 

 

LIBRARY
Mlchlgan State
UnIversIty

 

 

 

PLACE IN RETURN BOX to roman this chockom Imam your mead.
TO AVOID FINES mum on or bdon dd. duo.

DATE DUE DATE DUE DATE DUE
,,,,, 2-2,7,‘ , ,,,'|

      

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Walla-9.1

, k“

 

ﬂ

SUBSTRATE INTERACTIONS DURING THE DEGRADATION OF BENZENE,
TULUENE, AND p—XYLENE AND THEIR EFFECTS ON SORPTION IN A
FIXED BED BIOLOGICAL ACTIVATED CARBON REAC'POR

By

Myung-Keun Chang

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

1994

ABSTRACT

SUBSTRATE INTERACTIONS DURING THE DEGRADATION OF BENZENE,
TOLUENE, AND p-XYLENE AND THEIR EFFECTS ON SORPTION IN A
FIXED BED BIOLOGICAL ACTIVATED CARBON REACTOR

By

Myung-Keun Chang

Two Pseudomonas species (designated strains Bl and X1) were isolated from an
aerobic pilot-scale ﬂuidized-bed reactor treating groundwater containing benzene,
toluene, and p-xylene (BTX). Strain Bl grew with benzene and toluene as the sole
sources of carbon and energy, and it cometabolized p-xylene in the presence of toluene.
Strain X1 grew on toluene and p-xylene, but not benzene. Competitive inhibition and
cometabolic transformation were observed during batch rate experiments using paired
substrates (BT, TX, or BX) and the two strains. These were mathematically modeled and
rate parameters were determined. Coupling was achieved by deﬁning two transformation
capacity terms for the cometabolizing culture. Cometabolism increased decay rates, and

the observed yield for strain B1 decreased in the presence of p-xylene.

The biodegradation kinetic study showed that cometabolism resulted in the
formation of p-xylene dihydrodiol. In the absence of growth substrate, p-xylene
dihydrodiol was further degraded to 3,6-dimethyl pyrocatechol. These intermediate and
byproduct compounds were identified and quantified. The effect of these byproduct
compounds on activated carbon sorption revealed that they compete with the target

compounds.

Fixed-bed biological activated carbon (BAC) columns and biofilm columns were
seeded with strain BI and operated until steady-state conditions were reached. p-xylene
dihydrodiol was initially adsorbed by the activated carbon, but eventually saturated the
adsorption capacity of the column and appeared in the column effluent. Perturbation
studies showed the extended removal capacity of target compounds in BAC systems. The
response to the perturbation was buffered in the BAC systems as a result of the

adsorption capacity of the activated carbon.

The operation of columns treating toluene and p-xylene showed different patterns
of degradation depending on the bacterial species present. Shock loading to the reactors
by increasing the favorably degraded compound resulted in a significant increase in the
effluent concentrations of both compounds, while the increase of effluent concentrations
resulting from the increase of the less favorably degraded compound was smaller. The

formation of byproducts was reduced in both cases.

Solio Deo Gloria

ACKNOWLEDGMENTS

T he fear of the LORD is the beginning of wisdom,
and knowledge of the Holy One is understanding. (Proverbs 9:10)

Whenever I think about the last six years (1986-1992), I will remember the faces of

those who helped me to make the long journey to complete my Ph. D.

My committee members have my greatest appreciation. My major advisor, Dr.
Thomas Voice, encouraged me to start my life as a Ph.D. candidate and supported me
throughout the program. He provided me with new ideas in many ways. Dr. Craig Criddle
motivated me to think microbiologically and thereby gave me new energy to step forward to
the goal. His time and guidance were essential to this accomplishment. Dr. Mackenzie
Davis, who ﬁrst introduced me to environmental engineering during my master's program,
always reminded me to think of real world applications. His traditional Friday green suit
will be remembered for a long time. Dr. Stephen Boyd, also helped me in many different

ways, and I offer him my thanks for serving on my committee.

I also want to thank the many faculty and staff members at Michigan State
University who helped me and gave me back up. Dr. Patrick Oriel always answered my
questions about micro-organisms and pathways. Dr. Sadhana Chauhan, who is also my
GC injection partner, encouraged and helped me during the most difﬁcult phases of my
research. I deeply appreciate her suggestions and instructions in TLC work and byproduct

identiﬁcation. 1 especially thank Ms. Shirley Owens of the Pesticide Research Center for

V

vi

providing the SEM picture of my bacteria cultures and activated carbon particles. 1 thank to
Paul Groll of the Harzadous Substance Research Center, for his effort to prepare scanned
images of pictures on computer ﬁles. I also enjoyed talking with Dr. Daewon Pak about
many different topics in environmental engineering. I give my special thanks to Yanlyang
Pan, who was once my fellow graduate student, for his help running all the equipment and
solving analytical problems. I also would like to express my sincere thanks to Hung
Nguyen who provided important support for my experiment. There are also many other
names I did not list here, even though they aided me very much in many different ways, I

sincerely ask their forgiveness that I could not list them one by one.

I greatefully acknowledge the ﬁnancial support of the National Institute of Health

and the US. Environmental Protection Agency.

1 also enjoyed meeting my fellow graduate students who came from different parts
of the world to share their ideas and to struggle for the goals. They were an essential part of
my life at MSU. Chronologically speaking, I have to mention Nancy Hayden ﬁrst. I
appreciate her friendship over the last eight years. She inﬂuenced me to look at nature from
a different point of view, and she helped me to quit my bad smoking habbit so I can qualify
as a real environmental engineer. Dave Filipiak always gave me informal English while we
were shaking soils in the lab. Xianda Zhao and Jing Shi worked closely with me, and we
continuously exchanged ideas. And fellows like Munjed Maraqa, Mark Dixon, 1.]. Yao,
Chien-Chun Shih, ..... and many more who have also gone through the program will long

remain in my memory.

I would like to mention relatives who have provided moral support. My uncle and

aunt in Ann Arbor always took care of me with attention and consideration. I especially

vii

thank my aunt for her suggestions about my written English. I also thank my uncle and

aunt in Dayton, Ohio for stufﬁng me with turkey during every Thanksgiving holiday.

My father, who was one of the ﬁrst Korean sanitary engineers, influenced me very
much in my choice of environmental engineering. I appreciate his ﬁnancial support and
encouragement. Without my mother’s prayers, I could not have accomplished my research.

I thank my parents-in-law for their support, continued understanding and patience as well .

I saved the last few lines, even though it is not enough to describe everything, for
my wife Eunjin and my daughter Yewan. This work would not have been possible without
their presence, love, and encouragement. I can not say enough for Eunjin's sacriﬁces in
many parts of her life. I also appreciate her extra efforts to fulﬁll Yewan's needs when dad
was busy. Yewan was born and has grown with my research. I hope she does not have
bad feelings about her twin sister named ‘Research’ and understands why ‘She’ took so
much of daddy’s time. I am very glad now and feel well prepared to go back home to work

in Korea. Good Bye everybody and Thank You, Lord!

TABLE OF CONTENTS

List of Tables ................................................................................ xi

List of Figures ............................................................................... xii

Nomenclature ................................................................................ xv
CHAPTER

I. INTRODUCTION AND RESEARCH OBJECTIVES ...................... 1

1.1 Introduction, Background, and Problem Statement ................... 1

1.11 Introduction ....................................................... 1

1.1.2 Engineering Background ........................................ 2

1.1.3 Biological Activated Carbon Process .......................... 2

1.1 4 Problem Statement ............................................... 4

1.2 Research Objectives and Overview ..................................... 4

1.3 Production, Use, and Properties of BTX .............................. 6

II. BACKGROUND ................................................................ 9

2.1 Bioﬁlm Activated Carbon ................................................ 9

2.1.1 Development of Bioﬁlm on Granular Activated Carbon ..... 9

2.1.2 History and Definition ........................................... 10

2.1.3 Research on Bioﬁlm Activated Carbon System ............... 11

2.1.4 Adsorption, Desorption and Biological Activity .............. 15

2.2 Multi-Component Adsorption ........................................... 17

2.3 Microbiological Degradation of Organic Compounds ................ 20

2.3.1 Kinetics of Microbiological Degradation ...................... 20

2.3.2 Multisubstrate Degradation ..................................... 21

2.3.3 Degradation of BTX and Its Pathway ......................... 24

2.4 The Effect of Byproducts on Sorption .................................. 27

viii

III.

IV.

2.5

ix

Multicomponent BAC System .......................................... 28

KINETICS OF COMPETITTVE INHIBITION AND COMETABOLISM
IN THE BIODEGRADATION OF BENZENE, TOLUENE, AND

p-XYLENE BY TWO PSEUDOMONAS ISOLATES ...................... 29
3.1 Introduction ............................................................... 29
3.2 Materials and Methods ...................................................... 31
3.2.1 Growth Medium and Chemicals .................................. 31
3.2.2 Isolates ............................................................... 31
3.2.3 Inoculum Preparation .............................................. 32
3.2.4 Kinetic Experiments ................................................ 33
3.2.5 Identification of Byproducts ....................................... 34
3.2.6 Estimation of Decay Rates and Maximum Yields ............... 35
3.2.7 Parameter Estimation for Single Substrates ..................... 35
3.2.8 Quantification of Competitive Inhibition ......................... 38
3.2.9 Quantification of Cometabolism .................................. 39
3.3 Results ........................................................................ 39
3.4 Discussion .................................................................... 47
3.5 Conclusions .................................................................. 54
KINETICS OF COMETABOLIC DEGRADATION OF p-XYLENE BY

PSEUDOMONAS SP. STRAIN Bl AND THE INTERACTION OF
DERADATTON AND ADSORPTION IN FIXED-BED BIOLOGICAL

ACTIVATED CARBON SYSTEMS ............................................. 55
4.1 Introduction .................................................................. 55
4.2 Materials and Methods ...................................................... 58
4.2.1 Isolates .............................................................. 58
4.2.2 Inoculum Preparation .............................................. 58
4.2.3 Batch Kinetic Experiments ........................................ 59
4.2.4 Identiﬁcation and Quantification of Byproducts ................ 59
4.2.5 Pure Compound Adsorption Isotherm Experiments ........... 60

4.2.6 Effect of Strain Bl Byproducts on Sorption of Target
Compounds ......................................................... 61
4.2.7 Fixed-Bed Column Experiments ................................... 61
4.3 Results ........................................................................ 63
4.4 Discussion ................................................................... 78
4.5 Conclusion ................................................................... 82

X

V. EFFECT OF MICROBIAL PROCESS ON THE OPERATION OF
BIOLOGICAL ACTIVATED CARBON SYSTEM DURING SHOCK

LOADINGS ......................................................................... 83

5.1 Introduction .................................................................. 83

5.2 Materials and Methods ...................................................... 85

5.2.1 Isolates ............................................................... 85

5.2.2 Inoculum Preparation .............................................. 85

5. 2. 3 Batch Kinetic Experiment for Degradation ...................... 86

5. 2. 4 Packing Media ...................................................... 86

5. 2. 5 Batch lsotherrns for Toluene and p-Xylene ..................... 86

5.2.6 Fixed Bed Reactor .................................................. 87

5. 2. 7 Perturbation Experiment ........................................... 88

5.2.8 Analytical Method .................................................. 88

5. 2. 9 Scanning Electron Microscopy (SEM) ........................... 89

5.3 Results ....................................................................... 89

5.4 Discussion ................................................................... 100

5.5 Conclusion .................................................................. 110

VI. DISSERTATION SUMMARY AND RECOMMENDATIONS ............. 112

6.1 Summary ..................................................................... 112

6.2 Recommendations for Future Studies .................................... 114

APPENDIX A. Media Preparation .................................................... 116

APPENDIX B. Isolation Procedures ................................................. 117
APPENDIX C. Scanning Electron Microscopy Sample Preparation

Procedures ............................................................ 121

APPENDIX D. Code for Cometabolic Degradation Modeling .................... 122

LIST OF REFERENCES ............................................................... 126

TABLE

1-1.

1-2.

1-3.

2-1.
2-2.
2-3.
3-1.

3-2.
3-3.

3—4.

B-l.

LIST OF TABLES

Sites of environmental contamination listed on the priority lists in

the States of Michigan - ACT 307 ....................................... 4
Annual production of benzene, toluene, and p-xylene in the United
States during 1987-1990 ................................................. 6
Physical and chemical properties of benzene, toluene, and p-

xylene ....................................................................... 8
Comparisons of models by different researchers ...................... 14
Comparisons of enzyme inhibition ...................................... 22
Degradation pathway of BTX compounds ............................. 25
Single compound biodegradation kinetic parameters for strains B1

and X1 ...................................................................... 40
Degradation patterns for each combination by each strain ............ 4O

Decay rate of strain Bl with different initial p-xylene concentra-
tions ......................................................................... 46

Measured and estimated parameters for modeling degradation of
toluene and cometabolism of p-xylene by strain Bl ................... 46

Growth of isolated colonies in each carbon source .................... 118

xi

FIGURE

2-1.

3-1.

3-2.

3-3.

3-4.

3-5.

3-7.
3-8.

3-9.

LIST OF FIGURES

Example of efﬂuent jump from p-chlorophenol replaced by
p-nitrophenol ............................................................... 18

Sensitivity test for different initial biomass X0 (mg) with the same
half saturation consrant Ks of 1 m g/l .................................... 37

Sensitivity test for different half saturation constanth (mg/l) with
the same initial biomass of 30 ug ........................................ 37

Typical curves illustrating competitive inhibition in the degradation
of benzene and toluene by strain Bl .................................... 42

Lineweaver-Burke plot for the degradation of toluene alone and with
benzene by strain Bl ...................................................... 43

Typical curves illustrating cometabolic degradation of xylene in the
presence of toluene by strain X1 ........................................ 45

Comparisons between experimental data and model (estimated by
Runge-Kutta numerical solution) for competitive inhibiton of
benzene and toluene by strain Bl (Ks,benzenc = 3.17 mg/l,

KS,toluene = 1-96 mg/l, xo = 50 H8, KI,benzene = 3-10 "lg/1’
Kl’toluenc = 1.71 mg/l, and time lag = 70 min.) ...................... 48

Pathway proposed for the oxidation of p-xylene ..................... 50
Modeling of cometabolism - degradation of toluene and xylene by

strain Bl with parameters on table 4. Initial concentration of each
substrate in the liquid phase is approximately 10 mg/l ............... 52
Formation of byproducts monitored by absorbance at 279 nm for
transformation of p-xylene with toluene as a growth substrate by

strain B1 .................................................................. 53

Schematic diagram of the column apparatus .......................... 62

xii

4-9.

4—10.

4-12.

5-1.

5-2.

5-3.

xiii

HPLC chromatograms of two solutions: a) solution containing
byproducts from p- xylene degradation with toluene as growrh
substrate and b) ﬁnal solution after degradation of toluene by
Pseudomonas sp. strain B1. Peaks A, B, and C are byproducts

of p—xylene cometabolism ............................................... 64

Byproduct formation kinetic (initial 60 hrs) for degradation of
toluene and p-xylene ...................................................... 66

Byproduct formation kinetics (up to 300 hrs) for the transformation

of p-xylene (extended result of Figure 4-3) bottles were compared.
Bottle No. 1 contained biomass until the end of the experiment. Cells
were removed from bottle No. 11 after 30 hours .................... 67

Mass balance between p—xylene and byproducts based on the mass
of molecular carbon ....................................................... 68

Sorption isotherms of toluene and p-xylene prepared in control
solution and in solution containing byproducts ....................... 70

Comparison of isotherms for toluene, p-xylene and byproducts ;
p-xylene di-hydrodiol, 3,6-dimethyl pyrocatechol, and 2,5-dimethyl
phenol ...................................................................... 71

Efﬂuent proﬁles of target compounds and byproducts from BAC
column seeded with strain B1 ........................................... 72

Efﬂuent proﬁles of target compounds and oxygen consumption and
byproducts from bioﬁlm column grown on Baker product seeded
with strain Bl .............................................................. 73

Efﬂuent proﬁles of target compounds and byproducts from BAC
column and bioﬁlm column seeded with strain Bl after disconnecting
the supply of toluene (only p-xylene was provided during 24 hrs
period - shaded area) ..................................................... 75

Efﬂuent proﬁles of target compounds and byproducts from BAC
column and bioﬁlm column seeded with strain B1 after disconnecting
the supply of p-xylene (only toluene was provided during 22.5 hrs
period - shaded area) ..................................................... 76

Efﬂuent proﬁles of target compounds and byproducts from BAC
column and bioﬁlm column seeded with strain Bl after disconnect-
ing the supply of oxygen for 5 hrs (shaded area) ..................... 77

Degradation curves illustrating the degradation of toluene and p-
xylene by strain Bl ...................................................... 9O

Degradation curves illustrating the degradation of toluene and p-
xylene by strain X1 ...................................................... 91

Isotherms for toluene and p—xylene with Freundlich parameter
estimation ................................................................. 93

5-7.

5-8.

5-9.

5-10.

5-11.

5-12.

5-13.

5-14.

B-l.

xiv

Efﬂuent proﬁles of toluene and p-xylene for 1) BAC seeded with
strain B1, 2) BAC seeded with strain X1, and 3) GAC .............. 94

Efﬂuent proﬁles of byproducts, p-xylene dihydrodiol and 3,6-
dimethyl pyrocatechol from 1) BAC and 2) Bioﬁlm from columns
seeded with strain Bl .................................................... 96

Efﬂuent proﬁle of p—xylene from GAC column and accumulated

area graph of p-xylene, p-xylene dihydrodiol, and 3,6-dimethyl
pyrocatechol from BAC based on the concentration of molecular
carbon ...................................................................... 97

Efﬂuent proﬁles of target compounds and byproducts during shock
loadings by doubling toluene concentration in BAC column seeded
with strain B1 ............................................................. 98

Efﬂuent proﬁles of target compounds during shock loadings by
doubling toluene concentration in BAC and bioﬁlm columns seeded
with strain X1 ............................................................. 99

Efﬂuent proﬁles of target compounds during shock loadings by
doubling p-xylene concentration in BAC and bioﬁlm columns seeded
with strain B1 ............................................................. 101
Efﬂuent proﬁles of target compounds during shock loadings by
doubling p-xylene concentration in BAC and bioﬁlm columns seeded
with strain X1 ............................................................. 102

Efﬂuent proﬁles of target compunds in BAC columns after the
microbiological activity was eliminated ................................ 103

Scanning electron micrograph of carbon particles obtained near the
inﬂuent side of the reactor ................................................ 104

Scanning electron micrograph of carbon particles obtained near the
efﬂuent side of the reactor ................................................ 105

Scanning electron micrograph of the matrix of carbon particles near
inﬂuent side of the reactor ................................................ 106

Schematic details of isolation procedures .............................. 119

SEM of Pseudomonas sp. strain BI and strain X1 .................. 120

NOMENCLATURE

Fundamental quantities
L length
M mass, in general

MC mass of non-growth substrate
Mg mass of growth substrate

Mb mass of bacteria

English symbols

b ﬁrst order endogenous decay constant (T'l)
Cc concentration of solute at adsorption equilibrium (ML'3)

Hc Henry's constant

IL concentration of inhibitor in liquid phase (MgL'3)
kc maximum speciﬁc rate of utilization of the non-growth substrate in the absence of
growth substrate (Mch‘lT‘l)

kg maximum speciﬁc rate of utilization of growth substrate (MgMb‘lT‘l)
K1: constant representing the sorptive capacity

K1 inhibition coefﬁcient (MgL'3)

K33 half saturation coefﬁcient of the non-growth substrate (MCL'3)

Ks’g half saturation coefﬁcient of the growth substrate (MgL'3)

XV

xvi

n constant representing the sorption afﬁnity

qc mass of solute adsorbed per mass of adsorbent (MM‘I)
Sc concentration of non-growth substrate (MCL'3)

S g concentration of growth substrate (MgL'3)

SL concentration of substrate in liquid phase (MgL'3)

t time ('1')

TE" true biomass uansformation capacity in the absence of endogenous decay (Mch'l)
T5 growth substrate transformation capacity (McMg '1)

VL volume of liquid phase (L3)

VG volume of gas phase (L3)

X active organism concentration (MbL'3)

X0 initial concentration of active organisms (MbL'3)

Ym maximum yield or true growth yield (Mng‘l)

Y observed yield (Mng-l)

Greek symbols

11 speciﬁc growth rate (T '1)

CHAPTER 1
INTRODUCTION AND RESEARCH OBJECTIVES

1.1 Introduction, Background, and Problem Statement

1.1.1 Introduction

Discovery of pollutants in aquifers and water supply wells has led to the realization
that groundwater supplies can no longer be assumed to be pristine sources of drinking
water requiring little or no treatment. Groundwater quality has historically been largely
dependent upon nature. However, there are now numerous identiﬁable threats of serious,
widespread contamination. These threats are generally classiﬁed by their causes : 1) waste
disposal activities, 2) indusuial and commercial release due to accidents, 3) agricultural
operations, and 4) artiﬁcial recharge of ground water with contaminated water. Among the
many contaminants responsible for deteriorating groundwater quality, organic contaminants
are the most common health-threatening chemicals detected. The greatest difficulties in
groundwater remediation have been encountered at organic contamination sites. Much of
the groundwater contamination of this type is caused by leakage, spilling, or disposal of
organic liquids originally immiscible with water (nonaqueous-phase liquids-NAPLs) into
the ground. Subsequent dissolution of the NAPL and transport of the dissolved
constituents by groundwater is thought to generate many plumes. Dissolved organic
contaminants that are less soluble in water and that have a tendency to bind to aquifer
media, and thus retarded and move at rates slower than the rate at which groundwater

ﬂows.

1.1.2 Engineering Background

Once its quality has deteriorated, groundwater is either unusable or has to undergo
treatment before use, depending on the extent of the problem. However, in most instances
it is not practical to restore a contaminated aquifer to its former condition. Many
remediation techniques are available such as pump-and-treat, venting , and biorestoration.
It is difﬁcult to return aquifers to drinking-water quality, but restoration methods are often
able to lower the contaminant concentration significantly. Almost all remediation of

contaminated groundwater is based on groundwater extraction from wells or drains.

Among the many remediation techniques available, carbon adsorption is currently
the focus of renewed investigation because of its potential for removing many of the
contaminants being regulated under the 1986 amendments to the SDWA (Safe Drinking
Water Act). The US. Environmental Protection Agency (EPA) designated granular
activated carbon (GAC) treatment as a "Best Available Technology (BAT)" for treating a
variety of regulated synthetic organic chemicals. "BAT" implies that while the treatment

technique is not perfect, it is one of the best approaches currently available.

1.1.3 Biological Activated Carbon Process

In many cases, biological growth is observed during the operation of fixed-bed
granular activated carbon columns. This growth has previously been deemed undesirable,
and steps have been taken to combat it. Efforts to prevent the growth of microorganisms in
activated carbon adsorbers are often unsuccessful (Bishop et al. 1967). On the other hand,

many researchers now view biological activity on activated carbon as beneﬁcial.

 

3

The possibility of combining biological and physicochemical treatment was ﬁrst
proposed in the early 1970's by Weber et al. (1970; 1971; 1972) and has since been an
active research area. The European practice of BAC involves the deliberate growth of
microorganisms in the adsorber for the purpose of treating drinking water. This has lead to
the common use of the terms "Bioﬁlm Activated Carbon (BFAC)" or "Biological Activated
Carbon (BAC)" which describe these combined processes (Rice and Robson 1982). Since
BAC is not normally preceded by ﬁltration, it can function as both ﬁlter and adsorber.
When BAC does follow ﬁltration, its primary function is adsorption. In any case, the
carbon particles serve as a support medium for the biomass, and biological activity is the

major means of contaminant removal.

One of the major advantages claimed for BAC is that it can treat a broader range of
dissolved organics than granular activated carbon alone, including highly biodegradable but
non-adsorbable or slightly adsorbable organic compounds (Benedek 1980). A removal
technique that is right for one set of circumstances is not necessarily right for another. It is
imperative that time be taken to choose the right tool for the job and then to reﬁne and
optimize the selected technology for the particular conditions before implementation. BAC
is advantageous because it contains both physicochemical and biological treatment

elements.

Another proposed advantage is that there is an in-situ regenerating capability of the
carbon by the biological activity (bioregeneration of carbon). This can extend the operating
period, which is a most critical factor in deciding the cost of operation. BAC can also
effectively cope with shock situations by adsorbing high concentrations, then releasing
sorbed contaminants back into solution where they become available for microbial

degradation. One important application is the treatment of contaminated groundwater.

1.1.4 Problem Statement

Benzene, toluene and xylene (BTX) are among the many contaminants regulated
under the 1986 SDWA amendments. These three organic compounds are commonly
detected at sites contaminated by petroleum products. Of the 567 sites of environmental
contamination listed on the proposed priority list in the State of Michigan in 1989, 234 sites
(41.3%) were contaminated by benzene, toluene, or xylene or by some combination of the
three (DNR 1989). The number of sites contaminated by BTX has increased annually as
more contaminated sites are identiﬁed (Table 1-1). BTX are less strongly adsorbed to the
soil matrix than the aliphatic components in fuel. BTX are known to biodegrade aerobically
(Kincannon et al. 1983a; 1983b; Richards and Shieh 1986; Tabak et al. 1981) and to be
sorbed by activated carbon. For these reasons, BAC treatment can be an excellent approach

to take advantage of integrated physicochemical and biological methods.

Table 1-1. Sites of environmental contamination listed on the priority lists
in the States of Michigan - ACT 307 GDNR 1989)

 

sites contamina— total contamina-

 

year ted by BTX ted sites listed % BTX/total
1989 234 567 41 .3
1991 1198 g 2848 42.1
1992 1663 3396 49.0

 

1.2 Research Objectives and Overview

The overall objective of this study is to investigate the role of multi-substrate

degradation in the operation of packed-bed biological activated carbon systems under

 

 

5

various combinations of substrates and with different populations of organisms. Specific
objectives include characterizing the different biodegradation patterns for each organism
used and mathematically quantifying substrate interactions. Another objective is studying
the effect of a metabolic byproduct on sorption. A further objective is the qualitative
description of the integrated result of substrate interaction and sorption. This research is
divided into three phases. Each phase constitutes one chapter in the dissertation. A brief

description of each chapter, with its particular objectives, follows.

The ﬁrst phase, presented in chapter 3, covers the isolation of microorganisms and
degradation kinetics. The objectives of this phase of the study are: 1) to isolate BTX-
degrading organisms from a biological aetivated carbon system, 2) to determine kinetic
parameters for single substrate degradation, 3) to categorize the pattern of multiple
substrate utilization, and 4) to quantify substrate interactions in the biodegradation of paired
substrates (BT, TX, and BX). The information obtained from this phase was used in the

interpretation of later experimental data .

The second phase, presented in chapter 4, focuses on cometabolism of p-xylene
and the byproducts produced. The effect of these byproducts on the removal of target
compounds is studied. For this purpose, the identiﬁcation of byproducts is essential. This
identiﬁcation is accomplished by GC-MS analysis and by comparative studies using other
organisms producing known byproducts. The effect of the formation of byproducts on the
sorption of the major target compound is investigated using a bottle-point adsorption
isotherm method with and without byproducts. The formation of byproducts and their
effect in the reactor is studied in a packed-bed column. The role of adsorption in the
cometabolism-base BAC operation is investigated by a comparison of the dynamic loading

of growth substrate with that of non-growth substrate.

6

The third phase, presented in chapter 5, emphasizes the investigation of the effects
of different patterns of degradation on general BAC Operation. Also, the p0pulation effect
is studied to show the importance of speciﬁc populations of microorganisms in treating
particular compounds. How the treatment mechanism reacts with the dynamic loadings in

the system under different conditions of substrate degradation is also investigated.

The engineering signiﬁcance of the ﬁndings put forth in this dissertation is

discussed in the last chapter, along with recommendations for future studies.

1.3 Production, Use, and Properties of BTX

Benzene, toluene, and p-xylene are listed among the top 50 chemicals produced in
the United States during the last several years. Annual production of these three

compounds during 1987 through 1990 are shown in Table 1-2.

Table 12 Annual production of benzene, toluene, and p—xylene in the
United States during 1987-1990 (Chemical and Engineering News 1991).

unit : billions of lb.

 

Compound 1987 1988 1989 1990
benzene 11.57 11.81 11.94 11.86
toluene 7.00 6.45 5.8 3 6.10

J-xylene 5.16 5.60 5.34 5.20

 

Benzene is present n0t only in all crude oils but also in reﬁned oils. Benzene was

discovered by Faraday in 1825 and named "bicarburet of hydrogen". It was produced

 

 

7

commercially primarily by the coal carbonization industry until the late 1950's, when the
production of petroleum-derived benzene became its leading source (Ayers and Nuder
1964). At the present time, more than 90% of benzene is produced from this latter source
(Howard and Durkin 1974). Benzene is derived from petroleum through two processes:
catalytic reforming followed by solvent extraction and fractional distillation to remove
benzene and other aromatic hydrocarbons; and dealkylation of toluene or xylene (Ayers and
Nuder 1964). Benzene is an intermediate compound in the production of other chemicals,
including phenol, chlorobenzene, ethyl benzene, and cyclohexane. Benzene and chemically
transformed benzene products are used to make plastics, synthetic rubber, and numerous
polystyrene compounds. Benzene is also used as a solvent in paints and coatings, and in
the preparation of artificial and natural leathers. The major sources of benzene
contamination occur in (1) the production of chemicals and processings (2) coating
operations and (3) storage and transportation. Contamination can be caused by spills,
leaks, and efﬂuents from industries (U SEPA 1976). Benzene can be found in virtually any

waste from industries producing or utilizing benzene .

Toluene is a natural component of coal and petroleum (Kirk and Othmer 1963).
The primary producers of toluene in the US are the coke and petroleum industries. The
primary direct uses of toluene in indusuy are as a gasoline component and as a solvent. An

important indirect use of toluene is its conversion to benzene (Kirk and Othmer 1963).

Xylene is produced primarily from petroleum by the reforming process (Kirk and
Othmer 1963). Xylene is used mainly as a gasoline additive and as a solvent. It is not used
as extensively as benzene or toluene. There are three isomers of xylene: para-, ortho-, and

meta-xylene.

8

Benzene, toluene, and xylene act as neurotoxins. All three compounds affect cell
permeability with possible concomitant shifts in blood ions. Possible carcinogenic,
teratogenic, and mutagenic effects have been observed. Details of BTX toxicity can be

found elsewhere in toxicological studies (Dean 1985).

The physical and chemical properties of these three compounds are summarized on

Table 1-3.

Table 1-3. Physical and chemical properties of benzene, toluene, and p-xylene
(Verschueren 1983)

 

 

Compound benzene toluene p-xylene
CH3 CH3
Structural Formula Q Q
CH3
benzol methacide p-xylol
cyclohexatriene methyl benzene 1,4-dimethyl
Synonyms phene phenyl phenylmethane benzene
hydride toluol
Molecular Weight 78.1 1 92.10 106.17
Speciﬁc Gravity 0.8786 0.867 0.86
Solubility (20°C)
in water (mg/l) 1780 515 198
log Poet at 20°C 2.13 2.69 3.15
Henry's Constant
.2 4 .2601 .2967
(mg/l / mg/l at 20°C) 3 5
Melting Point (°C) 5.5 -95.1 13.26
Boiling Point (°C) 80.1 1 10.8 138.4

 

CHAPTER 2
BACKGROUND

This chapter provides a general summary of important ﬁndings regarding biological
activated carbon (BAC), the adsorption and biodegradation of organic compounds, and the
characteristics of benzene, toluene, and p-xylene relevant to their removal from
contaminated groundwater. The concepts underlying BAC and its research history are
described in the ﬁrst section, followed by two sections on removal mechanisms: adsorption
and degradation. The section on adsorption is not intended to be comprehensive; it is
focused on multi-component adsorption in granular activated carbon systems. On the other
hand, the section on degradation is divided into three sub-sections : 1) kinetics of microbial
degradation, 2) multi-substrate degradation, and 3) degradation of BTX and its pathway.

The last section brieﬂy reviews current research most directly related to this study.

2.1 Bioﬁlm Activated Carbon

2.1.1 Development of Bioﬁlm on Granular Activated Carbon

The development of a bioﬁlm can be observed easily on any surface submerged in
an aqueous environment. A bioﬁlm life cycle involves 1) transport and adsorption of
organic molecules and microbial cells to the surface, 2) attachment of microorganisms, 3)
microbial growth and metabolism, and 4) the detachment of the bioﬁlm (Trulear and

Characklis 1982; Characklis 1984). The surface structure of activated carbon is suitable for

10

the ﬁrst two steps in the bioﬁlm life cycle because of its morphological and adsorptive
characteristics (Pirbazari et al. 1990). Since activated carbon in water and wastewater
applications invariably comes in contact with microorganisms, the development of
microbial aetivity on activated carbon occurs naturally. The growth of bioﬁlm depends on a
variety of biological, chemical, and physical factors. Transport and interfacial transfer
processes probably control the ultimate thickness of a bioﬁlm in high hydrodynamic shear
-stress environments. Microbial activity within the bioﬁlm can inﬂuence the microbial
population distribution and product formation. Detachment mechanisms can be divided into
erosion (continuous removal of small particles from the bioﬁlm) and sloughing (detachment
of large fragments of bioﬁlm). Erosion is the result of hydrodynamic shear stress at the
bioﬁlm-water interface. On the other hand, the sloughing is triggered by transitions in the
environment. Sloughing can be observed at high substrate loadings or low shear stresses

(Characklis et al. 1990).

2.1.2 History and Deﬁnition

Weber et al. (1970) demonstrated that physicochemical treatment could meet the
more stringent demands of tertiary treatment. In their study, they suggesred the direct
application of physicochemical processes, speciﬁcally granular activated carbon (GAC), for
the treatment of wastes containing dissolved organic matter. With high organic loading,
biological activity occurred in the GAC systems. Their results demonstrated that biological
activity did not hinder adsorption, and it apparently enhanced the overall capacity for
removal of organics. Further work by Weber et a1. (1972) revealed the possibility of a
biologically extended-bed physicochemical treatment. They operated an aerobic expanded
adsorption system for nine months without regeneration. They attributed the adsorption bed

longevity to "bioregeneration" - the utilization of adsorbed organic materials by bacteria.

11

Organics adsorbed on adsorption sites are used by organisms, and the sorption sites
recover their adsorption capacity. They compared anthracite and activated carbon as
packing materials for support of biomass growth and concluded that activated carbon was
superior because of its shock-absorbing capability and its extended residence time for
acclimation. The early ﬁndings of Weber and co-workers have been the subject of
considerable debate by researchers in the subsequent years. Over the time, the term
biological activated carbon or BAC has come to be deﬁned as the simultaneous combination
of GAC adsorption of dissolved organic materials coupled with aerobic biological oxidation

of organic materials (Rice and Robson 1982).

2.1.3 Research on Bioﬁlm Activated Carbon System

Several publications have reviewed the technology of BAC (AWWA Committee
1981; Miller and Rice 1978; DiGiano 1981; Rice and Robson 1982). Research on this
subject started in late 1960's and continues today. Among the many directly and indirectly

related studies, the most important works have been chronologically reviewed below.

Studies regarding the microbial activity on the surface of activated carbon can be
found prior to 1970 but these are not generally considered to be part of the BAC literature.
Rather, they report difﬁculties in creating a bacteria-free activated carbon column and
suggest approaches for dealing with the bioﬁlm. Bishop et al. (1967) reported unexpected
increases in output turbidity and TOC from carbon beds. They explained that biological
activity, caused by the heavy load of BOD, was responsible. To prevent column failure,
predisinfection using chlorination or ozonation were recommended. However, Weber et al.
(1970) incidentally observed improvement of the treatment by the microbial growth on

activated carbon surfaces. The overall capacity for organic removal was enhanced and the

12

effective operation period was extended. This observation indicated the possibility for the

co—existence of microbiological and sorptive activities in the adsorption column.

After Weber et al. reported the positive effect of microbial activity in the adsorber,
other researchers started to examine these processes and demonstrated the practical
application of combined processes to various kinds of waste removal. Perrotti and Rodman
(1973; 1974) showed enhanced TOC removal from textile dye wastewater in a reactor
containing both mixed activated carbon and activated sludge as compared to either system
alone. They proposed that extracellular enzymes promoted high rate degradation of the
organic matter adsorbed on the carbon. Other applications include the treatment of reﬁnery
and petrochemical wastewaters by Ford and Buerklin (1972). They predicted that
'biological-carbon systems will probably be the most prevalent application of carbon for

reﬁnery wastewater treatment in the immediate future.

During mid-19705', BAC became the focus of increased research attention for
industrial waste applications. Techniques to improve the performance were developed by
including preoxidation steps before aetivated carbon treatment. Preozonation oxidizes a
fraction of the total organic carbon (TOC) there by reducing the organic loading and
rendering the load more readily degradable. In addition, the ozonation saturates the solution
with oxygen and subsequently satisﬁes the condition for aerobic biodegradation (Miller and

Rice 1978).

About the same time, many European water treatment practices applied ozone
coupled with activated carbon for reducing organics concentrations in drinking water (Van
Der Kooij 1976a; Van Der Kooij 1976b; Sontheimer 1978; Sontheimer et al. 1979). The

application of BAC to treat drinking water raised a question concerning the contamination

13

by bacteria. However, with post-chlorination, sufﬁcient disinfecting residuals to prorect

distribution system was achieved.

The addition of powdered activated carbon (PAC) to activated sludge was
developed during this period and commercialized under the name "PACT" (DeWalle and
Chian 1977; Rodman, Shunney and Perrotti 1978; Nayar and Sylvester 1979). PAC is
used to remove soluble organics and to aid clariﬁcation. The addition of PAC to primary
clariﬁers reduced the load of BOD and solids to downstream process and the addition to
activated sludge system controlled organic loading. The PACT process differes somewhat
from other BAC approaches, however, the use of two mechanisms in one reactor may

involve similar mechanism.

Studies during the 1980's focused on speciﬁc mechanisms and different working
conditions. A number of mathematical models were developed (Andrews 1979; Andrews
and Tien 1976; Andrews and Tien 1977; Andrews and Tien 1981; Chang and Rittrnann
1987; Tien and Wang 1982; Ying and Weber Jr. 1979) and operating experiences were
reviewed (AWWA Committee 1981; DiGiano 1981). Table 2-1 compares models
developed by different researchers. These models differ in their description of; 1) bioﬁlm
kinetic (growth and decay), 2) substrate transport, and 3) adsorption equilibrium. Each

model, however, yields similar patterns of target compound concentrations in the efﬂuent

from BAC systems.

Among the studies on mechanisms, Speitel used a unique approach to study
bioregeneration in the ﬁxed-bed adsorption column (Speitel 1985; Speitel and DiGiano
1987; Speitel et al. 1987; Speitel and DiGiano 1988; Speitel et al. 1988; Speitel Jr. et al.
1989). Radio-labeled chemicals were partially adsorbed on certain depths of activated

l4

 

 

 

0.38.. 8.8 .85.
é? c: 82.: maria 8» o: c: .590 3.— 2.353 _ a. 3292
Bees: .85.
8505 coast—U mczwsoa SEES—M
BEE o: 5:952“. 8&3 8a 8% e: an >88 03%; _ a. mew—U
b20358 .352
8.2 5635.
3. BE 2. 38: waste 8 e: 8 :8 £88 £55 _ .228.
53:: .858—
88._ 25m 388
Be Bx... o: Boa: 9.3.5 8» 2:38 8» execs 35n— _ x898
.85.
news: a: e: o: 8» 8» mo» _ 2 35m QC a. can d
8% 8.5 a. 25a: . .828 m .2523
O—HYAB— =O_m=.=6 hg ash—n—
53» can e: 5:235 8.5 w mg 8» mo» oEnta> _ . .E.
6863: t: N ”:02 d .v.
E; .55
Be .55.. we» gagged :MWEWB 8» 8.» e: a. ,8qu 03§> m 36%
6252
:23 . >88 30>»
Be Bx... o: gauged oﬁvwhe o: 8» o: a 8:02 03> — a. 9.;
w 52% . 862:6 852: EE comp—0&6
.538: :50 03:09:50 38:. was BEBE 5:256 35.5 3.8 83:3 806.25 2.333 3
mega:
SEEM 5.233 .8598“? ham—ab mam—z 832m .0 62 .822

 

Eng—9838 East? 3 £08.: .«0 2892.800 ._ -N 29mg.

 

 

15

carbon and regeneration was studied by measuring labeled carbon contents of efﬂuent
substrate, C02, and cells. Consumption of pre-adsorbed labeled carbon was observed as

the bioﬁlm developed

Some researchers questioned the enhanced performance in BAC and sometimes did
not observe enhanced removal (Lowry and Burkhead 1980; Maqsood 1977; Olmstead
1989; Peel and Benedek 1983).

2.1.4 Adsorption, Desorption and Biological Activity

Sorption and biological degradation are the basic removal mechanisms in the BAC
process. The beneficial effect of BAC results from a combination of adsorption and
biodegradation in a single reactor, processes which are complementary and synergistic.
This can be understood by considering the different types of organic compounds :
1)adsorbable and biodegradable, 2) adsorbable but non-biodegradable, 3) biodegradable
but non-adsorbable, and 4) non-adsorbable and non-biodegradable (Benedek 1980). Most
organic contaminants are in categories 1), 2), and 3) suitable for treatment in a BAC
system. BAC, which has both adsorption and biodegradation in one system, can be

claimed to remove broader spectrum of organic compounds than the alternative systems.

Adsorbing potentially toxic compounds onto activated carbon may also protect the
microbial system (Gardner, Suidan and Kobayashi 1988) As a general rule, many toxic
compounds are adsorbed easily while the compounds which are not adsorbable tend to be
readily biodegradable. The bioﬁlm may thus function as a ﬁltering device to remove easily
biodegradable compounds before adsorption. Speitel et al. (1989) conducted experiments
with mixtures of biodegradable and nonbiodegradable compounds in columns containing

pre-equilibrated GAC. Utilization of biodegradable compounds by microbial activity

16

reduced competition between biodegradable and nonbiodegradable compounds for
adsorption sites. They concluded that removal of biodegradable compounds from the

mixture through microbial activity may extend the service life of the GAC.

The adsorption capacity of activated carbon can smooth out periods of high inﬂuent
concentration, even in the later stages of operation when little adsorption capacity remains.
The effectiveness of BAC systems in buffering the ﬂuctuation of inﬂuent concentration or
shock loading has been proven experimentally (Chudyk and Snoeynik 1984). Buffering
occurs because of the adjustment in adsorption equilibrium brought about by an increase in
the feed concentration. During such a loading, biodegradable organics may pass through
the bioﬁlm and become adsorbed by activated carbon. When baseline conditions are again
established, the accumulated organics desorb and degrade as they pass through the bioﬁlm.

Thus, some adsorption sites can be regenerated.

Bioregeneration has been a controversial t0pic in BAC studies and its deﬁnition
may be different among different researchers. Some researchers hypothesized and noted
that there was no bioregeneration (Lowry and Burkhead 1980). Maqsood et al. (1977)
for example, compared the isotherms of the spent carbon before and after bioregeneration.
No difference in isotherm data was observed. However, more researchers have conﬁrmed
bioregeneration. Mechanisms which are most likely to make bioregeneration possible

appear to be desorption and the subsequent diffusion to an external bioﬁlm.

Speitel (1985) examined the rate and extent of bioregeneration of 1"‘C-labeled
phenol and p-nitrophenol that had been preadsorbed in a granular activated carbon column.
A recovery of radioactivity in the efﬂuent from the GAC column indicated that preadsorbed
compounds were desorbed and subsequently biodegraded. Biore generation was observed

to range from 8 to 15 percent for phenol and 5 to 22 percent for p-nitrophenol over a 10

17

day period. The potential for bioregeneration was observed to have increased as sorption
capacity and surface diffusivity increased. The author suggested two causes of substrate
desorption: (1) a reduction in bioﬁlm/GAC interfacial concentration resulting from substrate
uptake within the bioﬁlm, and (2) a reduction in the surrounding bulk liquid concentration

caused by biological activity in upstream portions of the GAC column.

Li and DiGiano (1983) observed higher biodegradation rates and speciﬁc growth
rates on GAC than on sand or coal. Their evidence for enhanced biodegradation and
higher speciﬁc growth rates pointed to utilization of adsorbed substrate rather than better
attachment on GAC. The enhanced speciﬁc growth rate also increased as particle size was
decreased. Dobrevski and Zvezdova (1989) investigated the effect of activated carbon
pore structure on the process of carbon regeneration. Based on their results, it seems that
the process of bioregeneration depends on the volume of pores with radii of 5-50 nm.
This study implied that the internal diffusion was an important factor for resupplying
sorbed substrate to the bioﬁlm. Other authors such as Kim et al. (1986), Andrews and
Tien (1976), Suidan et al. (1978) and Schultz and Keinath (1984), observed a similar

biore generation.

Understanding the combined mechanisms of biodegradation and adsorption
occurring in the same reactor is still limited. Whether and to what extent there is a

synergistic effect needs further study.

2.2 Multi-Component Adsorption

If each component in solution is adsorbed independently of the others, a

multicomponent adsorption system can be described by simply using single-solute theory.

18

In actuality, the solutes generally interfere with each Other. This requires an understanding
of equilibrium equation processes including interference effects (Cooney and Lightfoot
1966). Multicomponent adsorption was ﬁrst studied for gas-phase adsorption (Gariepy and
Zwiebel 1971; Kidnay and Myers 1966; Thomas and Lombardi I971). Gariepy and
Zwiebel (1971) studied the binary adsorption of a gas mixture onto ﬁxed beds. Even
though they showed some agreement between the experimental and predicted data, their
mathematical model did not include the effect of diffusional resistance within the adsorbent

particles which has been shown experimentally to be rate controlling for activated carbon.

One of the special characteristics in multicomponent adsorption is that there can be
an efﬂuent jump of the less adsorbable component, as shown on Figure 2-1. This jump is
caused by the displacement of the previously adsorbed component by a compound that is
more strongly sorbed. Gariepy et al. (1971) termed this less adsorbable compound the

"non-key component" as opposed to the more strongly sorbed compound "key

 

 

 

. l .l
We

05‘

C/Co

 

 

 

 

 

 

 

"a "m 200 300 £00 500
time

Figure 2-1. Example of efﬂuent jump from p-chlorophenol replaced by
p-nitrophenol (after Merk et al. 1980)

l9

component". The height of jump is affected by the inlet compositions of the two
components, and it increases with increasing bed length. Thomas and Lombardi (1971)
studied the binary adsorption of a benzene and toluene mixture onto a fixed bed of
activated carbon. They used a simple semi-empirical model with the assumption that
surface diffusion is a rate-controlling mechanism within the particle and is governed by
linear driving force kinetics. They correlated the adsorption of both components using a
separation factor to describe the relative afﬁnity of the adsorbent for toluene and benzene.
Cooney and Strusi (1972) tried to solve the two solute problem by considering two single
solute problems (which they called a pseudo-binary approach) and by solving them
independently. They used the linear driving force approximation to describe the diffusion

within the adsorbent.

Many researchers have used a Langmuir Isotherm model modiﬁed to describe
multi-component adsorption. However, Jain and Snoeyink (1973) proposed a model of
bisolute systems with the hypothesis that a certain amount of adsorption occurs without
competition. The basic assumption in the Langmuir equation is that the surface of the
adsorbent is homogeneous with respect to energy and that all sites are equally likely to
adsorb either solute. Adsorption onto activated carbon does not comply with the
assumption of constant site energies. Jain and Snoeyink's modiﬁed version provided a
reasonably good description of the data. Fritz et al. and Merk et al. (Fritz et al. 1980; Fritz
and Schlunder 1980; Merk et al. 1980) examined different models using experimental
results they obtained. They used a Freundlich isotherm. This equation has some
disadvantages such as the restriction of applicability to a limited concentration range, but it
is also suitable for highly heterogeneous surfaces. They used both single component batch
experiments and ﬁxed bed experiments, and compared the results of these experiments to

model predictions based on the assumption that diffusion is rate-controllin g.

20

The prediction of multicomponent adsorption performance was made (Crittenden et
al. 1985) using the simpliﬁed form of the ideal adsorption theory (IAT) which was
originally developed by Myers and Prausnitz (1965). The simpliﬁed form allowed the rapid
prediction of bi-solute adsorption. The deviation between the prediction and the
experimental result was signiﬁcant using only single solute data if there existed signiﬁcant
internal diffusion resistance and marked displacement. Predictions for more weakly
adsorbed solutes in bi-solute systems were less accurate than predictions for more strongly
adsorbed compounds. The possible prediction of bi-solute performance using single solute
data was evaluated. The deviations were reduced using data for preloaded carbon, which

gives the information about the displacement effect, along with single solute data.

2.3 Microbiological Degradation of Organic Compounds

2.3.1 Kinetics of Microbiological Degradation

Many factors inﬂuence the biodegradability of individual organic pollutants. One
way of comparing biodegradability is by comparing kinetic parameters. The most

commonly employed rate equation is that of Monod (1949), which can be expressed as:

__4_S__ #.....-S-X
dt Y-(KS+S)

(2-1)

where, rim/Y (which is often represented by k) is the maximum speciﬁc rate of substrate
utilization (mg substrate per mg biomass-min) at time t. X and S are active biomass
concentration (m g/L) and substrate concentration (mg/L). umax and Ks can be obtained by

integrating the equation (2-1) with known S and X values over a certain time period. Batch

21

experiments are generally easy to perform by measuring the biomass increase and substrate
utilization (Sundstrom and Klei 1979). In a batch experiment, the reactors containing the
pollutant of interest at different concentrations as the sole organic constituent are inoculated
with small quantities of biomass capable of performing the desired biodegradation. The
increase in biomass concentration in each reactor can be measured as a function of time and

analyzed to obtain um.

2.3.2 Multisubstrate Degradation

In the metabolism of multiple substrates, frequently reported degradation patterns
include diauxie (Monod 1942), simultaneous utilization (Deshpande, Chakrabartl, and
Subrahmanyam 1987), and competitive inhibition (Law and Button 1977; Strand, Bjelland,
and Stensel 1990). Diauxie or sequential utilization results from catabolite repression or
catabolite inhibition (Yoon, Klinzing, and Blanch 1977). The substrate consumption rate of
multiple-substrates is often affected by the interactions between enzymatic processes. An
enzymatic inhibitor will reduce the rate of an enzymatically catalyzed reaction by binding
with either free enzyme or with the enzyme-substrate complex (Grady and Lim 1980).
Three different types of inhibition are presented on Table 2-2. The mechanism, the

mathematical expression, and Lineweaver-Burk plots are compared.

Simultaneous utilization of substrates may or may not result in growth.
Simultaneous utilization of substrates that do not support growth is termed cometabolism
(Law and Button 1977). As deﬁned by Dalton (1982), cometabolism is the transformation
of a non-growth substrate in the "obligate presence of a growth subStrate or anOther
transforrnable compound". More generally, cometabolism also includes transformations by

resting cells if no growth results. The term "cometabolism" is often used with

22

5.55558 5.52.: n .0 .5888 .5252... u a. .5528 ass—25:025.: . Ev.

55.55.5255 38.3.... “no 5.8.2. 2.28% Eases... . E > .8... 5.53. H 5. 528055 5258055... 5 s. +

.8558 ESEEQEBEbESE . .mm .5558 5.3256835 . Hm
£5.32... N _ .5255 “m £29.50 2.3336535 . mm 65.53.... H m .2:an 8c . m 65:53 .5382 E J.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

”Q.
2. 9. .m. I .+ E
A.|UI+C.AD+ v: a .mu _+m 5:32.:
a n H .. a+mummum+m 03.595252
0 >
5:
\ Z. .W+....o+..e.
u . e u 6 .mm H .+ mm 553......
\ b; a+mummum+m 938.88:
.9 .
c.
>\ U + $1 + c. J. .m u .+ m 833......
. .9: u a. a + m u m. u m + m. 22.2.58
U >
.5232...
so... {Baggaocﬁ 2.56855 5525:5532 +5583— .5 on»...

 

8.3. a... es. 580. 8:32... 2:55 .e 88.538 .3. 2......

23

"cooxidation". Cooxidation is deﬁned as the process in which microorganisms oxidize a
substance without utilizing the energy derived from this oxidation to support growth
(Leadbetter and Foster 1959). The proof of cometabolism can be easily observed by the
disappearance of substrate and the accumulation of end product without an increase in
bacterial numbers (Horvath I972). The inability to grow on some substrates is not the
result of an organism's inability to attack the substrate but often results from its inability to
assimilate the products of oxidation (Foster 1962). Oxidation of the substrate may induce

the formation of enzymes capable of oxidizing other substrates.

Recently, several mathematical models have been developed for cometabolism
(Alvarez-Cohen and McCarty 1991; Criddle et al. 1990; Saez and Rittmann 1991; Schmidt
et al. 1985). Most of these expressions describe resting cell transformation. The important
concept in these models is introduction of transformation capacity (Alvarez-Cohen and
McCarty 1991; Saez and Rittmann 1991). Transformation of non-growth substrate gets its
reducing power from cells grown previously on growth substrate. Transformation of non-
growth substrate in the presence of growth substrate is not well understood. In the
presence of growth substrate, reducing power can be obtained from the growth substrate
itself (Saez and Rittmann 1991). Recently Criddle (1993) proposed a single unstructured
model to describe non-growth substrate transformation in the presence of growth substrate
by coupling the transformation of non-growth substrate to consumption of growth

substrate and biomass.

24

2.3.3 Degradation of BTX and Its Pathway

Degradation as a Single Component

Benzene and toluene have been known as moderately to easily biodegradable compounds
(Kincannon et al. 1983a; Kincannon et al. 1983b; Richards and Shieh 1986; Tabak et al.
1981). The ability of bacteria to utilize aromatic hydrocarbons for growth was ﬁrst
demonstrated in 1908 by Storrner who isolated bacteria which could grow on toluene and
xylene. In 1913, Sohngen reported the utilization of benzene by microorganisms. In 1968,
Gibson et al. isolated a strain of Pseudomonas putida that would grow with ethylbenzene
as the sole source of carbon and energy. This organism would also grow with toluene and
toluene-grown cells rapidly oxidized benzene to catechol. Claus and Walker (1964) isolated
a Pseudomonas species and an Achromobacter species that would grow with toluene as the
sole source of carbon and energy. p-xylene is also known to be oxidized by Nocardia

corallina (Jamison, Raymond, and Hudson 1969).

Degradation Pathway for BTX

Many researchers have studied biodegradation of BTX compounds (Dagley et al. 1964;
Dagley, Evans and Ribbons 1960; Davey and Gibson 1974; Davies and Evans 1964;
Davis, Hossler and Stone 1968; Evans 1963; Franklin et al. 1983; Gibson 1968; Gibson,
Mahadevan and Davey 1974; Hayaishi and Nozaki 1969; Hou ; Kitagawa 1956; Nozaka
and Kusunose 1968; Nozaka and Kusunose 1969; Omori, Horiguchi and Yamada 1967;
Omori and Yamada 1969; Omori and Yamada 1970a; Omori and Yamada 1970b; Ribbons
and Eaton 1982). The pathways for BTX are summarized on Table 2-3 along with the
microorganisms involved and references. The degradative pathway of benzene to catechol

and cis,cis -muconic acid was ﬁrst demonstrated by Marr and Stone (1961). In 1968,

25

 

 

 

 

 

 

 

 

2.3.8..» .0 8:08 0... 5.3 08.508058 0...; 000. x _.
02.09.02... 05.0.- s
5:00.893 .0..0-N. 7000630555
0 150560.. - -2 ._....ee...-e.m- a. - 25.0.. a
. . r \ E»
43. .020 2.. $5.02.... 55.0 ...m...m.wmm=e£ :0 20 I so so
\
:0 a.
:0 .
:0 I,
25 I 010 0:0
e=e=_o.e.;.._..-m.~
SEER 5:00.00 150.0% . -xx05»:€-m.~-n.5 - 0:030.
o3. .0 .0 003.0 .daeﬁegmi I
.30. 000033. 0.5 00.302 IO :0 \xt.
$0. .8203 .00.. 000.0 533.53.... :0 :0 la.
mggegmi f0 I 2.0 $6
$0. 00.00.53. .00.. 00.302 00.9.5020.» E Q Q RV Q
03%... 3:050 :00. IO
03. .0. .0 n. :0 1000 0:0 10.:0 £0
3.050.350 :0 \s.
0.5. S. .5 . . :0 Q
.00. m 0 .2 0023:0033“. :06 :0 In
I
02.0.0.0“ 00.00% 5.90.0.2 000000. 000025

 

 

 

8:000E00 Em .0 .3358 000.5055 .m-m 0.00...

26

 

 

 

8.§... 0.5. - 00200358 0.885..aos-0-:§.£0 -
.2808 0.8 0.0.50.0 8.8.0.52.
0.50.0-0 - 0.0.0.- 0 - 00.0.- Q - .050...- Q - 000.....- Q
ARK
:0. 0000.0 .000 A025 000050§00k 0:0 01m :85 :0010 m
0:0 0:0 0:0
0.00 20006000.. 00.202023 308.0305:
0.2.8.0090... - 5.0.0»..- Q - 4.0.06.9.- Q - b.0000...- Q -
IO :0
:0 n 0 W 0
OS. 0000.; B0 .0080 :000 :80 :00:0
$0. 0.00.5; .00 0.00300: 00.00 00.00.00.200
80900300.. 0.8 828.0 .283 .022.
wca. 0000000! .000 00.082 .0020- 0 - 0.0.0.- 0 - 000...»...0E- a - .5500.- Q .. 000.....- a
1m IONIO 10010 W
010 0:0
0009.003. 00.00% 30000.2 000000.. 00.00205

 

 

 

8.00000 .m-N 030-.-

27

Gibson found that cis - benzene glycol was an intermediate in the conversion of benzene to

catechol.

Microbial degradation of substituted benzene including toluene and xylene is initiated by
oxidation of the methyl group or the aromatic ring, depending on the organism. Oxidation
of the aromatic ring is similar to the degradation of benzene. Toluene is oxidized to 3-
methyl catechol. This pathway is reported by Claus and Walker (1964) and supported by
others (Gibson et al. 1970; Nozaka and Kusunose 1969).

The pathway of p -xylene degradation is also well documented. Two pathways for bacterial
oxidation of p -xylene have been reported. Gibson et al. (1974) showed the initial oxidation
products formed from p- and m- xylene by Pseudomonas putida 39/D by aromatic ring
oxidation. Gibson's proposed pathway of p-xylene produces diol and catechol. Similar
observation was made from Nocardia corallina (Jamison, Raymond, and Hudson 1969).
However, this strain further transformed 3,6-dimethyl pyrocatechol to a,0t'- dimethyl—
cis,cis muconic acid by ring cleavage. Oxidation of the methyl substituent is the second
case of degradation. This pathway is investigated more thoroughly than the aromatic ring
oxidation. Even though proposed pathways are described slightly differently by different
researchers, all researchers commonly found that p-xylene produces p-toluic acid as a

byproduct.

2.4 The Effect of Byproducts on Sorption

The presence of background organic matter can significantly affect the behavior of
target organic compounds in GAC columns (Summers et al. 1989). Metabolic byproduCts,

which are formed as a result of biodegradation, can also affect the sorption of target

28

compounds. Schultz and Keinath (1984) studied the adsorption of metabolic end product
(MEP) by powdered activated carbon during the degradation of phenol substrate. They
found 62 % of the MEP could be adsorbed by PAC and also observed this adsorbed
MEP was not desorbed by dilution or displaced by the target compound, phenol.
Chang's study (Chang 1985) with 14C labeled compound also showed competition
between microbial products and the target compound, acetate. Olmstead (1989) showed the
effect of microbial products on adsorption using bovine serum albumin (BSA) as the
representative protein. His results did not indicate a signiﬁcant difference between
isotherms in the absence or presence of BSA. On the other hand, the adsorption with GAC

preloaded with soluble microbial product (SMP) did show the different sorption patterns.

2.5 Multicomponent BAC System

Little is known regarding the performance of solute mixtures in BAC systems.
Multi-component systems involve both competitive adsorption and interference in
biodegradation. Speitel, Jr. et al. (1989) have studied the performance of bisolute mixture
in a BAC column using p-nitrophenol and trichloroethylene (TCE), under the assumption
that TCE is not aerobically biodegradable. This study showed the displacement effect of
one chemical by the others, but did not show biological interactions in the removal of these
two compounds. The assumption that TCE is not aerobically degradable may not be valid,
however. In 1985, Wilson and Wilson showed that aerobic metabolism of TCE occured
when soil microflora were exposed to natural gas in air. Recent studies by Nelson et al.
(Nelson, Montgomery, and Pritchard 1988; Nelson et al. 1987) showed that TCE can be
cometabolized in the presence of phenolic compounds. To study substrate interactions in
the BAC systems, kinetic characterization of each species should precede the BAC

experiment.

CHAPTER 3

KINETICS OF COMPETITIVE INHIBITION AND COMETABOLISM IN THE
BIODEGRADATION OF BENZENE, TOLUENE, AND p-XYLENE BY TWO
PSEUDOMONAS ISOLATES

published in Biotechnology and Bioengineering, Vol. 41, pp. 105 7-1065 (1 993)

3.1 Introduction

Benzene, toluene, xylene (BTX) are commonly found in soils and groundwater at
sites contaminated by gasoline and other petroleum products. Laboratory investigations of
biodegradation have often focused on single substrates alone, neglecting the fact that
environmental contamination by single substrate is uncommon. The vast majority of ﬁeld
situations involve degradation of multiple substrates. Complex degradation patterns are
frequently observed in such cases. Techniques are needed to quantify these kinetic

interactions to permit more rational design and operation of engineered systems.

In the metabolism of multiple substrates, frequently reported degradation patterns
include diauxie (Monod 1942), simultaneous utilization (Deshpande, Chakrame and
Subrahmanyam 1987), and competitive inhibition (Law and Button 1977; Strand, Bjelland
and Stensel 1990). Diauxie or sequential utilization results from catabolite repression or
catabolite inhibition (Voice et al. 1992). Simultaneous utilization of substrates may or may

not result in growth. Simultaneous utilization of substrates that do not support growth is

29

3O

termed cometabolism (Grady 1985). As deﬁned by Dalton and Stirling (1982), co-
metabolism is the transformation of a non-growth substrate in the "obligate presence of a
growth substrate or another transformable compound". More generally, cometabolism also

includes transformations by resting cells if no growth results.

A number of substrate interactions have been observed during hydrocarbon
degradation. Bauer and Capone (1988) reported a stimulatory effect of benzene on the
degradation of anthracene. Meyer, Marcus and Bergman (1984) observed diauxie
degradation of benzene in the presence of phenol. Growth on toluene induced non-speciﬁc
dioxygenase activity, permitting cometabolism of trichloroethylene (TCE) (Nelson et al.
1987). Goldsmith Jr. and Balderson (1988) evaluated BTX degradation kinetics in an
enrichment culture. They obtained ﬁrst-order kinetic parameters for individual compounds
and demonstrated differences in the rates of degradation among substrates. Interactions
between benzene and other aromatic carbons were evaluated by Arvin et al. (1989). These
researchers concluded that benzene degradation was enhanced in the presence of either
toluene or o-xylene. The effect of different substrate combinations were evaluated
statistically using factorial experiments. Recent studies by Alvarez and Vogel (1991)
revealed interactions during BTX degradation by organisms from aquifer enrichments and
by two pure-culture organisms. Factorial experiments were designed and interactions
evaluated by comparing changes in lag period and in pseudo zero-order biodegradation
rates. Induction, inhibition, and cometabolism were observed. These researchers observed
enhanced degradation of benzene and p-xylene in the presence of toluene by a
Pseudomonas strain. In none of these studies was a detailed quantitative interpretation of
substrate interactions attempted Quantiﬁcation is needed to facilitate comparison of cultures

from different sources, and to advance the design and operation of engineered systems

(Grady 1990).

31

The objectives of this study were: 1) to isolate BTX-degrading organisms with
different degradation characteristics from a biological activated carbon fluidized bed
treatment system, 2) to determine kinetic parameters for single substrate degradation, 3) to
categorize the pattern of multiple substrate utilization, and 4) to quantify substrate

interactions in the biodegradation of paired substrates (BT, TX, and BX).

3.2 Materials and Methods

3.2.1 Growth Medium and Chemicals

HCMM2 mineral salts medium was prepared as described elsewhere (Ridgway et
al. 1990). This medium has been previously used for the isolation of gasoline degrading
bacteria. Solid medium was prepared by addition of 15 g/L Bacto agar (Difco laboratories,
Detroit, Mich.) to the mineral medium. Reagent grade benzene, toluene, and p-xylene were

obtained from Aldrich Chemical Co., Milwaukee, Wis.

3.2.2 Isolates

BTX-degrading isolates were obtained from an aerobic pilot-scale activated carbon
ﬂuidized bed reactor located at Michigan State University. Details of reactor operation and
design are described elsewhere (Voice et al. 1992). After a three month period of
continuous operation treating BTX—contaminated water, activated carbon particles with well
developed bioﬁlms were removed from the fluidized-bed reactor. Approximately ﬁve
particles were placed in a 25-mL glass vial containing 10 mL of phosphate buffer solution
and ten 4 mm diameter glass beads. The vial was vigorously shaken and allowed to stand

for 15 minutes until all large ﬂoc had settled. A dilution series was then prepared from the

32

supernatant. Spread plates for each dilution were exposed to vapors of benzene, toluene,
or p-xylene in sealed chambers at room temperature (~24'C). After 48 hours, isolates were
picked, streaked, and re-incubated. This procedure was repeated 5 to 8 times. Colonies
isolated by growth upon vapors of a single pure compound, such as benzene, were then
tested for growth in chambers containing other substrates, such as toluene or p-xylene.
Growth of isolates was also evaluated in separate BTX vapor chambers in which the

substrates were paired (BT, TX, BX) or combined (BTX).

Two isolates, designated strains BI and X1, were selected for further investigation
based on their distinctive degradation patterns. Strain B1 did not degrade p-xylene. and
strain X1 did not degrade benzene. Each strain was characterized and identiﬁed on a GN
MicroPlate® test panel using Biolog's Microl.og® software. Strain B1 was identiﬁed as
Pseudomonas fragi, with a high conﬁdence rating in a MicroI.og® search; strain X1 was
identiﬁed as Pseudomonas ﬂuorescens, but with a low conﬁdence rating. Fatty acid
analysis performed by Microbial 1D, Inc., Newark, Del., conﬁrmed the Pseudomonas
classiﬁcation. Strain X1 had a high correlation with the fatty acid proﬁle for a library strain

of Pseudomonas ﬂuorescens.

3.2.3 Inoculum Preparation

Picked colonies were transferred to sealed 250-mL glass bottles with 100 ml.
HCMM2 liquid medium containing in the liquid phase: 10 mg/L of either benzene, toluene,
or p-xylene, or 5 m g/L of paired combinations of these substrates. Immediately after these
substrates had completely degraded (as indicated by GC analysis), the resulting cell

suspension (~15 mg/L) was used as an inoculum for batch experiments.

33

3.2.4 Kinetic Experiments

In all kinetic experiments, benzene, toluene, and p-xylene were assayed using a
headspace sampling method (Gossett 1985). Ninety-eight milliliters of HCMM2 mineral
media were added to a 250-mL glass bottle (VWR catalog no. 16151-300) and sealed with
a pressure-tested Teﬂon Mininert valve (Alltech catalog no. 95326). The bottle was
covered with aluminum foil and autoclaved for 20 minutes at 121°C . After cooling to room
temperature, BTX substrates were added to the bottles from stock solutions. The bottles
were then sealed and shaken for 12 hours. This procedure ensured complete dissolution of
the hydrocarbons and equilibrated the headspace with the solution. The liquid-phase
concentration at the initiation of an experiment was approximately 10 mg/L for each
compound. For the paired substrate experiments, the total carbon source was doubled.
Batch kinetic experiments were conducted using the inoculum directly (~15 ug/mL) or

immediately after degradation of a second spike of growth substrate (~30 ug/mL).

In experiments to assess substrate interactions and effects on growth, two milliliters
of the 30 ug/mL inoculum was transferred to the bottle leaving 150 mL of headspace. The
bottles were shaken on a rotary shaker at 100 rpm. Rates of mass transfer of BTX between
the gas and aqueous phases under these conditions (0.2 min-1, 0.16 min-1, and 0.12 min-1
for B,T, and X, respectively) were higher than the maximum biodegradation rates ensuring
rapid equilibration with the liquid phase and justifying use of gas—phase measurements for
biodegradation experiments. After inoculation, headspace gas samples (0.5 ml.) were
withdrawn from the bottle using a 1.0- mL Precision gas-tight syringe (Alltech catalog no.
050033) equipped with a side-port needle (Alltech catalog no. 943052), and assayed by
injection onto a Hewlett Packard 5890 gas chromatograph equipped with a DB-624

capillary column (30m x .549 mm, J & W Scientiﬁc) and a ﬂame ionization detector. The

34

carrier was helium (15 mL/min). Oven temperature began at 40°C and increased at a rate of

30°C/min to 200°C, where it remained constant for 2 minutes.

Gas sampling continued until growth substrate concentrations dropped below the
detection limit (~0.01 ppm). A sterile control was also assayed to assess abi0tic losses of
the target compounds. Growth was monitored with optical density measurement at 560 nm
using a Shimadzu UV spectrophotometer (path length = 1 cm). Biomass concentration was
estimated from a correlation of optical density (0. D.) to dry weight. For strain B1,
biomass concentration (mg/L) = (O.D. - .0046) / .0026 (98% correlation); for strain X1,
biomass concentration (mg/L) = (O.D. - .0060) / .0025 (99% correlation).

In experiments to assess competitive inhibition, the 15 ttg/mL inoculum was used
directly in the same bottle in which it had grown. After depletion of the original growth
substrate, paired combinations of substrates were added. Two substrate combinations,
benzene-toluene and toluene-xylene, were tested with strain B 1. In both cases, the amount
of benzene or xylene was held constant while the amount of toluene was varied (700,
1400, 2800 pg corresponding to ~5, 10, 20 mg/l in the liquid phase respectively). Three
levels of inhibitor were used (700, 1400, 2800 pg). The inhibitory effects of benzene and
xylene on toluene were evaluated by comparing the initial degradation rate of toluene.
Changes in the concentration of biomass and inhibitor over the initial period of degradation

(60 minutes) were insigniﬁcant (<5%).

3.2.5 Identiﬁcation of Byproducts

UV spectroscopy at 279 nm revealed the presence of one or more byproducts
arising from transformation of p -xylene by Pseudomonas strain B 1. Three UV-absorbing

products were subsequently separated on a Gilson high pressure liquid chromatograph

35

equipped with a fraction collector and a UV detector. These fractions were identified by
comparing their HPLC run time and elution patterns on thin layer chromatography with
known standards. Pseudomonas putida strain 39/D and Pseudomonas putida strain F1, two
strains known to transform p-xylene, were used to provide standards matching two of the
three peaks. Pseudomonas putida strain 39/D converts p-xylene to p—xylene dihydrodiol
and Pseudomonas putida strain F1 converts p-xylene to 3,6—dimethyl pyrocatechol (Gibson
1968; Gibson et al. 1970; Gibson et al. 1968; Gibson, Mahadevan and Davey 1974). The
third peak was identiﬁed as 2,5-dimethyl phenol using an authentic standard (Aldrich
Chemical Co., Milwaukee, Wis.). This identiﬁcation was conﬁrmed by ether extraction

and gas chromatography/mass spectroscopy as per Gibson et al. (1974).

3.2.6 Estimation of Decay Rates and Maximum Yields

Decay rates were measured in the absence of non- growth substrate (p-xylene) and
in its presence, at three different aqueous- phase concentrations (6, 9 and 12 mg/l) added
immediately after the growth substrate (toluene) had disappeared. Optical density and p-
xylene concentrations were monitored throughout the decay period (120 minutes). The
decay rate was obtained from the slope of a ln(X) vs. time plot. The maximum yield was
obtained by incorporating the sin gle-substrate toluene degradation data into the expression
ti = Ym(-dS/th)-b. This expression was solved by varying Ym to determine the best fit to

the growth curve.

3.2.7 Parameter Estimation for Single Substrates

A Monod expression was used to estimate kinetic parameters for single-substrate

biodegradation in batch experiments:

36

-dM = 1(5va
dt Ks + SL (3'1)

Equation (3-1) was modiﬁed using Henry's constants to correct for the continuous supply

of target compound from the headspace:

M=SL-(VL+HC-VG) (3-2)

Substituting equation (3-2) into equation (3-1) yields:

dM_ k-M-X-VL
dt KS-(VL+HC-VG)+M

 

 

(3-3)

The half saturation coefﬁcient, Ks, and the initial biomass, X0, were estimated by
non-linear parameter estimation using an integrated form of equation (3-3), where
X=Xo+Y-(Mo-M)/VL. Approximate values for X0 were obtained by measuring the optical
density of the inoculum. However, because model predictions were sensitive to the choice
of X0 (Figure 3-1), X0 was selected as a ﬁtting parameter, but was constrained to the 95%
conﬁdence interval for the inoculum as determined by 0D. measurements. Statistical
analysis was performed by the statistical package SYSTAT 5.0 (SYSTAT, Inc.). Ks values
obtained by non-linear parameter estimation using the analytical solution to equation (3-3)
were the same as values obtained by a least-squares minimization procedure using a

numerical solution for equation (3-3). Figure 3-2 illustrates the sensitivity of model

predictions to Ks values for a ﬁxed X0.

SUBSTRATE MASS (ug)

SUBSTRATE MASS (pg)

37

1200 -

1000 -

I

800 unl-
600 v-
400 4

200 -

I

 

 

 

TIME (hr)

Figure 3-1. Sensitivity test for different initial biomass XO (ug) with the same half
saturation constant Ks of 1 mg/l.

1200 a

1000 -

U

800 ‘-
600 '1-
400 --

200 -l-

 

 

 

TIME (hr)

Figure 3-2. Sensitivity test for different half saturation constant KS (mg/l) with the
same initial biomass of 30 ug.

38

3.2.8 Quantiﬁcation of Competitive Inhibition

Competitive inhibition was modeled with the following expressions (Machado and

Grady 1989) :

 

 

 

 

 

- X - V
_ (1:1 = (”31.1 ) . SLJ I L (3_4)
I Ks.1 -(1+ I—(L—i) + SM
[.1
S - X - V
-dM2 = (”maxj ). L,2 L (3-5)
at Y2

I
K -(1+ JiH S
5.2 KLZ L.2

To account for partitioning of substrate from the gas phase into the liquid phase,
equation (3-2) was substituted into equations (3-4) and (3-5). Ks, umax, and Y were
previously obtained from single-substrate experiments. KI,benzene was measured in a batch

experiment; Kmlme was estimated using the best ﬁt to the degradation curve.

For competitive inhibition between benzene and toluene, growth was described
using equation (3-6). A lag term was introduced numerically into this expression to

account for the observed lag between substrate removal and biomass production (see

Discussion).

X = x + Ybarune . (Mmmc — Mbenzme) + Ytoluaie ' (Mo,toluarc - Mtolucnc)

Vt VI. (3-6)

39

3.2.9 Quantiﬁcation of Cometabolism

The following expressions were used to quantify cometabolism. Details of this

model are provided elsewhere (Criddle 1993).

 

 

dM S
——J-=k . _§._ .)(.\/L (3-7)
dt ‘ KM+SII
-dM
_c1M.= Tg. ———£— +kc - —-§9—-— .x.v, (3-8)
dt X - VL -dt Ks... + S. ‘
-—dM _
gar", ——8- —b- 1., M (3-9)
dt VL-dt Tc VL-dt

Equations (3-7) - (3-9) were modified using equation (32) to account for
partitioning from the gas phase and solved using a second order Runge-Kutta numerical

solution. T: was estimated using non-linear curve ﬁtting procedures. ’1‘: was obtained by

monitoring biomass decay and loss of the non- growth substrate in the decay period.

3.3 Results

The parameter estimation results for single-substrate degradation experiments are
summarized in Table 3-1. Multi-substrate degradation studies revealed three different
substrate utilization patterns: no interaction, competitive inhibition, and cometabolism.

These patterns are given in Table 3-2 for each organism and for each substrate pair.

40

Table 3-1. Single compound biodegradation kinetic parameters for strains B1 and X 1.

 

 

 

Strains Substrates ( (1:381) (“3539 (rrlfgS/l)
Benzene(3) 8.05:3.10 1.043309 3. 1710.82
Bl Toluene(3) 13.03:} .83 1.22:.01 1.96:0.91
Xylene - - -
Benzene - - -
X1 Toluene(6) 10.84i2.77 0.99:.25 1.88:1.26
Xylene(3) 11851-173 0.52:.03 4.55:; 0.37

 

"‘ Numbers in parenthesis are the number of experiments run. Parameters estimated are shown 3; one
standard deviation.

Table 3-2. Degradation patterns for each combination by each strain.

 

 

Substrate Combination*
B/F T/X B/X
Strain B1 Competitive Cometabolic Cometabolic
inhibition of B by T degradation of X degradation of X
T was not affected Competitive X was not affected
Strain X1 and B was not inhibition of X by T and B was not
degraded degraded

 

’ B=benzene, T=toluene, and X=p~xylene

41

As shown in Figure 3-3, strain Bl removed benzene and toluene to concentrations
below the detection limit (.01 ppm) in a 12-hour period. Strain B1 grew faster on toluene
than on benzene, as evidenced by the larger limax values in Table 3-1.
Consequently, the degradation rate of toluene was faster than that of benzene as shown on
Figure 3-3. The rate of degradation of either benzene or toluene in the presence of the other
was slower than the degradation rate of either substrate alone. To evaluate the nature of
inhibitive interactions, a double-reciprocal plot (l/M vs. -1/(dM/dt)) was prepared for
toluene degradation in the presence of varied concentrations of benzene. As shown in

Figure 3-4, benzene and toluene are competitive inhibitors for strain B 1.

Substituting equation 3-2 into equation 3-4 and inverting the resulting expression yields:

 

1
K - 1+—L"— - V +11 -V
1 1 8.1 [ K11]( L C G) 1
———-—. ' + (340)
CM M1 1<,-x-vL k,.x.vL
dt

The competitive inhibition parameter KLbemnc was obtained graphically from the
Lineweaver-Burke plot described mathematically by equation (3-10) and illustrated in
Figure 3-4. A value of 3.10:0.12 mg/L was obtained. This value compares well with the
Ks,bemene value obtained in single substrate experiments (3.17i0.82, Table 3—1).
KLwlume was estimated by curve ﬁtting, using values provided in Table 3-1. A value of
1.71 mg/L was obtained. This value compares well with the value for KSMMC (1.963391
mg/L) obtained in single substrate experiments. The Ksjoluene value obtained from

competitive inhibition experiments (Figure 3-4) was 1.23 m g/L.

Unlike strain B1, strain X1 did not degrade benzene, whether it was added alone or

together with toluene or p—xylene (data not shown). In addition, the presence of benzene

BENZENE (119) SUBSTRATE MASS (119)

TOLUENE (pg)

42

1400
1200
1000 -'
800 a
600 -
400 -

1-2500

-- 2000

  
   
 
 
   
 

—I-— benzene ' 1500

U
1

—<>— toluene L 1000

V

+ biomass “500

(5") SSVWOIB ABC]

to
8

 

 

.4-
O

 

 

 

—l— benzene alone

—-0— with toluene

  

N #
O O
O O
1 1
V I

 

 

0 2 4 6 8 10 12 14 16

1 400
1 200
1 000
800
600
400

 
 
 
  

-—0-— toluene alone

—I— with benzene

 

 

o
‘
di-
un-
d
un-
b

TIME (hr)

Figure 3-3. Typical curves illustrating competitive inhibition in the degradation of
benzene and toluene by strain B1.

 

43

 

 

 

0.25 q-
0.2 ~-
2800 pg benzene
:3 0'15 I" 1400 pg benzene
'9
2
g .
'T 0.1 .. 700 ug benzene
/° ,_
ﬂ ,4. control
.05 --
p O l l J I
l l 1 I ﬂ
-0.0004 0 0.0004 0.0008 0.0012 0.0016

1/M

Figure 3-4. Lineweaver-Burke plot for the degradation of toluene alone
and with benzene by Strain B 1.

44

did not affect the degradation rate of toluene or p-xylene. Mutual inhibition was observed
in the degradation of toluene and xylene by strain X1. Strain X1 degraded p-xylene faster
than toluene. However, toluene had a greater effect on the degradation of p-xylene than p-

xylene had on toluene degradation, as evidenced by the smaller value of K5 for toluene.

Figure 3-5 illustrates cometabolic degradation of p-xylene in the presence of
toluene by strain B1. In single substrate experiments, strain B1 was incapable of
transforming p-xylene. However, p-xylene was transformed when combined with a
growth substrate, such as benzene or toluene. Transformation of p-xylene slowed
signiﬁcantly when the growth substrate disappeared, but increased when a growth
substrate (either benzene or toluene) was again added. Competitive inhibition experiments
with toluene and p-xylene provided no evidence for competitive inhibition between these
compounds in the concentration range studied. Products of the cometabolic transformation
were tentatively identiﬁed as p-xylene dihydrodiol , 3,6-dimethyl pyrocatechol, and 2,5-
dimethyl phenol.

Table 3-3 summarizes decay measurements for different concentrations of p-xylene
in the absence of growth substrate. The decay rate increased with increasing concentrations
of p-xylene, but appeared to reach a maximum value. Table 3-4 summarizes measured or
estimated parameters for modeling degradation of toluene and p-xylene by strain B1.

Similar observations were achieved for the degradation of benzene and p-xylene by strain

B1.

TOLUENE (pg) SUBSTRATE MASS (pg)

XYLENE (pg)

45

 

 

    
    
    

 

 

 

 

toluene
spiked
1400 T 3000
1200 - --2500
1000 blomass q _ 2000
800
##1500
600
400 - #- 1000
200 \ w 500
0 ° 3 r o
0 5 10 15 20 25
1400 .,
1200 -- —<>— toluene alone
1000 w
—o— toluene and xylene
800 «-
600 --
400 --
200 ~-
0 i
0 5 10 15 20 25
1400 T
1200 «-
1000 w
800 .l.
600 a —O— xylene alone
400 " + xylene and toluene
200 -ll-
0 I i i # ﬁ‘
0 5 10 15 20 25

TIME (hr)

Figure 3-5. Typical curves illustrating cometabolic degradation of xylene in
the presence of toluene by strain X1

(511) ssvw0le ABC]

 

46

Table 3-3. Decay rate of strain B1 with different initial p-xylene concentrations.

 

 

Initial p—xylene concentration (mg/l) biomass decay rate (day '1)
o 0.41*
6 0.68
9 1.21
12 1.19

 

* Endogeneous decay rate

Table 3-4. Measured and estimated parameters for modeling degradation of toluene and
cometabolism of p-xylene by strain B1.

 

 

Parameters Value

Measured parameters

pmax for toluene 13-03 day'l

Ks for toluene 1.96 mg/l

Endogeneus decay coefﬁcient, b 0.41 day'1

k for p-xylene a 0.71 mg xylene/m g biomass day'1
Estimated parameters

Ym for toluene b 1.27 mg biomass/m g toluene

T3 _ toluene transformation 0.45 mg xylene/mg toluene

capacity C

0

TE - biomass transformation 2-4 mg Xylene/mg biomass

capacity ‘1
Ks for xylene e ‘

a. measured graphically from the initial disappearance of non-growth substrate (xylene).

b. estimated from equation 3-10.

c. estimated by curve ﬁtting model of the degradation data for non-growth substrate (xylene) disappearance
while growth substrate (toluene) was present.

d. estimated by curve ﬁtting of degradation data for the non-growth substrate (xylene) after growth
substrate (toluene) had completely degraded.

e. Ks was not determined, but was assumed to be much less than the xylene concentration.

47

3.4 Discussion

The two Pseudomonas isolates demonstrated very different substrate preferences
for BTX. This observation calls into question the extent to which sweeping generalizations
regarding the ease of degradability of BTX can or should be made for mixed cultures,
especially in the absence of an intimate understanding of the processes and populations
present (Arvin et al. 1989). In this study, three different patterns of degradation were
observed in dual substrate experiments (no interaction, competitive inhibition, and

cometabolism), and these patterns required different modeling approaches for their

quantiﬁcation.

In conventional Monod modeling approaches, substrate utilization is tightly
(instantaneously) coupled to biomass production, as indicated by equation (3-4).
However, in the single-substrate batch BTX experiments, the disappearance of substrate
actually preceded biomass production. The same phenomenon was observed in some of
the dual-substrate experiments. Figure 3-6 shows how experimental data and model
predictions compare for the degradation of benzene and toluene by strain Bl. Using single-
substrate yield coefﬁcients and assuming a tight coupling of substrate disappearance to
biomass formation, the model predicted biomass appearance at an earlier point in time than
observed (solid line in Figure 3-6), suggesting that a time lag occurs between the uptake of
substrate and the production of biomass. The possibility that this lag was due to slow
air/water mass transfer was eliminated based on the high measured mass-transfer rates (data
not shown), and the fact that hindered air/water mass transfer would result in an apparent
lag in substrate consumption, not biomass production. An alternate explanation for the
time lag is that biomass formation results from the slow degradation of a BTX metabolite.
This hypothesis was supported by optical density measurements showing continued

increase in biomass after all of the substrate had disappeared from the system. By

48

 

 

 

 

biomass predicted __
1400 without time lag J 3000
- a"
1200 . "I _”_ 2500
'A
’3 1000 -- .;
c7? ' I . . -- 2000 g
to ,° biomass predicted _<
g 300 -- A: with time lag Q
' O
E I benzene .' -e 1500 g
< __ .’ >
03 60° ° toluene , g;
l— , A
2’3 A biomass ._ 1000 E
D 400 -- . V
U) .0
'I' A cl-
200 "" ’1 0 500
I
” ‘
0 5 10 15 20
TIME (hr)

Figure 3-6 . Comparisons between experimental data and model (estimated by Runge-
Kutta numerical solution) for competitive inhibition of benzene and
toluene by strain B1 (KS9!mzene = 3.17 mg/l, K5,,oluenc = 1.96 mg/l,

X0 = 50pg, KLberrzene = 3.10 mg/l, KI,toluene = 1.71 mg/l, and time lag =
70 min)

49

numerically incorporating a time lag into the model (dotted line of Figure 3-6), good ﬁts to
the biomass data were obtained using equations (3-4) - (3-5). Sharma and Ahlert (1977)
noted that while the Monod equation predicts an instantaneous coupling of substrate
degradation with growth, such a relationship is not observed in non-steady state systems.
Storer and Gaudy (1969) reported growth rate hysteresis in their system, and showed that

the Monod equation does not fully describe transient behavior.

In this work, Ks values obtained in single substrate experiments agreed remarkably
well with K] values obtained in competitive inhibition experiments. This observation
suggests that the inhibition coefﬁcient of a competitive inhibitor may be approximated by its
single substrate half saturation coefﬁcient as suggested by other researchers (Machado and

Grady 1989; Voice et al. 1992).

An important aspect of this work was the successful quantiﬁcation of cometabolic
transformation rates in the presence of growth substrate. Many studies have addressed
cometabolic transformation by resting cells in the absence of growth substrate (Alvarez-
Cohen and McCarty 1991a; Alvarez-Cohen and McCarty 1991b; Saez and Rittmann 1991).
This situation rarely occurs in practice, however. More often, the growth substrate is added
to enhance the degradation of non- growth substrate by supplying electrons and energy for
more rapid and extensive transformation. An important observation was the reduction in
yield for cells grown in the presence of p-xylene. This effect can also be expected for
cometabolizing populations (Bauer and Capone 1988), and agrees with predictions for the
integrated model used to quantify cometabolism (Criddle 1993).

The principal UV-absorbing byproducts were identiﬁed as p-xylene dihydrodiol
and 3,6-dimethyl pyrocatechol. The proposed pathway agrees with that proposed by

Gibson, et al. (1974) and is illustrated in Figure 3-7. After a long period of incubation or

50

CH3

OH

 

CH3

XYLENE 3,6-DIMETHYL PYRO-
P' p-XYLENE DIHYDRODIOL CATECHOL

l NONENZYMATIC

CH3
OH

CH3
2,5-DIMETHYL PHENOL

Figure 3-7. Pathway proposed for the oxidation of p—xylene (Gibson et al. 1974).

51

upon lowering the pH, 2,5-dimethyl phenol was produced. The formation of 2,5-dimethyl
phenol continued even after cells were removed from the solution by centrifugation (5,000

rpm) and ﬁltration through a 0.22 pm ﬁlter, and appears to be abiotic as suggested by

Gibson et al. (1974).

The integrated model for cometabolism used here predicts that the rates of decay are
dependent upon the concentration of non-growth substrate. Similar observations were
made by Janke and Ihn (1989) with aniline (non-growth substrate) on cells grown with
sodium acetate as a growth substrate. Séez and Rittmann (1991) observed that biomass
decay could be correlated with transformation of non-growth substrate during the decay
period. The same observation was noted in this work, that is, when the growth substrate
(toluene or benzene) was no longer present, further disappearance of non-growth substrate
(p-xylene) correlated with biomass disappearance (Figure 3-8). Production of UV-
absorbing byproducts also correlated with the loss in biomass (Figure 3-8 and Figure 3-9).
The non-growth substrate did not contribute to biomass formation, only to its loss. A loss
in biomass upon transformation of the non-growth substrate seems reasonable given the
dependency of dioxygenase systems upon cell reducing power. In initial attempts to
describe the transformation of p-xylene, a single transformation capacity term was used for
both the growth and decay periods. However, this approach did not describe the
accelerated transformation of p-xylene observed during the growth period. By adding a
transformation capacity term from the growth substrate (equation 3-8), a reasonable ﬁt to

the data for the whole degradation curve was obtained (Figure 3-8).

52

(on) ssvw0le AUG

 

 

 

 

 

1600 -- 1600
xylene
1-- o .
1400 "We" - -~ 1400
\ -. 0 . biomass

1200 e- 1200
a
3 1000 e- 1000
(D
to
<
5 800 -- 800
Lil
l-
E
,_ 600 -- 600
to
S
a) 400 -- 400

200 e- 200

0 0
0 5 10 15 20

TIME (hr)

Figure 38 Modeling of cometabolism - degradation of toluene and xylene by
strain B1 with parameters on Table 3—4. Initial concentration of each
substrate in the liquid phase is approximately 10 m g/l.

53

    
 

 

 

 

1600 ' . .. 0.5
1400 p-xylene byproduct o .P 0.45
0

-- 0.4
A 1200 up (<2
3 u. 0.35 I?)
w 1000 d- 0 (D
(2 0 h- 0.3 g
2 U)
E-J 800 -l- -i- 0.25 E
g toluene toluene 02 in)
i— 600 .. Spiked " ' 9.
g .. 0.15 '3
(D
ca 400 .. 3
.. 0.1 3

20° " ' .. 0.05

0
0 r, . ° : I i 0
0 5 10 15 20

TIME (hr)

Figure 3-9. Formation of byproducts as monitored by absorbance at 279 nm for
transformation of p-xylene with toluene as growth substrate by strain B1

54

3.5 Conclusions

Two Pseudomonas species with distinctive patterns of BTX degradation were
isolated. Kinetic expressions, incorporating competitive inhibition, a lag time, and coupled
cometabolic degradation were developed, and these expressions were used to quantify
complex substrate interactions. The results suggest that paired competitive substrate
interactions can be quantiﬁed using single-substrate kinetic parameters and batch
degradation experiments. Cometabolic transformation during the growth phase can be
modeled using two transformation capacity terms: one for the growth substrate, and a
second for the biomass. The results also suggest that modeling and interpretation of BTX

degradation in mixed-culture systems should be done with care, as differences in
transformation may be due to differences in the microorganism present and to the complex

enzymatic interactions possible within a single organism.

CHAPTER 4

KINETICS OF COMETABOLIC DEGRADATION OF p-XYLENE
BY PSEUDOMONAS SP. STRAIN Bl AND THE INTERACTION OF
DEGRADATION AND ADSORPTION IN FIXED-BED
BIOLOGICAL ACTIVATED CARBON SYSTI'EMS

4.1 Introduction

Benzene and methyl substituted benzene compounds are commonly detected in
soils and groundwater at sites contaminated by gasoline and other petroleum products.
Among the many remediation techniques available, carbon adsorption is widely used
because of its potential to remove many of the contaminants regulated under the 1986
amendments to the Safe Drinking Water Act (SDWA). The US Environmental Protection
Agency designated granular activated carbon (GAC) treatment as the "'best available
technology" for treating a variety of regulated synthetic organic chemicals. In operation,
it is commonly found that microorganisms readily colonize the carbon surface, a factor
which must be considered in maintaining such systems. Biological activated carbon
(BAC), exploits this phenomenon by encouraging organism growth and the resultant
degradation of organic compounds delivered to the column. By combining adsorption
and biodegradation in a single system, BAC has been shown to have the efﬂuent

protection capability of GAC and the efﬁciency of biological systems.

55

56

Most BAC research has focused on single substrates, despite the fact that
multiple substrate contamination is common in ﬁeld systems. It is known that complex
interactions occur between substrates in both adsorption and degradation processes.
Speitel et al. (1989) have investigated the performance of a BAC column receiving a
mixture p-nitrophenol and trichloroethylene (TCE). This study showed the sorption
displacement effect of one chemical by the other, however, complex interactions in the

degradation processes were not considered.

Adsorption interactions can be viewed in two ways. First, a variety of approaches
have been developed to describe adsorption dynamics of multiple substrates or solutes
(Fritz et al. 1980c; Digiano et al. 1980; Frick et al. 1980). These typically require an
understanding of both the mass-transfer processes and adsorption energies of all species
and a description of the interactions involved as solutes compete for adsorption sites. An
alternative approach is to focus only on one compound of interest and study its behavior
in a characterized background solution. For example, water known to contain humic
materials measured only as total organic carbon, has been shown to reduce the capacity of

activated carbon for certain target compounds (Summers et al. 1989).

Simultaneous biological utilization of substrates in multi-substrate systems may or
may not result in growth. When growth is not supported, the process is termed
cometabolism (Grady 1985). As deﬁned by Dalton and Stirling (1982), cometabolism is
the transformation of a non-growth substrate in the "obligate presence of a growth
substrate or another transformable compound". More generally, cometabolism also
includes transformations by resting cells if no growth results. The inability to grow on
some substrates is not the result of an organism’s inability to attack the substrate but often

results from its inability to assimilate the products of oxidation (Foster 1962). Oxidation

57

of the substrate may induces the formation of enzymes capable of oxidizing other

substrates.

Cometabolic transformation typically results in the production of byproducts that
are not further utilized by the cometabolizing population. These compounds have the
potential to interfere with adsorption in BAC systems, but no evidence was found in the
literature that this has been studied. In a previous study in this laboratory, the formation
of byproducts from the cometabolic degradation of p-xylene with toluene as a growth
substrate was shown (Chang et al. 1993). These by-products were not identiﬁed and
details of transformation kinetics were not investigated. The current study was designed
to investigate these unresolved issue relating to the cometabolic degradation of p-xylene

in BAC systems.

The aerobic degradation pathways for p-xylene have been well established by
others and can be classiﬁed into two groups depending on which part of the structure is
oxidized ﬁrst. The ﬁrst class involves the oxidation of the aromatic ring (Gibson et al.
1974; Jamison et al. 1969) while the second involves oxidation of a methyl substituent
(Davey and Gibson 1974; Davis et al. 1968; Evans et al. 1991; Nozaka and Kusunose
1968; Omori et al. 1967; Omori and Yamada 1969; Omori and Yamada 1970a; Omori
and Yamada 1970b). Gibson et al. (1974) showed that the initial oxidation products
formed from p-xylene by Pseudomonas putida 39/D were produced by aromatic ring
oxidation. This transformation was achieved by cooxidation in the presence of succinate
as a growth substrate. The proposed pathway is shown in Figure 3-7. Presently, more is
known about oxidation of methyl substituents than about aromatic ring oxidation. Even
though the pathways proposed by different researchers differ slightly, they all

hypothesized the formation of p-toluic acid and further degradation by ring ﬁssion.

58

The objectives of this study were to (1) identify and quantify byproducts from the
cometabolic degradation of p-xylene by Pseudomonas sp. strain Bl with toluene as a
growth substrate, (2) study the kinetics of byproduct formation, (3) study the effect of
byproducts on adsorption in batch and ﬂow-through systems, and (4) investigate the role

of cometabolic degradation in the performance of a ﬁxed-bed BAC system.

4. 2 Materials and Methods

4.2.1 Isolates

Strain Bl was isolated from an aerobic pilot—scale aetiated carbon ﬂuidized-bed
reactor, as described in the previous paper reporting the degradation kinetics of two
different BTX degrading isolates (Chang et al. 1993). This organism used benzene and
toluene as growth substrates, and it cometabolized p-xylene, in the presence of toluene.
The strain was characterized and identified as a Pseudomonas based on a GN
MicroPlate® test panel. This identification was confirmed by fatty acid analysis

(Microbial ID, Newark, Delaware).

4.2.21noculum Preparation

Picked colonies were transferred to sealed 250—mL glass bottles with 100 mL
HCMM2 liquid medium (Ridgway et al. 1990) containing 10 mg/L of toluene. After the
toluene had completely degraded, a second addition was provided. Immediately
following the disappearance of the second addition, an inoculum was withdrawn for

kinetic experiments. Two milliliters of grown cells (~ 0.3 mg/mL) served as the inoculum

59

in the batch degradation experiments. Five milliliters of concentrated pure culture of

strain B1 (~ 90 mg/mL) were used for column inoculation.

4.2.3 Batch Kinetic Experiments

Pseudomonas sp. strain B1 was transferred through a Mininert® valve into a
sealed 250-mL glass bortles with buffered HCMM2 mineral salt medium containing both
toluene and p-xylene. pH of the mineral medium was 6.9~7.1 before substrates were
added. The bottle was shaken on a rotary shaker at 100 rpm. Headspace gas samples (0.5
mL) were periodically withdrawn from the bottle using a 1.0 mL gas-tight syringe
equipped with a side-port needle, and assayed by injection onto a gas chromatograph
equipped with a DB-624 column (30 m x .549 mm, J & W Scientiﬁc) and a ﬂame
ionization detector using 15 mL/min of helium as a carrier gas. The oven temperature
began at 40°C and was increased at a rate of 30°C/min to 200°C, where it remained
constant for 2 minutes. Liquid samples (200 pL ) were also withdrawn and measured by a
high pressure liquid chromatograph equipped with a C13 reverse-phase column and a UV
detector at 280 nm. The mobile phase consisted of 40 % water and 60 % methanol with

0.1 % acetic acid and was delivered at a rate of 1.2 mL/min.

4.2.4 Identiﬁcation and Quantiﬁcation of Byproducts

The byproducts of cometabolism were collected from the batch kinetic
experiments by centrifugation at of a liquid aliquot at 5000 rpm for 20 min followed by
ﬁltration through a 0.22 pm Millipore ﬁlter to remove cells. The solution was then
concentrated by one of two methods. The ﬁrst involved the extraction of the byproduct
solution with ethyl acetate followed by drying in a rotary vacuum dryer as described by

Gibson et al. (1974). The concentrated sample was then redissolved in 1 mL of

60

methanol. In the second method the sample was frozen using dry ice and placed in a
vacuum freeze dryer until the water portion was completely removed. The dried sample
was redissolved in 1 mL of deionized water. Concentrated samples were injected into a
GC-MS for further analysis. Thin layer chromatography (TLC) was used to compare the
byproducts from strain B1 to those produced by Pseudomonas putida 39/D and
Pseudomonas putida F1 (provided by David T. Gibson, University of Iowa). The former
organism has been shown to accumulate p-xylene dihydrodiol (p-XDHD), while the latter
accumulates 3,6-dimethyl pyrocatechol (3,6—DMPC) (Gibson et al. 1974). Details of the
TLC method are provided elsewhere (Gibson et al. 1974; Zylstra et al. 1990).

Because of difﬁculties in obtaining commercially available p-XDHD and 3,6-
DMPC quantiﬁcation was made by producing byproducts from Pseudomonas putida
39/D and Pseudomonas putida F1. A known amount of p-xylene was added to batch
systems and the resulting concentration of the accumulated byproduct was
stoichiometrically calculated. This allowed for preparation of a standard curve for HPLC

analysis.

4.2.5 Pure Compound Adsorption Isotherm Experiments

Activated carbon (Filtrasorb-400, Calgon Corporation) was prepared by sieving to
obtain a 30x40 mesh fraction (average particle diameter of 0.5 mm). After sieving, the
GAC was rinsed with distilled water to remove oil and ﬁnes and then dried overnight at

100°C in an oven. The GAC was then stored in a sealed container until use.

Pure byproduct isotherms were performed by spiking a series of bottles containing
either Pseudomonas putida 39/D or P. putida F1 with different concentrations of p-

xylene. When complete degradation of p-xylene was observed, the solution in each bottle

61

was ﬁltered to remove cells, and the concentration of byproduct quantified. These
solutions were then used in bottle-point isotherms by adding a ﬁxed amount of activated
carbon (~5 mg), equilibrating by shaking at 100 rpm for 72 hours, and measuring the
ﬁnal byproduct concentrations. Isotherms for toluene and p-xylene were also performed

using stock solutions of these compounds.
4.2.6 Effect of Strain Bl Byproducts on Sorption of Target Compounds

Two liters of solution containing byproducts were prepared by degrading
approximately 28 mg of p-xylene by Pseudomonas sp. strain B1 with toluene as a growth
substrate. Approximately 50 mg of toluene was added incrementally until p-xylene
completely disappeared from solution. A control solution was prepared by adding the
same amount of toluene to a 2-liter bottle without p-xylene seeded with strain B1. Bottle
point isotherm experiments with activated carbon were performed by adding a series of
concentrations of toluene and p-xylene to a ﬁltered solution containing the byproduct and

a control (which did not contain any byproduct compound).

4.2.7 Fixed-Bed Column Experiments

Adsorption column experiments were conducted in 1.5 cm diameter columns to
study the effect of byproduct formation on the removal of target compounds in a packed
bed system. A schematic of the. experimental set-up is shown in Figure 4-1. To isolate
the effects of adsorption from biodegradation, parallel systems containing activated
carbon and baker product or non-activated carbon (provided by Calgon Corporation)
were employed. Baker product has been shown to provide a similar environment for the
growth of microorganisms, but has little adsorption capacity (Voice et al. 1992). This

medium was prepared in the same manner as the GAC.

62

+ efﬂuent

efﬂuent sampling port

(SS-way valve)

 

 

packed-bed
coluimn

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

inﬂuent sampling port
(CS-way valve)

®

 

in-Iirie
mixer
ﬁner peristaltic
(.22pm pore size) pump
. . .. .. vent (P'UGW
Ieieleiel with cotton)
substrate fed by
syringe pump

mineral media
(1/10 of HCMM2 media)

filter
(.22pm pore size)

‘—

org/gen

Figure 4-1. Schematic diagram of the column apparatus.

63

A bed depth of 2 cm and a ﬂow rate of 1 mL/min were used to maintain a
hydraulic residence time of approximately 2 min. The whole system was autoclaved
before inoculation and kept Sterile using ﬁlters. Prepared inoculum was injected into the
column and left overnight without ﬂow for attachment. Column operation was initiated
by supplying both substrates (toluene and p-xylene) and mineral medium (1/5
concentration of original HCMM2 medium). Unattached microorganisms were ﬂushed
out as soon as the column operation started. The columns were operated until steady-state
efﬂuent concentrations were reached. Additional column experiments involved various

inﬂuent perturbations such as the removing the supply of one of the target compounds or

oxygen.

4. 3 Results

Figure 4-2 shows a HPLC chromatogram of the solution containing byproducts
from the cometabolic degradation of p-xylene and the end product solution obtained from
the degradation of toluene with strain Bl. As the chromatograms show, the byproducts
were formed only when xylene was degraded. These peaks were initially designated
compounds A, B, and C corresponding to their order of appearance in the chromatogram.
Each of the three peaks were collected from the HPLC efﬂuent in separate aliquots and
compared to previously identified p-xylene degradation products using thin layer
chromatography after samples were concentrated. It was found that compound A
corresponded to the byproduct of p-xylene degradation by P. putida F1, both having an
Rf factor of 0.53. Compound B matched the byproduct produced by P. putida 39/D with
an Rf factor of 0.33. Compound C, which was obtained by decreasing the pH of a

solution containing compound B, produced the formation of a new spot at Rf = 0.84,

A->

‘3

*c

 

 

 

 

 

 

 

\"L—«wkv

Figure 4-2. HPLC chromatograms of two solutions: a) solution containing
byproducts from p-xylene degradation with toluene as growth
substrate and b) ﬁnal solution after degradation of toluene by
Pseudomonas sp. strain Bl. Peaks A, B, and C are byproducts
of p-xylene cometabolism.

65

which corresponded to that produced by 2,5-dimethyl phenol obtained commercially.

This identiﬁcation was conﬁrmed by compariSon of GC-MS spectra.

Based on these observations, it is proposed that compound A is 3,6-dimethyl
pyrocatechol (3,6-DMPC), compound B is cis-3,6-dimethyl-3,5-cyclohexadiene-1,2-diol
(p-xylene dihydrodiol, p-XDHD), and compound C is 2,5-dimethylphenol (2,5-DMP).

Figure 4-3 shows substrate and byproduct concentrations over the initial 60 hours
of a rate experiment. It can be seen that when the growth substrate (toluene) is
completely removed from solution, the rates of p-xylene degradation and p-XDHD
formation decrease. The formation of 3,6-DMPC began after the disappearance of
toluene and lagged behind the formation of p-XDHD. The same experiment is plotted
over the course of 300 hours in Figure 4-4, where it is compared to a similar study in
which the cells were removed from solution after 30 hours. With the cells present, the
concentration of p-XDHD slowly decreased from the maximum value reached during the
ﬁrst 60 hours, and the 3,6-DMPC continued to increase. In the system where cells were
removed, the 3,6—DMPC remained constant following removal, and the p-XDHD
decreased only slightly. After 30 days of extended observation, the p-XDHD was
completely removed from the system containing cells and the 3,6-DMPC and 2,5-DMP
increased signiﬁcantly. HPLC analysis of a sample in which the pH was lowered by
adding one drop of 4N sulfuric acid revealed that the peak of p-XDHD disappeared, and a
peak of 2,5-DMP appeared. The rate of formation of 2,5-DMP appears to correspond to
the rate of p-XDHD degradation.

A mass balance on p-xylene and byproducts during the kinetic experiment is
shown in Figure 4-5. Analysis of ﬁve replicates of this experiment indicated that the

measured concentrations of these four compounds represented 105 i 10% of the p-xylene

SUBSTRATE MASS (pg)

1 600

1400

1 200

1 000

800

600

400

200

66

 

 

 

Figure 4-3. Byproduct formation kinetics (initial 60 hrs) for
degradation of toluene and p—xylene.

-- 1000
. ,P‘, p-xylene dihydrodiol
I. . I - -
o‘ o I, b "DD _______ 0‘--- n- 800
l- , 53 Tail
0 I ’ ’
I ’ ’
I, ’ . ’
P I I” -- 600
. p I." 3.6-dimethyl
’ toluene ' P pyrocatechol
. I
i- D. o I. ‘ ‘ -' q .. 400
‘ I
' I
' I
'- ' o I
~ I
- ,’ , p-xylene ‘r 200
‘ 0 ,I . ,
' ’ .
I I .
I I I I I W1 : § 0
0 10 20 30 40 50 60
TIME (hrs)

(on) ssvw .LOFIOOUdAQ

p-XYLENE MASS (pg)

67

cells were removed
‘ from bottle II at 30 hrs.

 

 

 

 

 

1600 -- 1000
1400 0"
'- - 800
1200
1000 . l- 600
800
600 -- 400
400
‘- 200
2001
O iill ++:'+;+‘ :3 5%-.-.‘19.’ IL 0
0 50 100 150 200 250 300
TIME (hrs)

Figure 4-4. Byproduct formation kinetics (up to 300 hrs) for the
transformation of p-xylene (extended result of Figure 4—3).
bottles were compared. Bottle No. I ( ) contained
biomass until the end of the experiment. Cells were removed
from bottle No. II ( ------ ) after 30 hours (0 ; p-xylene,
D ; p-xylene dihydrodiol, I ; 3,6-dimethyl pyrocatechol, ;
2,5-dimethyl phenol).

 

(611) ssvw IOnOOUdAB

MASS BASED ON MOLECULAR CARBON (pg)

68

1400

1200

woo . . .
800 '

600

400 I I di Ithylvpyrocatecholi I

200

 

O 50 100 150 200 250
TIME (hrs)

Figure 4-5. Mass balance between p-xylene and byproducts based on the
mass of molecular carbon.

69

added on a molecular carbon basis. This conﬁrms the contention that further degradation

did not occur in these systems.

The effect of the byproducts on the adsorption of toluene and p-xylene by
activated carbon is shown in Figure 4-6. A slight decrease in the sorption of both
compounds was observed in the presence of byproducts, when compared to the control
solution. Adsorption isotherms for all of the individual byproduct and target compounds,
and the best-ﬁt Freundlich equations are shown in Figure 4—7. This data indicates that
2,5-DMP is readily adsorbed, comparable to that found for the most adsorbable target
compound, p—xylene. p-XDHD was sorbed somewhat less, falling below toluene. 3,6-

DMPC was not adsorbed to any signiﬁcant extent by activated carbon.

Inﬂuent and efﬂuent concentrations for ﬁxed-bed columns seeded with strain BI
and containing activated carbon (BAC system) and baker product (bioﬁlm system) are
shown in Figures 4-8 and 4-9, respectively. In the BAC system (Figure 4-8), both
toluene and p-xylene were almost completely removed during the initial period of
operation. As expected on the basis of adsorption alone, toluene breakthrough occurred
ﬁrst, followed by p-xylene. p-Xylene concentrations did not fall after breakthrough but
appear to increase at a slower rate and did not reach saturation (i.e. inﬂuent) levels as
would be expected from adsorption alone. In contrast, toluene and p-xylene were found

in the efﬂuent of the bioﬁlm system (Figure 4-9) from the onset of operation.

The behavior of the byproducts in the two systems was similar, in that both 3,6-
DMPC and p-XDHD were initially not found in the BAC system efﬂuent. In the bioﬁlm
system, p-XDHD appeared at the beginning of column operation, followed by 3,6-
DMPC, as would be expected on the basis of the proposed mechanism (Figure 3-7). It is
interesting to note that 3,6-DMPC appears before p-XDHD in the BAC system.

SORPTION CAPACITY (mg SUBSTRATE / 9 OF ACTIVATED CARBON)

250

70

 

zoo .. toluene

150.

100-

 

50

  
   

I control solution
[I byproduct solution

 

500

 

"II-

S 10 IS 20

 

400 .. WWW"e

300 -

200 u

 

 

 

 

100 A ‘ control solution
A A byproduct solution
0 . i i i :
O S 10 15 20 25

EQUILIBRIUM CONCENTRATION (mg/L)

Figure 4-6. Sorption isotherms of toluene and p-xylene prepared in control solution
and in solution containing byproducts.

SORPTION CAPACITY
(mg SUBSTRATE/g OF ACTIVATED CARBON)

600

500

400

300

200

100

71

 

 

,.
-521"
”1.0
‘r ":19”
A",‘._”. o
’0';"
I"'o’”6
-. , b,/
""’ #’
.oo'I’I' r—"”" '_—'
p' / ”’—’ "'-
-- 0" o/ D’J” ’0"'-’
‘ a I 'D’ l",-
AIO" II! "'I'
d", ’I D ’.,’ D toluene
I In A Ar'/'- .. A p-xylene
'/ A ' / ' A pxylene dihydrodiol
A Ar /A 2 . 0 3,6-dimethyl pyrocatechol
{ . 0:. 0 2,5-dimethyl phenol
O
r “ r. .r.. r r 1 r i
I I T I I I I I 1
0 10 20 30 40 50 60 70 80

EQUILIBRIUM CONCENTRATION (mg/L)

Figure 4-7. Comparison of isotherms for toluene, p-xylene and byproducts ; p-xylene

dihydrodiol, 3,6-dimethyl pyrocatechol, and 2,5-dimethyl phenol.

CONCENTRATION (mg/l.)

CONCENTRATION (mg/L)

72

 

  

 

 

    

 

 

  

 

 

 

 
 

 

 

 

 

 

 

 

 

 

 

 

    

 

 

° ‘°'”°"° (inﬂuent)
A p-xylene
8 , .
O toluene ’0‘
ettluent
A p-xylene( ) f M
0 100 200 300 400 500 600 700 800 900 1000
16
I 3,60imethyl pyrocatechol
0 p-xylene dihydrodiol
12 .. .. ............................................ .. _.,....,............... F
i ii\ -. , a
I./ Y
. Ji’
7 I,
II ' I
I
4 .,,..../....IIT ”TH“ x 1‘
° ‘ I
O 100 200 300 400 500 600 700 800 900 1000
TIME (hrs)

Figure 4-8. Efﬂuent proﬁles of target compounds (top) and byproducts (bottom)

from BAC column seeded with strain B1.

73

 

 

 

 

 

 

 

 

 

 

 

 

 
  

 

 

 
 

 

 

 

 

16 20
2 I33%;:(iniluent) - 00 uptake .o’q
A . o ' I. o / “
:12 ’:J ...... Q... -.. 8.. .. .. .Uo/ ‘ . ...... . ..... V .. “Ina .................... 15
E . J " ‘ " ‘II’ °
2 - O toluene
Q I . (efﬂuent)
,. 8 . A Ap-xylene 10
E
I—
z
w
3
o 4 5
0
° 0
0 1 00 200 300 400 500 600 700 800

16
'S, 12 ........... -....._
E
z I 3,6-dimethyl pyrocatechol
.Q o p-xylene dihydrodiol
I- ..
E 8 “ha! ........... v .................
" ’I
E I \ I,
cZJ f \-_‘ I-I
8 4 . ..................................................... . ................... ‘. .

I
0 a i I i i i I I
0 1 00 200 300 400 500 600 700 800
TIME (hrs)

Figure 4-9. Efﬂuent proﬁles of target compounds and oxygen consumption (top)
and byproducts (bottom) from bioﬁlm column grown on Baker product

seeded with strain B1.

(1/6w) NOIIde‘lSNOO 00

74

To further investigate the interactions between biode gradation and adsorption in
ﬁxed-bed columns, a series of experiments involving temporary changes in the inﬂuent
concentrations of columns that had reached steady-state operating conditions were
conducted. The effect of removing the supply of toluene for 24 hours is shown in Figure
4-10. A substantial increase in p-xylene concentrations occurred during this period in the
bioﬁlm system, whereas only a very minor increase was found for BAC. Efﬂuent 3,6-
DMPC concentrations fell during this interval but signiﬁcant levels were still observed in
the efﬂuents of both BAC and bioﬁlm systems. The decrease is somewhat more rapid
with the bioﬁlm system. Upon resumption of the toluene feed, a sharp increase in
efﬂuent toluene was observed in the bioﬁlm system, whereas this change and that of

other compounds monitored changed only slowly in the BAC system.

Removing the supply of p-xylene primarily affected the concentrations of
byproducts, as shown in Figure 4-11. Both 3,6-DMPC and p-XDHD levels dropped off
rapidly in the bioﬁlm system and somewhat more slowly in the BAC system. The
concentrations of both byproducts also returned to their previous levels more rapidly in
the bioﬁlm systems upon restoration of the p-xylene feed. It can be also be noted that the
concentration of p-xylene in the BAC system efﬂuent decreased only slightly during the
period of no p-xylene supply and then increased only slightly upon resumption. This is in
contrast to the rapid drop followed by a sharp rise in the bioﬁlm system.

When the supply of oxygen to the BAC and biofilm systems was interrupted,
byproduct concentrations fell in both systems, as shown in Figure 4—12. The decrease
was slightly slower in the BAC system. Both target compounds rose rapidly to inﬂuent

levels in the bioﬁlm system but increased only slightly in the BAC system.

CONCENTRATION (mg/L)

CONCENTRATION (mg/L)

75

 

10
Toluene
‘l—Not SuMM—ap BAC Column
8 .. _________ M‘ _____________________ .--- O toluene ______________
o p-xylene
TN I 3.6-dimethylpyrocatechol
' ' D pxylene dihydrodiol

 

 

 

 

 

 

 

 

 

 

 

 

10
Toluene .
‘—Not Sumw—F Biotilm Column
8 .. ____________________________________ ____ O toluene ____________ l
o pxylene
‘ I 3,6-dimethyl pyrocatechol
I, a p-xylene dihydrodiol
5 "' """" 5 ﬂ """""""""" "".,' """""""""""
+5n‘ ' h . .0 . o .
4 -i -------- ---.‘ 00 --------------------------------------------
_, ‘ E I--I-.‘_ a. . 'l- -------
2 m.- m..-- :3. ------.L’! ......... T :2} £-.k!-- - . -------:ﬁ--.
. 9 o u ".m “ 'ﬂ-"
%a'D-—~—- II-D-D‘D"--.
o - -- - ‘3 - -- -- . i J.
0 10 20 30 40 SO 60
TIME (hrs)

Figure 4-10. Efﬂuent proﬁles of target compounds and byproducts from BAC

column (top) and bioﬁlm column (bottom) seeded with strain B1 after
disconnecting the supply of toluene (only p-xylene was provided during
24 hrs period - shaded area).

CONCENTRATION (mg/L)

CONCENTRATION (mg/L)

10

76

 

on

 

 

O toluene
"" o pxylene '-

I 3.6-dimethyl pyrocatech0I

a p-xylene dihydrodiol
................... --.---------------------=;,.;1---
n- o— D- - —t:i-- ' ""

~‘~\ J BAC Column
‘\

 

 

 

 

 

 

 

_ D’
M

 

 

 

 

 

 

 

30 40 50 60
10
H4
' ‘I' N Tgybnfied ' \
o O u ' ‘
8 ..-s ................. ea ......... ,i ....... i, ...................
0 \ ,-
l \‘ -’o’
6 q--—------- I ---------------------------------- 1::- ............
l Biotilm Column
i O toluene
4 - ----------------------------------- o p-xylene
5 I 3.6-dimethylpyrocatech0l
.‘ I R a p-xylene dihydrodiol
2 "$— ----- . ' '

'P---D:---- i --------------------- I ----- '~ . -"' . -----------
U 0"? . o . ‘ﬂ- —, - _ _ _. - I
.AOJ °.::‘:" O'D-D- ''''''' '

0 . ° ° ° ° ° -° 4* i
0 10 20 30 40 50 60
TIME (hrs)

Figure 4-11. Efﬂuent proﬁles of target compounds and byproducts from BAC

column (top) and bioﬁlm column (bottom) seeded with strain B1 after
disconnecting the supply of p-xylene (only toluene was provided during
22.5 hrs period - shaded area).

77

 

‘l O
L I —‘—-’-_‘-F'—-'-—-"
8 I, f ’-
3 II ------------------- r- ----- 4 ------------------------------------ q
B NO 02 I
z q ____________ J. ________ ._ _ O toluene
Q 6 r ' '- ' i """" o p-xylene " ' ' ‘
z. 1 ,' I 3,6-dimethyl pyrocatechol
o: \ ' n pxylene dihydrodiol
*- i
2 h
LIJ
0
Z
O
O

 

 

 

 

 

 

 

 

 

CONCENTRATION (mg/L)

Biofllm Column

toluene

p-xylene .............
3,6-dimethyl pyrocatechol

p-xylene dihydgdiol _ __ _ __ _ _.

 

 

 

 

 

 

3O 40 50
TIME (hrs)

Figure 4-12. Efﬂuent proﬁles of target compounds and byproducts from BAC

column (top) and bioﬁlm column (bottom) seeded with strain B1
after disconnecting the supply of oxygen for 5 hrs (shaded area).

78

4.4 Discussion

When cells acquire proper enzyme and transformation capacity in the presence of
growth substrate, then they can utilize other non-growth substrate by cometabolism.
Recently, Criddle (1993) developed a model which considers cometabolic degradation
kinetics in both the presence and absence of the growth substrate. This conceptual
framework, in conjunction with the p—xylene degradation mechanism proposed by Gibson
(1974), provides the basis for interpreting the biodegradation rate data.

The degradation of toluene, the growth substrate, proceeds rapidly after an initial
acclimation period, as expected (Figure 4-3). The degradation of p-xylene to 3,6-DMPC
is a sequential two-step process (Figure 3-7). p-xylene degradation lags that of toluene
only slightly, and proceeds at a similar rate, until the supply of growth substrate is
depleted. This is matched by the production of p-XDHD, the first product in the
degradation sequence. Formation of the second degradation product, 3,6-DMPC, does
not begin until the growth substrate is depleted (Figure 4-4). In other experiments, it was
observed that with the continuous addition of toluene to solution, only p-XDHD
accumulated in the system and the addition of p- xylene to resting cells produced 3,6-
DMPC without accumulating p-XDHD even though the reaction was slow. It thus

appears that the second step occurs only in the presence of resting cells.

Two potential explanations can be offered for this behavior. 1) Enzymes obtained
in two situations are different. In the presence of growth substrate the organism can only
convert p-xylene to p-XDHD. In the absence of growth substrate, however, the organism
can convert either p-xylene or p-XDHD to 3,6-DMPC. 2) Enzymes are same in both

cases. However, the transformation of p-XDHD to 3,6-DMPC is competing with the

79

degradation of toluene. This competition diminishes when toluene is completely

degraded, so the enzyme can devote itself to the conversion of p-XDHD to 3,6-DMPC.

The formation of both p-XDHD and 3,6-DMPC is clearly the result of microbial
activity. Gibson et al. (1974) proposed that the formation of 2,5-DMP from p-XDHD
proceeds via a non-enzymatic pathway. It was found that this reaction was accelerated by
the addition of acid. 2,5-DMP was not observed at significant levels until after the
concentration of p-XDHD reached its maximum value. In systems containing cells for
the entire period, only 3,6-DMPC and 2,5-DMP were found in solution after 30 days.
2,5-DMP also appeared in the bottle without cells, however, the rate of the formation was
not as fast as that in the bottle with cells. It is not clear that the formation of 2,5-DMP

from p-XDHD is strictly an abiotic process as claimed by Gibson et al. (1974).

The slight reduction in sorption of both target compounds in the presence of
byproducts, suggests that the effects of degradation and adsorption in BAC systems will
be primarily additive rather than synergistic over short periods of time such as that used
to produce the byproduct solution for the adsorption experiments. It should be noted that
2,5-DMP is equally adsorbable as p-xylene, and thus may compete with this target
compound if it is formed in significant quantities. Since the rate of formation of 2,5-
DMP is somewhat slower than that of the other byproducts, we might expect accelerated

loss of adsorption capacity over longer periods of time.

In column studies, the additive effects of degradation and sorption can be inferred
by comparing the BAC and bioﬁlm systems (Figures 4-8 and 4-9). Adsorption serves to
remove both target compounds in the BAC system prior to development of the bioﬁlm.
The spike of toluene results from adsorption breakthrough, followed by development of

the degradation capacity. This spike does not occur with p-xylene, presumably because

80

the higher adsorption capacity of the carbon for this compound allows it to be retained
until degradation capacity is fully developed. It is interesting to note that the bioﬁlm
system appears to plateau at slightly better removal rates. This may result from the better
bioﬁlm development throughout the depth of the column due to the longer period of
exposure to the target compounds. Alternatively, the elevated toluene levels in the BAC
system may result from release of previously adsorbed material desorbing due to

displacement by the more adsorbable xylene.

Adsorption of the target compounds also appears to prevent the formation of
byproducts during the start-up period. In the bioﬁlm systems, we observe the formation
of p-XDHD shortly after start-up with the first evidence of p-xylene degradation. As
would be expected from the mechanism, this is followed by the formation of 3,6-DMPC.
It is interesting to note that in the BAC system, 3,6~DMPC is observed in the column
efﬂuent prior to p-XDHD, even though the former compound is the degradation product
of the later. This can be explained by considering that the carbon exhibits some
adsorption capacity for p-XDI-ID but not 3,6-DMPC, and thus the p-XDHD is retarded in

its release from the column.

The interactions between cometabolic degradation and adsorption under pseudo-
steady-state conditions (well developed bioﬁlm) can be seen more clearly in the studies in
which one of the target compounds or oxygen is removed from the inﬂuents of the two
columns (Figures 4-10 - 4-12). Because p-xylene degradation requires a growth
substrate, this reaction could not proceed during the period of no toluene feed in the
bioﬁlm system, as evidenced by the rapid increase of p-xylene in the efﬂuent (Figures 4~
10). Resumption of the toluene feed restored xylene degradation, although a signiﬁcant
lag was observed in the bioﬁlm column before either p-xylene or toluene degradation

returned to previous levels. During this period p-xylene degradation continued at nearly

81

the previous rate in the BAC system, presumably because the previously adsorbed
toluene desorbed and provided the means for the cometabolic process to continue. This
result must be inferred, however, because desorbed toluene was degraded and thus similar
low levels of toluene were found in the two systems. Other concentration changes were
also dampened by adsorption in the BAC system, and only a slight increase in toluene

concentration was observed upon resumption of the feed.

This mechanistic interpretation is substantiated by the study in which the p-xylene
feed was temporarily interrupted. In the bioﬁlm system the levels of both p-XDHD and
3,6-DMPC decreased substantially, but stabilized at non-zero levels. This may be caused
by the utilization of p-xylene trapped in the system either by the slight adsorption
capability of Baker products or by cells as biosorption. The concentrations of p-xylene
and both byproducts decreased only slowly in the BAC system. This conﬁrms that the
activated carbon serves as a reservoir, in this case of p-xylene, which allows the
degradation reaction to proceed. Resumption of the p-xylene feed produced
concentration spikes in the bioﬁlm but not the BAC system, similar to those observed for

the toluene interruption.

The conclusion that activated carbon dampens concentration changes is further
supported by the study in which biodegradation is stopped completely for an interval by
not supplying dissolved oxygen to the columns (Figure 4-12). In the bioﬁlm system, both
target compounds increased to inﬂuent levels and byproducts decrease to zero, very
rapidly after the oxygen supply is eliminated. In contrast, only a slight increase in
toluene and p-xylene concentrations was observed in the BAC system because in the
absence of biodegradation, adsorption becomes operative and dampens the system

response. It was noted that the byproduct concentrations fell somewhat more slowly in

82

the BAC system, perhaps because the p-XDHD desorbed from the carbon and
degradation to 3,6-DMPC proceeded in the absence of oxygen.

4.5 Conclusion

Pseudomonas sp. strain Bl cometabolically degraded p-xylene in the presence of
toluene as a growth substrate. Three different byproduct compounds were produced. p-
XDHD was formed by growing cells and a further degraded compound, 3,6-dimethyl
pyrocatechol was formed by resting cells. 2,5-dimethyl phenol was also formed. The

formation was faster in the bottle with cells than the one in the bottle without cells.

The byproducts of cometabolic poxylene degradation reduced the sorption of the
target compounds only slightly. It was found that the performance of BAC column
systems could be explained by considering the effects of degradation and adsorption to be
additive but not synergistic. Activated carbon serves as a reservoir for those compounds
and byproducts that adsorb, and thus induced concentration changes are dampened. This
was exhibited in 1) removal of target compounds and byproducts prior to full
development of a biofilm, 2) continued cometabolic degradation in a period when the
growth substrate was not supplied to the column, 3) dampening of concentration spikes
following periods where the supply of target compounds was interrupted and 4) reduced
efﬂuent levels of target compounds when biodegradation was eliminated altogether due

to lack of oxygen.

CHAPTER 5

EFFECT OF MICROBIAL PROCESS ON THE OPERATION OF BIOLOGICAL
ACTIVATED CARBON SYSTEM DURING SHOCK LOADINGS

5.1 Introduction

Biological activated carbon (BAC) systems have been shown to provide superior
performance to conventional activated carbon adsorption systems. The capability to
biologically degrade substrates can reduce the sorption load and extend the service time of
the reactor. The adsorption capability, in return, can add the capacity to minimize the effects

of shock loading on microorganisms during a perturbation event.

The deﬁnition by Rice and Robson (Rice and Robson 1982) describes BAC as a
simultaneous combination of granular activated carbon adsorption of dissolved organic
materials coupled with aerobic biological oxidation of organic materials. In most BAC
studies, a strictly aerobic environment is maintained by providing excessive amounts of
oxygen throughout the system (Chang and Rittmann 1987; Olmstead 1989; Speitel et al.
1988; Speitel et al. 1989; Voice et al. 1992). As a result, the adsorptive characteristics of
the BAC system are often neglected by the active biological degradation. During the steady-
state operation of a BAC ﬂuidized-bed system with excess oxygen, Voice et al. (1992)
observed that the major removal mechanism in the system is biodegradation. They also

found that sorption became important during changes in inﬂuent concentration. This is

83

84

clearly one of the advantages of the BAC system. However, how different types of
substrate degradation processes can affect the operation of BAC systems treating more than

a single substrate has not previously been shown.

Different types of microorganisms have different capabilities for degrading
substrates. These capabilities can be quantitatively expressed by kinetic parameters which
show how fast they utilize substrate, how efﬁciently they grow by utilizing substrate, or
how fast they grow. Pseudomonas sp. strain BI and strain X1 were previously isolated
from BAC reactors treating benzene, toluene, and p-xylene (Chang et al. 1993). Strain B1
degrades benzene and toluene as growth substrates and also cometabolically degrades p-
xylene in the presence of either growth substrate. Resting cells of strain B1, which were
previously grown on benzene or toluene, can also degrade p—xylene by cometabolism.

Strain X1 degrades toluene and p-xylene; however, it can n0t degrade benzene.

In this research, the effect of different degradation patterns on the operation of BAC
systems was studied. Strain Bl and strain X1 were inoculated in two reactors and two
substrates were fed to show how the BAC system seeded with different species of
microorganisms would respond to the same combination of multiple substrates. A control
was run by inhibiting any microbiological activities (designated GAC). The appropriateness
of the application of batch degradation data to column operation was compared.
Experiments to investigate the role of adsorption during perturbation were performed by
applying different shock loadings and observing the performance of the reactor. The shock
loading was applied by increasing one of the substrate concentrations two-fold. The role of
adsorption during the failure of microbial activity in the system was studied by removing
the supply of oxygen. Cross-sectional matrices of the column after 40 days of operation
were examined by the scanning electron microscope (SEM) to investigate the population

distribution of microorganisms at different locations in the column.

85

5.2 Materials and Methods

5.2.1 Isolates

Two Pseudomonas isolates, designated strains BI and X1, were previously
isolated based on their distinctive degradation patterns (Chang et al. 1993). Each strain was
characterized and identiﬁed on a GN MicroPlate® test panel using Biolog's MicroLog®
software. Strain B1 was identiﬁed as Pseudomonasfragi, with a high conﬁdence rating in
a MicroLog® search; strain X1 was identiﬁed as Pseudomonas ﬂuorescens, but with a
low conﬁdence rating. Fatty acid analysis performed by Microbial ID, Inc. (Newark, Del.),
conﬁrmed the Pseudomonas classiﬁcation. Strain X1 had a high correlation with the fatty

acid proﬁle for a library strain of Pseudomonasﬂuorescens.

5.2.2 Inoculum Preparation

Pure colonies of strain B1 and strain X1 were picked from the previously grown
nutrient agar plate and transferred to sealed 250-mL glass bottles with 100 mL HCMM2
liquid medium (Ridgway et al. 1990) containing toluene and p-xylene, respectively. After
each substrate had completely degraded, a second addition was provided Additions of each
substrate were continued until concentrated pure cultures of strain BI and strain X1

(approximately 90 ug/mL) were obtained. Five milliliters of culture were seeded in the

column.

86

5.2.3 Batch Kinetic Experiment for Degradation

A detailed method for the kinetic experiment is described elsewhere (Chang et al.
1993). In this research, degradation of toluene and xylene by strain B1 and X1 were

compared. Each substrate was assayed as single substrate or combination.

5.2.4 Packing Media

Two different media, GAC and a nonadsorbent cabon material (Baker product),
were used in this study. The GAC used was Filtrasorb 400. Baker product is the same
material as GAC, but has not been activated and has therefore little adsorption capacity.
Both media were provided by the Calgon corporation (Pittsburgh, PA). Both media were
prepared by sieving to obtain a 30x40 mesh fraction. The average particle diameter was 0.5
mm. After seiving, the GAC was rinsed with distilled water to remove oil and ﬁnes and

then dried overnight at 100°C. The GAC was then stored in sealed container until use.

5.2.5 Batch Isotherms for Toluene and p-Xylene

Isotherm assays for toluene and p-xylene were conducted using serum bottles (160
mL) sealed with teﬂon-coated septa. Twenty milligrams of sterilized activated carbon were
used with the desired initial concentration of toluene and p-xylene dissolved in HCMM2
mineral medium. The serum bottles were shaken (6 rpm) at 24'C for 5 days to reach
equilibrium. Controls were run without activated carbon to ensure that there was no
biological degradation. The liquid-phase concentration was then analyzed to determine the
residual concentration of toluene and p-xylene. The solid-phase concentration was

calculated using a mass balance. Isotherms for the Baker produCt were not prepared.

87

However, previous studies (Voice et al. 1992) showed that sorption capacity by the Baker

product was negligible.

5.2.6 Fixed Bed Reactor

A ﬂow-through experiment was conducted using 1.5 cm diameter columns
(OmniFit USA, Atlantic Beach, NY). A complete description of the set-up is provided in
Chapter 4 and shown in Figure 4-1. Two different packing media, activated carbon and
Baker product, were used to compare biofilm on both activated carbon and inert media. The
bed depth was 2 cm and the ﬂow rate was 1 ml /min to maintain a hydraulic residence time
of approximately 2 min. The whole system was autoclaved before seeding and kept sterile
using ﬁlters and autoclaving. Prepared inoculum was injected through the column, left
overnignt for attachment, and ﬂushed with mineral medium to remove residual inoculum
solution. Pumping was started by supplying mineral medium solution (1/5 concentration of
original HCMM2 midium) in which toluene and xylene had been dissolved using a syringe
pump (Harvard Apparatus, Model 22) and inline mixer. The commencement of pumping
was deﬁned as time zero. Five columns were operated simultaneously for parallel
comparisons. One GAC column was operated abiotically to evaluate adsorption. Two sets
of two columns each were packed with activated carbon and Baker product. One set of two
columns was seeded with strain B1 and other two were seeded with strain X1. The steady-
statc columns were operated until the GAC column reached complete exhaustion and the
biological columns reached steady-state (both BAC and bioﬁlm). During this experiment,
the effect of different microbial populations on the operation of BAC and bioﬁlm column
was studied by monitoring the efﬂuent concentrations of the target compounds. Also,
concentrations of byproduct from columns seeded with strain BI were measured to monitor

cometabolism.

88

5.2.7 Perturbation Experiment

Several perturbation experiments were performed to obtain information regarding
the role of each substrate, the role of each mechanism (adsorption and biodegradation), and
the role of different degradation patterns (cometabolism and competitive inhibition) in the

system. The following shocks were applied to each column :

1. The effect of increased inﬂuent concentration of toluene ; which showed a
different degradation rate and a different effect compared to p-xylene during the
degradation by each strain ; was studied in this experiment. Toluene is a growth
substrate for both strains. Strain B1 utilizes toluene faster than it does p-xylene.

However, toluene is less preferably utilized to p-xylene, by strain X1.

2. A similar experiment to the above shock loading with toluene was performed by
doubling the inﬂuent concentration of p-xylene, which is sorbed more strongly than

toluene.

3. Nitrogen was used to supply the mineral media instead of oxygen to eliminate

obligate aerobic activity.

5.2.8 Analytical Method

BTX liquid samples (1 ml) from column efﬂuents were collected in 22 m1 head-
space vials and assayed by automatic headspace analysis. The analysis system included a
Hewlett Packard HP 19395A automatic headspace sampler and a Perkin-Elmer Autosystem
gas chromatograph, equipped with a DB-624 capillary column (30 m x .549 mm, J & W

89

Scientiﬁc) and a ﬂame ionization detector. The carrier gas was helium (15 ml/min). Oven
temperature began at 60'C and increased at a rate of 10°C/min to 160’C, then increased at a
rate of 25'C/min to 200'C, where it remained constant for 1 minute. Liquid samples (4 ml)
were also collected from columns seeded with strain B1 and ﬁltered using sterile 0.22 pm
ﬁlters. The concentrations of two byproducts - 3,6-dimethyl pyrocatechol and p-xylene
dihydrodiol (Chang et al. 1993) - were then measured by a Gilson Gradient System high
pressure liquid chromatograph (HPLC) equipped with a C13 reverse-phase column and a
UV detector at 280 nm. The mobile phase was composed of 40 % water and 60 %

methanol with 0.1 % acetic acid. The ﬂow rate was 1.2 mein.

5.2.9 Scanning Electron Microscopy (SEM)

Samples of both virgin and bioﬁlm coated activated carbon and baker product (from
the ﬂuidized bed reactor) were ﬁxed at 4'C for 1-2 hours in 4% glutaraldehyde buffered
with 0.1 m sodium pnosphate (pH 7.4). Following a brief rinse in the buffer, samples
were dehydrated in a ethanol series (25%, 50%, 75%, 95%) for 10-15 min at each
gradation followed by three 10 min shifts in 100% ethanol. A Balzers critical point dryer,
using liquid carbon dioxide as the transitional ﬂuid, was used to critical point dry the
samples. Samples were sputter coated with gold (28 nm thickness) in an Emscope Sputter
Coater model SC 500 purged with argon gas, and examined in a JEOL JSM-35CF

Scanning Electron Microsc0pe (Japan Electron Optics Ltd.).

5.3 Results

Typical degradation curves for toluene and xylene by the two strains are shown in

Figure 5-1 and Figure 5-2. Figure 5-1 illustrates cometabolic degradation of p-xylene by

TOLUENE (I19) SUBSTRATE MASS (ug)

XYLENE(ug)

 

 

1400 .,
1200 ..
1000 a.
800 ..
600 ..
400 ..
200 ..

   
    
  

—O— xylene alone
—0— xylene and toluene

90

toluene
Spiked

1-3000

4 2500
biomass r2000

F1500
u-IOOO
elsoo

 

-0

—'°— toluene alone
—0— toluene and xylene

  
  

5 10 15 20 25

  

 

 

J 41
I I

q»
cu-
d—

5 10 15 20 25
TIME (hr)

Figure 5-1. Degradation curves illustrating the degradation of toluene and p-

xylene by strain B1.

(on) ssvwoue AUG

400
200

SUBSTRATE MASS (pg)
0)
8

1 400
1 200
1 000
800
600
400
200

TOLUENE (pg)

1 800
1 600
1 400
1 200
1 000
800
600
400
200

PXYLENE (no)

 

       
 

 

 

 

    

.0— toluene alone

91

toluene

 

—°— toluene and xylene

—°— xylene alone

+ xylene and toluene

 
  
  

 

«L-
m-
-l-

TIME (hr)

biomass .

T 2000
a- 1500
- 1000

- 500

 

db

18

18

Figure 5-2. Degradation curves illustrating the degradation of toluene and p-

xylene by strain X1

(511) SSVWOIQ AUG

92

strain B1. In single substrate experiments, strain Bl was incapable of transforming p-
xylene. However, p-xylene was transformed when combined with a growth substrate,
such as toluene. Transformation of p-xylene slowed signiﬁcantly when the growth
substrate disappeared, but increased when toluene was again added. Products of the
cometabolic transformation were previously identiﬁed as p-xylene dihydrodiol, 3,6-
dimethyl pyrocatechol, and 2,5-dimethyl phenol. Strain X1 utilizes both compounds as
growth substrates. The rate of degradation of p-xylene was faster than that of toluene. The
presence of either compound with the other affects the rate of degradation of each substrate

as shown in Figure 5-2.

The single-solute adsorption isotherms for toluene and p-xylene were described

using the Freundlich isotherm model;

Qe=KFX(Ce)" (54)

where, qc = mass of solute adsorbed per mass of adsorbent, Ce = concentration of solute at
equilibrium, and K}: and n are constants representing the sorptive capacity and afﬁnity
(intensity), respectively. The values of the constants were determined by a non-linear curve
ﬁtting procedure and tabulated in Figure 5-3 with the two iSOtherms. The larger K}: value

and the shape of the isotherm indicates p-xylene is more readily adsorbed than toluene.

Figure 5-4 compares efﬂuent proﬁles of toluene and p-xylene from three different
activated carbon columns ; 1) BAC seeded with strain B1, 2) BAC seeded with strain X1,
and 3) GAC as a control. BAC columns show slower appearance of target compounds in
the efﬂuent than GAC. In BAC columns, efﬂuent proﬁles show distinctively different
patterns for each different microorganism with which they were seeded. In the strain B1

seeded BAC column, toluene was observed in the efﬂuent earlier than p-xylene, however,

ADSORPTION CAPACITY
(mg SUBSTRATE / G of ACTIVATED CARBON)

500

400

300

200

 

93

p-xylene

     

toluene o

 

 

 

 

 

 

 

 

 

. ' ﬂ ‘— - v III. ‘
. ’ ‘- a- "' ‘I’ ‘— -’
’-
I / /
r ’ O
F reundlich isotherm
Compound
K F n
toluene 88.49 0.31
p-xylene 156.66 0.29
41 l I l l l L I
l l I l l l T l
S 1 O 1 5 2 0 2 S 30 3 S 40

EQUILIBRIUM CONCENTRATION (mg/L)

Figure 5-3. Isotherms for toluene and p-xylene with Freundlich parameter estimation.

94

 

 

 

 

 

 

 

 

 

 

 

20
Q
‘5” 15 " BAC with Strain 81 I toluene
E o pxylene
9 12 ..
.—
E
5 8 1" ' 000 o
0
0 I o I I
I 0° I I...
04 ‘
0 200 400 600 800
20
:T I toluene
\ 15 ..
g BAC with Strain X1 0 pxylene
(Z) 12 d»
< II I
E 8 T I. 'l l I II '
E I. I -
o 4 -- I
z I o
8 W000 °
04 .
0 zoo 400 600 300
20
I . I '
=1; 16 .. GAC . ' . I I 'I
E ' I l o I
<2; 12 .4.- I I . 0 0 I0 0
I I 0
E 8 -~ ' ' ° 0 ° 0
5 0° 0
8 °° ° :‘°'::°.2:.
c2) I 0 0 P Y
0 omen-noon? : g ;
0 200 400 600 300
TIME (hrs)

Figure 5-4. Efﬂuent proﬁles of toluene (I ) and p—xylene (0 ) for 1) BAC
seeded with strain Bl, 2) BAC seeded with strain X1, and 3) GAC.

95

it eventually decreased. p-xylene in the efﬂuent shows a typical breakthrough pattern. The
BAC column seeded with strain X1 shows low efﬂuent concentration of p-xylene,
however, toluene shows a breakthrough pattern. The GAC column shows typical

breakthrough curves found for bi-solute systems.

Figure 5-5 shows the byproducts in the efﬂuent from the cometabolic degradation
of p-xylene in columns seeded with strain B1. In the bioﬁlm column, p-xylene dihydrodiol
was formed immediately after starting the operation of the column. 3,6-Dimethyl
pyrocatechol was not formed until 100 hrs of bed operation. Then, eventually, the
formation of 3,6-dimethyl pyrocatechol increased and after approximately 200 hrs of bed
operation, it bypassed the formation of p-xylene dihydrodiol. However, in the BAC
column, byproducts were not observed until 200 hrs of operation. Figure 5-6 compares the
efﬂuent proﬁle of p-xylene from the GAC system and the accumulated area graph of p-
xylene, p-xylene dihydrodiol, and 3,6-dimethyl phenol, in the BAC system all on a

molecular carbon basis. There was a good agreement between the two lines.

The results of a shock load resulting from increased toluene concentrations are
provided in Figure 5-7 and Figure 5-8. Figure 5-7 shows efﬂuent proﬁles of target
compounds and byproducts during the shock in columns seeded with strain Bl. In these
columns, doubling the inﬂuent concentration of toluene not only increased the efﬂuent
concentration of toluene but also that of p-xylene. However, the formation of byproducts
was retarded dming this shock. Figure 5-8 compares efﬂuent proﬁles of target compounds
from BAC and bioﬁlm columns seeded with strain X1. Doubling the toluene signiﬁcantly

affected the efﬂuent toluene concentration. However, p-xylene concentration was only

slightly affected.

12

96

 

 

 

 

 

 

 

 

o
g BAC with strain er ' . g
g 9 .. o p-XDHD .
I: o
< 6 In-
E . .oo‘ o
E .0 C 8000
5 0 0000 O 00
O C 00 00000
0 -Ioooooooc. : ,1 :
0 200 400 600 800 1 000
12
3 Biotilm with strain 81 .
\ o. 0
°’ oIo
E 9 .. o p-XDHD O
5 o 3.6- DMPC
: . .
< 6 ‘-
E 9 o 0
Lu 0 o oo 00 '0 o oo
‘2’ 3 ‘* o a”. 0 0. ° 8 o 000 o o o
o o o
o (5’
0 l” 4 i I i
0 200 400 600 800 1 000
TIME (hrs)

Figure 5-5. Efﬂuent proﬁles of byproducts, p-xylene dihydrodiol ( o ) and 3,6-
dimethyl pyrocatechol (I ) from 1) BAC and 2) Bioﬁlm from columns
seeded with strain B1.

CONCENTRATION BASED ON MOLECULAR CARBON (mg/L)

    
 

20

 

 

 

—

 

0 200 400 600 800
TIME (hrs)

Figure 5-6. Effluent proﬁle of p-xylene from GAC column (-II-) and
accumulated area graph of p-xylene, p-xylene dihydrodiol, and 3,6-
dimethyl pyrocatechol from BAC based on the concentration of
molecular carbon.

CONCENTRATION (mg/L)

CONCENTRATION (mg/L)

20

.5
O)

.a
N

on

20

98

 

 

 

  

 

 

 

 

 

 

 

100

BAC with Strain 81 ”MM Target Compound
..................... _---.‘!99'3'.°9.-----j.----- "'- toluene ---.
1 —A— p-xylene
7 v.17;- """"""
:‘o'
‘0
I-u.----.=::.—::=-=.;£L ................ .- ------------------------
-50 -25 o 25 50 75 100
BAC with Strain B1 (Wane Byproducts
doubled -I-- 3.6-dimethyl pyrocatechol
—L'.I— p-xylene dihydrodiol
I ------- -...___,_.q
E ’ ~‘x
..................... r’rrx-i
I ' l \ I
‘F‘ﬂ ‘\ I
WIDE—WW0
-50 -25 0 25 50 75
TIME (hrs)

Figure 5-7. Efﬂuent proﬁles of target compounds and byproducts during shock
loadings by doubling toluene concentration in BAC column seeded

with strain B1.

CONCENTRATION (mg/L)

CONCENTRATION (mg/L)

99

 

 

 

 

 

 

 

20
BAC with Strain X1 Target Compound
.0 :
1 J, ‘.‘ i "'." toluene
6 ----------------------- :' ------ 1' """ —A— xylene
12
8
4
o : : . i I
-so -25 o 25 so 75 100
20 3': :
Biotilm with Strain x1, toluene , Target Compound
doubled : ' '0 ' ' toluene
16 >-----------—--------—‘------------------ ------
. . I + xylene

 

 

 

 

TIME (hrs)

Figure 5-8. Efﬂuent proﬁles of target compounds during shock loadings by
doubling toluene concentration in BAC and bioﬁlm columns seeded

with strain X1.

100

Figure 5-9 shows p-xylene shock-loading result for columns seeded with strain Bl.
These columns showed increased efﬂuent concentrations of p-xylene. There was a slight
increase in toluene concentrations as well. During the shock period, the formation of
byproducts was diminished. Figure 5-10 shows efﬂuent proﬁles of target compounds in
columns seeded with strain X1. In the BAC columns, efﬂuent concentrations were only
slightly affected by the shock. However in the bioﬁlm columns, concentrations of both

compounds were increased considerably.

Figure 5-11 compares efﬂuent proﬁles of both BAC columns (seeded with strain
B1 and strain X1) after the oxygen supply was terminated. Concentrations of toluene and
p-xylene from the column seeded with strain B1 increased and ultimately reached the same
concentration as the inﬂuent. The concentration of toluene from the column seeded with
strain X1 reached inﬂuent levels faster than p-xylene, which also increased. However,

there were more sorption capacity left for p—xylene.

Scanning electron microscopy of the carbon particles from different depths of the
column are compared in Figure 5-12 through 5-14. It can be seen that carbon particles
obtained near the inﬂuent are more throughly covered by biomass. However, those from
the efﬂuent side still show bare carbon surfaces. A Column section collected near the

inﬂuent at low magniﬁcation suggests that the pores are kept open even with active

biological growth.

5.4 Discussion

The two Pseudomonas isolates demonstrated very different substrate preferences

for toluene and p-xylene. Especially during multi-substrate degradation, their degradation

CONCENTRATION (mg/L)

CONCENTRATION (mg/L)

20

_I
a)

_I
N

20

101

 

 

 

 

 

 

 

   

BAC with Strain B1 p-xylene Target Compound
. doubled I - - '0 - - toluene
+ xylene
-50 -25 o 25 50 75 100
Target Compound
"" toluene___.
+ xylene

 

 

 

 

TIME (hrs)

Figure 5-9. Effluent proﬁles of target compounds during shock loadings by
doubling p-xylene concentration in BAC and bioﬁlm columns
seeded with strain Bl.

CONCENTRATION (mg/L)

CONCENTRATION (mg/L)

102

20

 

BAC with Strain x1 Target Compound
., ‘ pnxylana . I
1; doubled : .. --

16 . toluene
» + xylene

 

 

 

 

20

 

Biofilm with Strain x1 “mm Target Compound
16 .--------------------..L ..... 9999199."---5 ....... u." toluene -.

+ xylene

 

 

 

 

TIME (hrs)

Figure 5-10. Efﬂuent proﬁles of target compounds during shock loadings by
doubling p-xylene concentration in BAC and bioﬁlm columns seeded
with strain XI.

103

20

 

-—

BAC with Strain B1

4

O)
O
X
‘<
(O
O
3
0)
i
S'——

 

   
 
 
 

 

 

 

 

 

 

   
 

 

 

 

3 no oxygen supplied
2’ H,< H
v I A
i- I o" ""“th-"v.
< I ,
E 8 l
5
(2J ""‘"' toluene
8 4 —¢r— xylene
I
I
0 r l 1 f 4'
~20 0 20 40 60 80
20
} BAC with Strain x1
oxygen suppliedl no oxygen supplied
A 16 W >
31 ."“ '_.-~ --.'-.--.o.
g I ,-o"- \. ."O' .‘-'
Z I I ‘. 0 '
f: '0 s. ' ,"
E r. s‘ We
i- 8 -
z
I“ I
‘z’ i
8 4 .l. ""‘"' toluene
I + xylene
0 l A. i i
-20 0 20 40 60 80

TIME (hrs)

Figure 5-11. Effluent proﬁles of target compounds in BAC columns after the
microbiological activity was eliminated.

104

 

Figure 5-12. Scanning electron micrograph of carbon particles obtained near
inﬂuent side of the reactor.

105

 

Figure 5-13. Scanning electron micrograph of carbon particles obtained near
effluent side of the reactor.

106

 

Figure 5-14. Scanning electron micrograph of matrix of carbon particles near
inﬂuent side of the reactor.

107

pattern exhibited not only opposite preferences for one compound over the other, but also
the effect of one compound on the degradation of others. From the adsorption study, the
single compound isotherms show that p-xylene is more adsorbable than toluene. These

kinds of complexities make the prediction of the operation of BAC difﬁcult.

In this study, the initial seeding of different species of microorganisms showed
considerable differences in degradation patterns and in consequence, the importance of a

dominant population of microorganisms in the treatment of BAC was observed.

As Figure 5-4 shows, the efﬂuent proﬁles of p-xylene were greatly inﬂuenced
depending on which microorganisms were present in the columns. During the start-up
period, the major removal mechanism for BAC is adsorption. The measurement of oxygen
consumption (data not shown) showed that microbiological removal in the BAC system
developed more slowly than in the bioﬁlm columns. Toluene started to appear in the
efﬂuents of both BAC columns at about the same time, although it was decreased as the
biomass increased in the BAC column seeded with strain Bl. p-xylene is a more
adsorbable compound; however, because it is a less favorably degraded compound than

toluene by strain B1, it ultimately reached higher efﬂuent concentrations than toluene.

The BAC system seeded with strain X1 demonstrated different performance with
respect to p-xylene. p-xylene is more favorably degraded by strain X 1 than toluene (Figure
5—2) and it is also more favorably adsorbed. Consequently, both adsorption and

biodegradation effectively removed p-xylene and it was not observed in the efﬂuent until

600 hrs of operation.

Based on the previous batch studies, p-xylene dihydrodiol is formed when there is

a enough growth substrate, and 3,6-dimethyl pyrocatechol is formed from p-xylene

108

dihydrodiol under the limiting growth substrate condition. During the start-up period, only
p-xylene dihydrodiol was formed. This is evidenced from the result of the bioﬁlm column
(Figure 5-5). As the biomass increased, it depleted toluene quickly and caused a growth
substrate limitation. This resulted in the formation of 3,6-dimethyl pyrocatechol. In the
BAC system, the slightly adsorbable p-xylene dihydrodiol reached breakthrough at about
200 hrs of bed operation. It is not clear why the profile of 3,6-dimethyl pyrocatechol
changed (increased, decreased, then re-increased) until it reached Steady-state. It is

noteworthy that the proﬁle matched the toluene efﬂuent proﬁle which also increased, then

decreased (Figure 5-5).

The concentrations of byproducts and p-xylene were converted to a molecular
carbon basis to graphically show a mass balance on these compounds. Good matches
between these accumulated concentrations with the p-xylene proﬁle from GAC (Figure 5-6)
sugget that the difference between proﬁles of p-xylene in the two columns can be

interpreted as the result of microbiological degradation.

The addition of toluene to the bottle containing p-xylene with strain B1 caused
enhanced degradation of p-xylene in the batch experiment (Figure 5-1). It was
hypothesized that the increased toluene concentration in the inﬂuent would cause enhanced
degradation of p-xylene with increased amount of byproducts formation. However, the
column experiment revealed a different result. The increased inﬂuent concentration of
toluene caused an increase in toluene concentration as well as p-xylene in the efﬂuent
(Figure 5-7). This can be explained as the result of competition. The increase of the more
favorable substrate (toluene) caused the microorganisms to utilize toluene instead of the
increasing cometabolic degradation of p-xylene. Byproduct formation also decreased as
less p-xylene was degraded. The bioﬁlm system (data not shown) showed an immediate

increase of both target compounds as well as an immediate decrease of byproducts. The

109

BAC system showed sorptive buffering, which is suggested to be an advantage of the BAC

systems (Voice et al. 1992).

The shock loading of the less favorably degraded compound had a smaller impact
on the more favorably degraded compound. The shock loading of toluene in columns
seeded with strain Xl showed p-xylene was only slightly disturbed by the presence of
increased toluene (Figure 5-8). The response of shock loading was more rapid for the
bioﬁlm. The increased inﬂuent concentration of toluene instantly raised the efﬂuent

concentration as well.

For strain B1, p-xylene is the less desirable substrate. The increased inﬂuent p-
xylene concentrations caused increased efﬂuent concentrations of both compounds.
However, the amount of increase in toluene concentration was less signiﬁcant than that of
p-xylene. Figure 5-7 and Figure 5-9 shows that the effect on the more degradable
compound by the increase of the less degradable compound is not as signiﬁcant as the

reverse situation.

In the BAC system with strain Xl, more adsorption sites were available, because
the microbiological activity removes the more adsorbable compound (p-xylene) from the
system. This was advantageous for later shock loadings. The p-xylene shock for the
bioﬁlm column seeded with strain X1 showed that concentrations of both compounds
increased signiﬁcantly. This result was consistent for the shock loading of toluene in
columns seeded with strain B1. In other words, shock loading of the favorably degraded
compound affects the efﬂuent concentrations of both compounds signiﬁcantly. However,
the BAC system with strain X1 shows a much diminished shock effect in the efﬂuent. The

shock loading was buffered by adsorption.

110

Without microbiological activity, the only removal mechanism in BAC is
adsorption. As observed during the shock-loading studies, the BAC system with strain X1
had more adsorption capacity than the BAC system with strain Bl. During the oxygen
depravation degradation study, both BAC systems showed a slow increase of target
compounds in the efﬂuent . Even though others reached equilibrium with inﬂuent after 60

hrs, p-xylene was continuously removed by adsorption in the BAC system seeded with

strain X1 .

The observations by SEM differentiate carbon particles obtained near both the
inﬂuent and efﬂuent sides of the reactor. Activated carbon particles collected near the
inﬂuent, where substrate is abundant, show complete coverage of biomass. However, the
efﬂuent side gets less substrate and this causes less biomass to grow. If the reactor is deep
enough, this phenomena should be more distinct. Biomass will develop up to the height
providing sufficient growth substrate and dissolved oxygen. Beyond this level, the
uncolonized activated carbon will remain and can be reserved as backup for an unusual case
of inﬂuent loading. As the column section shows, the pores between the carbon particles
remain open. The shear force of the ﬂow is believed to take some biomass away from the

surface of the bioﬁlm surrounding the activated carbon particles.

5.5 Conclusion

BAC and bioﬁlm columns seeded with two Pseudomonas species - strain B1 and
X1 - were compared with GAC in treating two substrates; toluene and p-xylene. The
effluent proﬁles of the target compounds were greatly inﬂuenced by the microorganisms in
the systems. Shock loading was applied to the reactors by increasing one of the target

compounds. Shock loading of the favorably degraded compound resulted in a signiﬁcant

111

increase in the efﬂuent concentrations of both compounds. The effect of shock loading by
the less favorably degraded compound was smaller. Both cases of shock loadings affected
the degradation of p-xylene and resulted in the reduced formation of byproducts. The BAC
systems were less affected by perturbations. It is recommended that BAC columns be
designed with enough longitudinal length to provide residual adsorption capability to

handle variations in unusual inﬂuent loading.

CHAPTER 6
DISSERTATION SUMMARY AND RECOMMENDATIONS

6.1 Summary

The purpose of this research was to develop a better understanding of BAC systems
for the treatment of BTX as multiple substrates. For this purpose, microbiological kinetic
studies were performed. Two Pseudomonas species (designated strains B1 and X1) were
isolated from an aerobic pilot-scale ﬂuidized-bed reactor treating groundwater containing
benzene, toluene, and p-xylene (BTX). Strain Bl grew with benzene and toluene as the
sole sources of carbon and energy, and it cometabolized p-xylene in the presence of
toluene. Strain X1 grew on toluene and p-xylene, but not benzene. In single substrate
experiments, the appearance of biomass lagged the consumption of growth substrates,
suggesting that substrate uptake may not be growth rate-limiting for these substrates.
Batch tests using paired substrates (BT, TX, or BX) revealed competitive inhibition and
cometabolic degradation patterns. Competitive inhibition was modeled by adding a
competitive inhibition term to the Monod expression. Cometabolic transformation of non-
growth substrate (p-xylene) by strain B1 was quantified by coupling p-xylene
transformation to consumption of growth substrate (toluene) during growth and to loss of
biomass during the decay phase. Coupling was achieved by deﬁning two transformation
capacity terms for the cometabolizing culture: one that relates consumption of growth
substrate to the consumption of non-growth substrate, and a second that relates

consumption of biomass to the consumption of non-growth substrate. Cometabolism

112

113

increased decay rates, and the observed yield for strain Bl decreased in the presence of p-

xylene.

The formation of byproduct observed during the cometabolic degradation of p-
xylene was extensively investigated. Three byproducts - p-xylene dihydrodiol, 3,6-
dimethyl pyrocatechol, and 2,5-dimethylphenol - were identiﬁed. The byproduct formed
during cometabolic degradation was found to depend the conditions under which the cells
obtained transformation capacity. During the presence of the growth substrate, the primary
byproduct formed was p-xylene dihydrodiol. However, after the complete degradation of
toluene (growth substrate), the major byproduct was 3.6-dimethyl pyrocatechol.

p-xylene dihydrodiol was found to be very adsorbable to activated carbon, while
3,6—dimethyl pyrocatechol was not. Batch isotherm experiments showed that the presence
of byproducts in solution interfered with the sorption of the target compounds. Benzene
was the most signiﬁcantly affected of the three compounds. The formation of byproducts
during the operation of ﬁxed-bed biological activated carbon (BAC) system was studied by
comparing BAC reactors to bioﬁlm reactors during the absence of either the growth
substrate or the non-growth substrate. In the BAC reactor, cometabolic transformation
continued in the absence of both substrates, while the transformation stopped immediately

in the bioﬁlm column.

BAC system clearly showed extended service time when compared to GAC
systems. However, two BAC columns seeded with different species showed signiﬁcant
differences in efﬂuent proﬁles of the target compounds. During the start-up period, the
major removal mechanism was adsorption. As the biomass increased, biodegradation
became more important than adsorption. BAC also showed a buffering effect during the

inﬂuent perturbations. The effect of a shock-load by the growth substrate on the non-

114

growth substrate was more signiﬁcant than that by the non-growth substrate on the growth
substrate. Designing the BAC reactor with enough depth to retain adsorption capacity is

recommended to cope with variations in inﬂuent loading.

6.2 Recommendations for Future Studies

Continuation of this work should consider the following issues that require further

investigation.

- Degradation and competition of other substrates with BTX by the two species (strain B]

and X1) should be studied.

0 Perform kinetic studies on a mixed culture. The two (strain Bl and X1) species could be
combined and the degradation of target compounds and byproduct formation be

monitored.

. Investigate the toxicity of the byproduct and the long-term effects of shock loads on the

formation of byproducts.

- Develop model to describe multi-substrate bioﬁlm and adsorption dynamics to describe

different patterns of degradation such as competitive inhibition or cometabolic

degradation.

APPENDICES

APPENDIX A

MEDIA PREPARATION

C .. [IICIIIIZ . ]_ l I'
Add the following and take volume to 1 L with deionized water :

Major ions

KHzPO4 1.36 g
NazHPO4 1.42 g
KNO3 0.50 g
(NH4)2804 2.38 g
MgSO4-7H20 0.05 g
CaC12 0.01 g
Trace elements

H3BO4 2.862 mg
MnSO4-H20 1.538 mg
Fe(NH4)2(SO4)2-6H20 3.529 mg
CuSO4-5H20 0.0392 mg
ZnC12 0.0209 mg
CoC12-6H20 0.0405 mg
Na2M004-2H20 0.0252 mg

 

After Ridgway er al. (1990)
Check pH (6.9-7.1). Adjust if needed.

Solidiﬁed medium is prepared by adding 18 g Bacto agar.
Medium for column expeirment is diluted from above HCMM2 medium by 1/10.

C "[11 I 'lEﬁ

Add the following and take volume to l L with deionized water :

 

KH2PO4 8.5 mg
K2HPO4 21.75 mg
NazHPO4 17.70 mg
MgSO4 11.0 mg
After Chang (1987)

Phosphate and magnesium buffer was used for the dilution of cell containing solution.

116

APPENDIX B

ISOLATION PROCEDURES

Pure culture microorganisms were isolated from the laboratory-scale activated
carbon ﬂuidized reactor located at Michigan State University. Details of reactor operation
and design are described elsewhere (Voice er al., 1992). After a three month period of
continuous operation treating BTX-contaminated water, activated carbon particles with well
developed bioﬁlms were removed from the ﬂuidized-bed reactor. Approximately ﬁve
particles were placed in a 25-ml glass vial containing 10 ml of phosphate buffer solution
and ten glass beads 4 mm in diameter. The vial was vigorously shaken and allowed to

stand for 15 minuites until all large ﬂoc had settled.

Test tubes with 9 ml phosphate buffer solution were prepared after autoclaving.
Then 1 ml of solution from the above 25-ml glass vial was transfered to make a solution of
dilution 10“. This step was repeated to succeeding tubes until a solution of dilution 10'7
was made. From each tube of this dilution series, 0.2 ml of solution were spreaded on

plates of HCMM2 solid media (Appendix A). Three plates were prepared from each

dilution.

The plates were exposed to benzene, toluene, or p-xylene in completely sealed
chambers . After 48 hours, isolates were picked, streaked, and re-incubated. This
procedure was repeated 5 to 8 times. One to three distinctive colonies were identiﬁed from
each carbon source and were named appropriately by using the abbreviation of each
substrate followed by a sequence number. Colonies isolated by growth upon vapors of a
single pure compound, such as benzene, were then tested for growth in chambers

containing other substrates, such as toluene or p-xylene. The extent of growth was

117

118

recorded with regard to the formation of colonies. The results from the observations are

provided in Table B-1. Details of the isolation procedure are described on Figure B-l.

Table B-1. Growth of isolated colonies in each carbon source

 

 

Strain benzene toluene p-xylene
B1 ++ ++ -
Tl + ++ +
T2 + ++ +
X1 - ++ ++
X2 + ++ ++

 

++ : Active growth 4» : Moderate growth - : No growth

Among these strains, strain BI and strain X1 showed distinct differences in
degradation characteristics. Strain B1 grew on either benzene or toluene; however, it did
not grow on p-xylene. On the contrary, Strain X1 grew on xylene, but it did not grow on
benzene. Because of their different degradation patterns, these two strains were selected for

further investigation.

The results of scanning electron microscopy (SEM) of these two strains are shown

in Figure B-2. Materials and methods for SEM are provided in Appendix C.

119

transfer1 ml to next test tube

A mnmmmn

test tube contains
9 ml of autoclaved
phosphate buffer

25 ml vial

lass beads -1 -7
e 10 ~ 10 dilutions in phosphate buffer sout/
transfer 0.2 ml on HCMM2 solid media
and evenly spread @

pick up colonies and streak on HCMM2
solid midium. Repeat 5-8 times

C 211 C

incubate 48 hrs at 24 C in completely Incubate plates in chamber with
sealed chamber containing BTX single substrate

ll|||||llll|||llllllllIllllllllllllllllll||||||||||||||||| llllllllllllllllllllllllllllllllllIllllllllllllllllllllllll

   

— hypodermic needle

_ tench-lined septum
with cap

saturate cotton

L— with oTx

   

 

 

 

 

 

 

Figure B-l. Schematic details of isolation procedures

120

 

Pseudomonas sp. strain Bl

 

Pseudomonas sp. strain X1

Figure B-2. SEM of Pseudomonas sp. stain B1 and strain X1

APPENDIX C

SCANNING ELECTRON MICROSCOPY SAMPLE PREPARATION PROCEDURES

. Fix samples at 4 C for 1 hour in 4% glutaraldehyde buffered with 0.1 M sodium
phosphate at pH 7.4. Following a brief rinse in the buffer, dehydrate samples in a
ethanol series (25%, 50%, 75%, 95%) for 10-15 min. at each gradation followed by
three 10 min. changes in 100% ethanol.

. Critical point dry samples in a Balzers critical point dryer using liquid carbon dioxide as
the transitional ﬂuid.

. Coat samples with gold (28 nm thickness) in an Emscope Sputter Coater model SC 500
purged with argon gas.

. Examine samples in a JEOL J SM-35CF Scanning Electron Microscope (Japan Electron
Optics Ltd.)

121

CLS
DIM
PRINT

GOSUB

INPUT

APPENDIX D

CODE FOR COMET ABOLIC DEGRADATION MODELING

MODELING COMETABOLISM

Developed using 2nd order Runge-Kutta
numerical method by Myung K. Chang

March 22, 1992

Input parameters for growth substrate

SG(5000), SN(5000), X(5000), GS(5), NS(5), XC(5)

"Modeling of Cometabolism":PRINT

1000

"Do you want to input data from : 1. Key-board or 2. Input data ";YN

IF YN = 1 THEN GOTO 50

REM
READ

GOTO 100

50 CLS
PRINT
PRINT
INPUT
INPUT
INPUT

REM

PRINT
PRINT
INPUT
INPUT

INPUT

PRINT
INPUT
INPUT
INPUT

PRINT

Read data from File.
MMX, Y, KSG, TCG, k, KSN, FCO, B, TCB, SG(l), SN(l), X(l), dt

"Parameters for growth substrate"

" 1. Maximum speciﬁc growth rate : Urnax (l/day) = ";MMX
" 2. Yield coefﬁcient : Y (mass biomass/mass growth substrate) = ";Y
" 3. Half saturation coefﬁcient : Ks (m g/l) = ";KSG

Input parameters for non- growth substrate
"Parameters for-non- growth substrate"

" 4. Transformation capacity : Tcg (mass of NGS/mass of GS) = "; TCG

" 5. Maximum speciﬁc growth rate of utilization : k (mass of NGS/mass
biomass-day) = "; k

" 6. Half saturation coefﬁcient : Ks (mg/l) = ";KSN

Input other parameters
"Other parameters"

" 7. Fractional coefﬁcient : fco = ";FCO

" 8. Decay coefﬁcient: b(1/day) = ";B

" 9. Biomass transformation capacity : ch (mass of NGS/mass of
biomass) = "; TCB

122

123

REM Information on initial conditions

PRINT "Information on initial conditions":PRINT

INPUT " 10. Initial growth substrate mass : Sg (ug) = ";SG( 1)

INPUT " 11. Initial non-growth substrate mass Sng (ug) = "; SN(l)

INPUT " 12. Initial biomass amount: Xo (ug) = "; X(l)

INPUT " 13. Time interval for analysis : delta T (min) = "; dt

GOTO 100

REM Print all information on the screen for check

100 CLS

PRINT "Summary of input parameters":PRINT

PRINT " 1. Maximum speciﬁc growth rate : Umax = ";MMX; "(l/day)"

PRINT " 2. Yield coefﬁcient: Y = ";Y; "(mass biomass/mass growth substrate)"

PRINT " 3. Half saturation coefﬁcient : Ks = ";KSG; "(mg/l)"

PRINT " 4. Transformation capacity : Tcg = "; TCG; "(mass of NGS/mass of
GS)"

PRINT " 5. Maximum speciﬁc growth rate of utilization : k = "; k; "(mass of
NGS/mass biomass-day)"

PRINT " 6. Half saturation coefﬁcient: Ks = ";KSN; "(mg/l)"

PRINT " 7. Fractional coefﬁcient : fco = ";FCO

PRINT " 8. Decay coefﬁcient : b = ";B; "(l/day)"

PRINT " 9. Biomass transformation capacity : ch = "; TCB; "(mass of
NGS/mass of biomass)"

PRINT " 10. Initial growth substrate mass : Sg = ";SG(l); "(ug)"

PRINT " 11. Initial non-growth substrate mass Sng = "; SN(l); "(ug)"

PRINT " 12. Initial biomass amount : X0 = "; X(l); "(ug)"

PRINT " 13. Time interval for analysis : delta T = "; dt; "(min)"

90 GOSUB 1000

INPUT "Do you want to change any parameter"; YNS

IF YN$ = "n" OR YN$ = "N" THEN GOTO 300
IF YN$ 0 "y" AND YN$ <> "Y" THEN GOTO 90

80 GOSUB 1000

INPUT
PRINT

"Print number to change"; num

IF num = 1 GOTO 110
IF num = 2 GOTO 120
IFnum =3GOTO 130
IFnum =4GOTO 140
IF num = 5 GOTO 150
IFnum =6GOT0160
IFnum =7 GOTO 170
IFnum=8GOTO 180
IFnum =9GOTO 190
IFnum = lOGOTO 200
IFnum = 11 GOTO 210

124

IF num = 12 GOTO 220
IF num =13 GOTO 230

GOTO 80

110 INPUT " 1. Maximum speciﬁc growth rate : Urnax (l/day) = ";MMX

GOTO 100

120 INPUT " 2. Yield coefﬁcient : Y (mass biomass/mass growth substrate) = ";Y

GOTO 100

130 INPUT " 3. Half saturation coefﬁcient : Ks (mg/I) = ";KSG

GOTO 100

140 INPUT " 4. Transformation capacity : Tcg (mass of NGS/mass of GS) = "; TCG

GOTO 100

150 INPUT " 5. Maximum speciﬁc growth rate of utilization : k (mass of NGS/mass
biomass-day) = "; k

GOTO 100

160 INPUT " 6. Half saturation coefﬁcient : Ks (mg/I) = ";KSN

GOTO 100

170 INPUT " 7. Fractional coefﬁcient : fco = ";FCO

GOTO 100

180 INPUT " 8. Decay coefficient : b (l/day) = ";B

GOTO 100

190 INPUT " 9. Biomass transformation capacity : ch (mass of NGS/mass of biomass)
= "; T CB

GOTO 100

200 INPUT" 10. Initial growth substrate mass : Sg (ug) = ";SG(l)
GOTO 100

210 INPUT" 11. Initial non-growth substrate mass Sng (ug) = "; SN(l)
GOTO 100

220 INPUT " 12. Initial biomass amount : Xo (ug) = "; X(l)

GOTO 100

230 INPUT " 13. Time interval for analysis : delta T (min) = "; dt
GOTO 100

REM Start calculations

300 MMX = MMX / (24*60)
k = k / (24*60)
B = B / (24*60)

REM Decide the number of loop.

NL = INT(12(X)/dt)

CLS

FOR i = 2 TO NL

GS (1) = dt*(-MMX/Y*SG(i-1)*X(i-1)/(KSG*139.015+SG(i—1)))

NS (1) = dt*(TCG*GS(1)-k*X(i-1))*SN(i-l)/(KSN*144.505+SN(i-1))

XC (1) = dt*(Y*(-GS(1))*(1-FCO)-B*X(i-1)+NS(1)/TCB)

GS (2) = dt*(-MMX/Y*(SG(i-l)+GS(1)/2)*(X(i-1)+XC(1)/2)/(KSG*139.015+(SG(i-
1>+GS(1)/2)» ,

NS (2) = dt*(TCG*GS(2)-k*(X(i-1)+XC(1)/2))*(SN(1-
l)+NS(1)/2)/(KSN*144.505+(SN(i-1)+NS(1)/2))

XC (2) = dt*(Y*(-GS(2))*(1-FCO)-B*(X(i-1)+XC(1)/2)+NS(2)fF CB)

GS (3) = dt*(-MMX/Y*(SG(i-1)+GS(2)/2)*(X(i-1)+XC(2)/2)/(KSG*139.015+(SG(i-
l)+GS(2)/2)))

125

NS (3) = dt*(TCG*GS(3)-k*(X(i-1)+XC(2)/2))*(SN(i-
1)+NS(2)/2)/(KSN*144.505+(SN(i-1)+NS(2)/2))

XC (3) = dt*(Y*(-GS(3))*(1-FCO)-B*(X(i-1)+XC(2)/2)+NS(3)/I' CB)

G8 (4) = dt*(-MMX/Y*(SG(i-1)+GS(3))*(X(i-l)+XC(3))/(KSG*139.015+(SG(i-
1)+GS(3))))

NS (4) = dt*(TCG*GS(4)-k*(X(i-1)+XC(3)))*(SN(i-1)+NS(3))/(KSN*l44.505+(SN(i-
1)+NS(3)))

XC (4) = dt*(Y*(-GS(4))*(1-FCO)-B*(X(i-1)+XC(3))+NS(4)/I‘ CB)

REM Calculation of New Step

SG(i) = SG(i-l)+(GS(1)+2*GS(2)+2*GS(3)+GS(4))/6
SN(i) = SN(i-l)+(NS(1)+2*NS(2)+2*NS(3)+NS(4))/6
X(i) = X(i-1)+(XC(1)+2*XC(2)+2*XC(3)+XC(4))/6
NEXT i

REM Open output ﬁle.

PRINT :PRINT :PRINT

GOSUB 1000

INPUT "What ﬁle name you want for output data";ﬁle$
OPEN ﬁle$ FOR OUTPUT AS #1

WRITE #1, "Time(hr)", "GS", "NGS", "Biomass"

FOR j=1 TO 50

KM=j*NL/50

k1=KM*dt

t=kll60

WRITE #1, t, SG(KM), SN(KM), X(KM)
NEXT j

CLOSE

INPUT "Do you want to repeat calculation with different parameters"; YN$
IF YNS <> "Y" AND YN$ <> "y" THEN GOTO 500
MMX = MMX * (24*60)

k = k * (24*60)

B = B * (24*60)

GOTO 100

DATA 15.00,].28, 3.17, .48, .5, .8, .026, .41, 2.4, 1390.708, 1437.74, 50, 1
500 STOP

END

lOOOFORII=1TO3

SOUND 500,1:SOUND 400,1
NEXT 11

RETURN

LIST OF REFERENCES

AWWA Committee, 1981. An assessment of microbial acrivity on GAC. Journal AWWA,
73:81-83.

Alvarez, P. J. J. and T. M. Vogel, 1991. Substrate interactions of benzene, toluene, and
para-xylene during microbial degradation by pure cultures and mixed culture aquifer
slurries. Appl. and Env. Microbiol., 57(10) :2981-2985.

Alvarez-Cohen, L. and P. L. McCarty, 1991a. A cometabolic biotransforrnation model for
halogenated aliphatic compounds exhbiting product toxicity. ES & T, 25(8) :1381-1387.

Alvarez-Cohen, L. and P. L. McCarty, 1991b. Product toxicity and cometabolic
competitive inhibition modeling of chloroform and trichloroethylene transformation by
methanotrophic resting cells. Appl. and Env. Microbiol., 57(4) :1031-1037.

Andrews, G. F. and C. Tien, 1976. The interreaction of bacterial growth adsorption and
ﬁltration in carbon columns treating liquid waste. AIChE Symposium Series, 71:164—175.

Andrews, G. F. and C. Tien, 1977. New approach to bacterial kinetics in wastewater.
Journal of the Environmental Engineering, 103(6) :1057- 1074.

Andrews, G. F., 1979. The expansion of a ﬂuidized bed containing biomass. AIChE
Journal, 25:720-723.

Andrews, G. F. and C. Tien, 1981. Bacterial ﬁlm growth in adsorbent surfaces. AIChE
Journal, 27(3) 2396-403.

Arvin, E., B. K. Jensen and A. T. Gundersen, 1989. Substrate interactions during aerobic
biodegradation of benzene. Appl. and Env. Microbiol., 55(12) :3221-3225.

Ayers, G. W. and R. E. Nuder, 1964. Benzene. New York, NY: John Wiley and Sons.

Bauer, J. E. and D. G. Capone, 1988. Effects of co—occurring aromatic hydrocarbons on
degradation of individual polycyclic aromatic hydrocarbons in marine sediment slurries.
Appl. and Env. Microbiol., 54(7) :1649-1155.

Benedek, A., 1980. Simultaneous biodegradation and activated carbon adsorption- a
mechanistic look. In M. J. McGuire (Eds),

Was; (pp. 273- 302). Ann Arbor, MI: vAnn Arbor Science Inc.

Bishop, D. F., L. S. Marshall, T. P. O'Farrell, R. B. Dean, B. O'Connor and R. V.
Villiers, 1967. Studies on activated carbon treatment. Journal WPCF, 39(2) :188-194.

Chang,H. T. 1985. 11.. m
WUnivemity of Illinois, Urbana-Charnpaign, IL

 

126

127

Chang, H. T. and B. E. Rittmann, 1987. Mathematical modeling of bioﬁlm on activated
carbon. , 21:273-288.

Chang, M.-K., T. C. Voice and C. S. Criddle, 1993. Kinetics of competitive inhibition
and cometabolism in the biodegradation of benzene, toluene, and p-xylene by two
pseudomonas isolates. Biotechnology and Bioengineering, 41(1 1):1057-1065

Characklis, W. G., 1984. Bioﬁlm development: a process analysis. Microbial Adhesion
and Aggregationzl37-157.

Characklis, W. G., G. A. McFeters, and K. C. Marshall, 1990. Physiological ecology in
bioﬁlm systems. In W. G. Characklis and K. C. Marshall (Eds), Bioﬁlms (pp. 341-394).
Wiley Interscience.

Chemical and Engineering News, 1991. Facts and ﬁgures. 69(25) :28-33.

Chudyk, W. A. and V. L. Snoeynik, 1984. Bioregeneration of activated carbon saturated
with phenol. ES & T, 18(1) :1-5.

Claus, D. and N. J. Walker, 1964. The decomposition of toluene by soil bacteria. Journal
Gen. Microbiol., 36: 107-122.

Cooney, D. and E. Lightfoot, 1966. Multicomponent ﬁxed-bed sorption of interfering
solutes. , 5(1) :25-32.

Cooney, D. and F. Strusi, 1972. Analytical description of fixed-bed sorption of two
langmuir solutes under nonequilibrium conditions. Ind. Eng. Chem. Fundam, 11(1) :123-
126.

Criddle, C. S., 1993. The kinetics of cometabolism. Biotechnology and Bioengineering,
41(11):1048-1056

Criddle, C. S., J. T. DeWitt and P. L. McCarty, 1990. Reductive dehalogenation of
carbon tetrachloride by escherichia coli.. Appl. and Env. Microbiol., 56(11) :3247-3254.

Crittenden, J., P. Luft, D. Hand, and L. Orvitz, 1985. Prediction of Multicomponent

Absorption Equilibria Using Ideal Absorbed Solution Theory, ES & T, 19(11) : 1037-
1043

Dagley, S., P. J. Chapman, D. T. Gibson and J. M. Wood, 1964. Degradation of the
benzene nucleus by bacteria. Nature, 202(4934) :775-778.

Dagley, S., W. C. Evans and D. W. Ribbons, 1960. New pathways in the oxidative
metabolism of aromatic compounds by micro-organisms. Nature, 188(4750) :560-566.

Dalton, H. and D. I. Stirling, 1982. Co-metabolism. Philos. Trans. R. Soc. London Ser.
B, 297:481-496.

Davey, J. F. and D. T. Gibson, 1974. Bacterial metabolism of para- and meta-xylene:
oxidation of a methyl substituent. Journal of Bacteriology, 119(3) :923-929.

Davies, J. I. and W. C. Evans, 1964. Oxidative metabolism of naphthalene by soil
pseudomonads. Biochemical Journal, 91:251-261.

128

Davis, R. S., F. E. Hossler and . W. Stone, 1968. Metabolism of p- and m-Xylene by
species of pseudomonas. Canadian Journal of Microbiology, 14:1005-1009.

Dean, B. J., 1985. Recent ﬁndings on the genetic toxicology of benzene, toluene, xylenes
and phenols. Mutation Research, 154: 153-181 .

DNR, 1989. Michigan sites of environmental contamination ACT 307. The State of
Michigan.

Deshpande, S. D., T. Chakrabartl and P. V. R. Subrahmanyam, 1987. Mixed-substrate
utilization by acclimated activated sludge in batch and continuous-ﬂow stirred tank reactors.
ES & T, 21(10) :1003-1008.

DeWalle, F. B. and E. S. K. Chian, 1977. Biological regeneration of powdered activated
carbon added to activated sludge units. Water Research, 11:439-446.

Digiano, F. A., 1981. Inﬂuence of biological activity on GAC performance. In W. W.
Eckenfelder(Eds.), in - ..ti~ ' "l -. {n 2010 I OIHOII

WWI). 285- 305). Nashville, TN: Enviro Press

Dobrevski, I. and L. Zvezdova, 1989. Biological regeneration of activated carbon. Water
Science Technology, 21(1) : 141 - 143.

Evans, W. C., 1963. The microbiological degradation of aromatic compounds. Journal
Gen. Mirobiol., 32:177-184.

Evans, P. J., D. T. Mang and L. Y. Young, 1991. Degradation of toluene and m-xylene

and transformation of o-xylene by denitrifying enrichment cultures. Appl. and Env.
Microbiol., 57(2) :450-454.

Ford, D. L. and M. A. Buercklin, 1972. The interrelationship of biological and carbon
adsorption systems for the treatment of reﬁnery and petrochemical wastewaters. In S. H.

Jenkins (Ed) Adxarmunﬂarmllmioaﬁcmh Pergamon Press

Foster, J. W., 1962. Hydrocarbons as substrates for microorganisms. Journal Microbiol.
Serol, 28:241-274.

Franklin, F. C. H., P. R. Lehbrach, R. Lurz, B. Rueckert, M. Bagdasarian and K. N.
Timmis, 1983. localization and functional analysis of transposon mutations in regulatory
genes of the to] catabolic pathway. Journal of Bacteriology, 154(2) :676-685.

Fritz, W. and E. Schlunder, 1980. Competitive adsorption of two dissolved organics onto
activated carbon -1. Chemical Engineering Science, 36:721-730.

Fritz, W., W. Merk and E. Schlunder, 1980. Competitive adsorption of two dissolved
organics onto activated carbon - 11. Chemical Engineering Science, 36:731-741.

Gardner, D. A., M. T. Suidan and H. A. Kobayashi, 1988. Role of GAC activity and
particle size during the ﬂuidized-bed anaerobic treatment of reﬁnery sour water stripper
bottoms. Journal WPCF, 60(4) :505-513.

Gariepy, R. L. and I. Zwiebel, 1971. Adsorption of binary mixtures in fixed beds. AIChE
Symp. Series, 67:17-24.

129

Gibson, D. T., J. R. Koch and R. E. Kallio, 1968. Oxidative degredation of aromatic

hydrocarbons by microorganisms.I.Enzymatic formation of catechol from benzene.
Biochemistry, 7(7) :2653-2662.

Gibson, D. T., J. R. Kock, C. L. Schuld and R. E. Kallio, 1968. Oxidative degradation of
aromatic hydrocarbons by microorganisms. II. Metabolism of halogenated aromatic
hydrocarbon. Biochemistry, 7(11) :3795-3802.

Gibson, D. T., 1968. Microbial degradation of aromatic compounds. Science, 161(3846)
:1093-1097.

Gibson, D. T., M. Hensley, H. Yoshioka and T. J. Mabry, 1970. Formation of (+)-cis-

2,3-dihydroxy-l-methylcyclohexa-4,6—diene from toluene by pseudomonas putida.
Biochemistry, 9(7) :1626- 1630.

Gibson, D. T., V. Mahadevan and J. F. Davey, 1974. Bacterial metabolism of para- and
meta-xylene: Oxidation of the aromatic ring. Journal of Bacteriology, 119(3) :930-936.

Goldsmith Jr., C. D. and R. K. Balderson, 1988. Biodegradation and growth kinetics of

enrichment isolates on benzene, toluene and xylene. Water Science Technology, 20(11)
:505-507.

Gossett, J. M., 1985. Anaerobic degradation of Cl and C2 chlorinated hydrocarbons.
Tyndall Air Force base: Air Force Engineering and Service Center.

Grady, C. P. L., Jr. and H. C. Lim, 1980. Biological wastewater treatment-theory and
application. New York: Marcel Dekker, Inc.

Grady, C. P. L., Jr., 1985. Biodegradation : Its measurement and microbiological basis.
Biotechnology and Bioengineering, 27:660-674.

Grady, C. P. L., Jr., 1990. Biodegradation of toxic organism: Status and potential.
Journal of Environmental Engineering, 116(5) :805-828.

Hayaishi, O. and M. Nozaki, 1969. Nature and mechanism of oxygenases. Science,
164(3878) :389-396.

Horvath, R. S., 1972. Microbial co—metabolism and the degradation of organic compounds
in nature. Bacteriological review, 36(2) :146-155.

Hou, C. T. Microbial transformations of important industrial hydrocarbons. In J. P.

Rosazza (Edsq) WWW 82-107) Boca
Raton, Florida: CRC press.

Howard, P. H. and P. R. Durkin, 1974. Sources of contamination, ambient levels, and
fate of benzene in the environment. USEPA. EPA 560/5-75-005.

Jain, K. S. and V. L. Snoeyink, 1973. Adsorption from bisolute systems on activated
carbon. Journal WPCF, 45(12) :2463-2479.

130

Jamison, W. V., R. L. Raymond and J. 0. Hudson, 1969. Microbial hydrocarbon co-

oxidation - III. Isolation and characterization of an or, a' - dimethyl - cis, cis -muconic acid
producing strain of Nocardia corallina. Appl. Microbiol., 17:853-856.

Janke, D. and W. Ihn, 1989. Cometabolic turnover of aniline, phenol and some of their
monochlorinated derivatives by the rhodococcus mutant strain AM 144. Archies of
Microbiology, 152:347-352.

Kidnay, A. J. and A. L. Myers, 1966. A simpliﬁed method for the prediction of
multicomponent adsorption equilibria from single gas isotherms. AIChE Journal, 12(5)
:981-986.

Kim, R., E. S. K. Chian, W. H. Cross and S. S. Cheng, 1986. Adsorption, desorption,

and bioregeneration in an anaerobic, granular activated carbon reactor for the removal of
phenol. Journal WPCF, 58(1) :35-40.

Kincannon, D. F., E. L. Stover, V. Nichols and D. Medley, 1983a. Removal mechanisms
for toxic priority pollutants. Journal WPCF, 55(2) :157-163.

Kincannon, D. F., A. Weinert, R. Padorr and E. L. Stover, 1983b. Prediction treatability

of multiple organic priority pollutant wasterwaters from sin gle-pollutant treatability studies.
:641-650.

Kirk, R. E. and D. Othmer, 1963. Kirk-Othmer encyclopedia of chemical technology (2nd
ed.). New York: John Wiley and Sons.

Kitagawa, M., 1956. Studies on the oxidation mechanism of methyl group. Journal
Biochemistry, 43(4) :553-563.

Law, A. T. and D. K. Button, 1977. Multiple-carbon-source-limited growth kinetics of a
marine coryneform bacterium. Journal of Bacteriology, 129(1) :115-123.

Leadbetter, E. R. and J. W. Foster, 1959. Oxidation products formed from gaseous
alkanes by the bacterium pseudomonas methanica. Arch. Biochem. Biophys., 82:491-492.

Li, Y. L. and F. A. Digiano, 1983. Availability of sorbed substrate for microbial
degradation on granular activated carbon. Journal WPCF, 55:392-399.

Lowry, J. D. and C. E. Burkhead, 1980. The role of adsorption in biologically extended
activated carbon columns. Journal WPCF, 52:389-398.

Machado, R. J. and C. P. Leslie Grady Jr., 1989. Dual substrate removal by an axenic
bacterial culture. Biotechnology and Bioengineering, 33:327-337.

Maqsood, R., 1977. Low-temperature organic removal and denitriﬁcation in activated
carbon columns. Journal WPCF, 49:2107-2117.

Marr, E. K. and R. W. Stone, 1961. The bacterial oxidation of benzene. Journal
Bacteriol., 81:425.

Merk, W., W. Fritz and E. U. Schlunder, 1980. Competitive adsorption of two dissolved
organics onto activated carbon - 111. Chemical Engineering Science, 36:743-757.

131

Meyer, J. S., M. D. Marcus and H. L. Bergman, 1984. Inhibitory interactions of aromatic

organics during microbial degradation. Environmental Toxicology and Chemistry, 3:583-
587.

Miller, G. W. and R. G. Rice, 1978. European water treatment practices - the promise of
Biological Activated Carbon. Civil Engineering, 48(2) :81-83.

Monod, J., 1942. Recherches sur la croissance des cultures bacteriennes. Paris: Hermann
and Cie, Paris.

Monod, J., 1949. The growth of bacterial cultures. Annual Review of Microbiology,
3:371-394.

Nayar, S. C. and N. D. Sylvester, 1979. Control of phenol in biological reactors by
addition of powdered activated carbon. Water Research, 13:201-205.

Nelson, M. J. K., S. O. Montogomery, W. R. Mahaffey and P. H. Pritchard, 1987.
Biodegradation of trichloroethylene and involvement of an aromatic biodegradative
pathway. Appl. and Env. Microbiol., 53(5) :949-954.

Nelson, M. J. K., S. 0. Montgomery and P. H. Pritchard, 1988. Trichloroethylene

metabolism by microorganisms that degrade aromatic compounds. Appl. and Env.
Microbiol., 54(2) :604-606.

Nozaka, J. and M. Kusunose, 1968. Metabolism of hydrocarbons in microorganisms part
I. Oxidation of p-xylene and toluene by cell-free enzyme preparations of pseudomonas
aeruginosa. Agr. Biol. Chem., 32(8) :1033-1039.

Nozaka, J. and M. Kusunose, 1969. Metabolism of hydrocarbons in microorganisms Part
II. Degradation of toluene by cell-free extracts of Pseudomonas mildenbergii. Agr. Biol.
Chem., 33(6) :962-964.

Olmstead, K. P., 1989.
W University of Michigan Ann Arbor MI

Omori, T., S. . Horiguchi and K. Yamada, 1967. Studies on the utilization of
hydrocarbons by Microorganisms Part X. Screening of Aromatic Hydrocarbon-
Assimilating Microorganisms and p-Toluic Acid Formation from p-Xylene. Agr. Biol.
Chem., 31(11) :1337-1342.

Omori, T. and K. Yamada, 1969. Studies on the utilization of hydrocarbons by
microorganisms Part XIII. Oxidation of m-xylene and Pseudocumene by Pseudomonas
aeruginosa. Agr. Biol. Chem., 33(7) :979-985.

Omori, T. and K. Yamada, 1970a. Studies on the utilization of hydrocarbons by
microorganisms Part XVI. Detection of metabolic intermediates of xylene and
pseudomonas. Agri. Biol. Chem., 34(5) :659-663.

Omori, T. and K. Yamada, 1970b. Studies on the Utilization of hydrocarbons by
microorganisms Part XVII. Metabolism of p-xylene and related compounds. Agr. Biol.
Chem., 34(5) :664-669.

Peel, R. G. and A. Benedek, 1983. Biodegradation and adsorption within activated carbon
adsorbers. Journal WPCF, 55:1168—1173.

132

Perrotti, A. E. and C. A. Rodman, 1973. Enhancement of biological waste treatment by
activated carbon. Chemical Engineering Progress, 69:63-64.

Perrotti, A. E. and C. A. Rodman, 1974. Factors involved with biological regeneration of
activated carbon. AIChE Symp. Series, 70(144) :317-325.

Pirbazari, M., T. C. Voice and W. J. Weber, Jr., 1990. Evaluation of bioﬁlm development

on various natural and synthetic media. Hazardous Waste & Hazardous Materials, 7(3)
:239-250.

Ribbons, D. W. and R. W. Eaton, 1982. Chemical transformations of aromatic
hydrocarbons that support the growth of microorganisms. In A. M. Chakrabarty (Eds. ),
- (PP 59- 84). Boca Raton,

 

Florida: CRC Pre$s,1nc. .

Richards, D. J. and W. K. Shieh, 1986. Biological fate of organic priority pollutants in the
aquatic environment. Water Research, 20(9) :1077- 1090.

Rice, R. G. andC. M. Robson., 1982.
MW. Ann Arbor. Ann Arbor Sciene Publisher

Ridgway, H. F., J. Safarik, D. Phipps and D. Clark, 1990. Identiﬁcation and catabolic
activity of well-derived gasoline-degrading bacteria from a contaminated aquifer. Appl.
and Env. Microbiol., 56(11) :3565-3575.

Rodman, C. A., E. L. Shunney and A. E. Perrotti, 1978. Biological regeneration of
activated carbon. In P. N. Cheremisinoff and F. Ellerbusch (Eds), W (pp.
449-483). Ann Arbor, MI: Ann Arbor Sciene Publisher

Saez, P. B. and B. E. Rittmann, 1991. Biodegradation kinetics of 4-chlorophenol, an
inhibitory co-metabolite. Journal WPCF, 63(6) :838-847.

Schmidt, 8. K., S. Simkins and M. Alexander, 1985. Models for the kinetics of
biodegradation of organic compounds not supporting growth. Appl. and Env. Microbiol.,
50(2) :323-331.

Schultz, J. R. and T. M. Keinath, 1984. Organic removal mechanisms in biophysical
treatment systems. Water Science Technology, 17: 1043- 1054.

Shanna, B. and R. C. Ahlert, 1977. Nitriﬁcation and nitrogen removal. Water Research,
11:897-925.

Sohngen, N. L., 1913. Centr. Bakteriol. Parasitenk. Abt. 11, 37:595.

Sontheimer, H., E. Heilker, M. R. Jekel, H. Nolte and F. H. Vollmer, 1978. The
mulheim process. Journal AWWA, 79:393-396.

Sontheimer, H., 1979. Applying oxidation and adsorption techniques: A summary of
progress. Journal AWWA, 79(11) :612-617.

Speitel,G.E.,Jr.,1985.,'o - '

 

i h ' 1 I I. 1 University of North Carolina,

Chaple Hill, NC

133

Speitel, G. E., Jr. and F. A. DiGiano, 1987. The Bioregeneration of GAC Used to Treat
Micropollutants. Journal AWWA, 87:64-73.

Speitel, G. E., Jr., K. Dovantzis and F. A. Digiano, 1987. Mathematical modeling of
bioregeneration in GAC Columns. Journal of Environmental Engineering, 113(32):48-56.

Speitel, G. E., Jr., C. J. Lu and M. H. Turakhia, 1988. Biodegradation of synthetic
organic chemicals in GAC beds. Wat. Sci. Tech., 20(1 1/ 12) :463-465.

Speitel, G. E., Jr., and F. A. DiGiano, 1988. Determination of microbial kinetic

coefficients through measurement of initial rates by radiochemical techniques. Water
Research, 22(4) :829-835.

Speitel, G. E., Jr., C. J. Lu, M. Turakhia and X.-J. Zhu, 1989. Biodegradation of trace
concentration of substituted phenol in granular activated carbon columns. Environ. Sci. &
Tech., 23(1) :68-74.

Speitel, G. E., Jr., C. J. Lu, X. J. Zhu and M. H. Turakhia, 1989. Biodegradation and
adsorption of a bisolute mixture in GAC Columns. Journal WPCF, 61(2) :221-229.

Storer, F. F. and J. Gaudy A. F., 1969. Computational analysis of transient response to
quantitative shock loadings of heterogeneous p0pulations in continuous culture. Environ.
Sci. & Tech., 3(2) :143-149.

Stormer, K., 1908. Zentralbl. Bakteriol. Parasitenk. Infek., 20:282.

Strand, S. E., M. D. Bjelland and H. D. Stensel, 1990. Kinetics of chlorinated

hydrocarbon degradation by suspended cultures of methane-oxidizing bacteria. Journal
WPCF, 62(2) :124-129.

Suidan, M. T., P. Fox and J. T. Pfeffer, 1978. Water Science and Technology, 19:229-
236.

Summers, R. S., B. Haist, J. Koekler, J. Ritz, G. Zimmer and H. Sontheimer, 1989. The
influence of background organic matter on GAC adsorption. Journal AWWA, 89(5) :66-
74.

Sundstrom, D. W. and H. E. Klei, 1979. W. New Jersey: Prentice-
Hall.

Tabak, H. H., S. A. Quave, C. I. Mashni and E. F. Barth, 1981. Biodegradability studies
with organic priority pollutant compounds. Journal WPCF, 53(10) :1503-1518.

Thomas, W. J. and J. L. Lombardi, 1971. Binary adsorption of benzene-toluene mixtures.
Trans. lnstn Chem. Engrs., 49:240-250.

Tien, C. and S. C. Wang, 1982. Dynamics of adsorption columns with bacterial growth
outside adsorbents. The Canadian Journal of Chemical Engineering, 60:363-376.

Trulear, M. G. and W. G. Characklis, 1982. Dynamics of bioﬁlm processes. Journal
WPCF, 54(9) :1288.

134

USEPA. 1976. Health effects of benzene: a review, part VI: Occupational exposure to
benzene. Wm, 42(103) :27452-27459 EPA 560/5-76—003.

Van Der Kooij, D., 1976a. Process during biological oxidation in ﬁlters. In Sentheiner

(Ed.,) WW 9 (pp 689- 699) EPA
600/9 - 76 - 030.

Van Der Kooij, D. 1976b. Some investigations into the presence and behaviour of bacteria

in activated carbon ﬁlters. In Sentheiner (Ed.,) WW
muemmmteﬂechnologx 9 (pp 348- 354)

Verschucren. K.. 1983. WEE New
York: Van Nostrand Reinhold Company.

Voice, T. C., D. Pak, X. Zhao, J. Shi and R. Hickey, 1992. Biological activated carbon in
ﬂuidized bed reactor for the treatment of groundwater contaminated with volatile aromatic
hydrocarbons. Water Research, 26(10) :1389-1401.

Weber, W. J., Jr., C. B. Hopkins and R. Bloom Jr., 1970. Physico-chemical treatment of
wastewater. Journal WPCF, 42:83-99.

Weber, W. J., Jr., C. B. Hopkins and R. Bloom Jr., 1971. Expanded-bed adsorption for
treatment of sewage efﬂuents. Water-1970:541-553.

Weber, W. J., Jr., L. D. Friedman and R. Bloom Jr., 1972. Biologically-extended
physicochemical treatment :.

Wilson, J. T. and B. H. Wilson, 1985. Biotransformation of trichloroethylene in soil.
Appl. Env. Microbiol., 49(1) :242-243.

Ying, W. C. and W. J. Weber Jr., 1979. Bio-physicochemical adsorption systems model.
Journal WPCF, 51(11) :2661-2677.

Yoon, H., G. . Klinzing and H. W. Blanch, 1977. Competition for mixed substrates by
microbial populations. Biotechnology and Bioengineering, 19:1193—1210.

Zylstra, G. J, S. Chauhan and D. T. Gibson Degradation of chlorinated biphenyls by

escherichia coli containing cloned genes of the pseudomonas putida f1 toluene catabolic

pathway. In 311' 01 o 1' 1 _11_-_ ‘ -_ 1110 1101 10 D 0".
‘ - . ' - - - .USEPA.