EXPERIMENTAL AND NUMERICAL EVALUATION OF COMETABOLISM IN
SEQUENCING BATCH REACTORS

By
Chien-chun Shih

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Civil and Environmental Engineering

1995

A

ABSTRACT

EXPERIMENTAL AND NUMERICAL EVALUATION OF COMETABOLISM 1N
SEQUENCING BATCH REACT ORS

By

Chien-chun Shih

The potential use of sequencing batch reactors (SBRs) for cometabolism was
evaluated both experimentally and numerically. Mass balance considerations were used to
develop a numerical computer-based methodology for SBR design capable of
accommodating cometabolic transformation terms. The experimental system was a phenol-
fed enrichment capable of trichloroethylene (T CE) cometabolism. This system was used to
evaluate the effects of growth substrate feeding pattern on community structure and
cometabolism; the stability of cometabolism; the toxicity of transformation products; and
strategies for recharge of cometabolic activity. Modeling efforts were made to predict TCE

transformation in bench-scale SBRs.

Phenol feeding patterns altered the structure and function of phenol-fed
enrichments. Cells from a pulse fed reactor exhibited high phenol utilization rates, high
diversity, high levels of cometabolic activity, and good settleability. Cells from a
continuously fed reactor were predominately filamentous, exhibiting slow rates of TCE
transformation and poor settleability. In this culture, isolates capable of degrading TCE

were rare, and TCE transformation capacities varied greatly.

Stable and consistent TCE removal was obtained in a bench-scale SBR modified to
include a recharge stage to rejuvenate cometabolic activity. Over the TCE concentration

range from 0.5 to 5 mg/l, an average of 85-95% of the input TCE was biotransforrned and

about 5-10% of TCE was air stripped. Toxic effects, including a decrease in total biomass
and reduced TCE transformation activity, were observed for a short period at the beginning
of TCE addition. Subsequently, biomass levels recovered, and TCE transformation activity
resumed. The adapted system exhibited consistent TCE transformation capacity, suffering
no additional toxic effects for the remaining test period (about 150 days). A numerical

model capable of predicting TCE removal in the SBR was developed and verified.

 

ACKNOWLEDGMENTS

I would like to give my greatest appreciation to my advisor Dr. Craig Criddle of the
Department of Civil and Environmental Engineering for his encouragement, understanding,

and assistance in my research.

A special thank you is given to Dr. James Tiedje, Dr. Mark Worden and Dr. Robert
Hickey for their reviews of the manuscript and suggestions. They enlarged my views of

biological processes with different viewpoints.

Over past five years, many fellow students and friends gave me considerable
assistance and made my Spartan life more enjoyable. I would like to thank all of them from
my heart. I appreciate Dr. Jizhong Zhou and Ms. Mary Ellen Davey for their help in culture
isolation and characterization, Mr. Yanlyang Pan for his assistance in instrument operation,
Dr. Xianda Zhao and Myung-Keun Chang for their valuable comments and assistance in

computer programming and environmental engineering.

I would also like to thank my lab coworkers, Dr. Mike Dybas, Sadhana Chauhan,
Greg Tartara and Blake Key for their help in microbiological assays and sharing of
laboratory resources, Mr. Wang-kuan Chang and I-yuan Hsu for their knowledge of mass
transfer and modeling issues. Discussion with them is always enjoyable and has enriched

my knowledge in different fields.

Special thanks go to my parents and my sisters, whose patient, unfaltering support
and encouragement have made the task an enjoyable one. As a poor rural grower, I would

like to express my greatest appreciation to Taiwan government for scholarship assistance.

A

Without this support, I would not have had the opportunity to purse my Ph.D. degree in
the United States.

TABLE OF CONTENTS

List of Tables ...................................................................................... xi
List of Figures ..................................................................................... xiii
Nomenclature ...................................................................................... xvii
CHAPTER
I. INTRODUCTION AND RESEARCH OBJECTIVES .............................. l
1.1 Introduction .......................................................................... l
1.2 Research objectives ................................................................. 2
1.3 Overview of the chapters ........................................................... 2
II. BACKGROUND ......................................................................... 4
2.1 Selection of a model system for cometabolism .................................. 4
2.2 Properties of trichloroethylene ..................................................... 5
2.3 Causes and status of TCE contamination ......................................... 5
2.4 Biodegradation of TCE ............................................................. 8
2.5 Aerobic TCE degradation pathway ................................................ 9
2.6 Microbial degradation kinetics of TCE ............................................ 12
2.7 Biodegradation of phenol ........................................................... 17
2.8 Cometabolism in engineered systems ............................................. 21
2.9 Choice of microbial systems for cometabolism of TCE ........................ 24
2.10 Choice of reactor systems for cometabolism of TCE ......................... 25

III USE OF NUMERICAL SIMULATION FOR DESIGN OF SEQUENCING

BATCH REACTOR ..................................................................... 36
3.1 Introduction .......................................................................... 36
3.2 General approach .................................................................... 37
3.2.1 Constraints on SBR design ............................................. 37
3.2.2 Mass balance equations ................................................. 40

3.2.3 Use of the CSTR steady state solution to obtain initial

condition estimates .............................................................. 41
3.2.4 Mean solids retention time and wasting schedules ................... 42
3.2.5 Dynamic cell age ......................................................... 43
3.2.6 Generic solution procedure ............................................. 44
3.3 Application to domestic wastewater treatment ................................... 47
3.3.1 Choice of wastewater classification scheme .......................... 47
3.3.2 Choice of kinetic parameters ............................................ 49
3.3.3 Choice of treatment objectives .......................................... 53
3.3.4 Program development and execution .................................. 54
3.3.5 Simulation results ........................................................ 57
3.4 Summary ............................................................................. 67

IV. EFFECTS OF GROWTH SUBSTRATE FEEDING PATTERN ON MICROBIAL

COMMUNTTY AND COMET ABOLISM OF TRICHLOROETHYLENE ........ 69
4.1 Introduction .......................................................................... 69
4.2 Materials and methods .............................................................. 70
4.2.1 Chemicals ................................................................. 70
4.2.2 Media and culture ........................................................ 70
4.2.3 Reactors and operating conditions ..................................... 70
4.2.4 Inoculum .................................................................. 71
4.2.5 Evaluation of TCE transformation kinetics ........................... 71

4.2.6 Measurement of pH, dissolved oxygen, optical density, phenol,

and MLSS ........................................................................ 74
4.2.7 Identification and quantification of filaments ......................... 74
4.2.8 Rotifers enumeration .................................................... 74
4.2.9 Enumeration of phenol utilizers and TCE degraders ................. 75
4.2.10 Pure culture isolation and characterization ........................... 75
4.2.11 Assay for catechol ring fission pathway ............................. 76
4.3 Results and discussions ............................................................ 77
4.3.1 TCE transformation ...................................................... 77
4.3.2 Biomass variation ........................................................ 79
4.3.3 Phenol degradation ...................................................... 82

4.3.4 Phenol utilization patterns for reactors with different feeding

pattern ............................................................................. 82

4.3.5 Effect of feeding pattern on reactor pH and dissolved oxygen ..... 84

4.3.6 Microscopic inspection .................................................. 87

4.3.7 Enumeration of phenol utilizers and TCE degraders ................. 90

4.3.8 Pure culture isolation and characterization ............................ 92

4.4 Conclusions .......................................................................... 97
V. EFFECTS OF TCE TOXICITY ON CELL VIABILITY ............................ 101
5.1 Introduction .......................................................................... 101
5.2 Background .......................................................................... 103
5.3 Material and methods ............................................................... 105
5.3.1 Organisms and conditions .............................................. 105
5.3.2 Assays of TCE and phenol concentrations and cell viability ........ 105
5.3.3 Toxicity effects experiment ............................................. 106
5.3.4 TCE transformation capacity measurement ........................... 107

 

A

5.4 Results and discussions ............................................................ 107

5.4.1 Effect of TCE concentration on cell viability ......................... 107
5.4.2 Effect of cell type on its viability caused by TCE toxicity ........... 107

5.4.3 Quantification and modeling of TCE toxicity on P. cepacia G4 . . .. 110
5.5 Conclusions .......................................................................... 110

EXPERIMENTAL EVALUATION AND NUMERICAL SIMULATION OF
TRICHLOROETHYLENE REMOVAL IN PI-IENOL—FED SEQUENCING

BATCH REACTORS ................................................................... 115
6.1 Introduction .......................................................................... 115
6.2 Modeling consideration ............................................................ 116

6.2.1 Substrate removal mechanisms ......................................... 116
6.2.2 Mass balance equations ................................................. 119
6.3 Materials .............................................................................. 122
6.3.1 Chemicals ................................................................. 122
6.3.2 Bioreactor ................................................................. 122
6.3.3 Media and culture ........................................................ 124
6.4 Experimental ......................................................................... 124
6.4.1 Reactor operating conditions ........................................... 124
6.4.2 Batch assay of TCE transformation activity .......................... 125

6.4.3 Measurement of pH, dissolved oxygen and phenol, and

MLSS ............................................................................. 128
6.4.4 Measurement of TCE in reactor off-gas ............................... 128
6.4.5 Mass balance experiment ................................................ 129
6.5 Results and discussions ............................................................ 130
6.5.1 Mass balance experiment ................................................ 130
6.5.2 Batch TCE degradation experiment .................................... 134
6.6 Bench-scale SBR modeling ........................................................ 139

6.6.1 Growth substrate concentration ........................................ 139

6.6.2 Nongrowth substrate concentration .................................... 141

6.7 System optimization and design consideration ................................... 144

6.8 Conclusions .......................................................................... 148

VII. SUMMARY AND ENGINEERING IMPLICATION .............................. 152
7.1 Summary ............................................................................. 152

7.1.1 Use of a numerical design approach for sequencing batch
reactors ........................................................................... 152

7.1.2 Effect of growth substrate feeding pattern on community

structure and TCE cometabolism .............................................. 154
7.1.3 Effects of TCE on cell viability ......................................... 154
7.1.4 Bench-scale SBR for TCE cometabolism ............................. 155
7.2 Engineering implications ........................................................... 155
7.3 Proposed researches ................................................................ 156

APPENDIX A. Derivation of CSTR steady-state solution used as the Initial

condition for solving SBR modeling equations ......................... 159

APPENDIX B. Curve fitting for TCE degradation kinetics .............................. 162

LIST OF TABLES

Table 2—1 Chemical/physical properties of TCE ............................................... 6
Table 2-2 Water soluble products of TCE degradation by whole cells of

methanotrophs ..................................................................................... 12
Table 2-3 Summary of cometabolizing kinetic expressions .................................. 14
Table 2-4 Reported TCE degradation kinetic patterns and related kinetic coefficients .... 16

Table 2-5 Phenol degradation kinetic patterns and related kinetic coefficients ............. 20
Table 2-6 Growth kinetic for TCE degraders ................................................. 22
Table 3-1 Typical values of the required time for each operating stages in SBRs ......... 38
Table 3-2 Minimum cell age for different group of microbial populations ................. 45

Table 3-3 Assumed rate expressions for each type of solids and substrate

categorized in Figure 3-2 ......................................................................... 50
Table 3-4 Mass balance equations and solutions for SBR in fill period .................... 50
Table 3-5 Mass balance equations and solutions for SBR in react period .................. 51

Table 3-6 Steady-state solutions for solids and substrate in a CSTR with sludge

recycle .............................................................................................. 52
Table 3-7 Summary of input parameters and design outputs for design examples ........ 59
Table 4-1 Curve fitting results for estimating zero-order phenol degradation kinetic
coefficients ......................................................................................... 83
Table 4-2 Results for quantification of filaments and rotifers in reactors (sample

was collected on day 275 and day 290) ........................................................ 90
Table 4-3 Numbers of phenol and TCE degraders in mixed liquid from different

feeding pattern reactors (sampling on day 192) ............................................... 91

Table 4-4 Characteristics of pure culture isolates from different feeding pattern
reactors ............................................................................................. 93
Table 5-1 TCE transformation capacity of different cultures ................................ l l 1

Table 6-1 Biodegradability, Henry's constant and solid partition coefficient of

phenol and TCE ................................................................................... 1 17

Table 6-2 Summary of bench-scale SBR operating parameters ............................. 126

Table B-1 Kinetic models examined for TCE degradation ................................... 162
xii

LIST OF FIGURES

Figure 2—1 Pathways of transformation for TCE by methanotrophic mixed cultures ..... 1 1
Figure 2-2 Aerobic phenol degradation pathways ............................................. 18
Figure 2-3 Typical SBR operating mode and its flexibilities in choosing or combining

static fill, mixed fill, aerated fill, mixed react and aerated react. ............................. 26
Figure 3-1 Flowchart for numerical simulation and design of an SBR system ............ 46
Figure 3-2 Waste characteristics and their conversions in biological processes ........... 48

Figure 3-3 Repeating solution for dynamic cell ages and substrate concentrations in

an SBR system designed for carbon oxidation ................................................ 60
Figure 3-4 Repeating solution for mass of each type of solids present during two

SBR solid storage periods ....................................................................... 61
Figure 3-5 Repeating solution for substrate concentrations and dynamic cell ages in

an SBR designed for simultaneous oxidation of organic matter and ammonia. ............ 62
Km 3-6 Repeating solution for mass of each type of solids accumulated during

two solids storage perindc 63
Figure 3-7 Effects of solids wasting percentage (W) on the repeating solution for

total solid, dynamic cell age, reactor volume and cycle time ................................. 64
Figure 3-8 Effects of reactor number (N,) on the repeating solution for total suspend
solids, dynamic cell age, reactor volume and cycle time 65
Figure 3-9 Effects of solids wasting frequency (UN) on the repeating solution of

total solids, dynamic cell age, reactor volume and cycle time ................................ 66
Figure 4-1 Schematic of experimental setup ................................................... 72

Figure 4-2 Second-order TCE removal coefficients for cells from different feeding

pattern reactors ........................... 78

 

Figure 4—3 TCE degradation observed in batch experiments using cells from

different feeding pattern reactors ................................................................ 80
Figure 4-4 Changes of biomass during reactor operating period, each sample was

taken at the end of growth substrate fill period. ............................................... 81
Figure 4-5 Phenol degradation in reactor P for daily spiking with phenol from day

31 to 38 ............................................................................................. 83
Figure 4-6 Effects of phenol feeding pattern on phenol degradation in batch

 

systems ............................................................................................. 85
Figure 4-7 Effect of feeding pattern on dissolved oxygen concentration. .................. 86
Figure 48 Microphotograph of sludge in reactor SC5. ...................................... 88
Figure 4-9 Microphotograph of sludge in reactor P 88
Figure 4-10 Microphotograph of sludge in reactor SC2 ...................................... 89
Figure 4~11 Microphotograph of sludge in reactor P. ........................................ 89
Figure 4-12 REP-PCR patterns of isolates from reactor P and C. .......................... 95

Figure 4-13 TCE transformation capacity profiles of isolated from different feeding

pattern reactors .................................................................................... 96
Figure 5-1 The minimal steps involved in a suicidal inactivation of an enzyme ........... 104
Figure 5-2 Concentration of viable P. cepacia G4 after exposure to different initial

TCE concentrations. .............................................................................. 108

Figure 5-3 Concentration of different cell types after exposure to 2.5 mg]! of

aqueous TCE in a 24 hours incubation period. ................................................ 109
Figure 6-1 Experimental setup for bench-scale SBR ......................................... 123
Figure 6-2 Operating mode of bench-scale SBR for TCE cometabolism ................... 127

Figure 6-3 Compare the observed aqueous TCE concentration in non inoculated
samples and samples that contained 840 mg/l of autoclaved cells ........................... 131

Figure 6-4 Observed TCE concentration during fill and react period in an SBR with
influent contained 0.5 mg/l TCE (a) liquid phase, (b) gas phase ............................ 132

Figure 6-5 Percentages of TCE removed by biodegradation and air stripping

 

observed in mass balance experiment: 133
Figure 6-6 Biomass variation observed in a bench-scale SBR conducting TCE
transformation. .................................................................................... 133
Figure 6-7 Variation of specific TCE degradation rate during reactor operating

period ............................................................................................... 135
Figure 6-8 Variation of Monod kinetic coefficients Kc and k, for TCE degradation

observed during reactor operation period ...................................................... 137
Figure 6-9 Changes of TCE degradation activity during different stages of SBR
operation ............................................................................................ 138
Figure 6-10 Typical observed and modeled phenol concentrations in an SBR reactor

for cometabolism (data was taken on day 25) 140
Figure 6—11 Observed and predicted TCE concentration during fill and react stage in

an abiotic SBR ..................................................................................... 142
Figure 6-12 Liquid phase TCE concentrations observed and modeled in a bench-scale

SBR for TCE cometabolism. .................................................................... 143
Figure 6-13 Gas phase TCE concentrations observed and modeled in a bench-scale

SBR for TCE cometabolism ..................................................................... 143
Figure 6-14 Effects of K La on simulated liquid and gas phases TCE concentrations

present in SBRs ................................................................................... 145
Figure 6-15 Effect of kc on simulated liquid and gas phases TCE concentrations

present in SBRs ................................................................................... 146
Figure 6-16 Effect of K, on simulated liquid and gas phases TCE concentrations

present in SBRs ................................................................................... 147
Figure A-l Schematic of a typical CSTR activated sludge system .......................... 159

XV

Figure 8-1 Typical curve fitting results for TCE degradation observed in batch
experiments using cells from reactor P ......................................................... 163
Figure B-2 Typical curve fitting results for TCE degradation observed in batch
experiments using cells from reactor C ......................................................... 163
Figure B-3 Typical curve fitting results for TCE degradation observed in batch
experiments using cells from reactor SC2 ...................................................... 164
Figure B-4 Typical curve fitting results for TCE degradation observed in batch

experiments using cells from reactor P ......................................................... 164

xvi

NOMENCLATURE

English Symbols

A dynamic cell age, time'1

b endogenous decay coefficient, time'l

C concentration of nongrowth substrate, mass per volume

Es efficiency of substrate removal, as a percentage

Es” efficiency of TKN oxidation, as a percentage

E efficiency of solid stabilization, as a percentage

f4 biodegradable fraction of active biomass

f" organic nitrogen fraction of biodegradable suspended solid
maximum rate of substrate utilization per unit weight of microorganism,
time'l

kc maximum rate of nongrowth substrate utilization per unit weight of
microorganism, time'1

KLa overall mass transfer coefficient, time'1

k}, first order rate coefficient for hydrolytic conversion of degradable
substrate to soluble substrate, time‘1

K3 half velocity coefficient, equal to substrate concentration when rs=(1/2)k

M Mass

Ma mass of active cells

N, number of reactors

NW number of cycles between wasting events

Q influent flowrate, volume per time

rXa

SN
SRT
S VI

to
’c
1d
t;
It

V0

VsaR

3‘

X1.
Xi
Xrn

ratio of volume of reactor at the beginning of fill to the volume of reactor
at the end of fill.

rate of substrate utilization, mass per volume per time
rate of active biomass production, mass per volume per time
substrate concentration, mass per volume

TKN concentration, mass per volume

mean solid residence time, time

sludge volume index

time

time to fill reactor to the volume V0, time

time to complete full SBR cycle

time for decanting

fill time

idle time

time for the react period

time for the settling period

specific utilization rate, time-l

reactor volume

the volume in an SBR at the beginning of the fill period
the volume in an SBR at the end of the fill period
volume of wasted suspended solids

total suspended solids, mass per volume

active biomass concentration, mass per volume
degradable suspended solid concentration, mass per volume
aerobic heterotrophic biomass concentration

inert biomass concentration, mass per volume

inorganic suspended solid concentration, mass per volume

xviii

XN
Xr
Xv

YN

nitrifying biomass concentration, mass per volume
refractory suspended solid concentration, mass per volume
Volatile suspended solids, mass per volume

growth yield coefficient, mass per mass

growth yield coefficient for nitrifiers, mass per mass

ratio of wasted sludge to total suspended solids (=100VWNO)

Superscripts and subscripts

o superscript denoting reactor influent stream

0 subscript denoting initial condition

e subscript denoting reactor effluent stream

L liquid phase

g gas phase

Greek Symbols

0 hydraulic residence time, time

Or solid residence time, time

7,“ COD to weight ratio for Xd, mass of substrate per mass of biodegradable
solid

7,. COD to weight ratio for Xa, mass of substrate per mass of active biomass

u specific cell growth rate, time'1

um maximum specific cell growth rate, time'l

xix

CHAPTER 1
INTRODUCTION AND RESEARCH OBJECTIVES

1.1 Introduction

Over the past two decades, contamination of groundwater by volatile organic
compounds (V OCs) has been reported with increasing frequency. More than 40% of the
US population uses groundwater as the drinking water supply, often without any treatment
other than disinfection. Groundwater contamination is thus a serious public health concern.
Methods for remediation of contaminated groundwater include carbon adsorption, air
stripping, chemical oxidation, and biological processes. The advantage of chemical and
biological processes is that these methods result in partial or complete destruction of the
contaminant. The advantage of biological process over chemical processes is typically

economic.

Within the wide range of biological processes now available, sequencing batch
reactors (SBRs) have proven to be cost-effective and flexible, and are established
technology for the removal of nutrients, BOD, and hazardous substances. In an SBR
system, reactors pass through a repeating sequence of operations, including fill, react,
settle, and decant. Conditions during the fill and react periods can be manipulated to enable
selection of microbial communities with desirable properties. In principle, VOC removal
can be accomplished in an SBR, but the conventional cycle must be modified to enable
degradation of highly halogenated compounds, such as trichloroethylene (TCE). Such
compounds are "nongrowth substrates" and are degraded by cometabolism. Nongrth

substates cannot support microbial growth and can only be broken down if a growth

substrate capable of inducing and sustaining the desired cometabolic activity is added at
some point during the SBR cycle. The time oriented operation of an SBR is well suited for

addition of growth substrate in a manner that can induce and sustain cometabolic activity.
1.2 Research Objectives

This project was designed to investigate and exploit the operational flexibility of
SBRs for cometabolic transformations. Lack of design criteria and standardized design
methodology have prevented wide spread use of SBR technology (Arora et al., 1985). This
deficiency became especially apparent in attempting to apply SBR technology for
cometabolism. Accordingly, one of the first goals of this work was development of an
SBR design theory that could accommodate kinetic expressions for cometabolism. This
was accomplished using a classification scheme for suspended solids and substrates,
developing mass balance equations for each substrate and solid type, and iteratively solving
the resulting system of differential equations. Control strategies and design procedures
were also explored. A second objective of this work was to explore operating strategies
that might enrich for organisms capable of cometabolism. The effects of different growth
substrate (phenol) feeding patterns and selective effects of repeated exposure to a
nongrowth substrate (trichloroethylene) were evaluated. A final objective was to
experimentally evaluate use of SBR technology for cometabolism. Concepts and experience
acquired in accomplishing the earlier objectives were applied in the design, operation, and
simulation of a bench scale SBR.

1.3 Overview of the chapters

This project was design in four phases. Each phase constitutes one chapter in the
dissertation. SBR simulation and design were the major focus of the first phase. In the
second phase, selection of a cometabolizing population by varying the pattern of growth

substrate addition was investigated. Community structure and isolate studies were

conducted in this phase. The third phase consisted of a brief investigation of TCE toxicity.
Cell viability after TCE transformation was the major concern. The effects of TCE
transformation toxicity on cell selection and recharge of cometabolic activity were
evaluated. In the fourth phase, a bench-scale SBR conducting TCE transformation was
monitored and simulated numerically. The major concern in this phase was the stability of
TCE transformation activity. Removal mechanisms and degradation kinetics were also
evaluated. The dissertation concludes with a summary of the most important results and
recommendations for design and operation of SBRs for cometabolism. Suggested future

studies are also indicated.
Literature cited

Arora, M. L., E. F. Barth, and M. B. Umphres. “Technology evaluation of SBRs.” L
m 57 (8 1985): 867-875.

CHAPTER 2
BACKGROUND

2.1 Selection of a model system for cometabolism

The overall objective of this work is to evaluate the potential use and design of
sequencing batch reactors for cometabolic transformation of hazardous substances. Many
experimental systems can be envisioned for this purpose. The first important design
decision is selection of a matching pair of growth and nongrowth substrates. The growth
substrate must induce nonspecific enzymes that can attack the target nongrowth
substrate(s), it must support a cometabolizing microbial population or community, and it
must supply reducing power for sustained transformation of the nongrowth substrate(s).
From the standpoint of ease of laboratory operation and establishment of mass balances,
substrates that are nonvolatile and are highly soluble in water are advantageous.
However, most growth and nongrowth substrates are volatile and dissolve poorly in
water, so some compromise is necessary. In selection of a model system, aerobic
conditions are also useful because of the relative ease of culture manipulation. Phenol is a
growth substrate that satisfies many of the desired experimental needs - it is essentially
nonvolatile and induces oxygenase activity toward halogenated alkenes under aerobic
conditions. Accordingly, phenol was selected as the growth substrate. Several

halogenated alkenes can be transformed by phenol-induced oxygenase activity. Of these,
the most common groundwater contaminant is trichloroethylene (T CE). Thus, TCE was

selected as the model nongrowth substrate. In the following sections, the properties of

TCE and phenol are described further, as are the microbial communities that degrade

these compounds.
2.2 Properties of trichloroethylene

Trichloroethylene is a synthetic chlorinated organic chemical that fulfills all of the
requirements for the ideal degreasing solvent (Table 2-1). It is only slightly soluble in
water and forms an azeotrope with water, resulting in a mixture with a lower boiling
point and vapor density. It is also highly volatile. When released to the atmosphere, it is

destroyed by photooxidation with a half-life of about one day (Schaumburg, 1990).

TCE is generally inhibitory to microbial growth. Exposure of a growing batch
culture of Escherichia coli to 0.009 mg/L of TCE increased doubling time by 131% and
induced 22 stress proteins (Blom et al., 1992). It is mutagenic to Salmonella typhimurium
and Escherichia coli K-12 when activated by liver microsomes, and it is a suspected

carcinogen in some tested animals (Miller and Guengerich 1982).

The World Health Organization (WHO) recommends a drinking water guidance
level of 30 rig/l based on a carcinogenic end point. In the United States, the current
maximum concentration in drinking water under the Safe Water Drinking Act is 5 ppb
(US EPA, 1987). For effects other than cancer, the US EPA recommends a water quality

criterion of 6.77 mg/L.
2.3 Causes and status of TCE contamination

TCE was widely used from the forties to the seventies as a degreasing agent, dry
cleaning solvent, extractant for food processing, refrigerant, heat-exchange liquid,
fumigant, reactant and solvent for organic synthesis, and general anesthetic in medicine
and dentistry. In 1976, it was identified as a suspected carcinogen, and its uses were

severely curtailed. Nevertheless, for the nearly 30 year period of widespread use, there

A

Table 2-1 Chemical/physical properties of TCE

 

 

Property Value Reference
Chemical formula C12C=CHC1 ( l )
Molecular weight 13 1.40 (1)

Physical state Colorless liquid (1)

Boiling point 86.7 °C (1)

Melting point - 73 °C (1)

Density 1.4 g/mL at 25 °C (1)

Vapor pressure 77 mm Hg at 25 °C (1)

Vapor density 4.5 (air = 1) (2)

Lower explosive limit in air 12.5% (2)

Upper explosive limit in air 90.0% (2)

Autoiginition temperature 410 °C (2)

Henry '5 ”“5“” 0.356 at 25 °c (3)
(dimensionless)

Water solubility 1000 mg/L at 20 °C (1)

log K0,, 2.29 (4)

K: for algae 2.4 mL HZO/mg VS (5)

log ECfl 3.16 umol/L (6)

 

Reference: (1) Weast 1992, (2) Fan 1988, (3) Mackay and Leinonen
1975, (4) Burcell 1978, (5) Smets and Rittmann 1988, (6) Kamlet et al.,
1986.

was little control on TCE release and disposal. The most common disposal method was
landfrlling or direct disposal to land. When applied to unsaturated soil, TCE volatilizes
and simultaneously migrates downward to the water table. Because it is denser than
water, organic phase TCE arriving at the water table continues to sink until further
vertical migration is prevented by a confining clay or rock layer. Trapped or pooled TCE
below the water table serves as a long-term slow-release source of TCE. Under anaerobic
conditions, TCE that dissolved in water can be reductively dechlorinated, yielding
dichloroethylene isomers, vinyl chloride, and ethyléire or ethane. TCE can also be
produced anaerobically, by reductive dechlorination of tetrachloroethylene. Little aerobic
degradation of TCE is observed unless growth substrates are present that induce the
requisite oxygenase activity. These factors help to explain why TCE is such a pervasive
groundwater contaminant. Nationwide surveys of drinking Water supplies revealed that
TCE was present in up to 34% of water supplies tested in'the U.S. (Conglio et al., 1980;
Westrick et al., 1984). The highest levels of TCE were associated with leaching from
landfill waste disposal sites, where it was detected at 28% of waste sites examined

(Josephson, 1986).

As described above, TCE has had many applications. As a result, it has been
discharged to surface waters and groundwater by industry, commerce, and individual
consumers. About 90% of the TCE produced in 1974 was used in industrial degreasing.
The US EPA estimated that approximately 310,200 tons of wasted solvent were produced
by degreasing operations in 1974. Lesser amounts were used and discarded by dry
cleaners, septic tank cleaners, and other cleaning operations. By 1980, TCE solvent waste
production had decreased to 234,000 tons (Vogel et al., 1987), and, by 1985, industrial
TCE usage had dropped to 90,000 ton/yr. Until recently, many common consumer
products contained TCE. Some of these entered the environment by way of the septic
tank, sewer, or municipal landfills. Among these were pipe and drain cleaners, shoe

polish, spot removers, upholstery cleaners, paint remover, and septic tank cleaners.

2.4 Biodegradation of TCE

Until the early 19805, TCE was considered non-biodegradable under aerobic
condition and slowly transformable under anaerobic conditions (Bouwer et al., 1981).
This perception changed in 1985, when Wilson and Wilson (1985) observed aerobic
degradation of TCE in a soil column exposed to natural gas. It is now known that TCE
will degrade either anaerobically or aerobically, if an appropriate growth substrate is
present. To date, anaerobic processes have found limited acceptance for bioremediation
applications, in large part because of concerns that partial dechlorination products, such
as vinyl chloride, would accumulate. Vinyl chloride is a known human carcinogen. The

products of aerobic TCE biotransformation are believed to be less harmful.

Early observations of aerobic TCE degradation eventually established that TCE
was cometabolized by methanotrophic microorganisms. The nonspecific enzyme methane
monooxygenase (MMO) was responsible for fortuitous oxidation of the TCE. Similar
TCE-degrading monooxygenase systems were discovered in the ammonia-oxidizing
bacteria (Arciero et a1, 1989; Vanilli et al., 1990) and the propane-oxidizing bacteria
(Wacket et al., 1989). However, TCE cometabolism was not found in all microbial
populations possessing broad-specificity microbial oxygenase activity. Bacteria that
contained nitropropane dioxygenase, cyclohexane monooxygenase, cytochrome P-450
monooxygenase, 4-methoxgbezoate monooxygenase and hexane monooxygenase did not

degrade TCE (Wacket et al., 1989).

Another group of microorganism that exhibited TCE cometabolism activity were
the bacteria that degraded aromatic compounds. Some Pseudomonas species produce
aromatic oxygenases that can degrade TCE (Nelson et al., 1987; Wackett and Gibson
1988). Phenol and toluene are the typical inducing agents for this activity.

Isopropylbenzene, isoprene and propylene utilizing bacteria also exhibit TCE

A

cometabolism activity. The active species include Rhodococcus,Alcaligenes, and

Xanthobacter (Dabrock et al., 1992; Ensign et al., 1992; Ewers et al., 1990).

Under aerobic condition, TCE may be oxidized according to the following
equation:

CHC1=CC12 + H20 + 1.502 —> 2C02 + 3HC1 (2-1)
Reaction 2—1 is highly favorable (AG°'=-1065 KJ/mole) but reducing equivalents and
ATP are not made available to the organism by the oxidation (Dabrock et al., 1992). In

fact, the oxygenase requires reducing power. Therefore, it is not surprising that aerobic

growth on TCE has yet to be reported.

2.5 Aerobic TCE degradation pathway

The pathway for aerobic degradation of TCE has been investigated in detail for
mammalian cytochrome P450 and for soluble methane monooxygenase (sMMO). It is
expected that ammonia monooxygenase (AMO) and aromatic oxygenase (AO) systems

function in much the same way, with similar degradation products.

Oxidation of TCE by sMMO was investigated in detail as a model for

 

understanding _, g ..... ‘ I ‘ ‘ chloroolefm dechlorination (Fox et al., 1990).
Little et al. (1988) concluded that TCE degradation is a cometabolic process that provides
little or no benefit to the organisms. Methanotrophic bacterium strain 46-1 initiated the
degradation of TCE, but was unable to metabolize some intermediates like dichloroacetic
acid and glyoxylic acid. The use of purified enzyme components, obtained from
Methylosinus trichosporium OB3b, allowed an unambiguous determination of the
reaction products. sMMO oxidized haloalkenes largely to epoxide intermediates. This

was demonstrated by the formation of diagnostic adduct upon reaction with 4-(p-

nitrobenzy)pyridine. Furthermore, the predicted stable decomposition products of

A

10

epoxides were identified in the enzyme reaction mixtures. As shown in Figure 2-1, the
predominant fate of the epoxide appears to be hydration followed by carbon—carbon bond
scission. These spontaneous reactions led to the formation of the major products carbon
monooxide and formic acid. Methanotrophs can oxidize carbon monooxide and formic
acid further to carbon dioxide (Colby et al., 1977). Under moderately acidic conditions,
TCE epoxide hydrolyzes to dichloroacetic acid and glyoxylic acid (Little et al., 1988).
Chloral has also been observed as a minor (6%) product. Chloral is derived from TCE via
an intramolecular chlorine migration reaction (Miller and Guengerich, 1982). Further
oxidation of chloral produces trichloroacetate and reduction yields 2,2,2-trichloroethanol.
Products analysis of reaction mixtures containing synthetic TCE-epoxide exhibited no
detectable chloral formation, indicating that chloral was likely generated at the sMMO
active site. Chloral hydrate has been shown to be a product for TCE cometabolism in four
different methanotrophs expressing sMMO. Chloral hydrate is biologically transformed
to trichloroethanol and trichloroacetic acid. Mixed cultures containing methanotrophs
were capable of more extensive mineralization of TCE to carbon dioxide than a pure

culture (Little et al., 1988).

In aromatic oxygenase systems, formate, carbon monoxide, glyoxylic acid were
detected in pure cultures (Wacket and Householder 1989; Winter et al., 1989). For P.
cepacia G4, TCE completely degraded to C02, Cl- and unidentified, nonvolatile products
(Nelson et al., 1986). A summary of the reported water soluble products of aerobic TCE

biotransformation is provided in Table 2-2.

A turnover- and time-dependent inactivation of the sMMO protein components
occurred during oxidation of TCE in vitro (Fox et al., 1990). Inactivation was caused by a
diffusible reaction product as demonstrated by covalent modification of sMMO with

1,2,14C-TCE. In-vivo experiments with toluene diuxweuwe ‘ ' ' 9 strain of

L

 

V
ginger
0

C13C-C\H

/\

ClaC—CH2_0H Clac-c’OH

 

—_—>

H ,0
)Z=c\ TgE
c1 c1

NADH +H++1l202

 

oxygenase
NADH-Hzo
TCE emxide'
H‘C’EE'C
cf I21
mid.
OH OH
Cl-C—C- H ClsCH -C:Cl HCOOH CO
61 61
* O
0C—‘ C4
C1 CH
HC’ Ch 2 _C"OH

0\
0.2M"

 

 

W
C02 + Cells ‘—

Figure 2-1 Pathways of transformation for TCE by methanotrophic mixed culture.
*indicates possible or know carcinogenic metabolites

12

Pseudomonas putida indicated that toxic products can profoundly disrupt metabolism and

diminish observed rates of TCE degradation (W ackett and Householder, 1989).

Table 2-2 Water soluble products of TCE degradation by whole cells of

 

 

 

 

 

methanotrophs
Organism Water soluble products Reference
M. strain 46-1 glyoxylic acid, dichloroacetic acid Little et al., 1988
. Methylpsinus chloral, 1,1,l-trichloroethanol Oldenhuis et al., 1989
mchosponum OB3b trichloroacetic acid
. Methylpsinus chloral, 1,1,l-trichloroethanol Newman 1991
mchosporrum OB3b trichloroacetic acid
1 , l , 1 -trichloroethanol
Methylocystrs sp. stain M glyoxylic acid, dichloroacetic acid Uchryama 1992
trichloroacetic acid

 

 

 

 

 

2.6 Microbial degradation kinetics of TCE

One of the most widely accepted kinetic expressions for cell growth is the Monod
equation (Monod 1949):

_fl_mS
K,+s

p (2-2)

where, S = the concentration of substrate (mg/L), u = the specific growth rate (d'l), fl... =
the maximum specific growth rate constant (d‘l), and K, is the half-saturation constant
(mg/L). Since TCE does not support growth, equation 2-2 can not be used to describe

changes in microorganism concentration during TCE transformation. However, saturation

i

l3

kinetics using a Michaelis-Menton or Monod-like expression have been used successfully

to describe TCE degradation by resting cells:

=_’££_

q, - Kc + C (2-3)
where, C = the concentration of nongrowth substrate (mg/L), q, = the specific rate of
nongrowth substrate degradation (mg nongrowth substrate/mg cell-d), kc = the maximum
specific non-growth substrate degradation rate (mg nongrowth substrate/mg cell-d) and

Kc = the half-saturation constant (mg/L).

Several reports also describe TCE degradation kinetics that are first order with
respect to TCE concentration (Dabrock et al., 1992; Henry and Grbic'-Galic' 1990;
Oldenhuis et al., 1989; Strand et al., 1991; Wackett and Gibson 1988):

1C- : -k;CX (24)
dt

where, k; = the second order rate coefficient (L/mg-d).

Equations 2-3 and 2-4 fail to capture certain features of cometabolism, notably the
loss of transformation activity in resting cells caused by a depletion of reducing power (in
the absence of growth substrate) and product toxicity. In recent years, a set of related
models have emerged that can account for such effects. A summary of these models is
provided in Table 2-3. Also shown are simplifications of the Monod-like model for C <<
K c and C >> Kc. In these models, an important concept is the idea of a limited
transformation capacity (Tc) for the nongrowth substrate (Alvarez-Cohen and McCarty
1991a). T, is defined as:

_ g _ mass of contaminant transformed (2_5)
‘ dX mass of cells inactivated

 

To evaluate the loss of cell activity during the cometabolism, first order decay of biomass

is commonly used. The rate of cell decay can be expressed as

Table 2-3 Summary of cometabolizing kinetic expressions

 

 

 

 

 

 

 

 

 

 

 

 

 

—-
Differential equations for substrate
Model uti tion rate Integgted form Ref.
-:—f=1::CXC and???” C kX
+ _
la Kcln(E-]+C-Co=—-Lb ”(I-fl”) (1)42)
0 __d£= kECX e-bt 0
dt K + C
u, 129-k CX and%=—bx [x
dc b (fie-1)] (2). (3)
C<< K so —7 =k,’CX¢' ' C = Cue
1c —£=chand £=—bX k
d’ dC ‘1‘ Same as model 3 where Tc = 72L (4)
C>> K :0 -7 =1: CX,Ce—bt
_d_C_ kECX mi? _Tc 1
d: K + C 1 K I CX
:=— ___ hi +r ["14“
2, 1 kc C, /Tc — X, FC 5
k X,-F(Ca-C) ()
so -£=—‘— whereF=X0-—(C,-C)
dt K, + C Tc
dC . dC
2b -; = chX and a = Tc F'C-U':
dC l C=Co C kF whereF= -Xo -CH/T (4)
C<<K ' x- —°e ' '
-—=k x -— C -C
C so dt C4 0 Tc( 0 )J 0 Tce
dC dC
2C -5. = ch and E = Tc k
Same as model 3 where T = J“ (4)
C>>Kc 30—%=kc[xo_?:(co_c)) C b
_d_X_ dC
= —bX and — = T
3 d! dX C = C, — T,X,(1 - e‘b') (a)
so —%9 = chX,e'b’
_d_C_ k CX
(T: K_:— + C _
4 N0 integrated form (4)
51‘ =-bX—£‘TX C
d: T. K, + C

 

 

 

 

Reference: (1) Galli and McCarty, 1989, (2) Schmidt, 1985, (3) Criddle et al., 1990, (4) Criddle, 1993, (5)
Alvarez-Cohen and McCarty, 1991a, (6) Saez and Rittmann, 1991.

15

dr

where X = cell concentration and b = decay coefficient.

Models found in the literature prior to 1993 are obtained from various
combinations of equations 2-4, 2-5 and 2-6. Criddle (1993) proposed a fourth model that
unified the earlier models. Model 4 incorporated endogenous decay and product toxicity
into the cell decay term and assumed that non growth substrate degraded with saturation

kinetics.

A summary of reported kinetic models and respected rate coefficients for
microbial TCE transformation is provided in Table 24. Since not all TCE cometabolism
studies are designed to evaluate TCE degradation kinetics, some of the data presented in

Table 2-4 is provided as initial rates for a specified concentration of TCE.

Methanotrophic organisms exhibit substantial variation in TCE oxidation rate due
to differences in species and the type of MO expressed. Methanotrophs possess two
types of MMO: the membrane-associated or particulate form (pMMO) oxidizes TCE
slowly and is induced at high copper concentrations; the cytoplasmic or soluble MMO
(sMMO) oxidizes TCE at high rates and is induced at low copper concentrations (Dalton
et al., 1984; Stanley et al., 1983). The copper concentration for transition from sMMO to
pMMO was around 0.25 umol copper/g biomass (Anderson and McCarty, 1994; Tsien et
al., 1989). Differences in culture media composition also cause differences in the rate of
TCE oxidation (Bowman and Sayler, 1994; Henry and Grbic-Galic, 1990). A current
"ranking" of TCE degraders by rate is :

M. trichosporium OB3b > Methanotrophic mixed culture > P. cepacia G4 >
P. putida F1 ~ M. vaccae JOBS ~ N. europera

16

Table 2-4 Reported TCE degradation kinetic coefficients

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Kinetic Rate constant
Culture pattern It, (d‘l) Kc (mg/L) reference
Weadomonas cepacia G4 0.74 0.40 Folsom 1990
Methylosingthgighosporium 42 18 Brusseau 1990
51333 gogflyé‘ggfi)” Monod 64 4.83 Fox 1990
Me‘hflosmg‘ggfhospo'm’" 19.1 +/. 8.0 55 +/- 18 loldcnhuis 1991
Methylomonas sp MMZ 0.046 - 0.29 0.51-1.35 Henry 1990
kc'ang-d) TCE (mg/L)
Pseudomonas putida F1 0.0162 1.05 - 10.5 Wackett 1988
m’hflosmgsggfho-‘PO'W 3.08 0.026 - 13.0 Oldenhuis 1989
Methylomonas sp MMZ first order 0.003 - 2.3 Henry 1990
Methanotrophic mixed culture 0.007 0 - 3 Strand 1990
Pseudomonas sp JRl 0.42 3 -26 Dabrock 1992
Rhodococcus. erythroplis BDl 0.42 3-26 Dabrock 1992
rate (d‘l) TCE (mg/L)
Methanotrophic mixed culture 0.0032 0.75 Fogel 1986
Mycobacterium vaccae JOBS 0.113 2.62 Wackett 1989
Pseudomonas putida F 1 0.112 10.5 Wackett 1988
Nitrosomonas europea 0.027 - 0.064 1.0 Vannelli 1990
Alcaligenes eutrophus JMP 134 Initial rates 0.019 3.3 Harker 1990
Nitrosomonas europaea (mg TCE! 0.208 1.31 Arciero 1989
Thodococcus erythroplis If mg cell-d) 0.0141 03 Ew e rs 1990
Alcaligem “WWW 0.0170 0.8 Ewers 1990
Xanthobacter strain Py2 0.81 15 Ensign 1992
T. k, (d'l) W)
Methanotrophic mixed culture 0.036 0.54 0.42 Alvafgzgfm“
Methanotrophic mixed culture 0.042 0.84 1.5 “va‘fgzgfww
Nitrosomonas europaea 0.004 Hyman 1993—
ote: Biomass assumptions used to standardize rate datazmr mg cell, Mg

protein/mg cell, 3.3 mg wet weight/mg cell.

17

Although methanotrophic bacteria generally exhibit higher TCE degradation rates, they

grow slowly and TCE transformation rates are sensitive to copper concentrations.
2.7 Biodegradation of phenol

Microorganisms with phenol-degrading capacity include bacteria, such as
Pseudomonas (Beltrame et al., 1980; Yang and Humphrey, 1975), Nocardia (Rizzuti
and Augueliaro, 1982), and Bacillus (Buswell 1975); yeasts, such as Trichosporon (Gaal
and Neujahr, 1979), Candida (Krug et a1. 1984); and multicellular fungi, such as
Fusarium (Anselmo et al., 1985). ‘

Phenol is degraded through the intermediate catechol (Bayly and Barbour, 1984),
but opening of the aromatic ring can proceed by either ortho or meta ring cleavage
(Figure 2-2). The ortho pathway uses 1,2-oxygenase to open the catechol ring,
generating succinate and acetyl-CoA via B-ketoadipate (Feist and Hegeman 1969). The
1.2 monooxygenase and the two subsequent enzymes of the pathway are induced by
cis,cis-muconate in P. putida and P. aeruginosa. The meta pathway uses a 2,3—oxygenase
and is less specific. In contrast to the ortho pathway, all of the enzymes of the meta-
cleavage pathway are induced by phenol, including phenol hydroxylase (Bayly and
Barbour, 1984; Bayly et al., 1977; Feist and Hegeman, 1969). Induction of the meta
fission pathway is required for the degradation of TCE by P. cepacia G4 (Nelson et al.,
1987). Other TCE degraders such as P. putida PpFl and P. putida B5, also exhibit meta
cleavage when using toluene as growth substrate. However, not all organism using meta

pathway for ring cleavage are able to degrade TCE (Nelson et al., 1988)

Evidence indicates that meta pathway may be plasmid encoded (Bayly and
Barbour, 1984). The initial conversion of phenol to catechol is catalyzed by phenol

hydroxylase, a NADH-dependent monooxygenase. As such, this is an energy-requiring

18

 

 

 

 

phenol
NADH + 0, +H'
H,O + NAD’
OH
W cute: 01 W
1 ion
/
C00: <\— ‘Coon
.\. C00 CH0
cups muconate 2-hydmymueonic semialdehyde
H,O + NADA
COOH 11,0
mueonolactene HCOOH A \ COOH
4-oxaloerotcnale (end (can)
I coon 1 j
-0
O
B-ketoadipue enol lactone < <
COOH

rho-j

0 COOH
COOH
B-ketoedipate

HSCoA AD’
NADH + H‘

o 0
HO OSCoA
0 b-keloadipyl GM

“'00.:

HONOH

O 8060M“

H(())>OSC0A

aeetleoA

4-oxalocmtonate (keto form)

CO, 0
\ coon
CH

2-oxopent-4-enoete

CH3-< COOH

$11130”- ~2-oaovalerate

Figure 2-2 Aerobic phenol degradation pathways (modified from Allosop 1991)

19

step consuming the equivalent of three moles of ATP. Degradation of catechol by the

ortho and meta pathways regenerates the NADH needed for the first step.

The most commonly used equation used to describe microbial growth on phenol
is the Andrews or Haldane equation, representing substrate-inhibited growth (D'Adamo et
al., 1984; Kotturiet al., 1991; Pawlowsky and Howell, 1973; SoKol, 1988; Szetela and
Winnicki, 1981):

u... -S

2
KS+S+£—
K

 

p = (2-6)

1
where K, = inhibition constant (mg/L). Some researchers have concluded that phenol is
not inhibitory and have modeled its degradation with Monod kinetics (Auteinrieth et al.,
1991; Beltrame et al., 1980). Zero order kinetics with respect to phenol concentration
have also been reported (Rizzuti and Augueliaro, 1982; Tischler and Eckenfelder, 1969).
Reported kinetic expressions and kinetic parameters for aerobic phenol degradation are
summarized in Table 2-5. Arguments on the inhibitory properties of phenol and its
‘ degradation kinetics might be explained by the work of Templeton and Grady (1989).
They noted that cells physiologically adapt to their previous growth environment
(Templeton and Grady 1989). As the dilution rate at which the cells are grown in a
chemostat is increased they contain higher levels of RNA and enzymes. Consequently,
when those cells are removed from chemostat and placed into batch reactors in which the
substrate limitation is removed, the cells with the higher RNA and enzyme levels shift to
higher specific degradation rates. A decrease in K, with increasing dilution rate was also
reported. They suggested that K, may depend on cell geometry and physical
characteristics, and that it is influenced by the rate at which the bacteria are growing.
Autenrieth et al. (1991) reported zero order degradation for starved bacteria at low phenol
concentration (100 mg/L), Monod kinetics for starved bacteria at high phenol

concentration (650 mg/L, and Haldane kinetics for non-starving cells at high initial

l8

 

 

phenol
NADH + 0, +H’
11,0 «1» NAD'
OH
W care: 01 W
1 ion
/
COOS <\— ‘Coon
\ COO CHO

cis,eis automate

oCOOH 11,0
\0 -o \

mueonolactene

moon 4

COOH
l o

B-ketoadipate enol lactone

 

2-hydroxymuconie semialdehyde
H,0 + NADA
NADH + IT OH
<\ COOH
COOH
4-oxaloerotonate (enol form)

110

>- COOH

 

Ito-j

0 COOH
COOH
B'kcwflfipflc

HSCoA AD’
NADH + H’

HO 0 0

OSCoA
0 b-ketoadipyl CoA

"100‘

HONOR

O IIICClMIC

“8% OSCoA

aeetleoA

NADH + H NAD C02 COASH H3 C

4-oxalocrotonate coatoeto form)

23;...

2011051114an

C334 COOH

4-htgoxy- -2-oxovalerate

H3 C 1" H
agetaldehbde

Y‘on

Figure 2-2 Aerobic phenol degradation pathways (modified from Allosop 1991)

19

step consuming the equivalent of three moles of ATP. Degradation of catechol by the

ortho and meta pathways regenerates the NADH needed for the first step.

The most commonly used equation used to describe microbial growth on phenol
is the Andrews or Haldane equation, representing substrate-inhibited growth (D'Adamo et
al., 1984; Kotturiet al., 1991; Pawlowsky and Howell, 1973; SoKol, 1988; Szetela and

Winnicki, 1931):
-S
u= ”"' , <2-6)

KJ+S+S—
K

 

l

where K, = inhibition constant (mg/L). Some researchers have concluded that phenol is
not inhibitory and have modeled its degradation with Monod kinetics (Auteinrieth et al.,
1991; Beltrame et al., 1980). Zero order kinetics with respect to phenol concentration
have also been reported (Rizzuti and Augueliaro, 1982; Tischler and Eckenfelder, 1969).
Reported kinetic expressions and kinetic parameters for aerobic phenol degradation are
summarized in Table 2-5. Arguments on the inhibitory properties of phenol and its
degradation kinetics might be explained by the work of Templeton and Grady (1989).
They noted that cells physiologically adapt to their previous growth environment
(Templeton and Grady 1989). As the dilution rate at which the cells are grown in a
chemostat is increased they contain higher levels of RNA and enzymes. Consequently,
when those cells are removed from chemostat and placed into batch reactors in which the
substrate limitation is removed, the cells with the higher RNA and enzyme levels shift to
higher specific degradation rates. A decrease in K, with increasing dilution rate was also
reported. They suggested that K, may depend on cell geometry and physical
characteristics, and that it is influenced by the rate at which the bacteria are growing.
Autenrieth et al. (1991) reported zero order degradation for starved bacteria at low phenol
concentration (100 mg/L), Monod kinetics for starved bacteria at high phenol

concentration (650 mg/L, and Haldane kinetics for non-starving cells at high initial

20

63236 .o: u <22.
.m=oo 3.3.8.33 we. u > ecu E5 *

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$2 .232 Re @382
S2 535. 8... 28.834 .4
$2 628:. :83 Se sacred 23:6 852
§15e§=< Ed , 23.8 852
> 905.5358 382.86 ed
82 .Baeum 98 new 32 new : 2. 23.8 Be:
§_.5Eue< <2 been 32 :83 Rd 882 83.8 Be:
. . . . :25
«5 36.32%; as c 48 a n mm a swig km?
E: gagged 3.3 2.. 3a 83 fiafifimfl
:2 £25 <2 an «.2 £3 2336 852
52 49.8 <2 8.» 8.72. 32-2... 2.3.“: 633,4 .8.
32 £33 3: Km 3n 2 3 a 823m .4
82 6:53 3-3 3.8 we. 8.: 23.8 852
$2 acacia :<z 8:4: 83 8032... 23:6 Be:
8:263. 4» 3mm}. Amman u. 2.31: a; 23.5

 

 

 

BEEN—am new 533.. cone—canoe .985— n-~ 03¢...

 

21

substrate concentration (600 mg/L). These different kinetic responses from the same
organisms suggest that the appropriate kinetic model for phenol degradation is variable,
and may depend upon the physiological state of the phenol-degrading culture. The
different kinetic patterns reported in literatures may be also caused by different values for
kinetic coefficients and by the phenol concentrations applied in kinetic experiments.
When 52< <K, Haldane equation can be simplified to the Monod equation. When SZ<< K,
and S >>K,, Haldane equation can be further simplified to the zero order (with respect to

phenol concentration) form.

Phenol utilizing bacteria grow faster than methanotrophs by one or two orders of
magnitude (Table 2-6). The faster growth kinetics and relative ease of addition of phenol
give phenol-degrading organisms certain practical advantages in reactor systems. High
growth rates enable more rapid start-up, smaller reactor volumes, and more rapid

recovery from TCE product toxicity and enzyme inactivation during TCE transformation.
2.8 Cometabolism in engineered systems

As defined by Dalton and Stiring (1982), cometabolism is the "transformation of a
nongrowth substrate in the obligate presence of a growth substrate or another
transformable compound" (Dalton and Stirling 1982). In a more general sense, however,
cometabolism can be defined as any "transformation of a nongrowth substrate that
depends upon the concurrent or previous metabolism of a growth or energy substrate"
(Criddle, 1993). This definition includes the case of resting cells. Nitrifiers,
methanotrophs, and some pseudomonads can degrade TCE in the absence of growth
substrate. Although rates of cometabolism decline in the absence of a growth substrate,
the presence of growth substrate may be inhibitory as a result of competitive inhibition

between the growth and nongrowth substrates.

22

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.2 «as 8.9 3.8
32 €55 3c -3 3.3 ea 2.32. €955 8:88.52
2805
an
32 .525 a. .a $8 83.2 abuses»: .22 m
32 .525 ”.2 mm... 232 fignuugfiesaumusn
. 8a
32 .623 new 2.4. 9.3.3 eewumfiawfl
oak
32 .325 on... 2 .e 6829 nuances” _
c3— . e0 3855
.883"? me a c me 3:23 neceEeVauuR
32 . . ammo gremaefitc
$2520 :4 _ hm w 0552: 35.83:.»3
32
€662 “no 6558. £36559:
e5 5:50 23.8 .55.:
18:92
com. . . . . . apnea—«59:
o3. . . . . . amp—8:550:—
.E_2_8m N. c N c 2 e on c cm a 0559: 23.3 @058
as. mar—.8 BER—am
Thaw ARE 4‘»:— égmeam m8 Ba 53on an. 88.3%
00:20qu 538» ME 8.533. can :3 an. Egg
a J— .M .85 :8 me 538» an. > 5380
I: x

 

 

enema m2. .8 8:25. 5365 3 use...

 

 

 

 

 

23

To design cometabolic reactor systems, organisms capable of conducting
cometabolic transformations must be identified and cultured. Medium conditions and
toxicity related to the target compounds and their transformation products should also be
evaluated. Substrate competition should be minimized. Growth and nongrowth substrates
may be competing for the same active site. If the growth substrate is supplied at a high
concentration or has high affinity for the enzyme, transformation of nongrowth substrate
may be inhibited or prevented altogether. To reduce the effects of competitive inhibition,
target compounds and growth substrates may be supplied in different time periods, at
different locations, or at concentrations that allow for simultaneous cell growth and

cometabolism.

Product toxicity may affect the long-term stability of cometabolism. Before
adaptation to a nongrowth substrate can occur, loss of microorganisms and diminished
transformation can be expected. Transformation capability could be lost entirely. For
TCE, toxicity is believed to be due to interaction of TCE reaction byproducts with cell
macromolecules (Alvarez-Cohen and McCarty 1991b; Henry and Grbic-Galic 1991).
Cells that do not transform TCE do not suffer toxic effects when exposed to TCE itself.
As a result, populations with the capability for TCE degradation might be expected to
lose that capability as the TCE—transforming cells within the population die. Pure cultures
are likely to be especially susceptible because byproducts can accumulate (Little et al.
1988; Uchiyama et a1 1992). In a well-balanced consortium, the products of
cometabolism may be further degraded by consortium members (Uchiyama et al. 1992).
More extensive biodegradation and reduced toxicity are possible in mixed culture
systems. On the other hand, the selective pressure imposed by TCE toxicity may enhance

the competitive advantages of non-TCE degraders.

For mixed culture systems, the survival of cometabolic species and the balance

between different populations are important. Cometabolic species may be unable to

24

compete with non-cometabolizing species due to product toxicity. The effects of product
toxicity are most severe when reactive intermediates exert their toxic effect by reacting
with cell macromolecules. Cometabolizing cells can be damaged, providing an advantage
to organisms that do not transform TCE. However, if product toxicity is caused by other
TCE transformation products, like trichloroacetic acid (TCA) or dichloroacetic acid
(DCA), toxic effects may be more nonspecific. Populations capable of degrading the
TCA or DCA may be favored.

2.9 Choice of microbial systems for cometabolism of TCE

Many factors can influence the choice of a system for cometabolism of TCE.
Ideally, the cometabolizing microorganisms should be easily grown, using growth
substrate efficiently. The transformation should be fast and stable. Growth substrate
should be cheap, easy to handle and non-toxic or hazardous. .End products from
transformation of growth and non growth substrates should be non-hazardous or easily
handled for further treatment. None of the major TCE-cometabolizing microbial groups
(nitrifiers, methanotophs, and aromatic degraders) satisfies all of these requirements. A
comparative summary is provided in Table 2-6. Because nitrifiers are autotrophs, they
have low yields and grow slowly. They are also sensitive to environmental conditions,
with a narrow pH optimum (7.5-8.6) and a need for relatively high levels of dissolved
oxygen. They are also susceptible to a variety of inhibitors. Their oxidation products,
nitrite or nitrate, are potential common groundwater contaminants, and may require
further treatment. As to methanotrophs, the growth substrate, methane, is flammable and
explosive in high concentration. Methane also has low solubility in the aqueous phase (24
mg/L at 20°C, Weast, 1992) complicating its delivery. Costs involved in methane mass
transfer limitations may hinder its usage as a growth substrate. The common aromatic
growth substrates for cometabolism are phenol and toluene. Application of toluene is

limit because of its volatility and toxicity. Phenol is less toxic and has high water

25

solubility. Although phenol is also a regulated compound, it is readily biodegradable. It is
often found in leachant of hazardous waste landfills and in the industrial wastes of the
coke, petroleum and the chemical industries. Where wastes contain both phenol and TCE,
or where phenol is available from other waste stream, phenol is likely to be the favored
candidate for growth substrate. In other situations, phenol may still be the favored choice
because it is easy to handle and does not have the mass transformation limitations of

toluene, methane or other gaseous substrates.
2.10 Choice of reactor systems for cometabolism of TCE

Reactor configurations that have been evaluated for cometabolism of TCE include
conventional completely-stirred tank reactors (Folsom and Chapman 1991; Coyle et al.,
1993), fixed beds (Arvin 1991; Bilbo et al., 1992; Strand et al., 1991; Strandberg et al.,
1989) expended beds (Phelp et al., 1990) and multi-stage systems. Muti-stage systems are
designed to deal with toxicity and substrate competition (Alvarez-Cohen and McCarty
1991b; Mcfarland et al., 1992; Speital and Leonord 1992).

Sequencing batch reactors (SBRs) are a multi-stage reactor system that seems
well suited for the conduct of cometabolic reactions. Conventional SBR systems consist
of one or more tanks, each capable of substrate biodegradation and solids separation.
Each tank processes influent wastes through a success of five cyclic operating periods:
fill, react, settle, draw and idle (see Figure 2-3). Completion of the five periods
constitutes a cycle. During the fill period, the liquid volume inside the tank increases
from a set minimum volume in response to a predetermined maximum volume. Mixing or
aeration or both can be carried out during all or part of the fill period to provide distinct,
selective growth conditions. During the react period, flow is discontinued either by
cessation of wastewater generation or by diversion of flow to another tank. In aerobic
SBRs, aeration is provided to complete substrate removal. During the settling period,

energy inputs to the tank are stopped and the suspended solids are allowed to flocculate

26

Mixing

Influent

4'.—

 

 

 

 

Influent

Growth

substrate

~D~I§O~l~l
I’.’-fif{f.

I
{50‘05

 

\
w
m

Static Fill

I

 

 

 

 

__ Wasted sludge

  

Aerated Fill

Idle/Recharge

l

 

t
n
c
u
m
E

 

 

 

 

 

MiXing/ Mixed React

 

 

 

 

 

 

Aerated React

Settle

Figure 2-3 Typical SBR operating mode and its flexibilities in choosing or combining

static fill, mixed fill, aerated fill, mixed react and aerated react. Proposed modification for

cometabolic transformation was using idle period as the recharge stage that growth

substrate was supplied to regenerate the cometabolic activity of cells.

27

and settle under quiescent conditions. The treated supernatant is then decanted to the
minimum liquid volume level. Solids may be wasted from the tank as required to
maintain the biomass at manageable levels. The tank remains as such either until
wastewater production resumes (for a single tank system) or until other tanks making up
the system are full and flow is diverted back to the first tank, initiating another cycle. An
SBR accomplishes in time what traditional activated sludge systems accomplish in space
in a series of continuous-flow reactors. The potential advantages of SBRs for hazardous
waste treatment include
0Ability of processes to handle periodic flows;
°Possibility of tanks on- and off-lines to meet either short-term or seasonal
variations;
-Ability to periodically change environmental conditions, selecting or enriching
specific microbial populations; and
OAbility to better ensure biomass retention, as supernatant withdrawal occurs in

nearly ideal quiescent conditions.

Conventional SBR operation can be modified for cometabolism by addition of a
recharge stage after the decant period (see Figure 2-3). The recharge period rejuvenates
cometabolism activity lost by toxicity or decay. During recharge, growth substrate is
supplied to enrich or restore the cometabolizing populations. By incorporating a recharge

stage, the SBR may be capable of stable and continuous cometabolism.
Literatures cited

Allsop, P. J. , M. Moo-Young , and G. R. Sullivan. “The dynamics and control of

substrate inhibition in activated sludge.” 23;; gm; Bfll‘gm in Enrzitrznmenta] 9mm]
20 (1990): 115.

28

Alvarez-Cohen, L. M. and P. L. McCarty. “A cometabolic biotransformation model for
halogenated aliphatic compounds exhibiting product toxicity,.” W 25
(1991a): 1381-1387.

Alvarez-Cohen, L. M. and P. L. McCarty. “Product toxicity and cometabolic competitive

inhibition of chloroform and trichloroethylene transformation by methanotrophic resting

cells.” MW 57 (1991b): 1031—1037.

Alvarez-Cohen, L. M. and P. L. McCarty. “Two-stage dispersed-growth treatment of
halogenated aliphatic compounds by cometabolism.” WM 25 (1991c):
1387-1393.

Anderson, J. E. and P. L. McCarty. “Model for treatment of trichloroethylene by

methanotrophic biofilms.” [MW 120 (2 1994): 379-400.

Anselmo, A. M., M. Mateus, J. M. S. Cabral, and J. M. Novais. “Degradation of phenol
by immobilized cells of Fusarium flocciferum.” W 7 (12 1985): 889-
894. '

Arciero, D., T. Vannelli, M. Logan, and A. B. Hooper. “Degradation of trichloroethylene
by the ammonia-oxidizing bacterium Nitrosomonas europaea.” MW
Comm 159 (1989): 640-643.

Arvin, E. “Biodegradation kinetics of chlorinated aliphatic hydrocarbon in an aerobic
fixed biofilm reactor.” m 25 (7 1991): 873-881.

Auteinrieth, R. L., J. S. Bonner, A. Akgerrnan, and M. Okagun. “Biodegradation of
phenolic wastes.” W5 28 (1991): 29-53.

29

Bayly, R. C. and M. G. Barbour. “The degradation of aromatic compounds by the meta

and genntistatc pathways - biochemistry and regulation.” In Microbial d: 1 .
M. ed. D. T. Gibson. New York: Marcel Dekker, 1984.

Bayly, R. C., G. J. Wigmore, and D. I. McKenzie. “Regulation of enzymes of the meta-

Cleavage pathway of Pseudomonas putida: the regulation is composed of two operons.”

W 100 (1977)! 71-79.

Beltrame, P., P. L. Beltrame, P. Carniti, and D Pitea. “Kinetics of phenol degradation by

activated sludge in a continuous-stirred reactors.” M 52 (1 1980): 126-133.

Bilbo, C. M. , E. Arvin, H. Holst, and H. Spliid. “Modeling the growth of methane-
oxidizing bacteria in a fixed biofilm.” M 26 (3 1992): 301-309.

Blom, A., W. Harder, and A. Matin. “Unique and overlapping pollutant stress proteins of
Escherichia Coli.”ADnLEn!1m..Mi§mhinl. 58 (1 1992): 331-334.

Bouwer, E. J ., B. E. Rittmann, and P. L. McCarty. “Anaerobic degradation of
halogenated 1- and 2-carbon organic compounds.” mm 15 (1981): 596-
599.

Bowman, J. P. and G. S. Sayler. “Optimization and maintenance of soluble methane
monooxygenase activity in Methylosinus trichosporium OB3b.” Biodegradation 5 (1994):
1-1 1.

Buswell, J. A. “Metabolism of phenol and cresols by Bacillus stearothennophilus.” 1,

mm 124 (3 1975): 1077-1083.

Conglio, W. A., K. Miller, and D. Mackeever. Wm
WM Environmental Protection Agent, 1980.

30

Criddle, C. S. “The kinetics of cometabolism.” W 41 (1993): 1048-
1056.

Criddle, C. S., J. T. Dewitt, and P. L. McCarty. “Reductive dehalogenation of carbon
tetrachloride by Escherichia coli k-12.” MW 56 (1990): 3247-3254.

D'Adamo, P. D. , A. F. Rozich, and A. F. Jr. Gaudy. “Analysis of growth data with
inhibitory substrate.” W 26 (1984): 397.

Dabrock, B., J. Riedel, J. Bertram, and G. Gottschalk. “Isopropylbenze (cumene) - a new
substrate for isolation of trichloroethene-degrading bacteria.” Archives Qf MlthinQg}:
158 (1992): 9—13.

Dalton, H., S. D. Prior, D. J. Leak, and S. H. Stanley. “Regulation and control of methane

monooxygenase.” In MW, ed. R. L. Crawford and R. S.
Hanson. 75-82. Washington, DC: Ammerican Society for Microbiology Press, 1984.

Dalton, H. and D. I. Stirling. “Co-metabolism.” W 297 (1982): 481-496.

Ensign, S. A., M. R. Hyman, and Arp. D. J. “Cometabolic degradation of chlorinated
alkenes by alkene monooxygenase in a propylene-grown Xanthobacter strain.” App},
W 58 (9 1992): 3038-3046.

Ewers, J ., D. Freier-Schroder, and H. J. Knackmuss. “Selection of trichloroethene (TCE)

degrading bacteria that resist inactivation by TCE.” MW 154
(1990): 410-413.

Feist, C. F. and G. D. Hegeman. “Phenol and benzoate metabolism by Pseudomonas
putida: regulation and tangential pathway.” M 100 (2 1969): 869.

31

Fox, B. G., J. G. Bomeman, L. P. Wackett, and J. D. Lipscomb. “Haloalkane oxidation
by the soluble methane monooxygenase from Methylosinus trichosporium OB3b:

mechanism and environmental implication.” Biochemistry 29 (27 1990): 6419-6427.

Gaal, A. and H. Y. Neujahr. “Metabolism of phenol and resorcinol in Trichosporon
cutaneum.” m 137 (1 1979): 13-21.

Galli, R. and P. L. McCarty. “Biotransformation of 1,1,l-trichloroethane,
trichloromethane, and tetrachloromethane by a Clostridium sp.” AppLjnyjmn,
Microbial. 51 (1989): 720-724.

Henry, S, M. and D. Grbic-Galic. “Inhibition of TCE oxidation by the transformation
intermediate carbon monooxide.” WWI. 57 (1991): 1770-1776.

Henry, S. M. and D. Grbic-Galic. “Effect of mineral media on trichloroethylene
oxidation by acquifer methanotrophs.” W1, 20, (1990): 151-169.

Josephson, J. “Implementing superfund.” W 20 (1 1986): 23.

Kamlet, M. J ., R. M. Doherty, G. D. Veith, R. W. Taft, and M. H. Abraham. “Solubility
properties in polymers and biological media. 7. an analysis of toxicant properties that

influence inhibition of bioluminescence in photobacterium phosphoreum (The microtox

test)” WW 20 (7 1986): 690.

Kotturi, G., C. W. Robinson, and W. E. Inniss. “Phenol degradation by a psychrotrophic

strain of Pseudomonas panda.” AnaheiMicmbicloaLaniBimhnnlm 34 (1991):
539-543.

Little, C. D., A. V. Palumbo, S. E. Herbes, M. E. Lidstrom, R. L. Tyndall, and P. J.
Gilmer. “Trichloroethylene biodegradation by a methane-oxidizing bacterium.” Anal,
W 54 (4 1988): 951-956.

32

Mackay, D. and P. J. Leinonen. “Rate of evaporation of low-solubility contaminants from
water bodies to atmosphere.” WM 9 (13 1975): 1178.

Mcfarland, M. J ., C. M. Vogel, and J. C. Spain. “Methanotrophic cometabolism of TCE
in a two-stage bioreactor system.” £31,393, 26 (2 1992): 259-265.

Miller, R. E. and F. P. Guengerich. “Oxidation of trichloroethylene by live microsomal P-

450 evidenCe for chlorine migration state not involving trichloroethylene oxide.”

W511! 27 (5 1982): 1090-1097.

Monod, J. “The growth of bacterial cultures.” MW 3 (1949): 371-394.

Nelson, M. K., S. O. Montgomery, E. J. O'Neill, and P. H. Pritchard. “Aerobic
metabolism of trichloroethylene by a bacterial isolate.” W191, 52 (2
1986): 383-384.

Nelson, M. J. K., S. O. Montgomery, W. R. Mahaffey, and P. H. Pritchard.

“Biodegradation of trichloroethylene and involvement of an aromatic biodegradative

pathway.” WM 53. (5 1937): 949-954-

Nelson, M. J. K., S. 0. Montgomery, and P. H. Pritchard. “Trichloroethylene metabolism

by microorganism that degrade aromatic compounds.” AW 54 (2
1988): 604-606.

Oldenhuis, R., R. L. J. M. Vink, D. B. Janssen, and B. Witholt. “Degradation of
chlorinated aliphatic hydrocarbons by Methylosinus trichosporium OB3b expressing
soluble methane monooxygenase.” AW 55 (11 1989): 2819-2826.

Pawlowsky, P and J. A. Howell. “Mixed culture biooxidation of phenol. I. determination

of kinetic parameters.” WM 15 (1973): 889-896.

33

Phelp, T. J., J. J. Niedzielski, R. M. Schram, S. E. Herbes, and D. C. White.

“Biodegradation of trichloroethylene in continuous-recycle expended-bed bioreactor.”

WW1. 56 (1990): 1702-1709.

Rizzuti, L. and V. Augueliaro. “Kinetic parameter estimation in monoculture and

monosubstrate biological reactors.” W 60 (1982): 608.

Saez, P. B. and B. E. Rittmann. “Biodegraation kinetics of 4-chlorophenol, an inhibitory

co-metabolite.” W 53 (15).
1991): 838-847.

Schaumburg, F. D. “Banning trichloroethylene: responsible reaction or overkill?”

W24 (1 1990): ”-22.

Schmidt, S. K., S. Simkins, and M. Alexander,. “Models for kinetics of biodegradation of

organic compounds not supporting growth.” W50. (1985): 323-
331.

- Speitel, G. E. and J. M. Leonard. “A sequencing biofilm reactor for the treatment of

chlorinated solvents using methanotrophs.” WM 64 (5 1992):
712.

SoKol, W. “Uptake rate of phenol by Pseudomonas putida grown in unsteady state.”
W 32 (1988): 1097-1103.

Stanley, 8. H., S. D. Prior, D. J. Leak, and H. Dalton. “Copper stress underlies the
fundamental change in intracellular location of methane mono-oxygenase in methane-
oxidizing organisms: studies in batch and continuous cultures.” W 5 (7
1983): 487-492.

34

Strand, S. E., J. V. Wodrivh, and H. D. Stensel. “ Biodegradation of chlorinated solvents
in a sparged, methanotrophic biofilm reactor.” W 63 (6 1991): 859-
867.

Strandberg, G. W., T. L. Donaldson, and L. L. Farr. “Degradation of trichloroethylene

and trans-1,2-dichloroethylene by a methanotrophic consortium in a fixed-film, packed-
bed bioreactor.” W 23 (1989): 1422-1425.

Szetela, P. W. and T. Z. Winnicki. “A novel method for determining the parameters of
microbial kinetics.” 31mm 23 (1981): 1485-1490.

Templeton, L. L. and C. P. L. Grady. “Effect of culture history on determination of
biodegradation kinetics by batch and fed batch techniques.” M 60 (1989): 651.

Tischler, L. F. and W. W. Jr. Eckenfelder. “Linear substrate removal in the activated

sludge process.” In W, 361-374. 2. Oxford, England:
Pergamon, 1969.

Tsien, H. C., G. A. Brusseau, R. S. Hanson, and L. P. Wackettet. “Biodegradation of
trichloroethylene by Methylosinus trichosporium OB3b.” A991,_Enx1r99,_M191'99191, 55
(12 1989): 3155-3161.

Vanilli, T., M. Logan, and A. B. Hooper. “Degradation of halogenated aliphatic

compounds by the ammonia-oxidizing bacterium Nitrosomonas europaes.” A991,

W 56(1990): 1169-1171.

Vogel, T. M., C. S. Criddle, and P. L. McCarty. “Transformation of halogenated aliphatic
compounds.” WM 21 (1987): 722-736.

35

Wacket, L. P., G. A. Brusseau, S. R. Householder, and R. S. Hanson. “Survey of

microbial oxygenases: trichloroethylene degradation by propane-oxidizing bacteria.”

W91. 55 (1 1989): 2960-2964.

Wackett, L. P. and D. T. Gibson. “Degradation of trichloroethylene by toluene
dioxygenase in whole cell studies with Pseudomonas putida F1.” W
9419119th54 (1988): 1703-1708.

Wackett, L. P. and S. R. Householder. “Toxicity of trichloroethylene to Pseudomonas

putida F1 is mediated by toluene dioxygenase.” WM 55 (10 1989):
2723-2725.

Weast. R. C.. ed. CREhandlxmkfaLchemistnLaninhxsics 73 ed.. Boca Raton. Florida:
CRC Press, Inc., 1992.

Westrick, J. J ., J. W. Mello, and R. F. Thomas. “The groundwater survey.” mm
mm 76 (1984): 52-

Wilson, J. T. and B. H. Wilson. “Biotransformation of trichloroethylene in soil.” App],
W 49 (1985): 242-243.

Winter, R. B., K.-M. Yen, and B. D. Ensley. “Efficient degradation of trichloroethylene
by a recombinant Escherichia Cali.” W 7 (3 1989): 282.

Yang, R. D. and A. E. Humphrey. “Dynamic and steady state studies of phenol
biodegradation in pure and mixed culture.” W199“ (1975): 1599-1615.

CHAPTER 3
USE OF NUMERICAL SIMULATION FOR DESIGN OF SEQUENCING BATCH
REACTOR

3.1 Introduction

Sequencing batch reactors (SBRs) are established alternatives to conventional
continuous flow systems. SBRs are flexible in design and operation, well-suited for
nutrient removal, and give excellent solids separation (Irvine 1989) Substrate gradients
are easily established, redox conditions changed, reaction periods modified, and
substrates added as needed. These manipulations make it possible to enhance the growth
of organisms that carry out useful functions such as nutrient removal, while preventing
the growth of organisms that can cause problems, such as bulking. The creation of
selective niches within SBRs offers exciting possibilities for innovative engineering

design.

Current design procedures for SBRs are highly empirical - relying on rule-of-
thumb criteria, such as hydraulic detention time and substrate loading rates (Arora and
others 1985). Such approaches typically produce conservative designs, are limited in
range of application, and lack predictive or simulation capabilities. Simulation
capabilities are only achievable using dynamic models that distinguish between the
different types of organisms, substrates, and solids that may enter or be produced within
an SBR. Development and verification of such models is complex because SBRs are
inherently non-steady state, and tools for the verification of changes in microbial

community structure have been lacking. It now appears likely, however, that the latter

36

37

problem will be at least partially overcome by improved methods of microbial
community analysis (signature molecules, gene probes, antibody probes, image analysis,
etc.). These methods should provide insight into factors affecting the activity of key
populations and the relationships between different microbial groups. If this
understanding can be captured in verifiable kinetic expressions, computer simulations of

a range of reactor design and operating scenarios seems possible.

In order to deve10p a reasonable computer simulation tool, a wastewater
classification scheme is needed that distinguishes between substrates, microbial
populations, and non-viable suspended solids. For demonstration purposes, we adopted a
simplified version of the IAWPRC classification scheme (Henze and others 1987). Mass
balance expressions are developed for each constituent and organism type. A solution
representing long-term stable operation of the SBR is obtained by repeated numerical
solution of the mass balance equations, repeated application of the selected solids wasting
schedule, and repeated application of constraints on SBR operation. This long-term
repeating pattern is used as the basis for SBR design. Predictions of changes in the
concentration and age of specific cell types once the long term operating pattern is
established should facilitate simulation of age-related enzyme activities within specific

populations.

3.2 General approach

3.2.1 Constraints on SBR design
Time constraints

The operating cycle for a conventional SBR consists of five distinct time periods:
fill (if), react (t, ), settle (t, ), decant (rd) and idle (t,- ), where:

r, = tfrtrI-tsnd +t,- (3-1)

38

For N , reactors treating a continuous stream of wastewater, the time required for the first
reactor to complete its react, settle, decant and idle periods must equal the time required

to sequentially fill the remaining N r -1 reactors:

(N, -1 )tf =tr+ts+td +2,- (3-2)
Substituting equation 3-2 into equation 3-1,

if = to/N, (3-3)

As indicated by equation 3-3, selection of cycle time and the number of reactors fixes if.
React time depends upon the treatment requirements and the kinetics of degradation.
Settling time depends upon the solids settling characteristics, with a normal value of 0.5 -
1 hour for aerobic municipal wastewater treatment. Settle and decant time should be
chosen so that t, + td < 3 hr to avoid sludge rising and undesired odor caused by
anaerobic reactions. Sufficient idle time t,- is needed to accommodate changes in fill and
react times when flow and substrate loadings vary or when solids are wasted. Typical

values for each of the different time periods are provided in Table 3-1.

Table 3-1. Typical values of the required time for each operating stages in SBRs

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Parameter Unit Typical value Reference
tf hr 25% E Arora et al. 1985.
hr 0 tc Arora et al. 1985.
tr 0.5 .. 2* Irvin 1989
at hr 0.5-1.5 Irvin 1989
ts 20%‘9 Arora et al. 1985.
__‘.d" 111' 15%;; Arora et al. 1985.
ti hr 5% t9 Arora et a1. 1985.
9 hr 12-50 Mecalf & Eddy.1991
ec day NA
r dimensionless 25~75 % Irvin 1989.

 

"' For domestic waste only.

" To avoid rising sludge, Irvine (1989) suggested 1, + td < 3 hr.

 

39

Volume constraints

Volume restrictions arise because each SBR vessel is both a reactor and as a
settling tank. Consequently, part of the volume (V0) is reserved for storage or "recycle"
of settled solids from one cycle to the next. The recycled volume contains the settled
solids after decanting. The total volume of a reactor is made up of the volume of settled

solids and the wastewater volume added during the fill period:
VSBR = 9Q + Vo (3’4)
where: VSBR = reactor volume and Q = influent flowrate.

The relationship between V0 and VSBR is given by:

VSBR = Vo/r = Q(to+t}l (3-5)

where: r = V0 IVs-BR = recycle ratio and to = VO/Q = the time theoretically required to
fill the SBR to a volume V0 at a flowrate Q .

Substituting Q = Volta into equation 3-5 and solving for to gives
r
to = tf(—) (3-6)
I - r
Substituting equation 3-6 into equation 3-5 gives:

VSBR = (If “”75?” 04)

Equation 3-7 indicates that once if and r are chosen, the volume of an SBR is specified.
Another volume relationship of interest in SBR design is the hydraulic residence time, 9 .

For an SBR system, 9 can be defined as:

Nt
9=MS£R=_'_L=_‘_€_ (3-3)

Q l-r l—r

Typical values for 0 and r are provided in Table 3-1.

40
3.2.2 Mass balance equations

Success of the general design approach used in this work depends upon the
development of accurate mass balance equations for dissolved and particulate matter
entering the SBR, produced within it, or consumed within it. For the fill period, the mass

balance expressions are:

51.5; =_Q_ۤ:_~Z2+,3 (3-9)
dr V '
4X1: 90‘} -Xl) +

dt V rx’ (3-10)

 

where: X 1: concentration of suspended solid of type j , X} = influent concentration of

X j, Q = influent flowrate, r Xi = rate of production or consumption of suspended solid of

type j (mass of solid type j produced or consumed per unit volume per day), V = volume
of water in the SBR. To solve equations 3-9 and 3-10, r51. and rxj must be specified, and

initial conditions provided for S,- and X j .
- For the react period, the mass balance expressions are:

_J'.= 5' (3-11)
—L= rxl (3.12)

To solve equations 3-11 and 3-12, kinetic expressions for rs, and rxj must be defined,
and initial conditions provided for S,- and X j . Because the end of the fill period is the
beginning of the react period, initial conditions for the react period are obtained from

solution of equations 3-9 and 3-10 when r = if.

41
3.2.3 Use of the CSTR steady state solution to obtain initial condition estimates

The choice of initial conditions for equations 3-9 and 3-10 is not readily apparent.
The initial conditions of interest are those that occur once the reactor has achieved a
stable pattern of operation: at the beginning of each period of solids accumulation, just
after a wasting event. Repeated solution of the mass balance equations along with
repeated application of a specified solids wasting schedule eventually yields a repeating
pattern. In the repeating solution, each solid type accumulates during the solids storage
period, but is restored to its original mass at the beginning of the next solids storage
period, immediately after wasting. If the first guess of initial concentrations at the
beginning of the solids storage period is far from the values obtained for the repeating
pattern, many computations will be needed to arrive at the repeating solution. To avoid
this problem, a reasonable set of initial conditions is obtained by considering a limiting
case, where tc approaches zero and N , approaches 1. Under such conditions, SBR
operation approaches that of a continuous stirred tank reactor (CSTR). Solution of the
steady state CSTR equations for solids and substrate at a specified solids residence time
provides a reasonable first guess of initial conditions for the SBR. This approach has the
added advantage of facilitating comparison of CSTR and SBR volume requirements. For
a CSTR operating at a specified mean cell residence time, the steady state solution for

each solid type is derivated in Appendix I and given by:
Go
(Xj)ss = ‘9"on +9crxj (3'13)

An initial guess for tc can be obtained from the steady state CSTR solution by computing

0 for the CSTR and relating it to re by equation 3-8:

“Jest = 9C5??? (1’7) (3'14)

42
3.2.4 Mean solids retention time and wasting schedules

Solids tend to accumulate within an SBR and must be removed periodically.
Assuming negligible loss of solids in the effluent, the mass of solids wasted at the end of
a solids storage period must equal the mass of solids that accumulated over that same
period. Mathematically, this can be expressed as:

_ wVOX
N t

WC

QWX

 

(3-15)

where Q” = equivalent daily wasting rate (wasting rate averaged over the entire storage
period), w = volumetric fraction of settled solids wasted at the end of each solids storage
period, NW = number of cycles per solids storage period, so that Nwtc = tsp = solids
storage period. Equation 3-15 can be reformulated to give the mean solids retention time
for an SBR (Irvine 1989):

SRT = Vo/QW = Nwtc/w = tsp/w (3-16)

If SRT, and solid storage time are known or specified, equation 3-16 can be used to
calculate the fraction of the settled solids that must be wasted at the end of each solids

storage period.

The frequency of solids removal events may vary from once a cycle to once a
month or more. Wasting becomes mandatory if either of the following conditions occur:
(1) solids concentration during the react period exceeds some maximum allowable
concentration, X max, (2) the settled solids volume, Vsenled. exceeds the space available for
solids storage, V0. Xm exists because of limitations on oxygen mass transfer (Benefield
and Randall 1980). A typical upper limit when oxygen is supplied by aeration is ~ 5000
mg/L; when oxygen is supplied by pure oxygen, Xm ~ 8000 mg/L (Metcalf & Eddy
1991). Limits on Vsettled can be predicted from settleability tests, such as the sludge

volume index (SVI ):

43

SW 0 X
V tiled = 6 VS” .-
10 (3 l7)

5!

V0 must be greater than Vsettled; otherwise, suspended solids will be lost from the reactor

during the decant period.
3.2.5 Dynamic cell age

In a steady state CSTR with recycle, mean cell residence time Be is a constant and
is controlled by solids wasting. Solids residence time and mean cell residence time are
equal and constant for all types of organisms and suspended solids. In an SBR, the
average age of a given microbial population varies with time: decreasing during periods
of feast and growth and increasing during periods of famine and decay. Consequently, the
concept of dynamic cell age is more meaningful measure of cell physiology than SRT. In
principle, cell age can be linked to the level of expression of enzyme and/or RNA

activities of a specific microbial population (Shu 1961).

In the present article, we use dynamic cell age (A ) to describe the mean cell age
of a designated microbial population. Vaccari et a1. (1985) have previously described
application of the dynamic sludge age (DSA) concept to non-steady state activated sludge
systems. The derivation of equations for DSA is complex, and the reader is referred to the
original references for details (V accari and others 1985; von Foerster 1959). As derived
by Vaccari et al. (1988), computation of A for the case where no solids are wasted or lost

over the period t is given by (V accari and others 1988):
A = (Ao+t/2)Mao/Ma+r/2 (3-18)

where A = dynamic cell age at time t, A0 = dynamic cell age at time 0, Mao = active
biomass at time 0, M0 = activate biomass at time t. Equation 3-18 can be used to
calculate A in the fill and react periods when Mao and M, are known. During the settle,
decant, and idle periods, we assume that Maoma = 1 so that A = Ao+t.

44

Different microorganism types have different growth rates and different limits on
cell age. A limiting age occurs when cells are growing at their maximum specific growth
rate um so that dMa/dt = pmMa. A minimum age Ami" occurs when the cells are growing
at this rate. Its value can be computed by taking the time derivative of equation 3-18,
setting it equal to zero, solving for A0, then substituting the result back into equation 3-
18. The resulting expression is:

1 Mao t

Amin=2m(m+1)+2 (3-19)

In the limit, as t approaches 0 and Ma approaches Mao, Ami" approaches 141m:

1
(Ami-win. = fi- (320)
m

If a design choice for SRT is selected which is less than (Am-"mm for a given desired
microbial population, then that population will not be maintained in the SBR during long-
term operation. Accordingly, design values for SRT should be chosen that are
significantly larger than (A ”mi," for the slowest growing population required for proper
reactor performance. Table 3-2 summarizes (Amnllim values for common organisms types

used in wastewater treatment.
3.2.6 Generic solution procedure

The flowchart of Figure 3-1 summarizes computational procedures used to obtain
a repeating solution representing long-term SBR operation. Input parameters include
waste characteristics, SBR operating parameters, and rate coefficients for degradation or
production of solids and substrates. The CSTR steady state solution is calculated and
used as the first guess for initial conditions to solve the differential equations describing
solids and substrate concentrations during the fill and react periods. Wasting requirements
are satisfied at the end of each cycle. The most recent calculations of solid and substrate

concentrations are used to revise initial conditions, and the calculations are repeated.

45

Table 3-2 Minimum cell age for different group of microbial populations

 

 

Function of Microbial
Populations Pm (daY' 1) (Alain)lim Reference
Aemlzic
Organic removal 8 0.13 (1)
Nitrification 0.72 1.39 (2)
Sulfur oxidation 1.5 0.67 (3)
Ferrous oxidation 0.45 2.22 (3)
Hydrogen oxidation 1.8 0.56 (3)
Anaerobic
Denitrification 0.3 3.3 (2)
Sulfate respiration 0.9 1.11 (2)
Methane
fermentation
Fats 0.26 3.85 (3)
Proteins 0.68 1.47 (3)
Carbohydrates 1.9 0.53 (3)
Me 0.68 1.47 (3)

 

Reference: (1) Grady and Lim 1980, (2) Metcalf and Eddy 1992, (3) Lawrence and
McCarty 1970.

46

 

 

Establish:
ange desi - Waste characterization scheme
arameters gai—F - Rate express1ons for SOlldS and substrate

 

 

 

— SBR design parameters, SRT, tsp. N,., r, t, id, re.

1

CalcrIate: ' J

 

- CSTR steady state solution base on solid removal
and substrate stabilization éequirements.

Solve:

- Modeling equations for solids and substrate
during the fill period by using CSTR steady state
solution as the initial solution as the initial

49119111011

 

 

 

 

 

 

Solve:
— Modeling equations for solids and
substrate during the fill period using the
solutions at the end of previous cycle as
the initial condition

1 Y

 

 

 

 

 

Solve:

- Modeling equations for solids and substrate during the
react period by using the solutions at the end of fill
period as the initial condition

- determine t, when S = S,

 

 

 

 

 

 

 

 

 

 

 

 

Figure 3-1 Flowchart for numerical simulation and design of an SBR system

 

47

Eventually, the solids and substrate concentrations at the beginning of a new solids
storage period equal the solids and substrate concentrations at the beginning of the
previous solids storatge period. This repeating solution represents the long-term stable

pattern of operation.

Overall, two levels of iteration control are involved. The first is the number of
operating cycles (N). Solid and substrate concentrations iterate until the wasting occurs
(i.e., N =Nw). The second iteration control parameter is the solid and substrate

concentrations that continue to iterate until the repeating solution is obtained.
3.3 Application to domestic wastewater treatment
3.3.1 Choice of wastewater classification scheme

In order to implement the solution strategy described in the previous sections,
mass balances must be obtained on the different types of suspended solids and substrates
within the SBR. Many possible classification schemes are possible, and the appropriate
. choice will depend upon the system under consideration. Figure 3-2 illustrates a scheme
used for organics oxidation by aerobic heterotrophs, along with ammonia oxidation by
nitrifiers. The classification approach used is similar to that of IAWPRC (Henze and
others 1987). The first level of classification entails separation of suspended solids from
dissolved matter. This distinction is made by filtration of the wastewater through glass
fiber filters (APHA 1989). Suspended solids are further classified as either fixed or
volatile by igniting them at 550°C for one hour in a muffle furnace (APHA 1989). In
general, fixed suspended solids constitute the inorganic fraction of suspended solids (Xin)
while the volatile suspended solids represent the organic portion of suspended solid.

Biodegradability is an important characteristic in the classification scheme of

Figure 3-2. If dissolved organic matter is refractory, it passes through the biological

48

5:8 5555088 £55355 u 5N 6N be .3555 5.5....
8.2555 «.2: u 2k 5:8 5555588 b58852 u c». 82.8 5055588 535583555 .1. wk .8283 953.5?
5.58 n 2% "255553.555 8.5.53 258 u .5 6.55885 u an 6.5.858 guano 535533553 382 u
a. 58858 :2: N am. .223 888555 Bane—53 5 85385255 :55 55 88:25.85 88? at... Bani

 

 

 

 

 

 

Ski‘.
.x
V i ‘I 85:58 55555526
vuu— LN
35 All 223
8:82:
IAIDVI gk ”1‘-
. .‘luoocsm 8 o 5 5m

 

..l. gal
525 x all
.3559 3558:. E5883:

 

49

treatment process unchanged; if it is biodegradable, it is consumed for cell growth and
maintenance. Degradable suspended solids (X4) are removed by wasting and are
converted into dissolved biodegradable substrate by the hydrolytic action of extracellular
enzymes; refractory suspended solids (X r) are only removed by wasting. Active
heterotrophic biomass (X1. ) is produced by growth on soluble biodegradable organics and
is lost by decay. Active nitrifying biomass (X N ) is produced by growth on ammonia.
Decay is treated as a lumped parameter, including endogenous decay, death, predation,
and lysis. Decay is assumed to convert active biomass into biodegradable particulate

matter (X4) and inert particulate matter (X ,- ).
3.3.2 Choice of kinetic parameters

The selective pressures in an SBR are significant and can result in communities
with kinetic properties that are not predictable. Accordingly, bench- and pilot-scale

testing are recommended to establish appropriate kinetic expressions and coefficients.

For illustration purposes, we assumed saturation kinetics for substrate removal
and first order kinetics for the hydrolysis and decay of suspended solids. Rate expressions
for each type of solid and substrate are summarized in Table 3-3, and the corresponding
differential equations for the fill and react periods are summarized in Tables 3-4 and 3-5,
respectively. Steady state CSTR solutions are provided in Table 3-6. The steady state

CSTR solutions are used as the initial conditions for fill period of the first cycle.

Although saturation kinetics are used in these examples, the design and simulation
procedure outlined here is readily adaptable to different rate expressions, appropriate to
the system under consideration. Mass balance equations are obtained by substituting the

appropriate rate equation into equations 3-9 - 3-12.

50

Table 3-3 Assumed rate expressions for each type of solids and substrate
categorized in Figure 3-2.

Wastewater constituent ‘ Rate egression
0
0
—k X +fde +fdeXN

-bX + 1152!,
K +S
(I’fd)th +(1-fd)bNXN
YustNXN
KN 4’ SN

+thd

 

 

 

 

 

-b~X~ +

 

_ ksx.
K. + s

 

 

fi_X:-X,
dt — r+t.,
£X£_Xg,-X,-,,
th - Ht.

dXd X3-
= — X + bX + bX
dt 14% ka 4 fd 1. f4 N N

a). =Xh -Xh -th + W].

dt t+t., K,+S
dXN = XX, -Xfl “bNXN + YnstNxN
dt Hr, K~+S~

dX—' =‘X —+(1-X --f.)bx.+(1 fdleXN

d_t r + r,
-S _ kSX,
t +t. K, + S
"SN _kNSNXN
-KN 4' SN

 

 

 

 

+ thd

 

 

 

4' kath

 

 

51

Table 3-5 Mass balance equations and solutions for SBR in react period

.. ance egatron

er=0
dt

“in

(It
dX
—1 = -k,.x., + fdbx, + fdeXN

dt
_& = _th + YkSXi.
dt K, + S

dXN Y NkNSNXN

 

=-b X +
dt N N KN+SN

$3" = (1 -f.)bx. +(1 -fd)b~X~

 

+ thd

 

 

 

52

Table 3-6. Steady-state solutions for solids and substrate in a CSTR with sludge

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

recycle
0 t or
‘ substrate Steady-state solution
9
—‘—X°
X. 0 .
9
. 4X?
XIII 9 m
X, gal X; + fdb9 Y(S°- +‘fde9 YN(S"— —S,, N),
9 (1+k,9) (1+k39)(1+b59c+) (I+k,cc9)(1+b,,9)
‘ Y(S"- S)
X _9__c
" _[_1+b9 I
X" .nglJM]
1+ b,,9
X. 9,[Y(S"— S)(I—fd)b9, + Y,,(S°— —S,, )(I- -fd)b,9]
' “be, 1+b,,9
(1+b6c)K,
S
9C(Yk-b)-1
SN (1+bN9c)K,,

 

9,(Y,,k,, -b,,) - 1

 

MLSS

 

53
3.3.3 Choice of treatment objectives

Treatment criteria include effluent standards and requirements for stabilization.
Effluent standards for domestic wastewater typically include standards for BOD removal,
ammonia removal, and suspended solids removal. For an SBR, the effluent suspended
solids standards are typically satisfied without difficulty. As a result, soluble substrate
removal and overall waste biological stabilization are the critical design objectives. The

efficiency of substrate removal (BS) is calculated from the expression:
Es (%) = 100(S°-S,,)/So (3-21)

Overall waste biological stabilization (E) can be defined as the percentage of
biodegradable material removed during treatment (including soluble and suspended
forms). In general terms,

= 100 [(input total CODb + input NOD) - (output total CODb + output NOD)]

E input total com, + input NOD

 

Total CODb is total biodegradable soluble and suspended COD. For a CSTR,

C

V 1

100':Q(So + XZYXd) - E—(Xfi’de + X371“)- Qse

E (%) =

CSTR Q(So + X: YXd) (3-22)

 

 

where: YXd = COD to weight ratio for biodegradable suspended solids (mg COD
lmgVSS); YXa = COD to weight ratio for biomass = 1.42 mg COD/mg VSS.

For an SBR,

N=Nw
1%Qthw(S° + XSYXd) -( ZQtfse + WV0(XdYXd + XaYXa»
N=I
EsaR(%) ’-

Qthw(So + XSYXd)

 

] (3-23)

54

3.3.4 Program development and execution

A computer program based on the computational procedures of Figure 3-1 and the
classification scheme of Figure 3-2 was developed using QuickBasic. The system of
differential equations for the fill and react periods are solved using fourth-order Runge-
Kutta approximations. The program is initiated from an interactive input screen.
Simulation results are saved as a text file and can be exported to a spreadsheet program
for data manipulation. Information provided from the simulation output includes
concentrations of different types of solids and substrates, reactor size, required time for

each operating stage and mean cell ages of specific microbial groups.
Input parameters

Input parameters that must be specified include the number of reactors Nr (=2, 3,
4...), r (=Vo/VSBR). SVI, :5, rd, tc, Xm, SRT and tsp. Empirical values for those

parameters are summarized in Table 3-1.

. Initial conditions

Design of an SBR is an iterative process, but concentrations are needed for X, and
S.- at the beginning of the fill period to initate calculations. A reasonable set of initial
conditions can be obtained from the steady state CSTR solution. When solids residence
time and effluent concentrations for a CSTR are chosen, the designer can calculate (X,- )ss

and a hydraulic residence time (9) after selecting a desired mixed liquor suspended

 

 

 

solids concentration (X):
{X;+Xf+ X3 + fdbOcY(S°'5) +ftb~9.Y~(S§-S~) +Y(S"—S)‘|
-9; 1+9ck, (1+ k,9c)(1+b9c) (1+k,9,)(1+b,,0.) 1+bo, I
cm-
X — 0 _ o _
+YN(S; S”)+(1-f.)b9.y(s S)+(1_fd)bN9cY(S,, SN) |
1+ M. 1+b0. 1+b,9 J

(3-25)

55

As indicated by equation 3-14, the value of 9CSTR obtained from equation 3-25 can be

used to estimate an upper bound for re . In general, a reasonable working range for for to

is defined by:
t +1
—-'—4— s t s a 1 - 3-26
I-I/Nr c,max CSTR( r) ( )
Calculation steps

The following calculations are performed in sequence:

1. Calculate tf using equation 3-3.

2. Calculate to using equation 3-6.

3. Calculate V0 (= Qto ) and VSBR (= Vo/r ). The total volume is NrVSBR.

4. The volumetric sludge wasting fraction (w ) is calculated from w = tsp/SRT. The

number of operating cycle between two sludge wasting events (NW) is obtained from NW

= tsp/te-

5. Calculate the substrate and solids concentrations throughout fill period (until t= tf) by
solving the mass balance equations for the fill period (Table 3-4). To solve the mass
balance equations, initial conditions are first calculated using the CSTR steady state
solution (Table 3-6). The CSTR steady state solution for solids concentrations is divided
by r to estimate the initial solids concentration for the first cycle. Dynamic cell age, A, is
calculated throughout the fill period using equation 3-18.

6. Calculate solids and substrate concentrations throughout the react period by solving
the differential equations for the react period. The substrate and solid concentrations at

the end of fill period are the initial conditions for the react period. Calculations continue

56

until all treatment objectives are satisfied, where t =tr. Dynamic cell age, A, is calculated

throughout the react period using equation 3-18.

7. If solids are not scheduled for wasting, a new operating cycle is initiated by repeating
steps 5 to step 6, where the initial concentrations for a new operating cycle are the
concentrations at the end of the previous operating cycle, adjusted for settling (divided by
r). Between wasting events, biomass accumulates as it is carried over from one cycle to
the next, making it possible to satisfy the treatment objectives with a shorter t,. In
practice, however, a constant I, is used. Consequently, t, is computed for the first cycle
and maintained constant thereafter. This results in some excess level of treatment in

cycles after the first cycle of the solids storage period.

8. Wasting is executed when one of the following conditions is satisfied: (1) the operating
cycle N equals the wasting cycle NW, (2) X > Xmu , or (3) Vsludge > V0. Assuming solids

are removed during the idle period, the volume to be wasted (V was”) is
Vwaste = WVo (3-27)

After wasting, the liquid volume is reduced from V0 to V0 - Vwam. Because V0 is
changed, tf and to also change. Idle time, t,- , decreases to cover the increase of tf.
Because t,- is intended to provide operating flexibility and safety, a decrease in t,- will
not affect reactor operation, provided that peak flows can be accomodated and ti > 0 .
After mathematically altering V0 to account for wasting, the reactor is ready for another

solids storage period.

9. Divide suspended solids and substrate concentration obtained from step 8 by r and use

them as the initial condition for next cycle calculation.

10. Calculate ti from ti = tc-(tf+t,+tg+td). If t,- < 0 , calculations should be repeated using

a longer cycle time.

57

11, Compute oxygen requirements for the repeating solution. Oxygen utilization is

proportional to substrate utilization;

r02=75fer3 (3-28)

where r02 = oxygen utilization rate, mg OzlL-d, 7,, = COD to weight ratio for the

substrate, mg 02/ mg S, rs = substrate utilization rate, mg S/L-d, f, =1-1.42%=

fraction of electrons removed from substrate and transferred to oxygen for energy. The

oxygen requirement is obtained by multiplying r02 by the reactor volume, where

V( t) = V0 + Qt for the fill period and V( t) = VSBR for the react period.

12. Compute nutrient requirements. The accumulation of active biomass can be computed
during each solids storage period, between wasting events. Nitrogen and phosphorus
requirements are estimated by assuming the cell contain 2% of phosphorus and 12% of

nitrogen in the basis of dry weight.

Many possible design configurations can satisfy the treatment goals. Thus, a range
of conditions should be explored. The final design decision should consider capital and

operating costs, space requirements, flexibility, etc.
3.3.5 Simulation results

Two design examples were evaluated to illustrate the proposed design procedure.
Conditions and design comeouts for both examples are summarized in Table 3-7. The
first example is for aerobic heterotrophs removing biodegradable organics. Changes in
substrtate concentration and DSA during the repeated solution are shown in Figures 3-3.
As indicated in Figure 3-3, substrate concentration increased in the fill period and rapidly
decreased during react period. Suspended solids concentrations increase during the fill
period, and remain constant or undergo slight changes during the react stage. Dynamic

cell age decreases during the fill period, decreases until substrate is removed during the

58

react period, and increases after substrate is removed. Figure 3-4 demonstrtated the
accumulation of different type of solids during reactor operatiion. Solids concentration
accumulated during the normal operating cycle and decreased when sludge wasting took

place.

The second design example was for an SBR containing both heterotrophs for
organics removal and nitrifiers for ammonia oxidation. The scheme of waste
characterization and the system of modeling equations for this case were similar to the
first example except some extra terms derived from nitrification were added. Figures 3-5
and 3-6 summarize changes in substrate concentration, DSA, and each type of solids..
Trends for substrate removal, solid accumulation, and dynamic cell age variation were
similar to the first example, but longer tf, t, and tc are required. The required reactor

volume and dynamic cell age are also larger.

In the proposed SBR design procedure, the designer specifies SRT and tsp to
determine solid wasting parameters (NW and w ). The effects of these parameters on
SBR behavior were evaluated by executing the program under a default condition in
which only one parameter was changed. Simulation results are summarized in Figures 3-7
and 3-9. For low values of w, the SBR accumulated more solids, and required less reactor
volume and a shorter cycle time (Figure 3-7). Higher values for N... also had the same
effect on the concentration of accumulated solids, reactor volume and cycle time (Figure
3-9). Choosing a higher N,, or lower w in SBR design results in older cell ages but the cell
age never exceeds the design SRT. This result is ensured by selection of an SRT that is
sufficiently greater than (Amin)lim for the slowest growing population that is needed for

efficient reactor operation.

59

Table 3-7. Summary of input parameters and design outputs for design examples

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Carbplriiuggrggroonn and 023;; Carbon oxidation and nitrification £3282“
"”Kinetic Parameters"** “Influent Waste Characters"
Ks(glm3) 60 60 Q(m3/d) 10000 10000
KN(g/m3) 0.4 NA“ S°(g[m3) 300 300
k(d") S 5 {fig/m3) 40
kN(d°‘) 1 NA“ X g(g/m3) 0 O
kh(d”) 0.06 0.06 X g, ( g/m3) 0 0
b(d") 0.06 0.06 X g (g/m3) 130 130
bN(d“) 0.02 0.02 X," (g/m3) 0 0
Y(g/g) 0.6 0.6 X g, (g/m3) 60 60
mag/3) 0.36 0.36 x?(g/m3) 40 4o
fd(g/g) 0.8 0.8
*"“"* Desi ParametersIMHMI
E(%) 10 10 N, 4 4
Es(%) 90 90 r 0.5 0.5
ESN(%) 90 NA’ ts (d) 0.04 0.04
Xmax(mgll) 8000 8000 td (d) 0.04 0.04
SVI(mllg) 50 50 tc (d) 0.48 0.23
*tfitt Wasting Strategy fitttt
SRT(day) I 6.4 3 tsp(day) 1.9 0.9
Outputs for design examples
Required time Volume and wastifl requirement
tf(d) 0.0564 0.1 193 Nw 4 4
tr(d) 0.0482 0.2380 w 0.3 0.3
to(d) 0.0564 0.1 193 Vo(m3) 564 1 193
ti(d) 0.04 0.04 Vsnn(m3) l 127 2387

 

* NA = Not apply.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o : , r j
o 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
Time(day)
2.8
g» 2.75
‘6
v 2.7
0
3‘ 2.65 -
“'7'; 2.6 ..
9
.2 2.55 4-
E 2.5 ~-
5. 245
5 ° "
2.4 : k : : : 4 : :
o 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
Time (day)

Figure 3-3 Repeating solution for dynamic cell age and substrate concentration in an SBR

system designed for carbon oxidation (example 1)

61

 

 

4000
3500 .-
3000 .-

\ \

g 2500 -- m \‘fi\\\\\ \\

 

M —' N — V’ V co m N O —' V N 55

sassazcnecaaasznaxcfincss

O O O O O O O O O O O —- —- —- —- — - -- - —
Time (day)

Figure 3-4 Repeating solution for mass of each type of solids present during two

solid storage periods.

62

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

30 *
.1, m
"' r‘
\
g 15., fl 1‘ D
8 10..
0.
0 05 1 15 2 2.5 3 3.5 4
Time (day)
20

 

 

SN(m8’|)

 

 

 

 

 

 

 

4

g, 6
‘; 5.8
a, .
z 5.6-
Ti
° 5.4-+
i!
a 5.2--
E.
g 5 : 4 : : i 4 :

0 0.5 1 1.5 2 2.5 3 3.5 4

Time (day)

Figure 3-5 Repeating solution for substrate concentration and dynamic cell age in an SBR
system designed for simultaneous oxidation of organic matter removal and ammonia .

63

 

Ame—V852

Time (day)

Figure 3-6 Repeating solution for mass of each type of solids accumulated during two

solids storage periods.

.euszaacifisu{henceucziEcnn—ZSE
88 "5x85 58.312: u...8.$8n.m.8_ u 5.858uex858u8x8552 usxdusxusx85 8n
n 5m 5205555— 0 O .55 new 6.5 u > .75 55.5 n 5 u .5. .75 n u x .558 55 n J— 93 5555388 E 58: 825885.— 685 5.28 58
5823 .055 :55 555553 62.8 555588 .85. 55.. 5555.8 5555552 25 :5 93 55355855 55:83 55:8 55 38.5w gum 255E

 

 

 

 

 

 

 

 

3.5 05559.2— usuuakr 3.5 once—.582— 9.233

cm 3 am ON 5. 5 5n cc an an O— 5
u v n u o . u n u u c a
.. 8.5 m
:88 m
1... —.c a A
) :88. m.
52.5 .w m
“Emu .-
\\ 4. «.5 I ttttttttttttttttttttttt [#5552 m
2.5 88 (

 

A35 «55:09.2— 9.3553

 

 

   

 

 

 

8 9. 8 8 2 5
. u p u u .5 u
.m m
a
m a.“
S 'l
cm W».
n) m.
cm w
h D.

 

 

65

.8nu3.¢usz...8.ou_.u:u{Sufifiacnu

.8 .58. 88 u Ex .8... 88 u x a c. u $8 .88 u .m .8. n .m .8... 5.. u ex .3... 8 "ea. .8... 8. tax .5 no... u ax
.8... 8n n 8 52588. n 5 .8 n... a... u w .1. 8... u a u s. .2. n u U. .3... 8 u .5. ca 8.5.8.... c. .58.. 2288.... .28.9'

55.552 555 .5? :55 5.855% 82.8 5555588 35. 55.. 5555—8 wagon—5.. o... 55 5.25 .8552 55 .38.... 55 55 anaem— m-m 8:5."—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Auzv 52.55.. 535-8— 825 boa-5.. 55.5.08
5 h e n v m a a h o n v m N
t. t. u n w 5 . u n G
11 c s
n55 V
at —.c a m
: n... m s
8
IT . t «.5 .0 ‘H I.
/ nN c m
n5
€25 52.8.... 58553— 5.25 52.5.... 55.5.3—
5 N. 5 n v m n a h 5 n v n a
. .1 n u n u . 58. n t. n n J. 5 1
S o
L.
88 .m 85. m
i" I ) a 1 I g
83 m m. m, m
m w .. 89. a m
( IT 1.-
l I 88. w. .8. m
m. .. 88
. 55mm 855 m

 

 

.38"B...8.eu_.u...u...n..u..vu.z.3...8u.>m.3...
88 u .2... .3... 88 u x ... 2 u .55 .88 u .m .3... u .m .3... 8 u 2.. .3... 8 us... .3... 8. "8x ... "c... u 2.. .3... 8n
n om .82.. 88. u o ...... u... .8... u w .1. 8... u .. u 5. ..-.. n u .. .3... 8 n J. 2.. 8.3.2.... ... 8... 5.23.... .2... 0.... 8..
2:23 .0? :3 2.5.5:. .82.... 329%.... .38 .8 sous—o... 2:83. 2.. .8 1.7:: mucus—.0... mag...» 2...... .8 28...... dim 0...»...

32 ’2
N. c. a c v N N_ c. a c v N
T T u u u T u c

 

 

.d u

an...

.1.
O
(5'?) 3:
é
(gm) ammoA Joana}!

 

 

 

 

 

 

 

 

 

 

 

 

 

 

:3...

m... 88

32 32
N. 2 m c v N N. 2 .. e v N

. . u .r .. . . . . .. 1.
0
... :88. m
llN m m IU§N ) m
. v m\ :8... u u
l' ) i u
' I 1'; W p
3.. X\ -8... .
I :n I :88 w

.. [III 88..

67

Increasing the number of SBRs reduces the total volume requirement, required
SRT, and cycle time (Figure 3-8). This is because the SBR performs more like a plug
flow reactor as N, increases. The decrease of total reactor volume is not so apparent when
N, > 4. This is because the dilution of substrate becomes more pronounced during the fill
period (reducing the initial substrate concentration) as tfdecreases. As expected, the SBR

required significantly less volume than a comparable CSTR system.

3.4 Summary

Numerical simulation appears to be an effective tool for SBR design and
simulation. The ability to simulate substrate and solids concentrations, and to predict
oxygen utilization rates and nutrient requirements enables improved understanding of the
complex interactive processes within the SBR and should facilitate design and operation.
The dynamic cell age fluctuations of active microbial populations was constrained by the
choice of SRT. Monitoring of cell age provides a means whereby cell age-dependent

phenomena might be simulated and controlled.
Literatures cited

APHA.
Washington, D. C.: APHA, 1989.

 

Arora, M. L., E. F. Barth, and M. B. Umphres. “Technology evaluation of SBRs.” .L
m 57 (8 1985): 867-875.

Beneficld. L. D. and C. W. Randall. W
1st ed., Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1980.

Grady, C. P. L. and H. C. Lim. Wm. New York: Marcel
Dekker, Inc., 1980.

68

Henze, M., C. P. L. Grady, W. Gujer, G. V. R. Marais, and T. Matsuo. W
W. IAWPRC, 1987.

Irvine, R. L. “SBR for biological wastewater treatment.” 98;; Emma] Bflil’s in
Enximnmentalslonml 18 (4 1989): 255-266.

Lawrence, A. W. and P. L. McCarty. “Unified basis for biological treatment design and

operation.” W 93 (SA 3 1970): 757-
778.

Metcalf & Eddy, Inc.

 

McGraw-Hill series in water resources and environmental engineering, ed. R. Eliasen, P.

H. King, and R. K. Linsley. New York: McGraw-Hill, Inc., 1991.

Shu, P. “Mathematical models for product accumulation in microbiology processes.” 1,

W 3 (l 1961): 95-100.

Vaccari, D. A., A. Cooper, and C. Christodoulatos. “Feedback control of activated sludge
. waste rate.” 1m 60 (11 1988): 1979-1985.

Vaccari, D. A., T. Fagedes, and J. Longtin. “Calculation of mean cell residence time for
unsteady-state activated sludge systems.” W 27 (5 1985): 695-703.

von Foerster, H. “Some remarks on changing populations.” In W3:
proliferation, ed. F. Stohlman. 382-407. New York: Grune & Stratton, Inc., 1959.

CHAPTER 4
EFFECTS OF GROWTH SUBSTRATE FEEDING PATTERN ON COMMUNITY
STRUCTURE AND COMET ABOLISM

4.1 Introduction

A wide range of chemicals can be detoxified by cometabolism. To sustain
cometabolic reactions, a growth substrate must be added, either continuously or
periodically. In mixed cultures, it is not clear how different methods of growth substrate
addition will affect community structure and cometabolic populations within the
community. The pattern and timing of growth substrate addition is likely to change the
structure of the microbial community and its propensity for cometabolism. This could be of
particular importance in the design and operation of reactor systems for cometabolism. To
explore this issue, trichloroethylene (TCE) was selected as nongrowth substrate, and
phenol as the growth substrate. Certain phenol- and toluene-degrading Pseudomonas
species rapidly cometabolize TCE (Coyle et al., 1993; Nelson et al., 1988; Nelson et al.,
1987; Shields et al., 1986; Shields et al., 1991; Wackett and Gibson 1988), but this
ability is not common to all phenol- and toluene-degraders (Dabrock et al., 1992).
Consequently, enrichments that are fed phenol or toluene may differ in their capabilities for
cometabolism due to differences in the species present. To test this hypothesis, phenol and
TCE degradation activities were monitored in four reactors fed the same daily mass of
phenol, but with different feeding patterns (frequency and feeding rate). The structure and

composition of microbial community for these four reactor was subsequently analyzed.

69

70
4.2 Materials and methods
4.2.1 Chemicals

Trichloroethylene (TCE, 99% purity) was obtained from Aldrich Chemical Co.,
Milwaukee, Wisconsin. Chemicals for preparation of media and analyses were ACS

reagent grade and were purchased from Aldrich or Sigma Chemical Co.
4.2.2 Medium and culture

Phenol feed medium contained (per liter of deionized water): 2 g of phenol, 2.13 g
of Na2HP04, 2.04 g of KH2P04, 1 g of (NH4)2SO4, 0.067 g of CaC12 2H20, 0.248 g of
MgC126H20, 0.5 mg of FeSO4 7H20, 0.4 mg of ZnSO4 7H20, 0.002 mg of MnClz
4H20, 0.05 mg of CoClz 6H20, 0.01 mg of MCI; 6H20, 0.015 mg of H3BO3, and 0.25
mg of EDTA. The pH of the medium was 6.8.

4.2.3 Reactors and reactor operating conditions

Four glass vessel reactors were used to examine the effects of phenol addition: a
pulse fed reactor (reactor P); a continuously fed reactor (reactor C); and two semi-
continuously fed reactors (reactors SC2 and SCS). Each reactor received 200 mL of phenol
feed medium daily, resulting in a phenol loading rate of 0.2 g per liter per day and an
average dilution rate of 0.1 d‘1 (hydraulic residence time of 10 days). Each reactor was
operated so as to maintain a constant liquid volume of 2 L. For reactor P, 200 mL medium
was provided as a single daily pulse immediately after removing the same volume of liquid
from the reactor. For reactor C, 200 mL of medium was supplied continuously. Reactor
SC5 received phenol semi-continuously - alternating between 5 hours of feed and 3 hours

of no feed. Reactor SC2 alternated between 2 hours of feed and 6 hours of no feed.

Reactor vessels (Wheaton 2-L Double-Sidearm Celstir, Wheaton No. 356806) were
vigorously stirred by teflon paddles attached to a spinning shaft and driven by a teflon-

71

coated stir rod coupled magnetically to a magnetic stir plate under the vessel (Figure 4-1).
Defined phenol medium was provided by syringe pumps. An air pump supplied oxygen
(dissolved oxygen maintained > 2 mg/L) through a glass diffusion tube. All reactors were
operated at 21.5il.O °C.

4.2.4 Inoculum

Each reactor was inoculated on the same day with a subsample from the same
phenol-degrading enrichment. The enrichment was obtained by seeding a chemostat with
activated sludge from a municipal wastewater treatment plant (East Lansing, Michigan) and
providing phenol feed medium for two months at a dilution rate of 0.1 d'l. The enrichment
was maintained at 21.5:tl.0 °C. Microscopic examination revealed a diverse microbial
community, including flocs of spherical and rod shaped bacteria, filaments, and distinctive

predators, including protozoa and rotifers.
4.2.5 Evaluation of TCE transformation kinetics

Samples were periodically removed from the 4 phenol-fed reactors just before the
beginning of feed to evaluate TCE transformation activity. Two to five milliliters of culture
was transferred to a 20 mL glass vial, crimp-sealed with Teflon coated butyl rubber
stoppers, and spiked with 10~50 uL of aqueous TCE stock solution to give the desired
initial TCE concentration. Vials were incubated at room temperature on a rotatory shaker at
120 rpm. Periodically, 0.1 mL of headspace gas was withdrawn using a Precision gas tight
syringe and injected into the injection port of a Hewlett-Packed 5890A gas chromatograph
equipped with a 30 m (L), 0.53 mm (ID) DB624 capillary column (Alltech No. 93532) and
a flame ionization detector (helium carrier flowrate of 12 mL/min). The GC oven was
operated isothermally at 90 °C, and both the detector and the injector were maintained at
250°C. TCE concentrations were determined by comparing the peak areas of samples to

the peak areas of external standards. Aqueous phase TCE concentrations were calculated

72

Reactor operated at a
constant liquid volume of
2fimm

    
 

/

 

 

 

 

 

 

 

 

 

 

L___-__-____-__--—--—-| Pump controller

Figure 4-1 Schematic of experimental setup

73

using a dimensionless Henry's law constant of 0.333 at 21°C (Gossett 1987). The initial
concentration of TCE added was estimated fiom triplicate sterile controls after equilibration.
The total sample volume withdrawn from each vial was less than 5% of the total headspace
volume. In general, the coefficient of variance on TCE measurements was 1 to 5% for

triplicate samples.

Data from the above assays were fit to three different kinetic models using the
statistical package SYSTAT 5.2.1. A first-order rate expression (first order in TCE
concentration) fit the data for each of the four reactors with an average r2 > 0.95; a zero-
order rate expression did not fit the data; and saturation kinetics only applied at high TCE
transformation rates. Typical curve fitting results for TCE degradation kinetics were
provided in Appendix B. Transformation rates were also proportional to the biomass
concentration. Based on these observations, TCE transformation kinetics are assumed to
be first-order with respect to TCE concentration and first order with respect to biomass
concentration. A second-order rate coefficient kc' (first order with respect to TCE and
biomass concentrations) was used for comparison of TCE transformation rates. The

corresponding mass balance expression was:

dC -
dt ‘ ( )

where C = concentration of TCE in the aqueous phase and X = concentration of biomass.
Assuming that cell concentration does not change appreciably during the assay, the

integrated form of equation 4-1 is:

lnC= In c, -k;x: (4-2)

where Co = initial concentration of aqueous TCE.

74
4.2.6 Measurement of pH, dissolved oxygen, optical density, phenol, and MLSS

Dissolved oxygen, pH, and optical density (absorbance at 600 nm) were routinely
monitored with a dissolved oxygen probe (Orion model 97-08), pH meter (Orion 720) and
spectrophotometer (Shimadru UV-l60), respectively. Phenol was assayed by liquid
chromatograph on a Gilson HPLC (Gilson Medical Electronics, Inc., Middleton,
Wisconsin) equipped with a 250 mm (L) x 4.6 mm (ID) Whatman C-18 column (Alltech
NO. 46211502). The flow rate of the mobile phase (40% acetonitrile + 60% deionized
water containing 0.1% H3PO4) was 1 mL/min. A 20 pL injection loop was used for
injecting sample. Phenol was detected by a UV detector at 235 nm. The detection limit was
0.5 mg/L. A typical coefficient of variance for phenol measurement for 5 samples was less
than 5%. Suspended solids were measured by dry weight according to the procedure of
W (APHA 1989), but 0.2 pm membrane filters were substituted for glass
fiber filters in the filtration step.

4.2.7 Identification and quantification of filaments

Identification of filaments was performed according to the identification keys of
Eikelboom and van Buijsen (1981) and Storm and Jenkins (1984). Filamentous organisms
were quantified by total extended filament length (TEFL), as Sezgin and Jenkins (1978)
suggested.

4.2.8 Rotifer enumeration

Rotifers are distinguished from protozoa by their size (100~500 pm) and
morphological characteristics (Calaway 1968). To enumerate rotifers, a 10 uL cell
suspension was evenly spread over an area of 1.5 cm2 on a glass slide and examined under
100x magnification by a phase contrast microscope (Olympus BH2) equipped with
micrometer scale in ocular. The numbers of rotifers present in 0.5 cm2 viewing area were

counted. The number of rotifer in 1 mL sample (#lmL) was calculated and corrected for

75

viewing area and dilution factor. For triplicate samples, this method gave an average
coefficient of variation of 30% for the number of rotifers in a sample containing 1,000 to

10,000 per mL.
4.2.9 Enumeration of phenol utilizers and TCE degraders

A most probable number (MPN) procedure was developed to estimate the number
of phenol and TCE degraders. Phenol utilizers were identified by their ability to grow on
phenol, and TCE degraders were identified by their ability to degrade TCE after growth on
phenol. It was assumed that all TCE degraders are phenol utilizers. Some support for this
assumption was obtained in preliminary test results: phenol-grown cells degraded TCE, but
catechol-grown cells from the same inoculum did not. To enumerate cells by MPN, 10
milliliters of cell suspension were removed from the reactors at the end of the phenol
feeding period and briefly sonicated (30 sec, 25 watt) to disperse floc and to dislodge
attached cells from the filaments (Banks and Walker 1977). Dilutions (10'3 to 10's) of the
dispersed cell suspension were then prepared in mineral medium. A one milliliter inoculum
was added to a 20-mL glass vial with 9 mL of phenol feed medium (phenol conc. = 100
mglL). TCE was added to give 1 mg/L in the aqueous phase. Vials were sealed with
Teflon-coated septa and incubated at 21 °C on a rotary shaker at 120 rpm. Control vials
were prepared with sterile medium. After 10 days, optical density at 600 nm, phenol
concentration, and TCE concentration were measured. A vial was scored "phenol positive"
when OD5oo > 0.1 or more than 50% of the phenol had disappeared. A vial was scored
"TCE positive" when the remaining TCE in vials was over 50% less than that in the control

vials.
4.2.10 Pure culture isolation and characterization

Ten milliliters of samples were collected from each of the four reactors on day 425.
The samples were diluted in phosphate buffer (KzHPO4, 1.0 g and KHzPO4, 0.75 g; pH

76

= 7.2). Ten-fold dilutions were prepared in a total volume of 10 mL. The samples were
vortexed for five minutes before diluting. The samples were also vortexed for one minute at
each dilution step before transfer. The dilutions were plated on R2A agar and incubated for
four days at 25°C. Representatives of the various morphologies were counted, picked, and
transferred to R2A plates for isolation. To ensure the purity of each isolate, colonies were
transferred five times before REP-PCR was performed. Each isolate with a different REP-
PCR pattern was regrown in 20-mL glass vials with 10 mL mineral medium containing 100
mg/L phenol and 1 mg/L TCE for 10 days. Increase in optical density at 600 nm over the
incubation period was used to determine whether the isolate can grow on phenol. TCE
cometabolism was evaluated by monitoring TCE removal. For the phenol and TCE

degradation assays, each isolate was regrown and tested in five vials.
4.2.11 Assay for catechol ring fission pathway

The production of the yellow product, a-hydroxymuconic semialdehyde (a-
HMS), from catechol was used as a test for the presence of a catechol meta ring fission
. pathway (Folsom et al., 1990; Gerhardt 1994). Ten milliliters of cell suspension was
concentrated to two milliliters (~ 5000 mg/L) by centrifugation at 4,000 rpm. Concentrated
cells (0.5 mL) were resuspended in 2 mL of 0.2 M Tris buffer (pH = 8), supplemented
with 0.5 mL toluene to solubilize the cell membranes, and shaken with 0.2 mL of a 1.0 M
catechol solution. Appearance of a yellow color within a few minutes is indicative of meta

cleavage activity.

Production of B-ketoadipate from catechol was used as a test for the presence of a
catechol ortho ring fission pathway (Folsom et al., 1990; Gerhardt 1994). Ten milliliters
of cell suspension was concentrated to 2 mL (~5000 mg/L) by centrifugation at 4,000 rpm.
Concentrated cells (0.5 mL) were resuspended in 2 mL of 0.2 Tris buffer (pH = 8),
supplemented with 0.5 mL toluene to solubilize the cell membranes, and spiked with

catechol (1.0M, 0.2 mL). After 1 hr of incubation at 30°C, the following chemicals were

77

added: 1 g of (NH4)2SO4, 1 drop of 1% sodium nitroprusside (nitroferricyanide) and 0.5

mL of ammonia solution (28 to 30%). A purple color indicates ortho cleavage activity.
4.3 Results and Discussions
4.3.1 TCE transformation

The reactor communities displayed significantly different TCE removal rates during
the period of operation (Figure 4-2). TCE degradation rates in reactor P increased over
time. For the initial 75 days, kc' was as low as 0.00535zt0.00108 L/mg-day. From day 75
to 220, kc' doubled to 0011041000101 Umg-day. After day 250, removal rates increased
tenfold to kg: 01134100260 L/mg-day. At this point, saturation kinetics (kc' =
010361003174 mg TCFng cell d.w.-day and K, = 0.35 +/- 0.08 mg/L provided a more
accurate description of TCE transformation kinetics. Degradation rates observed after day
250 were comparable to rates of methanotrophic consortia (Arciero et al., 1989; Henry and
Grbic-Galic 1990). During the period of increasing TCE transformation rates, biomass

concentration did not increase.

For reactor C, during the initial 25 days, cells exhibited moderate TCE degradation
rates (kc’= 0.01349i0.004398 Umg-day). After 25 days, transformation rates decreased,
with kc'=0.00358610.001817 L/mg-day, and rates remained low for the remainder of the
operating period. The rates ultimately observed in reactor C were far below than those
observed in reactor P. From days 25 to 40, reactor C changed color from white to yellow,
and biomass decreased from 750 mg/L to 250 mg/L. The yellow color may be due to the
accumulation of a-hydroxymuconic semialdehyde (a-HMS), a yellow intermediate of
phenol degradation. After two more weeks operation, the yellow color disappeared. The
yellow color was not removed over a 3 day period when a filtered sample of reactor C was
incubated with cells drawn from other reactors (reactor P, SC2 and SC5). Brief

interruptions in aeration to reactor C also triggered appearance of a yellow color. These

78

 

 

 

 

 

 

 

 

 

 

 

 

 

9, 0'03 ReactorC
a 0.025
0.02 I
E 0.015
53 o... r
a? 0.005 I I I
0 ‘ffi'l - gnu-ugh. .. .
0 so 100 150 200 250 300 350 400
$2 0'03 Reactor-SCS
‘3 0.025 f I———-—-l
0.02-l
90
i 0015+- '
0.014-
- H
or .L A '3'. .:'l ill-:Jl:
0 50 100 150 200 250 300 350 400
’3‘. 0.03 R SC2
'6 0.025 "m"
0.02
I
g 0.015 I I
0.01 ’.I . I I... I I
. I I
A, 0.005 I. . I'-. I'I
0 . : : : l. : : :
0 50 100 150 200 250 300 350 400
1
A ReIctorP
>5 I . .
‘6 0.1 I.- ' HI '
g 0.01 I' I" II I
a.» ‘ I '
0.001 . : : i 1* : : ;
0 50 100 150 200 250 300 350 400

Figure 4—2 Second-order TCE removal coefficients for cells from different feeding

pattern reactors

79

observations suggest that the yellow color resulted from incomplete oxidation of phenol.
Appearance of the yellow color correlated with decreases in cell concentration and TCE

transformation rate, possibly indicating some form of product toxicity.

For cells harvested from reactor SC2, kc' for TCE removal varied from 0.015 to
0.005 Umg-d (average 0.011 L/mg-d). Changes in TCE removal rates (kc' s) roughly
correlated with changes in biomass concentration. For organisms from reactor SC5,
removal of TCE was optimal from day 10 to day 35 with kc'= 00211921000266 L/mg-
day. After day 35, TCE degradation rates and biomass levels decreased. After day 80, kc'
decreased to 0.002625:1:0.000906 Umg-day, and biomass concentration varied cyclically
from 500 mg/L to 1000 mg/L. TCE transformation rates were low and maintained at 0.002
to 0.003 L/mg-day after day 80. Second order rate coefficients for reactor SC5 were

similar to those of reactor C.

Typical TCE degradation curves for each reactor community after day 250 are
illustrated in Figure 4-3. TCE was transformed at high rates by cells from reactor P, at
intermittent rates by cells from reactor SC2, and at low rates by cells from reactors C and

SC5.
4.3.2 Biomass variation

As shown in Figure 4-4, total biomass concentrations remained relatively stable in
reactors P (suspended solids concentration of 9131126 mg/L) and SC2 (suspended solids
concentration of 990:1:170 mglL). In contrast, significant fluctuations were observed in
reactors C (suspended solids concentration range of 300 to 1400 mg/L, with an average of
835:1:248 mg/L) and SC5 (8501200) Reductions in the concentration of suspended solids
were accompanied by decreased TCE degradation rates in reactors C, SC2 and SC5.
Possible explanations for these differences include toxicity of phenol degradation

intermediates or bacterial predation by protozoa and rotifers. Evidence that predation was

80

 

TCE (mg/L)

 

 

 

 

Time (hr)

Figure 4—3 TCE degradation observed in batch experiments using cells from
different feeding pattern reactors

81

 

1500 rReactor C I

 

 

 

 

 

 

E. 1250 .l. I I
1000 I I '-
v ~' I. I . I
3 500 I. I I.
r 250 I'
“'5 0. : : : : 'r : : :
0 50 100 150 200 250 300 350 400 450
[Eeactor £25]
A 2“” 7
3 1750 I
a 1500 I .
v 1250 I I
. I I I
3 1% h.‘ I. II .' l'.I- I III . I I.‘
500 . I I II
E’ 250
'° 0. : : : .- : : : 5
0 50 100 150 200 250 300 350 400 450

IEeactor EEZI
1500 * f

 

 

 

 

 

 

A
I
31250 I .- I
I If
51w) ‘ fl -. ~*.II-‘ I'.. I I'-
. 750 ' I
‘3’ 500
r 250
'° 0. : ‘ : : : : . :

0 50 100 150 200 250 300 350 400 450
12007 eactor
£1000 'IIII- . “ .IIfl I P

E 800 w I -' -' -'II .
v - I ~
.3 600
I 400
E‘ 200
-e

0. : 4 : 3. : J. J. :

0 50 100 150 200 250 300 350 400 450

Time (days)

Figure 44 Changes of biomass during reactor operating period, each sample was taken at
the end of growth substrate fill period.

82

important was subsequently obtained by microscopic analysis. High levels of protozoa and

rotifers were observed in reactors SC5 and C.
4.3.3 Phenol degradation

Results of the catechol ring fission assay indicate that phenol degradation proceeded
by the meta-cleavage pathway. The occasional appearance of a-HMS in the reactors
supported this observation. Phenol concentrations in the continuous-fed reactor (reactor C)
and the two semi-continuously-fed reactors (SC2 and SC5) were below the phenol
detection limit (below 0.5 mglL). In the pulse-fed reactor (reactor P), phenol concentration
decreased from 200 mg/L to 0 mg/L in 2~6 hours after the pulse. On day 30, phenol
degradation in reactor P was zero-order in phenol concentration (Figure 4-5) with a rate
constant of 0.051 :t 0.008 hr'1 and r2 > 0.99 (Table 4-1). After day 250, the rate constant of
phenol degradation in reactor P had increased to 0.135 t 0.010 hr'l. The zero-order pattern
of phenol utilization in reactor P was similar to the results of Tischler and Eckenfelder
(1969). Substrate inhibition kinetics, such as those reported by Grady (1985), were not

observed.

Repeated spikes of phenol into reactor P did not affect the ability of the community
to degrade phenol, in contrast with the results of Okaygun et al. (1992). In their studies,
each spike of phenol delayed the degradation of subsequent phenol spikes (Okaygun and
Akgerman 1992) Apparently, the pattern of phenol addition in reactor P selected for
organisms with high phenol utilization rates and resistance to phenol toxicity. Phenol

degraded repeatedly without any detectable inhibition.
4.3.4 Phenol utilization patterns for reactors with different feeding pattern

Control of phenol feeding pattern appeared to select for cells with higher specific
substrate utilization rates. On day 155, 100 mL of culture was removed from each reactor,

spiked to an initial phenol concentration of 100 mg/L, and monitored for phenol utilization.

83

Table 4-1. Curve fitting results for estimating zero-order phenol degradation kinetic
coefficients.

 

X0 = initial microorganism concentration, mg/L.
So: initial phenol concentration, mg/L.

n = sample number in curve fitting procedure.
Data obtained on days 31 to 38.

 

 
  
 
 
 
 
 
  

 

250
‘ 1st Pulse
0
200 it 2nd Pulse
I 4th pulse
A A Pulse
3 150 .. 7‘“
E O
V 0 8th Pulse
.5 .
s —— zero-Oder kinetics
i 100 7 " predicted result
50 .-
'4 o
0 l l l l .3 .1.

 

 

0 l 2 3 4 5 6
Time (hr)

Figure 4-5. Phenol degradation in reactor P for daily spiking with phenol from day 31
to 38.

84

As shown in Figure 4-6, organisms from reactor P degraded phenol faster than organisms
from the other reactors. Cells obtained from continuously and semi-continuously fed
reactors exhibited a lag phase and slow rates of phenol degradation. these observations
might be explained by differences in community structure or in the enzymes present. A lag
period for phenol degradation was also observed for cells harvested from reactor P when
the spiked phenol concentration was greater than 400 mg/L. In general, a lag period
occurred whenever cells were exposed to phenol concentrations that were significantly

greater than the levels to which the cells were preadapted.
4.3.5 Effect of phenol feeding pattern on reactor pH and dissolved oxygen

There was little difference in the pH values of the four reactor communities. During
startup, pH decreased from 6.9 to 6.3. After 50 days, pH values stabilized as follows:
reactor C - 6.47:1:0.10; reactor SC5 - 6.51:0.06; reactor SC2 - 6.60:1:0.08; reactor P -
6.5 1:009. The pH decrease is probably explained by the acidic character of phenol and its
transformation products. Ammonia-oxidizing bacteria can also contribute to pH reduction
and TCE removal, but their role was apparently minor, as indicated by the low levels of

nitrate (~ 0-10 mg/L) and nitrite (~ 0-3 mg/L) detected in the reactors.

Aeration levels for all reactors sustained dissolved oxygen levels in excess of 2
mg/L at all times. These levels are considered sufficient to avoid oxygen limitation (Bailey
and Ollis 1986). However, variations in phenol feeding patterns did mirror changes in
dissolved oxygen (DO) concentrations. As shown in Figure 4-7, the continuously fed
reactor had the highest and most stable DO of 7.5 mg/L. DO levels in the other reactors
varied from 7.5 to 2.0 mg/L depending on the operating mode. The lower DO

concentrations were observed during periods of phenol feed.

85

 

 
 
  
  
 
  
 
 

X X I X
+ reactor C

a —I_ Reactor SC5
E, + Reactor SC2
8

.3 "'—.— Reactor P

a:

X Control

     

 

 

 

 

Time (hr)

Figure 4-6 Effects of phenol feeding pattern on phenol degradation in batch system.
Experiments were conducted in 250 mL bottle filled with 100 mL of cells obtained from
each reactor at the end of feeding stage on 3-24-94. Control bottle contained mineral
medium only. Initial suspended solid concentration in each bottle was adjusted to

1000150 mg/L.

86

 

( Reactor P )
10

 

 

 

DO (mg/l)

CN§O~®

 

 

 

( Reactor SCZ )
0

l

 

Donne/I)
cubaeo

°-

 

 

l l
T I I I V T I

4 8 12 16 20 24 28 32 36

34
t

48

 

( Reactor SOS )

 

 

 

 

 

 

 

 

O
.1.-

 

11-
q-
I

24 28 32 36 40 44 48
Time (hr)

0
a
00-0-
fl
N
H
O
N
C

Figure 4—7 Effect of feeding pattern on dissolved oxygen concentration

87
4.3.6 Microscopic inspection

The microbial community of all four reactors contained bacteria, fungi, protozoa and
rotifers. The microbial community of reactor P was predominantly flocculent; those of the
other reactors were predominantly filamentous. Shifts in community structure were

established by monitoring functional changes and changes in the morphology of dominant

species.

Prior to day 35, the dominant species in reactor SC5 were unattached 1-2 pm motile
gram negative rod-shaped bacteria. After day 80, the dominant organisms were filaments
with an average length of 500-900 um, 2-4 pm in diameter with visible septa and true
branches (Figure 4-8). Some bacteria were attached to the filaments. During the transition
period (days 35 - 80), flocculate biomass appeared. For reactors C and SC2, a similar
morphological change was observed. A color switch occurred in reactors C and SC2.
Reactor C changed from white to yellow then back to white. In reactor SC2, the color
switched from straw yellow to white, and floc structure changed from flocculent to
filamentous (from day 35 to day 40). After day 50, reactors SC5, C and SC2 were
dominated by filamentous microorganisms (Figure 4-8 ~ 4-10). The filamentous organisms
were identified as fungi. The filaments present in reactors were comprised of real branches,
straight or slightly bent, > 200 um in length, sheathless with visible crosswalls and
containing rectangular cells, > 2.5 pm in diameter, with a negative Neisser stain. The
prevalence of fungi might have been due to the slightly acidic conditions. In reactor P,

fungi were much less common, and a distinctive floc structure was observed (Figure 4—11).

Fungi were quantified by measuring the total extended filaments length (TEFL). As
shown in Table 4—2, the TEFLs for reactors ranged from 1.7 x 107 to 2.21 x 103 um/mg of
dry weight. Reactor C and SC5 had 10 times higher TEFL values than reactor P. The ‘
estimated mass filaments comprised 70% or 95% of the total biomass in reactor C and

88

 

Figure 4-8 Microphotograph of sludge in reactor SC5

as, .

% WW“W~MW
A -.

it. "a.

 

Figure 4-9 Microphotograph of sludge in reactor C

89

 

Figure 4-11 Microphotograph of sludge in reactor P

90

Table 4-2. Results for quantification of filaments and rotifers in reactors (sample was
collected on day 275 and day 290)

Parameter eactor P eactor eactor eactor

“if 11:11 ent, 1.9 +/. 0.5 x 104 1.7 +/— 0.7 x 105 2.2 +1. 0.6 x 105 2.87 +1. 0.7 x 105

L; magi: 1.7 +1. 0.5 x 107 8.03 +1. 4.2 x 107 2.21 +/- 1.02 x 103 1.7 +/- 0.5 x 108

arnent en ,

 

 

 

 

 

 

 

 

 

pm 900 +I- 400 500+l- 400 1000 +/- 700 600 +/- 300
mglmotgis 5'10 30-50 70-95 60-95
biomassa
5:25: 820 +/- 350 2500 +/- 750 4100 +/- 1300 11600 +/- 2100

a The % of filaments in total biomass was estimated using culture dry weight and an
estimation for fungal dry weight assuming filament diameters of 2 - 3 pm and density of 1

g/L.

SC2, 30-50 % of the total biomass in reactor SC2, and 5-10 % of the total biomass in
reactor P.concentration, and cyclic changes were observed. Rotifers, principally of the
order Bedellodia, were present in a relative stable number (1,000 to 100,000 per mL)
' depending on the feeding pattern. As indicated in Table 4-2, rotifers were present in much
higher concentrations in reactors C, SC2 and SC5 than in reactor P. This might explain the
biomass stability of reactor P in comparison with reactors C, SC5 and SC2. The stable
suspended solids concentration of reactor P might be attributed to a temporal substrate
gradient flavoring flocculent (bacterial) growth over filamentous growth or to the inhibitory

action of high phenol concentrations on predators.
4.3.7 Enumeration of phenol utilizers and TCE degraders

Most Probable Number (MPN) values for phenol utilizers ranged from 2.2 x 106 to
60 x 106 mL‘1 for all reactors, and the MPN values for TCE degraders ranged from 0.34 x
106 to 40 x 10‘5 mL'1 (Table 4-3). The percentage of TCE degrading phenol-degraders

Protozoa and rotifers were present in all reactors. Their existence might be responsible for

91

the fluctuation of biomass and TCE transformation. The dominant protozoa were ciliates
and their population varied (102 ~ 107 per mL) over the period of operation. The number
of ciliates appeared to be inversely proportional to total suspended solids ranged from 7 to
70%. Reactor P had the highest number of phenol utilizers and TCE degraders (~10-100
time higher than other reactors). Reactors C, SC2 and SC5 were not significantly different,
with approximately 2 x 10‘5 mL'1 phenol utilizers and 0.3 x 105 mL'1 TCE utilizers. MPN
enumeration results were consistent with the measured TCE transformation rates. Reactors
with highest TCE transformation rates also had highest concentration of TCE degraders.
The MPN method provided an approach for estimating the number of cometabolizing
organisms, but the method was time consumed and required exhaustive effort to draw
statistically valid conclusions. For samples with same order of magnitude MPN value, each
dilution required 10, 20 or even 100 tubes to provide statistically distinguishable results.
MPN measurements for reactor C, SC2 and SC5 were indistinguishable in the typical 5
tube counting procedure recommended by W (APHA 1989). A further
disadvantage of the MPN method was the need for a completely dispersed sample for

dilution. This was problematic for filamentous or flocculent growth.

Table 4-3 Numbers of phenol and TCE degraders in mixed liquid from different
feeding pattern reactors (sampling on day 192)

 

 

 

MPN (cells/g of dry wt.)
Reactor Phenol degraders TCE degraders
Cell no. 95 % CIa Cell no. 95 % CIa

 

P 6.7 x 1010 2.2 x 1010 - 2.0 x 1011 4.5 x 1010 1.5 x 1010 - 1.4 x 1011

C 5.4 x 109 1.8 x 109 - 1.6 x 1010 1.7 x 109 5.4 x 108 - 5.1 x 1010
SC2 3.3 x 109 1.1 x 109 - 1.0 x 1010 2.2 x 108 7.3 x 107 - 6.6 x 108
SC5 3.3 x 109 1.1 x 109 - 1.0 x 1010 4.5 x 103 1.5 x 107 - 1.4 x 109

aCI, confidence interval.

92
4.3.8 Pure culture isolation and characterization

Fifty bacterial isolates with distinct Rep-PCR patterns were obtained. Reactors P
and SC2 reactors appeared to have more diverse microbial communities than reactors C and
SC5 reactors (Table 4-4). Only 13 isolates possessed the ability to degrade phenol. Of the
phenol utilizers, 10 were also able to cometabolize TCE. Three species from reactor P (P-4,
P-21 and P-23) had the highest TCE transformation rates. The three TCE-degrading
isolates from SC5 and C (C-1, C-2 and SC5-l) grew slowly in phenol medium and
transformed only 10~15% of the TCE present in the TCE transformation assay. Isolates C-
l and SC5-1 are similar in Rep-PCR pattern as well as in phenol and TCE degradation rates
and cell morphology (filament). Similar REP-PCR pattern isolates among reactors are rare.
As indicated in Figure 4-12, the REP-PCR patterns for isolates from reactor C and P are

entirely different.

Characterization of isolates from the four reactors indicates that the microbial
communities in the four reactors are different and complex. As summarized in Table 44
and Figure 4-13, only 20% of the community members were capable of transforming TCE
(10 out of 50). Even the ability to utilize phenol was uncommon (13 out of 50). Over 75 %
of isolates were unable to grow with phenol as the carbon source in mineral medium.
Several explanations are possible for this observation. It is possible that organisms that
were unable to degrade phenol in the assay used in this work might still be able to degrade
phenol under other assay conditions, for example if growth factors provided by other
community members were available, or if the phenol concentration used in the assay were
reduced. The phenol concentration used might have been excessive for some isolates,
especially those from reactors C and SC5 because these communities were not adapted to
high concentrations of phenol. Finally, it is possible that some of the isolates were unable
to use phenol because they use other substrates for growth, such as intermediates in the

phenol degradation pathway or substances released by cell lysis.

93

Table 4-4. Characteristics of isolates from different reactor communities.

Reactor isolate

Continuous

<
<
<
<
<
<
<
<
<

 

94

Table 4—4. (cont'd)

isolate utilization

+

+(yellow)

SC5

 

a. Phenol utilization and TCE degradation assays were conducted in 20 mL glass vials.
Each vial filled with 10 mL phenol medium (phenol concentration = 80 mg/L) and
spiked to 1.0 mglL of TCE. One loopful inoculum was transferred from R2A agar to
vials and incubated on a 120 rpm rotatory shaker under room temperature for 10
days. Phenol utilization ability was determined by measuring the changes of optical
density before and after incubation.

b. TCE degradation capacity was determined by monitoring TCE disappearance in a 10-

day incubation period.

. Each isolate was tested in triplicate.

. Dominant colony morphology.

e. Similar REP-PCR patterns for isolates between reactor: C-1 and SC5-1, C-4 and SC5-
2, C-8 and SC2-1.

0.0

95

.338 e. _

.338 e. _

 

Figure 4-12 REP-PCR patterns of isolates from reactor P and C

96

 

 
  
  
  
  
   

100

90 i.

80 db +ReactorP
7O .. ""9"" Reactor SC2

60 ‘- --+--ReactorSC5
50 J-

40 ..

30 ‘r

20 T

10 ..

0 I .

O 0.2 0.4 0.6 0.8 1
Cumulative Frequency

—-+--ReactorC

% TCE degradation

 

 

 

Figure 4-13 TCE transformation capacity profiles of isolated from different feeding
pattern reactors.

97

The results of the isolates studies were consistent with measures of community
function, such as TCE transformation and phenol utilization. The microbial communities of
reactors P and SC2 exhibited higher rates of TCE and phenol degradation than reactors

SC5 and C; the same pattern was observed in the bacterial isolates.

We conclude that communities enriched with continuous or protracted feeding
intervals (reactor P and SC5) exhibited limited long-term capacity for TCE transformation.
Microbial communities enriched under conditions of pulse or abbreviated feeding intervals
(reactors P and SC2) maintained higher TCE transformation rates. An unexplained increase
in TCE transformation rates was observed in reactor P. In addition, cyclic changes in TCE

transformation rate and biomass were observed in reactor SC2.

4.4 Conclusions
Based on the above results, the following conclusions can be made

1. Growth substrate feeding patterns can have a profound effect on the selection of
microorganism types and concentrations, and can dramatically change the cometabolic

capabilities of a microbial community.

2. During start-up, conditions in reactors C and SC5 were initially favorable for the
selection of TCE degraders. However, this feeding pattern also selected for filamentous
microorganisms. Loss of TCE degradation activity during long term operation was

probably due to the outgrowth of fungi which did not degrade TCE or did so slowly.

3. The pulse feeding pattern resulted in a more stable biomass concentration and a higher
long-term capacity for TCE degradation.

98

4. The feeding pattern affected the reactor substrate level, DO profile, prevalence of
predators and biomass yield. All of these factors may contribute to the final community

structure and function .

5. Complex communities formed in the phenol fed reactors: many of the isolates (~70%)
failed to degrade both phenol and TCE under the conditions of the assay. The pulse fed

reactor exhibited higher diversity and had isolates with higher TCE transformation rates.

literatures cited

APHA.

 

Washington, D. C.: APHA, 1989.

Arciero, D., T. Vannelli, M. Logan, and A. B. Hooper. “Degradation of trichloroethylene
by the ammonia-oxidizing bacterium Nitrosomonas europaea.” W
Comm 159 (1989): 640—643.

Bailey. J. E. and D. F. Ollis. W5 2 nd ed., New
York: McGraw-Hill, Inc., 1986.

Banks, C. J. and I. Walker. “Sonication of activated sludge flocs and the recovery of their
bacteria on solid media.” WM 98 (1977): 363-368.

Calaway, W. T. “The metazoa of waste treatment processess - rotifers.” W
40 (11 1968): R415.

Dabrock, B., J. Riedel, J. Bertram, and G. Gottschalk. “Isopropylbenze (cumene) - a new
substrate for isolation of trichloroethene-degrading bacteria.” Archilgs gf Microbiology
158 (1992): 9—13.

Eikelboom, D. H. and H. J. J. van Buijsen. WW.
Deft, Neth.: TNO Res. Inst. Environ. Hygiene, 1981.

99

Folsom, B. R., P. J. Chapman, and P. H. Pritchard. “Phenol and trichloroethylene

degradation by Pseudomonas cepacia G4: kinetics and interactions between substrates.”

mm 56 (1990): 1279-1285-

Fox, B. G., J. G. Borneman, L. P. Wackett, and J. D. Lipscomb. “Haloalkane oxidation
by the soluble methane monooxygenase from Methylosinus trichosporium OB3b:

mechanism and environmental implication.” Biochemistry 29 (27 1990): 6419-6427.

Gerhardt. P.. ed. W- Washington. DC:
American Society for Microbiology, 1994.

Gossett, J. M. “Measurement of Henry's law constants for C1 and C2 chlorinated

hydrocarbons,.” WM 21 (1987): 202-208.

Grady, C. P. L. Jr. “Biodegradation: Its measurement and Microbiological basis.”

Wm 27 (May 1985): 660—674.

Henry, S. M. and D. Grbic-Galic. “Effect of mineral media on trichloroethylene oxidation
by acquifer methanotrophs.” W 20, (1990): 151-169.

Hopkins, G. D. and P. L. McCarty. “Field evaluation of in situ aerobic cometabolism of
trichloroethylene and three dichloroethylene isomers using phenol and toluene as the

primary substrate.” W 29 (6 1995): 1628-1637.

Maltoni, C. and G. Lefemine. “Carcinigenicity bioassays of vinyl chloride.” W
7 (1974): 387-398.

Nelson, M. J. K. , S. 0. Montgomery, W. R. Mahaffey, and P. H. Pritchard.

“Biodegradation of trichloroethylene and involvement of an aromatic biodegradative

pathway.” MW 53 (5 1987): 949-954-

100

Okaygun, M. S. and A. Akgerman. “Microbial dynamics in a continuously stirred tank
reactor with 100% cell recycle.” W 64 (6 1992): 811-816.

Oldenhuis, R., R. L. J. M. Vink, D. B. Janssen, and B. Witholt. “Degradation of
chlorinated aliphatic hydrocarbons by Methylosinus trichosporium OB3b expressing
soluble methane monooxygenase.” AW 55 (11 1989): 2819-2826.

Sezgin, M. and D. Jenkins. “A unified theory of filamentous activated sludge bulking.” L
m (Feb. 1978): 362-381.

Storm, P. F. and D. Jenkins. “Identification and significance of filamentous

microorganisms in activated sludge.” 1m 56 (5 1984): 449-458.

Tsien, H. C., G. A. Brusseau, R. S. Hanson, and L. P. Wackettet. “Biodegradation of
trichloroethylene by Methylosinus m'chosporium OB3b.” AW 55 (12
1989): 3155-3161.

Vanilli, T., M. Logan, and A. B. Hooper. “Degradation of halogenated aliphatic,

compounds by the ammonia-oxidizating bacterium Nitrosomonas europaes.” AppL

W 56 (1990): 1169-1171.

Vogel, T. M. and P. L. McCarty. “Biotransformation of tetrachloroethylene to

trichloroethylene, dichloroethylene, vinyl chloride, and carbon dioxide under methanogenic
conditions.” WM 49 (1985): 1080-1083.

Wackett, L. P. and D. T. Gibson. “Degradation of trichloroethylene by toluene
dioxygenase in whole cell studies with Pseudomonas putida F1.” AW
W54 (1988): 1703-1708.

Westrick, J. J ., J. W. Mello, and R. F. Thomas. “The groundwater survey.” LAmJEam
Wm 76 (1984): 52.

CHAPTER 5
EFFECTS OF TCE TOXICITY ON CELL VIABILITY

5.1 Introduction

Aerobic TCE biodegradation can be mediated by methanotrophs (Alvarez-Cohen
and McCarty 1991a; Henry and Grbic'-Ga1ic' 1991; Oldenhuis et al., 1989; Tsien et al.,
1989), propane oxidizers (W acket et al., 1989), ethylene oxidizers (Henry and Grbic'-
Galic' 1991), toluene, phenol, or cresol oxidizers (Folsom et al., 1990; Harker and Kim
1990; Shields et al., 1986), and ammonia oxidizers (Hyman and Arp 1992; Hyman et al.,
1995; Rasche et al., 1991). These organisms possess catabolic oxygenases with broad
substrate specificity and can catalyze fortuitous oxygenation of TCE when stimulated by

their respective substrate.

Regardless of the microorganism type, certain features of oxygenase-mediated
cometabolism must be considered in any engineered system: competitive inhibition,
intermediate and product toxicity, and reductant supply. For oxygenase systems, TCE and
the growth substrate are competitive inhibitors at the active cite of the oxygenase. For
methanotrophs, TCE concentrations in the range of 50 mg/l are toxic, inhibiting TCE
transformation and methane utilization (Henry and Grbic'-Galic' 1991; Oldenhuis et al.,
1989). In pure and mixed cultures studies with methanotrophs, Henry and Grbic'-Galic'
(1991) observed that oxidation of 6 mg/L TCE reduced subsequent methane utilization and
decreased the number of viable cells in the pure culture by an order of magnitude. They
hypothesized that toxicity was caused by the attack of reactive intermediates on cellular

macromolecules, not from TCE itself or its degradation intermediates.

101

102

In evaluating TCE oxidation by Pseudomonas putida Fl, Wacket and Gibson
(1988) found evidence of toxicity and proposed that a toxic intermediate was responsible
(Wackett and Gibson 1988). Wacket and Householder (1989) concluded that the toxic
effects of TCE oxidation in P. putida F 1, such as decreased growth rate and increased
death rate, depended upon metabolic activation of TCE by toluene dioxygenase. In E. coli,
TCE exposure led to increased doubling time and induction of stress proteins (Blom et al.,
1992). The TCE concentration causing a 50% inhibition of growth (IC50) for
Nitrosomonas, methanogens, and aerobic heterotrophs were 0.81 mg/L, 13 mg/L and 130
mg/L, respectively (Blum and Speece 1991). By comparison, LC50 values for fathead
mirrows and the Microtox assay were 44 and 960 mg/L, respectively. Apparently, TCE is

more toxic to Nitrosomonas rather than to methanogens and aerobic heterotrophs.

Broholm et al. (1990) suggested that competitive inhibition by TCE may reduce
methane consumption rates (Broholm et al., 1990). Another factor affecting rates is
toxicity. To estimate toxic effects, Alvarez-Cohen and McCarty (1991) introduced the idea
of a limited transformation capacity (Tc) by resting cells. The concept is that a given mass
of cells has a finite capacity for transformation of a nongrowth substrate, such as TCE. By
coupling the idea of a finite transformation capacity with saturation kinetics, a kinetic
expression was obtained that could account for product toxicity and limited reductant

supply (Alvarez-Cohen and McCarty 1991b).

In the following section, we evaluate the viability of TCE-amended phenol utilizers.
Viability of cells exposed to TCE is found to be dependent upon TCE degradation and the
microbial species evaluated. Modeling efforts were taken to evaluate the loss of cells during
TCE cometabolism. After selection of a model for estimatiion of the loss of cells, the
required time and growth substrate for cell recharge can be determined and applied to
design of the recharge stage for SBRs. The purpose of the studies was to verify TCE

103

toxicity effects and to explore possible models fort loss of cells during the TCE

transformation.
5.2 Background

The mechanism of oxygenase catalyzed oxidation is similar to that of cytochrome p-
450 catalyzed oxidation: both produce the same reactive species of oxygen (Miller and
Guengerich 1'982). Monooxygenase-mediated TCE oxidation results in epoxide formation,

which is chemically and biologically converted to mineral end products:

Hydrolysis & oxidation 2 CO;

 

C1 C1 Ox enase C\
C: c yg c1 / Cl
C— C
Cl H

 

 

L >
c1 H K ’ {- 3HC1
02 q H20 H20 02
NADH NAD-l- ' '
+H+

TCE epoxide is an extremely short-lived ( ~ 20 sec), highly reactive molecule that is toxic
in mammalian systems. Thus it appears likely that the toxicity to whole cells observed
during TCE transformation results from the epoxide or its degradation products rather than
TCE itself.

Upon transforming 1“C-TCE, methanotrophs generate radiolabeled cellular protein
(Oldenhuis et al., 1991). Wacket (1991) demonstrated l4C incorporation into DNA, RNA,
small molecules and lipids. Soluble methane monooxygenase and ammonia
monooxygenase were radiolabeled during l4C-TCE degradation (Hyman and Arp 1992;
Oldenhuis et al., 1991). These observations suggest that nonspecific covalent binding of
degradation products to cellular molecules causes the TCE-mediated inactivation of cells.

The mechanisms of TCE transformation toxicity can be more clearly understood by

considering the minimal steps involved in a suicidal inactivation of an enzyme, as shown in

Figure 5-1.

104

k
E+I : 131 —->[Ex]—b—>E+x—‘-‘4—>P(stable)
2

k4
EX' . . .
. . non-specific binding
(inactive) to additional proteins
Figure 5-1 The minimal steps involved in a suicidal inactivation of an enzyme (Hyman
and Arp 1991)

The initial reversible binding of the inhibitor (1) to the enzyme (E) to form an
enzyme inhibitor complex (BI) is followed by a catalytic step (EI—>EX) which transforms
the inhibitor to an activated reactive species (X). This reactive species can either bind to the
enzyme to produce a covalently modified inactive enzyme complex (EX') or it can fully
dissociate from the enzyme to give free X and active enzyme. As the free form of the
reactive species diffuses from the active site of the enzyme it may either react with other
cellular components or undergo further transformations to a stable unreactive product (P).
. The ratio of inhibitor activation events to the enzyme inactivation events is known as the
partitioning ratio and is normally defined by the ratio of k3/k4. An ideal suicide substrate
would have a partitioning ratio of 0. TCE can be regard as a nonspecific suicide substrate
so the partitioning ratio should be larger than 0. The activated TCE degradation product
covalently modifies numerous polypeptides or cellular components.

The loss of cells during cometabolism has been modeled by first order decay (Saez
and Rittman 1989), transformation capacity (Alvaez-Cohen and McCarty 1991) and an
unified model proposed by Criddle (1991). The expressions used in these models were
summarized in equations 5-1, 5-2 and 5-3 respectively.

dX
—=—b'X 5-1
dt ( )

105

LC = T. (5-2)
dX

fl = -bX - 3.; x (5'3)
dt T

Where X = active cells, b' = first order cell decay coefficient, include endogenous
decay, and product toxicity (time'l), b = endogenous decay coefficient (time '1). Tc =
transformation capacity (mg nongrowth substrate/mg cell) and qc = specific transformation
rate for nongrowth substrate (time'l). These models are all empirical and can be applied to

estimate the loss of cells during cometabolism.
5.3 Material and methods
5.3.1 Organisms and conditions

Pseudomonas cepacia G4 and Pseudomonas putida F 1 were obtained from the
laboratory of Dr. James Tiedje. Strain P2 was isolated from the phenol pulse fed reactor, as
described in chapter 4. Strain P2 was able to use phenol but was unable to cometabolize
TCE. Strains G4, F1, and P2 were batch cultured in a defined phenol medium. The
composition of the medium was (per liter of deionized water): 100 mg of C6H§OH, 2.13 g
of Nazi-IP04, 2.04 g of KH2PO4, 1 g of (NH4)2SO4, 0.067 g of CaClz 2H20, 0.248 g of
MgC126H20, 0.5 mg of FeSO4 71120, 0.4 mg of ZnSO4 7H20, 0.002 mg of MnC12
4H20, 0.05 mg of CoC12 6H20, 0.01 mg of MCI; 6H20, 0.015 mg of H3BO3, and 0.25
mg of EDTA. The pH of the medium was 6.8. Cells used in toxicity and viability studies

were harvested in the stationary stage.
5.3.2 Assays of TCE and phenol concentrations and cell viability

TCE concentration was assayed by injecting a 1 mL headspace gas sample into a

Hewlett-Packed 5890 gas chromatograph equipped with a 30 m (L), 0.53 mm (ID) D3624

106

capillary column (Alltech No. 93532) and a flame ionization detector. The flowrate of
helium carrier was 12 mIJmin. The GC oven was operated isothermally at 90°C. Sampling
injection port and detector temperatures were both 250°C. Liquid phase TCE concentrations
were calculated assuming equilibrium between TCE concentrations in the gas and liquid
phases. A dimensionless Henry's law constant of 0.333 was determined for the growth
medium at 21°C. The initial concentration of TCE was estimated from triplicate sterile
controls after equilibration. In general, the coefficient of variance of TCE determination
was 1 to 5% for triplicate samples. Phenol was analyzed by the 4-aminoantipyrene method
(Martin 1949).

Viability experiments were conducted in triplicate using a series of lO-fold dilutions
and R2A spread plates. Diluted cell culture was placed onto each plate and incubated for
one week under 21'C. Colony forming units were counted daily until a constant number

was obtained. In general, a 3 to 5 days incubation was required to obtain a constant

reading.
5.3.3 Toxicity effects experiment

One hundred milliliters of cells were grown in a 250 ml glass vessel containing 100
mg/L phenol medium, capped and sealed with a Mininert valve. Periodically, grown cells
were withdrawn and assayed for phenol concentration. As phenol concentration decreased
to nondetectable levels, 10 mL of cell suspension was transferred to a 20 mL glass vial and
spiked with TCE, and sealed with teflon-coated septa. The vials were then incubated at 21
'C in an 120 rpm shaker. After specified incubation periods, cell viability and TCE
concentration were assayed. Two sets of control vials were prepared: TCE-free vials as

controls for cell viability and sterile vials as controls for abiotic TCE losses.

107
5.3.4 TCE transformation capacity measurement

Ten milliliters of cell suspension (biomass concentration = 50 ~ 100 mg dry wt IL)
was spiked with 2 mg TCE sealed with mininert valve in 20 mL vials. Those vials were
incubated at 21°C in an 120 rpm shaker. TCE concentrations were periodically monitored
until TCE degradation ceased. The viable cell concentrations before and after TCE
degradation were measured by spread plating cell suspensions on R2A agar. Controls that
without TCE addition were used to correct for loss of viable cells due to endogenous
decay. The ratio of the degraded TCE to the decreased number of viable cells was the TCE
transformation capacity. For unit conversion, it was assumed I g cell (as dry wt) is

approximately equivalent to 2 x 1012 cell.
5 .4 Results and discussions
5.4.1 Effects of TCE concentration on cell viability

P. cepacia G4 (10 mL of culture at cell concentration of 50 mg/l) were incubated in
the presence of aqueous phase TCE concentrations ranging from 0 to 13 mg/L. Cell
viability after 16 u g of TCE had been transformed. The concentrations of viable cells are
summarized in Figure 5-2. A 56 % to 65% decrease in viable cells was observed for all
TCE concentrations examined. Apparently, for strain G4, cell viability was independent of

the applied TCE concentration over the range examined.
5.4.2 Effects of cell type on viability following TCE transformation

The role of TCE transformation in toxicity was established by comparing TCE
-transforming strains - P. cepacia G4 and P. putida F1, with the phenol-degrading isolate
P2. Strain P2 does not degrade TCE. All cultures were amended with an aqueous phase
concentration of TCE of 2.5 mg/L and incubated for 24 hours. The concentration of viable

cells remaining was determined. Results are provided in Figure 5-3. P. putida F1 was

108

5.00E+08

 

4.50E+08 .. -

4.00E+08

3.50E+08

3.00E+08

2.50E+08

2.00E+08

Number of viable cells after 2
days' incubation (CFU/ml)

l .50E+08

 

 

 

1 .00E+08

0 1.27 3.68 6.35 10.30 12.70
Initial aqueous TCE concentration (mg/l)

Figure 5-2. Concentration of viable P. cepacia G4after exposure to different initial
TCE concentrationsd. represent average viable cell concentration +/- standard
derviation.

l.00E+09

 

1 .00E+08

1 .00E+07

(CFU/ml)

1 .00E+06

 

 

 

Number of viable cells in 24 hours incubation

1.00E+05

.231 "' “rs: N in
’§ :0 “ .310 " 0
9.? 3“ ,3 :1- g
30 8+ "3 3': N
. In
a. “a fi- fi- 9-

a:

Figure 5-3. Concentration of different cell typesafter exposure to 2.5 mg/l of
ueous TCE in a 24 hours incubation period. represents average concentration
of viable cell +/- one standard derviation.

110

highly sensitive to TCE, experiencing a 99.5% decrease in viable cells. P. cepacia G4 was
less sensitive, with a 47% decrease in viable cells. Strain P2 did not undergo an appreciable
loss. These results were supportive of the generally held view that: (l) TCE toxicity is
linked to TCE transformation, inasmuch as strain P2 cells exhibited no loss of cell viability
and (2) TCE toxicity is species dependent. P. cepacia G4 was more resistant to TCE

transformation toxicity than P. putida F 1.
5.4.3 Quantification and modeling of TCE toxicity on P. cepacia G4

The transformation capacity of P. cepacia G4 was also measured to assess decrease
in active organisms during TCE transformation. A Tc of (2.67:0.42) x 10'11 mg TCE / cell
was obtained for cell concentrations ranges from 108 to 109. Assuming 1 g cell dry weight
approximately to 2 x 1012 cells, a Tc of 0.0534 i0.0084 mg TCE / mg cell dry wt. was
obtained. This value is similar to experimental values ( ~ 0.06 ) and is higher than the
reported transformation capacity of methanotrophs (0.036 mg TCE / mg cell (Alvaez-
Cohen and McCarty 1991)) and Nitrosomonas (0.004 mg TCE / mg cell (Hyman 1995)).
A summary of measured Tc for different cultures is provided in Table 5-1. As indicated in
Table 5-1, ( 1) the transformation capacity model adequately predicted the loss of viable cell
during TCE transformation, and (2) phenol enrichment from pulse fed reactor and P.
cepacia G4 have higher TCE transformation capacities compared to methanotrophs, P.

putida F1 and Nitrosomanas.
5.5 Conclusions

For the TCE concentrations range of 1 to 10 mg/L, degradation of TCE causes the
loss of viable cells. Results indicated that TCE transformation resulted in cell death, or at
least a loss of culturability on spread plates. It might be argued that the loss of cell

culturability is the result of damage to critical cell enzyme systems, and that repair of those

111

systems would enable regrowth. In these experiments, however, there was no evidence of

damaged cells, capable of slowly growing on spread plates.

Table 5-1 TCE transformation capacity of different cultures

 

 

 

Culture Tc (mg TCE lmg cella)
phenol enrichment 0.12 +/- 0.015
P. cepacia G4 0.054 +/- 0.008
P. putida F1 0.012 +/- 0.002
methaotrophs 0.03613
Nitrosomonas europaea 0.0040

a mg cell as dry wt

b Alvarez-Cohen and McCarty, 1991
C Hyman et al., 1993

The degree of TCE toxicity evaluated by cell reproduction ability was species
dependent. Two TCE degraders -P. cepacia G4 and P. putida F 1- exhibited different
sensitivity to TCE toxicity. P. cepacia G4 might be advantageous for bioremediation
applications. Death of P. cepacia G4 resulting from TCE transformation was well
modeled by use of a transformation capacity term and was log-linearly related to the
number of cell that were not exposed to TCE. Thus, it appears that modeling and
quantification efforts may be applied to assess the loss of active cells during TCE
cometabolism. To compensate for TCE toxicity, additional growth substrate must be
supplied for regeneration, and models are needed to predict the required time for cell

regeneration or regrowth.

112
literatures cited

Alvarez-Cohen, L. and P. L. McCarty. “TCE transformation by a mixed methanotrophic

culture - effects of toxicity, aeration and reductant supply.” MW 57
(l 1991a): 228-235.

Alvarez-Cohen, L. M. and P. L. McCarty. “A cometabolic biotransformation model for
halogenated aliphatic compounds exhibiting product toxicity.” WW1, 25
(1991b): 1381-1387.

Blom, A., W. Harder, and A. Matin. “Unique and overlapping pollutant stress proteins of

Escherichia Cali.” MW 58 (l 1992): 331-334.

Blum, D. J. W. and R. E. Speece. “A database of chemical toxicity to environmental
bacteria and its use in interspecies comparisons and correlations.” W
63 (3 1991): 198-207.

Broholm, K., B. K. Jensen, T. H. Christensen, and L. Olsen. “Toxicity of 1, 1, 1-

trichloroethane and trichloroethene on a mixed culture of methane-oxidizing bacteria.”

WM 56 (8 1990): 2488-2493.

Folsom, B. R., P. J. Chapman, and P. H. Pritchard. “Phenol and trichloroethylene

degradation by Pseudomonas cepacia G4: kinetics and interactions between substrates.”

WM 56 (1990): 1279-1285.

Harker, A. R. and Young Kim. “Trichloroethylene degradation by two independent
aromatic degrading pathways in Alcaligene eutrophus JMP134,.” W
M91, (1990): 1179-1181.

Henry, S. M. and D. Grbic-Galic. “Influence of endogenous and exogenous electron

donors and trichloroethylene oxidation toxicity on trichloroethylene oxidation by

113

methanotrophic cultures from a groundwater aquifer.” WM 57 (1
1991): 236-244.

Hyman, M. R. and D. J. Arp. “14C2H2- and l"C02-labeling studies of de nova synthesis

of polypeptides by Nitrosomonas eurapaea during recovery from acetylene and light

inactivation of ammonia monooxygenase.” mm 267 (3 1992): 1534-1545.

Hyman, M. R., S. A. Russell, R. L. Ely, K. J. Williamson, and Arp. D. J. “Inhibition,

inactivation, and recovery of ammonia-oxidizing activity in cometabolism of

trichloroethylene by Nitrosomonas europaea.” AW 61 (4 1995):
1480-1487.

Miller, R. E. and F. P. Guengerich. “Oxidation of trichloroethylene by live microsomal P-

450 evidence for chlorine migration state not involving trichloroethylene oxide.”

Wm 27 (5 1982): 1090-1097.

Oldenhuis, R., J. Y. Oedzes, and D. B. Janssen. “Kinetics of chlorinated hydrocarbon by
Methylosinus trichosporium OB3b and toxicity of trichloroethylene.” AW
mgr-ohm. 57 (1 1991): 7-14.

Oldenhuis, R., R. L. J. M. Vink, D. B. Janssen, and B. Witholt. “Degradation of
chlorinated aliphatic hydrocarbons by Methylosinus trichasporium OB3b expressing
soluble methane monooxygenase.” WM 55 (11 1989): 2819-2826.

Rasche, M. E., M. R. Hyman, and D. J. Arp. “Factors limiting aliphatic chlorocarbon

degradation by Nitrosomonas europaea: cometabolic inactivation of ammonia

monooxygenase and substrate specificity.” AW 57 (10 1991): 2986-
2994.

114

Shields, M. S., S. 0. Montgomery, P. J. Chapman, 8. M. Cuskey, and P. H. Pritchard.

“Novel pathway of toluene catabolism in the trichloroethylene-degrading bacterium G4.”

WW1. 55 (6 1986): 1624-1629.

Tsien, H. C., G. A. Brusseau, R. S. Hanson, and L. P. Wackett. “Biodegradation of
trichloroethylene by Methylosinus trichosporium OB3b.” W91, 55 (12
1989): 3155-3161.

Wackett, L. P., G. A. Brusseau, S. R. Householder, and R. S. Hanson. “Survey of
microbial oxygenases: trichloroethylene degradation by propane-oxidizing bacteria.” A991,
W 55 (1 1989): 2960-2964.

Wackett, L. P. and D. T. Gibson. “Degradation of trichloroethylene by toluene
dioxygenase in whole cell studies with Pseudomonas putida F1.” M1199,
W54 (1988): 1703-1708.

Wackett, L. P. and S. R. Householder. “Toxicity of trichloroethylene to Pseudomonas
putida F1 is mediated by toluene dioxygenase.” WM 55 (10 1989):
2723-2725.

CHAPTER 6
EXPERIMENTAL EVALUATION AND NUMERICAL SIMULATION OF
TRICHLOROETHYLENE COMETABOLISM BY A PHENOL-FED ENRICHMENT
IN SEQUENCING BATCH REACT ORS

6.1 Introduction

Many hazardous compounds can be detoxified by cometabolic reactions. Use of
such reactions will likely lead to novel wastewater treatment and groundwater
remediation processes. To date, however, development of engineered systems for
cometabolism has proceeded slowly. In part, this can be attributed to challenges inherent
to cometabolic transformations, including competition for enzyme between the growth
and nongrowth substrates, the requirement for continual or periodic inputs of growth or

energy substrate, and loss of activity due to product toxicity.

Sequencing batch reactors (SBRs) offer several advantages for cometabolic
transformations. An SBR can alternate between periods of growth on growth substrates
and periods of cometabolism of nongrowth substrates, eliminating the possibility of
competitive inhibition for enzyme between the growth and nongrowth substrates. In an
SBR, growth substrates can be added as needed to rejuvenate the cometabolizing biomass
and to select for microbial populations with favorable transformation kinetics and settling
properties. The ability to mix during the fill period is a favorable feature for
cometabolism and for hazardous waste treatment. Chemicals entering during this period

are readily diluted and degraded by a concentrated microbial suspension. This enables

115

116

treatment of contaminants at concentrations that would be toxic in many other reactor

configurations.

For the bench-scale SBR described in this chapter, cometabolic oxygenase
activity was induced by the addition of phenol as growth substrate. Phenol and
trichloroethylene (T CE), the nongrowth substrate, were provided in separate stages,
preventing competitive inhibition. Addition and degradation of phenol occurred during
the "recharge" period, a period that replaces the idle period of the conventional SBR
sequence of operations. Periodic phenol addition rejuvenated and sustained TCE

cometabolism during the fill and react periods.

Design of SBRs for cometabolism is not obvious or straight forward. We adopted
a mechanistic approach using mass balance relationships. In this paper, we illustrate the
development and verification of these procedures for a bench-scale SBR. The proposed
approach is shown to successfully describe cometabolism and volatilization in this

system.
6.2 Modeling considerations
6.2.1 Substrate removal mechanisms

Mechanisms of contaminant removal include biodegradation, air stripping, and
sorption. Not all of these mechanisms are significant for every compound. The relative
importance of the different removal mechanisms depends upon the kinetics of
biodegradation, mass transfer coefficients, Henry's constant, and solid partition
coefficients. As indicated in Table 6-1, for volatile substrates, such as TCE, the major
removal mechanisms in an aerated SBR are biodegradation and air stripping, with minor
removal by sorption. Less volatile substrates, such as phenol, are removed by
biodegradation, and negligible phenol is removal by air stripping and sorption.

Mathematical expressions must accurately account for both physical and biological

117

removal to successfully simulate biomass and substrate concentrations. Development of

these expressions is described in the following sections.

Table 6-1 Biodegradability, Henry's constant and solid partition coefficient of
phenol and TCE

 

 

 

 

Compounds phenol TCE
Readily biode graded in Cometabolism by some
Biodegradability acclaimed activated sludge (l) oxygenase system (2)
Henry's constant
H (L water/ L 515) 1.64 x 10'5 at 25’C (3) 0.33 at 21°C (4)
Partition coefficient in
activated sludge 95 at 25C (5) 295 at 25°C (5)
K, (5%)

 

 

 

 

 

Reference (1) D'Adamo et al., 1984), (2) Oldenhuis et al., 1989; Wacket et al.,
1989, (3) Mackay and Shin 1981, (4) Gossett 1987, (5) Estimate from log K,
=1.14 + 0.58 log K9,, (Dobbs et al., 1989).

Biodegradation

For hazardous contaminants, experimental evaluation of kinetic expressions is
recommended. Batch experiments on the phenol-fed enrichment of the SBR operated in
this study established that removal of phenol was zero-order with respect to phenol

concentration and first order with respect to biomass concentration for phenol

concentrations less than 200 mg/L. Thus, r = -k°Xa, where r, = phenol removal rate

(mg/L-d), Ic° = maximum specific rate of phenol utilization (mg/mg-d) and X,= active
biomass concentration (mg/L). This result is similar to the findings of Thischler and

Eckenfielder (1969) and Rizzuti et a1. (1982).

118

For cometabolism of TCE, saturation kinetics provided the best description of
degradation :

rc = ——chX0 (6'1)

Kc + C
where C = aqueous TCE concentration (mg/L), kc: maximum specific rate of TCE
transformation (mg/mg-d), and Kc: half saturation coefficient for TCE (mg/L). At low

TCE concentration (C << Kc):

rc = -k;CXa (6-2)
where k,’ = second-order rate coefficient (l/mg-d).
Air stripping

Delivery of oxygen by aeration results in the mass transfer of volatile compounds
(TCE, in this case) to the gas phase:

r3' = KLa(C- C“) (6-3)

where C = aqueous phase contaminant concentration (mg/L); C* = aqueous phase
contaminant concentration at equilibrium (mg/L); and K La: overall mass transfer

coefficient (d'l).
Sorption

Sorption of contaminants by microorganisms is a relatively rapid process (Matter-
Muller et al., 1980). For a liquid/solid system at equilibrium, the concentration of a
contaminant on a solid sorbent can be expressed as C, = KPCL. where K, = sorption
partition coefficient (L/kg), C, = concentration of contaminant on the solid (mg/kg) and
CL = concentration of contaminant in the liquid (mg/L). This relationship is generally

valid at the low contaminant concentrations, such as those encountered in wastewater

119

treatment. Sorption partition coefficients can be related to octanol/water partition

coefficients (KW). For activated sludge the following relation was proposed:
log K p = 96710ng — 2. 61 (Matter-Muller et al., 1980) (6-4a)

long = 0. 58log KW + 1.14 (Dobbs et al., 1989) (6-4b)
For a given K, , the percentage of contaminant partitioning into solids (P) can be
estimated by:

K,,X x 1045

P(%)= X
KPXX10‘6 +1

 

(6-5)
where X = suspended solid concentration, mg/L.
6.2.2 Mass balance equations

Equations describing removal of nongrowth substrate were obtained by
performing a mass balance on nongrowth substrate for the fill and react stages. Negligible

nongrowth substrate was removed during the settle, decant, and recharge periods.
Mass balance on TCE

Because TCE is volatile (Hc = 0.33 at 21°C, Gossett, 1987), TCE mass balance
expressions are needed for both the gas and liquid phases. Theoretical considerations
indicate that adsorption of TCE into biomass is insignificant, and this was confirmed

experimentally. The following liquid and gas phase TCE mass balance equations were
derived for the fill period:

dC o o C
(IV C C
‘71,?" = KLa.TCE(CL - 715‘)”. ' Qng (6'7)

where V, = reactor liquid volume (mL) at the beginning of the fill period (t = 0),V, = gas
phase volume (mL), Q, = air flowrate (mL/min), Q0: influent flowrate (mL/min), CL =

120

liquid phase TCE concentration (mg/mL), C0 = influent TCE concentration (mg/mL),

KLam = TCE overall mass transfer coefficient (min'l), C, = gas phase TCE

concentration (mg/mL), and If: fill time (min). Rearranging equations 6-6 and 6-7 gives:

 

461 Q” . C k C x

= C -C -K C __L V -__£_L_a__ 6-8
dt (V0+Q"t )( ‘) Lam“ ’- HC) " (K,+C,) ( )
dC C v C

dt (VT-VL) (VT—VL) '

C

where V, = total reactor volume, including both liquid phase and gas phase (mL), and V,
= liquid volume of reactor (mL). Substituting to = V0 /Q", V,_ = V0 + Q”: and

t, = (VT - V0 )lQ" into equations 6-8 and 6—9 gives:

 

dc, cue, C kCX

dr ro+r La'm( " H.) KC+CL ( )
dC C t H C

_r.=.._r_ / -1 +J_K C ——L 6-11
dt tt-I(Q' Q0 ) tg-t La.TC£( L He) ( )

Assuming that the influent contains an insignificant concentration of cells and that
decay of active cells is first-order with a decay coefficient b, then the biomass mass

balance equation is:

g: -—}£9——bXa (6-12)
dr to + r

Integrating equation 6-12, gives

«(flan

X, = Xwe ‘° (6-13)
where X“ = active biomass concentration at the beginning of fill (mg/L). When bt <<

ln[(to+t)/tol. equation 6-13 simplifies to

X - ---9—X (6-14)

121

For the react stage, mass balance expressions for substrate and biomass in the

liquid phase are:

C k C X
EL=_K C ___£ _L_L£ 6-15
4: La:TCE( L He) KC+CL ( )
X, = Xwe’b' (6-16)

Substituting equations 6-16 into equation 6-15 gives:

dC C kC X e'b'
_L=_ a, (C -_L)__LL—00 (6-17)
dr ‘ T“ L H, K, + C,

For substrate in gas phase, the mass balance is:

dC V C grcr
dt La' TCE ( V: X L Hr: ) Va ( )
Mass balance on phenol

During the recharge period, phenol was fed for a fixed interval (30 minutes) to the
settled biomass. Due to the low Henry's law constant for phenol (~1.64 x 10'5 at 25‘C
(Weast 1992)). removal of phenol by air stripping can be neglected. During recharge, the

mass balance for phenol is:

 

 

d8 5" - S
— = + 6-19
dr r, + r r’ ( )
For biomass,
dX -X
_ = 6-20
dt t + r ’ ( )

After completion of phenol addition, phenol degradation and biomass variation is
described by:

dS
Z=rs (6'21)

4“
—. = 6’ 2

122
6.3 Materials
6.3.1 Chemicals

Trichloroethylene (TCE, 99+%) was obtained from Aldrich Chemical Co.,
Milwaukee, Wis. Chemicals for preparation of media and analyses were ACS reagent

grade from Aldrich or Sigma.
6.3.2 Bioreactor

A 3.7 L water jacketed Pyrex reactor was operated sequentially in fill, react,
settle, decant and recharge modes (Figure 6-1). Mixing during the fill, react and recharge
periods was accomplished using a magnetic stir plate set at 100 rpm. Temperature was
maintained constant at 23 :t 1°C. Concentrated aqueous TCE (concentration = 12.5 mg/L
~ 125 mg/L) was delivered by a syringe pump and in-line mixed with mineral medium
stream that was delivered by a peristaltic pump. The combined flow was then supplied to
the reactor during the fill period. TCE volatilization losses were reduced by introducing
TCE into the reactor below the liquid level. Oxygen was supplied by bubbling air through
the reactor during the fill, react, and recharge periods. Phenol was fed by a syringe pump
during the recharge period. During the decant period, supernatant was withdrawn by a
peristaltic pump. A programmable timer was connected to the power supply to control the
on-off status of pumps, the aerator, and the mixer. The solids retention time of the
enrichment was maintained at 10 days by daily wasting of 10% of the reactor liquid
volume during the recharge period. All tubing and fittings in contact with TCE were
made of tefion, glass, or stainless steel to reduce losses of TCE. Before initiating reactor

operation, all fittings and connections were leak tested by pressurizing the reactor.

123

Minearl Medium Tank

    
 
 
 
 
 
       
     
 

TCE Feed
Syringe
Pyrex-Glass Reactor Vessel

Off-Gas Collection Bed

Headspacc

Sampling Port Medium Pump

   

DO Meter
E o 0

pH Meter

         
    
    
   
 
 

 

 

Liquid
Sampling

        

Power Controller

E

Recycle Pump

Magnetic Stirrer

Flow Meter

Effluent Tank

Water Bath (23 °C)

Figure 6-1 Experimental setup for bench-scale SBR

124
6.3.3 Media and inoculum preparation

The influent media contained phenol, macro and micronutrients, and phosphate
buffer. Each liter of deionized water contained: 5 g of phenol, 2.13 g of NazHPO4, 2.04
g of KH2PO4, 1 g of (NH4)2804, 0.067 g of CaClz 2H20, 0.248 g of Mng 6H20, 0.5
mg of FeSO4 7HzO, 0.4 mg of ZnSO4 7H20, 0.002 mg of MnC12 4H20, 0.05 mg of
C0C12 6H2O, 0.01 mg of NiClz 6H20, 0.015 mg of H3BO3, and 0.25 mg of EDTA.
Variable TCE levels were added to achieve influent TCE concentrations ranging from
0.5 to 5 mg/L, depending upon experimental requirements. The pH of the medium was
6.8.

The reactor was inoculated with organisms (200 mL) from a stable phenol-fed
enrichment. The phenol acclimated enrichment was obtained by fed-batch operation of a
chemostat for 300 days at a hydraulic residence time of 10 days. In this reactor, 200 mL
of medium (phenol concentration = 2000 mg/L) was pulse fed to the reactor daily before
the same amount of mixed cells were removed. Inoculum for the fed batch enrichment
was a sample of return activated sludge from the East Lansing wastewater treatment plant
(East Lansing, Michigan). Microscopic examination of the stable phenol enrichment
revealed a diverse microbial community, including cocci- and rod-shaped bacteria, fungi,
protozoa, and rotifers. A more complete description of this reactor microbial community

is provided in chapter 4.
6.4 Experimental conditions
6.4.1 Reactor operating conditions

Behavior of the SBR can be divided into three distinct operating phases: (1) a
start-up period in which the community adapted to cyclic SBR operation in the absence
of TCE, (2) a period of acclimation to TCE addition, and (3) stabilized TCE removal.

During start-up, TCE was not added for over 100 cycles (2 months). The reactor was

125

operated with 2 cycles per day. Each cycle consisted of a 1 hr fill period, 3 hr react
period, 1 hr settling period, 1 hr decant period, and 6 hr recharge period. After the reactor
had stabilized with respect to biomass and phenol concentrations, TCE feed was initiated
during the fill period. At the beginning of the fill period, the liquid volume was 1250 mL.
TCE and mineral medium was supplied at a flowrate of 20.8 mL/min to give a final liquid
volume of 2500 mL at the end of the fill period. Oxygen was provided by bubbling air
through the reactor at 220 ~ 280 mL/min (fiowrate decreased slightly as liquid level
increased). During the react stage, aeration continued, but no additional influent entered
the reactor. During the settling period, air flow and mixing ceased. After one hour of
settling, a decant pump was activated to withdraw supernatant from the reactor at a rate of
19.3 mL/min. Aeration and mixing resumed during the recharge step. At the beginning of
the recharge period, a 5 gll phenol solution was supplied to the reactor at a flowrate of 1.5
mL/min for 30 min. Mixing and aeration continued for 5.5 hrs to completely remove the
added phenol. The operating pattern and parameters are summarized in Table 6-2 and

Figure 6-2.

TCE, phenol, pH, dissolved oxygen (DO) and mixed liquid suspended solids
(MLSS) were monitored periodically. TCE transformation activity was evaluated using

cells removed for control of the solids residence time.
6.4.2 Batch assay of TCE transformation activity

Batch TCE transformation assays were conducted in 20 mL glass vials sealed
with teflon coated butyl rubber stoppers and aluminum crimp caps. A 5 mL volume of
cell suspension was dispensed into the vial, sealed, and spiked with 10~50 ul of aqueous
TCE stock solution. The vial was incubated in an 120 rpm rotatory shaker under 23°C.
Periodically, 0.1 ml. gas phase samples were withdrawn and injected into a Hewlett-
Packed 5890A gas chromatograph (GC) equipped with a DB624 capillary column

(Alltech No. 93532) and a flame ionization detector. The flowrate of carrier gas, helium,

Table 6-2. Summary of bench-scale SBR operating parameters

 

 

Parameter value
Reamflolume
Total volume, VT 3700 mL

Liquid volume, VL
Headspace volume, V g

1250 mL - 2500 mL
1200 mL - 2450 mL

Initial volume, V0 1250 mL
Elmer.

Influent flow rate, Q0 20.83 mL/min
TCE concentration, C0 0.5 - 5 mg/L
Recharge flow rate, an 1.5 mL/min
phenol concentration, 80 5000 mg/L
Airflow, Qg 220 - 280 mg/L
Minimum oxygen 2 mg/L
concentration rn reactor

Warned:

Fill time, if 1 hr
React time, t, 3 hr
Settle time, ts 1 hr
Decant time, td 1 hr
Recharge time, trc 6 hr
-phenol feed mode 0.5 hr
-phenol react mode 5.5 hr
operating cycle time, t; 12 hr/cycle
Sludge age, SRT 10 day

 

127

TCE

/ .
mf’ Mr
eé

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1
Air I Air
phenol 1 1
Air Effluen
<—t—
Wm Will!

 

 

 

Figure 6-2 Operating mode of bench-scale SBR for TCE cometabolis

128

was 12 mL/min. Oven temperature was 90 °C, and detector and injector temperatures
were 250 °C. TCE concentrations were determined by comparing the peak areas of
samples to the peak areas of standards. The coefficient of variance ranged from 1 to 5%
for triplicate samples. The total sampled volume for each vial was less than 5% of the
total headspace volume. All reported TCE concentrations are aqueous concentrations
calculated under the assumption that TCE concentrations in gas and liquid phases were in
equilibrium. A dimensionless Henry’s law constant of 0.355 at 23 °C was used in liquid
phase TCE calculation (Gossett 1987). The initial mass of TCE added to the vial was
estimated from triplicate uninoculated controls calculated using Henry's law. Preliminary
work (Clowater 1992) established that equilibrium was achieved after 30 seconds of hand

shaking and 3 min incubation in an 120 rpm rotatory shaker.
6.4.3 Measurement of pH, dissolved oxygen, phenol, and MLSS

Culture pH and dissolved oxygen were measured with a pH meter (Orion 720)
and a dissolved oxygen probe (Orion model 97-08), respectively. Phenol was measured
by the 4-aminoantipyrene derivative method (Martin 1949). The detection limit for
phenol was approximately 0.5 mg/L. Coefficients of variance for 5 samples were
typically less than 5%. Suspended solids were measured by the dry weight method of
Standard Methods (APHA 1989) but a 0.2 pm membrane filter was used instead of a
glass fiber filter.

6.4.4 Measurement of TCE in reactor off-gas

A modification of National Institute for Occupational Safety and Health (NIOSH)
method for measuring airborne contaminants was used to measure TCE concentration in
reactor off-gas. NIOSH approved glass tubes containing coconut-shell derived activated
carbon was attached to the off-gas line to capture volatilized TCE. The adsorbed TCE

was extracted with carbon disulfide, diluted into isooctane, and injected into a gas

129

chromatograph (Hewlett-Packard 5890) equipped with a DB5 capillary column (J&W
Scientific). The flowrate of carrier gas (nitrogen) was 1.5 mL/min. Detection was
accomplished with a “Ni electron-capture detector. The GC oven temperature program
was: 1 min at 45°C increasing at 5°C/min to 70°C with a final holding period of 1 min at

70°C. Injection port temperature was 250°C, and detector temperature was 350°C.

Known quantities of TCE were injected onto activated carbon in off-gas
collection tubes to determine TCE recovery efficiency. 95 i 5% of the TCE was
recovered. Accordingly, a correction factor of 0.95 was applied to off-gas TCE

measurements.
6.4.5 Mass balance experiments

Mass balance experiments allow TCE removal to be divided into air-stripped and
biodegraded fractions. Parameters measured during mass balance studies include flow
rate of the feed solution, TCE concentrations in both the feed stream and the reactor
(including both gas and liquid phases), and mass of TCE trapped on activated carbon.
Mass balance experiments were conducted from the beginning of the fill period and
ended when reactor TCE concentration was no longer detectable ( < l jig/l ). Gas phase
TCE samples were withdrawn from a mininert valve mounted on the reactor using a gas-
tight syringe and injected into the GC for TCE measurement. The gas phase sample
volume varied from 0.1 mL to 0.5 mL depending upon the TCE concentration. For
aqueous phase TCE sampling, 2 mL of liquid sample was withdrawn from the reactor and
preserved by injection and scaling in a 20 mL vial containing 30 pl H3PO4 (85%). These
samples were then analyzed by GC/ECD.

 

130
6.5 Results and discussion
6.5.1 Mass balance experiment

Mass balance experiments provided a means of assessing TCE removal by
different mechanisms. The major removal mechanisms were air stripping and
biodegradation. Partitioning of TCE to biomass was examined theoretically and
experimentally. For a suspended solids concentration of 0.1%, about 0.0084% of the TCE
was expected to partition to cells, as calculated from equation 6-4a, assuming logKm, for
TCE = 2.29 (Weast 1992). Experimental results confirmed that TCE partitioning to cells
was insignificant. Comparing the aqueous TCE concentrations of autoclaved samples to
the aqueous TCE concentrations of uninocculated samples (Figure 6-3), no significant
difference was observed. However, TCE adsorption by biomass might be significant at

higher solids concentrations, as predicted by equation 6-5.

TCE removal by air stripping was evaluated by comparing TCE removal in an
abiotic SBR to the TCE removal in the biotic reactor. Both reactors were operated under
identical except that the abiotic reactor was not inoculated with microorganisms. Figure
6-4 illustrates observed TCE removal in both reactors. The presence of organisms
enhanced TCE removal significantly. In the abiotic reactor, higher TCE concentrations
were observed, and a longer react time was required to achieve the same TCE removal

achieved in the biotic reactor.

The fate of influent TCE was determined by monitoring TCE concentrations in
the reactor and in the off-gas. The portion of influent TCE removed by biodegradation
was calculated from mass balance equations. In most measurements, biodegradation
accounted for 80 to 95 % of TCE removal and air stripping for 5 to 10% TCE removal
(Figure 6-5). Less than 5% of influent TCE was not accounted for and may have leaked

from the reactor or adsorbed onto tubing or fittings. Results from the mass balance

131

 

 

 

 

.5 12

8 a

g 8 10 --

'8

8 g 8 ..

3 5'3 ED 6 --

a '3 v

5 g 4 ..

e .5. 2

a E 0 l I l l l
0 2 4 6 8 10 12
Equalibrium aqueous TCE concentration in uninoculated

vial (ma/l)

Figure 6-3 Comparison of the observed aqueous TCE concentration in uninoculated
samples and samples that contained 840 mg/L of autoclaved cells. TCE concenn'ation
was measured after the sample was incubated for 2 hr at 120 rpm on a rotatory shaker .
Error bars denote the standard derivation of triplicate samples.

132

 

 

 

 

 

 

 

 

 

 

 

0.0003
3 0.00025 -
O
9 0.0002 -
H A
U E
E“ h 0.00015 -
g e
a O-Wl '-
3 0.00005 “ m
=
0 a
0 60 120 180 240
Time (min)
0.00004
0.000035 -»
é 0.00003 ..
N: (1W '1!- MW
2% 0.00002 «p
35 0.000015 «-
‘5' 0.00001 -» biotic reactor
‘5 0.000005 --
0 : : .
0 60 120 180 240
Time (min)

Figure 6-4 Observed TCE concentrations during fill and react period in an SBR with
influent containing 0.5 mg/L TCE in (a) liquid phase and (b) gas phase.

133

 

 

linfluent'l‘CE: 0.5 mg: I linfluent'ICE: 2.5mg1 I IinfluentTCE= 5.0 mg: I

% Biodegraded

    

% removal of TCE
8

 

 

 

 

Figure 6-5 Percentages of TCE removed by biodegradation and air stripping as
determined in mass balance experiments.

5

 

as
:2
:'

F
i
1.

§

«I
1000 ‘

l
T
I

dry weight (ma/l)
§

 

 

cage

 

I.
‘D

10 20 30 40 50 60 7O 80

C

Figure 6-6 Biomass variation observed in a bench-scale SBR conducting TCE

transformation.

134

experiments are summarized in Figure 6-5. Initially, the dominant TCE removal
mechanism was biodegradation (up to 86% removal by biodegradation). When TCE feed
was initiated, biomass concentration decreased (Figure 6-6) and the portion of TCE
removed by air stripping increased. At the point of greatest biomass loss, up to 55% of
the influent TCE was removed by stripping. As organisms acclimated to TCE, the
observed cell concentration gradually returned to the initial level, and biodegradation

again became the major removal mechanism of TCE.

The period of decreased TCE degradation prior to TCE acclimation might be
explained by loss of total biomass or changes in community structure caused by the
toxicity of TCE or its transformation products. Changes in specific rates of TCE
transformation suggest initial changes in overall biomass followed by changes in
community structure. Since specific rates are normalized by biomass, changes in these
values are suggestive of change of community structure. As shown in Figure 6-7, loss of
TCE degradation activity before day 5 was likely due to loss of biomass since specific
rates of TCE transformation decreased only slightly. After day 10, however, biomass
levels had recovered to pre-TCE exposure levels, but specific TCE transformation rates
remained low. This suggests that loss of TCE transformation capacity was likely due to
changes of community structure. After day 15, specific TCE degradation rates gradually
recovered to the levels found in the inoculum, and TCE degraded without loss of
transformation activity and biomass. TCE exposure apparently caused loss of biomass
and TCE transformation activity for cells that had never been exposed to TCE, but 40
cycles of repeated exposure (20 days), resulted in the emergence of a new community

that was stable with respect to TCE degradation.
6.5.2 Batch TCE degradation experiment

Cells harvested in the recharge stage after phenol had totally degraded were used in batch
TCE degradation studies. Batch data were fitted to the integrated forms of

135

 

 

 

 

0.10 1 T v I ‘ I
0.09 r '1
0.08 . ‘
r? 0.07; .
>} I U U 1
g 0.06:. I Im an I In I I . :4
3 0.05 .' ..
E T I ‘
3 004 ’ "
i “'03 r 1
0.02 .. a ..
0.01 L 1
0m . 1 J L r . J 1 ‘
0 20 40 6O 80

Time (days)

Figure 6-7 Variations in specific rate of TCE degradation during reactor operation.
The initial aqueous phase TCE concentration was 1 mg/L.

136

equations 6-1 and 6-2, assuming negligible change in biomass concentration during the
assay period ( ~ 30 to 60 min.). The integrated expressions were:

Kc 1

In(&)+

 

 

 

lax. C ax. (6-23)

1 C.
= — 6-24
’ k,’X.. W C) ( )

where, C, = initial aqueous phase TCE concentration (mg/L) and X, = initial biomass

concentration.

Initial specific rates of TCE degradation were also calculated. Data from these
studies confirmed biotransformation of TCE in the bench-scale SBR and provided an
independent measure of kinetic coefficients used in system modeling. Observed kinetics
coefficients are summarized in Figure 6-8. Saturation kinetics adequately described TCE
degradation. The maximum TCE transformation rate (k,) was between 0.03 to 0.1 mg
TCE/mg dry weight-day and the half saturation coefficient (K,) was 0.3 ~l.0 mg/L. A
typical second order rate coefficient (Ir/IQ) was ~ 0.1 l/mg-day, indicating a TCE half life
on the order of 7 days for a cell concentration of 1 mg/L. There was considerable
variability in the kinetic data, possibly due to changes in community structure or enzyme
activity. In general, though, the observed TCE degradation rates were comparable to or
greater than reported values for methanotrophic and toluene-oxidizing enrichments
(Fogel et al., 1986; Strand et al., 1991; Wackett and Gibson 1988), but slower than some
pure cultures like Pseudomonas cepacia G4 (Folsom et al., 1990), Methylosinus
trichosporium OB3b (Brusseau et al., 1990; Fox et al., 1990; Oldenhuis et al., 1991)
and certain methanotrophic mixed cultures (Alvarez-Cohen and McCarty 1991; Henry
and Grbic-Galic 1991). Typical changes in the initial rates of TCE degradation in an SBR
operating cycle are provided in Figure 6-9. Decline of TCE transformation rates was
observed during the fill, react, settle and decant periods and recovery was observed

during the recharge period.

137

 

 

 

 

 

 

 

 

0.10 , . .
I I a I
I II n ' I,
I I u a I
0.08 a a .1
I
I
:2"
g, 0.06 _ 1
i
U - 1
.8
0.04 k .I .1
0.02 - - - - - -
0 20 40 60 80
Time (day)
1.50 . 1 1 1 1
b 4
1.25 , ,
1.00 1, 1
- I
A 0.75 a a . .
E I- I . ' I
v 050'1 _ - - . . - . .
é ' '
. u .
0.25 _ I ' 1.
p 1
0.00 L 1 1 1
0 20 40 60 80
Time (day)

Figure 6-8 Variations in the saturation kinetic coefficients Kc and kc for TCE degradation
during reactor operation .

138

 

  

 

 

 

 

 

 

 

 

ant
Settle Extendedaeratio
. . .. .. ' v 1
mo' ! . . '. ! ! ' . . . . .
q: 0.09 ..
a“: 1 1
‘3 0.08- i
0.071 .1
i: _ 1
z; 0.0.. i .
3 L .
E 0.05. .1
. i .
0.04 . - 4 1 - 1 - - 1 - 4
8 9 10 ll 12 13 14 15 16 17 l8 19 20

Sampling Time (hr)

Figure 6-9 Changes of TCE degradation activity during different stages. of SBR
operation. TCE degradation activity was measured as initial rates for an initial TCE
concentration of 2.0 mg/L. Error bars represent the range of standard deviations.

139
6.6 Bench-scale SBR modeling
6.6.1 Growth substrate concentration

The major mechanism of phenol removal was biodegradation. Estimated phenol
partitioning to cells is only about 0.0025 % of the total phenol addition, as predicted by
equation 6-4a assuming logKow for phenol = 1.50 (W east 1992). Air stripping of phenol
from the reactor was also insignificant, as indicated by a constant phenol concentration
during the react period in the abiotic reactor. Biodegradation of phenol degradation was
zero-order with respect to phenol concentration and first-order with respect to biomass, as
shown in Figure 6-10. Typical zero-order kinetic coefficient (k0) were between 0.5 to 1
day‘ 1 . Observed phenol degradation patterns and kinetic coefficients were similar to those
reported for phenol-acclaimed activated sludge (Auteinrieth et al., 1991; Rizzuti and
Augueliaro 1982; Tischler and Eckenfelder 1969). Maximum phenol concentration (100
~ 120 mg/L.) was present at the end of phenol addition. Phenol degraded to nondetect
levels in 2~5 hr depending upon the influent TCE concentration. The higher the
. concentration of TCE in the influent, the longer the time required to consume the phenol.
The recharge time required for phenol removal (2~5 hr) was shorter than the time allotted
for recharge in each cycle (6 hr). This allowed for an idle period of 1 to 4 hours.

A decrease in phenol utilization rate was observed as TCE influent concentration
increased. A possible reason for the decrease in phenol utilization rate was product
toxicity. Others have reported that TCE breakdown products can covalently bind to cell
material and can inactivate oxygenase activity (Fox et al., 1990; Wackett and
Householder 1989). Inactivation of the oxygenase during TCE transformation may
explain reduced phenol utilization rates. Inactivation of the oxygenase may be a
reversible process (Rasche et al., 1991) or it may cause the death of cells (Heald and
Jenkins 1994; Henry and Grbic-Galic 1991). The extent of cell damage caused by TCE

140

 

200 . a . 1 - 1 . T -

 

  
  
  
  

150 P L i 1
‘ observed phenol in abiotic reactor
'— modeled phenol in abiotic reactor
1m 1- u

I observed phenol in biotic reactor

1...... modeled phenol in biotic reactor

50

Phenol Conc. (mg/l)

 

l

0.0 0.5 1.0 1.5 2.0 2.5
Recharge Time (hr)

 

Figure 6-10 Typical observed and modeled phenol concentrations in the SBR (data was
taken on day 25).

141

degradation is a fundamental issue related to how much growth substrate is needed for

activation and how long it takes to regenerate a given level of transformation activity.
6.6.2 Nongrowth substrate concentration

Modeling TCE concentration in the SBR is more complex due to the volatility of

TCE. To solve TCE mass balance equations, the overall TCE mass transfer coefficient

(K La, TCE) and the TCE degradation kinetic coefficients must be evaluated first. Observed

TCE concentrations in the abiotic reactor were used to estimate K La, TCE. A mass balance

for the react period in the abiotic reactor can be derived from equation 6-17 and 6-18 by

eliminating the biodegradation terms:

dC C
dc, V1 C, Q,
dt M.TC£( V: )( 1- HC V: g ( )

For an assumed K Lam, equations 6-23 and 6-24 can be solved simultaneously.
K La, TCE was estimated by minimizing the difference between observed and modeled TCE
concentrations (C, and CL). A K Lama value of 0.0175 min:1 was obtained for the react
period. As shown in Figure 6-11, using the K La, TCE value obtained from the react period
to model the fill period resulted in the C, and C, values for fill period that were consistent

with measured values. This suggests that changes in K Lam: during the fill period were

not significant for this experimental system.

Batch TCE degradation assays were used to determine 1:, and K,, and these results
cross checked well with reactor measurements. After K La, TCE, K, and k, were obtained,
equations 6-13, 6-14 and 6-16 were solved simultaneously using the software package
Mathematica 2.0. As shown in Figure 6-12 and 6-13, the observed TCE concentrations in

the reactor matched modeling predictions. TCE concentration increased in the fill period

142

 

 

 

 

 

 

 

 

0

8

1:

fl-

8

8

5

5'

a

.5

ca

0

[-

0 50 100 150 200 250
Time (min)
0.00004

z 0.000035 .. I

‘3. 0.00003 “F I

8 55‘ 0.000025 -- .

any, 0.00002 «-

.5 5 0000015 -- '

a: V '

0 0.00001 .-

E" 0.000005 1

0 : a : :
0 50 100 150 200 250
Time (min)

Figure 6-11. Observed and predicted TCE concentration during the fill and
react stage in the abiotic reactor. Fill period was for the period t = 0 - 60 min,
and the react period began at t = 60 min and continued until the end of the
experiment. Predicted TCE concentrations were obtained by assuming a
constant KLa = 0.0175 min'1 for both the fill and react stages.

CL(mg/ml)

143

 

 

  
  

   

   

 

  

 

 

0.0015
Co = 5 m
0.0012 -- 3“
0.0009 --
0.0006 -- Co = 2.5 mg/l
ti 1..
‘1" 1 1‘11
0.0003 4 /
Co = 0.5 mg/I I
0 ..-.'.'.'?'.'.'.-.-.-9
0 30 60 90 120
Time (min)

Figure 6-12 Liquid phase TCE concentrations observed and modeled in a
bench-scale SBR.

Ca (me/ml)

 

0.0002

0.00015 4

0.0001 -

I

0.00005 «-

 

   
   
 

Co=5mgll

 

Co = 0.5 mgll
Aaqaaawmqa-

 

 

30 60 90 120
Time (min)

Figure 6-13 Gas phase TCE concentration observed and predicted in a bench-
scale SBR.

144

and rapidly decreased during the react period. The typical react time for total
disappearance of TCE from reactor was 40 to 90 minutes depending on the influent TCE
concentration. Figures 6-12 and 6-13 also show that, in general, the model adequately
predicted reactor performance. The model slightly overpredicted TCE concentration in
liquid and gas phase. This might be due to losses of influent TCE by adsorption to or
leakage from tubing and fittings (representing up to 5~10% of influent TCE losses, as
indicated by mass balance experiments) or losses of TCE due to biodegradation in the
inlet tubing (some unavoidable biofilm developed in the submerged part of influent tube).
In spite of these limitations, the results indicate that the approach used provided a fair

approximation of reactor performance.

The effects of K Laws, K, and k, on reactor performance were evaluated and

compared in Figures 6-14, 6-15 and 6-16. Increases in k, or decreases in K, and K Lam,

enhanced TCE removal by biodegradation and reduced the importance of air stripping.
6.7 System optimization and design consideration

The bench-scale cometabolic SBR system was not optimized for its performance
although it consistently removed TCE. Several optimization procedures are possible.
Aeration rates may be decreased to levels that just satisfy the oxygen requirements for
TCE transformation and cell respiration . This would reduce TCE losses by air stripping.
The required time for each stage might be reduced to accommodate more wastewater.
The mass of phenol addition during recharge period might be reduced. The optimal tr
may range from 1.5 to 2.5 hr depending upon influent TCE concentration. One hour of
settling time was adequate as the effluent suspended solids ranged from 10~30 mg/L and
no bulking was observed. The required recharge time was found to be dependent upon the
influent TCE concentration and might range from 2.5 to 5 hr. From batch studies, the
phenol required for cell regeneration was about 15~20 mg of phenol per mg of treated

TCE. Phenol addition was designed to accommodate 8~10 mg/L influent TCE- and

145

 

 

 

 

 

 

0.0001
0.00009 KLa = 0.1 min-1
0.00008
0.00007
E 000000
a: 0.00005
5 0.00004
(3 0.00003
0.00002
0.00001
0
0 20 40 60 80 100 120
Time (min)
0.0006
KLa = 0.01 min-l
0.0005 .. “nu-f, KLa=0.05 min-1
,' -‘ KLa = 0.1 min-l
0.000. .. , , - , f
‘ \

    

 

 

 

0 20 40 60 80 100 120
Time (min)

Figure 6-14 Effects of KLa on simulated liquid and gas phases TCE
concentrations in an SBR. Parameters used in the simulation: k, = 0.1 d‘l, K, =
0.5 mg/L, C, = 2.5 mg/L, X, =1500 mg/L and k, = 0.05 d'l.

146

0.00012

 

00001 kc = 0.05 d-l

0.00008

0.00006

Ca (mg/ml)

0.00004

0.00002

 

 

 

 

 

0.0005
0.00045
0.0004
0.00035
0.0003
0.00025
0.0002
0.00015
0.0001
0.00005

kc = 0.05 d-l

 

Cl (mg/ml)

 

 

 

Time (min)

Figure 6-15 Effect of k, on simulated liquid and gas phases TCE concentrations
present in an SBR. Parameters used in simulation: K, = 0.5 mg/L, KLa, TCE =
0.05 min-1, C, = 2.5 mg/L, X, =1500 mg/L and k, = 0.05 d

147

 

0.00012

Cums/ml)
i i

g

 

 

 

 

0 20 40 6O 80 100 120

0.0005

 

0.0004

0.0003

0.0002

Cl(mg/ml)

0.0001

 

 

 

 

Time (min)

Figure 6-16 Effects of K, on simulated liquid and gas phases TCE concentrations in
an SBR. Parameters used in simulation: k, = 0.1 d'l, KLa, TCE = 0.05 min-1, C, = 2.5
mg/L, X, =1500 mg/L and k, = 0.05 d'l.

148

actually enabled degradation of 5 mg/L influent TCE without serious toxicity effects.

Thus, the SBR received more phenol than needed. Batch recharge studies also indicated

that phenol addition could be reduced.

6.8 Conclusions

In light of the above observations and modeling, the following conclusion may be

drawn:

1. An SBR microbial community incorporating a phenol recharge stage can cometabolize

TCE for extended periods without loss of transformation activity .

2. The major mechanisms involved in TCE removal from the SBR were biodegradation

and air stripping, accounting for up to 95% of the TCE reduction. Slower TCE
transformation kinetics and higher aeration rates increase the magnitude of air
stripping losses. Faster TCE kinetics and lower aeration rates enhance biodegradation
efficiency. For the bench scale SBR tested, 80 to 90% TCE was biodegraded and 5 to
10% was air stripped.

. A shift in community structure occurred when microorganisms were first exposed to
TCE. Prior to acclimation, declines in biomass and specific rates of TCE degradation

were observed prior to eventual recovery.

. Performance of the bench-scale SBR was adequately simulated by the proposed
modeling procedure. The modeling process is able to predict TCE concentration and
may be helpful in system design and simulation.

5. Increasing influent TCE concentration necessitates longer recharge periods because of

reduced rates of phenol utilization. It is not clear whether the reduced phenol
utilization rates were attributable to death of cells or inactivation of enzyme. Further

research into the effects of nongrowth substrates on community structure and enzyme

149

inactivation will facilitate improvements in the design and operation of SBRs capable

of cometabolic transformations.

Literatures cited

Alvarez-Cohen, L. and P. L. McCarty. “TCE transformation by a mixed methanotrophic

culture - effects of toxicity, aeration and reductant supply.” WM 57
(1 1991): 228-235.

 

Washington, D. C.: 1989.

Auteinrieth, R. L., J. S. Bonner, A. Akgerman, and M. Okagun. “Biodegradation of
phenolic wastes.” mmmmnm 28 (1991): 29-53.

Brusseau, G. A. , H. C. Tsien, R. S. Hanson, and L. P. Wackett. “Optimization of
trichloroethylene oxidation by methanotrophs and the use of a colorimetric assay to

detect soluble methane monooxygenase activity.” W 1(1990): 19-29.

Clowater, L. M. “Variables effecting trichloroethylene transformation by

methanotrophs.” Master, Michigan State university, 1992.

D'Adamo, P. D. , A. F. Rozich, and A. F. Jr. Gaudy. “Analysis of growth data with
inhibitory substlmef’ W 26 (1984): 397.

Dobbs, R. A., Leping Wang, and R. Govind. “sorption of toxic organic compounds on
wastewater solids: correlation with fundamental properties.” W 23 (9
1989): 1092-1097.

Fogel, M. M. , A. R. Taddeo, and S. Fogel. “Biodegradation of chlorinated ethenes by a
methane-utilizing mixed culture.” WM 51 (1986): 720-724.

150

Folsom, B. R., P. J. Chapman, and P. H. Pritchard. “Phenol and trichloroethylene

degradation by Pseudomonas cepacia G4: kinetics and interactions between substrates.”

Wanna]. 56 (1990): 1279-1285.

Fox, B. G., J. G. Borneman, L. P. Wackett, and J. D. Lipscomb. “Haloalkane oxidation
by the soluble methane monooxygenase from Methylosinus trichosporium OB3b:

mechanism and environmental implication.” W 29 (27 1990): 6419-6427.

Gossett, J. M. “Measurement of Henry's law constants for C1 and C2 chlorinated

hydrocarbons,.” WM 21 (1987): 202-208.

Henry, S. M. and D. Grbic-Galic. “Influence of endogenous and exogenous electron

donors and trichloroethylene oxidation toxicity on trichloroethylene oxidation by

methanotrophic cultures from a groundwater aquifer.” AW 57 (l
1991): 236-244.

Mackay, D. and W.-Y. Shiu. “A critical review of Henry's Law constants for chemicals of
environmental interest.” W 10 (4 1981): 1175-1197.

Martin, R. W. “Rapid colormetric estimation of phenol.” W 21 (11
1949): 1419-1420.

Matter-Muller, C., W. Gujer, W. Giger, and W. Stumm. “Non-biological elimination

mechanisms in a biological sewage treatment plant.” WM 12
(1980): 299-314.

Oldenhuis, R., J. Y. Oedzes, and D. B. Janssen. “Kinetics of chlorinated hydrocarbon by
Methylosinus trichosporium OB3b and toxicity of trichloroethylene.” A991,_Eny1r99,
M91, 57 (1 1991): 7-14.

151

Rasche, M. E., M. R. Hyman, and D. J. Arp. “Factors limiting aliphatic chlorocarbon
degradation by Nitrosomonas europaea: cometabolic inactivation of ammonia
monooxygenase and substrate specificity.” W21. 57 (10 1991):
2986-2994.

Strand, S. E., J. V. Wodrivh, and H. D. Stensel. “ Biodegradation of chlorinated solvents
in a sparged, methanotrophic biofilm reactor.” W 63 (6 1991): 859-
867.

Tischler, L. F. and W. W. Jr. Eckenfelder. “Linear substrate removal in the activated

sludge process.” In WWW, 361-374. 2. Oxford, England:
Pergamon, 1969.

Wackett, L. P. and D. T. Gibson. “Degradation of trichloroethylene by toluene
dioxygenase in whole cell studies with Pseudomonas putida F1.” M1199,
M19m9191,54 (1988): 1703-1708.

Wackett, L. P. and S. R. Householder. “Toxicity of trichloroethylene to Pseudomonas

putida F1 is mediated by toluene dioxygenase.” WM 55 (10 1989):
2723-2725.

Weast. R. 0. ed. Watts 73 ed., Boca Raton. Florida:
CRC Press, Inc., 1992.

CHAPTER 7
SUMMARY AND ENGINEERING IMPLICATIONS

7.1 Summary

The major objective of this work was to determine under what conditions
cometabolism can be accomplished in a sequencing batch reactor and to elucidate design
principles governing application of cometabolism. Theoretical considerations and
modeling efforts were undertaken to develop a simulation-based design approach.
Experiments were conducted to clarify the role of substrate feeding pattern on the
selection of TCE degrading organisms and on the long-term stability of the resulting
microbial community. Toxicity was also explored in cell viability experiments with
different organisms. Experience acquired in the course of these efforts was used to
operate and numerically simulate a bench-scale sequencing batch reactor with provisions

for cometabolism of TCE.
7.1.1 Use of a numerical design approach for sequencing batch reactors

The design approach proposed in this work is based on the derivation and
numerical solution of mass balance expressions for each phase of SBR operation. This
contrasts with traditional SBR design approach in which empirical loading rates are
chosen to determine reactor volume, with empirical guidelines used to establish wasting
strategies and time periods for each operating stage. A simulation program enables

simulation of various design options and strategies. It also enables simulation of SBR

152

153

performance, and it is adaptable for the addition of special features, such as a recharge

stage for cometabolism.

Because of its dynamic mode of operation, a steady state condition does not exist
for an SBR, but a repeating pattern of operation is expected for long-term operation if a
repeating or constant hydraulic and substrate loading pattern can be defined. In this work,
SBR design is based on the long-term repeating pattern of performance. Solids residence
time (SRT), a traditional design and control parameter for activated sludge systems, was
applied to SBR design, to arrive at a solids management schedule. However, SRT failed
to account for dynamic changes in mean cell age during SBR operation. A more realistic
estimation of average cell age (dynamic cell age) was proposed. By tracking changes in
mass of each designated organism type, changes in the average population age can be
predicted throughout the SBR operating cycle. Mean cell age decreases during periods of
substrate feed and increases as substrate levels are depleted and endogenous decay
becomes significant. SBRs must be designed so that the specified average solids
residence time exceeds the minimum cell age for the slowest growing microbial

- population needed for proper performance of the reactor (i.e., SRT > llum).

Solids control in an SBR is accomplished using a flexible solids wasting strategy.
The frequency of wasting events and the quantity of solids wasted per event control the
accumulation of solids. Solids concentrations in an SBR continue to increase until a
wasting event. The chosen solids wasting strategy affects the extent of solids
accumulation, the required time for the fill and react periods, and the volume of the
reactor. SBRs with frequent solids wasting have lower suspended solids concentrations,
shorter cell age, require more reactor volume and require more time for the react period.

Increasing the quantity of solids wasted per wasting event has similar effects.

The number of reactor used in an SBR system affects the required total reactor

volume, cell age, and the required cycle time. When more reactors are used, less total

154

reactor volume is needed, and shorter cycle times are possible. The incremental decrease
in total volume reduction is less when the number of reactors exceeds four. Reduced cell

age is also observed in SBR systems as the number of reactors is increased.

7.1.2 Effect of growth substrate feeding pattern on community structure and TCE

cometabolism

Community structure and TCE cometabolism activity can be manipulated by
changing growth substrate feeding pattern. For phenol, pulse feeding selected for
organisms with faster phenol and TCE degradation rates. Pulse feeding also selected for a
diverse bacterial community with a high concentration of TCE-degrading species and
floc-forming organisms. Fluctuations of biomass concentration were more extreme and
more frequent in reactors that were fed phenol continuously or semi-continuously,
apparently because of the greater prevalence of predators in these reactors. The
continuous and semi-continuously fed reactor communities were less diverse and
dominated by fungi that degraded TCE poorly or not at all. Of the total bacterial isolates
obtained, most were unable to degrade phenol or cometabolize TCE.

The experiments with growth substrate feeding pattern established that the
feeding pattern was critical to SBR operation - for phenol, a pulsed feeding pattern was
clearly the preferred method of feeding, both for cometabolism and for development of an

enrichment with favorable settling properties.
7.1.3 Effects of TCE on cell viability

That TCE cometabolism causes loss of cell reculturability rather than temporary
inhibit their growth was clearly demonstrated for P. putida F l and P. cepacia G4.
Toxicity effects were species dependent. TCE exhibited more toxic with strain F 1 than
with strain G4.. Loss of cell viability was not observed in a strain that grew on phenol,

but did not degrade TCE. For strain G4, the extent of inactivation of cells

155

was independent of the TCE concentration over the TCE concentration range from 1 to
10 mg/l. Loss of cell viability caused by TCE cometabolism was linearly related to the
quantity of TCE transformed. The concept of TCE transformation capacity proposed by
Alvarez-Cohen and McCarty adequate described the effects of TCE toxicity on cell
viability.

7.1.4 Bench-scale SBR for TCE cometabolism

A bench-scale SBR modified to include a recharge stage successfully
cometabolized TCE. Up to 90% of the influent TCE was removed by cometabolism and
about 10 % was air stripped. Evidence for TCE product toxicity was only noted upon
initiation of TCE addition, as biomass concentrations and TCE transformation activity
declined. However, after approximately 20 days of twice-daily exposure to TCE, biomass
and TCE transformation activity recovered to levels observed prior to TCE exposure. No
further toxic effects were observed. A mathematical model was used to simulate the
major TCE removal mechanisms (biodegradation and air stripping) after stable operating

conditions were achieved.
7.2 Engineering implications

The numerical, simulation-based SBR design approach described in this work
offers several advantages for the design and operation of conventional SBR systems and
for the extension of SBR technology to new applications, such as cometabolism. Many
design and operating strategies can be evaluated. It also reduces reliance on questionable
rule-of-thumb criteria and safety factors, and it provides insight into reactor performance.
The ability to numerically track mean cell age is of interest because certain reactor
functions are cell age dependent. A possible example is growth and nongrowth substrate
additions. To eliminate competitive inhibition, nongrowth substrate should be provided

when growth substrate levels are low and cells are young.

156

To apply SBR techniques for TCE cometabolism, a recharge stage is needed to
regenerate the TCE transformation activity of organism lost during TCE degradation.
Such a step enables stable operation and continuous long-term biodegradation of TCE.
Manipulating the substrate delivery mode appears to be an effective way of selecting cells
with the desired function. Addition of phenol in pulses worked well for phenol
degradation, TCE cometabolism and cell settleability. However, the interactions of
community members in the phenol enrichment is very complex and should receive
additional attention in reactor design. Traditional engineering frequently ignores the role
of community members, with simple steady state or empirical approaches to reactor
design. This tendency may lead to an impractical design because of changes in the
microbial community, especially during treatment of hazardous wastes. Dynamic and
unsteady state reactor systems, like the SBR, should focus on microbial community

structure, especially as it affects the stability of the reactor performance.

Transformation of TCE results in cell death. Therefore, regrowth of surviving
cells to desired population levels may require long time period, especially where large
quantities of TCE have been transformed. The surviving cells can be regrown during the
recharge stage. Thus, the recharge time is fixed by the concentration of surviving cells
and their growth kinetics. Prediction of the concentration of viable cells after TCE
exposure is a crucial step in establishing the required time for recharge. Previously
proposed cell viability models using the concept of transformation capacity may be

applied to estimate the number of cells that survive TCE exposure.
7.3 Proposed research

To better understanding the function and limitations of SBR systems for

cometabolism, the following additional studies are proposed:

157

1. Expansion of SBR design and simulation protocol to include a range of desired
community functions, including cometabolism, denitrification, and phosphate removal.
The effects of variable flows and substrate concentrations should also be explored in
simulation. Commercialized design and simulation tools can be developed to assist

engineers and operators in the rational design and operation of SBRs.

2. Selection and enhancement of the cometabolism activity might be achieved by better
understanding of the fundamental reasons why microorganisms engage in cometabolic

transformations in the first place.

3. The role of each community member should be clarified. What is the role of
community members that do not degrade the growth substrate or the nongrowth
substrate? Does their existence benefit cometabolizing organisms or do they interfere

with the desired function of the community?

4. Studies on TCE transformation toxicity and cell viability should be extended to
different types of TCE degrading organisms. This might enable selection of resistant
organisms for treatment applications. Further studies should also focus on TCE toxicity in
mixed cultures. Does addition of TCE to a mixed culture select for non TCE degrading
organisms? To what extent are the toxic the products of TCE cometabolism available to

other members of the community?

5. Operating parameters of SBRs for cometabolism should be optimized. This work
demonstrated the feasibility of SBR systems for cometabolism. To optimize reactor
treatment efficiency, the aeration rate may be altered, and the time allotted for each stage
may be varied. Further studies should be conducted to test the impact of such changes,

and a range of operating conditions should be examined.

6. As TCE degradation capacity and TCE toxicity resistance varies from organism to

organism, it may be possible to enhance treatment efficiency by the addition of foreign

158

species. Addition of TCE-degrading isolates that are resistant to toxicity such as P.
cepacia G4 or genetic engineered organisms, may increase treatment efficiency and
improve the stability of the process. Success of such an approach would depend upon the
survival of the added organism, or at least, the persistence of the added catabolic genes

within the community. These issues certainly merit additional attention.

APPENDICES

APPENDIX A
DERIVATION OF CSTR STEADY-STATE SOLUTIONS USED AS THE INITIAL
CONDITION FOR SOLVING SBR MODELING EQUATIONS

Consider a typical CSTR activated sludge system as indicated in Figure A-l.

r X)" r S Settling tank

S0, Q0 on S, Xj, V Q‘.Se.Xf

     
 

 

  
 
 

Aeration tank QW, ij' 5"”

Figure A-l Schematic of a typical CSTR activated sludge system

Mass balance equation when the system in steady-state is
.in-out+source-sink=0

Consider general suspended solids Xj,
Q°X;’At - [QWij + Q'Xj' ]At + (mass production rate)At - (mass reduction rateMt = 0
Q°X;’ + (mass production rate) -(mass reduction rate) = QWXIw + Q'X; (A-l)

where mass production rate - mass reduction = er].

 

Subtitute 0, = QWXW +1 Q‘X e and Vrd to equation A-l gives
j J'
0 0 VX.
Q Xj +Vrg. =——’- (A-2)
0

C

Equation A-2 divided by V and substitute V/Q° with hydraulic residence time ( 9 ) gives:

159

160

X.=—‘X‘.’+9r. (A-3)

which is the form present in equation 3-13.

For example 1 in chapter 3, consider each of the different types of suspended solids in the

 

reactor at steady state:

X4

Y(S" -S‘)Qo -bXaV = XaV

90

0 Y S" -S‘

X0 = _£[_(__._)]
0 1+b6,

X4

90x3 + fdeaV - thdV = V34

C

 

 

Xd= & I x; +__f_,bo,r(s°—S)

9 (1+kh0c) (1+kh0,)(1+b9,)]

X4
VX

(1 — f, )bX,V = 7L

C

 

 

X.- =&[Y(S°-S)(I-f,_1)b9,]
0 1+b9,
X;
o o _ VXr
Q Xr - 9c

161

 

On Xa mass balance:

wrxy _ bXaV = X,V
K, + S‘ 9,

_ K,(1+bo,)
0,(Yk-b)-1

 

APPENDIX B
CURVE FITTING FOR TCE DEGRADATION KINETICS

Three kinetic models were examined (Table B-l). The integrated forms were
obtainted by assuming the changes of cell concentration during the experimental period is

negelectable. A statistic package SYSTAT 5.2.1 was used to estimate the kinetic

 

 

 

 

 

 

 

 

parameters.
Table B-1. Kinetic models examined for TCE degradation
Kinetic model General form Integrated form Fitted parameters
k CX Kc C0 1
=— C t= ln(—)+—(Co-C)

M°n°d " K. + C k.x. C ex. "c’ Kc

- r - k'CX r— 1 100(9) k'
FlfSt-Ordcr c - c kéxo C c
Zero-order 1', ="k:X ‘=C° —C 1‘:

kZX.

 

 

 

 

 

 

Typical TCE degradation data after 300 days of operatiion were fit to different
kinetic models for each reactor. Results are summarized in Figures B1-B4. Estimated

parameters and correlation coefficients are provided in Table B-2.

162

163

 

 
 
   
   

TCE (mg/l)

0.2 1..

 

99
4803

I Measured values
Monod kinetics
. — ----- Zero order kinetic

"""" First order kinetic

l

 

 

 

0.2 0.4 0.6 0.8 1 1.2

1.4

Figure B-l Typical curve fitting results for TCE degradation observed in batch
experiments using cells from reactor P.

 

 

 

 

A l " “a
s M...

0.8 " u
g ~‘."~‘~-'

\ --..

‘6 0,6 1 I Measureddata T‘ :"'~~---_
B 0'4 - - ----- First-orderkinetic

0'2 " """" Zero-orderkinetrc

0 : 4. : 4. 1
0 2.5 5 7.5 10 12.5
Time (hr)

15

Figure B-2 Typical curve fitting results for TCE degradation observed in batch
experiments using cells from reactor C.

164

 

 
 
 
 
  
 
 

I Measured data

Monod kinetic

'00
L
t

\ """" First-order kinetic

''''' Zero-order kinetic

TCE (mg/L)
0 O C
O‘

It;
I
I

 

 

 

0.2 --
0 : -: ‘-3~"
0 2.5 5 7 5 10

Time (hr)

Figure B-3 Typical curve fitting results for TCE degradation observed in batch
experiments using cells from reactor SC2.

 

 

 

 

1 2 g.
“‘7‘”...
l -- I ’fia.‘
8"“\
,1 0 8 "‘--
§ . II- .‘ {-‘Q
I “and "3:3'»---
5 0.6 -- M d“ ~-...___'---.
it} ________ . . . "n
a 0.4 1. Frrst-orderkmetrc
"' ''''' Zero-orderkinetic
0.2 --
0 i i :
0 2.5 5 7.5 10
Time (hr)

Figure B-4 Typical curve fitting results for TCE degradation observed in batch
experiments using cells from reactor SC5.

165

Table B-2. Estimated kinetic parameters for TCE degradation

 

 

 

 

 

 

 

 

 

Kinetic model Reactor P Reactor C Reactor SC2 Reactor SC5
k, = 0.077 +/. 0.001 d-1 . k, = 0.011 d-1 b .
Monod K, = 0.29 +/. 0.01 mg/L F6113 K, =0.80 mg/L Falla
r2 = 0.9977 :2 = 0.9994
kc' = 0.13 +/— 0.02 10,: = 0.0024 +/- 11,: = 0.0089 +/. 10,: = 0.0018 +/.
First-order Umg-d 0.0087 Umg-d 0.0050 Umg-d 0.0071 Umg-d
r2 = 0.8571 :2 = 0.9543 :2 = 0.9856 r2 = 0.9741
- ° = 0.0024 +/- 0 = 0.0044 +/- 11,0 = 0.001844-
9 = 0.043 +/- 0.001 d 1 kc *9
Zem‘ordc’ kc r2 = 08950 00144 9" 0.0084 01-1 0.0095 04
r2 = 09131 r2 = 0.9631 :2 = 0.9542

 

a. Estimated parameters out of reasonable range. Parameters (kc and Kc) are highly

correlated (>0.99).
b. Singular Hessian. standard errors are not countable.
c. +/- 95 % confidence interval.

 

MICHIGAN STRTE UNIV. LIBRQRIES
lllllllllllllllllllllllllIllIll!illMIHIVIIWIIIUIIIll
31293014101822