the cell membrane as
well as in respect to DNA. The primary evidence for the involvement of the
cell membrane as a site for the lethal action of chlorine was gathered from
the results of the TOC release, UV-absorbing material release, potassium
uptake, protein and DNA synthesis, and respiration studies. The results of
the TOC release (Table 27) and the UV-absorbing material release (Table 26)
experiments indicated that, after exposure to chlorine, all three organisms
exhibited some level of release of either or both of these materials; the
only exceptions were in the UV-absorbing material release studies of
C. pata.p&'LloAJA with low dosages of chlorine, and M. fiovtuAJwn. Furthermore,
in the potassium uptake studies, both E. c.oLi and C. pajia.pt>iJLot>4A exhibited
lower levels of intracellular potassium after chlorine treatment (Figures 54
and 55). The increase in potassium levels observed with C. paAa.pt>JULot>Jj>
immediately after chlorine addition and with M. fioJvtnU.um (Figures 55 and 56)
may have been due to an initial degradation of the thicker and less permeable
cell wall in both of these organisms, permitting a greater permeability of
42K; after this initial phase with C. p4JLo&4A, the chlorine may have
degraded the cell membrane further, resulting in a loss of material and
greater levels of cell inactivation. The effects noted on protein and DNA
synthesis (Figures 57 and 58) were consistent with a disruption of the cell
membrane as a permeability barrier.
Supporting evidence for the involvement of the cell membrane as a lethal
target was seen in the growth experiments (Figures 51-53). The immediate
cessation of growth after chlorine treatment was consistent with a direct
physical damage, such as a membrane effect, rather than a lethal point muta-
tion, or similar lesions requiring growth prior to expression.
In summary, therefore, the lethal lesions caused by chlorine appear to
be a disruption of the cell membrane and cell permeability, and possibly a
124
-------
physical damage to the DNA of the cell. Based on the research results pre-
sented, the permeability hypothesis of differential action of HOC! and OC1"
of Chang (18) has been verified in vegetative microorganisms.
Permeability Hypothesis
The primary evidence for the acceptance of the permeability hypothesis
are the results obtained from the isotherm studies (Figures 49 and 50).
From these data, with both C. patiap&JJLotxU and M. ^ontu^utum, it was noted
that more chlorine was found in association with the cell after exposure at
low pH_than after exposure at high pH. Furthermore, the amount of chlorine
found in association with these cells appeared to be based predominantly
upon the theoretical concentration of undissociated hypochlorous acid in
solution, as would be predicted. The results for E. c.oti confirmed the
observations of Friberg (22,23) that more chlorine was found in association
with the cells at high pH than at low pH and, at first glance, would appear
to negate the permeability theory. However, since it has already been
demonstrated that the action of chlorine may have resulted in the loss of
considerable cellular material, especially with the more sensitive E. coti,
it is entirely possible that the increased efficiency of chlorine at low pH
promoted extensive loss of cell material, including associated chlorine,
leading to an underestimate of the latter quantity at low pH. Thus, the
current work appears to extend the permeability theory of chlorine action
to vegetative microorganisms in addition to those previously described.
To answer the questions posed at the outset of this phase of study, the
mode of action of free available chlorine in inactivating microorgansims
appears to be penetration into the cell, at least through the outer cell
wall. Following this, the chlorine may attack the cell membrane and cause
cell permeability disruption, including loss of cytoplasmic constituents,
as well as secondary metabolic disturbances. An additional primary lethal
lesion may be the physical degradation of DNA.
Resistance of Yeast and Acid-Fast Organisms
Considering the possible mechanism whereby chlorine inactivates micro-
organisms, the increased resistance of the acid-fast and yeast organisms may
be contrasted to that of E.
Comparing cell size, C. paMipt> has about 20-100 times the volume
and surface area per cell as E. coti. At pH 9.14 C. pa^ap^o-ix^ takes up
only about 2-10 times the amount of chlorine as E. c.oti\ after correction
for increased leakage, a similar ratio may be calculated for low pH values
as well. It thus appears that one possible reason for the increased resis-
tance of the yeast C. paAapA may be a decrease in its ability to take
up chlorine from solutions of equivalent concentration. This possibility
may be related to a difference in the nature and composition of the cell
wall of the two different organisms. From a priori considerations, Cooke
(40) predicted that yeasts would be more resistant to chlorine than E. c.oli
on the basis of decreased permeability. It would appear, therefore, that
the ability of yeast to take up less chlorine may be due to a heightened
permeability barrier.
125
-------
The decreased ability of bound chlorine to cause lethal events may also
be due to the distinctive features of the cell wall of yeasts, i.e., thick-
ness of the cell wall. A key event in cell inactivation would appear to be
the loss of material from the cell. The thickness of the cell wall in yeasts,
far in excess of that associated with E. coLL, may be much more resistant to
lesions causing cytoplasmic leakage (43).
E. coti and M. fioKtuLtum are of similar size and would, therefore, be
be expected to take up a similar quantity of chlorine. On the other hand, it
is clearly apparent from the results of this study that M. fioKtuitum takes up
considerably less chlorine than does E. coti. This, in all probability,
reflects the very high lipid level in the cell walls of the acid-fast bac-
teria, making them very impermeable to polar molecules, e.g., hypochlorous
acid (39,44).
The impermeable cell wall of the mycobacteria may significantly reduce
the loss of cytoplasmic material. In the UV release experiments, no release
of UV-absorbing material was detected even though 97 percent inactivation
with chlorine occurred (Table 26). One interpretation may be that the cell
wall plays an important role in resistance by retaining small amounts of
material lost through the cytoplasmic membrane.
As a result of the above studies, it may be concluded that the
increased resistance of yeasts, such as C. paAap^lo^,^ , to free available
chlorine is primarily a result of the thickness and rigidity of the cell
wall. The increased resistance to free available chlorine exhibited by
acid-fast organisms, such as M. fiovtuAtum, probably results principally
from the impermeability of the cell wall. Both of these organisms, there-
fore, may be expected to be more resistant than E. c.oLi to chlorine in a
wide variety of circumstances.
126
-------
REFERENCES
1. Bonde, G. J. Bacteriological Methods for Estimation of Water Pollution.
Health Lab. Sci., 3:124, 1966.
2. Engelbrecht, R. S. and E. 0. Greening. Chlorine-Resistant Indicators.
In:. Indicators of Viruses in Water and Food, G. Berg, ed., Ann Arbor
Science Publ., Inc., Michigan, 1978.
3. Shuyal, H. I. Detection and Control of Enteroviruses in the Water
Environment. In: Developments in Water Quality Research, H. I. Shuval,
ed., Ann Arbor Science Publ., Inc., Michigan, 1970.
4. Engelbrecht, R. S., D. H. Foster, E. 0. Greening and S. H. Lee. New
Microbial Indicators of Chlorination Efficiency. EPA-670/2-73-082,
U.S. Environmental Protection Agency, Cincinnati, OH, 1974.
5. Engelbrecht, R. S., B. F- Severin, M. T. Masarik, S. Farooq, S. H. Lee,
C. N. Haas and A. Lalchandani. New Microbial Indicators of Disinfection
Efficiency. EPA-600/2-77-052, U.S. Environmental Protection Agency,
Cincinnati, OH, 1977-
6. Engelbrecht, R. S. and C. N. Haas. Acid-Fast Bacteria and Yeasts as
Disinfection Indicators: Enumeration Methodology. Presented at AWWA
Technology Conference, Kansas City, MO, 1977-
7. Standard Methods for the Examination of Water and Wastewater. Amer.
Pub. Health Assn. Publication Office, Washington, D.C., 1975.
8. Shechter, H. Spectrophotometric Method for Determination of Ozone in
Aqueous Solutions. Water Res., 7:729, 1973.
9. Austin, J. H. and F. W. Sollo. Oxygen Relationships in Small Streams.
Sanitary Eng. Ser. No. 52, Dept. of Civil Eng., Univ. of Illinois at
Urbana-Champaign, 1969.
10. Moeller, T. Inorganic Synthesis. Vol. 5, McGraw Hill, New York, 1957.
11. Dove, M. F. A. and D. B. Sowerby. Isotopic Halogen Exchange Reactions.
In: Halogen Chemistry, V. Cutmann, ed., Academic Park Press, London,
Vol. 1, 1967.
12. Anbar, M. , S. Guttman and R. Rein. The Isotopic Exchange Between
Hypohalites and Halide Ions. II. The Exchange Between Hypochlorous Acid
and Chloride Ions. J. Amer. Chem. Soc., 81:1816, 1959.
127
-------
13. Surucu, F. and C. N. Haas. Inactivation of New Indicator Organisms of
Disinfection Efficiency. Part I. Free Available Chlorine. Presented
at the AWWA Annual Conference, New Orleans, 1976.
14. Sokal, R. R. and F- J. Rohlf. Introduction to Biostatistics. W. H.
Freeman and Co., San Francisco, 1973.
15. Hewes, C. G., H. W. Prengle, C. E. Mauck and 0. D. Sparkman. Oxidation
of Refractory Organic Materials by Ozone and Ultraviolet Light.
DAAK02-74-C-0239, U.S. Army Mobility Equipment R&D Center, Fort Belvoir,
VA, 1974.
16. Stumm, W. Ozone as a Disinfectant for Water and Sewage. J. Boston
Soc. Civ. Eng., 45:68, 1958.
17. Hewes, C. G. and R. R. Davidson. Renovation of Waste Water by Ozonation.
Amer- Inst. Chem. Engr. Symp. Ser., 69:129, 1973.
18. Chang, S. L. Destruction of Microorganisms. J. Amer. Water Works Assn.,
36:1192, 1944.
19. Skvortsova, E. K. and N. S. Lebedeva. Changes in the Activity of
Enzyme Systems in RacWtu.*, ayvth^a.c.o-id^ Spores During Germination and
Due to the Action of Calcium Hypochlorite. Mikrobiologiya, 44:791, 1975.
20. Dennis, W. The Mode of Action of Chlorine on f2 Bacterial Virus During
Disinfection. Sc.D. Thesis, School of Hygiene and Public Health, The
Johns Hopkins University, Baltimore, 1977.
21. Kaminski, J. J., M. M. Huycke, S. H. Selk, N. Bodor and T. Higuchi.
N-Halo Derivatives V. Comparative Antimicrobial Activity of Soft
N-Chloramine Systems. J. Pharmaceutical Sci., 65:1737, 1976.
22. Friberg, L. Quantitative Studies on the Reaction of Chlorine with
Bacteria in Water Disinfection. Acta Pathologica et Microbiologica
Scandinavica, 38:135, 1956.
23. Friberg, L. Further Quantitative Studies on the Reaction of Chlorine
with Bacteria in Water Disinfection. Acta Pathologica et Microbiolgica
Scandinavica, 40:67, 1957.
24. Green, D. E. and P. K. Stumpf. The Mode of Action of Chlorine. J. Amer.
Water Works Assn., 38:1301, 1946.
25. Knox, W. E., P. K. Stumpf, D. E. Green and V. H. Auerbach. The Inhibi-
tion of Sulfhydryl Enzymes as the Basis of the Bactericidal Action of
Chlorine. J. Bacteriology, 55:451, 1948.
26. Sykes, G. Disinfection and Sterilization. 2nd ed., E. and F. N. Spon
Ltd., London, 1965.
128
-------
27. Kulikovsky, A., H. S. Pankratz and H. L. Sadoff. Ultrastructural and
Chemical Changes in Spores of Bac-iliuA cetaoi after Action of Disin-
fectants. J. Appl. Bacteriol., 38:39, 1975.
28. Venkobachar, C. , L. lyengar and A. V. S. Prabhakara Rao. Mechanism of
Disinfection: Effect of Chlorine on Cell Membrane Functions. Water
Res., 11:727, 1977.
29. Rosenkranz, H. S. Sodium Hypochlorite and Sodium Perborate: Prefer-
ential Inhibitors of DNA Polymerase Deficient Bacteria. Mutation Res.,
21:171, 1973.
30. Shih, K.L. and J. Lederberg. Effects of Chloramine on Bac-MuA
Deoxyribonucleic Acid. J. Bacteriol., 125:934, 1976.
31. Shih, K.L. and J. Lederberg. Chloramine Mutagenesis in
Science, 192:1141, 1976.
32. Wlodkowski, T. J. and H. S. Rosenkranz. Mutagenicity of Sodium Hypo-
chlorite for SalmonMa typh-imuA^um. Mutation Res., 31:39, 1975.
33. Fetner, R. H. Chromosome Breakage in tAtc/a fiaba by Monochloramine.
Nature, 196:1122, 1962.
34. von Rosen, G. Breaking of Chromosomes by the Action of Elements of the
Periodical System and by Some Other Principles. Hereditas, 40:258, 1954.
35. Ingols, R. S. The Effect of Monochloramine and Chromate on Bacterial
Chromosomes. Public Works, 89(12) :105, 1958.
36. Mickey, G. H. and H. Holden, Jr. Chromosomal Effects of Chlorine on
Mammalian Cells in vitro. Newsletter of Environ. Mutagen. Soc., 4:39,
1971.
37. Lehninger, A. L. Biochemistry. Worth Pub!., Inc., New York, 1970.
38. Epstein, W. and S. G. Schultz. Ion Transport and Osmoregulation in
Bacteria. In: Microbial Protoplasts, Spheroplasts, and L-Forms,
L. B. Guze, ed. , Williams and Wilkins, Baltimore, 1968.
39. Rat! edge, C. The Physiology of the Mycobacteria. Adv. in Microbial
Phys., 13:115, 1976.
40. Cooke, W. B. The Role of Fungi in Waste Treatment. CRC Critical
Reviews in Environ. Control, 1:381, 1971.
41. Venkobachar, C. , L. lyengar and A. V. S. Prabhakara Rao. Mechanism of
Disinfection. Water Res., 9:119, 1975.
129
-------
42. Bernarde, M. A., W. B. Snow, V. P. Olivieri and B. Davidson. Kinetics
and Mechanism of Bacterial Disinfection by Chlorine Dioxide. Appl.
Micro., 15:257, 1967.
43. Albert, A. Selective Toxicity, 5th ed. Chapman and Hall, London, 1973.
44. Shen, T. H. The Li pine Contents of Mycobacteria and Their Resistance
Against Chemical Disinfectants. J. Shanghai Sci. Inst., Sect. IV,
1:157, 1934.
130
-------
APPENDIX A
THE RELATIVE RESISTANCE OF ACID-FAST AND OTHER
ORGANISMS TO CHLORINATION: A REVIEW
INTRODUCTION
In view of the apparently lower resistance of coliforms to disinfection
than certain pathogens, notably enteric viruses, many workers have attempted
to find a new and more appropriate indicator of fecal pollution and/or dis-
infection efficiency (1). One group of organisms proposed has been the
acid-fast bacteria.
In order to be useful as an indicator of disinfection efficiency, the
proposed organism must be at least as resistant as the most resistant patho-
gen. Studies in this laboratory have demonstrated that the acid-fast organ-
isms are considerably more resistant to chlori nation than either coliforms,
poliovirus type 1 or SalmoneMd. typhA*mu/Lium. Field work in progress also
appears to show that removal and/or die-off of acid-fast bacteria in waste-
water treatment processes and downstream from an outfall is less than con-
ventional indicator organisms, i.e., col i form organisms.
In many parts of the world, Mt/cofaac-CeA^um tubeAcaio^^ is an important
pathogen. Sanitoria wastes and domestic wastewaters can contain significant
densities of the tubercle bacilli (2-4). Natural waters with and without
wastewater effluents also contain significant acid-fast bacteria (5). It
appears that these organisms may be excreted by healthy individuals (6,7).
If acid-fast organisms, other than M. ^abe^cu£o4^6 , as a group are more
resistant to disinfection than M. iube/t-co&M^, then their usefulness as
indicators of disinfection efficiency would be enhanced. It is the purpose
of this review to determine this relative resistance.
RESISTANCE OF ACID-FAST ORGANISMS TO CHLORINATION
The resistance of acid-fast organisms to chlori nation has been studied
by Engelbrecht and co-workers (1,6). Several acid-fast species were isolated
from samples of chlorinated secondary wastewater effluent, including
M. fioKtusitum and M. pkl
-------
resistant to the effects of this form of chlorine than coliform organisms.
RESISTANCE OF MycobacteJtlum tubeAculoAiA TO CHLORINATION
In reviewing the early work on the subject, Greenberg (9) discussed the
survival of tubercle bacilli in wastewater treatment processes. Starting
from the date of his review, a comprehensive search of the literature was
performed for information on the sensitivity of M. tabeAculoA^Li, to chlori-
nation.
Rhines (10), using Avian tubercle bacilli, found that a dose of 91 mg/£
of chlorine effected 99.96 percent inactivation within 30 min. In this and
other examples, it will be assumed that the minimum quantifiable level of
M. tubeAcutoAiA is 1 organism/m£ in order to calculate inactivation levels.
Jensen and Jensen (11), working with secondary effluent obtained from
a tuberculosis sanitorium, observed that contact with a dose of 20 mg/£ of
chlorine for 40 min resulted in over 99.99 percent inactivation of the
tubercle bacilli; a dose of 9 mg/£ for 1 hr resulted in over 99.999 percent
inactivation.
In later work, Jensen (12,13) measured the chlorine dose and residual
necessary to inactivate 99.999 percent of the tubercle bacilli in raw waste-
water and secondary effluent. Chlorine residual measurements were performed
by the "OTD" method.
Heukelekian and Albanese (14) measured the orthotolidine residual neces-
sary to inactivate 99.9 percent of a culture of tubercle bacilli in sterilized
raw wastewater. Pramer at al. (15) measured the "free residual" and ortho-
tolidine chlorine residual needed to inactivate tubercle bacilli in secondary
effluent.
More recently, Bhaskaran it aJL. (3) studied the chlorination of treated
sanitoria effluents. Residuals were measured by an unspecified technique.
Dozanska and Manowska (16) found identical sensitivity of the HsyRv and
BCG strains of M. tubeAc.uXjo&4A to chlorine residuals. The effects of free
and combined chlorine were anlayzed separately. Free chlorine was measured
by the OT and OTA methods while total residuals were controlled using the
DPD procedure. Chlorination results were reported for "100 percent" inacti-
vation.
In a later study, using the BCG strain, Dozanska and Manowska (17)
studied the chlorination of untreated raw wastewater monitored by the OTA
method until "destruction of tubercle bacilli approached 100 percent."
A summary of the data reported by the above researchers appears in
Table 1.
Figure 2 summarizes the data from Table 1 for raw wastewater, Imhoff or
septic tank effluent, compared with the inactivation of M. fio/LtuLtum in
132
-------
TABLE 1
REPORTED CONDITIONS FOR TOO PERCENT INACTIVATION OF M. tubuuuuJbo&JA BY CHLORINE
Suspending Raw Wastewater, Imhoff or
Menstrum Septic Tank Effluent
Secondary Effluent
Tap Water or Buffer
Dose
CO
CO
Musehold
63 ppm
(18)
2
hr
Sollazzo (19)
20 ppm 2 hr
33.6 ppm 25 min
Jensen (12,13)
25 ppm 1 hr
Rhines (10)
91 ppm 30 min
Jensen & Jensen (11)
20 ppm 40 min
9 ppm 1 hr
Jensen (12,13)
10 ppm 1
hr
Kopeloff & Davidoff (20)
30-50 ppm 5 min
Green (21)
125 ppm
65 ppm
35 ppm
10 min
30 min
4 hr
Residual Jensen (12,13)
3.5 ppm 1 hr
Dozanska & Manowska (16)
(see Figure 5)
Dozanska & Manowska (17)
5 ppm 1 hr
Jensen (12,13)
0.4-0.95 ppm 1
hr
Heukelekian & Albanese
2 ppm 1 hr
1 ppm 2 hr
Bhaskaran
-------
phosphate buffer by both free-available chlorine (1) and chloramines (8).
Figure 3 depicts the resistance of the tubercle bacilli in secondary efflu-
ent and, again, compared to inactivation of M. ^on^tuJMm in phosphate buffer.
The most comprehensive study reported has been that of Dozanska and
Manowska (16). They investigated the inactivation of tubercle bacilli to
chlorine in tap water and wastewater, using four pH values and two tempera-
tures. The results of their work, compared with results from this laboratory
for both free-available chlorine and chloramines, are shown in Figures 4 and
5, respectively.
DISCUSSION OF COMPARATIVE RESISTANCE TO CHLORINE
As Figures 2 and 3 indicate, the data for the inactivation of
M. -tubetca&M-ci by chlorine in raw wastewater and secondary effluent
reported by others are bracketed by our observations with M.
This would seem to indicate that there is no evidence to support the propo-
sition that the tubercle bacilli are more resistant to chlorination than
the more resistant acid-fast organisms such as M. fiofctuAttm.
As would be expected, the data given for chlorine dosage in Table 1
fall significantly above our data using residual measurements. In most of
the studies with raw wastewater, Imhoff and septic tank effluent and secon-
dary effluent the dosages are in such a range that little or no free resid-
ual would be expected. The only inconsistent data are those of Green (21)
which seem to be at chlorine dosages in excess of what would be required to
reach the breakpoint in tap water or buffer. However, inasmuch as the exact
details of his experimental systems were not given, further comment is
unwarranted.
Dozanska and Manowska (16) appear to have carried out the most thorough
studies. Examination of Figures 4 and 5, which compares their data for 100
percent inactivation of M. tubeAcutoA'ii, with data of Severin (8) and
Engelbrecht at at. (1) for 99.9 percent inactivation of M. ^otvtuAMw and
M. pkleA., seems to indicate that while M. fiotituAtum and M. tabeAc.alo-i>-Li, show
similar slopes of time vs. concentration, M. tubnnc.uJLot> is more sensitive
to chlorine and chloramine than M.
As can be seen in Figure 4, M. phte^i, while obeying a similar time-
concentration slope as the other acid-fast bacteria, is less resistant than
M. ;£ubeAcu£o4'6i. For a given contact time, it appears that approximately
twice the free chlorine residual required for M. pht&L is needed for inacti
vation of M. tub for a given contact period.
In the case of combined chlorine, there is still greater discrepancy.
With a contact time in excess of 30 min, more than 2.5 times the chlorine
residual required for inactivation of M. pklni is necessary to inactivate
M. tu.beAcuZ.oA-u> . On the other hand, about 125 percent of the combined
chlorine residual required to inactivate M. tubeAc.aio^-Lf> is necessary to
inactivate M. fiofctuttum for a given contact time.
134
-------
RELATIVE RESISTANCE TO OTHER TOXIC AGENTS
In an extensive review of mycobacteria inactivation, Croshaw (22) noted
that a number of researchers believe that "atypical" mycobacteria have
greater heat tolerance than M. tu.beAc.ulo*iJ> . Croshaw, in analyzing the infor-
mation, restated a hypothesis that resistance to chemical antagonists was
approximately proportional to cellular lipid content. This would imply a
higher resistance by the high lipid tubercle bacilli than by the more sapro-
phytic species of mycobacteria.
At least one study appears to have shown that natural die-off of the
tubercle bacilli is greater than other acid-fast bacteria. Savov (23)
found that, under similar conditions in Lowenstein-Jensen medium, M. tub&ic.u-
Lot>JJ> survived up to 2.5 yr whereas M. avi.um and M. fiovtuLtum each survived
up to 13 yr.
Further support of this conclusion arises from the observation of
Beerwerth (7) that Runyon's Group I, which included the pathogenic myco-
bacteria, were not present when numerous other mycobacteria were isolated
from surface waters and "pasture watering tanks."
CONCLUSIONS
The data gathered in this analysis indicate that at least some acid-fast
organisms present in domestic wastewater may be more resistant to chlorina-
tion than M. tubeA.cu£.oA . If these organisms, such as M. ^o^taitum, were
present and detectable at significantly high levels in wastewaters and raw
water supplies, then the acid-fast group of organisms could be useful as
indicators of the disinfection of tubercle bacilli.
If, however, organisms such as M. fiositiutum were absent, and only less
resistant acid-fast organisms, such as M. phla-i, were present, the acid-fast
organisms may not be particularly useful as indicators of the disinfection
of tubercle bacilli .
The apparent resistance of M. ^o^taitum to chlorination, because of its
lower lipid content, is in apparent contradiction to the hypothesis pre-
sented by Croshaw (22). Thus, there should be no attempt to extrapolate
the findings of this review to relative resistance towards other disinfecting
agents, e.g., ozone, iodine, etc.
135
-------
BIBLIOGRAPHY
1. Engelbrecht, R. S., D. H. Foster, E. 0. Greening and S. H. Lee. New
Microbial Indicators of Wastewater Disinfection Efficiency. EPA-670/2-
73-082, U.S. Environmental Protection Agency, Cincinnati, OH, 1974a.
2. Saldanha, F. L., S. N. Sayyid and S. R. Kulkarni. Viability of M.
tubeAcuJtoA-u, in the Sanitorium Sewage. Ind. Jour. Med. Res., 52:1051,
1964.
3. Bhaskaran, T. R., M. N. Lahiri and B. K. Ghose Roy. Effect of Sewage
Treatment Processes on Survival of Tubercle Bacilli. Ind. Jour. Med.
Res., 48:790, 1960.
4. Ancusa, M. and W. Terbancea. Vorkommen von Tuberkulosebakterien in
Vorflutern. Zeirschrift fur die Gesamte Hygiene und ihne Grensgeb,
16:913, 1970.
5. Tacquet, A., H. Leclerc and B. Devulder. Epidemiologie des Mycobac-
teries Atypiques. Proc. 3rd Intern. Colloquium on the Mycobacteria.
The Genus Mycobacteria, Prince Leopold Inst. of Tropical Medicine,
Antwerp, 1973.
6. Engelbrecht, R. S., D. H. Foster, M. T. Masarik and S. H. Lee. Detec-
tion of New Microbial Indicators of Chlorination Efficiency. AWWA
Dallas Water Technology Conference, 1974b.
7- Beerwerth, V. W. Mykobakterien in Viehtranken und Oberflachengewassern.
Deutsche Tierarztliche Wochenschrift, 80:393, 1973.
8. Severin, B. M.S. Special Problem, Dept. of Civil Engineering, Univ.
of Illinois at Urbana-Champaign, 1975.
9. Greenberg, A. E. and E. Kupka. Tuberculosis Transmission by Wastewater-
A Review. Sewage and Industrial Wastes, 29:524, 1957-
10. Rhines, C. The Longevity of Tubercle Bacilli in Sewage and Stream
Water, Amer. Rev. of Tuberc., 31:493, 1964.
11. Jensen, K. A. and K. E. Jensen. Occurrence of Tubercle Bacilli in
Sewage and Experiments on Sterilization of Tubercle Bacilli-Containing
Sewage with Chlorine. Acta Tuberc. Scand., 16:217, 1942.
12. Jensen, K. E. Undersgelser Over Forrkomst og Uskadeliggorelse of
Virulente Tuberkelbacterier i Spildevand. I Kommiss. Hos. Gee. Gads
Forlung, Copenhagen, Denmark, 1948.
136
-------
13. Jensen, K. E. Presence and Destruction of Tubercle Bacilli in Sewage.
Bull. World Health Organization, 10:171, 1954.
14. Heukelekian, H. and M. Albanese. Enumeration and Survival of Human
Tubercle Bacilli in Polluted Waters. II. Effect of Sewage Treatment
and Natural Purification. Sewage and Industrial Wastes, 28:1094, 1956.
15. Pramer, p., H. Heukelekian and R. A. Ragotzkie. Survival of Tubercle
Bacilli in Various Sewage Treatment Processes. I. Development of a
Method for the Quantitative Recovery of Mycobacterium from Sewage.
Public Health Repts., 65(27):851, 1950.
16. Dozanska, W. and W. Manowska. The Requirements for the Disinfection of
Sewage Containing Tubercle Bacilli. Roczniki Panstwowy Zaklad Higieny,
21:253, 1970.
17. Dozanska, W. and W. Manowska. Study on the Effectiveness of Sewage
Disinfection Containing Tubercle Bacilli Without Biological Treatment.
Roczniki Panstwowy Zaklad Higieny, 23:353, 1972.
18. Musehold, P. Uber die Widerstandfahigkert der mit dem Lungenauswurfher-
ausbeforderter Tuberkelbazillan im Abwassern, im Flusswasser und im
Kultwerten Boden., Arb. Kaiser!. Gesundh., 17:56, 1900.
19. Sollazzo, G. La Disinfezione della Acque Luride Delgi Ospedali per
Tuberculotici Mediante Cloro Gassoso. Annali di Igeni, 41:745, 1931.
20. Kopeloff, N. and L. M. Davidoff. The Action of Chlorine Compounds on
B. Tuberculosis. Proc. Soc. Exp. Biol. and Med., 28:7, 1930.
21. Green, C. A. Cited in J. R. Trueman, 1971, The Halogens, W. B. Hugo,
(ed.), 1964.
22. Croshaw, B. The Destruction of Mycobacteria. In: Inhibition and
Destruction of the Microbial Cell, W. B. Hugo (ed.), Chap. 10, Academic
Press, London, 1971.
23. Savov, N. Studies on the Resistance of Mycobacterium. Veterinarno
Meditsinski Nauki, 10(2):39, 1973.
24. Berg, G. Virus Transmission by the Water Vehicle. III. Removal of
Viruses by Water Treatment Procedures. Health Lab. Sci., 3:170, 1966.
137
-------
10
O)
c:
S-
o
QJ
re
a>
OJ
0.1
0.01
pH 7
20°C
Coxsackie virus A2
M.
(Engelbrecht
(Engelbrecht at aJL.)
kAdenovirus 3
10 100
Contact Time (min)
1000
Figure 1
Free Chlorine Residuals and Contact Times Necessary for
99.9 Percent Kill of M. ^ofituuitwrn and M. pkl&i., Compared
with Data for Other Organisms Assembled by Berg (24)
138
-------
100
10
CO
CD
(O
3
-o
to
-------
100
o Jensen (1948, 1954)
A Bhaskaran (1960)
o Heukelekian (1956)
10
en
E
CO
O)
C£
C1J
S-
o
M. ^o^ta-itim pH 7
99.9 percent 20°C
combined chlorine
(Severin)
pH 7
99.9 percent 20°C
free available chlorine
(Engelbrecht &t at.}
0.1
Figure 3.
10 100
Contact Time (min)
Comparison of Data for Inactivation of M.
1000
s in Secondary Effluent
-------
10
pH 9
en
s-
o
o
O)
JD
(O
OJ
(U
M. pht&i pH 7
99.9 percent
(Engelbrecht at at.)
M. faoKtiuAum pH 7
99.9 percent
(Engelbrecht at at.)
pH 7
pH 6
0.1
10 100
Contact Time (min)
Figure 4. Comparison of Data of Dozanska and Manowska (1970) for
100 Percent Inactivation of M. tub&ic.utoi>iA by Free
Available Chlorine at 20°C
1000
141
-------
10
en
E
Ol
s-
o
-a
O)
-
O
O
M.
pH 7
99.9 percent
(Severin)
pH 6
M. pkt&i
pH 7
99.9 percent
(Severin)
0.1
I
10
Figure 5.
100
Contact Time (min)
1000
Comparison of Data of Dozanska
100 Percent Inactivation of M.
Chloramines at 20°C
and Manowska
tabeAc.ato^^ii>
(1970)
by
for
142
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
• REPORT NO.
EPA-600/2-79-091
3. RECIPIENT'S ACCESSI ON- NO.
4. TITLE AND SUBTITLE
ACID-FAST BACTERIA AND YEASTS AS INDICATORS OF
DISINFECTION EFFICIENCY
5. REPORT DATE
August 1979 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
Richard S. Engelbrecht, Charles N. Haas,
Jeffrey A. Shular, David L. Dunn, Dipak Roy, Ajit
Lalchandani, Blaine F. Severin, and Shaukat Farooq
8. PERFORMING ORGANIZATION REPORT NO
PERFORMING ORGANIZATION NAME AND ADDRESS
Department of Civil Engineering
University of Illinois
Urbana, Illinois 61801
10. PROGRAM ELEMENT NO.
1CC824; SOS 2; Task 04
11. CONTRACT/GRANT NO.
EPA-IAG-D6-0432
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory—Cin.,OH
Office of Research & Development
U. S. Environmental Protection Agency
Cincinnati, Ohio 45268
13.r
VERED
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer: Raymond H. Taylor (513) 684-7204
Prior report covering 5/1/72-4/30/75 published as EPA 600/2-77-052
16. ABSTRACT
Since the coliform group of organisms is considered to be less resistant
to chlorine than some bacterial and viral pathogens, the utility of both
yeast and acid-fast organisms as potential indicators of disinfection efficiency
was evaluated. In most laboratory studies these two groups of organisms were
represented by Candida parapsilosis and Mycobacterium fortuitum, respectively.
The relative resistance of the test organisms to free chlorine was: acid-fast>
yeast>coliforms. The increased chlorine resistance of these organisms appeared
to be the result of the thickness and impermeability of the cell wall. It was
concluded that the primary mode of action of chlorine in disinfection was
disruption of the cell membrane with a resultant change in cell permeability
and physical damage to the cell DNA.
Resistance to ozonation was also studied. Variations in pH between 5 and
10 did not significantly affect organism survival of either yeasts or acid-fast
organisms using constant ozone residuals, while increasing temperatures increased
the inactivation of both organisms. Large volume sampling and enumeration
techniques were developed for the yeasts and acid-fast organisms using membrane
filtration which enabled the enumeration of these organisms at the relatively low
densities found in finished drinking water.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Disinfection, Potable water, Microorganism
control (water), Coliform bacteria,
Escherichia coli. Chlorination, Mycobac-
terium, Yeasts, Indicator species,
Ozonization, Water treatment
Coliforms
13B
8. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report/
UNCLASSIFIED
!1. NO. OF PAGES
155
20. SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
143
JUS GOVERNMENT PRINTING OFFICE 1979 -6 57 - I 46 / 546 1
------- |