Mitigation of sewer crown corrosionvia competitive inhibition of thiobacilli

Item Type text; Thesis-Reproduction (electronic)

Authors Padival, Navnit Ajit, 1966-

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.

Download date 25/05/2018 10:34:09

Link to Item http://hdl.handle.net/10150/278007

INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some

thesis and dissertation copies are in typewriter face, while others may

be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and

continuing from left to right in equal sections with small overlaps. Each

original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

University Microfilms International A Bell & Howell Information Company

300 North Zeeb Road. Ann Arbor. Ml 48106-1346 USA 313/761-4700 800/521-0600

Order Number 1346425

Mitigation of sewer crown corrosion via competitive inhibition of thiobacilli

Padival, Navnit Ajit, M.S.

The University of Arizona, 1991

U M I 300 N. Zeeb Rd. Ann Arbor, MI 48106

MITIGATION OF SEWER CROWN CORROSION

VIA COMPETITIVE INHIBITION OF THIOBACILLI

by

Navnit Ajit Padival

A Thesis Submitted to the Faculty of the

DEPARTMENT OF CIVIL ENGINEERING AND ENGINEERING MECHANICS

In Partial Fulfillment of the Requirements For the Degree of

MASTER OF SCIENCE WITH A MAJOR IN CIVIL ENGINEERING

In the Graduated College

THE UNIVERSITY OF ARIZONA

1 9 9 1

2

STATEMENT BY AUTHOR

This thesis has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgement of source is made. Requests for permission of extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgement the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED:

APPROVAL BY THESIS DIRECTOR

This thesis has been approved on the date shown below:

f O 1 4 4 1

Dr. Robert G. Arnold Assistant Professor of Civil Engineering

and Engineering Mechanics

Date

ACKNOWLEDGEMENTS

I would like to thank the University of Arizona, especially the Civil Engineering

Department for their financial assistance during this project. Also, special thanks to the

Los Angeles County Sanitation Districts for their partial funding of the research.

I am indebted to my advisor Bob Arnold. He is an excellent person, friend, and

mentor.

Finally, I would like to thank my committee members Dr. Sierka and Bruce

Logan for their input.

Dedicated

to my

Parents

Rajni and Ajit

5

TABLE OF CONTENTS

Page

LIST OF FIGURES 10

LIST OF TABLES 19

ABSTRACT 20

CHAPTER

1. INTRODUCTION 21

1.1 Sewer Crown Corrosion 21

1.1.1 General 21 1.1.2 Sulfide Corrosion: The Problem 21

1.2 Sulfate Reduction and Cycle in Waste Collection System 24

1.3 The Los Angles County Case Study 32

1.3.1 Intensification of Corrosion Problem 32 1.3.2 Treatment Efforts at LACSD to

Reduce Corrosion 34 1.3.3 EPA Recommended Treatment 35 1.3.4 Investigation at Academic Institutions 35 1.3.5 Other Related Research 37

1.4 Thiobacilli: Sulfur-Oxidizing Bacteria 39

1.4.1 General 39 1.4.2 Environment and Habitat 40 1.4.3 Principles of Respiration and Phosphorylation 41 1.4.4 Oxidation of Sulfur Compounds in Chemolithotrophs .. 42 1.4.5 Practical Applications of Thiobacilli 42

6

TABLE OF CONTENTS (continued)

CHAPTER Page

1.5 Competition in Microbial Ecosystems 46

1.5.1 General 46 1.5.2 Competition: Survival of the Fittest 46 1.5.3 Basic Theoiy of Continuous Cultures 47 1.5.4 Competition and Selection in a Chemostat 47 1.5.5 Competition Related Research 50

1.6 Objectives 57

1.6.1 General 57 1.6.2 Isolation Experiments 58 1.6.3 Stoichiometric Experiments 58 1.6.4 Two-Organism Competition Experiments 58 1.6.5 Effect of pH on Competition Experiments 59 1.6.6 Effect of Different Heterotrophs in

Competition Experiments 59

2. MATERIALS AND METHODS 61

2.1 Microorganisms 61

2.2 General Description of Media Composition 61

2.3 Isolation of Yeast from the Sewer Crown 62

2.3.1 General 62 2.3.2 Sampling Techniques 62 2.3.3 Enrichment Steps 68 2.3.4 Serial Dilutions 68

2.4 Stoichiometric Experiments 70

2.4.1 General 70 2.4.2 Determination of "Critical " N/C or N/S Ratio 70

2.5 Experimental Procedures 72

2.5.1 The General Procedure 72

7

TABLE OF CONTENTS (continued)

CHAPTER Page

2.5.2 Variations in the General Procedure 73

2.5.2.1 Variations in Influent Ammonia-Glucose Ratio 73

2.5.2.2 Variation in pH 74 2.5.2.3 Variations in Competitor Identity 74

2.6 Analytical and Chemical Techniques 76

2.6.1 Microbial Growth Measurements 76 2.6.2 pH Measurement 77 2.6.3 Total Sulfate Measurements 78 2.6.4 Ammonium Ion Measurements 78 2.6.5 Dissolved Organic Carbon Analyses 79

2.7 Immunofluorescence: Fluorochrome Antibody Conjugates 80

2.7.1 General 80 2.7.2 Fluorescent Antibody Procedure 80

3. RESULTS 82

3.1 Isolation of Heterotrophic Competitor 82

3.1.1 Field Observations: Crown pH and Corrosion 82 3.1.2 Growth of Crown Samples in the Laboratory 82

3.2 Critical N/C and N/S Ratios for Nutrient Utilization in Batch Cultures 84

3.2.1 General 84 3.2.2 Ammonia and Glucose Utilization Rates 84 3.2.3 Ammonia and Thiosulfate Utilization Rates 85 3.2.4 Effect of Initial pH on Growth of Yeast 91

3.3 Titration Curves 93

I?

8

TABLE OF CONTENTS (continued)

CHAPTER Page

3.4 Competition Studies 94

3.4.1 General 94 3.4.2 Low-pH Competition Experiments 94

3.4.2.1 Variations in Influent Ammonia/Glucose 94 3.4.2.2 Variations in Culture pH 107 3.4.2.3 Variations in Competitor Identity 113

3.4.3 Mid-pH Competition Experiments 119

3.4.3.1 Variations in Influent Ammonia/Glucose Feed Ratios 119

3.4.3.2 Variations in Competitor Identity 136

4. DISCUSSIONS 143

4.1 Sample Collection and Cultivation 143

4.1.1 Crown Conditions 143 4.1.2 Enrichment and Isolation from Crown Samples 144

4.2 Stoichiometric Studies 146

4.2.1 Ammonia-to-Glucose Utilization Studies 146 4.2.2 Ammonia-to-Thiosulfate Utilization Rates 147

4.3 Two-Organism Competition Studies 147

4.3.1 Low-pH Competition Experiments 147 4.3.2 Mid-pH Competition Experiments 151

5. CONCLUSIONS 153

5.1

5.2

Isolation of Sewer Crown Samples 153

Stoichiometiy of Ammonia-to-Glucose Utilization Rates 152

9

TABLE OF CONTENTS (continued)

CHAPTER Page

5.3 Stoichiometry of Sulfur Oxidation 154

5.4 Competitive Inhibitory Control of Thiobacilli 155

REFERENCES CITED 157

LIST OF FIGURES

10

Figure Page

Figure 1-1. The physiological basis of hydrogen sulfide corrosion in concrete sewers 23

Figure 1-2. Steps in the assimilatory reduction of sulfate to hydrogen sulfide for use in bio-synthetic reactions 26

Figure 1-3. Pathway of dissimilatory sulfate reduction in Desulfovibrio species and "hydrogen cycling" hypothesis. 1, lactate dehydrogenase, membrane-bound, H-acceptor not known; 2, pyruvate ferredoxin oxidoreductase; 3, phosphotransacetylase; 4, acetate kinase; 5, ATP sulfurylase; 6, pyro-phosphatase; 7, APS reductase; 8, sulfite reductase; 9, cytoplasmic hydrogenase; 10, periplasmic hydrogenase 27

Figure 1-4. Recycling pathway of the dissimilatory reduction of sulfite to sulfide in an anaerobic environment 28

Figure 1-5. Sulfate reduction mechanism found in sewer systems 29

Figure 1-6. Effects of liquid-phase pH on the relative concentrations of sulfide species, E^S and HS" 31

Figure 1-7. Actual and predicted depth of corrosion of a 12-foot diameter reinforced concrete sewer pipe. Pipe measured section was a section of joint outfall "B" of the LACSD collection system 33

Figure 1-8. Principles of electron transport and phosphoiylation in thiobacilli 43

Figure 1-9. The reductive pentose phosphate or Calvin cycle for the conversion of C02 into glucose. For each turn of the cycle, one mole of hexose is synthesized from 6 COz at the expense of 18 ATP and 12 NADH, this energy being derived from the oxidation of inorganic substrate in chemolithoautotrophs 44

11

LIST OF FIGURES (continued)

Figure Page

Figure 1-10. Routes of the oxidation of sulfur compounds in chemolithoautotrophs. 1, oxidation of sulfide to polysulfide sulfur [S];, 2, conversion of elemental sulfur to polysulfide sulfur; 3, thiosulfate-oxidizing multi-enzyme complex; 4, sulfur oxidase; 5, sulfite oxidase; 6, APS reductase; 7, ADP-sulfurylase 8, electrons are transferred to oxygen via components of the respiratory chain 45

Figure 1-11. Hypothetical specific growth rates of organism 1 and 2. The text explains why organism 1 will wash out of the continuous culture .. 49

Figure 2-1. Steps used for the isolation and identification of sewer crown samples 69

Figure 3-1. Growth of E. coli on glucose in pure, batch cultures. Cell densities at stationary phase was a function of the initial nitrogen concentration when total ammonia added was < 2.0 x 10'3 M. At higher nitrogen concentrations, growth proved to be glucose-limited 86

Figure 3-2. Relationship between culture optical density at stationary phase and initial nitrogen concentrations in pure, batch culture of E. coli 87

Figure 3-3. Growth of acidophilic yeast on glucose in pure, batch cultures. Cell density at stationary phase was a function of the initial nitrogen concentration when total ammonia added was < 5.0 x 10"* M. At higher nitrogen concentrations, growth proved to be glucose-limited 88

Figure 3-4. Relationship between culture optical density at stationary phase and initial nitrogen concentrations in pure, batch culture of acidophilic yeast 89

Figure 3-5. Growth of T. neapolitamis on thiosulfate in pure, batch cultures. Cell number at stationary phase was a function of the initial nitrogen concentration when total ammonia added was < 3.0 x 10"* M. At higher nitrogen concentrations, growth proved to be thiosulfate-limited 90

12

Figure

Figure 3-6.

Figure 3-7.

Figure 3-8.

Figure 3-9.

Figure 3-10.

Figure 3-11.

Figure 3-12.

LIST OF FIGURES (continued)

Page

Relationship between culture cell density at stationary phase and initial nitrogen concentrations in pure, batch cultures of T. neapoliianus 92

Acid production as measured by pH decrease in medium containing pure cultures of T. neapoliianus using thiosulfate as substrate 93

Titration curve used for the periodic base addition for the maintenance of pH at 2.50 or 3.25 in continuous-flow cultures of T. thiooxidans and yeast 95

Titration curve used for the periodic base addition for the maintenance of pH at 6.80 in continuous-flow culture of T. neapoliianus, yeast, and E. coli 96

Cell number versus time. Numbers of yeast and T. thiooxidans in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the steady concentrations of T. thiooxidans. When growth was nitrogen-limited, effect on the population of T. thiooxidans were much more pronounced 98

EfQuent sulfate concentration as a function of time and cultures conditions in continuous-Qow cultures of T. thiooxidans and yeast plus T. thiooxidans. Notice that the addition of yeast and glucose greatly diminished the production of sulfate and presumably the activity of T. thiooxidans 99

EfQuent ammonia concentration as a function of time and culture conditions in continuous-Qow cultures of T. thiooxidans and yeast plus T. thiooxidans. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of yeast and high N/C feed conditions did not 100

13

LIST OF FIGURES (continued)

Figure Page

Figure 3-13. Periodic base addition for the maintenance of pH at 3.25 in continuous-flow cultures of T. thioaxidans and yeast Notice that the rate of acid production was adversely affected by the addition of glucose and yeast at about 250 hours. In the low N/C culture (low-nitrogen feed or culture that was expected to be nitrogen-limited), acid production was almost completely interrupted by the shift in feed conditions .... 101

Figure 3-14. Cell number versus time. Numbers of yeast and T. thioaxidans in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the steady concentrations of T. thioaxidans. When growth was nitrogen-limited, effect on the population of T. thioaxidans were much more pronounced 102

Figure 3-15. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. thioaxidans and yeast plus T. thioaxidans. Notice that the addition of yeast and glucose greatly diminished the production of sulfate and presumably the activity of T. thioaxidans 103

Figure 3-16. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. thioaxidans and yeast plus T. thioaxidans. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of yeast and high N/C feed conditions did not 104

Figure 3-17. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving yeast and T. thioaxidans. Results confirm that growth was glucose-limited in the high-N/C, but not the low-N/C experiments 105

Figure 3-18. Periodic base addition for the maintenance of pH at 3.25 in continuous-flow cultures of T. thioaxidans and yeast. Notice that the rate of acid production was adversely affected by the addition of glucose and yeast at about 250 hours. In the low N/C culture (low-nitrogen feed or culture that was expected to be nitrogen-limited), acid production was almost completely interrupted by the shift in feed conditions 106

14

LIST OF FIGURES (continued)

Figure Page

Figure 3-19. Cell number versus time. Numbers of yeast and T. thiooxidans in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the steady concentrations of T. thiooxidans. When growth was nitrogen-limited, effect on the population of T. thiooxidans were much more pronounced 108

Figure 3-20. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. thiooxidans and yeast plus T. thiooxidans. Notice that the addition of yeast and glucose greatly diminished the production of sulfate and presumably the activity of T. thiooxidans 109

Figure 3-21. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. thiooxidans and yeast plus T. thiooxidans. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of yeast and high N/C feed conditions did not 110

Figure 3-22. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving yeast and T. thiooxidans. Results confirm that growth was glucose-limited in the high-N/C, but not the low-N/C experiments Ill

Figure 3-23. Periodic base addition for the maintenance of pH at 2.50 in continuous-flow cultures of T. thiooxidans and yeast. Notice that the rate of acid production was adversely affected by the addition of glucose and yeast at about 275 hours. In the low N/C culture (low-nitrogen feed or culture that was expected to be nitrogen-limited), acid production was almost completely interrupted by the shift in feed conditions 112

Figure 3-24. Cell number versus time. Numbers of yeast and T. thiooxidans in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the concentrations of yeast. Notice the temporary effect on the population of T. thiooxidans. Increase in thiobacilli cell densities was seen after about 80 hours following the cross inoculation of yeast 114

15

LIST OF FIGURES (continued)

Figure Page

Figure 3-25. EfQuent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. thioaxidans and yeast plus T. thioaxidans. Notice that the addition of yeast and glucose temporarily reduced the production of sulfate and presumably the activity of T. thioaxidans 115

Figure 3-26. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. thioaxidans and yeast plus T. thioaxidans. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of yeast and high N/C feed conditions did not 116

Figure 3-27. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving yeast and T. thioaxidans. Results confirm that growth at high-N/C and low-N/C was nitrogen-limited, not glucose-limited 117

Figure 3-28. Periodic base addition for the maintenance of pH at 2.00 in continuous-flow cultures of T. thioaxidans and yeast. Notice that the rate of acid production rate in both low N/C and high N/C cultures were temporarily affected by the addition of glucose and yeast at about 300 hours 118

Figure 3-29. Cell number versus time. Number of T. thioaxidans in continuous-flow reactors during competition experiments. Addition of mixed seed lowered the steady concentration of T. thioaxidans at pH = 3.25, but did not at pH = 2.00 120

Figure 3-30. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. thioaxidans and a mixed-seed plus T. thioaxidans. Notice that the addition of the contaminant and glucose temporarily reduced the production of sulfate and presumably the activity of T. thioaxidans at pH = 2.00 121

16

LIST OF FIGURES (continued)

Figure Page

Figure 3-31. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. thiooxidans and yeast plus T. thiooxidans at pH 2.00 and 3.25. Results confirm that cultures were nitrogen-limited 122

Figure 3-32. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving the mixed-seed and T. thiooxidans. Results confirm that growth at pH 2.00 and 3.25 was nitrogen-limited, not glucose-limited 123

Figure 3-33. Periodic base addition for the maintenance of pH at 2.00 and 3.25 in continuous-flow cultures of T. thiooxidans and a mixed-seed from the sewer. At pH = 3.25 rate of acid production was adversely affected by the cross inoculation of mixed-seed, but not at pH = 2.00 124

Figure 3-34. Cell number versus time. Numbers of yeast and T. neapolitamis in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the steady concentrations of T. neapolitanus. When growth was nitrogen-limited, effect on the population of T. neapolitanus were much more pronounced 126

Figure 3-35. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. neapolitanus and yeast plus T. neapolitanus. Notice that the addition of yeast and glucose greatly diminished the production of sulfate and presumably the activity of T. neapolitanus 127

Figure 3-36. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. neapolitanus and yeast plus T. neapolitanus. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of yeast and high N/C feed conditions did not 128

HTv

17

LIST OF FIGURES (continued)

Figure Page

Figure 3-37. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving yeast and T. neapolitanus. Results confirm that growth was glucose-limited in the high-N/C, but not the low-N/C experiments 129

Figure 3-38. Periodic base addition for the maintenance of pH at 6.80 in continuous-flow cultures of T. neapolitanus and yeast Notice that the rate of acid production was adversely affected by the addition of glucose and yeast at about 325 hours. In the low N/C culture (low-nitrogen feed or culture that was expected to be nitrogen-limited), acid production was almost completely interrupted by the shift in feed conditions 130

Figure 3-39. Cell number versus time. Numbers of yeast and T. neapolitanus in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the steady concentrations of T. neapolitanus. When growth was nitrogen-limited, efFect on the population of T. neapolitanus were much more pronounced 131

Figure 3-40. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. neapolitanus and yeast plus T. neapolitanus. Notice that the addition of yeast and glucose greatly diminished the production of sulfate and presumably the activity of T. neapolitanus 132

Figure 3-41. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. neapolitanus and yeast plus T. neapolitanus 133

Figure 3-42. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving yeast and T. neapolitanus. Results confirm that growth was glucose-limited in the high N/C, but not the low N/C experiments 134

LIST OF FIGURES (continued) 18

Figure

Figure 3-43.

Figure 3-44.

Figure 3-45.

Figure 3-46.

Figure 3-47.

Figure 3-48.

Page

Periodic base addition for the maintenance of pH at 6.80 in continuous-flow cultures of T. neapolitanus and yeast Notice that the rate of acid production was adversely affected by the addition of glucose and yeast at about 325 hours. In the low N/C culture (low-nitrogen feed or culture that was expected to be nitrogen-limited), acid production was almost completely interrupted by the shift in feed conditions

Cell number versus time. Numbers of E. coli and T. neapolitanus in continuous-flow reactors during competition experiments. The addition of E. coli and glucose lowered the steady concentrations of T. neapolitanus. When growth was nitrogen-limited, effect on the population of T. neapolitanus were much more pronounced

135

137

Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. neapolitanus and E. coli plus T. neapolitanus. Notice that the addition of E. coli and glucose greatly diminished the production of sulfate and presumably the activity of T. neapolitanus 138

Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. neapolitanus and E. coli plus T. neapolitanus. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of E. coli and high N/C feed conditions did not 139

Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving E. coli and T. neapolitanus. Results confirm that growth was glucose-limited in the high-N/C, but not the low-N/C experiments 140

Periodic base addition for the maintenance of pH at 6.80 in continuous-flow cultures of T. neapolitanus and E. coli. Notice that the rate of acid production was adversely affected by the addition of glucose and E. coli at about 350 hours 141

19

LIST OF TABLES

TABLE Page

Table 1-1. Thiobacilli species identification 40

Table 2-1. General composition of defined chemical medium used for the growth of T. thiooxidans 63

Table 2-2. General composition of defined chemical medium used for the growth of T. neapolitanus 63

Table 2-3. Composition of defined chemical medium used for the isolation of low- and mid-pH organisms from the sewer crown samples 64

Table 2-4. Composition of defined chemical media used in stoichiometric experiments involving E. coli growing in the mid-pH range 64

Table 2-5. Composition of defined chemical media used in stoichiometric experiments involving yeast growing in the low and mid-pH range 65

Table 2-6. Composition of defined chemical media used in stoichiometric experiments involving T. neapolitanus 65

Table 2-7. Composition of defined chemical media used in T. thiooxidans and yeast competition experiments 66

Table 2-8. Composition of defined chemical media used in T. neapolitanus and yeast competition experiments 66

Table 2-9. Reactor details and design criteria for low-and mid-pH experiments 67

Table 2-10. A summary of experiment run, parameters measured, and independent variables used in low-and mid-pH competition experiments 75

Table 3-1. Comparison of measured parametric levels and percentage difference in CSTR competition experiments before and after glucose/competitor addition 142

20

ABSTRACT

Inhibition of biogenic acid production by thiobacilli was investigated by

encouraging the growth of potential competitors. Two-organism competition experiments

(Thiobacillus sp. vs. heterotrophic competitor) were conducted in a bench-scale,

continuous-flow reactor. Results were sensitive to the influent ammonia/glucose ratio.

Under nitrogen-limiting conditions, the cell concentration of T. thiooxidans or T.

neapolitanus and acid production rates were reduced by about two orders of magnitude.

Under carbon-limiting conditions, only modest reductions in the thiobacilli cell density

and effluent sulfate were observed. In general, encouragement of microbial competition

can interrupt acid production by thiobacilli and may inhibit crown corrosion in concrete

sewers.

21

CHAPTER 1

INTRODUCTION

1.1 Sewer Crown Corrosion.

1.1.1 General. Thiobacillus-induced corrosion of concrete sewer pipe, sewage

collection systems, and treatment plants has caused concern among agencies in the

United States, Germany, Japan, the USSR, and many other places in the world. Within

the Los Angeles County Sanitation Districts (LACSD) sewer system alone, unanticipated

repair and replacement activities attributable to crown corrosion will cost almost 200

million dollars prior to the turn of the century (Morton, 1989; Sponsellar, 1988).

1.1.2 Sulfide Corrosion: The Problem. In all corroded sections of the LACSD

collection system, the pH of the concrete surface is exceptionally low (pH < 4.0). In

general, the surface pH is lowest at the sewer crown and gradually increases with

distance from the crown apex (Milner, 1989). The regions of the pipe that are

periodically washed with bulk sewage during high flow exhibit the highest pH,

approaching neutral values (Milner, 1989; Price, 1989). The greatest degree of

degradation is generally observed at the crest of the sewer, where the pH is often less

than 4.0.

22

Corrosion of concrete pipe is linked to naturally occurring events involving the

transformation of inorganic sulfur. This four-step transformation mechanism is illustrated

in Figure 1-1: (i) Sulfide [S(-II)] is generated in an anaerobic slime layer by sulfur-

reducing bacteria (SRB), and (ii) S(-II) diffuses into the sewage. Among the reduced

sulfur species is hydrogen sulfide gas (HjS), which eventually volatilizes and is

transported to the sewer crown, (iii) At the crown, HjS is bacterially oxidized to sulfuric

acid (H2SO4) by the sulfur-oxidizing bacteria, and (iv) the acid chemically attacks the

concrete of the sewer pipe. This can lead to excessive corrosion, sewer deterioration,

and eventual sewer collapse (Holder et al., 1983; EPA, 1974; Parker, 1951).

The obligately aerobic, chemolithoautotrophic acidophile ThiobacUlus thioaxidans

is thought to be primarily responsible for the sulfide oxidation/acid production which

leads to rapid corrosion of concrete sewers. These bacteria assimilate carbon dioxide

(COj) for biosynthesis, and oxidize inorganic, sulfur-containing compounds (such as HjS)

for cellular energy generation.

T. thioaxidans are present in large numbers in the LACSD sewage collection

system (Price, 1989). Similar findings in the sewer systems of the Hamburg Municipal

Drainage Department have been reported (Milde et al., 1983). They noted that new

sewers coated with epoxy resin were corroding at alarming rates (3/4 inch/year). Large

numbers of T. thioaxidans were present at highly corroded sites. T. thioaxidans greatly

out-numbered the other thiobacilli in these highly corroded sections of the Hamburg

collection systems (Sand and Bock, 1984).

23

CONCRETE PIPE

HaS—»2H • SO,

OXIC SEWER ATMOSPHERE

BULK UQUlO

H,S (o)

HS' AND H,S HS AND HjS

SO*"* AND ORGANIC CARBON NUTRIENTS

SUME (ANOXIC ZONE)

Figure 1-1. The physiological basis of hydrogen sulfide corrosion in concrete sewers.

24

In order to establish environmental conditions favorable to the acidophilic

thiobacilli, the pH of the sewer pipe walls is gradually neutralized by the mid-pH species

of thiobacilli, such as T. neapolitanus. The mid-pH thiobacilli can decrease the concrete

pH from a value of 11 to 5.5 or so, upon which the acidophilic species can become

established and lower the crown pH further. However, it is believed that drop in pH

from 11 to neutral values is a non-biological process (weathering).

1.2 Sulfate Reduction and Cycle in Waste Collection System.

Sulfate (SO/2) is present in all natural waters, ranging from small concentrations

of 1.6 mg/L in groundwater found near Oakland, California, to 200 mg/L found in the

Colorado River waters near Lees Ferry, Arizona (Snoeyink and Jenkins, 1980). In

wastewaters, these concentrations increase by an average of 100 mg/L (Neal, 1964).

Consequently, wastewater SO/2 levels seldom limit HjS production by sulfate-reducing

bacteria or, ultimately, the activity of thiobacilli.

Under anaerobic conditions, such as those which prevail in slime layers beneath

the stagnant or slow-moving sewage, sulfate [S(VI)] is dissimilatively reduced by the

sulfate-reducing bacteria. During this dissimilative process, [S(VI)] serves as the terminal

electron acceptor for electrons derived from the oxidation of organic matter. This is a

distinct process from assimilative S(VI) reduction, which is necessary for the synthesis of

the amino acids cysteine and methionine. Dissimilative reduction is an energy-generating

process carried out exclusively by SRBs. All organisms are capable of assimilative sulfate

reduction (EPA, 1974).

25

The pathway for assimilative reduction of sulfate to hydrogen sulfide for use of

biosynthetic reactions is shown in Figure 1-2. A total of eight electrons are transferred

as S(VI) is reduced to S(-II). S(VI) is converted initially to adenylsulfate, and the

process is completed via the catalytic reduction of sulfite (SO, 2) to S(-II). The S(VI)-to-

S(IV) reaction utilizes energy stored in phosphate bonds, reductant, and catalysis from

three specific enzymes. The steps from S(IV) to S(-II), a six electron transfer, are

catalyzed by a complex flavometallo-protein, sulfite reductase (Stanier, 1976).

Dissimilative reduction is accomplished via successive transfer of four pairs of

electrons from organic substrate to SO/2. The oxidation of lactate to acetate coupled

with S(VI) reduction to S(-II) is schematically shown in Figure 1-3. Four enzymes are

required for this reduction, (i) ATP sulfurylase, (ii) pyrophosphorylase, (iii) APS

reductase, and (iv) sulfite reductase (Gottschalk, 1986). The last three electron-pairs are

transferred via sulfite reductase in a three-step recycling process as shown in Figure 1-4

(Stanier, 1976).

In the sewer environment, two genera (seven are recognized) of sulfate-reducing

bacteria, Desulfotomaculum and Desulfovibrio, are primarily responsible for S(-II)

production via dissimilative SO;2 reduction (Posgate, 1959; Pfennig, 1971; Stanier, 1976;

Sand, 1987). They are strict anaerobes and are unique due to their massive S(-II)

producing ability. Many other species can reduce partially oxidized sulfur intermediates,

but their importance to S(-II) production in sewers is considered to be relatively minor.

The mechanism by which hydrogen sulfide (HjS) is generated in sewers is depicted in

Figure 1-5 and by the simplified reactions (EPA 1974) that follow:

26 so4

2-

sulfate adenylyitrans­ferase

•ATP r'

,S»©-® adenylyi sulphate

•ATP

adenylylsulfate kinase r-

S*-ADP

3'-phosphoadenylvl sulphate

NADPH phosphoadenylyl-sulfate reductase

NADP

so,2-

suiphite reductase

3'-phosphoAMP

3NADPH

3NADP*

Figure 1-2. Steps in the assimilatoiy reduction of sulfate to hydrogen sulfide for use in bio-synthetic reactions.

mttabraar 27

2CHS—CHOH-COOH

2CHj—CO-COOH

bydrotenur 2CHj—CO—CoA

r*-:coA

2CH,—COO©

I 2CHj—COOH 3H,0-

ATT biltnct

2 Ucute

soj-

T" ZATP

_L

• 2 urate • 2CO}

dmuosnotie mcduntsn t t

3ATP

Figure 1-3. Pathway of dissimilatory sulfate reduction in Desulfovibrio species and "hydrogen cycling" hypothesis. 1, lactate dehydrogenase, membrane-bound, H-acceptor not known; 2, pyruvate ferredoxin oxidoreductase; 3, phosphotransacetylase; 4, acetate kinase; 5, ATP sulfurylase; 6, pyro-phosphatase; 7, APS reductase; 8, sulfite reductase; 9, cytoplasmic hydrogenase; 10, periplasmic hydrogenase.

28

3 SO," *2e . S,0l-+2' ~s,or-+2e' v \ sol" so r

1 )

»-s 2-

Figure 1-4. Recycling pathway of the dissimilatory reduction of sulfite to sulfide in an anaerobic environment.

29

AIR

H,S ENTERING AIR

WASTEWATER BULK LIQUID

DISSOLVED OXYGEN <0.1 mg/L. DISSOLVED SULFIDE PRESENT. HS" • HaS

or.

DEPLETION OF 0;

DIFFUSION OF S04"* AND NUTRIENTS

DIFFUSION OF SULFIDE INTO THE STREAM

Figure 1-5. Sulfate reduction mechanism found in sewer systems.

I

SO;2 + 2C + 2HzO <

30

> 2HCO," + HjS

HjS is relatively a weak acid. It can dissociate to form HS" and S"2 as shown by

the following reactions. Thus, the speciation of inorganic sulfide present in the waste

stream depends primarily on the pH of the bulk sewage. The pK,'s for the

deprotonation of HjS in

HjS <=========> H+ + HS* pK,, = 7.0

HS" <=========> H+ + S-2 pK^ = 13.9

t

water are 7.0 and 13.9, respectively (Stumm, 1981). The pH range of most wastewaters

is between 5.4 and 8.4 (Snoeyink, 1980). Therefore, S*2 comprises a negligible percentage

of the total inorganic sulfide present in typical wastewaters. Figure 1-6 illustrates the

distribution of inorganic sulfide between H2S and HS" for pH values between 5.0 and 9.0.

Clearly the dominant form of total inorganic sulfide is HjS at pH values below 7.0.

H2S is the only volatile form of inorganic sulfide and can escape from the liquid

waste stream and enter the head space above the wastewater. HjS (g) diffuses into

biofilms coating sewer walls where it is oxidized to sulfuric acid (H2SO4) by the thiobacilli.

This H2SO4 is neutralized by calcium carbonate in the sewer concrete as shown by the

following reaction (Sand, 1987):

HjSO, + CaCOj > CaSO< + HzO + C02

31

pH If pK « 7.0

' s

' • T

\ s

s V \ \ \ \

k-

\ i *

\ *

\ \

1 ^ \

1 -2JO -U) -j$ -.4 -Z 0 +2 t.4 +*

p H - p K *S *10 *ZJ0

Figure 1-6. Effects of liquid-phase pH on the relative concentrations of sulfide species, HjS and HS".

32

The CaSO< produced combines with H20 to form gypsum. The result is a loss

of concrete integrity. Initially, as the biofilm grows, the crown pH will remain fairly

stable (around pH 7.0), but as concrete alkalinity is titrated and the rate of bacterially

mediated biogenic acid generation increases, the crown pH decreases (around pH 4.0).

As a consequence, the observed rate of crown corrosion increases dramatically (around

pH 2.0). Eventually corrosion penetrates to reinforcing steel, undermining the stuctural

integrity of the pipe. Crown pH can drop as low as 1.0.

13 The Los Angeles County Case Study.

1.3.1 Intensification of Corrosion Problem. LACSD has observed an

acceleration of sewer crown corrosion rates in a significant proportion of its collection

system, over the past 10 to IS years. Subsequently, they also noticed that the sulfide

concentration in the sewage had increased more than twice from the late 1970s to early

1980s. The increase in conosion rate coincided with an increase in sulfide concentration,

as measured at the Joint Water Pollution Control Plant (JWPCP). Inspections indicated

that approximately 40 miles of large-diameter sewers were severely corroded and required

repair or replacement. The corrosion rate measured exceeded values, predicted by an

empirical model developed by EPA (1974), by a factor of 3 to 4 (Figure 1-7; EPA, 1974).

Recent reductions in metals concentrations might have permitted enhanced

growth and activity among sulfate-reducing bacteria within the County (Sponseller, 1988).

In other words, implementation of the categorical industrial waste pretreatment standards

have increased the rate of sulfate reduction, with subsequent implications for acid

Corrosion Comparison 144" Unit 1A

33

S 3.0-o c

z 2 2.0-<o o cc cc O

£ 1.0-o MEASURED

PREDICTED X H CL ili o

30 40 10 YEARS OF SERVICE

Figure 1-7. Actual and predicted depth of corrosion of a 12-foot diameter reinforced concrete sewer pipe. Pipe measured section was a section of joint out fall "B" of the LACSD collection system.

34

generation and corrosion rates within the collection systems. Removal of metals might

have also limited the formation of metal-sulfide precipitates, thus increasing both the

driving force for the evolution of HjS into the headspace, and the headspace

concentration expected at vapor-liquid equilibrium. Both these factors may have lead to

increased rates of bacterially-catalyzed crown acidification and corrosion.

Additionally, high metals concentration may have inhibited acid production by the

sulfur-oxidizing thiobacilli in the sewer crown, diminishing their ability to oxidize HjS to

HjSO^. Therefore, decreased metals concentration in the sewer waste stream may have

aggravated the problem of acid production within the sewer environment in several ways.

Other explanations for the increased crown corrosion rates are possible.

1.3.2 Treatment Efforts at LACSD to Reduce Corrosion. Attempts to impede

crown corrosion have included: (i) inhibition of sulfur-reducing bacteria, (ii) oxidation of

aqueous S(-IE) prior to its transport to the sewer head space, and (iii) mechanical

removal of acid-producing bacteria from the sewer crown surface. In-house research by

LACSD includes the use of a sulfide chamber to empirically determine rates of concrete

corrosion as a function of factors such as pH, HzS concentration, bacterial numbers, and

pipe material. They also have operated pilot-scale anaerobic digestors to anticipate the

possible effects of various corrosion-inhibiting additives on the downstream treatment

processes.

LACSD attempted to mitigate the problem specifically by four methods: (i) slug

dosing with sodium hydroxide (NaOH) to raise crown pH above 13, (ii) adding hydrogen

peroxide (H202) to the bulk liquid-phase; peroxide served both as a bactericide for

35

sulfate-reducers and a strong oxidant for liquid-phase sulfides, (iii) aerating sewage to

suppress growth of sulfate-reducers and chemically oxidize S(-II), and (iv) adding ferrous

chloride (FeClj) to the bulk-liquid. FeG2 promotes the formation of insoluble iron-

sulfide complexes. Results suggest that addition of FeCl2 provides an effective, though

expensive means for controlling the flux of HiS to the sewer crown (Redner, 1988; Jin,

1987). Other methods have' proven to be expensive and only marginally successful.

1.3.3 EPA Recommended Treatment. USEPA's recommended treatments for

controlling sulfide generation and liquid-phase S(-II) concentrations consists of air/oxygen

injection and chemical addition to wastewater, including the addition of an alternative

electron acceptor (to SO/2) for bacterial respiration. Air/02 increases the bulk liquid-

phase dissolved oxygen concentration above 1.0 mg/L and rapidly oxidizes S(-II) to

thiosulfate (EPA, 1985). Effective chemical additives include chlorine, hydrogen

peroxide, nitrate, and metal salts. Chlorine and hydrogen peroxide oxidize S(-II) and

prevent the release of HzS into the sewer headspace. Metal salts form insoluble metal-

sulfide complexes. Sulfate reduction is generally absent under denitrifying (bacterially

catalyzed N03' reduction) conditions.

13.4 Investigations at Academic Institutions. LACSD has funded several

academic institutions to investigate solutions to the problem. The California Institute of

Technology is looking at oxidation of S(-II) with cobalt tetrasulfophthalocyanine

(CoTSP). This compound in small amounts can catalyze the oxidation of large amounts

of sulfide. The kinetics of the reaction are important due to the volatility of HjS. If

CoTSP can catalytically oxidize S(-H) prior to its volatilization, the gaseous HjS

36

concentration in the sewer atmosphere could be held below the level required for cell

metabolism of the sulfur-oxidizing bacteria. Caltech has also plans for investigating the

catalytic S(-II) destruction with ultrasound (Hoffmann, 1989).

The University of California-Los Angeles is investigating the inhibition of sulfate-

reducing bacteria without affecting the downstream wastewater treatment processes.

Inhibitors such as oxyanions, arquad compounds, and formaldehyde have been tested.

UCLA conducted some SRB enumerations, compared lactate to propionate using SRBs,

and have identified Desulfobulbus propionicus in the system (Redner, 1988).

The University of Southern California is developing a corrosion-monitoring device

that will enable the employees at LACSD to quantitatively measure corrosion rates. The

research team at USC is also developing a mechanical system for periodically flushing the

crown in order to disrupt corrosion.

At the University of Arizona, researchers are concentrating on the ecology and

biochemistry of the acid-producing, sulfur-oxidizing bacteria. Inhibition of crown

corrosion using transition metals, weak organic acids, and strong oxidants was studied by

Milner, Sima, and Burris, respectively (1989; 1990; 1991). Results have proven

promising. Milner (1989) investigated the chemical inhibition of acid production by

acidophilic thiobacilli using a variety of metals and chelators (alone and in combination).

Hg(II) and Ag(I) showed toxic effects at concentrations of 10"5 M and below; Cd(II) and

Zn(II) and NTA were non-inhibitory at mM concentrations. Combination experiments

showed both synergistic and antagonistic effects in experiments involving Mo(VI); Cu(II);

37

Co(II); Zn(I3), and Cu(II)-EDTA; Cu(II)-Fe(III); Cu(II)-Co(II). Many of these

inhibitors either completely blocked or delayed the onset of acid production.

Burns (unpublished, 1990) investigated the chemical inhibition of acid production

by acidophilic thiobacilli using a variety of strong oxidants and organics (alone and in

combination). KMn04, and peroxymonosulfate showed inhibitory effects at

concentrations of 10'5 M and above. These oxidants delayed the onset of acid

production.

Sima (1990) investigated the inhibition of T. thiooxidans by externally provided

weak organic acids. The mechanism of toxicity was pursued using a battery of tests such

as measurement of intracellular ATP, 02 utilization rate, COz uptake, and growth rate.

Acetate, formate, pyruvate, succinate, oxaloacetate, malate, and citrate showed inhibitory

effects at concentrations of 10"3 M. The suspected mechanism of weak-acid inhibition

involves either lowering the cytoplasmic pH or interfering with the maintenance of a

transmembrane protonmotive force. A portion of the work conducted at U of A

laboratories is the subject of this report. Our primary objective was the elimination of

the thiobacilli via competition experiments. Details are provided in the chapters to

come.

13.5 Other Related Research. The first Thiobacillus-induced corrosion,

identification of acid-producing bacteria, and hypothesis of a corrosion mechanism were

attributed to Parker (1945, 1951). Parker studied the presence of S(-II) in sewage,

transport of HjS across the liquid-gas interface, acid production at the sewer crown, and

the chemistry of corrosion. Kempner (1966) studied acid production by Thiobacillus

38

thioaxidans in liquid-phase with elemental sulfur (S°) as the electron acceptor. He found

that they could lower the pH to 0.52.

Research conducted in the United States, Germany, Japan, and the USSR, was

motivated by the extent of crown corrosion reported in the sewer collection systems of

those countries. Milde (1983) isolated and enumerated thiobacilli from the crown and

bulk liquid phase from sewers. It was determined that T. thioaxidans was primarily

responsible for acid production and concrete. Milde also discovered that aeration of the

bulk liquid reduced the activity of thiobacilli and mitigated low-pH conditions in the

crown. Accumulation of S° at the crowns led to the conclusion that elemental sulfur was

the energy source for the thiobacilli as opposed to HjS, S203'2 or organic polysulfides.

It was suggested that the oxidation of HjS to S° was an abiotic process.

In 1984, Pohl studied the degradation of methionine (a sulfur-containing amino

acid) and volatile organic polysulfides such as methyl and ethyl mercaptan, etc. Previous

work done by Sivela (1975) had shown that thiobacilli (unidentified species) could grow

on such compounds. Pohl (1984) found that relatively high concentrations of protein in

sewage increased the availability of organic sulfides to the thiobacilli.

The results of Sand (1987) partially contradicted those by Pohl (1984). HjS was

considered the primary source of electrons for acidophilic thiobacilli in the Hamburg

sewer environment, and thiosulfate (S2Oj*) was the primary substrate for non-acidophilic

thiobacilli. Mercaptans and organic sulfides did not support the growth of thiobacilli

tested. Sand's experiments were conducted in a controlled breeding chamber into which

the substrate was added. Corrosion of concrete pipe core samples in the chambers could

39

be accelerated eight-fold over corrosion rates observed in the sewer (Sand, 1984 and

1987). The presence of S° on core samples placed in the breeding chamber was also

observed.

1.4 Thiobacilli: Sulfur-Oxidizing Bacteria.

1.4.1. General. Thiobacilli are rod-shaped, gram-negative cells that, with few

exceptions, utilize a single polar flagellum for motility. They are 1-2 uM in length (Sima,

1989). Fourteen species were identified in the Bergev's Manual of Determinative

Bacteriology (Vishniac, 1974 and Kelly 1981; 1982).

The genus Thiobacillus includes strict autotrophs, facultative autotrophs, and a

single heterotrophic species. The obligately aerobic, chemolithoautotrophic thiobacilli

obtain their energy by oxidizing reduced forms of inorganic sulfur. These include sulfide

IS(-H)], elemental sulfur (S°), thiosulfate (S2O32), polythionates, and sulfite (SO/2)

(Vishniac, 1974). This is accompanied by acid generation.

Experiments conducted and reported in this work involved two species of sulfur-

oxidizing bacteria. These are an acidophilic, autotrophic thiobacilli, Thiobacillus

thiooxidans and a mid-pH range species Thiobacillus neapolitanus. Because the mid-pH

thiobacilli are thought to be responsible for establishing environmental conditions for the

acidophiles, we decided to conduct experiments involving both the low and mid-pH

thiobacilli. If acid production by the non-acidophilic thiobacilli can be mitigated, the low-

pH communities will find it difficult to establish themselves. Tests involving these two

40

species simulated heavily corroded and lightly or non-corroded sewer conditions. Shown

below are growth conditions for the two species:

1.4.2 Environment and Habitat. Thiobacilli can generally be isolated from any

aerobic environment in which contains reduced inorganic compounds of sulfur and iron.

These organisms have been found in rivers, canals, estuaries, acid sulfate soils, hot acid

springs, sewer collection systems, mine drainage, etc. (The Prokarvotes. 1977).

Table 1-1. Thiobacilli species identification.

Species

Identification

pH Range

for

Growth

Carbon and

Energy Source

Generation time

at 25°C (Hrs.)

T. thioaxidans 0.5-5.5 Obligate autotroph

lithotroph (S°)

12-15

T. neapolitanus 4.5-8.5 Obligate autolithotroph

lithotroph (S203'2)

14-16

They flourish under a wide range of pH conditions. In most acidic environments

with low-metals concentrations, T. thioaxidans predominates. A notable exception is

acid-mine drainage, where T. ferroaxidans is the predominant species. Viable cultures

have been recovered from environments of pH < 2.0 (Price, 1989; Espejo and Romero,

1987; Espejo et al., 1988; Kempner, 1966). On the other hand, T. neapolitanus inhabits

41

environments in the pH range of 4.5 to 8.5 (Bergev's Manual of Determinative

Microbiology: Vishniac, 1974 and Kelly 1981; 1982).

1.4.3 Principles of Respiration and Phosphorylation. During the oxidation of

inorganic sulfur compounds, released electrons enter a system of membrane-bound

proteins, collectively called the electron transport chain (ETC). These proteins are

located in the cytoplasmic membrane of the organism. Energy liberated during electron

transport is converted into adenosine-5'-triphosphate (ATP) by oxidative

phosphorylation. The mechanism by which electron transport is coupled to oxidative

phosphorylation (chemosmotic hypothesis) was Erst proposed by Mitchell (1961). Energy

derived from the passage of electrons along the transport chain is thought to drive the

translocation of protons from the cell's cytoplasm to the periplasmic space or cell

surroundings. The process establishes an electrochemical protonmotive force which is

comprised of two components, (i) a gradient in proton concentration and (ii) a difference

in electrical potential, or charge gradient across the cytoplasmic membrane. The

protonmotive force drives hydrogen ions back to the cytoplasm via a protein complex

known as ATPase (F, F„ ATPase). Retranslocated protons drive membrane-associated

functions, including phosphorylation of adenosine-5'-diphosphate (ADP) to produce ATP

(Hinkle, 1978). The exact mechanism of the phosphorylation is not known. ATP is

useful form of stored energy for many endergonic biochemical functions (Gottschalk,

1986); reducing power (as well as ATP) is required for the assimilation of C02 by

autotrophic bacteria.

42

Figure 1-8 illustrates the mechanisms, principles, and importance of electron

transport in thiobacilli (Kelly, 1982). The reactions are summarized in Figure 1-9. The

products of these reactions are substrates for additional biosynthetic reactions. The

synthesis of one molecule of glucose from C02 involves the hydrolysis of 18 molecules

of ATP. Reducing power for this biosynthesis is provided in the form of NADH.

1.4.4 Oxidation of Sulfur Compounds in Chemolithotrophs. The biochemistry

of dissimilative sulfur oxidation is not entirely established A recently proposed pathway

of sulfide [S(-n)], elemental sulfur (S°), and thiosulfate (S2Of) oxidation is illustrated in

Figure 1-10 (Stanier, 1976). S(-II) is oxidized to polysulfide in a reaction that requires

glutathione (see step 1 in Figure 1-10). S° is also oxidized in this manner. All electrons

released from the oxidation are transferred to Oz via the proteins of the electron

transport chain, as described above. The system for thiosulfate oxidation yields an end

product of two sulfate molecules formed for every thiosulfate molecule oxidized

(Gottschalk, 1986). Figure 1-8 illustrates the mechanism of electron transport from

reduced sulfur compounds in thiobacilli (Kelly, 1982). ATP is generated from the

transfer of electrons and resulting proton motive force. Reducing power as NADH or

NADPH is produced by reversing the direction of electron transport, an ATP-dependent

process. Reductant is required by the Calvin cycle to reduce COz for cellular

biosynthesis.

1.4.5 Practical Applications of Thiobacilli. Jack et al. (1979) reported that oil

recovery from the Athabasca Oil Sands in Alberta, Canada resulted in the production of

coke rich in trace metals. This coke was found to be amenable to the recovery of

43

NAD *.02 OR NOj, NO J, N»0, NO

Thiobaciltus denitrificans

Thiobaciltus novetlus, Thiobaciltus Aj. T. neapolitanus, T. ferrooxidans

Figure 1-8. Principles of electron transport and phosphorylation in thiobacilli.

6ADP (6ATP)^_V6

6-Ribulose. SP

4 Xylulose SP

2 Ribose SP 2 Xylulose SP

[2 Glyctraldetiyde 3P*| 2 Sedonepiulose 7Pj"*

wA 2 Sedoheplulose 1,7I\

^S^f20HAPl-4-1.2E4P J

0

I e CO, I i—_-J ^—• 12 Glycerate 3P -

RuOPCase ^^•WADP

13 Glycerol* t, 3f^

[2 Xylulose SP 1 •|_2 Erythrose 4Pj

12 NADH+ I2H* 12N?

12 Glycerolde'hyde 3P (G3P)

•3 Glyceroldehyde 3P •3 OihydroKyocelone 3P

Fructose 1.0P.

3 Fructose 6P

I 1 Glucose 6P

Figure 1-9. The reductive pentose phosphate or Calvin cycle for the conversion of COz into glucose. For each turn of the cycle, one mole of hexose is synthesized from 6 C02 at the expense of 18 ATP and 12 NADH, this energy being derived from the oxidation of inorganic substrate in chemolithoautotrophs.

45

Figure 1-10. Routes of the oxidation of sulfur compounds in chemolithoautotrophs. 1, oxidation of sulfide to polysulfide sulfur [S];, 2, conversion of elemental sulfur to polysulfide sulfur; 3, thiosulfate-oxidizing multi-enzyme complex; 4, sulfur oxidase; 5, sulfite oxidase; 6, APS reductase; 7, ADP-sulfurylase 8, electrons are transferred to oxygen via components of the respiratory chain.

46

vanadium by a simple leaching technique that involved the growth of T. thioaxidans on

elemental sulfur and subsequent addition of microorganisms to the coke (Zajic et al.,

1978). T. thioaxidans detoxified and maintained vanadium in solution by the reduction

of vanadium from vanadium (V) to vanadyl ion, vanadium (IV).

In mining applications, ThiobacUlus ferroaxidans has been widely used to leach

metals such as copper, zinc, and uranium from ores (Smirnov, 1963; Postgate, 1982).

1.5 Competition in Microbial Ecosystems.

1.5.1 General. In nature, a multiplicity of different organisms are involved in

cycling the various elements. In order to come to some understanding of their place and

function, it has become an established practice to isolate microorganisms in pure cultures

and to study their properties under well-defined conditions in the laboratory. In these

studies, the application of continuous culture techniques has received considerable

attention since the approach allows the study of microbial growth under conditions

similar in almost all respects to those found in nature.

1.5.2 Competition: Survival of the Fittest Darwin's work on natural selection

emphasized the importance of competition between different species. Competition

results from dependence of two and/or more species on a common factor such as food,

light, space, or some other limiting resource. Consumption of this common factor by

each species limits its availability to the other, so that the growth rates of both the

organisms are affected negatively.

47

In a competitive situation, we are interested in investigating whether either

species enjoys a natural advantage. A little reflection suggests that the species with the

fastest growth rate should do better, since it will be able to utilize more of the limiting

factor than the slower-growing organism by virtue of its rapid growth. This faster-

growing organism will exhaust such a limiting factor and may, over a sufficient period of

time, eliminate the slower-growing organism.

1.5.3 Basic Theory of Continuous Cultures. A continuous culture system

consists of a culture vessel, the contents of which are well mixed, to which fresh medium

is added at a constant feed rate while the working volume is kept constant via continuous

removal of effluent at the same rate. The system is thus open. All the components

required for growth are present in excess accept for one: the growth-limiting substrate.

Ideally, this cultivation system allows continuous, exponential growth (though sub-

optimal) of microorganisms under constant environmental conditions, i.e., in a steady-

state, with no fluctuations in either the substrate concentration or the microbial

population (Monod, 1950; Novick and SzUard, 1950).

1.5.4 Competition and Selection in a Chemostat A chemostat is a reactor for

the cultivation of microorganisms in which steady-state conditions can be established. In

a chemostat, it is generally assumed (Monod, 1950; Herbert, Elsworth & Telling, 1956;

Powell, 1965) that, the relationship between the growth rate of organisms and the

concentration of growth-limiting nutrient can be represented adequately by a Michaelis-

Menten type of equation:

[D] = u = * [S/(K,+S)J

48

where, u is the specific growth rate; u,, is the maximum rate of growth of the organisms

in the growth medium; S is the extracellular concentration of growth-limiting substrate

in the culture and K, is a half-saturation constant (numerically equal to the extracellular

concentration of growth-limiting nutrient at u = 0.5 u ). Under steady-state conditions

in a chemostat, u is equal to the dilution rate (D). Dilution rate characterizes the

holding time or processing rate in a chemostat.

The following comments pertain to a two-organism competitive situation

established in a chemostat with dilution rate D under limiting-substrate concentration.

We shall assume that specific growth rates for organism 1 and 2 are u, and Uj,

respectively. As indicated in Figure 1-11, the concentrations of the growth-limiting-

substrate within the vessel will be Sv Now, if organism 2 is introduced into the

chemostat, it will grow initially at the rate u2. Since u2 > D, the population of species

2 will increase with time, causing the substrate concentration to fall. With S < S„ the

specific growth rate of species 1 is smaller than D, so that organism 1 will begin to

washout. This trend will continue until the substrate concentration reaches Sj, at which

point species 2 attains steady-state while species 1 continues to washout. Therefore, the

washout rate of the slower-growing organism is function of the dilution rate. In reality,

the situation is more complex since a population attached to a surface can survive in a

chemostat even when D exceeds its specific growth rate. Similarly, a slower-growing

organism, if present in an ecologically advantageous niche such as a biofilm, has been

observed to persist in competition with fast-growing species.

*2

Growth-limiting substrate conc., s

Figure 1-11. Hypothetical specific growth rates of organism 1 and 2. The text explains why organism 1 will wash out of the continuous culture.

50

1.5.5 Competition Related Research. Nathanson (1902) first isolated organisms

capable of autotrophic growth using reduced sulfur compounds as an energy source.

Since then, several different types of these organisms have been isolated. Starkey (1935)

showed that one group of these organisms, the genus Thiobacillus, includes not only the

obligately chemolithotrophic(auto)trophic organisms, but also facultative

chemolitho(auto)trophs, which possess the potential for heterotrophic growth in the

absence of an inorganic sulfur-containing substrate. The fact that (Starkey, 1935; Taylor

et al., 1971; Swaby and Vitolius, 1968) obligately chemolithotrophic thiobacilli are able

to maintain themselves in close proximity to almost any microbial cell in the natural

environment is remarkable. One of the characteristic properties of the thiobacilli is their

low maximum specific growth rate as compared to specialized heterotrophs. As growth

rate must be one of the decisive selective properties for competing species, the survival

of the thiobacilli would seem difficult to explain. However, it has been well established

that the outcome of competition under substrate-limiting conditions cannot always be

predicted on the basis of maximum specific growth rates but also depends on many other

parameters such as the actual specific growth rate (u), the substrate half-saturation

concentration (K,) (Veldkamp and Jannasch 1972; Harder and Veldkamp, 1971; Meers,

1971), the presence of predators (Jost et al., 1973), the availability of other limiting

substrates (Meers and Tempest, 1968; Megee et al., 1972), and probably a whole set of

still unrecognized parameters. In considering the potential of a facultatively

chemolithotrophic Thiobacillus to survive in competition with specialized heterotroph it

seems likely that metabolic versatility allows such organisms to survive in nature.

51

Rittenburg (1972) suggested that these chemolithotrophic thiobacQli-type organims

should be able to compete successfully with a "specialist" Thiobacillus under mixed

substrate conditions.

Gottschal (1981) found that when a chemostatic culture of Thiobacillus A2 was

grown under alternate limitation of acetate and thiosulfate (4h acetate/4h thiosulfate or

8h acetate/16h thiosulfate) at a dilution rate of 0.05 h'1, uninterrupted growth of the

culture was obtained. However, with 16h acetate/8h thiosulfate, Thiobacillus A2 required

several hours before it attained autotrophic growth at a rate of 0.05 h'1 following the

transition from acetate to thiosulfate supply. In a two-species, continuous flow, mixed

culture with the heterotrophic spirillum G7, Thiobacillus A2 outcompeted the

heterotroph when grown under alternating limitation of acetate (4h) and thiosulfate.

Under the same growth conditions, Thiobacillus A2 co-existed in approximately equal

numbers with T. neapolitanus, from which it was concluded that Thiobacillus A2 grew

only heterotrophically on acetate in this case. Analogous competition experiments with

three-species cultures grown under the same conditions, resulted in complete elimination

of Thiobacillus A2 and co-existence of T. neapolitanus and Spirillum G7 in almost equal

numbers. In an enrichment culture, again grown under the same regime, a facultatively

chemolithotrophic Spirillum-shaped Thiobacillus became dominant

Gottschal (1979), also investigated competition in a chemostat between the

versatile Thiobacillus A2 and the specialized 71 neapolitanus using thiosulfate as the sole

growth-limiting substrate. The specialized organism dominated at dilution rates > 0.025

h'1. Higher concentrations of acetate or glycollate in the thiosulfate medium caused

52

increased relative numbers of Thiobacillus A2 at steady-state(s) at dilution rates equal

to 0.07 h'1. Eventually, with 10-12 mM organic substrate per liter, complete dominance

of Thiobacillus A2 over T. neapolitanus occurred. In the same subset of work, mixed

cultures of Thiobacillus A2 and a specialized spirillum-shaped heterotroph competed for

acetate as sole growth-limiting substrate. The heterotroph dominated completely at

dilution rates of 0.07 and 0.15 h"1. In this case, increasing concentrations of thiosulfate

in the acetate medium, up to lOmM. eventually led to the elimination of the heterotroph.

These results were interpreted as evidence that Thiobacillus A2 was growing

mixotrophically. As the concentration of the second substrate was raised, the number

or ThiobacillusA2 cells increased, and as a result Thiobacillus A2 consumed an increasing

portion of the common substrate.

In mixed chemostat cultures containing all three organisms, Thiobacillus A2 was

maintained at all tested ratios of acetate and thiosulfate in the influent medium. The

heterotroph was excluded from the culture below a relatively low acetate-to-thiosulfate

ratio, whilst T. neapolitanus was completely eliminated at very high acetate/thiosulfate

values.

Smith et al. (1979), investigated the competition between T. neapolitanus and

Thiobacillus A2 in thiosulfate-limited chemostat cultures as a function of pH. He found

that, in pure cultures, pH <, 7.6, T. neapolitanus grew faster than Thiobacillus A2, but

in competition Thiobacillus A2 dominated at pH 7.35 and 7.6. At pH 7.1, T. neapolitanus

dominated, although a significant steady-state population of Thiobacillus A2 persisted

that apparently grew on organic nutrients excreted by T. neapolitanus. Co-existence of

53

both organisms occurred under all chemolithotrophic growth conditions tested with the

dominant organism comprising of 85 to 99% of the population, indicating that

competition was not the sole interaction between the species. However, at pH 7.1, the

inclusion of glucose in thiosulfate medium resulted in the rapid domination of the culture

by Thiobacillus A2, with the virtual elimination of T. neapolitanus. This study suggested

that the capacity for mixotrophy is a selective advantage to a facultative Thiobacillus in

competition with an obligately chemolithotrophic species.

Thingstad and Gottschal (1982) developed a mathematical model based on

Monod-type growth kinetics that described the growth of two bacterial species in mixed

chemostat cultures under dual-substrate limitation. In their experiments, competition

between a facultatively chemolithotrophic Thiobacillus and either a heterotroph or an

obligately chemolithotrophic Thiobacillus served as a model system. Furthermore, the

introduction of Monod-type growth kinetics in the model slowed an assessment of the

relative importance of the growth parameters for the outcome of the competition. In

addition, they showed how a mathematical description of the two-membered mixed

cultures can be used to predict the outcome of competition experiments among three

species competing for acetate and thiosulfate. In contrast to the experimental data,

Gottschal (1979) predicted that either two of the three species or the mixotroph alone

would survive in the culture. Within the framework of the model, he suggested two

possible explanations for the experimentally observed co-existence of three species: either

the veiy slow dynamics of the system did not allow the attainment of a true steady-state

within the time scale of the experiments or some parameters describing the mixed culture

54

were extremely sensitive to minor fluctuations in dilution rate. The results of this model

support the view that facultatively chemolithotrophic bacteria are able to survive under

appropriate mixed-substrate conditions in the presence of more "specialized" heterotrophs

and obligate chemolithotrophs, despite their relatively low specific growth rates.

Pol and Kuenen (1981), showed that during autotrophic growth, Thiobacillus A2

cells retained a considerable capacity to oxidize various organic sources.

Heterotrophically grown cultures, on the other hand, were completely devoid of the

capacity to fix C02 via the Calvin cycle and to generate energy from thiosulfate. During

transitions from organic media to inorganic thiosulfate-containing media in the

chemostat, a long lag-phase was observed before energy generation (thiosulfate

oxidation), C02 fixation and, consequently, measurable growth occurred. This lag-phase

was practically abolished if organic substrates were present at very low concentrations in

the thiosulfate mineral medium which could be used as an energy source. The same

result was obtained when the cells contained reserve material at the moment of the

transition. During transition from thiosulfate-limited growth to starvation, the Q02-max-

thiosulfate and the capacity to fix COz decreased very slowly, after an initial short (+4h)

increase in both enzyme systems. In contrast, these two metabolic functions were

inactivated relatively rapidly in the presence of an oxidizable organic carbon and energy

source. This process of inactivation was instantaneously stopped and reversed into rapid

enzyme synthesis upon replacement of the organic substrate by thiosulfate.

Goldberg et al. (1976) and Switzer (1977) showed that competition in microbial

cells is not only due to physiological flexibility and/or enzyme reactivity but also be

55

subject to selective inactivation under conditions in which the presence of the enzyme

is of no use or even particularly harmful. Hie authors studied both appearance and

disappearance of some enzyme activities during substrate transitions in a chemostat.

Attention was mainly focused on the enzyme system responsible for the oxidation of

thiosulfate and on the key enzyme of the autotrophic C02-fixation pathway in

ThiobacillusA2 (ribulose-1,5-bisphosphate carboxylase). They concluded that the ability

of a species to survive under various conditions may be related to its ability to conserve

energy by shutting down the biosynthesis of proteins that are not needed.

Jost et al. (1973) studied competition between Escherichia coli and Azotobacter

vinelandii for glucose (sole carbon source) by adding a suspension of E. coli to a steady-

state chemostat culture of A vinelandii. The time of addition was defined as the "zero

hour". The level of glucose at steady-state was about 0.005 mg/ml, the steady

concentration for the Azotobacter culture. In about 2 days, Azotobacter numbers were

reduced to less than 10% of their initial density, and the glucose level dropped to a point

where it was no longer measurable. Plate counts showed that there were about 10s

Azotobacter per ml. Similar data were obtained at other holding times ranging from 5.6

to 23 hours. In all cases E. coli displaced A vinelandii', this was true even when the E.

coli cells were cross-inoculated from stationary-phase batch cultures.

Meers and Tempest (1968) investigated several different magnesium-limited

mixed cultures, each containing two microbial species. Competition experiments were

initiated by exchanging small numbers of organisms between (previously) pure,

magnesium-limited chemostat cultures of Aerobacteraerogenes, Pseudomonasfluorescens,

56

Bacillus subtilis, Bacillus megaterium, Staphylococcus epidermis, and Torula utilis. The

fate of each species in these mixed cultures was determined following cross-inoculation.

Gram-negative organisms invariably outgrew the Gram-positive ones. The ability of B.

subtilis and B. megaterium to outgrow each other, or to outgrow the yeast (Torula utilis),

depended on the inoculum size. This dependence resulted from the presence of specific

extracellular products in the Bacillus cultures which stimulated their growth and uptake

of magnesium. The concentration of these extracellular growth-promoting substance(s)

in magnesium-limited cultures of B. subtilis varied with population density. Thus, when

the magnesium concentration in the influent medium was decreased from 0.9 to 0.15

ug/ml (thereby causing the steady-state population to be decreased to one-eightieth of

the initial value), B. subtilis could no longer maintain itself against proportionally lower

concentrations of T. utilis. But addition of extracellular fluids from dense magnesium-

limited B. subtilis cultures to magnesium-limited T. utilis cultures enabled small numbers

of the Bacillus to outgrow the yeast. From these two observations it was concluded that

magnesium assimilation by B. subtilis is more dependent on extracellular substance(s)

than is magnesium assimilation by either T. utilis or the Gram-negative organisms.

In another study, Meers (1970) investigated the influence of magnesium-limitation

on growth of Gram-positive organism-Bacillus subtilis-and the data compared with those

obtained with Aerobacter aerogenes grown under similar conditions. The magnesium

contents of both organisms varied with growth rate but were very similar at

corresponding growth rates. With magnesium-limited chemostat cultures of each

organism, uptake of magnesium was almost complete at specific growth rates less than

57

0.7max. Cellular magnesium was tightly bound, none being removed by res us pension of

the organisms at 20°C in 0.85% (W/V) NaCl. When magnesium-limited organisms were

suspended in environments containing excess magnesium, this ion was rapidly adsorbed;

the amount adsorbed varied with both the initial extracellular magnesium concentration

and the composition of the diluent. B. subtilis had a greater capacity for magnesium

adsorption than A. aerogenes but its affinity for this ion was less. The latter difference

correlated with the ability of A. aerogenes to rapidly outgrow B. subtilis in magnesium-

limited chemostat cultures containing both organisms.

1.6 Objectives.

1.6.1 General. This investigation was undertaken to determine if a steady-state

culture of Thiobacillus could be displaced by heterotrophic organisms and to define the

conditions under which such displacement occurred. A long range goal of this research

would be to determine if competitive displacement could be effected in corroding sewers.

In some respects Thiobacillus thiooxidans has a tenuous metabolism. The ATP output

from oxidizing sulfur compounds is low. The organism is also disadvantaged by having

to spend enormous amounts of energy to build all its cellular ATP by reducing C02. The

growth rates for thiobacilli are generally slow (12-16 hours), compared to other

organisms. However, T. thiooxidans is able to survive at very low pH and needs no

external source of organic carbon. Once established, it can drive out competitors by

acidifying the environment to a degree that is intolerable to most other species. Sewers

are carbon-rich environments. However, heterotrophic organisms residing on the crown,

58

may conceivably experience carbon limitation. That is, the availability of carbon to many

areas of the crown may be limited to volatile compounds that are not necessarily good

substrates (easily assimilated) for heterotrophic growth. Thus, if good growth substrates

were supplied, growth of heterotrophs would be encouraged. Under those circumstances,

heterotrophs might compete favorably with the thiobacilli for other limiting nutrients.

Such competitive relationships are the subject of this research.

1.6.2 Isolation Experiments. Sample collection and isolation of low-pH and mid-

pH heterotrophs capable of growth in a defined medium from a local sewer preceded all

other laboratory work. Sewer crown samples from the Roger Road Treatment Plant

collection system were subjected to growth and biochemical tests that led to the isolation

of the organisms. Isolation procedures generally followed those in common use.

1.63 Stoichiometric Experiments. Bench-scale, batch-culture experiments were

conducted to determine the stoichiometric relationships between nitrogen (ammonia) and

carbon (glucose) utilization by potential heterotrophic competitors. Nitrogen (ammonia)

to thiosulfate utilization ratios were also established for T. neapolitanus were tested.

Determination of specific growth rates and nutrient utilization rates for the sewer isolates

at various initial pHs was determined. This led to the determination of critical N/C ratio

defined as the nitrogen-to-glucose ratio at which transition from nitrogen-limitation to

glucose-limitation occurs. The critical N/C ratio is also the ratio of nitrogen assimilation

to carbon utilization during balanced growth of the subject heterotrophic competitors.

1.6.4 Two-Organism Competition Experiments. The acidophilic thiobacilli are

highly suited to occupy a rather constrained ecological niche that is characterized by low-

59

pH and low-organic conditions. Once established, they modify their environment to the

exclusion of less acid-tolerant organisms by dramatically lowering the ambient pH. On

the other hand, the thiobacilli are slow-growers and may not be at their best in organic-

rich environments.

To test this hypothesis, competition experiments involving thiobacilli and

heterotrophs were conducted under NH4+-limiting and carbon-limiting conditions. The

ammonia-to-glucose ratio in the feed-stream was varied in continuous-flow experiments

so as to bracket the critical N/C ratio for the growth of the heterotrophic competitors.

It was speculated that an adequate supply of carbon would enable the heterotrophic

organisms to out compete the thiobacilli for a limiting supply of nitrogen. Tests were

conducted at both low and mid-pH levels using T. thioaxidans and T. neapolitanus,

respectively, in order to simulate heavily corroded and lightly corroded or non-corroded

sewer conditions.

1.6.5 Effect of pH on Competition Experiments. The objective of this portion

of the work was to monitor the effect of bulk, liquid-phase pH on potential microbial

competitors that can interrupt acid production by thiobacilli. Experiments were

conducted at steady pH values of 2.00, 2.50, and 3.25, at influent feed N/C ratios above

and below the critical N/C ratio.

1.6.6 Effect of Different Heterotrophs in Competition Experiments. The object

of this work was to screen different heterotrophs that might be encountered in the actual

sewer environment as inhibitors of growth and acid production by the thiobacilli Tests

were conducted both under low and mid-pH conditions using sewer isolates including a

60

mixed seed the sewer and E. coli. Low-pH experiments were conducted at ZOO, 2.50,

and 3.25.

61

CHAPTER 2

MATERIALS AND METHODS

2.1 Microorganisms.

Two strains of thiobacilli were examined in terms of their ability to produce acid

in the presence of a heterotrophic competitor. Including the heterotrophic competitors,

the four species investigated were Thiobacillus thiooxidans, Thiobacillus neapolitanus,

Escherichia coli, and an unidentified strain of yeast.

Axenic cultures of thiobacilli and E. coli were obtained from the American Type

Culture Collection (ATCC, 1984). ATCC strain numbers 8085 (T. thiooxidans), 23638

(T. neapolitanus), and 15597 (E. coli) were purchased during the early phase of the

research. A species of yeast was isolated from the crown of an exceptionally corroded

local (Tucson) sewer.

2.2 General Description of Media Composition.

Four mineral salts media were used to grow the thiobacilli, E. coli, and the yeast.

The media used in all the experiments represent modifications of a Thiobacillus medium

described in the ATCC media handbook (medium #125; ATCC, 1984). Tables 2-1

through 2-8 list the chemical constituents and their concentrations in the low-pH and

62

mid-pH media. The basal salts media used in competition experiments were sulfate-free.

Sulfate was excluded to permit accurate measurements of sulfate ion as an indication of

the acid generation rate by the thiobacilli.

Media were autoclaved prior to bacterial inoculation for 15-20 minutes at 121°C.

Elemental sulfur (sublimed powder; EM Industries, Inc.) was sterilized via tyndallization

and added in excess (approximately 10 g/L) to the low-pH media as an energy source.

Iron and glucose were added from filter-sterilized stock solutions after heat-sterilization

of other ingredients was completed.

23 Isolation of Yeast from the Sewer Crown.

2.3.1 General. Crown samples at the Roger Road Treatment Plant in Tucson,

Arizona were collected in March, 1989. Within the premises of the plant area two

manholes were selected, one by the golf course and the other at the syphon outfall.

Samples withdrawn from the sewer crowns were inoculated into test tubes containing

growth media as in Table 2-3. pH was measured on the crowns as an indication of sewer

condition. Isolation of the bacterial strain (heterotrophs) from physical samples involved

liquid-culture enrichment followed by colony development on complex and defined solid

media.

23.2 Sampling Techniques. Safety procedures for collection of samples were

those followed by sewer-maintenance personnel. Sewer headspace was vented with fresh

air prior to entrance. Crown samples were taken using a pair of acid-resistant rubber

gloves. Samples were obtained by scraping material from the crown. Reasonable efforts

63

Table 2-1. General composition of defined chemical medium used for the growth of T. thiooxidans.

Grams per liter of distilled H20:

pH adjusted to 3.75

Table 2-2. General composition of defined chemical medium used for the growth of T. neapolitanus.

Gram per liter of distilled H20:

pH adjusted to 6.8

Notes for all media preparations:

(i) FeCl3. 6H20 was prepared at lOQx the final concentration and filter-sterilized prior to addition to previously cooled, autoclaved medium;

(ii) Desired initial pH was achieved by adding concentrated HQ.

Elemental sulfur KH2PO„ MgClj. 6H20 CaCl2

FeCl3. 6H20 NH<C1

10.0 3.0 0.5 0.25 0.01 0.5

Na^Oj. 5H20 Na2HPO< KHjP04

MgQj. 6H20 CaQ2

FeQ,. 6H20 NH4Q

10.0 1.2 1.8 0.1 0.03 0.01 0.5

64

Table 2-3. Composition of defined chemical medium used for the isolation of low and mid-pH organisms from sewer-crown samples.

Grams per liter of distilled H20:

KC1 0.05 KH2P04 0.1 MgSO< 0.5 (NH,)jS04 0.2 FeClj. 6H20 0.01 Glucose 5.0

pH adjusted to 3.75 and 6.8

Table 2-4. Composition of defined chemical media used in stoichiometric experiments involving E. coli growing in the mid-pH range.

Grams per liter of distilled H2Or

KC1 0.05 KH2PO, 0.1 MgSO< 0.5 (NH^O, (Various) 0.660 (5.0 x 10"3 M)

0.297 (2.25 x 10"3 M) 0.132 (1.0 x 10 ̂M) 0.101 (7.5 x Iff4 M) 0.067 (5.0 x lCT'M)

or 0.002 (1.0 x Iff5 M)

Glucose 0.405 (2.25 x 10"3 M)

pH adjusted to 6.80

65

Table 2-5. Composition of defined chemical media used in stoichiometric experiments involving yeast growing in the low and mid-pH range.

Grams per liter of distilled H20:

KC1 0.05 KH2PO4 0.1 MgSO, 0.5 (NH4)2SO, (Various) 0.925 (7.0 x 10"3 M)

0.463 (3.5 x 10-3 M) 0.106 (8.0 x 10"* M) 0.041 (3.0 x 10^ M) 0.007 (5.0 x 10 s M)

or 0.000 (control)

Glucose 0.405 (2.25xlO"3M)

pH adjusted to 3.25 and 6.8

Table 2-6. Composition of defined chemical media used in stoichiometric experiments involving T. neapolitanus growing in the mid-pH range.

Grams per liter of distilled H20:

Na2HPO< 1.2 KH2P04 1.8 MgCl2.6H20 0.1 CaCl2 0.03 FeClj. 6H20 0.01 Na^A. 5H20 1.24 (5.0 x 10"3 M) NH4C1 (Various) 0.187 (3.5 x 10"3 M)

0.043 (8.0 x 10"4 M) 0.016 (3.0 x 10"4 M) 0.003 (5.0 x 10-5 M) 0.0005 (1.0 x 10"5 M)

or 0.000 (control)

pH adjusted to 6.8

66

Table 2-7. Composition of deflned chemical media used in T. thiooxidans and yeast competition experiments.

Grams per liter of distilled H20:

Elemental sulfur 10.0 KH2PO< 3.0 MgCk 6H20 0.5 CaCl2 0.25 FeClj. 6H20 0.01 NH4C1 (Various) 0.535 (1.0 x 102 M)

0.187 (3.5 x 10-3 M) 0.028 (5.0 x 10^ M)

or 0.005 (1.0 x 10"4 M)

Glucose 0.405 (2.25 x 103 M)

pH adjusted to 2.0, 2.5, and 3.25 (experiment dependent)

Table 2-8. Composition of deflned chemical media used in T. neapolitanus and yeast competition experiments.

Grams per liter of distilled HzO:

NaH2P04 1.2 KH2PO4 1.8 MgCl2. 6H20 0.1 CaCl2 0.03 FeCl3. 6H20 0.01 Na&Oj. 5H20 1.24 (5.0 x ia3 M) NH4CI 0.535 (1.0xl02M)

0.187 (3.5 x 10* M) 0.028 (5.0 x ia4 M) 0.005 (i.o x ia< M)

Glucose 0.405 (2.25 x ia3 M)

pH adjusted to 6.80

67

Table 2-9. Reactor Details and Design Criteria for low and mid-pH experiments:

Detention time = 25 hours (approximately twice the doubling time of the slower-growing organism). Working liquid volume = 1000 ml Nutrient feed stream flow rate = 0.667 ml/min

68

to collect samples with sterile cotton swabs proved futile due to lack of surface moisture

at the sewer crown. Scraped samples were transferred into 20 ml test tubes containing

sterile media described in Table 2-3 and transported to the University of Arizona

laboratories. Crown samples were placed in an insulated container packed with ice

during transportation. Samples were processed within three hours of collection.

The crown surface pH was measured using pH paper indicator sticks (American

Scientific Products Inc.,). Due to the inadequacy of surface moisture on the crown

concrete, a small volume of distilled water was sprayed onto the crown surface prior to

the pH check. Samples from the tubes were transferred into 500 ml erlenmeyer flasks

containing media (200 ml volume) described in Table 2-3. Flasks were placed on a

shaker table and incubated at room temperature (23°C to 25°C).

2.3.3 Enrichment Steps. Inoculated media were incubated at room temperatures

for several days. Subsequently, samples were withdrawn aseptically and transferred from

each culture to fresh, identical media to eliminate growth that was driven by constituents

present in the original samples. These "first-transfer" enrichments were cultivated in the

same manner as the original cultures. Steps utilized for the isolation are summarized in

Figure 2-1.

2.3.4 Serial Dilutions. First-transfer cultures that indicated growth were used

to obtain pure strains of yeast via spread-plate technique. Dilution media were identical

to the original growth media (Table 2-3). When serial dilution immediately preceded

spread-plating, glucose was omitted. Serial dilution tubes were placed on a tube rotator

and/or shaker tables at room temperature (23°C to 25°C) for incubation and inspected

Collect sample for growth in initial selective

medium as in Table 2-3.

Transfer to fresh medium (first-transfer) for

additional enrichment.

Serial dilute and spread on plates with Bacto-agar and

medium. Monitor growth and monoclonal colonies.

Transfer from agar-plate to original medium and grow

axenic culture.

Restreak and grow bacteria on original medium and

cultivate isolates.

Figure 2-1. Steps used for the isolation and identification of sewer crown samples.

70

for growth. Monoclonal cultures of yeast were obtained via standard spread-plate

techniques from the highest dilution tube which showed growth.

Solid media were generally prepared by supplementing appropriate defined media

with 1.5% (WAO aSar P"or to sterilization. For low-pH media, pH was adjusted

approximately to 3.75 after sterilization at a temperature that was slightly higher than the

agar solidification point and before solidification. The process was structured in this

manner to avoid acid-catalyzed hydrolysis of agar at high temperatures (above 53°C).

Solidification was accomplished using Noble agar (DIFCO). When growth on solid

media was too dense, representative "colonies" were re-streaked onto fresh plates of

identical media in a manner designed to permit development of monoclonal colonies.

2.4 Stoichiometric Experiments.

2.4.1 General. Stoichiometric relationships among growth, nitrogen-to-glucose,

and nitrogen-to-thiosulfate utilization ratios were developed using two kinds of batch-

culture experiments. In the first, the ratio of nitrogen-to-carbon usage for the yeast

isolate and E. coli were determined. In the second, the nitrogen-to-thiosulfate usage ratio

for T. neapolitanus was determined. A N/S usage ratio for T. thioaxidans could not be

determined inasmuch as the acidophilic thiobacilli obtain carbon for growth by

assimilating carbon dioxide (obligate autotroph).

2.4.2 Determination of "Critical" N/C or N/S ratio. "Critical" nitrogen-to-

carbon ratios for growth of E. coli and yeast and a nitrogen-to-thiosulfate ratio for T.

neapolitanus were established in batch-culture experiments containing media described

71

in Table 2-4, 2-5 or 2-6. Following sterilization, 50-ml of media were dispensed into a

series of 125-ml erlenmeyer flasks containing ammonia-nitrogen at concentrations

provided in Table 2-4 and 2-5 or 2-6. The original nitrogen concentration was varied

over several orders of magnitude in the presence of a fixed concentration of glucose or

thiosulfate (experiment dependent). Flasks were inoculated with yeast, E. coli or T.

tieapolitanus from a dense overnight, 2.5% v/v, and shaken at room temperature (23°C

to 25°C) during the course of the experiment Growth was monitored into stationary

phase. Cell density was measured spectrophotometrically (culture absorbance at 600nM)

and directly using a PetrofF-Hauser counting chamber as described subsequently. Direct

counts were necessary when experiments involving thiobacilli formed suspensions of

secondary sulfide/sulfite/sulfate compounds within the medium.

In theory, the terminal optical density or cell number at which stationary-phase

growth was achieved, should be a function of the initial ammonia concentration until

sufficient ammonia is added to permit the culture to exhaust its glucose or thiosulfate

supply (nitrogen-limitation). At higher nitrogen concentrations, growth limits should not

depend on the initial nitrogen concentration. The nitrogen-to-glucose ratio at which

transition occurs from nitrogen-limited to glucose-limited would be the critical N/C ratio.

Similarly, the point at which nitrogen-limitation yields to thiosulfate-limited growth occurs

is the critical N/S ratio.

72

2.5 Experimental Procedures.

2.5.1 The General Procedure. The actual competition experiments were

conducted in 2-liter fermentors (Virtis, 1-liter working volume) operated in a continuous-

flow mode (CSTR). The theoretical detention times for these experiments were

established at 25 hours, or about twice the generation time for thiobacilli (slower-growing

species). The reactor design and operational details are summarized in Table 2-9.

Oxygen was supplied continuously as filter-sterilized compressed air at 1.5 ml/min, and

the reactor contents were continuously agitated at approximately 200 rpm. Temperature

was maintained at 23°C-25°C, and culture pH was held constant via the periodic additions

of 2.0 M or 3.0 M NaOH (experiment dependent). The selection of pH was dependent

on the experiment and designed to permit growth of both the Thiobacillus species and

E. coli or yeast at rates that were determined via feed rates of ammonia and glucose or

thiosulfate.

The conduct of competition experiments was as follows. T. thioaxidans or T.

neapolitanus was grown in a sulfate-free medium described in Table 2-7 or 2-8

respectively, under continuous-flow conditions until steady-state conditions were

achieved. Steady-state was indicated by stabilization of cell numbers/ml, effluent sulfate

concentration, rate of base addition for the maintenance of pH (experiment dependent),

and effluent ammonia concentration. When steady-state conditions were established,

heterotrophic competitors (E. coli or yeast; experiment dependent) were spiked into the

reactor. The cell concentration of the competitor after the spike was approximately an

order of magnitude lower than the steady-state concentration of thiobacilli. Glucose

73

addition to the reactor was initiated at that time. In separate experiments, the influent

ratio of glucose to ammonia nitrogen was an independent variable. Ratios were selected

both above and below the critical ratio determined in batch cultures (Section 2.4.2).

Following the shift in the feed-stream concentration of glucose and the

inoculation of the reactor with heterotrophs, the post-glucose-addition reactor

performance was monitored via the semi-continuous measurement of cell numbers/ml

(both heterotrophs and thiobacilli) and the concentrations of sulfate; ammonia; and

glucose (dissolved organic carbon) in the reactor effluent. Throughout the experiment

(before and after inoculation with the heterotrophs), a running account of the base (2.0

M or 3.0 M NaOH) added to maintain a constant culture pH was monitored. NaOH

added for the maintenance of pH was assumed to be a measure of biogenic acid

production and consequently, Thiobacillus activity. Because a sulfate-free medium was

fed to the reactors, effluent sulfate was also a measure of Thiobacillus activity. It proved

easy to differentiate between yeast and the prokaryotic populations.

Immunofluorescence techniques facilitated the differentiation between E. coli and T.

neapolitanus. The analytical procedures employed during the competition experiments

are described subsequently.

2.5.2 Variations in the General Procedures.

2.5.2.1 Variations in influent ammonia-glucose ratio. A variation in the

general experimental procedure was required to evaluate the effects of various nitrogen-

to-glucose ratios on the activity, populations, and acid production by thiobacilli and the

performance of competitors. Experiments were performed at both low and mid-pH.

74

Hie ammonia-to-glucose ratio was varied in each of the experimental runs

involving both low and mid-pH thiobacilli. The feed-stream N/C molar ratio was varied

from 0.0445 to 4.445 (NH/-N/glucose-C). This range bracketed the critical N/C ratio.

When nitrogen was provided above the critical N/C ratio, carbon-limited growth was

anticipated and effects on Thiobacillus activity and growth should have been minimal.

Below the critical N/C ratio, nitrogen was to be growth-limiting. A detailed summary of

growth media and N/C feed ratio is provided in Tables 2-7 and 2-8. A summary of

experiment run, parameters measured, and independent variables is illustrated in Table

2-10.

2.5.2.2 Variation in pH. These experiments were designed to explore the

effects of pH on the performance of microbial competitors. The ability of the acidophilic

thiobacilli to tolerate extremely low-pH conditions (>. 0.5 pH units) is well known. In

these experiments, the influent pH was adjusted to 2.00, 2.50 or 3.25 per Table 2-7.

Culture pH was maintained at the influent level via the periodic addition of 3.0 M

NaOH. Influent nutrient concentrations (glucose and ammonia) were manipulated above

and below the critical N/C ratio at pH 2.00 and 2.50.

2.5.2.3 Variations in Competitor Identity. Experiments were designed

to explore the performance of different microbial pairs (Thiobacillus/heteTotioph) in

competition experiments. Several heterotrophs were used in both the low (T.

thiooxidans) and mid-pH (71 neapolitanus) experiments. A summary of species pairs

investigated at various pHs and N/C ratios is presented in Table 2-10.

TABLE 2-101 A Summaty at Experiment Run, Fatameten Meatuted, end Independent Variables med to Low end Mld-pH OompetiUoo Experiments.

J Experiment

"

Species Fair

Growth Medium

Culture pH

Feed Stream

N/C Ratio

0.045 0.223 1556 4.445

Parameters Measured

Cell Counting

Technique

Independent Variable

Total | No. of 1

Experiments (

'

T. thioaddans vs.

Yeast

Tabfci-7 345

Feed Stream

N/C Ratio

0.045 0.223 1556 4.445

NH/ Carbon (DOC)

SO,1

Rate or 3.0M NaOH Added Cell Numbers/tal

(T. thioaddans and Yeast)

Direct Count

(PetrofT-Hauser

Counting Chamber)

N/C Ratio Four |

2. T. thioaddans vs.

Yeast

Table 2-7 ZOO 1556 4.445

NH/ • Carboa (DOC)

Rate of 3.0KJ NaOH Added Cell Numbers/tail

(T. thioaddans and Yeast)

Direct Count

(Petroff-Hauser

Counting Chamber)

Culture PH and

N/C Ratio

Six | 2. T. thioaddans vs.

Yeast

Table 2-7

250 1556 4.445 -

NH/ • Carboa (DOC)

Rate of 3.0KJ NaOH Added Cell Numbers/tail

(T. thioaddans and Yeast)

Direct Count

(Petroff-Hauser

Counting Chamber)

Culture PH and

N/C Ratio

2. T. thioaddans vs.

Yeast

Table 2-7

325 1556 4.445

NH/ • Carboa (DOC)

Rate of 3.0KJ NaOH Added Cell Numbers/tail

(T. thioaddans and Yeast)

Direct Count

(Petroff-Hauser

Counting Chamber)

Culture PH and

N/C Ratio

3. T. thioaddans vs.

Mixed Seed

Tbble 2-7 2.00 1556 NH/ Carbon (DOC)

SO Rate of 3.0H NaOH Added

Cell Numbers/ml (T. thioaddans Only)

Immuno­fluorescence

Culture PH

TSro 3. T. thioaddans vs.

Mixed Seed

Tbble 2-7

325 1556

NH/ Carbon (DOC)

SO Rate of 3.0H NaOH Added

Cell Numbers/ml (T. thioaddans Only)

Immuno­fluorescence

Culture PH

TSro

4. T. neapolitamts vs. "

Yeast

TbNe 2-8 &80 0.045 0.223 1556 4.445

NH,* Carbon (DOC)

SO/1

Rate of 20M NaOH Added Cell Numbers/ml

(T. neapoEtams and Yeast)

Direct Count

(Petroff-Hauser

Counting Chamber)

N/C Ratio Four

5. Tjieapolitamis vs.

£ coti

Table 2-8 6.80 1556 4.445

NH/ Carbon (DOC)

SO/* Rate of 2.0M NaOH Added

CeQ Numbers/ml (T. neapolitamts and E eoff)

Spread Plating

N/C Ratio Two

76

2.6 Analytical and Chemical Techniques.

2.6.1 Microbial Growth Measurements. Growth of bacterial cultures was

measured via four different methods throughout this research. In the first method,

absorbance of culture density in the ultra-violet (UV) light range was measured at a

wavelength of 600nm. Absorbance was measured using a Shimadzu UV-160A Visible

Recording Spectrophotometer. This analyzer was micro-computer controlled and utilizes

a double-beam photometer. A reduced volume quartz cuvette with a one-cm path length

was used for all absorbance measurements. The instrument was corrected for the

solution baseline absorbance.

The second method employed a Petroff-Hauser Direct Counting Chamber (light

microscopy) to determine cell densities. This counting chamber is a framed microscope

slide divided into three areas; the middle section is etched with a grid pattern surrounded

by raised ledges. The outer sections act as recess channels for draining excess liquid.

The etched grid section consists of squares of that are 0.05 mm in dimensions. A cover

slip placed on two raised ledges of the slide provides a uniform space, 0.02 mm thick,

between the top of the chamber glass to the cover slip. A drop of sample was placed on

the slide and cover slip set on the raised ledges. Excess liquid was displaced from the

grid section of the slide to recessed-channels adjacent to the divided grid. The number

of cells per square was determined by averaging cell counts from five or more squares.

This quantity was multiplied by 2 x 107 square volumes/ml to determine the number of

cells per ml.

The third method consisted of an eight-tube serial dilution and spread-plating

technique. This facilitated the determination of T. neapolitanus and E. coli cell numbers

in competition experiments that involved these organisms. One-ml samples of the

original mixed culture were transferred to the first tube and decimally diluted to

extinction. 100 ul samples from tubes corresponding to the 10"1 through 10"® dilutions

were spread-plated on nutrient agar (Difco) and thiosulfate solid media plates. The

nutrient agar and thiosulfate plates supported only the growth of E. coli and T.

neapolitanus, respectively. The difference in total cell counts obtained from the direct

counts and spread-plating was quantitatively equivalent to the cell numbers of E. coli or

T. neapolitanus. The fourth counting technique is described in section 2.7.

2.6.2 pH Measurement. Two methods of measurement were used throughout

this research: pH indicator strips (American Scientific Products Inc.) or a pH

meter/probe (Orion Research Model 811 millivolt/pH microprocessor or Beckman

Instruments Model-32 equipped with a Fisher Scientific pencil-thin Ag/AgCl electrode).

The pH sticks were used to measure culture pH when concern of contamination and/or

lack of liquid volume precluded other methods (e.g. crown pH during isolation).

Otherwise, the probe was used exclusively. The pH strips were considered accurate

within +0.2 pH units. Accuracy of the pH meter and probe were within +0.01 pH units.

Calibration of the meter/probe was accomplished with 4.00 and 7.00 pH buffers

(American Scientific Products Inc.).

Strong base (3.0 M NaOH) was added periodically for the maintenance of culture

pH in competition experiments. Titration curves were prepared to facilitate accurate

78

addition and accounting of base requirements. Such curves were constructed by plotting

the amount of base added to media versus the relative change in pH. The amount of

base necessary to return to the culture pH to target level was then determined from

titration curves adjusting for relative culture volume.

2.63 Total Sulfate Measurements. Sulfate ion was measured as a surrogate

parameter for the determination of acid/sulfate generation rate and metabolic activity of

thiobacilli. In this protocol, 3-ml samples were centrifuged (Eppendorf Centrifuge-5415

at 12000 rpm or 11750 RCF: reciprocal centrifugal force) for 10 minutes to remove

suspended elemental sulfur. To the centrate, 1-ml of conditioning reagent (consisting

of 50 ml glycerol; 30 ml HC1; 300 ml distilled water; 100 ml 95% ethyl or isopropyl

alcohol; and 75 g NaCl) was added and mixed in a stirring apparatus ^Standard Methods

for the Examination of Water and Wastewater. 1976). Following this, excess barium

chloride (BaCl2 crystals) was added and stirred for 60 seconds at constant speed.

Turbidity of the resulting barium sulfate (BaSO<) crystal suspension was measured at

420nm. Absorbance measurements were performed on a scanning Shimadzu 160A

UV/VIS spectrophotometer using a reduced volume one- centimeter pathlength quartz

cuvette. The sulfate ion concentration was determined by comparison of the reading

with a standard curve.

2.6.4 Ammonium Ion Measurements. In all of the experiments conducted,

ammonia measurements were carried out as an indication of steady-state conditions and

nitrogen-limitation. Samples were acidified and frozen until analyzed. Ammonium ion

79

measurements were performed using ammonium-gas/ion-specific probe (Fisher Scientific

Inc.) and an Orion Research Model 811 millivoIt/pH microprocessor.

The sample pH was increased to >. 11.0 by diluting with ionic strength adjuster

(ISA, Orion Research Products Inc.) to convert all ammonium ions to dissolved NH3.

The ionic strength adjuster was added until the sample remained blue in color

(colorimetric indication of required pH) and the measurement was completed. The

ammonium ion concentration was determined by conversion of the millivolt potential

reading with a standard curve.

2.6.5 Dissolved Organic Carbon Analyses. Total dissolved organic carbon was

considered a surrogate parameter for glucose concentration. Samples were periodically

removed and frozen until analyzed. DOC was measured using a Dohrmann carbon

analyzer, Model DC-80. This system measures the C02 produced via the UV-catalyzed

(chemical) mineralization of organic carbon. Samples were filtered through a 0.22 um

polycarbonate membrane filter (Poretics Corporation) and diluted to lower DOC

concentration into the most sensitive range of the analyzer. The samples were acidified

to pH 2.5 or below by adding 3 drops of concentrated phosphoric acid (American

Scientific Products Inc.,). COj was then stripped from solution using a stream of

nitrogen gas. The organic carbon present in the sample was oxidized to C02 and sparged

to a non-dispersive infrared detector. Calibration was accomplished using an

acidified/sparged potassium phthalate solution (Fisher Scientific Inc.,) preparation of

known concentration.

80

2.7 Immunofluorescence: Fluorochrome Antibody Conjugates.

2.7.1 General. This technique enables the determination of cell densities of a

specific organism present in a mixed culture. The antibody procedure was employed to

determine the populations of T. thioaxidans when a mixed seed from the sewer

environment was used to provide heterotrophic competitors.

Fluorochrome-antibody conjugates are prepared using affinity-isolated antibodies

or IgG fractions of antisera from human or animal immunoglobulins and are specific to

the organism against which they are prepared. Antibodies are conjugated to fluorescein

isothiocynate (FITC) and are tested for reactivity and performance by single radial

immunodiffusion (Becker, 1969), Immunoelectrophoresis, and immunofluorescent assays.

2.7.2 Fluorescent Antibody Procedure. Bacterial samples tested via the

fluorescent antibody procedure (FA) were vigorously shaken to dislodge T. thioaxidans

that colonized the sulfur particles. After sulfur was allowed to settle, floating particles

of sulfur were carefully removed with a pasteur pipet The remaining liquid was removed

without disturbing the sulfur at the bottom and centrifuged (Eppendorf centrifuge 5415)

at 12000 rpm or 11750 RCF for 10 minutes. The supernatant was discarded and the

pellet resuspended in phosphate buffered saline solution. Washing and centrifugation

were repeated at least two times. The composition of the phosphate buffered saline

(PBS) was, in grams per liter: NaCl, 8.0; KC1, 0.2; Na2HPO«, 1.15; KH2PO<, 0.2.

20-ul of the washed sample was heat fixed on a multi-well slide, and 20-ul of

antibody #1 (Sigma Chemical Company) was spotted on the slide containing the sample,

humidified (to prevent drying), and incubated for 20 minutes. Following gentle washing

81

with PBS and air drying, 20 ul of antibody #2 (Sigma Chemical Company) was applied

and incubated in the dark for 20 minutes. Finally, the slide was washed again with a

stream of PBS and air dried.

The cover slip was mounted with the FA mounting fluid (Olympus Inc.) by

applying a small amount (10 ul). Fluorescence microscopy was conducted using an

Olympus BH-2 microscope equipped with an oil immersion objective giving a

magnification of lOOOx. Thiobacillus thiooxidans will fluoresce green following antibody

attachment. Enumeration was done by counting the number of bright green bacteria in

at least 35 fields. The total number of cells per milliliter of sample was calculated by

multiplying the mean number of cells per field by 8.5 x 10\

82

CHAPTER 3

RESULTS

3.1 Isolation of Heterotrophic Competitors.

3.1.1 Field Observations: Crown pH and Corrosion. Crown pH and the degree

of corrosion were noted in sewers at the Roger Road Treatment Plant premises during

the March 1990 sampling trip. The crown surfaces with lower pH exhibited a much

greater degree of concrete degradation. The crown of the manhole at the syphon outfall

was highly corroded; pH was 0.8 to 1.0. Large concrete aggregate and reinforcing steel

of the RCP were visible. On the other hand, the manhole south of the golf course was

moderately corroded, pH was 4.8 to 5.0, and adjacent pipe sections were lightly

deteriorated. Interestingly, the cast iron cover of the syphon outfall manhole was

completely corroded to about 1/10 th of its original thickness due to the corrosive action

of sulfuric acid.

3.1.2 Growth of Sewer Crown Samples in the Laboratory. Growth of

heterotrophic organisms (sewer crown samples from the Roger Road Treatment Plant)

in the initial enrichment media (Table 2-3) and the first-transfer media (Table 2-3) was

observed in the first few days following inoculation. The pH of enrichment media was

stable, suggesting that thiobacilli were not active in those cultures.

83

Samples from the first-transfer cultures were diluted and cultivated on solid media

(solidified with 1.5% Bacto agar). Both the low- and mid-pH first-transfer plates

supported growth of a variety of organisms. Colonies were visually analyzed, noting

color, and extent of growth of over time. When faster-growing bacterial colonies were

transferred from plates back to the original liquid (selective) media, growth of

heterotrophs increased reasonably. The colonies selected for such transfers appeared to

be axenic. Transfer steps used for isolation of original samples were duplicated (Figure

2-1).

Colony size, direct visual appearance, and morphology of cells led to the

preliminaiy classification of the isolate (competitor) as yeast. Colonies that showed

identical characteristics believed to be yeast, were selected for subsequent isolation and

experimentation. Light microscopy, colony size, color, appearance, and morphology of

pure cultures that grew at both low and mid-pH were found to be identical in all

respects. The isolated organism was capable of balanced growth at reasonable rates (1

to 2 hours) in defined media between the pH range of 3 to 8 units (Figure 3-5).

Semitransparent, pale yellow colonies were observed on solid media within about 12

hours after inoculation and turned orange-pink after about 3 to 5 days. The orange-pink

pigment appeard in liquid media after the same period of time.

Originally, I was looking for an acidophilic and mid-pH heterotroph that could

grow in a defined media. However, isolation experiments yielded a heterotrophic yeast

capable of growing at both low- and mid-pH ranges in a defined meidia. Because

84

prototrophs could not be isolated from the crown environment, yeast was the only

heterotrophic competitor compatible in my experiments.

3.2 Critical N/C and N/S Ratios for Nutrient Utilization in Batch Cultures.

3.2.1 General. Three organisms were utilized in batch stoichiometric studies-the

yeast isolate collected from the sewer crown, E. coli (ATCC #15597), and T.

neapolitanus (ATCC #23638). Comparitive rates of utilization were established for (i)

ammonia and glucose (yeast or E. coli) and (ii) ammonia and thiosulfate (T.

neapolitanus). This kind of study could not be performed on T. thiooxidans since the

disappearance of elemental sulfur, supplied in excess in all cases, could not be quantified,

and carbon was assimilated as C02 in an open system.

The objective of these experiments was to establish generation times for

exponential growth and critical ammonia-glucose or ammonia-thiosulfate concentration

ratios. This led to the selection of hydraulic detention time for the continuous-flow

reactors and relative (critical) influent nutrient concentrations for the competition

experiments. During the determination of these critical ratios, the batch cultures were

inoculated with microorganisms in a log-phase state of growth.

3.2.2 Ammonia and Glucose Utilization Rates. Batch growth of the yeast isolate

and E. coli was measured spectrophotometrically into stationaiy-phase. Figures 3-1 and

3-3 represent growth as function of initial nitrogen concentration for E. coli and yeast,

respectively. The stoichiometry of nitrogen-to-glucose utilization was similar for yeast

85

at both low (2 to 4, data not included) and mid-pH (6 to 8, data not included).

Nitrogen/glucose uptake experiments for E. coli were conducted only at mid-pH.

At initial NH3 concentrations below ImM. cell numbers at stationary-phase were

proportional to the amount of nitrogen initially present. A doubling of the initial

nitrogen concentration resulted in a doubling of cell numbers at low concentrations of

ammonia. At higher concentrations, this proportionality detriorated. A five-fold increase

in initial nitrogen from 1 x 10"3 M to 5 x 10'3 M produced only 50% more cells. Below

ImM. cell populations were limited by the available nitrogen. As nitrogen levels were

increased above ImM. some other factor (presumably glucose) limited growth. The

transition point from nitrogen to carbon limitation is not abrupt but appears to occur.

This transition from nitrogen to glucose limitation occurred at about 8.0 x 10"4 M of

nitrogen (ammonia form) for yeast and about 2 x 10"3 M for E. coli.

The nitrogen-to-glucose ratio at which transition occurred, is the "critical" N/C

ratio (moles N/moles C). As illustrated in Figure 3-2 and 3-4, the critical N/C ratio for

both E.coli and yeast was at about 1:2 (moles N/moles C). Modest growth was seen in

the control (no nitrogen). This may be attributable to the presence of background

nitrogen in the inoculum and/or chemical impurities in the media constituents. Modest

growth in controls support the fact that cells store some extra nitrogen (ammonia) in

excess of their metabolic needs.

3.23 Ammonia and Thiosulfate Utilization Rates. Growth of T. neapolitanus

as a function of initial ammonia concentrations is summarized in Figure 3-5. From these

growth curves, the relationship between initial nitrogen concentration and cell numbers

86

1.00

I E S* 0.10

c o o (O •*-> CO <D a c ca n. o (0 A <

i i 10 12

i 14 16

Time (Hrs.)

NH4 = 5.0 E-03 -0- NH4 = 7.5 E-04

—I— NH4 = 2.25 E-03 NH4 = 1.0 E-03 -¥r- NH4 = 5.0 E-04 NH4 = 1.0 E-05

Figure 3-1. Growth of E. coli on glucose in pure, batch cultures. Cell densities at stationary phase was a function of the initial nitrogen concentration when total ammonia added was < 2.0 x 10'3 M. At higher nitrogen concentrations, growth proved to be glucose-limited.

87

0.35

Estimated Point of Nitrogen Limitation = 3.0 E-03M

Initial Glucose Concentration = 1 ?5 E-03M ^ 0.3-

£ O CD a £ a t a e o to " 0.15-n • o e a JO 0.1-

0.05-

Estlmated N/C Critical Ratio for E.coll at pH 6X0 «» 1

1E-03 3E-03 Initial Ammonia Concentration (M)

2E-03 Initial Ammonia Concentration

4E-03 5E-03

Figure 3-2. Relationship between culture optical density at stationary phase and initial nitrogen concentrations in pure, batch culture of E. coli.

88

i— i E o

0.1-

o o CO <

<0 0) o c CO £2 u. 0.01-o (0 A <

0.001 15 20 0 10 25 5

Time after Inoculation (Hrs.)

*- 7.0 E-03 M —I— 3.5 E-03 M —3K— 8.0 E-04 M

a- 3.0 E-04 M -K- 5.0 E-05 M -±- 0 M

Figure 3-3. Growth of acidophilic yeast on glucose in pure, batch cultures. Cell density at stationary phase was a function of the initial nitrogen concentration when total ammonia added was < 5.0 x 10"4 M. At higher nitrogen concentrations, growth proved to be glucose-limited.

0.7

E o ® as m JE a. £« a c 0 1 55 S u a •g 02-o a •D <

0.5-

0.4-

0.3-

0.1

Region of / Nitrogen Limitation/ Region of -——" Nitrogen Limitation/

Glucose UmttgltofP

-P

Initial Glucose Concentration & 2.25 E-03M

11 N/C Critical Ratio for Yeast Isolate = 1:2 (Moles N/Moles C)

1 For Both Low-pH (3.75) and Mld-pH (6.80)

2 3 4 5 Initial Ammonia Concentration (mM)

Figure 3-4. Relationship between culture optical density at stationary phase and initial nitrogen concentrations in pure, batch culture of acidophilic yeast.

90

1E+09q

i tm a 0) m a s. a & CO c 0 1 55 % 1 = ai O

1E+08:

1E+07:

1E+06

Ammonla-to-Thlosulfate Utilization Rates for

T. neapolitanus Stationary Phase of Growth

Log Phase 77/

i, S —I 1

Lag Phase

— —

50 100 150 Time (Hrs.)

200 250 300

NH4 = 0M NH4 = 1E-05M NH4 = 5E-05M

NH4 = 3.0E-04 M -X- NH4 = 8.0E-04 M ~±r NH4 = 3.5E-03M

Figure 3-5. Growth of T. neapolitanus on thiosulfate in pure, batch cultures. Cell number at stationary phase was a function of the initial nitrogen concentration when total ammonia added was < 3.0 x 10"4 M. At higher nitrogen concentrations, growth proved to be thiosulfate-limited.

91

in stationary- phase was determined (Figure 3-6). When initial ammmonia levels were

less than 3 x 10^ M, T. neapolitanus numbers at stationary-phase were directly

proportional to the amount of nitrogen provided. At higher nitrogen concentrations, cell

growth was thiosulfate-limited. At approximately 3 x 10"4 M of ammonia, the cells appear

to be thiosulfate-limited. This point was estimated as the transition from nitrogen to

thiosulfate limitation. The molar ratio of nitrogen-to-thiosulfate at which nitrogen

limitation occurred was about N/S = 1:1.5 (moles N/moles S). Figure 3-6 shows the

approximate transition in nitrogen-limitation to thiosulfate-Iimitation. Time-dependent

pH measurements are also summarized in Figure 3-7. The cultures with highest initial

nitrogen concentrations reached the lowest final pH. Results indicated that the cultures

were not pH-limited during the experiment (Figure 3-7).

3.2.4 Effect of Initial pH on Growth of Yeast Experiments were run that

depicted the growth of the yeast isolate in pure, batch cultures as a function of pH, when

all the nutrients required for bacterial growth were provided in excess (data not

included). In this experiment, the initial pH was varied from 2.00 to 4.25 at intervals of

0.25 pH units. At initial pHs of 2.00 and 2.25, growth was retarded in rate and extent.

In all these cultures, growth was generally apparent after a lag period of 24 hours.

Specific growth rates were essentially the same in cultures grown at initial pHs from 3.00

to 4.25 and reached an absorbance maximum of about 0.200 cm'1. Similar tests were

performed under mid-pH conditions. Specific growth rates were essentially identical

throughout the pH range of 3 to 8 units (data not shown).

92

1E+08-

S 8E+07 Si

CO c 0> E! 6E+07 5 D ol B L e? ca e o S

4E+07-

2E+07-

0E+00

Point of Nitrogen Limitation = 3.0 E-04 M

Initial Thlosulfate Concentration = 5.0E-04 M

Point of Ammonia Limitation

Estlr lated Critical N/S Ratio = 1:1.5

0E+00 1E44 2 E-04 3Ek>4 4E-04 5E-04 6E-04 Initial Ammonia Concentration (M)

7E-04 8 E-04

Figure 3-6. Relationship between culture cell density at stationary phase and initial nitrogen concentrations in pure, batch cultures of T. neapolitanus.

93

6.5-

3 3 u 5.5-

4.5 100 150

Time (Hrs.) 250 300 50 200

-m- NH4

-B- NH4

= 0 —

= 3.0E-04 -5 t— NH4

X- NH4

= 1.0E-05

= 8.0E-04 -»t€- NH4

NH4

= 5.0E-05

= 3.5E-03

Figure 3-7. Acid production as measured by pH decrease in medium containing pure cultures of T. neapolitanus using thiosulfate as substrate.

94

33 Titration Curves.

Figures 3-8 and 3-9 represent the titration of low- and mid-pH medium shown

in Tables 2-7 and 2-8. Titration curves provided an indication of media buffer capacity

over a wide range of pH values. The data were later used to estimate quantities of

biogenic acid produced by the thiobacilli (by back titrating the growth medium to its

original pH).

3.4 Competition Studies.

3.4.1 General. It was hypothesized that acid production by T. thiooxidans and T.

neapolitanus could be mitigated by encouraging conditions favorable to heterotrophic

competitors. Two-organism, continuous-flow experiments were conducted at low- and

mid-pH in sulfate-firee basal salts media. Variations in the general procedure included

alteration of (i) the influent ammonia/glucose ratio to include experiments above and

below the critical N/C ratio as determined above, (ii) the pH balance of the system, and

(iii) the identity of the heterotrophic competitor. A summary of experimental conditions

investigated is provided in Table 2-10. Elemental sulfur and thiosulfate were used as

substrates for T. thiooxidans (low-pH experiments) and T. neapolitanus (mid-pH

experiments), respectively. Yeast and E. coli were grown on glucose. Atmospheric C02

served as a carbon source for thiobacilli.

3.4.2 Low-pII Competition Experiments.

3.4.2.1 Variations in influent ammonia/glucose. The feed concentration

of nitrogen was varied in a series of experiments to study the effect of the N/C ratio on

95

Amount of 0.1M NaOH (ml)/100 (ml) Media

Figure 3-8. Titration curve used for the periodic base addition for the maintenance of pH at 2.50 or 3.25 in continuous-flow cultures of T. thiooxidans and yeast.

96

10 Amount of 0.5M NaOH (ml)/100 (ml) Media

Figure 3-9. Titration curve used for the periodic base addition for the maintenance of pH at 6.80 in continuous-flow culture of T. neapolitanus, yeast, and E. coli.

the outcome of the competition between yeast and T. thioaxidans. Results of low-pH

(T. thioaxidans and yeast cells) continuous-flow experiments run at feed concentrations

of ammonia and glucose that were above and below the critical N/C ratio are summarized

in Figures 3-10 through 3-18. Three experiments were run at feed ratios below the

critical N/C ratio, and only one experiment was conducted under what was projected to

be carbon-limited conditions. Experiments were run at feed N/C ratios of 0.045, 0.223,

1.556, and 4.445. All experiments were conducted at pH 3.25.

Results of experiments at (N/C)0 = 0.223 and 1.556 are summarized in Figures

3-10 through 3-13. The experiments conducted at (N/C)0 = 0.045 and 4.445 are

summarized in Figures 3-14 through 3-18. Comparison of the T. thioaxidans growth

curves prior to yeast and glucose addition (Figure 3-10 and 3-14) indicates that, at low

influent nitrogen levels, growth was infact nitrogen-limited. Cell numbers, effluent

sulfate concentrations, effluent ammonia concentrations, and rate of base addition

indicated that cultures of T. thioaxidans reached steady-state after about 240 hours.

Following the addition of yeast and glucose, steady numbers of yeast were a function of

the influent nitrogen concentration. Results are summarized in Table 3.1. It is apparent

that the growth was nitrogen-limited at the low N/C ratios. After this point, numbers of

thiobacilli decreased dramatically, from 2.8 x 107 to 7.21 x 10s, 8.5 x 107 to 4.57 x 10s, and

1.4 x 108 to 8 x 107, respectively, at all the three N/C critical ratios, but remained present

in low numbers.

Yeast and T. thioaxidans coexisted at high numbers at high N/C. At high N/C,

glucose depletion prevented the faster growing yeast from assimilating all the available

98

1E+09q

E

0) Si 1E+08; e 0 1 3 CL O a "3 a = 1E+07; o a a o

1E+06-

fit*

N/C Ratio = 1.556:1 f

/ ""—* fi

y N/C Ratio = 0.223:1 ^ k 11

Yeast Inlectlon at \B£NW- X

I T. thlooxldansvs. veast

pH = 3.25

... . Washout Steady State

50 100 150 200 Time (Hrs.)

250 300

1E+08

E * ai O e o

1E+07 1 a. o a ai O ts a a> >•

350 1E+06

NH4=5.0E-04M (T) -B- NH4=3.5E-03M (T) -Ar NH4=5.0E-04M (Y) NH4=3.5E-03M (Y)

Figure 3-10. Cell number versus time. Numbers of yeast and T. thiooxidans in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the steady concentrations of T. thiooxidans. When growth was nitrogen-limited, effect on the population of T. thiooxidans were much more pronounced.

99

700-

600- N/C Ratio = 1.556:1 -I ra E

500-

N/C Ratio = 0223:1 I 400-o e o 0 0) 1

300-

3 to € 200-a> 3

£ 100-

Wash out Yoast Injection at

T. thlooxldano/6. yeast

pH = 3.25 Steady State

100 250 300 50 150 200 350 Time (Hrs.)

-m- NH4=5.0E-04M -S- NH4=3.5E-03M

Figure 3-11. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. thiooxidans and yeast plus T, thiooxidans. Notice that the addition of yeast and glucose greatly diminished the production of sulfate and presumably the activity of T. thiooxidans.

100

0.01 q

0.001: N/C Ratio = 1.556:1

E e> u e o " 0.0001 w

N/C Ratio a 0.223:1

c 0 E E < e 01 3 1E-05:

Ysast Injection a t Washout - T. thlooxldans vs. yeast

pH = 3£5

1E-06 300 50 150 2c

Time (Hrs.) 250 100 350

-m- NH4<=5.0E-04M S- NH4»&5E-03M

Figure 3-12. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. thuxxddans and yeast plus T. thiooocidans. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of yeast and high N/C feed conditions did not.

101

TJ at •a •a < X O co Z

co "S S 3 O E <

2.5

2-

1.5-

1-

0.5-

T. lhlooxldans vs. yeast

pH = 3.25

Steady State

N/C Ratio = 1.556:1

Steady State

N/C Ratio = 0.223:1

Yeast Injection at

Steady State

jfll.ii.fl. •illVrl,

Wash out

•, •, M. 1 I I I I • I I T » I J I I » ' I I I I I I I I ' I 0 48 96 144 192 240 246.5 252.5 258.5 267 276.5 291

Time (Hrs.)

I 1 NH4=5.0E-04M Hi NH4=3.5E-03M

Figure 3-13. Periodic base addition for the maintenance of pH at 3.25 in continuous-flow cultures of T. thiooxidans and yeast. Notice that the rate of acid production was adversely affected by the addition of glucose and yeast at about 250 hours. In the low N/C culture (low-nitrogen feed or culture that was expected to be nitrogen-limited), acid production was almost completely interrupted by the shift in feed conditions.

102

1E+08 N/C Ratio = 4.445:1

* 1E+08:

N/C Ratio = 0.045:1 S- 1E+07: -1E+07 •§ I Q.

Wash out •5 1E+06:

Yeast ln]ectlon at

T. thlooxldana/8. yeast

pH = 3.25 Steady State

1E+05 1E+06 50 100 150

Time (Hrs.) 200 250 300 350

-m- NH4=1.0E-04M (T) —4— NH4=1.0E-04M (Y) -B- NH4=1.0E-02M (T) -jk- NH4=1.0E-02M (Y)

Figure 3-14. Cell number versus time. Numbers of yeast and T. thiooxidans in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the steady concentrations of T. thiooxidans. When growth was nitrogen-limited, effect on the population of T. thiooxidans were much more pronounced.

103

900-

T. thloo)ddan«V8. yeaat

pH = 3.25 800-

_f N/C Ratio = 4.445:1 ra 700-

2 600-

S 500-

400-

N/C Ratio = 0.045:1

200-

Wash out 100-

Steady State

50 100 150 200 250 300 350 Time (Hra.)

HB- NH4=1.0E-04M -S- NH4=1.0E-02M

Figure 3-15. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. thiooxidans and yeast plus T. thiooxidans. Notice that the addition of yeast and glucose greatly diminished the production of sulfate and presumably the activity of T. thiooxidans.

104

1E-04I

Yeast Injection a t

T. thiooxidans v«. yeast Steady State

Time (Hrs.)

-rn- NH4=1.0E-04M -e- NH4=1.0E-02M

Figure 3-16. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. thiooxidans and yeast plus T. thiooxidans. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of yeast and high N/C feed conditions did not.

105

"S 160- Yeast Injectloi

at Steady C 140-2 14(h State CO

S 120-

N/C Ratio = 4.445:1

Glucose Sufficient

N/C Ratio = 0.045:1 Glucose Deficient Wash o

h- • • • -P a n—a

T. thiooxidans vs. yea6t

°0 20 40 60 80 100 120 Time (Hrs.)

-m- NH4=1.0E-04M -H- NH4=1.0E-02M

Figure 3-17. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving yeast and T. thiooxidans. Results confirm that growth was glucose-limited in the high-N/C, but not the low-N/C experiments.

106

2.5-

1 21 •D <

1.5-O m 2 S o o o c 1-3 O E <

0.5-

Steady State

N/C Ratio = 4.445:1

Steady State N/C Ratio = 0.045:1

/ Yeast Injection at

Steady State

T. thlooxldans vs. yeast

pH = 3.25

Wash out

• i J J l l l m l . I I I I » I 1 l l l l l l l l l l l I I I I I I I I I I I I t I I

0 168 246 264 279 300 330 Time (Hrs.)

I 1 NH4=1.0E-04 M Hi NH4=1.0E-02 M

Figure 3-18. Periodic base addition for the maintenance of pH at 3.25 in continuous-flow cultures of T. thiooxidans and yeast. Notice that the rate of acid production was adversely affected by the addition of glucose and yeast at about 250 hours. In the low N/C culture (low-nitrogen feed or culture that was expected to be nitrogen-limited), acid production was almost completely interrupted by the shift in feed conditions.

107

nitrogen. Steady-state residual ammonia levels at an influent N/C = 4.445 are 10 to 400

times higher than those of other (low N/C critical ratio) experiments. Nonetheless, yeast

injection produced some degree of competition between yeast and T. thiooxidans.

Ammonia concentrations dropped 87% across the reactor. Thiobacilli grew more slowly,

causing their numbers to decrease 2-fold. This 2x reduction in numbers of T. thiooxidans

suggested that the glucose and/or yeast addition had a modest inhibitory effect on the

thiobacilli. These results were supported by measurements of effluent S04'2

concentration, which dropped almost three-fold in the high N/C feed ratio experiment

and virtually disappeared at all the three low N/C feed concentrations.

It is apparent that in the presence of excess glucose, the faster-growing yeast

outcompeted T. thiooxidans for a growth-limiting supply of ammonia. Nonetheless, the

thiobacilli did not completely wash out of the CSTR. A summary of levels and

percentage decreases in parameters measured before and after the addition of

glucose/competitor is presented in Table 3-1.

3.4.2.2 Variations in Culture pH. Additional competition experiments

involving T. thiooxidans and the yeast isolate were conducted at pH 2.00 and 2.50. Only

two experiments corresponding to different influent ammonium concentrations (above

and below the critical N/C ratio; 1.556 and 4.445) were run at each pH.

Results for these experiments are summarized in Figures 3-19 through 3-23. At

pH 2.50, steady-state was achieved after about 275 hours and the growth of T.

thiooxidans was nitrogen-limited at the low N/C feed ratio. After the addition of yeast

and glucose, yeast grew to moderately high numbers, and T. thiooxidans was again

108

1E+09q

.E j| 1E+08; "5 O e 0 1 §• 1E+07: a « O

o ca J3 o 1E+06:

1E+05

~

N/C Ratio = 4.445:1

/ •

N/C Ratio = 1.556:1

Yeast Injection at

T \

5K Waahout

« 1

" pH = 2.50 Steady State

i i i i i

5K Waahout

« 1 50 100 300 350

1E+08

400 1E+07

Time (Hrs.)

NH4=3JE-03 M(T) NH4=1 E-02 M(T) NH4-=3.5E-03 M(V) NH4=1E-02 M(Y)

Figure 3-19. Cell number versus time. Numbers of yeast and T. thiooxidans in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the steady concentrations of T. thiooxidans. When growth was nitrogen-limited, effect on the population of T. thiooxidans were much more pronounced.

109

900'

T. thiooxidans vs. yeast

pH = 2.50 800-

S> 700- N/C Ratio = 4.445:1

2 600-

R 500-

N/C Ratio = 1.556:1 400-

W 300-

200-

Yeast Injection at

100-

Steady State Wash out

50 200 Time (Hrs.)

250 100 150 400 300 350

NH4=3.5E-03 M -B- NH4=1.0E-02 M

Figure 3-20. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. thiooxidans and yeast plus T. thiooxidans. Notice that the addition of yeast and glucose greatly diminished the production of sulfate and presumably the activity of T. thiooxidans.

110

1E-02I

N/C Ratio = 4.445:1

c 1E-03:

1E-04:

N/C Ratio = 1.556:1

1E-05:

Yeast Injection at T. thiooxidans v». yeast pH = 2.50 Steady State Wash out

1E-06 100 250 300 150 200

Time (Hrs.) 350 400

NH4=1.0E-02 M NH4=3.5E-03 M

Figure 3-21. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. thiooxidans and yeast plus T. thiooxidans. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of yeast and high N/C feed conditions did not.

Ill

T. thlooxldnn»v». yeast

pH = 2J50

N/C Ratio = 1.556:1

Glucose Sufficient Yeast Injection at

Steady State

N/C Ratio = 4.445:1

Glucose Deficient

-m- NH4=3.5E-03 M -S- NH4=1.0E-02 M

Figure 3-22. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving yeast and T. thioaxidans. Results confirm that growth was glucose-limited in the high-N/C, but not the low-N/C experiments.

112

5-i

4.5-

4-

•S 3.5-U •a 1 3h O a Z Z5-2 o ri 2-

•s E 1.5 3 0 1 1

0.5-J

i I I i i I i j I i i I I I I I I V

Steady State T. thlooxldan* vt. yeart

N/C Ratio = 4.445:1 pH = 230

Steady Stale

N/C Ratio = 1356:1

Yoaat Injection at

/ Steady State

a Washout

nl 0 72 144 216 288 297 312 321 336 345 380 369 384 400

Time (Hrs.)

| NH4=3.5E-03 M • NH4=1.0E-02 M

Figure 3-23. Periodic base addition for the maintenance of pH at 2.50 in continuous-flow cultures of T. thiooxidans and yeast. Notice that the rate of acid production was adversely affected by the addition of glucose and yeast at about 275 hours. In the low N/C culture (low-nitrogen feed or culture that was expected to be nitrogen-limited), acid production was almost completely interrupted by the shift in feed conditions.

113

inhibited to some extent. Order-of-magnitude decreases in effluent sulfate concentration

was observed following yeast and/or glucose addition. In general results were about the

same as those obtained in experiments conducted at pH 3.25 at similar N/C influent

concentrations.

The results of the experiments conducted at pH 2.00 are summarized in Figures

3-24 through 3-28. Cell numbers and effluent sulfate concentrations indicate that the

yeast was completely inhibited at pH 2.00. Numbers of T. thiooxidans decreased for a

short period of time, after which cell numbers regained steady-state levels in about 60

hours following the injection of yeast. Growth curves (Figures 3-24) indicate that at low

nitrogen levels, growth was nitrogen-limited; at high nitrogen levels growth of yeast was

glucose-limited and/or failed to grow. These results were complemented by the

reductions in effluent ammonia concentration (Figures 3-26), base addition for the

maintenance of pH (Figures 3-25 and 3-28), and effluent DOC (Figures 3-27). Table 3-1

shows the percentage decrease of parameters measured at steady-state, before, and after

the addition of glucose and yeast.

3.4.2.3 Variations in Competitor Identity. Two sets of experiments were

conducted with a mixed-seed competitor, obtained directly from the sewer crown, at pH

2.00 and 3.25. The nitrogen-to-glucose feed ratio was selected to make the system

glucose sufficient (nitrogen deficient, N/C = 1.556:1) in all experiments. Results (Figures

3-29 through 3-33) are similar to those obtained using the yeast isolate.

At pH = 3.25, the steady numbers of T. thiooxidans prior to the introduction of

glucose were directly related to the influent ammonium concentration, on comparison

114

1E+09q

E 2? "5 2. 1E+08; e 0 1 a a. o a » O = 1E+07; u a xt o

1E+06

- -

T. thlooxidans vs. veast

' pH = 2.oo ra-e-ei

Iff r#*

. N/C Ratio = 4.445:/ /

v/j/ N/C Ratio = 1.556:1

Yeast Injection at -

i r-= • -

Steady State

1 1 1 1 I I 1 SO 100 150 200 250 300 Time (Hrs.)

;1E+08

E jjfc

:1E+07 a! Si c 0 1 3 a o a

1E+06 O

a £

350 400 450 500 1E+05

NH4 = 3.5E-3 M(T) NH4 = 1E-2 M(T) NH4 = 3.5E-3 M(Y) NH4 = 1E-2 M(Y)

Figure 3-24. Cell number versus time. Numbers of yeast and T. thioaxidans in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the concentrations of yeast. Notice the temporary effect on the population of T. thioaxidans. Increase in thiobacilli cell densities was seen after about 80 hours following the cross inoculation of yeast.

115

900'

T. thlooxldam va. yeast 800-

pH = 2.00

® 700-N/C Ratio = 4.445:1

600-

€ g 500-c o 0 « 1

400-

W 300-E o> S 200-Ul

N/C Ratio = 1.556:1

Yeast Injection at 100-

Steady State

50 100 150 250 Time (Hrs.)

300 350 400 450 200 500

NH4 = 3.5 E-03 M -B- NH4 = 1.0 E-02 M

Figure 3-25. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. thiooxidans and yeast plus T. thiooxxdans. Notice that the addition of yeast and glucose temporarily reduced the production of sulfate and presumably the activity of T. thiooxidans.

116

1E-01

N/C Ratio = 4.445:1 1E-02E

c o O m N/C RAtlo = 1.556:1

e i 1E-04: E < E Q) | 1E-05:

Wash out

T. thlooxldans vs. yeast

pH = 2.00 Steady State

1E-06 450 250

Time (Hrs.) 50 100 150 200 300 400 350 500

-m- NH4 = 3.5 E-03 M -B- NH4 = 1.0 E-02 M

Figure 3-26. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. thioaddans and yeast plus T. thiooxidans. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of yeast and high N/C feed conditions did not.

117

180-

160-

H/C Ratio = 1.556:1 Yeast Injection

at Steady State

Washout 100-N/C Ratio = 4.445:1

o O at 80-M o u - 60-O E 0) 3 40-

T. thlooxldans vs. yeast

pH = 2.00 20-

60 80 Time (Hrs.)

100 140 160 20 40 120

NH4 = 3.5 E-03 M S- NH4 = 1.0 E-02 M

Figure 3-27. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving yeast and T. thioaxidans. Results confirm that growth at high-N/C and low-N/C was nitrogen-limited, not glucose-limited.

118

6E+00-

T. thiooxidans vs. yeast Steady State

5E+00-

•a o> S 4E+00-< X o CO Z 3E+00-£ o CO O 2E+00-

3 O B < 1E+00-

pH = 2.00

OE+OO-ji

StMdySUt*

N/C Ratio »1.556:1

N/C Ratio = 4.445:1

0 72 144 216 288 312 336 360 384 408 432 Time (Hrs.)

I 1 NH4 = 3.5 E-03 M B| NH4 = 10 E~°2 M

Figure 3-28. Periodic base addition for the maintenance of pH at 2.00 in continuous-flow cultures of T. thiooxidans and yeast. Notice that the rate of acid production rate in both low N/C and high N/C cultures were temporarily affected by the addition of glucose and yeast at about 300 hours.

119

with previous results (Figures 3-29). According to the effluent ammonia and glucose

data, the cultures grew into nitrogen-limitation at the conclusion of both the experiments

(Figure 3-31). A significant amount of glucose residual remained at all times throughout

the experiment (Figure 3-32). Following the addition of the competitor and glucose,

effluent sulfate and acid generation rate were cut down by about 50% (Figure 3-30 and

3-33). In addition, T. thioaxidans cell density were cut by two orders of magnitude

(Figure 3-29).

At pH = 2.00, temporary reductions in T. thioaxidans cell population and effluent

sulfate concentrations were observed (Figures 3-29 and 3-30). Approximately 50 hours

after the introduction of the mixed culture competitor and glucose these parameters

returned to their pure-culture, steady-state concentrations. The rate of addition of 3.0

M NaOH also increased gradually from 1.5 to 2.5 ml following the increase in T.

thioaxidans cell concentration (Figure 3-33). However, effluent ammonia concentration

showed a decrease, but remained low following the injection of the mixed-seed (Figure

3-31). This suggested that heterotrophic organisms contained in the mixed-seed were

able to establish themselves at some steady, albeit low level. They did not compete

seriously with T. thioaxidans for the limiting nutrient, probably due to extreme pH. From

Figure 3-32, glucose in the effluent remained present throughout the experiment.

Results are summarized in Table 3-1.

3,43 Mid-pH Competition Experiments.

3.4.3.1 Variations in influent ammonia/glucose feed ratio. Thiobacillus

species such as T. neapolitanus, which grow optimally in the neutral pH range, are

120

1E+09q

T. thiooxidans vs. Mixed Seed

pH = 2.001> 3.25

£ 1E+08:

1E+07: N/C Ratio = 1.556:1

Wash out Injection of Mixed Seed

at Steady State

350 400 100 200 250 Time (Hrs.)

300 450 50

HH- pH = 2.00 -e- pH = 3.25

Figure 3-29. Cell number versus time. Number of T. thiooxidans in continuous-flow reactors during competition experiments. Addition of mixed seed lowered the steady concentration of T. thiooxidans at pH = 3.25, but did not at pH = 2.00.

121

600-

T. lhlooxldam vs. Mixed Seed

pH = 2.00 fc 3.25 500-

O)

400-

300- N/C Ratio = 1.556:1

200-

Injectlon of Mixed Wash out

Seed at Steady State

50 100 150 200 Time (Hrs.)

250 300 350 400 450

Figure 3-30. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. thioaxidans and a mixed seed plus T. thioaxidans. Notice that the addition of the contaminant and glucose temporarily reduced the production of sulfate and presumably the activity of T. thioaxidans at pH = 2.00.

122

c o « w £ 0> o e o O a e o E E < E 0) 3 £

1E-02q

1E-03:

1E-04:

1E-05:

1E-06-

I

N/C Ratio = 1.556:1

T. thiooxldansvs. Mixed Seed Injection of Mixed Seed

" pH = 2.00 & 3.25

1 1

at Steady State

1 1

Ammonia Deficient 1

1 1 50 100 150 200 250

Time (Hrs.) 300 350 400 450

pH = 2.00 -S- pH = 3.25

Figure 3-31. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. thioaxidans and yeast plus T. thiooxidans at pH 2.00 and 3.25. Results confirm that cultures were nitrogen-limited.

123

180-i Injection of Mixed Seed

t Steady State

N/C Ratio = 1.556:1 Wash out

Ammonia Deficient

-a

Ammonia Deficient

20-

T. thiooxidans vs. Mixed Seed

pH = 2.00 & 3.25

—i— 60

—i— 120 20 40 80 100 140

Time (Hrs.)

pH = 2.00 -B- pH = 3.25

Figure 3-32. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving the mixed seed and T. thiooxidans. Results confirm that growth at pH 2.00 and 3.25 was nitrogen-limited, not glucose-limited.

124

2.5-

T3 0> •o •o < X o a Z

o CO o E 3 O E <

T. thlooxtdanavi. Mixed Seed

pH = 2.00 It 325

N/C Ratio = 1.556:1

2-

1.5-

1-

0.5-

Steady State

n i 1 i ' i i i i 1 i 1 ^ f 1 " i 1 i i 1 i i 1 i 1 i 1 i 0 72 144 216 288 324 348 372 396 435

Time (Hrs.)

I 1 pH = 2.00 BH PH = 3J2S

Figure 3-33. Periodic base addition for the maintenance of pH at 2.00 and 3.25 in continuous-flow cultures of T. thioaxidans and a mixed seed from the sewer. At pH = 3.25 rate of acid production was adversely affected by the cross inoculation of mixed seed, but not at pH = 2.00.

125

thought to establish the sewer crown environment for succession to the acidophile T.

thioaxidans. Competition between T. neapolitanus and the yeast isolate was examined

at pH 6.80. Four experiments were run at feed N/C ratios of 0.045, 0.223, 1.556, and

4.445. Experimental results at feed concentrations less/greater than the critical N/C ratio

are summarized in Figures 3-34 through 3-43.

From Figures 3-34 and 3-39, it is apparent that the culture reached steady-state

at about 300 hours and was nitrogen-limited at the low influent nitrogen levels. Growth

of T. neapolitanus was restricted in the presence of glucose and a yeast population that

was capable of competing for nutrients other than glucose. Reductions in population

density of T. neapolitanus (from 4 x 108 to 2.5 x 107 cells/ml), sulfate concentrations (375

to 175 mg/1), and acid generation rate were observed in even the high-nitrogen

experiments (at an N/C = 4.445). Glucose was almost exhausted under these conditions,

limiting the extent of the growth of yeast (Figure 3-42). Sufficient ammonia remained

in solution to allow T. neapolitanus to persist in the reactor, even though the size of the

population was reduced 20-fold (Figure 3-41). At lower N/C ratios, the yeast had greater

success in lowering effluent nitrogen levels. The ammonia concentrations in the reactor

dropped to micromolar levels (Figure 3-36 and 3-41). In the high N/C experiment,

reactor concentrations were millimolar. The depression of nitrogen concentrations led

to a decrease in T. neapolitanus numbers by about 2 to 3 orders-of-magnitude (Figure

3-34 and 3-39). This trend is reflected by cell numbers, sulfate production, and acid

production. Nevertheless, bacterial attachment apparently played a significant role by

126

1E+09=

N/C Ratio = 1.556:1

1E+08

N/C Ratio = 0.223:1

Q- 1E+07

1E+06 Yeast Injection at

Steady State T. neapolltanusva. yeast

pH = 6.80

cl E+09

E *

1E+08 « O.

a 3 a. o a.

1E+07 o •& C0

Washout

150 200 Time (Hrs.)

250 300 350 400 1E+06

NH4=5 E-04M(T) NH4=5 E-04M(Y) NH4=3.5E-03(T) NH4=3.5E-03(Y)

Figure 3-34. Cell number versus time. Numbers of yeast and T. neapolitanus in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the steady concentrations of T. neapolitanus. When growth was nitrogen-limited, effect on the population of T. neapolitanus were much more pronounced.

127

O) & e o « w e V o c o 0

1 3 CO € 0) a

£

250-

200-

50

T. neaDolilanueva. veast

pH = 6.80 ja*0"4

/ N/CRAtlo = 1.556:1

CZi N/C Ratio = 0.223:1

Yeast Injection at

Steady State Wash out

50 100 150 200 Time (Hrs.)

250 300 350 400

NH4=5.0 E-04M NH4=3.5 E-03M

Figure 3-35. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. neapolitanus and yeast plus T. neapolitanus. Notice that the addition of yeast and glucose greatly diminished the production of sulfate and presumably the activity of T. neapolitanus.

128

1E-02q

N/C Ratio = 1.556:1 e 1E-03:

50 250 100 150 200 Time (Hrs.)

300 350 400

NH4=5.0 E-04M -B- NH4=3.5 E-03M

Figure 3-36. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. neapolitanus and yeast plus T. neapolitanus. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of yeast and high N/C feed conditions did not.

129

200

180-

Ol 160-1 ,E e o % E © o c o O 0) CO o o 3

5 E 0) = 40

80-

60-

20-

T. neaoolltanus vs. veast

pH = 6.80

Yeast N/C Ratio = 0223:1

Injection at •

Glucose Sufficient

Steady State a •

Wash out

-S- —Q Glucose Sufficient

-

N/C Artlo = 1.556:1

10 15 20 25 Time (Hrs.)

30 35 40 45

NH4=5 E-04M NH4=3.5 E-03M

Figure 3-37. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving yeast and T. neapolitanus. Results confirm that growth was glucose-limited in the high-N/C, but not the low-N/C experiments.

130

2.5-1

T. neapolltanusvs. yeast

pH = 6.80

2-

•o a> •a •a < 1* O a Z 5 0 01 1-•s E a o E "* 0.5-I

1 1 1 1 1 1

Steady State

N/C Ratio = 1.556:1

Steady State N/C RAtlo = 0^23:1

Yeast Injection at

Steady State

i i i ri i i i i i i < i "i r liii Wash out

xID l 0 72 120 168 216 264 312 336 342 348 354 364 375

Time (Hrs.)

I 1 NH4=5.0 E-04M HBB NH4=3.5 E-03M

Figure 3-38. Periodic base addition for the maintenance of pH at 6.80 in continuous-flow cultures of T. neapolitanus and yeast. Notice that the rate of acid production was adversely affected by the addition of glucose and yeast at about 325 hours. In the low N/C culture (low-nitrogen feed or culture that was expected to be nitrogen-limited), acid production was almost completely interrupted by the shift in feed conditions.

131

1E+09q

E * 1E+08:

1E+09

T. neapoHtanuo/B. yeast

pH = 6.80

N/C Ratio = 4.445:1 •e—Btw

N/C Ratio = 0.045:1

S- 1E+07 1E+08 •=

•g 1E+06

Yeast Injection

Wash out at Steady State

1E+05 1E+07

Time (Hrs.)

NH4=1.0E-04M (7) -B- NH4=1.0E-02M (T) -Jtt- NH4=1.0E-04M (Y) -A- NH4=1.0E-02M (Y)

Figure 3-39. Cell number versus time. Numbers of yeast and T. neapolitanus in continuous-flow reactors during competition experiments. The addition of yeast and glucose lowered the steady concentrations of T. neapolitanus. When growth was nitrogen-limited, effect on the population of T. neapolitanus were much more pronounced.

132

400-

T. neapolHanuavs. yeast

pH = 6.80 350-

O) E 300-

S 250- N/C Ratio = 4.445:1

200-

Yeast Injection at

Steady State

® 100-

N/C Ratio = 0.045:1 Wash out 50-

250 300 50 150 200 Time (Hrs.)

350 400

NH4=1.0E-04M NH4=1.0E-02M

Figure 3-40. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. neapolitanus and yeast plus T. neapolitanus. Notice that the addition of yeast and glucose greatly diminished the production of sulfate and presumably the activity of T. neapolitanus.

133

0.13

O) e. e o

0.011

£ 1E-05=

1E-06

N/C Ratio = 4.445:1

0.001:

0.0001 E B

N/c Ratio = 0.045:1

-&B-

T. neapolltanuevB. yeast

pH = 6.80

Yeast ln]ectlon •

at Steady Slate —i— 50

—i— 100

—I— 150 200

Time (Hrs.) 250 300

Wash out

350 400

NH4=1.0E-04M NH4=1.0E-02M

Figure 3-41. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. neapolitanus and yeast plus T. neapolitanus. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of yeast and high N/C feed conditions did not.

134

200-

180-

? 160-

N/C Ratio = 0.045:1

Glucose Sufficient Yeast Injection at

Steady State

140-

120-

100-N/C Rtlo = 4.445:1 Glucose Deficient

80-

60-

40- T. naapolltanusvB. yeaet

pH = 6.80 Wash out 20

20 Time (Hrs.)

25 30 35 40

-e- NH4=1.0E-04M -m- NH4=1.0E-02M

Figure 3-42. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving yeast and T. neapolitanus. Results confirm that growth was glucose-limited in the high-N/C, but not the low-N/C experiments.

135

3.5-

3-e=> E, •g 2.5-•o •D <

T. neaPolHanu»v». yeast

pH = 6.80

O at

2-

i 1.5 oi •s

1-o E <

0.5-

Steady State N/C Ratio = 4.445:1

Steady State

N/C Ratio = 0.045:1

/ Yeast Injection at Steady State

ilnii i ii i ii ii ii i i i i i i i ii i" r i ii i 0 48 96 144 192 240 288 324 339 345 351 360 368

Time (Hrs.)

I 1 NH4=1.0E-04M I— NH4=1.0E-02M

Figure 3-43. Periodic base addition for the maintenance of pH at 6.80 in continuous-flow cultures of T. neapolitanus and yeast. Notice that the rate of acid production was adversely affected by the addition of glucose and yeast at about 325 hours. In the low N/C culture (low-nitrogen feed or culture that was expected to be nitrogen-limited), acid production was almost completely interrupted by the shift in feed conditions.

136

preventing the complete elimination of T. neapoliianus via washout under nitrogen-

limited conditions. Experimental results are summarized in Table 3-1.

3.4.3.2 Variations in Competitor Identity. Experiments involving T.

neapoliianus and E. coli were run at pH 6.80. Only two experiments, corresponding to

different influent ammonium concentrations were run. The N/C feed ratios were

selected above and below the critical N/C ratio for E.coli. Results corresponding to

experiments conducted at 1.556 and 4.445 are summarized in Figures 3-34 through 3-49.

The steady numbers of T. neapoliianus prior to the introduction of glucose and E. coli

were directly related to the influent concentration of ammonia (Figure 3-44). Following

inoculation with E. coli, the high N/C cultures grew into glucose limitation, and low N/C

cultures were nitrogen-limited (Figures 3-46). A. significant glucose residual was

measured at all times at the low ammonium feed rate (Figure 3-48).

As illustrated in Figure 3-44, cell numbers provided direct evidence of the effects

of glucose/E. coli addition. In the high-ammonia-feed-rate experiment, the competition

lowered the numbers of T. neapoliianus by at least an order-of-magnitude. When

nitrogen limitation was established (Figure 3-44), numbers of thiobacilli dropped by more

than three orders-of-magnitude. E. coli grew rapidly by more than two orders-of-

magnitude in each case. Rates of base addition and effluent sulfate concentration are

summarized in Figures 3-49 and 3-45, respectively. In the ammonia-limited culture, rates

of acid generation were almost nil following the establishment of the E. coli population.

Numerical results are again summarized in Table 3-1.

137

1E+09q ; T. neapolHanuBV. E.coll

1E+06

N/C Ratio = 4.445:1

Time (Hre.)

p1E+10

pH = 6.80

o> y. ie+OS 1E+09 a)

N/C Ratio = 1.556:1

1E+08 O = 1E+07

E. coll Inlectlon

>t Steady Stat* •1E+07

NH4 = 3.5E-03 M (T NH4 = 1.0E-02 M (T —NH4 = 3.5E-03 M (E NH4 = 1.0E-02 M (E

Figure 3-44. Cell number versus time. Numbers of E. coli and T. neapolitanus in continuous-flow reactors during competition experiments. The addition oiE. coli and glucose lowered the steady concentrations of T. neapolitanus. When growth was nitrogen-limited, effect on the population of T. neapolitanus were much more pronounced.

138

40fr

350-

ra E, B O 2 250-

0> o e o 0 a» 1 3 w E a> 3 £

300-

200-

150-

100-

50-

T. neanolftanus v». E. coll

pH = 6.80 N/C Ratio = 4.445:1

J* N/C Ratio = 1.556:1

•Jr Ecoll Injection S<C~// •

at Steady State Washout

50 100 150 200 250 300 Time (Hrs.)

350 400 450 500

NH4 = 3.5 E-03 M NH4 = 1.0 E-02 M

Figure 3-45. Effluent sulfate concentration as a function of time and cultures conditions in continuous-flow cultures of T. neapolitanus and E. coli plus T. neapolitanus. Notice that the addition of E. coli and glucose greatly diminished the production of sulfate and presumably the activity of T. neapolitanus.

139

•e—a—• -€Lj

N/C Ratio = 4.445:1

Glucose I Deficient

E. coll Injection at Ammonia Deficient

T. neapolHanueva. E coll

pH = 6.8 Steady State

-m- NH4 = 3.5 E-03 M -B- NH4 = 1.0 E-02 M

Figure 3-46. Effluent ammonia concentration as a function of time and culture conditions in continuous-flow cultures of T. neapolitanus and E. coli plus T. neapolitanus. Results confirm that low N/C feed conditions did result in nitrogen limitation following the addition of E. coli and high N/C feed conditions did not.

140

180-

160 ̂

i

e o S

140-

120-

E. coll

Injection at Steady

State ̂

N/C Ratio = 1.556:1

« 01 o e o A • 00

100-

80-

\ \. QIUCOM Sufficient

\ "" *——m

o 3 O 60- N/C Ratio =» 4.445:1

E 0> 3

§

40-

20-

0-

T. neaDolttanu«v«. E.coll

pH = 6.80

Glucose Deficient

\. Wash out is— a

• i i 0 20 40 60 80 100 120

Time (Hrs.)

NH4 = 3JS E-03 M -EH NH4 = 1.0 E-02 M

Figure 3-47. Effluent glucose concentration versus time and culture conditions in continuous-flow competition experiments involving E. coli and T. neapolitanus. Results confirm that growth was glucose-limited in the high-N/C, but not the low-N/C experiments.

141

3.50-]

3.00-

S 2.50-XI XI <

O 2.00-« z 2 0 1.50-01 •s E 3 O E <

T. naapolltanua v*. E.coll

pH = 6.80

1.00-

0.50-

0.00- i

StMdy State

li Kwty State

weiutio>ijaM

N/C Ratio = 4445:1

Watiiout

i i i i i i i i i i i i i i i i ii II rr i II i i 0 48 96 144 182 240 288 336 386 384 396 414 432 456

Time (Hrs.)

I 1 NH4 ° 3J E-03 M BM NH4 - 1J E-02 M

Figure 3-48. Periodic base addition for the maintenance of pH at 6.80 in continuous-flow cultures of T. neapolitanus and E. coll Notice that the rate of acid production was adversely affected by the addition of glucose and E. coli at about 350 hours. In the low N/C culture (low-nitrogen feed or culture that was expected to be nitrogen-limited), acid production was almost completely interrupted by the shift in feed conditions.

TABLE 3-1. Comparison of Measured Parametric Leveb and Percentage Difference in CSTR Competition Experiments before and after Glucose/Competitor Addition.

Expt. Run

1.

Species Fair PH

Feed Stream N/C Ratio

Before Glucose Addition After Glucose Additkn Percentage Difference Expt. Run

1.

Species Fair PH

Feed Stream N/C Ratio

NH/ so4l DOC Cetl#/ml NH/ SO,"1 DOC CeD#Anl NH/ SO,' DOC Cell#/tal

Expt. Run

1. T. thioeoidans

vs. Yeast

325 0.04S 7E-06 200 200 4E+07 1E-06 25 130 6E+05 86 88 35 99

Expt. Run

1. T. thioeoidans

vs. Yeast

325

0223 1E-Q5 500 • 8E+07 1E«6 30 • 3E+06 90 94 • 96

Expt. Run

1. T. thioeoidans

vs. Yeast

325

1556 1E-04 680 0 1E+08 3E-05 100 0 5E+06 70 85 • 95

Expt. Run

1. T. thioeoidans

vs. Yeast

325

4.445 3E-03 800 180 4E+08 4E-04 350 10 1E+08 87 56 94 75

2. T. thioaddans

vs. Yeast

2.00 1.556 8E-Q5 550 165 1E+08 5E-06 510 115 8B+07 94 7 30 20 2. T. thioaddans

vs. Yeast

2.00

4.445 6E-03 850 165 4E+08 5E-04 810 85 3E+08 92 5 48 25

2. T. thioaddans

vs. Yeast

250 1556 1E-04 600 167 2E+08 3E-06 50 85 4B+05 97 92 49 100

2. T. thioaddans

vs. Yeast

250

4.445 6E-03 860 163 4E+08 1E-04 350 20 1E+08 98 60 88 75

2. T. thioaddans

vs. Yeast

325 1.556 1E-04 680 0 8E+07 1E-06 30 0 3B+06 90 94 • 96

2. T. thioaddans

vs. Yeast

325

4.445 3E-03 800 • 1E+08 3E-05 100 0 5E+06 70 85 • 95

3. 21 thioaddau vs.

Mixed Seed

2.00 1556 8E-05 580 165 1E+08 1E-05 550 105 1E+08 99 5 36 0 3. 21 thioaddau vs.

Mixed Seed 325 1556 8E-05 530 170 2E+08 5E-06 120 80 6E+05 94 77 53 100

4. T. neapoBtaw vs.

Yeast

6JB0 0.045 9E-06 50 170 1E+08 1E-06 <10 120 1B+07 89 85 30 90 4. T. neapoBtaw vs.

Yeast

6JB0

1X223 5E-05 120 165 8E+07 2E-06 20 125 1E+05 96 84 25 100

4. T. neapoBtaw vs.

Yeast

6JB0

1556 5E-04 220 170 25E+08 2E-06 60 90 1E+06 100 73 47 100

4. T. neapoBtaw vs.

Yeast

6JB0

4.445 6E-03 375 170 4E+08 1E-04 175 25 1E+Q5 98 53 85 100

'

T. neapotiltmui vs.

E-coB

6JB0 1556 7E-04 200 167 2E+08 1E-06 15 95 4E+07 100 93 43 80

'

T. neapotiltmui vs.

E-coB

6JB0

4.445 5E-03 360 163 4E+08 6E-05 80 18 3E+07 99 78 89 93

NOTE: • Analysis not performed. Units for NH/, SO/2, and DOC are mo)et/l, mg/1, and mg/U respectively.

143

CHAPTER 4

DISCUSSION

4.1 Sample Collection and Cultivation.

4.1.1 Crown Conditions. The surface pH and extent of corrosion of the sewer

crowns at sample collection locations were inversely related. It is virtually certain that

elevated rates of sulfuric acid consumed the alkalinity at the inner surface of the sewer

pipe. The remaining concrete, with less alkalinity, was consumed at a faster rate in order

to balance the same rate of acid production. This resulted in the conversion of concrete

to gypsum with loss of pipe integrity. Crown pH was lowest near the syphon outfall,

possibly as a result of H2S production in the syphon and/or turbulence at the outfall.

Moderate crown pH for the pipe by the golf course may have been caused by the

periodic submergence and washing of this area by sewage due to diurnal variations of the

sewage flow rate and depth.

The pipe crown was generally dry in the sampling locations. The lack of moisture

and low pH did not appear to mitigate the activity of thiobacilli. Numerous heterotrophs

were also present at the pipe crowns with low pHs, suggesting that they coexist with

thiobacilli in the acidic environments. Heterotrophic enrichments from the pipe crowns

grew well at low pH or mid pH, on solid or in liquid media.

144

The manhole by the syphon outfall showed heavy corrosion, and pH of the crown

was less than 1.0. The concrete securing the steel of the RCP and the manhole lid were

deteriorated, presumably via Thiobacillus-mduced corrosion. The pipelines by the golf

course appeared to be intact although they might have been exposed to HjS for a

considerable amount of time. This manhole (by the golf course) showed very little

corrosion; sewer crown pH was about 5.0. Great differences in corrosion rates might be

attributable to variations in pipe material, sewage age, velocity of flow, and chemical

constituents of the wastewater.

4.1.2 Enrichment and Isolation from Crown Samples. Initial heterotrophic

enrichment media used for the sample isolates produced dense, mixed cultures.

However, the first-transfer enrichments produced considerably fewer mixed growth. That

is, the number of slower-growing heterotrophs decreased. Nevertheless, all of the first-

transfer cultures showed signs of growth. Both bacteria and yeast were evident. Mixed

cultures may have contained both heterotrophs and thiobacilli, but pH stability in both

low (3.75) and mid-pH (6.80) cultures eliminated the possibility of lithotrophic growth.

The growth media was identical to that described in the ATCC Handbook of Bacterial

Medium (1984) for Thiobacillus organoparus, a facultative autotroph and should have

supported heterotrophic growth (Vishniac, 1972) among the facultative thiobacilli initially

present.

Many types of organisms grew on the Bacto-agar used to fortify solid media for

the isolation of heterotrophic organisms. However, successive isolation procedures,

consisiting at each stage of serial dilution, spread-plating, and transfer/growth of cultures

145

in liquid media progressively reduced the number of microbial species that remained.

Both low- and mid-pH plates yielded "monclonal" colonies of yeast identical in

appearance, color, and strain.

Strains of yeast were isolated from both pipe crowns, at points where the pH was

both very low and moderate. There is no reason to suspect that all yeast isoaltes were

identical. The species isolated from the acidified crown was capable of growth at a veiy

wide range of pH values (3 to 8). All strains were identical in appearence under phase-

contrast light microscopy and colony morphologies were indistinguishable on solid media.

This suggested that the strain isolated could function in the low and neutral pH

environment in conjunction with either acidophilic (T. thiooxidans) or mid-pH thiobacilli

(T. neapolitanus).

The yeast isolate from the Roger Road Treatment Plant (on the syphon outfall

side) was capable of growing adequately on glucose as substrate and ammonia as its

nitrogen source in a defined (selective) media. These isolates grew into round, pale

yellow colonies, which turned orange-pink in color with age (5 to 7 days). Their round

shape, growth via budding, and tendency to form clusters of 4 to 8 in liquid, led to the

classification of the sewer crown isolate as yeast This fulfilled an inital objective of the

research; isolation procedures had produced a pure, microbial culture that was capable

of heterotrophic growth under low- and neutral-pH conditions in a defined medium.

146

4.2 Stoichiometric Studies.

4.2.1 Ammonia-to-GIucose Utilization Studies. Yield coefficients, growth rates,

cell consumption, etc., were the same at both, low and mid pHs for yeast. In batch

cultures during balanced growth, the yeast isolate consumed nitrogen and carbon at

approximately a 1:2 ratio (moles N/moles C). The stoichiometry of ammonia and glucose

was estimated from the shapes of the curves in Figure 3-3 and 3-4.

The terminal optical density at stationary-phase growth was a function of initial

nitrogen concentration under ammonia limiting conditions. At low N/C, the dependence

of the optical density at stationary-phase on initial ammonia was stronger than that at

high N/C. A 16-fold increase of ammonia from 50 uM to 800 uM resulted in a 3-fold

increase in cell population. However, the nine-fold increase from 800 uM to 7000 uM

resulted only in a 30% increase in cell density. This suggested that at transition from

nitrogen independence to nitrogen limitation occured between 800 uM to 7000 uM.

Transition from nitrogen to carbon conditions is not sharp. Because, cells may find ways

to use ammonia more efficiently for growth as nitrogen limitation is approached.

Estimated critical N/C ratio for E. coli was about the same as yeast, that is, 1:2 (moles

N/moles C).

If organic carbon is the substrate, about 50% of it is utilized via respiration or

oxidation to C02; this results in enough ATP to convert the rest 50% into biomass

(Gottschal, 1978). For both E. coli and yeast, the N/C critical ratios closely related to

the N/C ratio in dry mass of a cell, adjusted for the percentage of available carbon which

is assimilated into biomass (Stanier, 1976).

147

4.2.2 Ammonia-to-ThiosuIfate Utilization Rates. When T. neapolitanus was

provided with thiosulfate, the stoichiometry of ammonia-to-thiosulfate utilization was

approximately 1:1.5 (moles N/moles S203'2). This was the critical N/S ratio. At this point,

where transition from nitrogen to thiosulfate limitation was encountered, growth rate

decreased abruptly by about 80%. This abrupt decrease suggested that oxidation of some

forms of sulfur continues after thiosulfate is exhausted-presumably oxidation of

metabolic intermediates in the way to sulfate (Kelly, 1982). Further identification of

steps along thiosulfate oxidation pathway and oxidation of intermediates would require

measurement of expected sulfur intermediates and thiosulfate. This oxidation process

is usually brought about by a complex thiosulfate-oxidizing multi-enzyme (Gottschal,

1986). This system of thiosulfate oxidation yields an end product of two sulfate

molecules formed for every thiosulfate molecule oxidized.

43 Two-Organism Competition Studies.

43.1 Low-pH Competition Experiments.

Variation in influent ammonia-to-glucose ratio. Results indicated that

growth and acid production by T. thiooxidans were restricted in the presence of glucose

and a yeast cell population that is capable of competing for limiting nutrients. It is also

evident that both T. thiooxidans and the acidophilic yeast behaved as predicted in the

presence of nitrogen-limiting conditions. That is, in carbon-rich environments, the faster-

growing species (heterotroph or yeast) outcompeted T. thiooxidans for a growth-limiting

substrate, in this case nitrogen. However, ammonia limitation was not entirely

148

responsible for decreases in T. thiooxidans numbers and activity. Reductions in their cell

numbers and activity were also evident in experiments in which the influent glucose

concentration limited yeast growth and excess ammonia was present in the reactor

effluent. From Figure 3-14, following the introduction of yeast and excess glucose (N/C

= 4.445:1), T. thiooxidans cell density reduced from 4 x 10® to 1 x 108 cells/ml and

effluent sulfate concentration dropped from 800 to 300 mg/1 (Figure 3-15). These

reductions occurred despite the persistence of ammonia-nitrogen in the reactor. Because

effluent sulfate decreased by no more (relatively speaking) than the numbers of

thiobacilli, it is evident that the population of T. thiooxidans was metabolically active

(continued to oxidize sulfur) in both the presence and absence of glucose.

The difference in generation rates of yeast (1 to 2 hours) and T. thiooxidans (12-

16 hours) gave yeast a tremendous competitive edge over T. thiooxidans. The advantage

of this edge is most evident in the results of experiments in which nitrogen proved to be

growth-limiting. T. thiooxidans was essentially washed out within 40 hours after the

introduction of the yeast isolate and excess glucose.

There is a physiological basis for difference in generation time of these organisms.

Glucose is energy-rich. Its energy yield has been estimated at 38 ATP/mole of glucose

(Stanier, 1976), and metabolic intermediates of glucose catabolism are backbones for

some amino acids (Kelly, 1982). Glucose can be converted to other sugars. Both these

process facilitate carbon assimilation. T. thiooxidans oxidizes reduced sulfur forms for

energy and must expend a significant fraction of that energy to build all its cellular

components.

149

Under nitrogen-limiting conditions, levels of ammonia and sulfate/acid generation

rates decreased drastically to a point where ammonia and sulfate were no longer

measurable (Figures 3-11; 3-12 and 3-15; 3-16). Under carbon-limiting conditions,

effluent sulfate, and ammonia were reduced by an order-of-magnitude and glucose in the

effluent was virtually nil (Figures 3-15, 3-16, and 3-17). However, ammonia-limitation

was not encountered. Periodic base consumption for the maintenance of the culture pH

at 3.25 complimented these results (Figure 3-18). That is, after the introduction of the

yeast, the amount of base added to neutralize the acid produced decreased by about

90%.

Variations in Culture pH. When experiments were conducted at pH

2.50, yeast experienced a lag period of about 24 hours and then outcompeted T.

thioaxidans for a limiting supply of nitrogen (Figure 3-19). From batch experiments (data

not included), the specific growth rate of yeast at pH 2.50 is considerably lower than at

pH 3.25. A 24 hour lag period was probably experienced due to the low competitive

edge yeast had at pH 2.50 when compared to its balanced growth at pH 3.25.

Consequently, a more gradual decrease in T. thioaxidans activity was observed following

initiation of growth by yeast. The concentrations of ammonia and glucose dropped to

non-measurable levels in nitrogen-limited and carbon-limited cultures, respectively

(Figure 3-21 and 3-22). Three-fold reductions in cell numbers of T. thioaxidans and

effluent sulfate concentration in the high N/C experiment (Figure 3-19 and 3-20) and

virtual disappearance of thiobacilli and sulfate at low N/C feed concentration were

observed (also Figures 3-19 and 3-20).

150

When the same experiment was conducted at pH ZOO, yeast seemed to suffer a

from excessively low pH although they persisted in very low numbers. Appreciable

decreases in T. thioaxidans activity were not observed. Thiobacilli cell numbers and

sulfate production decreased immediately after injection of yeast but regained their

original steady-state concentrations after about 18 hours (Figures 3-24 and 3-25). The

result was unexpected in light of results of enrichment experiments in which acidophilic

yeast were encountered in a sewer crown of pH less than 1.0. However, possible

explanations for both these observations would be — (i) at Iow-pH, yeast goes into

quiescent state, (ii) crown has lots of micro-environments which have pH greater than

what could be measured, and (iii) crown pH is very transient.

Although the results of the experiments are very encouraging from the standpoint

of mitigating crown corrosion, the thiobacilli did not completely wash out qf the CSTR.

T. thioaxidans seem to persist in low numbers (>. 1 x 10s cells/ml) in the nitrogen-limited

experiments. This probably might have been due to their ability to colonize sulfur

particles and/or attach to the reactor walls and mixing apparatus.

Variations in Competitor Identity. At pH 3.25, the mixed-seed

competitors seemed to outcompete within about 3 days after the cross inoculation. This

might have been due ~ (i) to the lower generation times of the various organisms present

in the mixed-seed as compared to the slower-growing T. thioaxidans and/or (ii) the yeast

might have been present in the mixed-seed which dominated the thiobacilli. At pH 2.00,

the potential heterotrophic competitors again seemed to be pH-limited. This might have

been due to the fact that the mixed-seed competitors were obtained from a higher pH

151

micro-environment. The competitors were active for about 50 hours after inoculation,

but numbers/activity of thiobacilli were eventually restored, presumably due to the

advantage enjoyed by T. thioaxidans under exceptionally extreme low-pH conditions.

4.3.2 Mld-pH Competition Experiments.

Variations in influent ammonia feed ratio. From the results, it is evident

that T. neapolitanus activity was mitigated by yeast under nitrogen-limiting conditions.

At high nitrogen concentrations, T. neapolitanus was thiosulfate-limited and the yeast

isolate was glucose-limited. At steady-state T. neapolitanus apparently oxidized all the

S203"2 (influent concentration is 325 mg/L as S) to SO/2 (effluent sulfate concentration

330 mg/1), which resulted in thiosulfate limitation. Following the inoculation of yeast,

moderate reductions in T. neapolitanus cell numbers, effluent sulfate concentration, and

nitrogen concentration were observed in the high ammonia experiments.

Under nitrogen-limiting experiments, yeast addition brought about appreciable

decreases in T. neapolitanus activity (one and two orders-of-magnitude in acid production

and cell density, respectively). Reductions in effluent sulfate and T. neapolitanus cell

numbers were due to the inability to keep up with the rapid growth of yeast. Yeast

assimilated nitrogen faster than T. neapolitanus, which resulted in the decrease of

Thiobacillus numbers and activity. After the introduction of yeast, nitrogen in the

effluent was nil and glucose dropped down by 30%. This indicated that the cultures were

nitrogen deficient and glucose sufficient. Failure to eliminate the slower-growing bacteria

may be again attributable to bacterial attachment and colonization on precipitated sulfur

particles and reactor surfaces.

152

Variations in Competitor Identity. E. coli seemed to outcompete T.

neapolUanus at a faster rate when compared to yeast. This might have been due to the

low generation time of E. coli (20-30 minutes) as opposed to yeast (1-2 hours). Due to

the low doubling time, E. coli rapidly outgrew T. neapolUanus. Reductions in T.

neapolUanus cell numbers and acid productions were almost the same in experiments

involving E. coli and yeast, when conducted at low nitrogen concentrations. However,

E. coli utilized both glucose and nitrogen sooner than the yeast isolate, under similar

experimental conditions. The pH (6.8 units) at which the experiment was conducted may

have favored the heterotroph unduly, inasmuch as crown pHs are frequently lower, even

when conditions are only modestly aggressive.

153

CHAPTER 5

CONCLUSIONS

5.1 Isolation of Sewer Crown Samples.

Sewer crown samples collected from the Roger Treatment Plant {syphon outfall)

were enriched on defined (selective), heterotrophic, low-(3.25), and mid-pH (6.80) media

containing glucose-carbon and ammonia-nitrogen. This glucose-based media supported

growth of original samples collected. The first-transfer cultures also grew in both low-

and mid-pH media. Diversity was significantly lower in the first-transfer cultures than

the original samples. Growth in the original enrichment and first-transfer cultures was

dependent on the nutrients in the designed media.

Heterotrophs grew in the initial enrichment media (Table 2-3) and in the first-

transfer media (also Table 2-3) in about 3 to 5 days. Thiobacilli activity was nonexistent

due to pH stability. Successive serial dilutions, transfers, and spread plating on solid

media yielded a isolate capable of growth in a defined media.

The monoclonal isolates were identified as yeast by phase-contrast light

microscopy, and colony morphology. They were round-shaped, refractile, formed orange-

pink colonies, and formed clusters of 4 to 8. Ouster formation was preceded by budding.

Colonies turned orange-pink after 3-5 days of growth in liquid and on solid media.

154

5.2 Stoichiometiy of Ammonia-to-GIucose Utilization Rates.

The yeast isolates and E. coli demanded about 1:2 molar ratio of nitrogen-to-

glucose, in batch experiments conducted at room temperature (25°C). At the point of

carbon or nitrogen limitation, yeast and E. coli stopped growing. The terminal optical

density at the onset of stationary-phase growth was a function of the initial ammonia

concentrations under ammonia limiting conditions. At the highest nitrogen

concentration, growth was dependent on initial carbon concentration. For both the

organisms, the nitrogen-to-glucose ratio at which transition from nitrogen-limited to

glucose-limited growth occurred was the critical N/C ratio~N/C = 1:2 (moles N/moles

C). The growth and nitrogen-based yield factors (stoichiometrics) for the yeast strains

used in these studies were identical at low- (3.25) and mid-pH (6.80) levels.

The specific growth rate of yeast decreased gradually with the decrease in pH in

the range 2 to 2.75. Growth was pH-limited at <. 2.25 pH units. The most favorable pH

condition for the growth of yeast was between 3.00 and 8.00 pH units. Persistence of

the experimental strain under field conditions (pH < 1.0) cannot be explained on the

basis of these data.

S3 Stoichiometiy of Sulfur Oxidation.

T. neapolitanus consumed 1.5 moles of thiosulfate for every mole of nitrogen

assimilated. The terminal cell numbers at stationary-phase was a function of the initial

nitrogen concentration under ammonia-limiting conditions. When cultures were

155

thiosulfate-limited, all S203'2 was oxidized to SO/2. The gradual decrease in growth at

stationary-phase suggested that the oxidation of sulfur intermediates was possible.

5.4 Competitive Inhibitory Control of Thiobacilli.

From this research, both T. thioaxidans and T. neapolitanus behaved as predicted

in the presence of yeast and nutrient-limiting conditions. That is, the thiobacilli are

probably not at their best in organic-rich environments where a growth-limiting nutrient

such as nitrogen can be identified. Under experimental conditions, acid production by

both T. thiooxidans and T. neapolitanus was substantially curtailed by encouraging

conditions favorable to yeast, the mixed heterotroph seed or E. coli. At least two order-

of-magnitude reductions in thiobacilli cell numbers and effluent sulfate concentrations

were observed under nitrogen-limiting conditions. There is a reason to suspect that

glucose enrichment of competitive populations will limit the activity of thiobacilli in a

sewer environment, even though specific nutrient limitations cannot be identified.

Reductions in these parameters (sulfate concentrations and Thiobacillus cell numbers)

were observed under carbon-limiting conditions, although results were considerably more

modest. The mechanism of this inhibition under glucose limitation remains speculative.

In conclusion, growth and acid production by the chemolithoautotrophic

thiobacilli in liquid media can be substantially mitigated via the encouragement of

microbial competitors, if the collective activities of these competitors result in the

elimination or near elimination of essential nutrients. Such measures could be

implemented in corroding sewers without risk to the general sewage biota or jeopardy

156

to the downstream wastewater treatment operations. This utility of competitive control

of thiobacilli acid production in the sewer crown environment remains to be established.

157

REFERENCES CITED

APHA, AWWA, and APCF (1976). Standard Methods for the Examination of Water and Wastewater. 427c. Sulfate, pp. 493-495.

Arnold, R. A. (1988). Asst. Professor, University of Arizona. Personal Communication.

ATCC (1984). American Type Culture Collection Media Handbook, 1st Edition.

Brock (1974). "Organisms Metabolizing Sulfur and Sulfur Compounds, Genus 2-5." Bergev's Manual of Determinative Bacteriology (8th Edition). Ed. R. E. Buchanan and N. E. Gibbons. The Williams and Wilkins Company, pp. 458-461.

Bailey, J. E. and D. F. Ollis (1986). Biochemical Engineering Fundamentals (2nd ed.,); McGraw-Hill Book Co., New York, pp. 854-902.

Burris, M. J (1990). Unpublished Data. University of Arizona, Civil Engineering.

Cassell A. E., Sulzer, F. T., and Lamb, J. C. HI. (1966). Population Dynamics and Selection in Continuous Mixed Cultures. Journal WPCF.. Vol. 38, pp. 1398-1409.

Chiu, S. Y., L. E. Erickson, L. T. Fan, and I. C. Kao. (1972). Kinetic Model Identification in Mixed Populations Using Continuous Culture Data. Biotech, and Bioeng., Vol. XIV, pp. 207-231.

Collins, Y. E. and G. Stotzky (1989). "Factors Affecting the Toxicity of Heavy Metals to Microbes", in Metal Ions and Bacteria, (eds.) Beveridge, T. J. and R. J. Doyle, John Wiley & Sons, New York, pp. 60-61.

Contois, D. E. (1959). Kinetics of Bacterial Growth: Relationship between Population Density and Specific Growth Rate of Continuous Culture. J. Gen. Microbiol.. Vol. 21, pp. 40-50.

Curds, C. R. (1973). A Theoretical Study of Factors Influencing the Microbial Population Dynamics of the Activated-Sludge Process-I. Water Research. Pergamon Press, Vol. 7, pp. 1269-1284.

158

Curds., C. R., and A. Cockburn (1971). Continuous Monoxenic Culture of Tetrahymma pyriformis. Journ. of Gen. Microbiol.. Vol. 66. pp. 95-108.

Espejo, R. T., and P. Romero (1987). Growth of ThiobacUlus feroaxidans on Elemental Sulfur. Appl. Environ. Microbiol.. Vol. 53, pp. 1907-1912.

Espejo, R. T., B. Escobar, E. Jedlicki, P. Uribe, and R. Badilla-Ohlbaum (1988). Oxidation of Ferrous Iron and Elemental Sulfur by ThiobacUlus feroaxidans. Appl. Environ. Microbiol.. Vol. 54, pp. 1694-1699.

Gibson, T., and Y. Abd-El-Malek (1957). The Development of Bacterial Populations in Milk. Can. Journ. of Microbiol.. Vol. 3, pp. 203-213.

Gottschal, J. C., and Kuenen, G. J. (1980). Selective Enrichment of Facultatively Chemolithotrophic Thiobacilli and Related Organisms in Continuous Culture. FEMS Microbiology Letters 7, pp. 241-247.

Gottschal, J. C., and Kuenen, G. J. (1980). Mixotrophic Growth of ThiobacUlus A2 on Acetate and Thiosulfate as Growth Limiting Substrates in the Chemostat. Arch. Microbiol.. Vol. 120, pp. 33-42.

Gottschal, J. C., Pol, A., and Kuenen, G. J. (1981). Metabolic Flexibility of ThiobacUlus A2 During Substrate Transitions in the Chemostat Arch. Microbiol.. Vol. 129, pp. 23-28.

Gottschal, J. C., Vries de Sacco, and Kuenen, G. J. (1979). Competition Between the Facultatively Chemolithotrophic ThiobacUlus A2, an Obligately Chemolithotrophic ThiobacUlus and a Heterotrophic Spirillum for Inorganic and Organic Substrates. Arch. Microbiol.. Vol. 121, pp. 241-249.

Gottschal, J. C., Nanninga, H. J., and Kuenen, G. J. (1981). Growth of ThiobacUlus A2 under Alternating Growth Conditions in the Chemostat. Journ. of Gen. Microbiol.. Vol. 126, pp. 85-96.

Gottschalk, G. (1986). Bacterial Metabolism (2nd Edition). New York: Springer-Verlag. pp. 293-300.

Herbert, D., Elsworth, R. and Telling, R. C. (1956). The Continuous Culture of Bacteria; A Theoretical and Experimental Study. J. Gen. Microbiol.. Vol. 14, pp. 601-622.

Hijmans, W., H. R. E. Schuit, and F. Klein. (1969). Techniques: An Immunofluorescence Procedure for the Detection of Intracellular Immunoglobulin. Clin. Exp. Immunol.. Vol. 4, pp. 457-472.

159

Holder, G. A., W. Drew, and G. Vaughan (1983). "Studies of Sulphate Removal in a Model Sewer." Tech. Pap. Conv. Aust. Water/Wastewater Assoc.

Jack, T. R., E. A. Sullivan, and J. E. Zajic (1979). Leaching of Vanadium and Other Metals from Athabasca Oil Sands Coke and Coke Ash. Fuel. Vol. 58, pp. 585-588.

Jack, T. R., E. A. Sullivan, and J. E. Zajic (1980). Growth Inhibition of Thiobacittus thioaxidans by Metals and Reductive Detoxification of Vanadium (V)". Eur. J. Appl. Microbiol. Biotechnol.. 9, pp. 21-30.

Jin, C. G. (1987). "Sulfide Control with Sodium Hydroxide in Large Diameter Sewers." LACSD In-house report.

Jost, J. L., J. F. Drake, A. G. Fredrickson, and H. M. Tsuchiya. (1973). Interactions of Tetrahymena pyriformis, Escherichia coli,Azotobactervinelandii, and Glucose in Minimal Medium. Journ. of Bacteriol.. Vol. 113, pp. 834-840.

Kelly, D. P. (1981). "Introduction to the Chemolithotrophic Bacteria" Chapter 79 in: The Prokarvotes. ed. M. P. Starr. Spring-Verlag, pp. 997-1004.

Kelly, D. P. (1982). "Biochemistry of the Chemolithotrophic Oxidation of Inorganic Sulphur." Phil. Trans. R. Soc. Lond.. 298, pp. 499-528.

Kempner, Ellis, S. (1966). "Acid Production by Thiobacittus thioaxidans." Journ. of Bacteriology. 92, pp. 1842-1843.

LACSD (1987). County Sanitation Districts of Los Angeles County. In-house Report.

LACSD (1988). County Sanitation Districts of Los Angeles County. In-house Report.

Meers, J. L. (1971). Short Communications: Effect of Dilution Rate on the Outcome of Chemostat Mixed Culture Experiments. Journ. of Gen. Microbiol.. Vol. 67, pp. 359-361.

Meers, J. L., and D. W. Tempest. (1968). The Influence of Extracellular Products on the Behavior of Mixed Microbial Populations in Magnesium-Limited Chemostat Cultures. J. Gen. Microbiol.. Vol. 52, pp. 309-317.

Megee, R. D., Ill, J. F. Drake, A. G. Fredrickson, and H. M. Tsuchiya (1972). Studies in Intermicrobial Symbiosis. Saccharomyces cerevisiae and Lactobacillus casei. Can. Journ. of Microbiol.. Vol. 18, pp. 1733-1742.

160

Meyer J. S., H. M. Tsuchiya, and A. G. Fredrickson (1975). Dynamics of Mixed Populations Having Complementary Metabolism. Biotech, and Bioeng.. Vol. XVII, pp. 1065-1081.

Milde, K., W. Sand, W. Wolff, and E. Bock (1983). "Thiobacilli of the Corroded Concrete Walls of the Hamburg Sewer System." J. of Gen. Microbiol.. 129, pp. 1327-1333.

Milner, L. (1988). Unpublished Data. University of Arizona, Civil Engineering.

Neal, J. H. (1964) "Advanced Waste Treatment by Distillation." U. S. Public Health Service Report. 999-WP-9.

Oberhofer, T. R., and W. C. Frazier (1960). Competition of Staphylococcus aureus with Other Organisms. Biometrika.

Parker, C. D. (1945). "The Corrosion of Concrete. 1. The Isolation of a Species of Bacterium Associated with the Corrosion of Concrete Exposed to Atmospheres Containing Hydrogen Sulfide. 2. The Function of Thiobacillus concretivorus in the Corrosion of Concrete Exposed to Atmospheres Containing Hydrogen Sulfide." Aust. Jour. Exptl. Biol, and Med. SC.. V23, pp. 81 and 91.

Parker, C. D. (1951). "Mechanics of Corrosion of Concrete Sewers by Hydrogen Sulfide." Sewape and Industrial Wastes. V23, nl2, pp. 1477-1485.

Pfennig, N., F. Widdel, and H. G. Truper (1971). Chapter 74 of The Prokarvotes. ed M. P. Starr. Spring-Verlag.

Pohl, M., E. Bock, M. Rinken, M. Aydin, and W. A. Konig (1984). "Volatile Sulfur Compounds Produced by Methionine Degrading Bacteria and the Relationship top Concrete Corrosion." Z. Naturforsch. 39c, pp. 240-243.

Postgate, J. (1959). "Sulfate Reduction by Bacteria" Ann. Rev. Microbiol.. 13, pp. 505-520.

Postgate, J. R. (1982). Economic Importance of Sulfur Bacteria. Phil. Trans. R. Soc. Lond. Series B, Vol. 298, pp. 583-600.

Rapport, D. J., and Turner, J. E. (1975). Predator-Prey Interactions in Natural Communities. J. theor. Biol.. Vol. 51, pp. 169-180.

Sand, W., E. Bock, and D. C. White (1984). "Role of Sulfur Oxidation Bacteria in the Degradation of Concrete." Corrosion 84: The International Corrosion Forum. New Orleans, LA., Paper No. 96, pp. 1-11.

161

Sand, W., and E. Bock (1984). "Concrete Corrosion in the Hamburg Sewer System." Environmental Technology Letters. 5, pp. 517-528.

Sand, W. (1987). "Importance of Hydrogen Sulfide, Thiosulfate, and Methylmercaptan for Growth of Thiobacilli During Simulation of Concrete Corrosion." Applied Environ. Microbiol.. 53, pp. 1645-1648.

Sand, W., K Bock, and D. C. White (1987). "Biotest System for Rapid Evaluation of Concrete Resistance to Sulfur-Oxidizing Bacteria." Materials Performance, pp. 14-17.

Sima, H (1989). Unpublished Data. University of Arizona, Civil Engineering.

Sivela, S., and V. Sundman (1975). "Demonstration of Thiobacillus-Type Bacteria, which Utilize Methyl Sulfides." Arch. Microbiol.. 103, pp. 303-304.

Smirnov, V. V. (1963). Role of the Microbiological Factors in Increasing the Corrosive Aggressivity of Mine Waters of the Kizel Coal Basin. Mikrobiologiva Vol. 32, pp. 695-699.

Smith, A L., and Kelly, D. P. (1979). Competition between an Obligately and a Facultatively Chemolithotrophic Thiobacillus. Journ. of Gen. Microbiol.. Vol. 115, pp. 377-384.

Snoeyink, V. L., and D. Jenkins (1980). Water Chemistry. New York: John Wiley & Sons.

Sponseller, M. P. (1988). Industrial Pretreatment Backfires in L.A. Pipes. Environ. News Res. No. 10, pp. 14-16.

Stanier, R. Y., E. A. Adelburg, and J. Ingraham (1976). The Microbial World Prentice-Hall, Inc. pp. 190-197 and 707-710.

Stumm, W., and J. J. Morgan (1981). Aquatic Chemistry (2nd Edition). New York: Wiley-Interscience Publication.

Tempest D. W., J. W. Dicks, and J. L. Meers (1967). Magnesium-Limited Growth of Bacillus subtilis, in Pure and Mixed Cultures, in a Chemostat. J. Gen. Microbiol.. Vol. 49, pp. 139-147.

Thingstad, F. T., and Gottschal, J. C. (1982). Mathematical Description of Competition between Two and Three Bacterial Species under Dual Substrate Limitation in the Chemostat: A Comparison with Experimental Data. Biotech, and Bioeng.. Vol. XXIV, pp. 1403-1418.

162

Tsuchiya, H. M., J. F. Drake, J. L. Jost, and A. G. Fredrickson. (1972). Predator-Prey Interactions of Dictyostelium discoideum and Escherichia coli in Continuous Culture. Joun. of Bacteriol.. Vol. 110, pp. 1147-1153.

U. S. Environmental Protection Agency, (1974). Process Design Manual for Sulfide Control in Sanitary Sewerage Systems. EPA 625/1-74/005, Washington, D.C.

U. S. Environmental Protection Agency (1985). Odor and Corrosion Control in Sanitary Sewerage Systems and Treatment Plants. Center for Environmental Research Information, Cincinnati, OH.

Vishniac (1974). "Organisms Metabolizing Sulfur and Sulfur Compounds, Genus 1. Thiobacilhis" Bergev's Manual of Determinative Bacteriology (8th Edition). Ed. R. E. Buchanan and N. E. Gibbons. Baltimore: The Williams and Wilikins Company, pp. 458-461.

Wilkinson T. G., H. H. Topiwala, and G. Hamer (1974). Interactions in a Mixed Population Growing on Methane in Continuous Culture. Biotech, and Bioeng.. Vol. XVI, pp. 41-59.

Wolfgang, B. (1969). Determination of Antisera Titres Using the Single Radial Immunodiffusion Method. Immunochemistrv. Pergamon Press, Vol. 6, pp. 539-546.

Zajic, J. E., E. A. Sullivan, T. R. Jack, and G. Silver (1978). Metals, Microbes, and Bituminous Sands Part n. Alberta Oil Sands Technology and Research Authority, Edmonton, pp. 38-74.