1 J... 4W...
59...? . .
fa... ..

I:
.5 m ,. an?

1’?"

. a :uet. .
.3; ¢ - .. .

twain». “ . . . V .
- if". . . . . . . . .
. J. n.3,» . _ . . . .. , . . .
. . , . . . . . ....u.,....f.l.v.v..1~1uu “.3..."
x > )5 lat»‘4nr.¢;\. I 4‘ ‘IVC‘IIb.
o v u A,‘ -
«rd..-

n... V .
-...-:. m

.WWWN. ...

 

kit
. flnfisl. t
3: w: 1 .

 

 

 

 

.lv.

 

 

 

n. -1..U.....w
ind. . . .\
... JIM!
. bra}...
.1..th

u.

 

 

 

.u?.. A .

.rl.

 

140' t. I;
ft.4ln|}~. L.

4:U-t.‘.ll r . ,

\l. .c l\!4 . .v .

. . ‘{?'.1

1r . . ..|lr\1

. 4:..- 1

v .1..-
it. .iv.

.1511

v .

 

u. l
.0‘1 u . ¢ .L IV. .5 117-.
lunvlhuflntavv .9“. ~-t..|' 910‘. L...

V!»- . “1.3! :u 1.. . . .1.» V‘.1\4.L‘....:

. . I . It fit I-Ioot

.. _
mm

“on“

. .H-........H..1...4.......

§$&. “mm"..kwflkhod «haunt r .6 ..

 

\

'THESlS

MISCHIGAN

l 'HWUHHIUIIIUlllilllllllllllllllllll“Ill

293 01581 2179

IN”

 

 

 

 

 

 

 

 

LIBRARY

Mlchlgan State
Unlverslty

 

 

 

This is to certify that the

dissertation entitled

THE MECHANISM OF THE REACTION OF OZONE
WITH PYRENE AND BENZ[a]ANTHRACENE
IN ACETONITRILE / WATER MIXTURE

presented by

J EHNG-JY UN YAO

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Environmental Engineering

 

 

/ M r professor

Date December, 1996

 

MS U i: an Affirmatiml Action/Equal Opportunity Institution 0-12771

 

PLACE ll RETURN BOX to romovo this checkout from your record.
FINES return on or before date duo.

TO AVOI
DATE DUE DATE DUE DATE DUE ll

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU loAn Affirmative Mon/Equal Opportunity Inukwon
WW1

THE MECHANISM OF THE REACTION OF OZONE
WITH PYRENE AND BENZ[a]ANTHRACENE
IN ACETONITRILE/WATER MIXTURE

By

Jehng-Jyun Yao

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

1997

ABSTRACT

THE MECHANISM OF THE REACTION OF OZONE
WITH PYRENE AND BENZ[a]ANTHRACENE
IN ACETONITRILE/WATER MIXTURE

By

Jehng-Jyun Yao

The ozonation of pyrene and benz[a]anthracene was studied. The toxicity of ozonation
products was assessed by monitoring gap junction intercellular communication (GJIC).
The intermediate ozonation pyrene products were more inhibitory to GJIC than pyrene
and the final products. Therefore, the identification of the ozonation products and an
understanding of the reaction pathways that occur in aqueous solutions is important in

minimizing the formation of toxic products during ozone treatment.

The ozonation products were identified and quantified using gas chromatography/mass
spectrometry (GC/MS). For the ozonation of pyrene, 14 products including aldehyde and
carboxylic acid substituted phenanthrene- and biphenyl-type oxidation products were
identified. Initial ring cleavage of pyrene at the 4,5 position produced phenanthrene-type
products. Prior to the complete disappearance of pyrene, secondary ring cleavage at the
9,10 position occurred. After the disappearance of pyrene, the concentration of the
phenanthrene-type products decreased dramatically and the secondary ring cleavage
biphenyl-type products predominated. A stoichiometric ratio of 1.68 moles ozone/mole
pyrene was required to completely destroy pyrene. For the ozonation of
benz[a]anthracene, 15 products including quinone type and phenyl-naphthyl type
products were identified. Ozone attacked either on the carbon at the 7 and/or 12 position
producing the quinone or hydroxyl functional groups, or at the 5,6 position resulting in
ring cleavage and the formation of phenyl-naphthyl type products. These products were
formed simultaneously.

Initial attack by ozone on either the bond or atom having the lowest localization energy
occurred preferentially. For pyrene, ring cleavage occurred at the bond having the lowest
energy, and resulted in a stepwise ring cleavage pathway. For the ozonation of
benz[a]anthracene, both atom attack and bond attack occurred, but higher preference was
for bond attack.

In aqueous systems, -OH radicals have greatest effect on the initial bond attack type of
ozonation reactions and have least effect on initial atom attack type of reactions. Free
radicals also reacted with ozonation products. Conducting the ozonation experiments in
aqueous solutions containing a non-participating solvent were more reliable in predicting
aqueous products than were organic solvent. Using the results of these studies we were
able to predict ozonation products for compounds that have not been previously studied.

In honor of the giant in my life - my courageous and intelligent father.

iv

ACKNOWLEDGMENTS

I wish to extend my sincere appreciation to all persons who provided me with
guidance and assistance during this study. In particular, I would like to thank Dr. Susan J.
Masten not only for her professional guidance and valuable suggestions as my adviser
also for her attentions and help as my friend throughout these years in Michigan State
University. Also, I wish to thank Dr. Zhi-Heng Huang, one of my committee members,
for his special instruction in GC/MS analysis technique and valuable suggestions
throughout this research. Appreciation is also extended to my other committee members,
Dr. Sharon J. Anderson, Dr. Douglas A. Gage, Dr. Simon H. Davies and Dr. William E.

Saul, for their helpful suggestions concerning this study.

I would like to acknowledge the following sources of funding: MSU-NIH Mass
Spectrometry Facility which is supported, in part, by a grant (RR00480) to J. Throck
Watson from the National Institutes of Health, National Center for Research Resources.
NIEHS-Superfund grant 2P42ESO4911-05; U. S. AF OSR (F49620-92-J-0293); Meridian
Instrument support ACAS grants fi'om Michigan State University; USDA Food Safety &
Toxicology Center, Michigan State University; National Science Foundation Grant BCS

9016994.

I would like to thank my best friend, Dr. Munjed A. Maraqa, for his
encouragement throughout these years of my study, Dr. Michael I-yuang Hsu for his
helpful discussions in ozonation work. I would also like to thank the assistance of Beth R.
Lockwood in the work of cell culture, Naxing Xu in the GC/CI work and the assistance of
Holly Homer in the work of NMR. Special thank to Dr. J. Throck Watson and Dr.

Muraleedharan Nair for their generously allowing me to use their facilities.

Finally, I would like to thank my parents, Meng-Han Yao and Ai-Hui Wang, and

all of my sisters for their encouragement, support and sacrifice for the duration of my
graduate studies at Michigan State University. I could not have accomplished any

achievements without our strong and extraordinary family tie.

vi

TABLE OF CONTENTS

 

 

 

 

LIST OF TABLES x
LIST OF FIGURES xi
LIST OF SYMBOLS xvii
CHAPTER 1 INTRODUCTION 1
1.1 Environmental Significance .................................................................................... 1
1.2 Choice of Model Compounds ................................................................................. 2
1.3 Background .......................................................................................................... 4
1.4 Research Objective .................................................................................................. 15
1.5 Organization of Thesis ............................................................................................. 16
1.6 References .......................................................................................................... 19

CHAPTER 2 DETERMINATION OF THE EFFICACY OF OZONE
TREATMENT SYSTEMS USING A GAP JUNCTION

INTERCELLULAR COMMUNICATION BIOASSAY ............ .24

Abstract ...................................................................................................... 24

2.1 Introduction .......................................................................................................... 25
2.2 Methods ...................................................................................................... 30
2.2.1 Cell Culture ............................................................................................. 30

2.2.2 Bioassay of GJIC .................................................................................... 30

2.2.3 Ozonation ................................................................................................ 32

2.2.4 I-IPLC Analysis ....................................................................................... 33

2.3 Results ...................................................................................................... 34
2.4 Discussion ...................................................................................................... 44
2.5 Conclusion ...................................................................................................... 48
2.6 References ...................................................................................................... 49

vii

CHAPTER 3 THE OZONATION OF PYRENE IN ACETONITRILE/

3.1
3.2
3.3
3.4
3.5
3.6

4.1
4.2
4.3
4.4
4.5
4.6

5.1
5.2
5.3

A.l
A2
A3
A.4

WATER MIXTURE: PATHWAY AND PRODUCT

 

 

 

 

IDENTIFICATION 62
Abstract ...................................................................................................... 52
Introduction ...................................................................................................... 53
Methods ...................................................................................................... 55
Results ...................................................................................................... 61
Discussion ...................................................................................................... 76
Conclusions .......................................................................................................... 82
References ...................................................................................................... 83
CHAPTER 4 THE OZONATION OF BENZ[a|ANTHRACENE IN
ACETONITRILEIWATER MIXTURE. PATHWAY AND
PRODUCT IDENTIFICATION 87
Abstract ...................................................................................................... 87
Introduction ...................................................................................................... 88
Methods ...................................................................................................... 89
Results ...................................................................................................... 92
Discussion ...................................................................................................... 103
Conclusions .......................................................................................................... 109
References ...................................................................................................... 110
CHAPTER 5 CONCLUSION AND RECOMMENDATIONS _ _ 113
Conclusion .......................................................................................................... 113
Recommendations ................................................................................................... 121
References .......................................................................................................... 122
APPENDIX A KINETICS STUDY 126
Introduction ...................................................................................................... 126
Kinetics Model .................................................................................................... 129
Method ...................................................................................................... 130
Results and Discussion ....................................................................................... 134
References .......................................................................................................... 138

A5

viii

APPENDIX B

APPENDIX C

APPENDIX D

MASS SPECTRA AND INTERPRETATION OF
SILYLATED PYRENE OZONATION PRODUCTS .................. 140

OZONATION PRODUCTS PURIFICATION 178

MASS SPECTRA AND INTERPRETATION OF
SILYLATED BENZ[a|ANTHRACENE OZONATION
PRODUCTS 184

 

ix

Table 1 .1

Table 1.2

Table 1.3

Table 2.1

Table 3.1

Table 4.1

Table A.1

Table C]

Table C.2

LIST OF TABLES

Target compounds studied .......................................................................... 3

Rate constants for the reaction of ozone and various PAHs in water ...... 10

Byproducts formed during the ozonation of various PAHs .................... 12
Effect of ozonated byproducts of pyrene on GJIC .................................. 43
GC/MS characteristics of pyrene ozonation products ............................. 64
GC/MS characteristics of benz[a]anthracene ozonation products .......... 96

Rate constants and stoichiometric coefficients for the ozonation of
PAHs ...................................................................................................... 127

Separation flow chart for sample reacted with ozone for 10 minutes ..... 179

Separation flow chart for sample reacted with ozone for 20 minutes ..... 180

Figtu'e 1.1

Figure 1.2

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

Figure 2.5

LIST OF FIGURES

Mechanism of the Ozonation Reaction
(Adapted fiom Bailey, 1972) .................................................................. 6

Reactions of Aqueous Ozone in the Presence of Micropollutant M
(Adapted from Staehelin and Hoigné, 1985) .............................................. 8

Reaction of ozone with an unsaturated carbon in the presence and
absence of water. The primary products in the presence of water
are H202, an aldehyde, and a carboxylic acid if R2= H ........................... 28

(A) Chromatographic separation by RP-HPLC of ozonated pyrene that
was dissolved in various concentrations of acetonitrile/water mixtures.
The solid, dashed, and dotted lines were mixtures containing 10%,

25%, and 60% water, respectively. This separation used linear gradient
system B.

(B) Chromatographic separation by RP-HPLC of various

concentrations of pyrene ozonated in a 90%l10%, acetonitrile/water
mixtures. The solid, dashed, and dotted lines were mixtures containing
initial concentrations of 5.0, 2.5, and 1.0 mM pyrene, respectively.

This separation used linear gradient system B ........................................ 35

(A) Ozone demand of the reactor for various concentrations of pyrene.
(B) An example of the raw data collected during the reaction of ozone
with 0.1 mM pyrene ................................................................................ 37

3-D plot of RP-HPLC chromatograms of 5 mM pyrene reacted with
ozone for different ozone contact time intervals. The average flow rate
was 1.25 mmol of ozone/min. This separation used linear gradient
system A .................................................................................................. 3 8

Determination of the ozone dosage required for the removal of pyrene

and the oxidation byproducts from the solution that were inhibitory to
GJIC. D = the concentration of pyrene, which was an average of two

xi

Figure 2.6

Figure 3.1

Figure 3.2

Figure 3.3

Figure 3.4

Figure 3.5

Figure 3.6

Figure 3.7

Figure 3.8

Figure 3.9

replicates, with a detection limit of 500 nM as determined using RP-
HPLC. O = the inhibition of GJIC, which was an average i 1 SD of

the distance the dye traveled fiom three different plates per

experiment ............................................................................................... 41

Representative chromatogram of the by-product peaks collected for
analysis of GJIC activity shown in Table 2.1. This separation used
linear gradient system B .......................................................................... 42

Configuration of ozonation system ......................................................... 56

Ozone breakthrough curve for ozonation of 5 mmoles/L pyrene
dissolved in 90% CH3CN/water solvent mixture. 0 Influent ozone
absorbance; I Effluent ozone absorbance .................................................. 57

The effect of ozone dosage on the residual pyrene concentration. The
residual pyrene concentration is determined by the GC area response

for pyrene (Aw) relative to the response for 1,8-naphthalaldehydic acid
(25 uL of 5 mM) used as an internal standard (Al). The solid line
represents the linear regression based on the first six data points. The
arrow corresponds to the point where 0.5 mole of pyrene had
completely reacted by a total ozone demand of 0.84 mole, i.e., a ratio
of 1.68 moles ozone/mole pyrene ........................................................... 62

Products formed from the ozonolysis of pyrene ..................................... 63

(a) The reaction of pyrene (0.5 mmole) with 0.38 mole of ozone.
(b) The reaction of pyrene (0.5 mmole) with 1.34 moles of ozone.
The Roman numerals corresponds to the products listed in Figure 3.4...65

Mass spectra and structural interpretation of fragmentation patterns for
4-carboxy-5-phenanthrenecarboxyaldehyde (111 MB) derivatized by
BSTFA + TMCS. The migration of Me3Si within a molecule is

described in References 24,25. The bold numbers correspond to the
important peaks on the mass spectra ....................................................... 67

Mass spectrum and structural interpretation of fragmentation patterns

for 2’,6,6’-biphenyltrialdehydo-2-carboxylic acid (X) derivatized by
BSTFA + TMCS ..................................................................................... 69
Mass spectra of the isomeric forms of XI A/B ....................................... 70

The relative concentrations (A/AI) of pyrene and its ozonation
products obtained under different ozone : pyrene ratios. The product

xii

Figure 3.10

Figure 3.11

Figure 4.1

Figure 4.2

Figure 4.3

Figure 4.4

Figure 4.5

Figure 4.6

concentrations were determined relative to 1,8-naphthalaldehydic acid
(25 uL of 5 mM) used as an internal standard ........................................ 72

The relative concentration (A/AI) of 4-CPA and its ozonation products
obtained under different ozone : 4-CPA ratio. The product

concentrations were determined relative to 1,8-naphthalaldehydic acid
(25 uL of 5 mM) which was used as an internal standard (AI) ............... 75

Proposed pathway for the first ring cleavage of pyrene during

ozonolysis. Compounds shown in brackets [ ] were detected only in
ozonated solutions which were concentrated by fieeze—drying and
reconstituted in acetonitrile. The compound shown in { } is
hypothesized, and would be formed by a sequence of reactions

involving -OH in the presence of O2 ........................................................ 78

Products formed from the ozonolysis of benz[a]anthracene ................... 93

Typical GC chromatogram. Figures (3), (b), (c), (d) correspond to the
reaction of benz[a]anthracene (0.1 mmole) with 0.05, 0.11, 0.20 and

0.42 mole of ozone, respectively. The Roman numerals corresponds

to the products listed in Figure 4.1 .......................................................... 95

(A) Mass spectrum of 5,6«dihydrochrysenediol obtained from NIST
library search. The abundant ions m/z 202, 215 and 231 are

characteristic of o, o’-disubstituted phenyl-naphthyls.

(B) Mass spectrum of an ozonation product, 2-(2’-formyl)phenyl-3-
naphthoic acid or 2-(2’-carboxy)phenyl-3-naphthaldehyde (V II A,B),
derivatized as the trimethylsilyl ester using BSTFA + 1 % TMCS. The
phenyl-naphthyl backbone was evidenced by the presence of triplet

ions m/z 202, 215 and 231, where m/z 215 was formed through a

Me3Si- migration mechanism (cf:Ref 18,19) followed by loss of CO2

and Me3SiO (= 44 +89 = 133) from the molecular ion m/z 348 ............. 98

The relative concentrations of benz[a]anthracene and its ozonation
products obtained under different ozone : BaA ratios. The product
concentrations were determined by the GC area response for

compound (A) relative to the response for 1,8-naphthalaldehydic acid

(50 [IL of 1 mM) used as an internal standard (AI) ................................ 100

Proposed pathway for the ozonation of benz[a]anthracene. Compounds
shown in brackets [ ] were detected at concentrations which were

below detection limit ............................................................................... 104

The effect of ozone dosage on the concentrations of two major products,
2-(2’-formyl)phenyl-3-naphthaldehyde (V) and benz[a]anthraquinone (X).

xiii

Figure 5.1

Figure A.1

Figure A.2

Figure A.3

Figure B]

Figure B.2

Figure B.3

Figure B.4

Figure B.5

Figure B.6

Figure 3.7

Figure 3.8

The residual concentration is determined by the GC area response for
compound (A) relative to the response for 1,8-naphthalaldehydic acid (50
uL of 1 mM) used as an internal standard (AI). 0 BaA; A Product V; I

Product X ................................................................................................ 108
Structure of the example PAHs .................................................................. 118
Configuration of kinetics apparatus ........................................................ 132

The stoichiometric coefficient for the reaction of ozone with pyrene.
Different symbols indicated different set of data; solid line is linear
regression result with r2 = 0.92 ................................................................... 135

The rate constant for the reaction of ozone with pyrene in aqueous solution.

AplotoflnA vs T. Where A=w and T=([03]o—§[M]o)t.

[03 I. '[M] ’
I = calculated k from each data point; solid line is the linear regression
result (r2 = 0.96) ........................................................................................... 137

Products formed fiom the ozonolysis of pyrene. Roman numbers
correspond to the ID assigned to each product. Molecular weights are

given in parentheses ................................................................................ 141

Common ions of mass spectra of ozonation products formed from
pyrene ...................................................................................................... 142

Interpretation of the mass spectrum for the silylation product 11.
Molecular ion weight = 234 .................................................................... 145

Interpretation of the mass spectra for the silylation product 111 A,B.
Molecular ion weight = 322 .................................................................... 147

Interpretation of the mass spectrum for the silylation product IV.
Molecular ion weight = 294 .................................................................... 150

Interpretation of the mass spectrum for the silylation product V.
Molecular ion weight = 396 .................................................................... 152

Interpretation of the mass spectrum for the silylation product VI.
Molecular ion weight = 206 .................................................................... 154

Interpretation of the mass spectrum for the silylation product VII.
Molecular ion weight = 294 .................................................................... 156

xiv

Figure 3.9

Figure 3.10

Figure B.11a

Figure B.11b

Figure B.12a

Figure B. 12b

Figure B.12c

Figure B.12d

Figure B.13a

Figure B.13b

Figure 3.14

Figure 3.15

Figure 3.16

Figure C]

Figure C.2

Interpretation of the mass spectrum for the silylation product VIII.
Molecular ion weight = 220 .................................................................... 158

Interpretation of the mass spectrum for the silylation product IX.
Molecular ion weight = 266 .................................................................... 160

Interpretation of the mass spectrum for the silylation product X.
Molecular ion weight = 354 .................................................................... 162

Interpretation of the mass spectrum for the silylation product X with
SiMe3 migration dm'ing GC/MS analysis. Molecular ion weight = 354.163

Interpretation of the mass spectrum for the silylation product XI A,B.
Molecular ion weight = 442 .................................................................... 165

Interpretation of the mass spectrum for the silylation product XI A,B with
SiMe, migration during GC/MS analysis. Molecular ion weight = 442.166

Interpretation of the mass spectrum for the silylation product XI A,B.
Molecular ion weight = 442 .................................................................... 167

Interpretation of the mass spectrum for the silylation product XI A,B with
SiMe3 migration during GC/MS analysis. Molecular ion weight = 442.168

Interpretation of the mass spectrum for the silylation product XII.
Molecular ion weight = 530 .................................................................... 170

Interpretation of the mass spectrum for the silylation product XII with
SiMe3 migration during GC/MS analysis. Molecular ion weight = 530..171

Interpretation of the mass spectrum for the silylation product XIII.
Molecular ion weight = 618 .................................................................... 173

Interpretation of the mass spectrum for the silylation product XIV.
Molecular ion weight = 426 .................................................................... 175

Interpretation of the mass spectrum for the silylation product XV.
Molecular ion weight = 336 .................................................................... 177

GC chromatograms of sample F3 from 10 min ozonation sample followed
by separation on a Cl8 coltunn ................................................................ 182

CC chromatograms of sample F4+5~3 from 10 min ozonation sample
followed by a second separation ............................................................. 183

XV

Figure D]

Figure D.2

Figure D.3

Figure D.4

Figure D.5

Figure D.6

Figure 13.7

Figure D.8

Figure D.9

Figure D.10

Products formed from the ozonolysis of benz[a]anthracene. Roman
numbers correspond to the ID assigned to each product. Molecular weights
are given in parentheses .......................................................................... 185

Interpretation of the mass spectra for the silylation product 11 A,B.
Molecular ion weight = 316 .................................................................... 188

Interpretation of the mass spectra for the silylation product 111 A,B.
Molecular ion weight = 404 .................................................................... 190

Interpretation of the mass spectrum for the silylation product IV.
Molecular ion weight = 334 .................................................................... 192

Interpretation of the mass spectra for the silylation product VI A,B.
Molecular ion weight = 332 .................................................................... 195

Interpretation of the mass spectra for the silylation product VII A,B.
Molecular ion weight = 348 .................................................................... 197

Interpretation of the mass spectrum for the silylation product VIII.
Molecular ion weight = 436 .................................................................... 199

Interpretation of the mass spectra for the silylation product IX A,B.
Molecular ion weight = 348 .................................................................... 201

Interpretation of the mass spectra for the silylation product XI A,B.
Molecular ion weight = 260 .................................................................... 204

Interpretation of the mass spectra for the silylation product XII A,B.
Molecular ion weight = 348 .................................................................... 206

xvi

[0310
PAHs

PY
[PY]

RP-HPLC

LIST OF SYMBOLS

GC area response of product

GC area response of internal standard
GC area response of pyrene
benz[a]anthracene

concentration of benz[a]anthracene, mM
ozone influent concentration, mM

ozone effluent concentration, mM
4-carboxy-5-phenanthrenecarboxyaldehyde
fluorene

gas chromatography

gap junction intercellular communication
the lowest atom localization energy, eV
the lowest bond localization energy, eV
reactant

reactant concentration, M

initial reactant concentration, M

mass spectrometry

naphthalene

ozone

ozone concentration, M or mM

average ozone concentration, M

initial ozone concentration, M

ozone consumption, mM

polycyclic aromatic hydrocarbon

phenanthrene

pyrene

pyrene concentration, mM

gas flow rate, mL/min

reverse phase-high performance liquid chromatography
aqueous solubility, ppb

xvii

k rate constant, M’1 s'1 or mM" s‘1

t reaction time, sec

t”, reaction half-life, sec

v/v volume to volume ratio
w/v weight to volume ratio

A concentration difference
é stoichiometric coefficient

Roman Numeral assigned for ozonation products ID

xviii

CHAPTER 1

INTRODUCTION

1.1 Environmental Significance

Polycyclic aromatic hydrocarbons (PAHs) are found in many petroleum products
including crude oils, motor oils, gasoline and heating fuels. The predominant routes by
which PAHs enter the environment are petroleum spillage and the release of PAHs during
the combustion of petroleum products with subsequent wet or dry deposition. It is estimated
that nearly 230,000 tons/year of total PAHs enter the aquatic environment from various
sources (Neff, 1979 p.227). As a result of the release of PAHs to the environment, these
compormds find their way into drinking waters, soils, plants and wastewater. In a review
article by Harrison et a1. (1975), it was noted that concentrations of benz[a]pyrene of 40 -
1300 mg/L have been found in dried soil taken from presumably uncontaminated forests
and fields. In plants and vegetables, concentrations of individual PAHs have been found to
range from 5 to 110 mg/kg of dry material. Raw sewage from domestic sources can contain
significant levels of PAHs (Bourcart et al., 1965; Han'ison et al., 1975). Rainfall may
increase the levels of PAHs in sewage by more than lOO-fold over that which occurs during
dry weather periods. Because of their low water solubilities (F utoma et al., 1981), PAHs
are rapidly adsorbed to organic and inorganic particulate materials or are accmnulated in the

tissues of aquatic organisms. PAHs released to the environment tend to remain close to the

l

2

sites of deposition. Therefore, lakes, rivers, estuaries, and coastal marine environments near
centers of human population are the primary repositories of aquatic PAHs. In Great Lakes
water, high concentrations of pyrene and fiuoranthene (11.2 and 10.6 ng/liter, respectively)
have been found (Eadie, 1983). The concentration of individual PAHs in sediments ranged

from 30 to 750 mg/kg dry wt.

PAHs are of significant environmental importance because they are not readily
biodegradable (Sims and Overcash, 1983). They are persistent in soils (Stevens et al.,
1989), bioaccumulated (Smith et al., 1984; Solbakken et a1. 1983; Hallett et al., 1983), and
either carcinogenic or potentially carcinogenic (Griest et al., 1987; Cavalieri et a1. 1980;
Thorton et al., 1981). Sixteen PAHs are listed among the 129 US. EPA Priority Pollutants

(EPA, 1984).

1.2 Choice of Model Compounds

Two PAHs will be used as model compounds (listed in Table 1.1); they are pyrene
and benz[a]anthracene. The target compounds are chosen because 1) they belong to
different classes of PAHs and should have different ozonation mechanisms, 2) they are
reactive with both ozone and the -OH radical, 3) they are recalcitrant (Howard et al., 1991),
and 4) they have a range of mutagenic activities but are not strongly carcinogenic to reduce

healths risks during laboratory studies.

Table 1.1 Target compounds studied

Compound . Pyrene Benz[a]anthracene
Abbreviation PY BaA

Relative - b + b
Carcinogenicityb

Most Reactive 4,5 5,6
Bond‘I

Lowest Bond- 1.06 1.03
Localization

Energy'

Most Reactive 3 7
Atom‘II

Lowest Atom- 1.51 1.35
Localization

Energy‘

Structure

 

a. Bailey, 1982. In: Ozonation In Organic Chemistry. 39(2): 43-76.
b. Harrison, 1975. War. Res, 92331-346.
Relative activity on mouse epidermis: +++, active; ++, moderate; +, weak; -, negative

1.3 Background

The most important processes that govern the fate of PAHs in the environment are
photooxidation, chemical oxidation and biodegradation. In atmospheric systems chemical
oxidation by ozone or the OH radical along with photooxidation are the predominant
mechanisms responsible for the degradation of PAHs (Nefl‘, 1979). In soil systems, only
low molecular weight hydrocarbons, such as substituted benzenes, can be biodegraded. The
higher molecular weight PAHs, however, are highly recalcitrant with half-lives on the order
of 210 days to 5.2 years (Howard et al., 1991). Both chemical oxidation and biodegradation
can be exploited in developing systems to treat PAHs. If the PAHs are first oxidized using
ozone or hydroxyl radicals resulting in ring cleavage, then the byproducts may be much
more easily biodegradable than the parent compounds. This process would be applicable for
wastewater treatment in which ozonation could be used as a pretreatment followed by some

form of biological treatment (WPCF, 1986; Holzer et al., 1991; Gilbert, 1988).

Ozone is a very strong oxidant, more powerful than other oxidants commonly used
in water and wastewater treatment. Numerous applications exist for ozonation in the area
of wastewater treatment, including the processing of shale oil (Masten and Davies, 1994).
While in the United States, ozone is primarily used in water treatment for disinfection and
DPB control, it is also used to control odor, improve suspended solids removal, improve
performance of granular activated carbon units and improve the biodegradability of

wastewater (Rice, 1985; Gasi et al., 1990; Robson and Rice, 1991). Ozone would be

5

highly useful for the remediation of wastewaters and industrial effluents containing
PAHs. For example, ozonation coupled with biodegradation could be used for the
ultimate destruction of PAHs in the enviromnent. Ozonation could be used to break the
ring structure, followed by biological degradation for the subsequent oxidation of the
ozonation byproducts. Therefore, an investigation, especially in terms of the
determination of aqueous phase rate constants, reaction byproducts and the effect of

competing substrates, is highly pertinent to the treatment of wastewaters.

Preliminary work on the use of gaseous ozone for the decomposition of PAHs in
soils has revealed very promising results (Yao and Masten, 1992). It has been shown that
ozone is capable of degrading several PAHs (including pyrene, chrysene and phenanthrene)

in the soil; however, it is still unclear what degradation products are produced.

Ozone reacts with olefinic compounds by a 1,3-dipolar cyclic addition across the
double bond to produce an trioxalane (also called a molozonide), which rapidly decomposes
to form aldehydes and organic acids (as shown in Figure 1.1). When ozone reacts with
aromatic compounds, three types of ozone attack can occur, resulting in either substitution
or ring cleavage. These are: 1) 1,3-dipolar cycloaddition (ozonolysis) at the bond or bonds
having the most double bond character (the bond with the lowest bond-localization energy);
2) electrophilic ozone attack on the atom with the lowest atom-localization energy; 3)

conjugate addition where there is a reactive diene system (a system of lowest para-

trioxalane

l

.30:
u "

R
\
C + /C=O
/ \ H
H R
carbonyl oxide zwitterion aldehyde
A20 \
O H
O—O
R-f—OH Rx x:
H O
H
hydroxyhydroperoxide criegee ozonide
or H202 .
R
\ R
/C==0 + \C=O
HO H/
carboxylic acid aldehyde

Figure 1.1 Mechanism of the Ozonation Reaction
(Adapted from Bailey, 1972)

localization energy).

When ozone reacts with PAHs, which type of attack ozone will occur depends on
where the PAHs has the lowest-localization energy. For example, ring cleavage occurred
when phenanthrene reacted with ozone (Bailey, 1982; Meineke and Klamberg, 1978). Gas
phase ozonolysis of aromatic compounds resulted in the formation of CH3COCHO as the
major product with small amounts of carbon monoxide, carbon dioxide and formic acid
(Atkinson, 1984). The mechanism for the formation of these products is not known. The gas
phase reaction of benz[a]pyrene and ozone has also been studied (van Cauwenberghe et al.,
1979; Finlayson-Pitts and Pitts, 1988). Such substitution reactions of benz[a]pyrene resulted
in the formation of quinones and epoxides, whereas ring cleavage reactions resulted in the
formation of dialdehydes, benzanthronedicarboxylic acid and other dicarboxylic acid

compounds. The mechanisms of these reactions is not fully understood.

In water, ozone degrades to form hydroxyl radicals (as shown in Figure 1.2). The
rate of degradation depends on pH, and on the concentrations, types of chemicals and ions
present. Hydroxyl radicals (OH), being less selective and a stronger oxidant than ozone,
will oxidize many compounds that do not or only very slowly react with ozone. The
mechanisms of these reactions are complex and the oxidations may occur via three
pathways: hydrogen abstraction from a -CH2 group, radical addition, and electron transfer.

The secondary radicals formed during these reactions can react with additional ozone, with

 

 

 

 

 

 

 

 

 

 

 

 

 

«are A
' Cor-gm;

.0;

 

 

Hog.

 

 

 

02
LEGEND

M - Micropollumnt (can also ac: as an initiator. scavenger, or promoter)
M “a - Oxid'md form of micropollutant
P - Promoter _
P oxid. - Oxidized form of promoter
1 - Initiator
I" - Protonated form of initiator formed by electron transfer reacn'on
S - Scavenger

S a.“ - Oxidized form of scavenger

 

Figure 1.2 Reactions of Aqueous Ozone in the Presence of Micropollutant M
(Adapted from Staehelin and Hoigné, 1985)

 

9

one another, or with the target compound(s). Organoperoxides can be formed by the
reaction of the carbonyl (C°) radical with any oxygen present. These compounds can enter
into a chain reaction involving ozone and the hydroxyl radicals resulting in the formation of
new radicals or superoxide. Other compounds can react with the free radicals produced to

terminate the chain reaction. (Staehelin and Hoigné, 1985).

The ozonation of PAHs in aqueous solution has been studied by a number of
researchers. Ozone reacts with PAHs with rate constants on the order of >103 M'l s". The
rate constants for the reaction of selected PAHs with ozone and half-lives are given in Table
1.2. Radding et a1. (1976) determined the rate constants for the reaction of ozone and
several PAHs in water at 25 °C. Hoigné and Bader (1983) determined the rate constant for
the ozonation of naphthalene in water. Their experiments were conducted at 20 °C, pH 2
and in water containing 1 mM tert-butyl alcohol. They assumed that the reaction order with
respect to ozone concentration is one, and found that the reaction is first order with respect
to organic compounds. Hoigne and Bader state that the stoichiometric coefficients usually
range flour 1 to 5 and calculated the rate constant by assuming a second order reaction and a
stoichiometric coefficient equal 2.5. Contrastingly, Legube et al. (1986b) determined the
rate constant for the reaction of ozone with naphthalene in organic solvent at 1 °C, pH 5.6.
They found that the stoichiometric coefficient for the initial reaction step is 2, and that the
reaction is second order. Other researchers assumed that the reaction order and stoichiome-

try were identical to that assumed by Hoigné and Bader (Butkovié et al., 1983; Corless et

10

Table 1.2 Rate constants for the reaction of ozone and various PAHs in water

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Compound Rate constant, k‘ Calculated t,,2 b Ref.
(M"S") (860)
Naphthalene 3,000 2.3 1
550 12.5 2
Pyrene 170 41 3
39,000 0.18 4
Benz[a]pyrene 110 63 3
- 6,200 1.1 4
Benz[a]anthracene 260 27 3
Dimethylbenzanthracene > 680 < 10 3
Dibenzanthracene 280 25 3
Fluorene fast completely 5
destroyed
Phenantlrrene rapid oxidation of center ring, then very 5
slow oxidation

15,000 0.46 4
Anthracene fast 5
Acenaphthene fast 5
Acenaphthylene fast 5
3,4-Benzofluoranthene intermediate 5
1 1,12-Benzofluoranthene intermediate 5

 

 

 

 

 

' Tested at pH = 1~7.

" Calculated fiom t.,, = 0.69/(k [53), where [5;] = 10* M.

 

Ref.: 1. Hoigné and Bader, 1983; 2. Legube et al., 1986; 3. Radding et al., 1976;
4. Butkovic et al., 1983; 5. WPCF, 1986.

 

11

al., 1990; Kuo, 1985). However, those researchers did not confirm the reaction order and

stoichiometry.

Pryor et al. (1983) found that in solutions of dichloromethane (a non-participating
solvent), linear free-energy relationships between the rate constants and the calculated
orbital parameters of the nine unsubstituted PAHs studied could be obtained. They also
found that electrophilic attack by ozone to yield a sigma-complex was the presumed
rate-limiting step in the oxidation of these compounds. Although Lee and Hunter (1985)
did not determine reaction rate constants, they did find that greater than 99%
decomposition of naphthalene, fluoranthene, benz[b]tluoranthene, benz[k]pyrene, and
benz[a]fluoranthene could be achieved when aqueous solutions of these compounds were
oxidized using an ozone dosage of 21 .3 mg/L and one hour contact time. The most
recalcitrant of the compounds studied was benz[a]anthracene; 54% of this compound was
degraded at the conditions given above. In the WPCF Manual of Practice No. FD-ll
(1986) relative rate constants are listed for a number of PAHs. Though no numerical
values for the rate constants are presented, qualitative indication of the reaction rates are

given.

Several experiments have been conducted to identify ozonation products of a
number of PAHs. The products identified are listed in Table 1.3. Dreher and Klamberg

(1988) studied the degradation of PAHs in dilute aqueous systems and identified the

12

Table 1.3 Byproducts formed during the ozonation of various PAHs

 

 

 

 

 

 

 

 

 

 

 

 

 

Compound Products Ref.
oxidized

Naphthalene orthophthaldialdehyde; phthalaldehydic, phthalic, oxalic, 1,6
formic, oxomalonic acids, hydrogen peroxide, a cyclic
peroxide
cis and trans-Z-formylcinnamaldehyde, 2
1,4-naphthoquinone, phthalic acid

Phenanthrene diphenic acid fiom center ring oxidation, then carbon 3;6
dioxide and oxalic acid; biphenyl carboxylic acid,
phthalic acid

Fluorene (F L) oxalic acid, F L-oxide, mono- and de-hydroxyl-FL, 3;7
7-hydroxy-FL and 7-hydroxy-F L oxide

Anthracene anthraquinone, phthalic acid 3;6

Naphthacene Naphthaquinone, anthraquinone, phthalic acid 6

Acenaphthene 7-formyl-l-indanone, 7 -hydroxy-1-indanone, 1- 4;7
indanone, l-indanone—7-carboxylic acid, indan-1,7-
dicarboxylic acid

Acenaphthylene 1,8-naphthalenedialdehyde, l-naphthoic acid, 1,2-epoxy- 4
acenaphthalene, naphthaldehydic acid

Fluoranthene FL-quinone, FL-acid 7

Pyrene short chain aliphatic compounds A 5

Chrysene biphenyl carboxylic acid, phthalic acid 6

 

 

 

Ref; 1. Legube et al., 1986; 2. Marley et al., 1987; 3. WPCF, 1986; 4. Kuo, 1985;

5. Corless, 1990; 6. Meineke and Klamberg, 1978; 7. Dreher and Klamberg, 1988.

 

13

major byproducts formed from the ozonation of these compounds. The ozonation of
fluorene, acenaphthene and fluoranthene resulted in ring cleavage and a number of
carboxylated benzenes, quinones and aliphatic compounds. Meineke and Klamberg
(1978) studied the ozonation of other PAHs in dilute aqueous systems. They found that
the main products of compounds, such as anthracene and naphthacene, are quinone or
quinone substituted with carboxylic acids. For the other compounds investigated, e.g.,
phenanthrene and chrysene, the final products were biphenyl carboxylic acid and phthalic
acid. Other researchers (NAS, 1972; Neff, 1979) found that the reaction of PAHs with
ozone could result in the cleavage of phenanthrene-like bonds producing various diacids,
quinones (by the oxidation of anthracene at the 9,10.like position), diones, and the

oxidation of alkyl side-chains.

The products formed from the reaction of ozone with low molecular weight
hydrocarbons (e.g., from the ozonation of naphthalene and substituted benzenes) have been
successfully identified (Legube et al., 1983, 1986). Legube et al. (1986a) extensively
studied the ozonation of naphthalene in aqueous solution and proposed three reaction
pathways for the initial attack of ozone on naphthalene. These pathways are: (1) one
electrophilic substitution of ozone on carbon 1 (or 2) of the naphthalene molecule, (2)
two simultaneous 1,3-dipolar cyclic additions on both the 1,2 and 3,4 bonds of the
naphthalene molecule, and (3) one 1,3-dipolar cyclic addition of ozone on the 1,2 bond of

the naphthalene molecule.

14

Several mechanisms for the reactions of ozone with PAHs have been proposed, but
have not been confirmed. The major focus of the research that has been conducted on the
ozonation of PAHs has been to determine the kinetics of the reaction of ozone with these
compounds in aqueous solutions (e.g., Hoigné and Bader, 1983; Staehelin and Hoigne’,
1985). While some work has focused on investigating the products formed from the higher
molecular weight compounds, because of the low aqueous solubility of PAHs much of this
work has also been accomplished in organic solvents. However, as water acts as a
participating solvent, different reaction products would be expected in water as compared to
organic solvents. As a result, our knowledge of the mechanisms by which ozone reacts

with PAHs in either natural waters or soils is severely lacking.

Although PAHs ozonation products may be more easily biodegradable than the
parent compounds, little research has been done to evaluate the toxicity of the ozonation
products. Some researchers (Pearson, 1982; Pelkonen and Nebert, 1982; Aryton et al.,
1990) have shown that oxidation products of PAHs, such as hydroxylated and epoxidized
forms of benz[a]pyrene which are produced biologically, are more carcinogenic form than
their parent compounds. Il’nitskii et a1. (1968) monitor the carcinogenicity of PAHs during
ozonation. They used 60 times the ozone : PAH ratio and a treatment time of 2.5 min; they
found the carcinogenic hydrocarbons were inactivated. However, Marley et a1. (1987) found
some of the ozonation products were shown to have significant toxicity. They found

naphthalene ozonation products, such as o-phthalaldehyde, were shown to inhibit yeast

15

growth or were mutagenic in the Ames test. Upham et al. (1994) also found that the

ozonation of PAHs resulted in the formation of initial byproducts that were more toxic than

the parent compound. Therefore, an investigation of toxicity of the ozonation products

needs to be addressed.

1.4 Relearch Objective

The major focus of this research is to investigate the pathways by which ozone

reacts with PAHs in an aqueous system. This is central for understanding and predicting the

role of ozonation in degrading PAHs in waters and other environment systems. To achieve

such goal the following questions need to be answered:

1.

2.

Are the ozonation products more toxic?

Are the mechanisms by which the three different classes of PAHs (phenanthrene-,
pyrene- and anthracene-types) react with ozone different? If so, are the byproducts
resulting fiom each of these types of reactions different?

Can a better understanding of the reaction pathways be used to predict which
products will be formed fi'om PAHs based upon their classification?

How does the presence of other reactive species affect the rates and pathways of the

reaction between the targeted PAH and ozone?

The following specific aims are designed in an attempt to answer the above questions:

1.

To determine the kinetics of the reaction of ozone with pyrene, chrysene and

benz[a]anthracene in the aqueous phase.

16

2. To assess the toxicity of ozonation products.

3. To identify the byproducts formed from the reaction of ozone with each of the target
compounds.

4. To propose and verify a reaction pathway for the primary reaction of ozone with the
target PAHs.

5. To investigate the effect of other reactive species on the pathway of the reaction of

ozone with the target compounds.

1.5 Organhation of Thesis

The objective of Chapter 2 was to investigate the toxicity of the ozonation
products formed from pyrene. A non-genotoxic bioassay that monitors gap junction
intercellular communication (GJIC) was used to assess the toxicity after the ozonation of
pyrene. The results show the importance of using a biological assay to assess the amount
of ozone required to reduce toxicity. The results also suggested that identifying the
ozonation products is necessary. By identifying these ozonation products and the reaction
pathways in aqueous solutions, treatment alternatives that would minimize the formation
of toxic products can be developed. Observations described in this chapter show the
feasibility of using ozone to remove pyrene and its toxic byproducts, as long as the ozone

dosage was greater than 4.5 mol of ozone/mol of pyrene.

In Chapters 3 and 4, the ozonation pathway of pyrene and benz[a]anthracene are

17

investigated. The ozone consumption data obtained for the initial reaction with the target
compound along with the identification of byproducts from the initial attack on the PAH
molecule by ozone was used to propose the ozonation pathway. The ozone dosages were
varied to result in different extents of reaction. The concentrations of target compounds
and their ozonation products were quantified upon different extents of reaction. The
ozonation products which were generated were identified. The effect of free radicals on the
oxidation of the target compounds were also studied. The oxidation products were
identified to determine if the presence of other reactive species affected the reaction
pathway. In Chapter 3, the ozonation of pyrene was observed to occur by a stepwise ring
cleavage, and in the oxidation of aldehydic moiety or functional group. Fourteen products
including aldehyde and carboxylic acid substituted phenanthrene- and biphenyl- type
oxidation products were identified. Two pairs of isomers were identified. In Chapter 4, the
results of studies investigating the reaction of ozone with benz[a]anthracene are reported.
Fifteen products including seven pairs of isomers were identified. Ozonation by bond
attack (resulted in formation of phenyl-naphthyl type products) and atom attack (resulted

in formation of quinone type products) occurred simultaneously.

In conclusion, a detailed discussion of ozonation pathways and an evaluation of the
predictability of ozonation for the degradation of PAHs in aqueous systems is provided in
Chapter 5. Bond- and atom-localization energies were used to predict the pathways by

which PAI-Is would react with ozone. In comparing our results with previously published

18

researches, most of which were performed in organic solvents, the importance of the 'OH
radical in aqueous reactions is emphasized. Based on these conclusions, the pathways by

which other similar type of PAHs would react with ozone have been predicted.

In Appendix A, a kinetic study for the reaction of ozone with pyrene is reported.
The reaction order of ozone with pyrene and the stoichiometric coefficient was investigated.
Confirming these parameters would allow for a higher degree of confidence in predicting
the rates of reaction of PAHs with ozone in other complex systems. An assumption of
ozone direct reaction with target compound was made. Therefore, a second-order reaction
was observed. However, the results observed show that, due to the extremely rapid reaction
of ozone with PAHs, other reactive species were involved in the reaction. As such, the
assumption that this reaction is overall second order reaction could not applied. In order to
obtain the kinetic data for the reaction of ozone with only the parent compound, a stopped

flow system should be used.

19

1.6 References

Andreozzi, R., A. Insola, V. Caprio, and MG. D'Amore. 1992. "Quinoline Ozonation in
Aqueous Solution". Wat. Res., 26(5):639-643.

Atkinson, R. and E. Carter (1984) "Kinetics and Mechanisms of the Gas-phase Reactions of
Ozone with Organic Compounds under Atmospheric Conditions". Chem. Rev. 84:
43 7-470.

Bader, H. and J. Hoigné. 1982. "Determination of Ozone in Water By The Indigo Method -
A Submitted Standard Method". Ozone: Sci. and Eng, 4: 169-176.

Bailey, PS. 1972. "Organic Groupings Reactive Towards Ozone - Mechanisms In
Aqueous Media". In: Ozone In Water And Wastewater Treatment. Evans, F.L. Ed..
Ann Arbor, MI: Ann Arbor Science. pp. 2959.

Bailey, PS. 1982. "Ozonation of Aromatic Compounds: Bond Attack on Benz-Fused
Carbocyclics". In: Ozonation In Organic Chemisg’y. Wassennan H.H. Ed.,
Academic Press, New York. 39(2): 43-76.

Butkovié, V., L. Klainc, M. Orhanovic, J. Ttu'k, H. Gusten. 1983. "Reaction rates of
polycyclic aromatic hydrocarbons with ozone in water". Environ. Sci. T echnol.
17(9): 546-548.

Calvalieri, E.L., E.G. Rogan, and W.G. Thilly. 1980. "Carcinogenicity, Mutagenicity And
Binding Studies Of The Environmental Contaminant Cyclopenteno[c,d]pyrene And
Some Of Its Derivatives". In: Chemical Analysis and Biological Fate: Polyguclear
Aromatic Hydrocarbons. Cooke M. and A.J. Dennis Eds., Battelle Press, Columbus,
Ohio, pp. 199-208.

Corless, C.E., G.L. Reynolds, N.J.D. Graham, and R Perry. 1990. "Ozonation of Pyrene in
Aqueous Solution". Wat. Res., 24(9): 1119-1123.

Dreher, W. and H. Klamberg. 1988. "Zum Abbau von polycyclischen aromatischen
kohlenwasserstoffen III. Reaktionsprodukte der umsetzung von weiteren
polycyclischen aromatischen kohlenwasserstoffen in wasser mit ozon". Fresenius Z
Anal. Chem. 331: 290-294.

Eadie, B.J.. 1983. "Distribution of Polycyclic Aromatic Hydrocarbons in The Great Lakes".
Toxic Contaminants in The Great Lges. Nriagu, J .O. and MS. Simmons Eds.. John
Wiley and Sons, New York. pp. 195-211.

20

EPA. 1984. “Guidelines Estalishing Test Procedures for the Analysis of Pollutants Under
the Clean Water Act; Final Rule and Interim Final Rule and Proposed Rule”. 40
CFR Part 136, US. EPA.

Finlayson-Pitts, B.J. and Pitts, Jr., J.N. 1986. Atrno 'c Chemi : Fundamentals and
Exm’ gntal Techniques. John Wiley and Sons, New York p. 938.

Futoma, D.J., S.R. Smith, T.E. Smith, J. Tanaka. 1981. "Solubility Studies of PAH in
Water". Polvcvclic Aromatic Hydrocarbons in Water Systems. Boca Raton, F 13.:
CRC Press. pp. 1-11.

Gasi, T.M.T., L.A.V. Amara], C.E.M. Pacheco, A.G. Filho, A.D. Garcia, Jr., S.M.M.
Vieira, R. Francisco, Jr., P.D. Orth, M. Scoparo, M.S.R. de S. Dias, and ML.
Magri. 1990. "Ozone Application for the Improvement of UASB Reactor Effluent -
I. Physical-Chemical and Biological Appraisal". Ozone Sci. Eng. 13(2): 179-193.

Gilbert, E.. 1988. "Biodegradability of Ozonation Products as a Function of COD and
DOC Elimination by the Example of Humic Acids". Wat. Res., 22(1): 123-126.

Griest, W. H., B.A. Tomkins, M.V. Buchanan, A.R. Jones, and RR. Reagan. 1987. "Level
of Polycyclic Aromatic Hydrocarbon Denna] Tumorigens In Various Liquid F uel".
Fuel 66: 1046-1049.

Hallett, DJ. and R.W. Brecher. 1983. "Cycling of Polynuclear Aromatic Hydrocarbons in
The Great Lakes Ecosystem". Toxic Contaminants in Th3 Great Lakes. Nriagu,
J.O. and MS. Simmons Eds.. John Wiley and Sons, New York. pp. 213-237.

Harrison, RM., R. Perry, and RA. Wellings. 1975. "Review Paper: Polynuclear Aromatic
Hydrocarbons in Raw, Potable and Waste Waters". Wat. Res., 9:331-346.

Hoigné, J. and H. Bader. 1983. "Rate Constants of Direct Reactions of Ozone with Organic
and Inorganic Compounds in Water-I. Non-dissociating Compounds". Wat. Res. 17:
993-1004.

Holzer, K. and J.F. Lawson. 1991. "LOPROX - A Flexible Way to Pretreat Poorly
Biodegradable Effluen ". Proc. of the 46th Annual Purdue University Industrial
Waste Conference, May 10-12, 1991.

Howard, P.H., R.S. Boethling, W.F. Jarvis, W.H. Meylan, E.M. Michalenko. 1991.

Handbook of Environmental Deggrfltion Rates Heather Taub Printup, Ed. Lewis
Publishers, Inc. Chelsea, MI.

21

Il'nitskii, A.P., A. Ya. Khesina, S.N. Cherkinskii, and L.M. Shabad. 1968. "Effect of
Ozonation upon Aromatic Hydrocarbons, Including Carcinogens". Hyg. Sanit. ,
33:323-327.

Kuo, CH. 1985. "Reactions of Ozone with Organics in Aqueous Solutions". US.
Environmental Protection Agency. Report No. EPA/600/3-85/031. 66 pp.

Lee, YS and J .V. Hunter. 1985. "Effect of Ozonation and Chlorination on Environmental
Protection Agency Priority Pollutan ". In: Water Chlorination: Chemigy,
Environmental [mat and Health Effects. Vol. 5. Chelsea, MI: Lewis Publishers.
1515-1526.

Legube, B., S. Guyon, H. Sugirnitsu, and M. Dore. 1983. "Ozonation of Some Aromatic
Compounds in Aqueous Solution: Styrene, Benzaldehyde, Naphthalene,
Diethylphthalate, Ethyl and Chloro Benzenes”. Ozone Sci. Engng. 5:151-170.

Legube, B., S. Guyon, H. Sugimitsu, and M. Dore. 1986a. "Ozonation of Naphthalene in
Aqueous Solution - 1. Ozone Consumption and Ozonation Products". Wat. Res.
20(2): 197-208.

Legube, B., S. Guyon, H. Sugimitsu, and M. Dore. 1986b. "Ozonation of Naphthalene in
Aqueous Solution - II. Kinetic Studies of the Initial Reaction Step". Wat. Res. 20(2):
197-208.

Levenspiel, O.. 1972. In: Ch;err_1ical Reaction Engine_e_r_rr1° g, 2nd Ed. John Wiley & Sons,
Inc., NY. pp. 47-50.

Marley, K.A., R.A. Larson, P.L. Stapleton, W.J. Garrison, C.M. Klodnycky. 1987.
"Ozonolysis of Naphthalene Derivatives in Water and in Kerosene Films". Ozone:
Sci. and Eng. 9(1): 23-36.

Masten, SJ. and S.H.R. Davies. 1994. "The Use of Ozonation to Degrade Organic
Contaminants in Wastewa ". Environ Sci. Technol., 28(4): 180A-185A.

Meineke, I. and H. Klarnberg. 1978. "Zum Abbau von polycyclischen aromatischen
kohlenwasserstoffen I. Reaktionsprodukte der Umsetzung von polycyclischen
aromatischen Kohlenwasserstoffen in Wasser mit Ozon". F resenius Z Anal. Chem.
293: 201-204.

National Academy of Sciences. 1972. Particulate Polycyclic Organic Matter. Washington,
DC, NAS: pp. 361.

22

Neff, J.M.. 1979. Polmuclear Aromatic Hydrocarbons In The Aquatic Environment:
Sources, Fates And Biologrc__al_' Effects. London: Applied Science Publishers, Ltd.

Pryor, W.A., G.J. Gleicher, and DP. Church. 1983. Reaction of polycyclic aromatic
hydrocarbons (PAI-D with ozone. Linear flee-energy relationships and tests of likely
rate determining steps using simple molecular orbital correlations. Manuscript.

Radding, 8.8., T. Mill, C.W. Gould, D.H. Liu, H.L. Johnson, D.C. Bomberger, and CV.
Fojo. 1976. The environmental fate of selged mlmuclear aromatic hydrocarbons.
EPA 88560/5-75-009. US. Environmental Protection Agency, Washington, DC.
pp. 122.

Rice, RG. 1985. In: Innovatiom in the Water and Wastewater Fields. Glysson, E.A., D.E.
Swan, and E.J. Way Eds.. Butterworth, Boston, MA, pp. 65-88.

Robson, CM. and RG. Rice. 1991. "Wastewater Ozonation in the USA. - History and
Current Status -1989". Ozone Sci. Eng. 13(1): 23-40.

Sims, RC. and MR. Overcash. 1983. "Fate Of Polynuclear Aromatic Compounds (PAHs)
In Soil-Plant Systems". Residue Reviews 88: 2-68.

Smith, J.D., J. Bagg, and EN. Bycrofi. 1984. "Polycyclic Aromatic Hydrocarbons In The
Clam Tridacna Maxima From The Great Barrier Reef, Australia". Environ Sci.
Technol. 18(5): 352-358.

Solbakken, J.E., A.H. Knap, C.E. Searle, and K.H. Palmork. 1983. "Uptake And
Elimination Of [9-"C]-Phenanthrene In The Turkey Wing Mussel (Arcazebra)".
Bull. Environ. Contam. Toxicol. 30: 420-423.

Staehelin J. and J. Hoigné. 1985. "Decomposition Of Ozone In Water In The Presence Of
Organic Solutes Acting As Promoters And Inhibitors Of Radical Chain Reactions".
Env. Sci. Technol. 19: 1206-1213.

Stevens D.K., W.J. Grenney, and Z. Yan. 1989. "Sensitive Parameter Evaluation For A
Vadose Zone Fate And Transport Model - Project Summary". US. Environmental
Protection Agency Report No. EPA/600/S2-89/03 93.

Thorton S.C., L. Diamond, M. Hite, and W.M. Baird. 1981. "Mutation Induction,
Metabolism And DNA Adduct Formation By Polycyclic Aromatic Hydrocarbon In
L5178Y Mouse Lymphoma Cells: Effects Of S-20 Concentration". In: Chemical

Analysis and Biolggical Fate: Polmuclear Aromatic Hydrocarbons. Cooke M. and
A.J. Dennis Eds., Battelle Press, Columbus, Ohio, pp. 199-208.

23

van Cauwenberghe, K., L. van Vaeck, and J .N. Pitts Jr.. 1979. "Chemical transformations of
organic pollutants during aerosol sampling". Adv. Mass Spectrom. 8: 1499

Verschueren, K.. 1977. "Handbook of Environmental Data on Organic Chemicals". 2"" Ed.,
pp. 393-394.

WPCF Manual of Practice No. FD-ll. 1986. "Removal of Hazardous Wastes in
Wastewater Facilities - Halogenated Organics". Facilities Development, Water
Pollution Control Federation, Alexandria, VA pp. 68-71.

Yao, J.J. and SJ. Masten. 1992. "Use of Ozone for the Oxidation of PAHs in Dry Soils".
Proc. of the 24th Mid-A tlantic Industrial and Hazardous Waste Conference. July
14-17, 1992. Morgantown, WV. pp. 642-651.

Upham, B.L.; SJ. Masten; B.R. Lockwood; J.E. Trosko. 1994. Fundam. Appl. T oxicol.
23: 470-475.

CHAPTER 2
DETERMINATION OF THE EFFICACY OF OZONE TREATMENT SYSTEMS

USING A GAP JUNCTION INTERCELLULAR COMMUNICATION BIOASSAY

ABSTRACT

A non-genotoxic bioassay that monitors gap junction intercellular communication (GJIC)
was used to assess the presence of potential tumor promoters after the ozonation of
pyrene, a polycyclic aromatic hydrocarbon (PAH). Water and ozone were not rate-
limiting reactants in a solution containing 10 % water/90 % acetonitrile (v/v) and 5 mM
pyrene during ozonation. Pyrene (5 mM) was oxidized at various ozone dosages, and the
resultant mixtures were assayed for their effect on GJIC. Approximately 4.5 mol of
ozone/mol of pyrene was required to remove all the intermediate byproducts found to be
inhibitory to GJIC. At 3.6 mol of ozone/mol of pyrene, the mixture, which was Slightly
inhibitory to GJIC, was fractionated by RP-HPLC into 13 components, of which only one
component was inhibitory to GJIC. Just 1.6 mol of ozone/mol of pyrene was needed to
remove > 90 % of the pyrene, which showed that monitoring only the removal of the

parent compound was insufficient for the assessment of the treatment efficacy.

24

25

2.1 Introduction

Environmental pollutants are a major concern from an ecological and human
health perspective. This concern has led to a massive effort to clean up toxic waste dump
sites and contaminated aquifers along with the development of numerous treatment
technologies. The efficacy of a treatment process is most often assessed by monitoring
the contaminant of interest. However, when a treatment process involves chemical
transformations, the detection of the parent compound provides no toxicological
information about the reaction products or the parent compound. Upham et al.(1) has
shown that the ozonation of PAHs resulted in the formation of products more toxic than
the parent compound. Also, numerous researchers (2-4) have shown that oxidation of
simple aromatic compounds such as benzene or 1,3,5-trichlorobenzene can result in the
formation of a very large number of byproducts. The identification and quantification of a
large number of byproducts, such as those formed during chemical or biological
processes is prohibitively expensive and time consuming. An evaluation of chemical and
biological treatment technologies needs to be made using toxicological screening
methods to assess the effects of the byproducts on biological systems. The treatment

process can then be engineered to minimize the toxicological effects of the byproducts.

Cancer is a primary health concern resulting from human exposure to
environmental pollutants. In the past, in vitro assays for mutagenicity (genotoxicity) of

environmental contaminants have been used for the detection of potential carcinogens.

26

Many chemicals appear not to be genotoxic (5 -7) but rather to modify gene expression by
their ability to alter the homeostatic control of multicellular organisms (8). lntercellular
communication is a crucial biological process that maintains the homeostatic health of
multicellular organisms (9, 10). A membrane structure called the gap junction permits the
transfer of low molecular mass ions and molecules (5 1200 Da) between contiguous cells
(11). Amino acids, CAMP, calcium, inositol, and triphosphates are regulatory molecules
known to diffuse through gap junctions (12-14). The gap junction has been linked to
many regulatory roles such as growth control, developmental and differentiation
processes, synchronization, and metabolic regulation (15). Chronic exposure to epigenetic
toxicants that have been implicated in tumor promotion during carcinogenesis (15-1 7), in
teratogenesis (18), in reproductive dysfunction (19-21), and in the alteration of muscle
contractions in the heart and uterus (22, 23) are also known to effect gap junction
intercellular communication (GJIC). Trosko et al. (24) also noted that neurotoxicological
effects could result from the inhibition of intercellular communication in the central
nervous system. A mutation in the connexin 32 gene that encodes for gap junction
proteins has been implicated in the cross-linked Charcot-Marie-Tooth syndrome, which is

a peripheral neuropathy (25).

It is clear that the inhibition of GJIC can result in pathological consequences
including carcinogenesis (16, 26). Thus, the effects of pollutants on the mammalian gap

junction can give invaluable toxicological information. Previous toxicological evaluation

27

of ozone byproducts has focused primarily on genotoxic events (e.g., see 27-29). The
objective of this research was to use a bench-scale, completely-mixed, semi-continuous
reactor to assess the efficacy of the ozonation of aqueous organic compounds to remove

products that are inhibitory to GJIC. Pyrene was used as the target compound.

As PAHs (e.g., pyrene) usually have low water solubilities, one must either
ozonate the chemicals in dilute aqueous solutions and concentrate the products for
bioassays or ozonate the compound in more concentrated organic solutions that can be
then directly used for biological assays. The latter was chosen in this study to avoid

potential losses or contamination of the products during the concentration step.

As water is a participating solvent in the reaction of ozone with unsaturated or
aromatic hydrocarbons (30 and Figure 2.1), the addition of water to an ozone reactor
containing a high concentration of an organic solvent is necessary. The primary products
of ozonating in the presence of water are H202 and compounds that contain carboxylic
acid and aldehyde functional groups (30, Figure 2.1). In the absence of water the primary
product is the Criegee ozonide. It was essential to choose a solvent that was nonreactive
with ozone and miscible with water. Therefore, it was also necessary to establish
conditions wherein water not a rate-limiting reactant to ensure that the ozonation pathway

favored the reaction of the carbonyloxide zwitter ions with water.

28

Figure 2.1 Reaction of ozone with an unsaturated carbon in the presence and
absence of water. The primary products in the presence of water
are H202, an aldehyde, and a carboxylic acid if R2= H.

29

2:55. comer—U

 

_
\o /\ cu: m/
UHO
\1 “w/J— «ESSA—96.3 ~m0\
0.10 408:3»:
+ J— OI: E
/U\ /U.|.O
\ / ..l_
83 NONE .
of
\/
E M: a: J. an m
/ \ /o\ _ _ a E
W + __ T Jul! llUi i0 Mm All /U|IU\
o 30 e\ /o max .I /.
O\:2.5:.BN /O\ K m

3 258 355.50 wee—382E m0

30

2.2 Methods
2.2.1 Cell Culture

WB-F344 rat epithelial cell lines were obtained from Drs. J. W. Grisham and M.
S. Tsao of the University of North Carolina (Chapel Hill, NC). Cells were cultured in 2
mL of D medium (Formula No. 78-5470EG, GIBCO Laboratories, Grand Island, NY),
supplemented with 5% fetal bovine serum (GIBCO Laboratories, Grand Island, NY) and
50 mg/mL gentamicin (Quality Biological, Inc., Gaithersburg, MD). The cells were
incubated at 37°C in a humidified atmosphere containing 5% CO2 and 95% air. The cells
were grown in 35 mm2 plastic petri dishes and the culture medium was changed every
other day. Bioassays were conducted with confluent cultures that were obtained after 2-3

days of growth.

2.2.2 Bioassay of GJIC

The scrape loading/dye transfer (SL/DT) technique was adapted after the method
of El-Fouly et al. (31). The cells were incubated with 20 pL of the target compounds for
60 min unless indicated otherwise. A primary product of the reaction of ozone with
aromatic hydrocarbons and alkenes is hydrogen peroxide (30, 32); therefore, hydrogen
peroxide was removed from the growth medium by the addition of catalase (1000
Units/mL; Sigma Chemical Co., St. Louis, M0) to cell cultures immediately preceding
the addition of the ozonated solutions. Also, GJIC assays were conducted at noncytotoxic
levels of the samples as determined by the neutral red uptake assay kit (Sigma Chemical

Co., St Louis, MO). Following incubation, the cells were washed with 3 mL of phosphate

31

buffered saline (PBS). Lucifer yellow was added to the washed cells and three scrapes
were made with a surgical steel-blade scalpel under low light intensities. The purpose of
the three scrapes was to ensure that the scrape traversed a large group of confluent cells.
After a 3-minute incubation period the cells were washed with 10 mL of PBS and then
fixed with 2 mL of a 5 % formalin solution. The distance the dye traveled from the scrape
was measured from photographs of microscopic images (magnification 200x) made of
the fixed cells using a Nikkon Diaphot-TMD epifluorescence phase-contrast microscope

illuminated with an Osrarn HBO 200W lamp and equipped with a 35-mm FA camera

(Nikkon, Japan).

The distance the dye migrated was measured from the cell layer at the scrape to
the edge of the dye front that was visually detectable. Due to the slight irregularity of the
dye fiont's edge, the distance along one scrape per dish was measured every 1 cm for a
total of 8 cm (at 200x, 1 em = 50 mm). The average cell length was 25 mm; therefore, the
l-cm measurement on the photograph was equivalent to two cells. Each scrape chosen for
photographic analysis was from a group of cells that were homogeneous in cell
morphology and confluency. The nine measurements of distance were averaged and
reported as the distance the dye traveled for the chosen group of cells representing the cell
population of one plate. The photographs were taken and developed at the same time.
Three plates per treatment were measured as described, and the values were reported as

an average :1: lSD. An analysis of variance (AN OVA) was used for a test of significance

32

for the results. A Dunnett's multimean range t-Test was used for a test of significance

between each treatment and the control (33).

2.2.3 Ozonation

Ozone was generated in dried oxygen by electric discharge using a Polymetrics
Model T-408 ozone generator (San Jose, CA). Ozone was bubbled into a 100 mL solution
of 5 mM pyrene (99% purity, Sigma Chemical Co., St. Louis, MO), dissolved in a
acetonitrile/water solvent mixture. Acetonitrile was chosen as the organic solvent because
it is miscible with water, has little effect on GJIC (1), is immediately amenable to RP-
HPLC analysis, and has a low reactivity with ozone with a half life of 18 years (34). The
acetonitrile (99.8%, purity) was purchased from EM Science (Gibbstown, NJ). The pH of
the reactor solution was approximately 2. The above reactions were carried out in a 125
mL gas washing bottle. The ozonated solutions were continually mixed with a magnetic
stirrer and stir bar. Effluent gaseous ozone was trapped in 2% (w/v) K1 in water. The flow
of ozone was regulated at 100 mL/min with a Sidetrack flow controller (Sierra
Instruments Inc, Monteray, CA). The i.d. of the tubing was 1/8". All the tubing,
connectors and valves were constructed of Teflon®. The concentration of ozone in the
influent and effluent gas stream was measured spectrophotometrically at 258 nm using a
UV-vis spectrophotometer (Model 1201, Shimadzu Scientific Instruments, Japan). The
molar absorptivity coefficient for ozone is 3000 M" cm" (35). The flow cells were quartz

cuvettes with a path length of 0.2 cm. Reactions were terminated by flushing the reactor

33

with helium for 3-4 min to remove detectable levels of ozone. Quenching agents were not
used to remove ozone so as to avoid the formation of radical species and the overloading
of the RP-HPLC columns and to simplify the evaluation of the biological effects of the

ozonated mixtures.

2.2.4 HPLC Analysis

A Gilson HPLC unit (Worthington, OH) equipped with two pumps (Model 303),
manometric module (Model 802b), dynamic mixer (Model 811), UV detector (Model
116), samme injector (Model 231) and diluter (Model 401), and a 5mm 4.6 x 250 mm C-
18 Partisphere RP column (Whatrnan, Clifton, NJ ) was used for the chromatographic
separations. The eluent was monitored at two wavelengths (210 and 240 nm). The parent
compound was separated from the oxidized products by two different linear gradient
systems. Linear gradient system A consisted of 50%:50% (V :V) acetonitrile/water at the
time of injection and increased to 95%:5% acetonitrile/water 15 min after injection. The
mobile phase was held at this composition for an additional 4 min and then decreased
back to 50%:50% acetonitrile/water mixture giving a total run time of 20 min. Linear
gradient system B consisted of 15%:85% (vzv) acetonitrile/water at the time of injection
and increased to 25%:75% acetonitrile/water 15 min after injection. The mobile phase
was held at this composition for an additional 4 min and then decreased back to a
15%:85% acetonitrile/water mixture giving a total run time of 20 min. Linear gradient

system A separated the less polar components found at the lower ozone dosages while

34

system B separated the polar components found at the higher ozone dosages. Fraction
collection was conducted manually. Each fraction was freeze-dried and reconstituted in

90% acetonitrile at a volume equivalent to the injection volume.

2.3 Results

Pyrene (500 mM) was ozonated in an acetonitrile solvent containing various
concentrations of water ranging from 5.6 (10%) to 33 M (60%). The ozonation
byproducts of these reactions were collected after the reactor reached reaction-lirnited
conditions and separated using RP-HPLC (Figure 2.2A). The chromatographic profiles
obtained for each concentration of water had essentially the same fingerprint. Various
concentrations of pyrene ranging from 0.5 to 5.0 mM were ozonated in solutions
containing 90 % acetonitrile/ 10% water (v/v). The ozonated mixtures were diluted by a
factor determined fi'orn the initial concentration of pyrene in the solution relative to the
lowest initial concentration of pyrene (0.5 mM); e. g., the solution containing an initial
concentration of 5.0 mM was diluted lO-fold and separated by RP-HPLC (Figure 2.2B).

These chromatographic profiles were also the same relative to retention times.

35

 

 

 

 

  

 

 
 

+1
.2
.93
a)
:1:
x
(U
o
a.
'U
2
(U
o o
to i
i
50
25

 

 

 

 

Retention Time (min)

Figure 2.2 (A) Chromatographic separation by RP-HPLC of ozonated pyrene that was
dissolved in various concentrations of acetonitrile/water mixtures. The solid,
dashed, and dotted lines were mixtures containing 10%, 25%, and 60%
water, respectively. This separation used linear gradient system B.
(B) Chromatographic separation by RP-HPLC of various concentrations of
pyrene ozonated in a 90%/10%, acetonitrile/water mixtures. The solid
dashed, and dotted lines were mixtures containing initial concentrations of
5.0, 2.5, and 1.0 mM pyrene, respectively. This separation used linear

gradient system B.

36

The ozone demand, which includes the actual chemical demand and the
dissolution of ozone into solution, exerted during the reaction of ozone with pyrene is
shown in Figure 2.3A. Figure 2.3B is an example of the raw data obtained during the
reaction of ozone with 0.1 mM pyrene. Ozone demand was calculated as the difference
between influent and effluent gaseous ozone concentrations as a function of time. Using
gas flow rates, the change in concentration of ozone was converted to a change in mass.
The ozone demand of the reactor in the absence of pyrene is represented by the curve
with the closed squares which showed that the ozone demand rapidly approached zero.
The addition of pyrene increased the ozone demand of the system. The amount of ozone
consumed remained constant and at a high level for a longer period at higher

concentrations of pyrene.

After the reaction involving pyrene with ozone was terminated, samples were
collected for HPLC analysis. Ozonation reactions were terminated by flushing the system
with helium, which resulted in the rapid removal of ozone fiom the system (Figure 2.33).
A three dimensional representation of chromatographic profiles with increasing contact
time of ozone with pyrene is shown in Figrne 2.4. As the contact time with ozone was
increased the products tended to elute from the column at shorter retention times (i.e.,

they are more polar).

37

 

(A)

 

 

 

 

 

 

 

O
‘1’

0 500 1000 1500
TIME (sec)

 

 

 

(B)

 

 

 

 

 

 

 

0 200 400 600 800 1000
TIME (sec)

 

 

 

Figure 2.3 (A)Ozone demand of the reactor for various concentrations of pyrene
(B)An example of the raw data collected during the reaction of ozone
with 0.1 mM pyrene.

38

Figure 2.4 3-D plot of RP-HPLC chromatograms of 5 mM pyrene reacted with ozone
for different ozone contact time intervals. The average flow rate was 1.25
mmol of ozone/min. This separation used linear gradient system A.

39

o:o~O Co $355.

25.3

25.2.

25.8

25.25
25.2:

Ema: 28¢

25.52
25.5:
25.52

    

cg? .
. teetrfi
.... ., gnwcmvww
. :3 . . {at .1. st
. “New Rangersweavtme
. ...uxs.fifitréshexammwkncg$mw

[.gfiqfi. .

            

       

     
                 

       
 

25.2

25.3

25.8
25.5
25.2:
25.52
25.5:

25.2:

40

Pyrene was ozonated at various dosages ranging from 0 to 4.4 mmol of
ozone/mmol of pyrene. The reactor reached reaction-limited conditions at 4.4 mmol of
ozone/mmol of pyrene. Reaction-lirnited conditions were determined by observing the
time it took for the effluent ozone gas to reach concentrations that remained fairly
constant over time (Figure 2.3B). Ozonated samples (20 mL aliquots) obtained at each of
the ozone dosages were assayed for biological effects on GJIC (Figure 2.5). These
samples were not cytotoxic as compared to the control and the distance the dye traveled
had a maximum standard deviation for toxicity was equivalent to 10% of the control.
Initially, inhibition of GJIC did not begin to decreased until 50% of the pyrene was gone.
The ozonated pyrene mixture remained inhibitory until approximately 2.2 moles of
ozone was added to the reactor. This was also when the reactor reached reaction-limited

conditions.

Samples were collected at an ozone dosage of 1.8 mmol and separated on RP-
I-IPLC into 13 different components (Figure 2.6). The effect of each fraction on GJIC was

determined (Table 2.1). It appeared that only peak 13 was inhibitory to GJIC.

41

 

 

 

A 10 0.5

E a

*" 08 - . 04
e 5 ' 3
.50 g
32 0.6 r . . 0.3 g-
..—. H O -5
E t... a . "d
0 g 0.4 - - 0.2 .g
>53 (to :3
Q *5 . E;

E 0.2 "' . " 0.1

La. I

0.0 ‘ I II I 0.0

 

0.5 1.0 1.5 2.0 2.5
Ozone Dosage (mmol)

P
c

Figure 2.5 Determination of the ozone dosage required for the removal of pyrene and the
oxidation byproducts from the solution that were inhibitory to GJIC.
E] = the concentration of pyrene, which was an average of two replicates,
with a detection limit of 500 nM as determined using RP-HPLC.
O = the inhibition of GJIC, which was an average i 1 SD of the distance the
dye traveled from three different plates per experiment.

42

 

 

 

 

 

 

100 11
E
or 75-
05 4
:r:
x
8 50-
Q
3 3 9 13
E
0 25' 12
a: 12 5 1o
1 l l
00 5 10 15

Retention Time (min)

20

Figure 2.6 Representative chromatogram of the by-product peaks collected for analysis

of GJIC activity shown in Table 2.1. This separation used linear gradient

system B.

43

Table 2.1 Effect of ozonated byproducts of pyrene on GJIC

 

Compound added to cells Fraction of control‘I
Control (acetonitrile) 1.00 :t 0.05
Pyrene (unozonated) 0.31 :t 0.10b
Ozonated pyrene, HPLC peak 1 0.94 i 0.06
Ozonated pyrene, HPLC peak 2 0.99 :t 0.06
Ozonated pyrene,HPLCpeak3 1.16:1:0.10
Ozonated pyrene, HPLC peak 4 0.75 i 0.22
Ozonated pyrene, HPLC peak 5 0.98 :t 0.05
Ozonated pyrene, HPLC peak 6 1.05 1: 0.17
Ozonated pyrene, HPLC peak 7 1.09 d: 0.07
Ozonated pyrene, HPLC peak 8 1.11 i 0.04
Ozonated pyrene, HPLC peak 9 1.22 i 0.11
Ozonated pyrene, HPLC peak 10 1.14 :1: 0.05
Ozonated pyrene, HPLC peak 11 0.99 i: 0.01
Ozonated pyrene, HPLC peak 12 0.93 :t: 0.13
Ozonated pyrene, HPLC peak 13 0.52 r 0.25”

' Each datum represents an average :1: 1 SD of the distance the dye traveled in three different
plates per experiment. All plates were from the same batch of cells. The average distance the
dye traveled in the control was 140 pm. The ozone dose was 1.8 mmol ozone/0.5 mmol of
pyrene. The above results were significant at p s 0.01 as determined by ANOVA.

b These averages were found to differ significantly from the control at p s 0.01 as determined by
a Dunnett's t-Test.

 

44

2.4 Discussion

One of our objectives was to determine the optimum conditions for ozonating a
contaminant in an organic solvent system in which the reaction was not rate limiting in
either ozone or water. Water is a participating solvent in the reaction of ozone with
unsaturated and aromatic hydrocarbons (30), and if water is not present in the system in
sufficient quantity, the primary product is the Criegee ozonide, which is not the primary
product formed in an aqueous environment (Figure 2.1). High concentrations of many
organic molecules can be attained in a 90% acetonitrile/10% water mixture; therefore, we
wanted to determine if ozonation in 10% water will result in the same kind of products as

ozonations at very high water concentrations (e.g., at 60%).

The lack of major differences between the RP-HPLC chromatograms at low and
high concentrations of water shows that water was not a rate-limiting reactant (Figure
2.2A). Similarly, there were no changes in the RP-HPLC chromatograms of the
byproducts at various initial concentrations of pyrene (Figure 2.28), indicating that ozone
was not rate-limiting in this system. This shows that a pure aqueous system can be
mimicked by ozonating high concentrations of organic compounds such as pyrene in a
relatively nonpolar solvent/water mixture. The mixture of byproducts formed in such a

system can be directly tested for biological effects without the need to concentrate the

products.

45

The samples described above were ozonated at reaction-limited conditions in
which the effluent ozone concentrations remained constant with time. These conditions
followed a period of a high ozone demand where the effluent concentrations were zero
(Figure 2.3B). This period of maximal ozone demand increased with increasing
concentrations of pyrene (Figure 2.3A). As apparent from the HPLC chromatograms (see
Figure 2.4), during this period of maximum ozone demand, pyrene disappears and
intermediate products form and also subsequently disappear or diminish in concentration.
Only the most polar compounds (shorter retention times) continued to increase in
concentration with ozonation time. This is expected since the ozonation of organic
compounds usually results in the formation of products containing aldehyde, carboxylic
acid and hydroxyl functional groups (refs 30, 38 and Figure 2.1). Increasing the water
solubility of an organic compound would probably increase the mobility of the
compound(s) in an aqueous environment. This is significant since the more water soluble

compounds would be more easily mobilized into aquifers, streams and lakes.

At steady state, the concentration of effluent ozone was not equivalent to the
influent ozone concentration (Figure 2.38), which was probably the result of
autocatalytic reactions of ozone in aqueous solutions (36) and pressure differences from
diffusion processes across the liquid/air interface. Autocatalytic processes were
minimized by ozonating the solution at an approximate pH of 2 where the decomposition

of ozone to hydroxyl radicals is slow (3 7); this allowed us to assess the toxicity of the

46

mixture that was primarily the result of ozonation reactions.

The reaction of pyrene, benz[a]pyrene and fluoranthene at low ozone dosages
results in the formation of products that are more inhibitory than the parent compound
(1). Therefore, our ultimate goal was to determine if higher ozone dosages can remove all
inhibitory byproducts formed from the ozonation of a polycyclic aromatic hydrocarbon
such as pyrene. The removal of all the inhibitory components was achieved at an ozone
dosage of approximately 4.5 mol of ozone/mol of pyrene (Figure 2.5). The ozonated
pyrene solution was most inhibitory at the point where 50 % of the pyrene was removed,
showing that the intermediate products enhanced the toxic effect of pyrene. Although,
only 1.6 mol of ozone/mol of pyrene was needed to remove most of the pyrene, the
solutions ozonated at this level were still inhibitory to GJIC. This showed that monitoring
only the removal of the parent compound could result in the production of intermediate
products that are also biologically unsafe. Hydrogen peroxide, which is formed during the
ozonation of unsaturated and aromatic hydrocarbons (Figure 2.1), may also play an
important role in the inhibition of GJIC because ozonated mixtures of pyrene were more

inhibitory when hydrogen peroxide was not removed from the mixture (1).

The ozonation byproducts formed during the reaction of pyrene at low ozone
dosages were separated by RP-HPLC into several components. All fractions were shown

to be inhibitory to GJIC (1 ). In this paper, the byproducts formed were

47
chromatographically separated (see Figure 2.2) at high ozone dosages (approximately 3.6
mol ozone/mol pyrene). At this ozone dosage the mixture of byproducts was still slightly
inhibitory to GJIC (Figure 2.5). Only one out of the 13 components was found to be
inhibitory to GJIC (Table 2.1) as determined by the Dunnet's test. This component (peak
13) was one of the less polar compound(s) as indicated by the longer HPLC-retention
time. Due to the lack of power in the Dunnett's test, another peak such as peak 4 might
have been significant if more than three replicates were obtained but the overall
conclusion remains the same, i.e., the intermediate byproducts of ozonated pyrene, which
were very inhibitory to GJIC, readily reacted with ozone resulting in the formation of

stable compounds, most of which are not inhibitory to GJIC.

Observations described in this paper show the feasibility of using ozone to
remove pyrene and many of its byproducts and the importance of using a biological assay
to assess the amount of ozone required to reduce toxicity. In particular, the use of a non-
genotoxic assay is unique compared to previous work which employed genotoxic and
cytotoxic assays. The use of a non-genotoxic bioassay should make a significant
contribution in the development of risk assessment models for water treatment
technologies that are concerned with removing carcinogenic compounds. This is
especially true since most carcinogens appear not to be genotoxic (5-7) but rather
demonstrate tumor promoting activity, which has been linked to the down regulation of

GJIC (16). The most common genotoxic assay employed has been the Ames test which is

48

based on a prokaryotic bacterial system that makes extrapolation to mammalian systems
difficult. Furthermore, Ames test results are often conflicting. For example, the treatment
of water by ozone has shown both an increase (28) and a decrease (27) in mutagenicity.
Tumor initiating properties determined in vivo also showed no effect (27). Also,
mutagenicity is no longer equated with carcinogenicity. For example, water treated with
either chlorine, chlorine dioxide or monochloroamine was concentrated 4000x by XAD
resin adsorption and tested for mutagenicity and lung adenomas (2 7). The samples were

all mutagenic but did not cause the formation of any tumors.

2.5 Conclusion

In conclusion, the simple monitoring of the disappearance of pyrene during
chemical treatment is insufficient to protect human health and the environment.
Bioassays, such as GJIC, must be employed during the design, testing, and monitoring of
the engineered processes. In our system, we specifically showed that 4.5 mol of
ozone/mol of pyrene were required to remove all components inhibitory to GJIC, whereas

only 1.6 mol of ozone/mol of pyrene was required to remove most of the pyrene.

49

2.6 References

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(3)

(9)

(10)

Upham, B. L.; Masten, S. J.; Lockwood, L. R.; Trosko, J. E. Fund. Appl.
Toxicol.1994, 23, 470-475.

Pan, X-M.; Schuchmann M. N.; von Sonntag, C. J. Chem. Soc. Perkin Trans.
1993, 2, 289-297.

Pan, X-M.; Schuchmann M. N.; von Sonntag, C. J. Chem. Soc. Perkin Trans.
1993, 2, 1021-1028.

Pelizzetti, E.; Minero, C.; Sega, M.; Vincenti, M. Advanced oxidation
technologies for water and air remediation, 1994, l, 116.

Cairns J. Nature 1981, 289, 353-357.

Tennant, R.W.; Margolin, B. H.; Shelly, M. P.; Zieger, E.; Haseman, J. K.;
Spalding, J .; Caspary, W.; Resnick, M.; Stasiewicz, S.; Anderson, B.; Minor, R.
Science 1987, 236, 933-941.

Trosko, J. E.; Chang, C. C. In Banbury Report 31: New directions in the
qualitative and quantitative aspects of carcinogen risk assessment. Hart, R. W.;
Hoerger, F. D., eds.; Cold Spring Harbor: New York, 1988; p 139-170.

Trosko, J. 13.; Chang, 0. c.; Madhukar, B. v. Toxic. In Vitro 1990, 4,‘ 635-643.
Trosko, J. E.; Chang C. C.; Madhukar, B.V.; Oh, S. Y. In Vitro Toxicology 1990,
3, 9-26.

Sheridan, J. D. In: Cell to Cell Communication, W. C. DeMello, ed; Plenum

Press: New Mexico, 1987; p 187-232.

(11)
(12)
(13)
(14)

(15)

(16)
(17)
(18)

(19)

(20)

(21)

(22)
(23)
(24)

(25)

50

Loewenstein, W. R. Biochem. Biophys. Acta 1979, 560, 1-65.

Pitts, J. D.; Finbow M. E. J. Cell Sci. 1986, 4(suppl), 239-236.

Lawrence T. S.; Beers, W. H.; Gilula, N. B. Nature 1978, 272, 501-506.

Saez, J. C.; Conner, J. A.; Spray, D. C.; Bennett, M. V. L. Proc. Natl. Acad. Sci.
1989, 86, 2708-2712.

Trosko, J. E.; Chang, C. C.; Madhukar, B. V.; Klaunig, J. E. Pathobiol. 1991,
58, 265-278.

Trosko, J. E.; Madhukar, B. V.; Chang, C. C. Life Sciences 1993, 53, 1-19.
Yamasaki, H. Carcinogenesis 1990, 11, 1051-1058.

Trosko, J. E.; Chang, C. C.; Netzloff M. T erat. Carcinogen. Mutagen. 1982, 2,
31-45.

Gilula, N. B.; Fancett, D. W.; Aoki, A. Dev]. Biol. 1976, 50, 142-168.

Larsen, W. J.; Wert, S. E.; Brunner, G. 1). Dev]. Biol. 1986, 113, 517-521.

Ye, Y. X.; Bombick, D.; Hirst, K.; Zhang, G. X.; Chang, C.C.; Trosko, J. E.;
Akera, T. Fund. Appl.. Toxicol. 1990, 14, 817-83.

Cole, W. C.; Garfield, R. E. Am. J. Physio]. 1986, 251e, 411-420.

DeMello, W. C. Prog. Biophys. Mo]. Biol. 1982, 39, 147-182.

Trosko, J. E.; Jone, C.; Chang, C.C. Mo]. Toxicol. 1983, 1, 83-93.

Bergoffen, J.; Scherer, S. S.; Wang, S.; Oronzi, S. M.; Bone, L. J.; Paul, D. L.;
Chen, K.; Leusch, M. W.; Chance, P. F.; Fischbrook, K. H. Science 1993, 262,

2039-2042.

(26)

(27)

(28)

(29)

(30)

(31)
(32)

(33)
(34)
(35)
(36)
(37)

(38)

51

Holder, J. W.; Ehnore, E; Barrett, J. C. Cancer Res. 1993, 53, 3475-3485.
Miller, G. R.; Kopfler, F. C.; Condie, L. W.; Pereira, M. A.; Meier, J. R.;
Ringhand, H. P.; Robinson, M.; Casto, B. C. Environ. Health Perspec. 1986, 69,
129-139.

Cognet, L.; Courtois, Y.; Mallevialle, J. Environ. Health Perspec. 1986, 69, 165-
175.

Bourbigot, M. M.; Hascoet, M. C.; Levi, Y.; Erb, F.; Pommery, N. Environ.
Health Perspec. 1986, 69, 159-163.

Bailey, P. S. In Ozonation in organic chemistry; Wasser'man, H. H., ed.;
Academic Press: New York, 1982; p 1-469.

El-Fouly, M. H.; Trosko, J. E.; Chang, C. C. Exp. Cell Res. 1987, 168, 422-430.
Corless, C. E.; Reynolds G. L.; Graham, N. J.; and Perry, R. Wat. Res. 1990, 24,
1119-1123.

Dunmtt, C. W. J. Am. Stat. Assoc. 1955, 50, 1096-1121.

Yao, C. C. D.; Haag, W. R. Water Res. 1991, 25, 761-773.

Bader, H.; Hoigné, J. Ozone Sci. Techno]. 1983, 4, 169-176.

Staehelin, J.; Hoigné, J. Environ. Sci. T echnol. 1985, 19, 1206-1213.

Hoigné, J. ; Bader, H. Water Res 1976, 377-386.

Glaze, W. H. Environ. Health Perspec. 1986, 69, 151-157.

CHAPTER 3
THE OZONATION OF PYRENE IN ACETONITRILE/WATER MIXTURE:

PATHWAY AND PRODUCT IDENTIFICATION

ABSTRACT

The pathway by which ozone reacts with pyrene, a polycyclic aromatic hydrocarbon
(PAH), is described in this paper. Pyrene was dissolved in a 90 % acetonitrile : water
(v/v) mixture to achieve an initial concentration of 5 mM pyrene. The ozone dosages
were varied to obtain different extents of reaction. A stoichiometric ratio of 1.68 moles
ozone/mole pyrene was required to complete destroy pyrene. The ozonation products
were identified by gas chromatography / mass spectrometry (GC/MS). Fourteen products
including aldehyde and carboxylic acid substituted phenanthrene- and biphenyl- type
oxidation products were identified. Several ring-cleavage reactions occurred sequentially:
i) one ring cleavage of pyrene occurred at the 4,5 position, with phenanthrene-type
products predominating; and ii) once secondary ring cleavage occurred, the concentration
of pyrene and the phenanthrene-type products decreased dramatically, while the
concentration of biphenyl-type products increased. The oxidation of pyrene by hydroxyl

free radicals was found to be involved in the reaction, even at pH 3.7.

52

53

3.1 Introduction

Polycyclic aromatic hydrocarbons (PAHs) are derived from the incomplete
combustion of organic matter. Mineral fuels, coal derived oils, tobacco smoke and
vehicle exhaust (1) contain PAHs. The release of PAHs to the environment has resulted
in the presence of detectable levels of PAHs in air, water and soil (2). As PAHs are
sparingly soluble or insoluble in water and recalcitrant, they will tend to accumulate on
solid phases, including on air-bome particulate matter and soil organic matter. The
accumulation of PAHs in the environment has resulted in the need for remediation
processes that will reduce the risk of human exposure to these chemicals, many of which

are carcinogenic or potentially carcinogenic (2).

Ozone, which is used both in water and wastewater treatment, is a very powerful
oxidant that can react with PAHs with rate constants on the order of >103 M" s" (3,4).
The products formed fiom the reaction of ozone with PAHs have been identified by several
investigators (5-10). In most studies (5 -8), the ozonation products were generated in pure
organic solvents. Neff (8) found that, in organic solvents, the reaction of PAHs with
ozone produces various diacids, quinones, and diones, and results in the oxidation of
alkyl side chains. However, limited data describing the oxidation of PAHs in aqueous
solutions are available (11). Additionally, when ozonation reactions were performed in
aqueous solutions excess ozone was used and only the frnal products were identified.
Dreher and Klamberg (9) found that in dilute aqueous solutions, the ozonation of
fluorene, acenaphthene and fluoranthene in the presence of excess ozone resulted in ring

cleavage with the formation of a number of carboxylated benzenes, quinones and

54

aliphatic compounds.

Pyrene, the target PAH used in this study, can be completely destroyed using
ozone treatment (11, 12). While the toxicity of pyrene has been investigated (13), to our
knowledge, no studies other than those of Upham et al. (12, 14) have investigated the
toxicity of the products formed from the ozonation of pyrene. These studies have shown
that the ozonation of pyrene results in the formation of several intermediates that are
more inhibitory to gap junction intercellular communication (GJIC) than is pyrene (14). It
was also shown that an ozone dosage of 1.6 moles ozone/mole pyrene was required to
oxidize pyrene entirely, whereas a dosage of 4.5 moles ozone/mole pyrene was required
to destroy all products inhibitory to GJIC (12). Based upon this work, it became evident
that research is needed to identify the products formed from the reaction of ozone with
pyrene. By identifying these products and the reaction pathway, it is our aim to develop
treatment alternatives that would minimize the formation of toxic products. The pathway
by which ozone reacts with PAHs has been studied in a non-participating solvent (15),
and Bailey (16) proposed a reaction pathway based upon the results. However, the

reaction pathway has not been investigated in water, which is a participating solvent.

The objective of this study is to identify the ozonation products and to propose a
pathway for the reaction of pyrene with ozone in aqueous solutions. As the intermediates
formed from the reaction of ozone and pyrene were found to inhibit GJIC more than
pyrene, pyrene was used as the model compound. The ozone dosages were varied to

result in different extents of reaction. The ozonation products obtained after pyrene had

55

reacted to different extents were identified. Also, the relative concentrations of pyrene

and its ozonation products were quantified at different extents of reaction.

3.2 Methods
Ozonation experiments

Ozone was generated fi'om dried oxygen by electric discharge using a Polymetrics
Model T-408 ozone generator (San Jose, CA). The system is illustrated in Figure 3.1.
Pyrene solution was added to a 125 mL gas washing bottle containing 100 mL mixture of
acetonitrile /water (90:10 v/v). The tubing (1/8" i.d.), connectors and valves were
constructed of Teflondb or stainless steel. Gaseous ozone was bubbled into the pyrene
solution which was continuously mixed using a magnetic stirrer and stir bar. Prior to
commencing ozonation, helium was passed through the system to clean it. After purging
with helium, ozone was allowed to flow into the system but by-passed the reactor to a
trap containing 2% (w/v) KI in water. When the influent ozone reached steady state, the
influent ozone was then passed through the reactor to commence the reaction. The flow of
ozone into the reactor was regulated at 100 mL/min using a Sidetrack flow controller
(Sierra Instruments Inc., Monteray, CA). The concentrations of ozone in the influent and
effluent gas streams were measured spectrophotometrically at 258 nm using a UV-vis
spectrophotometer (Model 1201 , Shimadzu Scientific Instruments, Japan). The
absorbance values were converted to concentration units using a molar absorptivity
coefficient for ozone of 3000 M" cm" (I 7). Quartz flow cells with a path length of 0.2 cm

were used. To avoid the formation of radical species, ozone was not quenched using

56

Gas Outlet

 

 

 

 

 

 

 

 

   
 

 

 

 

 

 

 

 

 

 

 

 

{—
(— v
\
.. ./ 1 +
.0. ; Flow
- m Controller
0 O I
Ozone 09‘— mo ‘
C t 1 ; *
UV on actor: UV V
' Flow
Flow Cell : Cell Ozone
V By-pass Gaseous
Ozone To KI Trap Ozone
By-pass Inlet
To KI Trap .

— Ozone Line
' ' He Line For System Clean Up
- - - Ozone By-pass

Figure 3.1 Configuration of ozonation system

57

chemical reagents but rather, reactions were terminated by purging the ozone fi'om the
reactor with helium. An average of 15 to 30 sec were required to purge the ozone to

below the detection limit (ABS < 0.001).

When the concentration of ozone in the effluent gas was constant, i.e., when the
byproducts did not react further with ozone (as illustrated in Figure 3.2), the reaction can be

considered to be complete.

0.4 “

0.35 W

0.3 4‘

0.25 1*
ABS 0.2 “r
0.15 ‘

0.1 ‘

0.05 ‘

 

 

1 L A I A I 41
0 . . 2 . r ‘ ‘

0 500 1000 1500 2000 2500 3000 3500

TIME (see)

Figure 3 .2 Ozone breakthrough curve for ozonation of 5 moles“. pyrene dissolved in 90%
CHgCN/water solvent mixture. 0 Influent ozone absorbance; I Efiluent ozone
absorbance.

58

In order to study the reaction pathway for the primary reaction of ozone with each of the
target PAHs, the byproducts fi‘om the initial attack of ozone on the PAHs need to be
identified. The initial reaction byproducts were generated in the same manner as that

described above except the reaction was stopped before ozone was detected in the off-gas.

Since pyrene has very low aqueous solubility (SW = 32 ppb at 24 °C), pyrene was
dissolved in acetonitrile so as to achieve sufficiently high concentrations of the ozonation
products to allow for their direct identification by gas chromatography/mass spectrometry
(GC/MS). Concentration procedures such as solvent extraction were avoided to minimize
losses of ozonation products. Acetonitrile (99.8% purity, EM Science, Gibbstown, NJ)

was chosen as the organic solvent because it is miscible in water and it has a low

reactivity with ozone (tl,2 2 18 years at pH 7 and [63] = 20.8 mM) (18). Water was added
to the acetonitrile at a sufficiently high concentration for the water to act as a
participating solvent. As evidenced in a previous study (12), the HPLC chromatographic
profiles for various concentrations of water ranging fiom 10% to 60% (volume ratio)
were identical. Therefore, a 90% acetonitrile/water (v/v) mixture was chosen for all the
experiments. In aqueous solutions, the reaction of ozone with CH will form °OH
radicals. To minimize side reactions resulting from the oxidation of pyrene by -OH
radicals, acidified distilled water was used in preparing the solvent mixture. Distilled
water was acidified to pH 2 using phosphoric acid (85% purity) and reduced the pH of the
solvent mixture to 3.7. Pyrene (99% purity, Sigma Chemical Co., St. Louis, MO) was

dissolved in 100 mL of the acetonitrile/water mixture (90:10) to achieve a pyrene

59

concentration of 5 mM.

Determination of hydrogen peroxide

The presence of hydrogen peroxide was determined by modifying the N,N-
diethyl-p-phenylenediamine (DPD) method (19). The ozonated solution (5.4 mL) was
mixed with 0.6 mL of sodium phosphate buffer solution to maintain a pH > 6. Afier 10
sec, the solution was transferred into a UV cell of 1 cm pathlength. The color was

measured at 554 nm after 45 sec.

Analysis of ozonation products

Ozonated samples were evaporated under helium and the residue was dried over
P20, in a vacuum desicator for 8 hours. The completely dried sample was derivatized by
silylation using bis-trimethylsilyl / trifluoroacetamide (BSTFA) + 1% of
trimethylchlorosilane (TMCS) (Regis Technologies Inc., Morton Grove, IL) at 100°C for 20
min to convert all fi'ee -OH and -COOH groups into their volatile TMS-ether (-OSiMe3) and

TMS-ester (~COZSiMe3) derivatives, respectively.

The ozonation products obtained were identified by GC/MS. GC/MS was
performed using a JEOL AX-505H double-focusing mass spectrometer coupled with a
Hewlett-Packard 5890] GC (Norwalk, CT). A DBSMS (30 m length x 0.32 mm i.d. x 0.25
pm film thickness) fused silica capillary column (J&W Scientific, Rancho Cordova, CA)

was employed for GC separation. A splitless injector was used with a column head

60

pressure of 10 psi using helium as the carrier gas, producing a flow rate of ca. 1 mL/min.
The initial column temperature was held for 2 min at 100°C, ramped at 20°C/min to
220°C, then ramped at 5°C/min to 260°C, and finally ramped at 20°C/min to 300°C. The
mass spectrometer was operated in electron impact mode. Mass calibration of the
spectrometer was performed using perfluorokerosine. The molecular ions of several
derivatized products (those whose molecular ions have a low intensity in the electron
impact mode) were confirmed by gas chromatography - chemical ionization (methane) -

mass spectrometry (GC-CI(CH4)-MS).

Quantitation

The relative amounts of pyrene and the ozonation products were assessed using
GC/MS by determining the area response of each compound relative to an internal
standard. The internal standard, 1,8-naphthalaldehydic acid (97% purity, Aldrich
Chemical Co., St. Louis, MO), was dissolved in pure acetonitrile to obtain a stock
solution concentration of 5 mM. The sample solutions were prepared by adding 25 uL of
the stock solution containing the internal standard into 50 uL of each of the ozonated

solutions prior to the drying step. The samples were then derivatized as described above.

The presence of the analytes was detected using GC/MS. A H? 5970 mass
spectrometer ChemStation (Norwalk, CT) was used. A SPB-l (30 m length x 0.25 mm i.d.
x 0.25 pm film thickness) fused silica capillary column (Supelco Inc., Supelco Park,

Bellefonte, PA) was employed for GC separation. A splitless injector was used with a

61

column head pressure of 7.5 psi using helium as the carrier gas, producing a gas flow rate
of ca. 1 mL/min. The initial column temperature was held for 5 min at 200 °C, ramped at
5 °C/min to 260 °C, then ramped at 20 °C/min to 300 °C. The mass spectrometer was
operated in electron impact mode. Mass calibration of the spectrometer was performed

using perfluorotributylamine.

3.3 Results

Pyrene was oxidized using ozone at dosages ranging from 0.02 to 2.35 moles of
ozone. In all experiments, the initial pyrene concentration was 5 mmoles/L. The effect of
ozone demand on the residual pyrene concentration, relative to the internal standard, is
shown in Figure 3.3. The residual pyrene concentration was determined using the GC
area response for pyrene (AW) relative to that for 1,8-naphthalaldehydic acid (25 uL of 5
mM) which was used as an internal standard (Al). The ozone demand was calculated for
each specific reaction time:

OzoneDernand = (C, — C,) x Q x t

where Ci and Ce are the influent and effluent gaseous ozone concentrations, respectively,
Q is the gas flow rate and t is the reaction time. Based on the regression line obtained
from the plot of the first 6 data points (see Figure 3.3), at an ozone/pyrene ratio of 1.68 :1:
0.05 moles ozone/mole pyrene, pyrene had completely reacted. This compares to 1.6

moles ozone/mole pyrene that was determined in an earlier study (12).

By varying the extents of reaction, 14 products were identified (see Figure 3.4).

62

 

 

 

Figure 3.3 The effect of ozone dosage on the residual pyrene concentration. The residual
pyrene concentration is determined by the GC area response for pyrene (AW)
relative to the response for 1,8-naphthalaldehydic acid (25 uL of 5 mM) used
as an internal standard (A1). The solid line represents the linear regression
based on the first six data points. The arrow corresponds to the point where
0.5 mole of pyrene had completely reacted by a total ozone demand of 0.84
mole, i.e., a ratio of 1.68 moles ozone/mole pyrene.

63

OH HO

 

11 [[1 MB [11 NB IV

o-C”

 

 

Figure 3.4 Products formed from the ozonolysis of pyrene.

64

The GC/MS characteristics of each product are given in Table 3.1. Two typical GC
chromatograms (for ozone dosages of 0.38 and 1.34 mmoles) are provided as Figure 3.5a

and 3.5b, respectively.

Table 3.1 GC/MS characteristics of pyrene ozonation products

 

GC molecular weight of
product retention parent TMS

 

 

ID time compd. deriv.‘ Important Ion peaks, m/z
I 8'25" 202 202 202 (M+.) 101
11 9'38" 234 234 234 (M+.) 205 176 151 102

IIIA/B 10'26" 250 322 322 (M+.) 293 205 189 176
IIIA/B 11'05" 250 322 322 (M+.) 307 294 205 189

IV 8'35" 222 294 294 (M+.) 293 205 176 151
V 904" 252 396 396 (M+.) 307 293 279 189
VI 8'05" 206 206 206 (M+.) 205 176 151 103
VII 8'50" 222 294 294 (M+.) 279 205 177 151
VIII 9'10" 220 220 220 (M+.) 193 163 110

IX 9'00" 266 266 266 (M+.) 237 209 181 152
X 923" 282 354 354 (M+.)” 325 295 237 221

XI AB 945" 298 442 442 (M+.)” 427 325 309 269
X] A/B 9'53" 298 442 442 (M+.)” 427 398 325 235
XII 9'58" 314 530 530 (M+.)” 515 471 385 323
XIII 1019" 330 618 618 (M+.)” 603 558 323 221
XIV 7'50" 282 426 426 (M+.) 411 337 249 175
XV 1013" 264 336 336 (M+.)b 295 246 237 221

 

‘ theoretical value based on molecular weight of proposed TMS derivative

b derivatived molecular ion confirmed using CI

65

CC Retention Time (min)
6 8 10 12

1m 4 1 1 1 L L 1 I a L L L L

I

 

soj

60

40

zoj

Relative Abundance (%)

 

 

 

 

100 ‘ r r r 1“
80

60

40

Relative Abundance (%)
a

 

 

 

 

 

Md M X“

"'300" '6ob'*"7ob' ' 80b"F9%'
ScanNumber

 

 

Figure 3.5 (a)The reaction of pyrene (0.5 mmole) with 0.38 mole of ozone.
(b) The reaction of pyrene (0.5 mmole) with 1.34 moles of ozone.
The Roman numerals corresponds to the products listed in Figure 3.4.

 

66

Although library searches of standard spectra in NIST MS Search (ver 1.5, 1996) and
BenchTop/BPM Search (ver 3.10d, Wiley & Son, 6th ed., 1994) were conducted, due to
the paucity of information in these data base, no matches were found. The tentatively
assigned identifications of the ozonation products were based on interpretations of the
mass spectra (see Appendix B). Prior to the disappearance of pyrene, one ring cleavage
phenanthrene-type products predominated. The major product of this type was 4-carboxy-
5-phenanthrenecarboxyaldehyde (4-CPA, IIIA/B); its mass spectra and the interpretation
of the mass spectra are shown in Figure 3.6. Once pyrene had completely reacted,
secondary ring cleavage occurred and biphenyl-type products predominated. The major
product of this type was 2’,6,6’-biphenyltrialdehyde-2-carboxylic acid (X); its mass

spectrum and the interpretation of the mass spectrum are provided in Figure 3.7.

Two pairs of structural isomers have been tentatively identified. The first pair, 111
A/B (mass spectra of TMS derivatives are shown in Figure 3.6), contains aldehyde and
acid functional groups. The second pair, XI A/B (mass spectra of TMS derivatives are
shown in Figure 3.8), contains dialdehyde and diacid firnctional groups. When an
aldehyde and a carboxylic acid functional group are adjacent to one another, a cyclization
may occur. For example, an equilibrium between the original and the cyclized structure

01“” —- 0:33

of 2-carboxybenzaldehyde coo" C’O exists. Afier derivatization of this
compound with TMS, two peaks were observed using GC/MS. Only one commercial

compound, 4-CPA (Sigma-Aldrich Library of Rare Chemicals, Milwaukee, WI), could be

67

Figure 3.6 Mass spectra and structure interpretation of fragmentation patterns for 4-
carboxy-S-phenanthrenecarboxyaldehyde (III A/B) derivatized by BSTFA +
TMCS. The migration of Me3Si within a molecule is described in References
24,25. The bold numbers correspond to the important peaks on the mass
spectra.

68

 

 

 

 

 

 

 

b

 

 

 

 

 

 

 

 

 

 

 

100
l 189 III A/B .
gs 80: h
23’ I 294
o 1
.8 .o.
a 4
=1 .
.2 ‘ I
o 40‘ 73 -
'5 ‘ 05 '
£3 1 176 2 +. I
é) 20‘ 233 279 322 (M ) _
< 131 151 263 307 :
ct"**r"f~fifi T_, If fit’
10050 100 150 200 M/Z 250 300 350 400
‘ 205 293 III A/B .
g; 80:
3;: :
5 60:
'8 . 73
S
'2 3 189
40.
.1.>’ .
g : 176 278
a: 20, +
; 322 (M ')
. 88 131 151 263
O J I i I .L [I L
0"'100'I'150'*‘z W250' -00' ‘3so'r'71
M/Z
CHO 028N131: 294 .OSiMea o
/ 8‘ C) c
Q ”\0“; 9; Q
3220“")
lMegSi migration
(ref24,25)

:fSiMe)
H .

02

@©©

 

©©© - ‘OSiM/ez91

69

 

100 325

80<

73

295

Relative Abundance (%)

  

 

 

 

 

 

 

 

   

chm —2°~m '—-~Iso 452%..”

'OSiMeg
.28}-

03mg: @.©-c —_’ I93 —" I65
/

221

' ' ' ttems for
F1 e 3.7 Mass spectrum and structural Interpretation of fiagmentauon pa
gm 2’,6,6’-biphenyltrialdehydo-2-carboxyhc acrd (X) denvafized by BSTFA +
TMCS

100

70

 

8 8 8
A l a L l 1 A a A 1 a A A

Relative Abundance (%)

 

73

147

 

269

325

 

XIA/BI

 

 

 

Relative Abundance (%)

N
o
l A a

 

Z A I

235

 

 

 

 

325

 

383
l
398
427 .
[442(M+-):
A v t V v 5m

 

 

Figure 3.8 Mass spectra of the isomeric forms of XI A/B.

7]

obtained but it contained impurities, including compounds VI, X, and XI NB. This
formulation was prepared so as to compare its GC/MS spectra with that of compound III
A/B. Using GC/MS, the TMS derivative of 4-CPA was found to exist in two isomeric

forms which were identical to those shown in Figure 3.6.

Several of the products identified appeared to have resulted from the oxidation of
pyrene by free radicals, most likely by -OH. These include one major product, IV, and
minor products, 11, V, IX, XIV and XV. To test this hypothesis, two experiments were
performed either in pure acetonitrile or in an acetonitrile/water solvent mixture that
contained t-butanol (l M) as an -OH radical scavenger. In solutions containing t-butanol,
the sample solutions contained qualitatively higher concentrations of hydrogen peroxide
than in the solvent mixture without t-butanol. Products V, XIV and XV were not
produced in either pure acetonitrile or in the presence of t-butanol; however, the relative
concentration of product IV increased. Products 11 and IX were formed in the acetonitrile
/water solvent mixture both in the presence and absence of t-butanol, but were not found

in pure acetonitrile.

The effect of the ozone : pyrene ratio (reported as moles ozone/mole pyrene) on
the distribution of the products and their relative concentrations is shown in Figure 3.9.
The product concentrations are given as the ratio of the GC area response of the analyte
relative to that of the internal standard (A/AI). Product 11 and phenanthrene-4,5-
dicarboxylic acid were found only during preliminary experiments in which the ozonated

solutions were concentrated by freeze-drying and the concentrate was reconstitutied in

72

Figure 3.9 The relative concentrations (A/AI) of pyrene and its ozonation products
obtained under different ozone : pyrene ratios. The product concentrations
were determined relative to 1,8-naphthalaldehydic acid (25 uL of 5 mM) used
as an internal standard.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o O O o O
‘0. ‘12 N. o- 00' ' ' .

d[03 ]/[PY]

 

Product H)

 

74

acetonitrile. As such, no quantitative data were obtained. At an ozone : pyrene ratio of
0.04 mole ozone/mole pyrene, only products III and IV were observed. Initially,
concentrations of these products increased as the ozone : pyrene ratio increased. When
the ozone : pyrene ratio was increased to 0.58 mole ozone/mole pyrene, several
intermediates, including compounds V, VI, VII and VIII, could be detected. Additionally,
the secondary-ring-cleavage products X and XI were detected. At an ozone : pyrene ratio
5 1.30 moles ozone/mole pyrene, reaction product 111 predominated; several of the
secondary-ring-cleavage products (e.g., IX~XII and XV) appeared. When the ozone :
pyrene ratio was increased to the point where pyrene had completely reacted, the
concentration of product III decreased dramatically. The concentrations of the other
phenanthrene-type products (e.g., IV, V, VI, VII and VIII) also decreased to below
detection limits. At this time, compound X became the predominant product. Upon
further increases in the ozone : pyrene ratio (to > 1.77 moles ozone/mole pyrene), the
concentration of product III decreased to below its detection limit and the concentrations
of the biphenyl-type products gradually increased. When the ozone : pyrene ratio reached
2.70 moles ozone/mole pyrene, the concentration of compound X reached a maximum
value. Upon further increases in the ozone : pyrene ratio, the concentrationof compound
X decreased and the concentration of other biphenyl-type products increased. At ozone :
pyrene ratios between 2.70 and 5.00 moles ozone/mole pyrene, compound XI A/B

became the predominant product .

75
To confirm the proposed ozonation pathway, the ozonation of commercial 4-CPA
was investigated. This mixture was prepared and ozonated in the same manner as that
used with pyrene. The effect of the ozone : 4-CPA ratio (reported as moles ozone/mole 4-
CPA) on the distribution of the products and their relative concentrations is shown in

Figure 3.10.

 

 

d[03]/[4-CPA]

 

Figure 3.10 The relative concentration (A/AI) of 4-CPA and its ozonation products
obtained under different ozone : 4-CPA ratio. The product concentrations
were determined relative to 1,8-naphthalaldehydic acid (25 uL of 5 mM)
which was used as an internal standard (AI).

76

Secondary ring cleavage products, such as products X ~ XII, appeared when the ozone
dosage was only 0.05 mole ozone/mole 4-CPA. Products IV, VII and VIII were not
found. Product V and VI were present, but below detection limits. Upon increases in the
ozone dosage, the concentration of products X ~ XII increased, but products V and VI
disappeared. Also, product IX was observed upon increasing the ozone dosage. At an
ozone dosage of 0.56 mole ozone/mole 4-CPA, 4-CPA had completely reacted, and only

products IX ~ XII were detected. Product X predominated.

3.4 Discussion

The addition of ozone at an ozone : pyrene ratio of 1.68 moles ozone/mole pyrene
resulted in the elimination of pyrene'. This value is much less than the stoichiometric
ratios of >100 moles ozone/mole pyrene determined by Corless et al. (11) for the
ozonation of pyrene in dilute aqueous solution at pH 4.5 - 6.0. The large stoichiometric
ratios used by Corless et al. may be necessary since in dilute aqueous solutions, the
aqueous self-decomposition of ozone is very important as compared to the consumption
of ozone by organic compounds. As such, a smaller percentage of the applied ozone
would go to the oxidation of the target chemicals than we would observe in our more
concentrated solutions. Based upon this study, at stoichiometric ratios < 5 moles
ozone/mole pyrene, rapid destruction of the ozonation intermediate products also

occurred. Based upon results from a previous study (12), at these ozone dosages, all the

 

‘ Note: The reaction of ozone with ozonation products occurs prior to the elimination of
pyrene; therefore, ozone would be consumed simultaneously by pyrene and the initial by-
products. This would be expected to result in a stoichiometric ratios greater than one
which is the value that would be expected if ozone reacted only with pyrene.

77

intermediate products which were found to have the potential to promote tumor grth
are removed. Thus, it is likely that at the ozone dosages used in water and wastewater
treatment for the oxidation of pyrene, ozonation would result in a reduction in epi genetic

toxicity.

Based upon the products identified along with the sequence of their appearance, a
pathway for the ozonolysis of pyrene is proposed (see Figure 3.11). Only the pathway for
the formation of one ring cleavage products is shown in the figure. The pathway by which
second ring cleavage occurs is similar. Ozone directly attacked the double bond causing
ring cleavage of pyrene at the 4, 5 (or 9, 10) position. At the early stages of the reaction,
ozone reacted with pyrene to form six one-ring cleavage phenanthrene-type products.
Pyrene reacted to form a dialdehyde , e. g., product II, at a concentration below its detection
limit. An internal rearrangement of II involving the dialdehyde groups could result in ring
closure with a ketone and a hydroxyl functional groups substituted on the cyclic ring. The
major ozonolysis product fi-om pyrene was a phenanthrene substituted with an aldehyde
and carboxylic acid ([11). This is consistent with the pathway proposed by Bailey (16).
Upon further ozonation, either the aldehydic functional group was oxidized to form a
carboxylic acid or the hydrogen on the carboxylic acid was released and decarboxylation
resulted in the formation of minor products VI, VII and VH1. The decarboxylation product
VIII is consistent with that shown by Davies and Waring (as cited in Sheldon and Kochi
(20)). When 4-CPA, the commercial compound of product III, was ozonated, products
VII and VIII were absent from the oxidized mixture. This suggests that ring cleavage

occurs preferentially over decarboxylation.

78

Figure 3.11 Proposed pathway for the first ring cleavage of pyrene during ozonolysis.
Compounds shown in brackets [ ] were detected only in ozonated solutions
which were concentrated by freezedrying and reconstituted in acetonitrile.
The compound shown in { } braces is hypothesized, and would be formed
by a sequence of reactions involving OH in the presence of Oz.

 

 

 

 

666% 66.1 .666

 

 

80

The direct reaction of ozone with pyrene could also result in the formation of
product IV. The ozonolysis of pyrene in solutions containing t-butanol as an -OH radical
scavenger or in pure acetonitrile, resulted in an increase in the concentration of product
IV (as compared to ozonolysis in the solvent mixture without radical scavenger). This
result is surprising since product IV would not be formed by the classical Criegee pathway.
While it is unclear exactly how product IV is formed, one possibility is that a trioxalane
forms, followed by a rearrangement to form an aldehyde group and a
hydroxyhydroperoxide which undergoes decarboxylation to form an alcohol. This
hypothesis was substantiated by the results of an ozonation experiment in which 4-CPA

(III) was used as the parent compound. Product IV was absent in the oxidized mixture.

In aqueous solutions (pH > 4), ozone autocatalytically decomposes to form
hydroxyl radicals. The -OH radicals, which enter into a cyclic reaction, enhance the rate
of ozone self-decomposition (21 ). They may also react with PAHs to form phenyl-type
compounds that can be further oxidized to form ketones. Although at pH 3.7 the
decomposition of ozone to form ~OH radicals would not be expected to be significant,
work by Masten et al. (22) has shown that, even at low pH, -OH radicals can still
responsible for degradation of TCB. Work by Xiong et al. (23) has shown that the
formation of ozonation products such as glyoxalic acid or other ketoacids can participate
in the initiation step for the decomposition of ozone, resulting in the formation of -OH
radicals. Based upon our results, it appears that even at pH 3.7, -0H radicals were
involved in the reaction of pyrene and the production of product V and the ketone type

products XIV and XV. When using t-butanol (1M) as an -OH radical scavenger or when

81

conducting the ozonolysis reactions in pure acetonitrile, the products V, XIV and XV
were not observed. Hydroxyl radical attack, with the addition of oxygen on pyrene, would
result in the formation of the diol intermediate. The reaction of this diol intermediate with
ozone and the subsequent dehydrogenation would form product V. Upon further
ozonation and dehydrogenation of product V, a dialdehyde would form, e. g., compound
II. This could explain why products II were found during the ozonolysis of pyrene in the
solvent mixture but not found in the pure acetonitrile since the diol would not be formed
in a acetonitrile. The ketone type products, XIV and XV, likely resulted fiom the direct
attack of -OH radicals on pyrene followed by dehydration. Ozonation products will also
react with -OH radicals. For example, product 111 reacts with ozone to form a dicarboxylic
acid intermediate. This intermediate can further react with ozone, followed by the
dehydrogenation of the carboxylic acid functional group. Alternatively, it can react with
°OH. Again dehydration of the carboxylic acid functional group would follow. In either
case, the formation of a phenantherene with one carboxylic acid group and a ~COZ radical
would result. This species can undergo electron transfer/rearrangement of -CO, and
subsequent the decarboxylation of one of the carboxylic acids, resulting in the formation of

product VIII.

Upon further ozonation, ozone reacted with the phenanthrene-type compounds at
the 9,10 position resulting secondary ring cleavage and the formation of biphenyl-type
products (IX ~ XV). This pathway is not shown in Figure 3.11, however the pathway for
the second ring cleavage is expected to be similar to that observed for cleavage of the first

ring. As the concentrations of the biphenyl-type products increased gradually, the

82

concentrations of the phenanthrene-type products decreased dramatically. After the major
phenanthrene-type product (III) had disappeared, the ring cleavage biphenyl-type

products X and XI predominated.

3.5 Conclusions

A pathway for the ozonation of pyrene in acetonitrile/water mixtures has been
described. As sufficient water was present for it to participate in the ozonation reaction
involving pyrene, this work is relevant to aqueous systems. Ozonation resulted in
stepwise ring cleavage and in the oxidation of aldehydic functional groups. Even at pH
3.7, the -OH radicals appeared to be involved in the reaction of the primary ozonation

products.

The identification of the reaction products was accomplished by GC/MS. Future
work will include preparatory TLC and NMR to confirm the product identification. Once
pure fractions of the products are obtained, these individual products will be ozonated to

further investigate the mechanistic pathway.

83

3.6 References

1. Yoshikawa, T.; L.P. Ruhr; W. Flory; D. Giamalva; D.F. Church; W. Pryor, "Toxicity
of Polycyclic Aromatic Hydrocarbons. 1. Effect of Phenanthrene, Pyrene, and Their
Ozonized Products on Blood Chemisu'y in Rates". Toxicol. Appl. Pharmacl. 1985, 79,
21 8-226.

2. Sontag, J .M. In Carcinogens in Industry and the Environment. New York: Marcel
Dekker, Inc., 1981, 169.

3. Hoigné, J.; H. Bader, "Rate Constants of Reactions of Ozone with Organic and
Inorganic compounds in Water-I Non-dissociating Organic Compounds". Wat. Res.
1983, 17, 173-183.

4. Butkovic', V.; L. Klasinc; M. Orhanovic; J. Turk; H. thsten, "Reaction Rates of
Polynuclear Aromatic Hydrocarbons with Ozone in Water". Environ. Sci. Technol.
1983, 17(9), 546-548.

5. Legube, B.; S. Guyon; H. Sugimitsu; M. Dore, "Ozonation of Naphthalene in Aqueous
Solution - 1. Ozone consumption and Ozonation Products". Wat. Res., 19863, 20(2):
197-208.

6. Marley, K.A.; R.A. Larson; P.L. Stapleton; W.J. Garrison; C.M. Klodnycky,
"Ozonolysis of Naphthalene Derivatives in Water and in Kerosene Films". Ozone:
Sci. and Eng. 1987, 9(1): 23-36.

7. Kuo, CH. and HM. Barnes, In Reactions of Ozone with Organics in Aqueous

Solutions. US. EPA Report No. EPA/600/3-85/031. 1985, 66.

84

8. Neff, J .M. In Polynuclear Aromatic Hydrocarbons in the Aquatic Environment:
Sources, Fates and Biological Eflects. London: Applied Science Publishers, Ltd., 1979,
227.

9. Dreher, W.; H. Klamberg, "On the Degradation of Polycyclic Aromatic Hydrocarbons.
III. Reaction Products of the Ozonolysis of Further Polycyclic Aromatic
Hydrocarbons in Aqueous Systems". Fresenius Z. Anal. Chem. 1988, 331, 290-294.

10. Meineke, 1.; H. Klamberg, "On the Degradation of Polycyclic Aromatic
Hydrocarbons. I. Reaction Products of the Ozonolysis of Polycyclic Aromatic
Hydrocarbons in Aqueous Systems". Fresenius Z Anal. Chem. 1978, 293, 201-204.

11. Corless, C.E.; G.L. Reynolds; N.J.D. Graham; R Perry, "Ozonation of Pyrene in
Aqueous Solution". Wat. Res., 1990, 24(9), 1119-1123.

12. Upham, B.L.; J.J. Yao; J.E. Trosko; SJ. Masten, "Determination of the efficacy of
Ozone Treatrnent System Using a Gap Junction lntercellular Communication
Bioassay". Environ Sci. T echnol. 1995, 29(12): 2923-2928.

13. National Academy of Sciences, "Biologic Effects of Atmospheric Pollutants". In
Particulate Polycyclic Organic Matter. Washington, DC, NAS. 1972, pp. 361.

14. Upham, B.L.; S.J. Masten; B.R. Lockwood; J .E. Trosko, "Nongenotoxic Effects of
Polycyclic Aromatic Hydrocarbons and Their Ozonation By-Products on the
lntercellular Communication of Rat Liver Epithelial Cells". Fundamental and Applied
Toxicology. 1994, 23, 470-475.

15. Pryor, W.A.; G.J. Gleicher; D.F. Church, In Reaction of polycyclic aromatic
hydrocarbons (PAI-D with ozone. Linear flee-energy relationships and tests of likely

rate determining steps using simple molecular orbital correlations, Manuscript, 1983.

85

16. Bailey, P. S. In Ozonation in Organic Chemistry, Wasserman, H. H., ed.; Academic
Press: New York, 1982, 1-469.

17. Bader, H.; J. Hoigné, "Determination of Ozone in Water By The Indigo Method - A
Submitted Standard Method". J. Ozone Sci. Technol. 1982, 4, 169-176.

18. Yao, C. C. D.; W. R. Haag, "Rate Constants for Direct Reactions of Ozone with
Several Drinking Water Contaminan ". Water Res. 1991, 25, 761-773.

19. Bader, H.; V. Sturzenegger; J. Hoigné, "Photometric Method for the Deterrninetion of
Low Concentrations of Hydrogen Peroxide by the Peroxidase Catalyzed Oxidation of
N,N,-Diethyl-p-Phenylenediamine (DPD)". Wat. Res. 1988, 22(9), 1109-1115.

20. Sheldon, R.A.; J .K. Kochi, "Oxidative Decarboxylation of Acids by Lead
Tetraacetate". In Organic Reactions, Dauben, W.G. et. al. ed.; John Wiley & Sons,
Inc., 1972, 19, 322.

21. Gurol, M.D.; P.C. Singer, "Kinetics Of Ozone Decomposition: A Dynamic
Approach". Environ. Sci. T echnol. 1982, 16(7):377-3 83.

22. Masten, S.J.; M. Shu; M.J. Galbraith; S.H.R. Davies, "Oxidation of Chlorinated
Benzenes Using Advanced Oxidation Processes". Hazardous Waste & Hazardous
Materials 1996, 13(2):265-282.

23. Xiong, F .; J .P. Croué; B. Legube, "Long-Tenn Ozone Consumption by Aquatic Fulvic
Acids Acting as Precursors of Radical Chain Reactions". Environ. Sci. Technol. 1992,
26(5): 1059-1064.

24. Petersson, G., "Mass Spectrometry of Hydroxy Dicarboxylic Acids As Trimethylsilyl
Derivatives. Rearrangement Fragrnentations". Organic Mass Spectrometry 1972, 6:

565-576.

86

25. Draffan, G.H.; R.N. Stillwell; J .A. McCloskey, "Electron Impact-Induced
Rearrangement of Trimethylsilyl Groups in Long Chain Compounds". Organic Mass

Spectrometry 1968, 1: 669-685.

CHAPTER 4
THE OZONATION OF BENZ[a]ANTHRACENE IN ACETONITRILE/WATER

MIXTURE: PATHWAY AND PRODUCT IDENTIFICATION

ABSTRACT

The pathway for the ozonolysis of benz[a]anthracene (BaA), a polycyclic aromatic
hydrocarbon (PAH), has been described in this paper. Benz[a]anthracene was dissolved in
a 90 % acetonitrile : water (v/v) mixture to achieve an initial concentration of 1 mM. The
ozonolysis of benz[a]anthracene was evaluated using different ozone dosages. Gas
chromatography followed by mass spectrometry (GC/MS) was used to separate and
identify ozonation products. Fifieen products including seven pairs of isomers were
identified. Ozone reacted simultaneously by bond and atom attack with
benz[a]anthracene. Ozone attack on the carbon at the 7 and/or 12 position produced the
quinone or hydroxyl functional groups. The bond attack type of reaction occurred at the
5,6 position and caused ring cleavage resulting in phenyl-naphthyl type products. Free
radicals appeared be involved in the bond attack type of reaction of benz[a]anthracene.
Ozonolysis of benz[a]anthracene occurred preferentially by bond attack rather than by the

atom attack type of reaction.

87

88

4.1 Introduction

Polycyclic aromatic hydrocarbons (PAHs) are found in many petroleum products
including crude oils, motor oils, gasoline and heating fuels. The predominant routes by
which PAHs enter the environment are petroleum spillage and the release of PAHs during
the combustion of petroleum products with subsequent wet or dry deposition during rainfall
or fallout (1). As a result of the release of PAHs to the environment, these compounds find
their way into drinking waters, soils, plants and wastewater. Preliminary work on the use of
gaseous ozone for the decomposition of PAHs in soils have revealed very promising results
(2). It has been shown that ozone is capable of degrading several PAHs (including pyrene,
chrysene and phenanthrene) in soils. Numerous applications exist for ozonation in the area
of wastewater treatment, including the processing of shale oil (3). However, it is still

unclear what products are produced from the ozonation of PAHs either in soils or water.

Ozone is highly reactive with the PAHs. The reaction rate constants of ozone with
PAHs range from 5 x 102 to l x 10’ M"s". The products formed fi‘om the reaction of ozone
with many PAHs have been identified prior to when gas chromatography/mass
spectrometry (GC/MS) was extensively used (4-9). Products were identified by mixture
melting points or by comparison with published infi'ared spectra. In most of the published
literature, the ozonation products were generated either in organic solvents, such as
methylene chloride, methanol, and chloroform, or in acetic acid. Nevertheless, only a
limited number of products have been identified using these systems. While several

mechanisms for the reaction of ozone with PAHs have been proposed, these mechanisms

89

are questionable since they are based upon the inconclusive identification of products and

results obtained in organic solvents.

In this study, benz[a]anthracene was chosen as the model PAH compound.
Benz[a]anthracene, a known carcinogen, is found in petroleum in concentrations of 1.2 to
90 ppm. The objective of this study is to identify the ozonation products of
benz[a]anthracene by GC/MS and to propose a pathway for the reaction of
benz[a]anthracene with ozone in aqueous solutions, and to verified results with the
previously published results identified by melting points or infrared spectroscopy (7-13).
The ozonation products obtained after benz[a]anthracene had reacted to varying extents
were identified. Also, the relative concentrations of benz[a]anthracene and its ozonation

products were quantified.

4.2 Methods
Ozonation experiments

Ozone was generated in dried oxygen by electric discharge using a Polymetrics
Model T-408 ozone generator (San Jose, CA). Gaseous ozone was continually bubbled
into a 125 mL gas washing bottle containing 100 mL of 1 mM benz[a]anthracene (99%
purity, Sigma Chemical Co., St. Louis, MO) in a acetonitrile/water mixture (90/10
volume ratio). As evidenced in a previous study (I 4), 10% water (volume ratio) was
sufficient to act as a participating solvent. Therefore, a 90% acetonitrile/water (v/v)

mixture was chosen for all the experiments. Acetonitrile was chosen as the organic

90

solvent because it is miscible with water and it has a low reactivity with ozone (tl,2 2 18

years at pH 7 and [5;] = 20.8 mM) (15). Phosphoric acid used to lower the pH. The pH

of the reactor solution was approximately 3.9.

Influent gaseous ozone was continually bubbled into the gas washing bottle
containing the benz[a]anthracene solution. The solutions were mixed with a magnetic
stirrer and stir bar. Effluent gaseous ozone was trapped in an aqueous 2% (w/v) KI
solution. The flow of ozone was regulated at 120 mL/min with a Sidetrack flow controller
(Sierra Instruments Inc, Monteray, CA). All the tubing (i.d. 1/8"), connectors and valves
were constructed of Teflon®. The concentration of ozone in the influent and effluent gas
stream was measured spectrophotometrically at 258 nm using a UV-vis
spectrophotometer (Model 1201, Shimadzu Scientific Instruments, Japan). The molar
absorptivity coefficient for ozone is 3000 M" cm’1 (16). The flow cells were quartz
cuvettes with a path length of 0.2 cm. An influent gaseous ozone concentration of 0.5
mM was used in all experiments. Reactions were terminated by flushing the reactor with
helium; 2 min was required to purge ozone to non-detectable levels. After ozonation, 2 g

of NaZSO3 was added into the solution to remove the residual organic radicals.

Identification
Separation and purification of the ozonation products were attempted using
reverse phase medium pressure column followed by thin layer chromatography (TLC).

However, this procedure failed because the silica on the TLC surface catalyzed the further

91

oxidation of PAHs by any free radicals present (see Appendix C). Ozonated samples were
then prepared for direct separation and identification using GC/MS. Ozonated samples were
dried under helium. The residue was further dried over P20, in a vacuum desicator. The
completely dried sample was derivatized by silylation using bis-trimethylsilyl/
trifluoroacetarnide (BSTFA) + 1% of trimethylchlorosilane (TMCS) (Regis Technologies
Inc., Morton Grove, IL) at 100 °C for one hour to convert all fi'ee -OH and -COOH groups

into their volatile TMS-ether (-OSiMe3) and TMS-ester (-COzSiMe,) derivatives,

respectively.

GC/MS was performed using a JEOL AX-505H double-focusing mass
spectrometer coupled with a Hewlett-Packard 5890J GC (Norwalk, CT). A DBSMS (30 m
length x 0.32 mm i.d. x 0.25 pm film thickness) fused silica capillary colmnn (J&W
Scientific, Rancho Cordova, CA) was employed for GC separation. A splitless injector was
used with a column head pressure of 10 psi using helium as the carrier gas, producing a
flow rate of ca. 1 mL/min. The initial column temperature was held for 2 min at 100 °C,
ramped at 20 °C/min to 220 °C, then ramped at 5 °C/min to 280 °C, and finally ramped at
20 °C/min to 300 °C. The mass spectrometer was operated in electron impact mode. Mass

calibration of the spectrometer was performed using perfluorokerosine.

Quantitation
The relative concentrations of benz[a]anthracene and the ozonation products were

quantified by determining the area response of each compound relative to an internal

92

standard. The same GC/MS as previously described was used. The internal standard, 1,8-
naphthalaldehydic acid (97% purity, Aldrich Chemical Co., St. Louis, MO), was
dissolved in pure acetonitrile to obtain a stock solution having a concentration of 1 mM.
The sample solutions were prepared by adding 50 [IL of the stock solution containing the
internal stande into 100 uL of each of the ozonated solutions prior to the drying step.

The samples were, then, derivatized as described above.

4.3 Results

The ozonation products that have been identified are illustrated in Figure 4.1.
Fifteen products including 8 pairs of structural isomers were identified. Typical GC
chromatograms for ozone dosages of 0.05, 0.11, 0.20 and 0.42 mole are provided
(Figure 4.2a, 4.2b, 4.2c and 4.2d, respectively). The GC/MS characteristics of each
product are given in Table 4.1. Library searches of standard spectra in NIST MS Search
(ver 1.5, 1996) and BenchTop/BPM Search (ver 3.10d, Wiley & Son, 6‘" ed., 1994) were
conducted. Matches of TMS-derivatized mass spectra of compounds V, X, XIII, XV, and
XVI were obtained. The CAS registered numbers for these compounds are listed in Table

4.1.

The spectra of compounds XIV and XVI were compared to that of the
commercially available preparations of 2-carboxybenzaldehyde and 2-phthalic acid (97%
and 99.5%, respectively. Aldrich, Milwaukee, WI). The tentatively assigned
identifications of the other ozonation products were based on interpretations of the mass

spectra and comparison with other known spectra from library searches.

93

Figure 4.] Products formed from the ozonolysis of benz[a]anthracene

94

EOOU

m. < EX m < an
IOOU OED

mi. 5 Ex

.66 .m6 ”6.6

6.6 6.6 6.6 .66. .66.

€465
OED

m<=>

EOOU
OED

:80

6.6 .8666 6.6 66.

CEO MONEU

.66. 666 66.

95

GC Retention Time (min)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o 12 14 16
10066.-- '. --‘.---4 ‘IAL -.-+
I (a)
A 1
as '°~
v a
g 60:
“g .
3
.D
< 40.
3 1 11A
.g ‘
2 2°: xm
1 xvr / v1.4.3
‘ A _l I
‘ c L .
a
A I
a? 80.
V
° 1
6 ,0.
'8 -r
53' .
0 4o.
.2. j
a 1
3 20. x" N
‘xrv
1 ”51% "“1
roo°- - -
1
A I
as so.
v 4
3 ‘ "
so
‘ anB
A
E V" 3 6.6
Q
< 1
0
1:): ms
3 III
e \ 6..
x' -
A I (d)
a? so.
V ‘ xvr
3 3 VIIAB
6 ‘°‘
‘2 3 v
g ‘0‘ D‘AJ
3- : A's xrr 93
o 20. VI
M ‘ |\XV [v
o F
403 600 6'0 rooo

 

Scan Number

Figure 4.2 Typical GC chromatogram. Figures (a), (b), (c), (d) correspond to the
reaction of benz[a]anthracene (0.1 rnmole) with 0.05, 0.11, 0.20 and 0.42
rnmole of ozone, respectively. The Roman numerals corresponds to the

products listed in Figure 4.1.

SN .2 N .mom n): so 538on mass. 53388 a 3..
0333.8: mzh 38:2: 2: mo Ems; 838—08 05 so :33 03? _8_~28£ _..

 

 

 

3m :2 S m 9:. $3: 3v 2:. :8 =5: E

am am 2 m as $3: 2:. 3v :8 3.2 E

8_ SN 2 m a $3: 3m ”3 ea .22 mg. 5

EN EN 8m Rm $3: ”3 3m :2 =82 m.< =x

:8 :2 am am $3: 2m 2 m V: 23.2 =

Me am 2 m 3 $3: 3m 3m EN :82 m.< 5

2m _8 2 m 5 $3: :3 3m :2 392 m.< in

333% z: 0: SN :3. $3: ”2 mm ”2 am: x

SN m _ m 5 2 m $3: xx 3m :2 =23 m.< =>

mom a: 5. $3: on. o? mom =23 :5

0% :8 am am $3: 2m :2 3m 25.2 =

0: EN EN SN $3: :8 SN :8 3.32 m.< an

6 2m ME mom : m $3: mmm mmm 8m 3.: m.< S

9 :NiASE SN 2 RN 2 m $3: 3m 3m EN :92 m.< =>

9: SN 2 am $3: :3 SN :8 =82 m.< an

93-222 z: m: mom i $3: 8m 8m SN 1 :2 >

30% z: E SN $3: ”mm ”mm mm .8: e

mg on 8m 2 m $3: «mm gm :3 =3: m.< 5

EN EN 3m 2 m $3: vmm 3H SN :9: Z

0-3-22 3: S N MR 23 $3: 2; 3m :2 5: Ex

33:8 me am mg mg $3: 2 m S m :2 :83 5x

m2 g E SN $3: NNN N8 :2 =5: m.< 2x

$2-32 m2 MS 08 SN $3: SN SN ”2 =3.“ >x

m2 m2 0: SN $3: mmm mmm :2 exam m.< Ex

& .wom m<o HE .830: :2 E8595 03:35: 9:59:00 2:: A:

seam 2mm m—ZH. 30an Got—~30.— «0360.5
:0 2303 338—05 00

 

3262a cogeono ocoomEEaENeon :0 83380820 mgUO 3‘ 033.

97

For example, the identical ions characteristic of 5,6-dihydrochrysenediol (see Figure
4.3A) listed in NIST library were compared to the mass spectrum of the TMS derivative
of product VII A,B (see Figure 4.33). The mass spectra and the interpretations of these

spectra of all other products are provided in Appendix D.

Two types of ozonation products were found. Compounds IIA, IIIA, and IV to
VIII are phenyl-naphthyl type ozonation products. The major product of this type was 2-
(2’-formyl)phenyl-3-naphthaldehyde (V). Compounds [[8, IIIB, and IX to XII are
quinone type ozonation products. The major product of this type was
benz[a]anthraquinone (X). Lower molecular weight products, such as dihydroxy-
naphthalene (XIII), 2-carboxybenzaldehyde (XIV), 2-hydroxybenzoic acid (XV) and 2-

phthalic acid (XVI), were also found.

The effect of the ozone : benz[a]anthracene (BaA) ratio (reported as moles
ozone/mole BaA) on the distribution of the products and their relative concentrations is
shown in Figure 4.4. Benz[a]anthracene was oxidized using ozone at stoichiometric ratios
ranging from 0.29 to 9.88 moles ozone/mole BaA. In all experiments, the initial BaA
concentration was 1 mM. The concentrations of the products are given as the ratio of the
GC area response of the analyte (A) relative to that for 1,8-naphthalaldehydic acid (50 uL
of 1 mM), which was used as an internal standard (AI). Due to their low concentrations,
products VIII and XI could not be quantified throughout all experiments. At an ozone :
BaA ratio of 0.29 mole ozone/mole BaA, products IIA, V, VI and VII, which are phenyl-

naphthyl type products, along with quinone type products IIB, IIIB, X and X11 were

98

Figure 4.3 (A) Mass spectrum of 5,6-dihydrochrysenediol obtained from NIST library
search. The abundant ions m/z 202, 215 and 231 are characteristic of 0,0’-
disubstituted phenyl-naphthyls.

(B) Mass spectrum of an ozonation product, 2-(2’-formyl)phenyl-3 -naphthoic
acid or 2-(2’-carboxy)phenyl-3-naphthaldehyde (V II A,B), derivatized as the
trimethylsilyl ester using BSTFA + l % TMCS. The phenyl-naphthyl
backbone was evidenced by the presence of triplet ions m/z 202, 215 and
231, where m/z 215 was formed through a Me3Si- migration mechanism
(cf:Ref 18,19) followed by loss of CO2 and Me3SiO (= 44 +89 = 133) from
the molecular ion m/z 348.

© . © WC), <A>

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2 2“ 215
100.l 2415231
262
78
5° ‘ 244
202
39 52 p 108 189 l
O .J l I . . . . l1 "s l i .
10 190 220 250
1+
M Si tion
003100 *3 r
100
2115 (B)
ga 80«
o I
'0
E: 60‘
'2 < 202
= a
g .
40.
319
.E’ : 35
H
2 .
a 20« L
4 101
, ‘ 348(M+.)
. 33 144
o . . . . . .
so 100 150 200 250 300 350 400

M/Z

100

Figure 4.4 The relative concentrations of benz[a]anthracene and its ozonation products
obtained under different ozone : BaA ratios. The product concentrations were
determined by the GC area response for compound (A) relative to the
response for 1,8-naphthalaldehydic acid (50 uL of 1 mM) used as an internal
standard (AI).

101

96d[03 ]/[BaA]

 

 

l I,
I (
Ell-lulu, HI» ”5..
III II!)

”(I ..
(5.,

       
 
  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

102

detected. Also, at this ozone dosage, products XIII and XVI had already formed. As the
ozone : BaA ratio increased, the concentration of all products present increased. When the
ozone : BaA ratio was 1.08 moles ozone/mole BaA, the concentration of BaA decreased
dramatically. Additionally, product IIA disappeared, and products IIIA, IV, XIV and XV
were found. At an ozone : BaA ratio of 1.99 moles ozone/mole BaA, the concentration of
compounds V and X reached their maximum values, and the relative concentration of
BaA decreased to 0.03. At this time, compound V predominated. Upon further increases
in the ozone : BaA ratio, the concentrations of all compounds present (except products
XIV, XV and XVI) began to decrease. The concentration of product V decreased faster
than that of product X, and product X became predominant. At an ozone : BaA ratio of
2.96 moles ozone/mole BaA, the parent compound and product XIII disappeared. Upon
fm'ther increases in the ozone : BaA ratio (to > 5.50 moles ozone/mole BaA), the
concentration of product V decreased to below its detection limit. At this ratio, compound
XVI predominated. At an ozone : BaA ratio of greater than 8.20 moles ozone/mole BaA,

only products XIV, XV and XVI were detected.

Several of the products identified appeared to have resulted from the oxidation of
benz[a]anthracene by free radicals, most likely by -OH. These include minor products, 11
A and B, IIIA and B, IV and XIII. To prove this hypothesis, experiments were performed
in pure acetonitrile with the ozone dosage varying from 0.05 to 0.50 mole. Product IIA
was not produced, and product IV was not detected until the ozone dosage had increased

to 0.50 rnmole. Although product IIIA was detected, it was found in low concentrations.

103

As observed in the acetonitrile/water mixture, products IIB, IIIB and XIII were also

detected in pure acetonitrile when a low ozone dosage were used.

4.4 Discussion

Ozone is known to react with aromatic compounds resulting in either substitution by
atom attack or ring cleavage by bond attack. The location of the attack of ozone can be
correlated with the lowest atom- and bond- localization energies of the target compound.
Based upon the products identified, along with the sequence of their appearance, a
pathway for the ozonation of benz[a]anthracene is proposed (see Figure 4.5). Both atom
and bond attack occurred during ozonation. Atom attack by ozone on the carbon at the 7
or/and 12 position produced compounds with hydroxyl functional groups, e. g., IIB and
IIIB. Upon further ozonation of these compounds, benz[a]anthraquinone (X) was
produced. Atom attack by ozone at the 7 and 12 position resulted in a trioxalane ring
structure; product IX and benz[a]anthraquinone (X) were produced. This pathway was
consistent with that proposed by Bailey (8). Upon further ozonation, the hydroxyl and/or
aldehydic functional group was oxidized to form a carboxylic acid, resulting in the
formation of minor products XI and X11. Additionally, the ketone bond can be cleaved
resulting in formation of low molecular weight compounds, e.g., XIII, XIV, XV and XVI.
Products IIB, IIIB, IX, X and XVI are consistent with products presented by previous

researchers (9-13).

104

Figure 4.5 Proposed pathway for the ozonation of benz[a]anthracene. Compounds
shown in brackets [ ] were detected at concentrations which were below
detection limit.

105

TewooooE> © 3.5 zeg+n ”6+ 256%! .86

© 88 no 3:?
.88

M7 a; =

zwmw 3.5 © © 3.?

\©© J.M.—w. \noQ . My. cam .

©.©WUEH§/ 8H ©©. I Q 4/8fi \“QQ. ©

\3 .7, .89 8%.
@8/ . W8
.rfim8iW8- .%8 \...

 

 

(=—

106

The bond attack type of reaction occurred on the bond having the lowest energy,
i.e., at the 5,6 position and resulted in ring cleavage. Benz[a]anthracene reacted to form a
dialdehyde, e.g., product V (2-(2’-formyl)phenyl-3-naphthaldehyde). An internal
rearrangement involving the dialdehyde groups on V could result in ring closure with
ketone and hydroxyl functional groups substituted on the cyclic ring (product VI). Upon
further ozonation, the aldehydic functional group was oxidimd to form a carboxylic acid
(product VII and VH1). However, further ozonation could also result in cleavage of
phenyl-naphthyl type products at the bond between the phenyl and naphthyl groups to
produce dihydroxy-naphthalene (XIII) and other low molecular weight compounds. Since
the ozonolysis of benz[a]anthracene would be expected to result in the direct formation of
products such as V, IX and X, the appearance of low molecular weight compounds, such
as XIII, XIV, XV, XVI, at low ozone dosages substantiated the hypothesis that these

compounds can be produced by the direct ozonolysis of products V, IX and X.

Free radicals may also be involved in these reactions to form phenyl-type
compounds. When benz[a]anthracene was ozonated in pure acetonitrile, the product IIA
did not form and the concentration of product IIIA and IV decreased (as compared to that
observed in 90% acetonitrile/water solution). The hydroxyl radical could attack at the 5 or
6 position, resulting in the formation of products 11A and IIIA. Also, hydroxyl radical
attack, with the addition of oxygen on benz[a]anthracene, followed by dehydrogenation
would result in the formation of product IV (I 1). Upon further ozonation and
dehydrogenation of -CH20H on product IV, a dialdehyde would form, i.e., product V. In

pure acetonitrile, product IIIA was produced at lower concentrations (as compared to that

107

observed in 90% acetonitrile/water solution) and product IV was not produced until the
stoichiometric ratios were greater than 3.77 moles ozone/mole BaA. These products may
result from the reaction of organic peroxides, generated during the ozonation auto-
decomposition chain reaction (I 7). Free radicals can also react with 2-phthalic acid
(product XVI), resulting in the formation of 2-hydroxybenzoic acid (product XV) (1 1-

13).

A comparison of the effect of ozone demand on the formation of products V (the
major product from bond attack pathway) and X (the major product from atom attack
pathway) is shown in Figure 4.6 (determination was done as shown on page 62).
Although bond and atom attack occur simultaneously from the beginning of the reaction,
the bond attack reaction occurs preferentially. Before benz[a]anthracene completely
disappeared, products V and X reached their maxinmm concentrations. However, the
maximum concentration of product V was 25% higher than product X. Upon further
increasing the ozonation dosage, product V appeared to decrease more rapidly than did
product X. Prior to the disappearance of product X, all of the phenyl-naphthyl type

products formed from ozone attack had disappeared.

108

0.70 ~—
0.60 +
0.50 4—

0.30 r % ~

0.20 “" \

0.10 “I \L
m

0.00 i
0.00 0.20 0.40 0.60 0.80 1.00

NA]

 

 

 

Ozone Demand (mmole)

Figure 4.6 The effect of ozone dosage on the concentrations of two major products, 2-
(2’-formyl)phenyl-3 -naphthaldehyde (V) and benz[a]anthraquinone (X). The
residual concentration is determined by the GC area response for compound
(A) relative to the response for 1,8-naphthalaldehydic acid (50 uL of 1 mM)
used as an internal standard (AI). 0 BaA; A Product V; I Product X.

109

4.5 Conclusions

A pathway for the ozonation of benz[a]anthracene in acetonitrile/water mixtures
has been described. Ozonation resulted in parallel reactions by both atom and bond
attack. The atom attack at the 7 and/or 12 position produced the quinone or hydroxyl
functional groups. The bond attack type of reaction occurred at the 5,6 position and
caused ring cleavage resulting in phenyl-naphthyl type products. The bond attack reaction

occurs preferentially.

The identification of the reaction products was accomplished by mass
spectrometry. Future work will include HPLC purification, and NMR to further confirm
the product identification. Once pure fractions of the products are obtained, these

individual products will be further investigated to assess their toxicity.

110

4.6 References

. Nefi‘, J.M. In Polynuclear Aromatic Hydrocarbons in the Aquatic Environment:
Sources, Fates and Biological Eflects. London: Applied Science Publishers, Ltd., 1979,

227.

. Yao, J.J.; S.J. Masten, "Use of in-situ Ozonation for the Removal of PAHs from

Unsaturated Soils". Proc. of the 24th Mid-Atlantic Industrial and Hazardous Waste
Conference. July 14-17, 1992. Morgantown, WV, 1992, 642-651.

. Masten, S.J.; S.H.R Davies, "The Use of Ozonation to Degrade Organic
Contaminants in Wastewaters". Environ. Sci. Technol. , 1994, 28(4): 180A-l 85A.

. Moriconi, B.J.; W.F. O’Connor; F.T. Wallenberger, "Ozonolysis of Polycyclic
Aromatics. VI. Benz[a]anthracene and Benz[a]anthracene-7,12-dione. Correlation of
Quinone-Hydroquinone Oxidation-Reduction Potentials with the Positions of
Predominant Ozone Attack". J. Am. Chem. Soc. 1959, 81, 6466-6472.

. Copeland, P.G.; R.E. Dean; D. McNeil, "The Ozonolysis of Polynuclear Aromatic
Hydrocarbons. Part I.". J. Chem. Soc. 1960, 3230-3234.

. Copeland, P.G.; R.E. Dean.; D. McNeil, "The Ozonolysis of Polycyclic

Hydrocarbons. Part II". J. Chem. Soc. 1961, 1232-1238.

. Bailey, P. S.; P. Kolsaker; B. Sinha; J.B. Ashton; F. Dobinson; J.B. Batterbee,

"Competing Reactions in the Ozonation of Anthracene". J. Org. Chem. 1964, 29,
1400-1409.
. Bailey, P. S.; J .E. Batterbee; A.G. Lane, "Ozonation of Benz[a]anthracene". J. Am.

Chem. Soc. 1968, 90, 1027-1033.

9.

111

Bailey, P. S. "Reaction of Aromatic Compounds: Bond Attack on Benz-Fused
Carbocyclics". In Ozonation in Organic Chemistry, Wasserman, H. H., ed.;

Academic Press: New York, 1982, 1-469.

10. Chen, P.N.; G.A. Junk; H.J. Svec, "Reactions of Organic Pollutants. 1. Ozonation of

11.

12.

13.

14.

15.

16.

Acenaphthylene and Acenaphthene". Env. Sci. T echnol., 1979, 13:451-454.

Larson, R.A.; H.L. Ju; V.L. Snoeyink; M.A. Recktenwalt; P.A. Dowd, "Some
Intermediates in the Wet Air Oxidation of Phenanthrene Adsorbed on Powdered
Activated Carbon". Wat. Res., 1988, 22:337-342.

Legube, B.; S. Guyon; H. Sugimitsu; M. Dore, "Ozonation of Naphthalene in Aqueous
Solution - 1. Ozone consumption and Ozonation Products". Wat. Res., 1986a, 20(2):
197-208.

Marley, K.A.; R.A. Larson; P.L. Stapleton; W.J. Garrison; C.M. Klodnycky,
"Ozonolysis of Naphthalene Derivatives in Water and in Kerosene Films". Ozone:
Sci. and Eng. 1987, 9(1): 23-36.

Upham, B.L.; J .J . Yao; J .E. Trosko; SJ. Masten, "Determination of the efficacy of
Ozone Treatment System Using a Gap Junction lntercellular Communication
Bioassay". Environ. Sci. Technol. 1995, 29(12): 2923-2928.

Yao, C. C. D.; W. R. Haag, "Rate Constants for Direct Reactions of Ozone with
Several Drinking Water Contaminants". Water Res. 1991, 25, 761-773.

Bader, H.; J. Hoigné, "Determination of Ozone in Water By The Indigo Method - A

Submitted Standard Method". J. Ozone Sci. T echnol. 1982, 4, 169-176.

112

17. Staehelin, J .; J. Hoigné, "Decomposition Of Ozone In Water In The Presence Of
Organic Solutes Acting As Promoters And Inhibitors Of Radical Chain Reactions".
Env. Sci. Technol., 1985, 19: 1206-1213.

18. Petersson, G., "Mass Spectrometry of Hydroxy Dicarboxylic Acids As Trimethylsilyl
Derivatives. Rearrangement Fragmentations". Organic Mass Spectrometry 1972, 6:
565-576.

19. Drafi‘an, G.H.; R.N. Stillwell; J .A. McCloskey, "Electron Impact-Induced
Rearrangement of Trimethylsilyl Groups in Long Chain Compounds". Organic Mass

Spectrometry 1968, 1: 669-685.

CHAPTER 5

CONCLUSION AND RECOMMENDATIONS

5.1 Conclusion

The overall objective of this thesis was to gain a better understanding of the
reactions involved in the ozonation of PAHs in order to predict the products formed from
the ozonation of PAHs in waters and other environmental systems. The major focus of this
study was to investigate the pathways of the reactions of ozone with pyrene and
benz[a]anthracene in a aqueous system. The target compounds were chosen from two
different classes of PAHs to compare different ozonation pathways. An review of the
literature related to general ozone chemistry has been presented in Chapter 1. The
subsequent three chapters contain descriptions of experimental investigations of two PAHs.
Each of these chapters can stand alone and is published or currently in the process of being

published. A conclusive discussion based on these research results is provided in this

chapter.

The aqueous solubilities of PAHs are very low. Therefore, the reaction of ozone
with PAHs has been previously conducted either in dilute aqueous solution (1- 7) or in
organic solvent (8-1 7). However, in dilute aqueous solution, excess ozone was usually used.

When excess ozone is involved in reaction, only final products are observed and the

113

114

reaction pathway is not predictable (4,8). To obtain more concentrated solutions of PAHs,
most research was conducted using organic solvents. Some organic solvents, such as
methanol or methylene chloride, will react with ozone and produce organic peroxides
(1,18). This poses an additional difference in reaction products as compared to using water.
In this study, acetonitrile was chosen as the organic solvent because it has a low reactivity
with ozone, it is miscible with water, it is immediately amenable to RP-HPLC analysis,
and it has little effect gap junction intercellular communication (GJIC) which was used as
a bioassay. In the water that was present (and dissolved in acetonitrile), 'OH radicals will
form from the decomposition of ozone. Additionally, in our studies, sufficient water was
present such that water can participate in the direct ozone reactions, through hydrolysis
type reactions. As such, reaction pathways involving both the direct reaction with ozone and

the direct reaction involving -OH radical reaction were proposed.

The toxicity of ozonation pyrene products was assessed by monitoring GJIC, a
tumor promoter bioassay. The observations indicated that the ozonation of pyrene
resulted in the initial formation of products which were more inhibitory to GJIC than was
pyrene, itself. Upon further ozonation, the toxicity of the reaction products was reduced.
Therefore, identification of the ozonation products became important. By identifying
these ozonation products and understanding the reaction pathways that occur in aqueous
solutions, treatment alternatives that would minimize the formation of toxic products can
be developed. Gas chromatography/mass spectrometry (GC/MS) used with library

searches of standard spectra and the GC/MS analysis of commercially available

115

preparations provided more convincing product identification than other studies using

melting point or IR determination.

A reaction pathway involving ozone bond attack which results in the stepwise
ring cleavage of pyrene and where ozonation byproducts compete with pyrene for ozone
was hypothesized. A stoichiometric ratio of 1.68 moles ozone/mole pyrene was required
to completely destroy pyrene, indicating that the reaction of ozone with ozonation
products occurs prior to the elimination of pyrene; therefore, ozone would be consumed
simultaneously by pyrene and the initial byproducts. This would be expected to result in a
stoichiometric ratios greater than one which is the value that would be expected if ozone

reacted only with pyrene.

The initial bond attack occurred during the reaction of ozone with pyrene.
Fourteen products1 including aldehyde and carboxylic acid substituted phenanthrene- and
biphenyl-type oxidation products formed from the reaction of ozone with pyrene were
identified. Ozonation resulted in stepwise ring cleavage and in the oxidation of aldehydic
functional groups. First, ring cleavage of pyrene occurred at the 4,5 position, with
phenanthrene-type products predominating. Similar results were found by researchers
who used organic solvents (IO-12,1546); except that they observed that the main product

formed was phenanthrene-4,5-dicarboxylic acid. In our studies, 4-carboxy-5-

 

' These products were numbered from 11 to XV as shown in Figure 3.4.

116

phenanthrenecarboxyaldehyde was the main product. Upon further ozonation, we found

that secondary ring cleavage at the 9,10 position occurred.

While the reaction of ozone with pyrene resulted initially in the formation of
phenanthrene-type products, further ozonolysis reaction resulted in the formation of
biphenyl-type products. Until pyrene had completely disappeared, the phenanthrene-type
products predominated (over the biphenyl-type products). Once the phenanthrene-type
products had reacted, the biphenyl-type products predominated. This sequence of reactions
further supports the hypothesis that pyrene competes with its ozonation byproducts for

020116.

Even at pH 3.7, -0H radicals appeared to be involved in the reaction of the
primary ozonation products. This is thought to be true, because: (i) compound VIII is
likely to have been formed from the decarboxylation of an adjacent dicarboxylic acid by
-OH radicals (19); (ii) the reaction of OH radicals with pyrene followed by ozonation
would result in formation of compound V(20); (iii) the presence of ketone type products,

XIV and XV, would result from -OH radical attack on pyrene followed by dehydration.

For the ozonolysis of benz[a]anthracene, the ozonation pathway differed from that
observed with pyrene. The initial bond and atom attack appeared to occur simultaneously

in the reaction of ozone with benz[a]anthracene. Fifteen productsz, which resulted from

 

2 These products were numbered from 11 to XVI as shown in Figure 4.1

117

these two pathways, were identified. The bond attack type of reaction occurred at the 5,6
position of benz[a]anthracene and caused ring cleavage resulting in phenyl-naphthyl type
products. Bond attack (e.g., ring cleavage) resulted in the formation of product V but not
VIII, and only a trace amount of VIII. Atom attack on the carbon at the 7 and/or 12
position produced the quinone or hydroxyl functional groups. Similar products were
observed in organic solvent (8,21 -25). The Ozonolysis of benz[a]anthracene appeared to
occur preferentially by the bond attack type of reaction. Although the oxidation of
benz[a]anthracene by hydroxyl free radicals did not result in high concentrations of
products, -OH radical was found to be involved in the reaction. Hydroxyl free radicals
appeared to result in the formation of initial products (IIA and IV), and subsequent
oxidation of product IIA to form a diol product (IIIA). Also, -OH radicals can be
involved in the further oxidation of ozonation products as described in the previous
paragraph. However, our work confirms the observations made by Bailey et al. (25) that

products IIB and IIIB could be formed in the absence of -OH radicals.

In aqueous systems, -OH radicals significantly affect the ozonolysis reaction of
compounds where ozone reacts by initial bond attack. For those compounds which react
with ozone by both attack on the bond and on an atom, the influence of -OH radicals is
less important. However, in any initial atom attack type of reactions, only ozone is
involved. For examme, Bailey (26) showed that, even in the presence of hydrogen
peroxide, only initial atom attack product was observed in the reaction of ozone with

naphthacene (shown in Figure 5.1).

©©©©

0

Figure 5.1 Structure of the example PAHs

119

Where ozonolysis occurs by initial bond attack and -OH radicals is involved in the
reaction of ozonation products, ozonolysis needs to be studied in the presence of water.
The results of studies conducted in non-participating solvents can only predict the
products (or pathway) formed from initial ozone attack but can not precisely assess the
importance of -OH radicals. Although some researchers (12, 15, 16,21,22, 25) have used
participating organic solvents, any further reaction to produce non-ozonation products
need to be proven (Appendix C). The use of a mixture containing water and a non-
participating solvent, as done in this study, is advantageous over using participating
solvents such as methanol. This is because in the acetonitrile/water mixture the
importance of the free radical reaction can be investigated, yet side reactions involving

the participating solvent (e. g., methylation reaction) can be avoided.

In comparing of the reactions of ozone with different PAHs, it is apparent that the
lowest localization energy of the target compound is important. Ozone attack either on the
bond or atom depends on which of them has the lowest localization energy. For pyrene, the
lowest bond localization energy (Lb) is much smaller than the lowest atom localization
energy (L,), i.e., L, - L, = 0.45; therefore, a bond attack occurs. Other compounds which
have L, << L, also show the initial ozone bond attack reaction only. For example, with
phenanthrene, which has L, - Lb = 0.72, ozonation resulted in ring cleavage by ozone attack
on the 9,10 position (2, 26,27). Moniconi et al. (28,29) found that bond attack occurred
during the ozonation of dibenz[aj]anthracene (L, - Lb = 0.40) and dibenz[a,h]anthracene (L,
- Lb = 0.46). However, for benz[a]anthracene, the localization energy between bond and

atom is smaller, i.e., L, - Lb = 0.32, resulting in competition between bond attack and atom

120

attack. Moniconi and Salce (30, 31) also found that bond and atom attack occurred when
ozone reacted with benz[r,s,t]pentaphene. Unfortunately the L, of benz[r,s,t]pentaphene is
not available. When the localization energy for the lowest bond was closer or even larger
than the localization energy for the lowest atom, initial atom attack by ozone occurred. For
example, the ozonation of benz[a]pyrene (L, - L, = 0.13) resulted in the formation of

benz[a]pyrene-3,6-dione and benz[a]pyrene-1,6-dione (32, 33) by initial atom attack.

Based on our research, some unknown reaction pathways of ozone with previously
unstudied PAHs can be predicted. For example, the ozonation of benz[b]chrysene, which
has L, - L, = 0.21, is expected occur by both bond and atom attack. However, because the
difference between L, and L, is smaller than benz[a]anthracene, products formed from atom
attack may have a higher yield than from bond attack, e.g., quinone type of products are
expected as the major products. The ozonation of chrysene, of which our study has not
been completed due to the problems separating the products, may also be predicted. For
chrysene, although the bond and atom having the lowest energy happen to be at the same
location, and L, - L, = 0.55 (34), ozone is expected to attack on the lowest energy bond and
produce phenyl-naphthyl type of products. This precise prediction of ozone bond attack
reaction is consistent with the work done by Copeland et al. (23,24). The phenyl-naphthyl
product is similar to the benz[a]anthracene products formed by ozone initial bond attack.
Therefore, the ozonation of chrysene in aqueous solution can be expected to occur by
similar pathway involving ring cleavage as observed for the ozonation of

benz[a]anthracene.

121

5.2 Recommendations

This study identified the products formed from the ozonation of selected PAHs

using GC/MS directly. It also opened many directions for future research. We

recommend that:

1.

the separation procedure be improved for purifying ozonation products. Preparatory
TLC and reverse phase medium pressure LC resulted in further oxidation of the
products;

the ozonation pathway for chrysene be investigated using a improved separation
procedure followed by GC/MS identification;

the identified reaction products be confirmed using NMR;

the mechanistic pathway for the reaction of ozone with the purified products be
further investigated to verify the proposed pathways;

the toxicity of ozonated products be investigated;

the effect of presence of other reactive substrates on the pathway of the reaction of

ozone with the target compounds be investigated.

122

5.3 Reference

9.

. Chen, P.N.; G.A. Junk; H.J. Svec, "Reactions of Organic Pollutants. 1. Ozonation of

Acenaphthylene and Acenaphthene". Env. Sci. Technol., 1979, 13:451-454.

Beltran, F.J.; G. Ovejero; J .M. Encinar; J. Rivas, "Oxidation of Polynuclear Aromatic
Hydrocarbons in Water. 1. Ozonation".1nd Eng. Chem. Res., 1995, 34:1596-1606.
Andreozzi, R.; A. Insola; V. Caprio; M.G. D’Amore, "Quinoline Ozonation in
Aqueous Solution". Wat. Res., 1992, 26(5):639-643.

Corless, C.E.; G.L. Reynolds; N.J.D. Graham; R Perry, "Ozonation of Pyrene in
Aqueous Solution". Wat. Res., 1990, 24(9), 1119-1123.

Helleur, R.; M. Malaiyandi; F .M. Benoit; A. Benedek, "Ozonation of Fluorene and 9-
Fluorenone". Ozone: Sci. and Eng. 1979, 1 1: 249-261.

Meineke, I.; H. Klamberg, "On the Degradation of Polycyclic Aromatic
Hydrocarbons. I. Reaction Products of the Ozonolysis of Polycyclic Aromatic
Hydrocarbons in Aqueous Systems". Fresenius Z. Anal. Chem. 1978, 293, 201-204.
Dreher, W.; H. Klamberg, "On the Degradation of Polycyclic Aromatic Hydrocarbons.
III. Reaction Products of the Ozonolysis of Further Polycyclic Aromatic
Hydrocarbons in Aqueous Systems". Fresenius Z. Anal. Chem. 1988, 331, 290-294.
Copeland, P.G.; R.E. Dean; D. McNeil, "The Ozonolysis of Polynuclear Aromatic
Hydrocarbons. Part 1.". J. Chem. Soc. 1960, 3230-3234.

Vollmann, H.; H, Becker, US. Patent 2, 127,096, 1938.

10. Fieser, L.F.; F.C. Novello, "9-Methyl-3,4-benzpyrene". J. Am. Chem. Soc., 1940,

62:1855-1859.

ll.

12.

13.

14.

15

16.

17.

l8.

19.

20.

21.

123

Newman, M.S.; H.S. Whitehouse, "The Preparation of 4,5-Dimethylphenanthrene". J.
Am. Chem. Soc., 1949, 71:3664-3667.

Dessy, R.E.; M.S. Newman, "5-Formyl-4-Phenanthroic Acid (4-
Phenanthrenecarboxylic Acid, 5-Formyl-)". Org. Synth, Collect, 1962, 4:484.
Badger, G.M.; J .E. Campbell; J .W. Cook; R.A. Raphael; A.I. Scott, "The Synthesis of
4:5-Dimethylphenanthrene". J. Chem. Soc., 1950, 2326-2328.

Volhnann, H.; H. Becker; M. Corell; H. Streeck, "Beitriige zur Kenntnis des Pyrens

und Seiner Derivate". Liebigs Ann. Chem, 1937, 531:1-159.

. Danheux, C.; L. Hanoteau; R.H. Martin; G. Van Binst, "Ozonolyse Du Pyrene En

Acide Phenanthrene-4,5-Dicarboxylique". Bull. Soc. Chim. Belg, 1963, 72:289-290.
Sturrock, M.G.; R.A. Duncan, "The Ozonation of Pyrene. A Monomeric
Monoozonide Formed in Polar Solvents". J. Org. Chem, 1968, 33:2149-2152.

Van Duuren, B.L.; G. Witz; S.C. Agarwal, "Synthesis and Photorearrangement of 4,5-
Epoxy-4,5-dihydropyrene". J. Org. Chem, 1974, 39:1032-1035.

Bader, H.; J. Hoigné, "Determination of Ozone in Water By The Indigo Method - A
Submitted Standard Method". J. Ozone Sci. T echnol. 1982, 4, 169-176.

Sheldon, R.A.; J.K. Kochi, In Organic Reactions, Dauben, W.G. et. al. ed.; John
Wiley & Sons, Inc., 1972, 19, 322.

Larson, R.A.; B.J. Weber, In Reaction Mechanisms in Environmental Organic
Chemistry, Lewis Publishers, 1994, 259-261.

Moriconi, B.J.; W.F. O’Connor; F.T. Wallenberger, "Ozonisation of

Benz[a]anthracene". Chem. Ind. (London), 1959, 22-23.

124

22. Moriconi, B.J.; W.F. O’Connor; F .T. Wallenberger, "Ozonolysis of Polycyclic
Aromatics. VI. Benz[a]anthracene and Benz[a]anthracene-7,12-dione. Correlation of
Quinone-Hydroquinone Oxidation-Reduction Potentials with the Positions of
Predominant Ozone Attack". J. Am. Chem. Soc. 1959, 81, 6466-6472.

23. Copeland, P.G.; R.E., Dean; D. McNeil, "The Ozonolysis of Polycyclic
Hydrocarbons". Chem. Ind. (London), 1959, 329-330.

24. Copeland, P.G.; R.E. Dean.; D. McNeil, "The Ozonolysis of Polycyclic
Hydrocarbons. Part II". J. Chem. Soc. 1961, 1232-1238.

25. Bailey, P. S.; J .E. Batterbee; A.G. Lane, "Ozonation of Benz[a]anthracene". J. Am.
Chem. Soc. 1968, 90, 1027-1033.

26. Bailey, P. S. In Ozonation in Organic Chemistry, Wasserman, H. H., ed.; Academic
Press: New York, 1982, 1-469.

27. Larson, R.A.; H.L. Ju; V.L. Snoeyink; M.A. Recktenwalt; P.A. Dowd, "Some
Intermediates in the Wet Air Oxidation of Phenanthrene Adsorbed on Powdered
Activated Carbon". Wat. Res., 1988, 22: 337-342.

28. Moriconi, B.J.; B. Rakoczy; W.F. O’Connor, "Ozonolysis of Polycyclic Aromatics.
IX. Dibenz[a,j]anthracene". J. Org. Chem 1962, 27:3618-3619.

29. Moriconi, E.J.; W.F. O’Connor; W.J. Schmitt; G.W. Cogswell; B.P. Fiirer, "
Ozonolysis of Polycyclic Aromatics. VII. Dibenz[a,h]anthracene ". J. Am. Chem. Soc.
1960, 82:3441-3446.

30. Moriconi, B.J.; L. Salce, "Ozonolysis of Polycyclic Aromatics. XIV. Ozonation of

Pentaphene and Benzo[rst]pentaphene ". J. Org. Chem. 1967, 32:2829-2837.

125

31. Moriconi, B.J.; L. Salce, "Ozonolysis of Polycyclic Aromatics. XV. Carcinogenicity
and K- and/or L-Region Additivity Towards Ozone ". Adv. Chem. Ser. 1968, 77:65.

32. Moriconi, B.J.; B. Rakoczy; W.F. O’Connor, "Ozonolysis of Polycyclic Aromatics.
VIII. Benzo[a]pyrene ". J. Am. Chem. Soc. 1961, 83:4618-4623.

33. Bailey, P. S.; P. Kolsaker; B. Sinha; J.B. Ashton; F. Dobinson; J .E. Batterbee,
"Competing Reactions in the Ozonation of Anthracene". J. Org. Chem. 1964, 29,
1400-1409.

34. Brown, R.D., "Theoretical Investigation of Reactivities of Conjugated Bonds". J.

Chem. Soc., 1950, 3249-3254.

APPENDICES

APPENDIX A

APPENDIX A

KINETICS STUDY

A.l Introduction

In recent years, the polycyclic aromatic hydrocarbons (PAHs) have caused much
concerned among environmental scientists and engineers. The PAHs are generated and
released to the environment by human activities. Low concentrations of these chemicals

have been shown to be carcinogenic to mammals (1,2).

Ozone is a more powerful oxidant than the other oxidants used in water and
wastewater treatment. The increasing interest in using ozone for treatment is based on the
successful operations, mainly in Europe, as well as in Canada, Japan and United States
(3). Therefore, the ozonation kinetics study for the degradation of PAHs is important for

Evaluating treatment efficiency.

Only a few studies related to the kinetics of the ozonation of non-
dissociating PAHs have been published (4-8). The rate constants and
stoichiometric coefficients for the reaction of ozone with four PAHs are presented
in Table A. l. Kuo (4) determined the rate constants for the reaction of ozone and
several PAHs in water at 25 °C. Kuo (4) and Hoigné and Bader (5) determined the
reaction rate constants of naphthalene by monitoring ozone consumption. On the

contrary, Butkovié et al. (6) and Kuo (4) measured the change in the PAH (except

126

127

Table A.l Rate constants and stoichiometric coefficients

 

 

for the ozonation of PAHs
Compound Stoichiometry Coeff. Rate Law k1 Ref
M' s‘1
Naphthalene Use 2 mole O3/mole NA -flfld = 1,0) [031M] k = 843 4
3' (25°C, pH=3)
k = 9845 4
(35°C, pH=3)
k = 3202 4
(25°C, pH=5~6)
1~5 mole 03 /mole M "0E1. = ‘0‘ [0311M] kO3 = 30001600 5
Use 2.5 in this paper ’ ' (20°C, pH=2)
2 mole O3/mole NA _ 5_N _ k [0 ll N] kN = 550180 8
5" (1°C, pH=5.6)
Phenanthrene Use 2 mole O3/mole PH __o"__[M] _ k [M] k = 1.9~4.7x104 4
at (25°C,
pH=2.21~7)
- _§_[_M] k = 150001120 6
a: = k [M] (25°C, pH=l~7)
A [0% _é’_l__M_ kN = 24001200 7
calCUIate A[ M] at M [QHM] (200C, pH=7)
Pyrene - - (18~21°C, pH =?) 1
- _ o”[M] _ k' [M] k = 4000015000 6
at ‘ (25°C, pH=1~7)
Benz[a]pyrene - _ d[M] = k' [M] k = 620011000 6
at (25°C, pH=7)

=k[03]

 

128

for naphthalene) concentrations in solutions where ozone was in excess; pseudo-first
order reaction conditions were used to determine the reaction rate constants. Beltran et al.
(7) and Legube et al. (8) measured both ozone and PAH consumption at different reaction
times. They also determined the stoichiometry coefficient of the initial reaction. This
method provides a better understanding of the initial reaction of ozone with a target

compound.

In order to study the initial ozonation reaction, the stoichiometry coefficient of the
reaction of ozone with PAH was determined. Hoigné and Bader (5) state that the
stoichiometric coefficients usually range from 1 to 5. They calculated the reaction order for
aromatic compounds by assuming that the stoichiometric coefficients were equal to 2.5. In
most subsequent papers, researchers assumed that the reaction order and stoichiometry
were identical to that assumed by Hoigné and Bader (5). However, those researchers did not
confirm the reaction order and stoichiometry. Contrastingly, Legube et al. (8) determined
the rate constant for the reaction of ozone with naphthalene in organic solvent at 1 °C, pH
5.6. They found that the stoichiometric coefficient for the initial reaction step is 2, and that a
second order reaction is verified. Beltran et al. (7) found the stoichiometric coefficient for
phenanthrene to be 1. This is the evidence for ozone direct reaction with phenanthrene. In
this study, the kinetics of the reaction of ozone with pyrene were investigated. Both ozone

and pyrene concentration were monitored.

129

A2 Kinetics Model
If the decomposition of the target compounds is due solely to the direct reaction
of the compounds with ozone (and not due to reaction with secondary oxidants), then the
direct reaction of ozone with solute M can be expressed as:
M+§O,-—)M,,,, [1]
where g is the stoichiometric coefficient (i.e., the number of moles of ozone consumed per
mole of M transformed to Moxide)- The change in concentrations of reactions with time

can be written as:

dM .. ..
--d-7-k[M] [03] [2]

where m, n are the reaction order for solute and ozone. Work done by Bader and Hoigné
(5) shows a second order reaction for some non-dissociating organic compounds;
therefore, an assumption of a second order reaction was made, i.e., m = n = l. The

solution to equation [2] can be derived as:

03 . o
l L—l-[fl-=([03].-€[M].)kt [31

n
[03 ]o ' [M]
This expression can be written as:

mA=kT

where A = w , and T = ([03], —6 [M ], )t. Values for E, can be calculated as:
_ [03]o— [03] _ A[03]

— [ML-[M] _ MM]

 

5 [4]

130

If In [03] ' [M10

vs ([O3]0-§[M]O)*t is linear, the second order reaction is confirmed and
[03 ]o . [M ]

k can be determined.

A.3 Method
Kinetics Measurements

This study was conducted at room temperature; and, since ozone has lower
degradation rate at lower pH (9), the reaction was controlled at pH 2. All aqueous ozone
solutions were prepared in acidic deionized (DI) water. Phosphoric acid (85% purity) was
added to acidify the DI water to a pH of 2. Ozone was produced fiom dried oxygen using
a ozone generator (Polymetrics, Inc., San Jose, CA, Model T-408). Aqueous ozone
solutions were prepared by bubbling gaseous ozone through a one liter vessel containing

the acidified water. The aqueous ozone concentrations ranged from 0.05 to 0.3 mM.

Because of the low aqueous solubility of pyrene (2), a stock solution was prepared
in pure acetonitrile. Since a detection limit for pyrene (using HPLC/ UV detection) of 0.1
mg/l was obtained, and the reaction of ozone and pyrene in aqueous solutions can be very
fast, the kinetics experiments were conducted in solutions containing an initial

concentration of at least 10 mg/L pyrene.

To terminate the reaction, any residual ozone in the reaction solution was quenched
using indigo. Work done by Bader and Hoigné (10) has shown that indigo reacts

immediately with ozone and can be employed in this manner. The indigo stock solution

131

contained 1 mM of potassium indigotrisulfonic acid ( ~95 % purity. Aldrich, WI) in pH 2
water. As the percentage of solvent was found to affect the solubility of pyrene, the
reagent solutions were prepared so as to maintain a constant acetonitrile/water mixture
during mixing. Thus, the stock indigo solution was diluted to 0.25-0.5 mM (depending on
the initial ozone concentration). This maintained the ratio of 50% acetonitrile to pH 2

water in all reaction solutions.

The ozonation reaction were studied in a continuously flowing apparatus
(illustrated in Figure A. 1). All solutions were pumped using a FMI RP pump (Model
RH/QG, Fluid Metering, Inc., Oyster Bay, NY). Acidified water was continuously
pumped into the ozone contactor to maintain a constant concentration of ozone in the
stock solution. Reactants were continuously pumped through a Y-shaped mixer. The
reaction retention times were controlled by the volume contained in the reaction coil. The
reaction was quenched by mixing the solution in the reaction coil with the indigo stream
in the second Y-shaped mixer. The residual ozone concentration was determined by the
indigo method (10). At the end of the system, the reaction solution was collected, and

residual pyrene concentration was determined using HPLC.

Analytical Techniques

Aqueous Ozone. Aqueous ozone was withdrawn from the ozone contactor using a gas-tight
glass syringe. In solutions that did not contain pyrene or the reaction products, the ozone
concentration was determined directly using UV spectrophotometer at 258 nm. A dio-array

spectrophotometer (Model # 8452A, Hewlett Packard, Palo Alto, CA) was used for this

132

Figure A.1 Configuration of kinetics apparatus

 

_’

 

 

:OU comgomum W ”

ease.
coziom 03?: .o__:_=o.ou< 5 each»;

 

 

 

0.5: o. 0%

336223.8on

8.22.50 0:80 N u :a .835 3523.

O O

1 a / .

133

 

 

 

 

 

 

 

13.3-20

 

 

 

 

 

 

 

 

 

 

 

80 cacao

 

134

purpose. An extinction coefficient of 3000 M‘1 cm" was used to convert the absorbance
units to concentration. In solutions containing pyrene or the reaction products, ozone was
measured using the indigo method (10). The final ozone concentration was determined by
the decrease in the absorbance at 600 nm as compared to the blank which contained no

OZOI'IC.

Pyrene Concentration. Aliquots of the sample solutions were collected before and after

ozone reacted with pyrene. HPLC was carried out using a Gilson system equipped with a
5pm 4.6 x 250 mm Partisphere RP C18 column (Whatman, Clifton, NJ). An UV detector

(Model No. 116) was used to measure the absorbance at 240 nm. The mobile phase

contained in 85% acetonitrile and 15% water mixture flowing at a flow rate of 1.5 ml/min.

A.4 Results and Discussion

The stoichiometric coefficient for ozonation of pyrene in aqueous solution can be
calculated according to Equation [4]. The experimental results for 5 are shown in Figure
A.2. Using a linear regression of all the data points, the stoichiometric coefficient, 1;, was
determined to be 2.0 1 0.02, r2 = 0.92. Based upon these experimental results, one pyrene
reacted with two ozone. However, based upon previous research (1 1), except when ozone
was present in excess (12), only one bond attack ozonation products are produced. As
such, 1; would be expected to be one. The fact that we obtained E = 2.0 suggests that
ozone was consumed either by self-decomposition of ozone or by the reaction of ozone

with products formed by the initial one bond attack on pyrene.

135

0.12 T

! ‘ A
0.14; . 8

l n
0.08 f

T

d[03]0.06 4
0.04 " g m

0.02 *”

 

0 0.01 0.02 0.03 0.04 0.05 0.06

Figure A2 The stoichiometric coefficient for the reaction of ozone with pyrene. Different
symbols indicated different set of data; solid line is linear regression result with r2 =
0.92.

136

When equation [3] was applied, the reaction rate constant, k, could be obtained by
plotting lnA vs T (shown as Figure A.3). A reaction rate constant, k of 3.718 1 0.066
mM'1 3", r2 = 0.96 was observed. On the contrary, Butkovic' et al. (6) obtaine a rate
constant of 40000 :I: 5000 M’1 s" for the reaction of ozone with pyrene. The large
difference in values of k may because in their studies ozone was in excess and they only
measured the change in the concentration of pyrene. However, any organo peroxide or
-OH free radicals generated during the reaction may react with pyrene, itself. As a result,
an over estimate of the rate constant may have occurred. In this study, both ozone and
pyrene concentrations were monitored. However, since the reaction between ozone and
pyrene is very fast, the ozone consumption measured during the experiment may not only
be due to the reaction of ozone with pyrene but by any intermediates formed along with
ozone self-decomposition. Therefore, the assumption that the decomposition of pyrene is

due solely to the direct reaction of ozone can not be applied in this case.

In order to get more accurate results for ozone direct reaction, a shorter reaction
time is required for this fast reaction. As the equipment required to do these experiments,
such as stop-flow reaction, was not available, we were not able to obtain the desired kinetic

data.

137

lnA

 

 

 

T (mM 5)

Figure A3 The rate constant for the reaction of ozone with pyrene in aqueous solution. A
[03]. [M]o
lotoflnAvsT. Where A=————- ,and T: O 0— M o t.
p [03]0‘[M] (I: 3] ;[ ] )
I = calculated R from each data point; solid line is the linear regression result
(r2 = 0.96).

138

A.5 References

. Neff, J .M.. In Polynuclear Aromatic Hydrocarbons in the Aquatic Environment:
Sources, Fates and Biological Eflects. 1979. Applied Science Publishers Ltd,
London.

. Futoma, D.J., S.R. Smith, T.E. Smith, J. Tanaka. In Polycyclic Aromatic
Hydrocarbons in Water Systems. 1981. Boca Raton, Fla.: CRC Press. pp. 1-1 1.

. F ochtrnan, E.G., R.G. Rice, M.E. Browning E. In Forum on Ozone Disinfection.
1977 . International Ozone Institute, NY.

. Kuo, CH. and HM. Barnes, In Reactions of Ozone with Organics in Aqueous
Solutions. US. EPA Report No. EPA/600/3-85/031. 1985, 66.

. Hoigné, J.; H. Bader, "Rate Constants of Reactions of Ozone with Organic and
Inorganic compounds in Water-I Non-dissociating Organic Compounds". Wat. Res.

1983, 17, 173-183.

. Butkovié, V.; L. Klasinc; M. Orhanovic; J. Turk; H. Glisten, "Reaction Rates of

Polynuclear Aromatic Hydrocarbons with Ozone in Water". Environ. Sci. Technol.
1983, 17(9), 546-548.

. Beltran, F.J.; G. Ovejero; J .M. Encinar; J. Rivas, "Oxidation of Polynuclear Aromatic
Hydrocarbons in Water. 1. Ozonation". Ind. Eng. Chem. Res., 1995, 34:1596-1606.

. Legube, B.; S. Guyon; H. Sugimitsu; M. Dore, "Ozonation of Naphthalene in Aqueous
Solution - 1. Ozone consumption and Ozonation Products". Wat. Res., 19863, 20(2):

1 97-208.

139

9. Gurol, M.D.; P.C. Singer, "Kinetics Of Ozone Decomposition: A Dynamic
Approach". Environ. Sci. Technol. 1982, 16(7):377-3 83.

10. Bader, H.; J. Hoigné, "Determination of Ozone in Water By The Indigo Method - A
Submitted Standard Method". J. Ozone Sci. Technol. 1982, 4, 169-176.

1 1. Bailey, P. S. In Ozonation in Organic Chemistry, Wasserman, H. H., ed.; Academic
Press: New York, 1982, 1-469.

12. Corless, C.E.; G.L. Reynolds; N.J.D. Graham; R Perry, "Ozonation of Pyrene in

Aqueous Solution". Wat. Res., 1990, 24(9), 1119-1123

APPENDIX B

APPENDIX B
MASS SPECTRA AND INTERPRETATION OF

SILYLATED PYRENE OZONATION PRODUCTS

In this appendix, the raw spectra obtained from GC/MS analysis is followed by
the interpretation of these spectra for pyrene ozonation products. Products were silylated
by BSTFA + 1% TMCS and heated at 100°C for 20 min prior to GC/MS analysis. The

silylation reagent converts -OH or -COOH functional groups to -OSiMe3 or -COOSiMe3.

Pyrene mass spectrum from library search in NIST MS Search (ver 1.5, 1996) is
illustrated to allow the comparison of the raw mass spectrum obtained fi'om GC/MS
analysis with published spectra. The commercial compound, 4-carboxy-5-
phenanthrenecarboxyaldehyde (4-CPA, Sigrna-Aldrich Library of Rare Chemicals,
Milwaukee, WI), was prepared so as to compare its GC/MS spectra with that of
compound 111 A13. From the GC chromatographs and mass spectra of 4-CPA before and
after TMS derivative was found to exist in two isomeric forms which were identical to

those shown in compound III AB.

140

I41

 

 

 

0 OH HO OHC COOH
I (202) [I (234) III MB (250) IV (222)
O
OHC coon O—C”
V (252) V1 (206) VII (222) VIII (220)
OH CHO OH OOH OH HO
OHC CHO OHC HO HOOC OOH
[X (266) X (282) XI NE (298) XI MB (298)

   

XII (314) XIII (330) XIV (282)

Figure B] Products formed from the ozonolysis of pyrene. Roman numbers correspond
to the ID assigned to each product. Molecular weights are given in
parentheses.

142

 

 

 

Figure B.2 Common ions of mass spectra of ozonation products formed from pyrene.

143

Mass Spectrum of pyrene (I)

MASS SPECTRUM Data File: [100,1001x09209604 3-0CT-88 19:24
Sample: 1H c—BuOH 20min 100-2-20-220-5-260-20-300-5

R? 9'52' 8! (Pos.) GC 450.6c BP: m/z 202.0000 Int. 110.3524 Lv 0.00
Scanl (659) - (656, 662) Icoet. 1.00]

 

 

 

 

 

100
2132 b

R P
e l .
I 80‘ r
a
C
i
V 60‘ *
e t
A 4
b 404 _
u 4
n
d
a 20~ 101 l .—
n
c l .
e 1 73 33 i

0,-0-1, V-,gnvvivr'fn. , .. ,--

50 100 150 200 250 300

Ht’Z

Mass spectrum of pyrene obtain from NIST database (NIST#: 9063, CAS#: 129-00-0)

100, 202

GO

101

50.

88 I
I

O h: A I .1
Inrrrr rprrTiWT-rprrflrirrlr...

20 4O 60 80 100 120 140 160 180 200

 

 

 

11 11 '
IFFW‘11TFIII.TII.IYIII.IIIT?tIYTIITIIIIIIIII‘TITTTIVII‘ '1

144

Mass Spectrum of the silylation product 11

MASS SPECTRUM Data F116: [100,IOOIX09169405 16-589-94 14:58
Sample: BSTFA‘485 DBSMS 100-2-20—1
RT 9'53” 81 (P03.) GC 450.6c BP: m/z 205.0000 Int. 184.4544 Lv 0.00

Scanl (660)

 

 

 

 

 

 

100
205 .
R b
a l i
1 80* r
a r
t b
L P
v 60' L
8 L
‘ 2 4 290 '
b
‘ ,0. 220 .
u" 275 >
n 1 t
d < -2o.o >
9 20: r
n r
C 88 >
. 75AJ 102 >
1 1 __ _A 11 1 .

U r *fi * r r r r v fr # f r ' V ' r ' ' ' ' Y ' f ' '

so 100 250 300 350 400

I45

CH0 CH0 ‘1?

234 205 177
\ILHO
23 l co I 29
CHO
206

 

 

Figure B.3 Interpretation of the mass spectrum for the silylation product II.
Molecular ion weight = 234.

146

Mass Spectra of the silylation product III A,B

MASS-SPECTRUM Data File: [100,1001x09209612 3-OCT-88 22:27
Sample: In CflJCN 2min 100-2-20-220-5-260-20-300-5

RT 13°19“ 21 (Pos.) GC 450.6c 8?: m/z 205.0000 Int. 1200.2550 Lv 0.00
Scan! (890) - (880, 892) (coat. 1.00]

 

 

 

 

 

 

 

 

 

 

100
205 293 r
a D
e ‘ ’
I 80‘ -
a l ’
t. ‘ .
i
v 60* '
e 73 ’
A 322 ’
b ‘04 I” 307 "
u t l
n < r . T44 . -*< .P
d * 176 -10.o .
‘ 20* '
n r
c 151 263 >
LA 11... .11 A. L A .L
uri' ' f ' r r * fir W ' '1 r V ' ' * T ' f' ' fi' ' ' ' I ' ' ' ' I

50 100 150 200 250 300 350 400
H/Z

HASS SPECTRUH Data File: [100,1001x09209612 3-OCT-88 22:27

Sample: in CHJCN 2min 100-2-20-220-5-260-20-300-5
R? 14°12“ El (903.) CC 450.6c 89: a]: 189.0000 Int. 1104.9910 Lv 0.00
Scan. (949) - (937, 951) (coat. 1.00)

 

 

 

 

 

 

 

 

100
139 .

R l
e ‘ r
1 so: _
a
c p
i ‘ 294 .
v 60* -
e l

t >
A . .
b 40‘ 3 9 -
u 73 .
n : - . . . . , v . . ..
d 176 2 5 3 2 -1o.o .
a 20'1 2 3 279 -
“ 263 3 1 r
c ‘ 1 1 r
e 1 1 r

0' V; ‘ ' r 'f V'r'fi' fi' 1 ng 1 ' *V

so 100 150 200 250 300 350 400

I47
+
CH0 COzSiMe3—l' _ CH0

322

lSi migration

. e
OSiMe 3 O“ Q
, /
0..

000

74l-OSiMe2
(9

-USiMe3
——>
8’ O
189

Figure B.4 Interpretation of the mass spectra for the silylation product III A,B.
Molecular ion weight = 322.

 

148

GC chromatograph of 4-CPA in pure solvent. The insert shows the mass spectrum of the
peak. ‘

TIC - Data File: [100,1001X09279603 27-SEP-96 12:26
Sample: 4-CPA in CHZCIZ OBS I00-2-20-220-5-260-20-300-5

Scan! 500 to 1200(1200) RT 7'29” :0 17'58“(17'58”) EI(P0:.) Lv 0.00
Operator: OQ

 

 

 

 

 

 

 

 

 

 

 

 

Max 5565.66 R.T.
10 12 14 mag. abund.
100 9 ‘ ‘ * ‘ ‘ : 1 : - L 1 - 1F - - a
A 1‘ IC
D CAI! I'tCTIUI Data '1101 1100.100|109219601 2,-800-90 12:10
tangle: new. In CIICII 00$ 1.0-l-10-110-0-100-10-100-3
u I? )‘11‘ at 1000.) 6C 090.60 I'; III 100.0000 lat. 010.1060 L: 0.00
Ira-I um - 16H. “11 (cool. 1.001
n 80 -4
d 4 19. .2
a I 111
n I u
C 60‘ f
e l v to
40* Z
‘ ‘1 I, .. r
. [10
i H J I no i
20 ‘ ' ' .s.
‘ III!
1 w...»
m r-
500 ' I ' ' ' I ‘ ' ' ' ' F r '1.0 5565.66
600 700 800 900 1000 1100 1200
Scan
GC chromatograph of 4-CPA in silylation reagent
TIC Data File: [100,1001x09279604 27-SBP-96 13:00

Sample: 4-CPA In BSTFA DBS 100-2-20-220-5-260-20-300-5

Scan! 500 to 1200(1194) RT 7'29“ :0 17'58“(17'53') EI(P03.) Lv 0.00
Operator: OQ

 

 

 

 

 

 

 

 

Haa 11338.6 3.1.
9 10 12 14 16 mag. abund.
100 ‘ A ‘ l ‘ * t 1 ‘ ~ - 1 - - L 1 . - -
A + 110 I IIIA,B
b 1
0 J
0 80-1
d 4
a 4
n 1
° ‘°‘ IIIA,B
e i
40:
20:
4M
4 W
fl I ' r ' 1 ' I ,— T ' F ' 'I.0 11338.6
50° 600 700 800 900 1000 1100 1200

Sean

149

Mass Spectrum of the silylation product IV

naaa SPECTRUH Data Ella: [100,1001x09209604 3-OCT-88 19:24
Sa-pla: 1H t-BuOH 20-ln 100-2-20-220-5-260-20—300-5

ar 10‘30' 2! (Pos.) cc 450.6c 3?: n/a 205.0000 Inc. 712.9849 Lv 0.00
Scan! (701) - (695, 703) (coat. 1.001

 

 

 

 

 

 

100
r 205 b
R
a ‘ .
1 so: 253 "
a l
t 4
l ‘ t
V 60‘ .
Q l L
A ‘ ,
b so: .
u ‘ F
n l
d 4 73
a 20" I76 .-
1'1 4
C 4
. ‘ 1 1 AL .
U . %-.Lf , . é . , . - . C.T f 111 , ,
50 100 150 200 250 300 350 400

Figure 8.5

 

.+_
OSiMe3 _l '

- ‘OSiMe3 - CH0
—" —“" 176
89 29

 

 

Interpretation of the mass spectrum for the silylation product IV.
Molecular ion weight = 294.

Mass Spectrum of the silylation product V

MASS SPECTRUM Data F110: [100,IOOIX09169405 IG-SEP-94 14:58
Sample: BSTPAO415 DBSMS 100-2-20-1
RT 9'01" 21 (P03.) GC 450.6C BP: (II/l 293.0000 Int. 93.1320 Lv 0.00

Scan! (602)

 

 

 

 

 

 

 

 

 

100
l 293 r
R t
a P
I 80" b
a D
t P
l h
V 50" ..
IE9
e 4
A 73 ’
' 40‘" ,.
u\’ .
n 4 r
d 279 r
a 204 __
n F
c 1 7 2 5 307 ,
26]
O 177 396
A “I A “L1 k L
U ' ' I '7 7 fl T V I v V I
100 200 300 400 500

 

+
Me3SiOH2C COzSiMe3 “I- C028iMe3
e
——>
396 293
COzSiMe3

’CHQOSiMe3)

( - H - OSiMe3
——> ———a-

Me3SiOH2

- OSiMe3
89 300

 

-'OSiMe3
-' GD
1.\H

 

®OSiMez

- 'OSiMez
-'OSiMe3 ;/
26

-6

Figure 8.6 Interpretation of the mass spectrum for the silylation product V.
Molecular ion weight = 396.

Mass Spectrum of the silylation product VI

 

 

 

 

 

MASS SPECTRUM Data F110: X02129410 2-DEC-94 23:03
Sanplo: ZHINSOUL OBS I00-2-20-220-S-260-20-]00-5
R? 8'02' El (Pos.) CC 450.6c 8?: all 205.0000 Int. 23.6619 Lv 0.00
Scan! (483) - (482) (coat. 1.00]
100

205 l
R 1
. .
1 80* .
a
L ‘ ,
1 .
v 60‘ l'
o .
A r
' 401 *
m_ .
n >
d 176 ’
a 20. 88 103 .
n 1
c . 151 ’
. 75

L..- -41- . A, rag- A. , #115 .mu_- - .A. . -1, f, ..
50 100 150 200 250 300

154

 

CH0 1’:
e
O O - H - CO
0 I 0
206 205 177

 

 

Figure 8.7 Interpretation of the mass spectrum for the silylation product VI.
Molecular ion weight = 206.

Mass Spectrum of the silylation product VII

MASS SPECTRUH Data F110: [100,1001x09209613 3-OCT-88 22:53
Sample: in CHJCN 16010 100-2-20-220-5-260-20-300-5

RT 10'46‘ El (903.) CC 450.6C BP: m/z 205.0000 Inc. 122.7060 Lv 0.00

 

 

 

 

 

 

 

 

Scanl (719) - (715, 721) (coet. 1.00)
100
205 ,
R .
e 294 F
1 80: p
a
r .
i
v 60‘ ”
e 177 t
279 >
A t ’
b 40‘ "
u .
n .
a l .
a - r
n 2°, 139
: ‘ 73 88 1 1 219235 C
50 100 150 200 250 300 350 400

156

ego '

294

 

 

 

Figure 8.8 Interpretation of the mass spectrum for the silylation product VII.
Molecular ion weight = 294.

157

Mass Spectrum of the silylation product VIII

HASS SPECTRUM Dara Elle: [100,1001x09209604 3-0CT-88 19:24
Sample: 1M r-BuOH 20010 100-2-20-220-5-260-20-300-5

RT 11'20' El (Poa.) GC 450.60 8?: ml: 220.0000 Int. 1350 0340 Lv 0.00
Scan! (757) - (750, 761) [c0et. 1.00]

 

 

 

 

 

100
220

a l ’
O
I 80'4 _
a
t .
l ‘ .
v 60‘ -
e
A t
b 40" P-
u P
n 4 r
d t >
a 204 163 '-
n 4 1 2 b
c 32 r
e . J 96 110 1

0, :‘fl‘. 41-71, ‘1“ . fr. 1 . vrA‘v y . fr 1 ' ' E ‘ T f r

50 100 150 200 250 300

H/Z

158

 

163

2611:2112

137

Figure 8.9 Interpretation of the mass spectrum for the silylation product VIII.
Molecular ion weight = 220.

Mass Spectrum of the silylation product IX

 

 

 

 

 

 

 

MASS SPECTRUM Data File: (100,1001X09209604 3-OCT-88 19:24
Saaple: 1M t-BuOH ZOmln 100-2-20-220-5-260-20-300-5
RT 10'59‘ El (903.) CC 450.6c a/z 237.0000 Int. 1040.6230 Lv 0.00
Scan! (734) - (725, 737) (coat.
100

37 .
a ‘ r
e L
1 80‘ -
a .
t ‘ .
1 .
v 60‘ _
e

266 .
345
A >
u .
n fi #VA , 4»
d 152 -10.0 »
a 20: 1 '
n t .
c 76 209 ,
° 53 104 l ’
(r #1? - r ,1, - 1 v1: .
50 100 150 200 250 300 350

Figure B.10

 

160

76

Interpretation of the mass spectrum for the silylation product IX.
Molecular ion weight = 266.

I61

Mass Spectrum of the silylation product X

MASS SPECTRUH Data Flle: [100,1001X09209613 3-OCT-88 22:53
Sample: tn CH3CN 16min 100-2-20-220-5-260-20-300-5

RT 11'59' 81 (P03.) GC 450.6c 89: m/z 325.0000 Int. 701.4540 Lv 0.00

 

 

 

 

 

 

 

 

 

 

 

 

Scanr (300) - (789, 804) (coat. 1.00)
100
325
1
R b
e I
1 so: .
l

a 4 354 .
F 73
1 237 >
v 60‘ ”
e b

221 .
A 152
b ‘0-1 i"
0 295 ’
" ‘ 193 F " ' ' " ' " "
d -10 o »
° 20‘ ”
n 1 5 2 7 .
c 2 7
e 1 1 1 1 L .

0 r 1 _ _ If . , . -
100 200 300 400

H/Z

.1.
CH0 (30231111237: C0251Me3 COzSiMe3
~H’ Ct)
-‘CHO -
-CH0
7: —».. o
CHO HO CHO o
354 325 29s

CHO COzSiMe3

. 0 0 0:0

 

CHO
081N183
325 ,CO
90
CHO COzSiMe3 \?
- H-
———>
118 .
@OH
H 207
30
CHO H0
12:
CHO
- COzSiMe3 _ CO _ CO
——"” 209 ———-’ 181
117 28
-H' -
——> 180 —C%> 152

Figure B.11a Interpretation of the mass spectrum for the silylation product X.
Molecular ion weight = 354.

 

- C02 - OSiMe3

or ——-> 310 ——>

44 89

 

Figure 8.1 lb Interpretation of the mass spectrum for the silylation product X with
SiMe3 migration during GC/MS analysis. Molecular ion weight = 354.

164

Mass Spectra of the silylation product XI A,B

MASS SPECTRUM Data File: (100,1001X09209604 3-OCT-88 19:24
Sample: 1H c-BuOH 20min 100-2-20-220-5-260-20-300-5
RT 12°46“I SI (903.) CC 450.6c 3?: ml: 325.0000 Int. 932.6188 Lv 0.00

 

 

 

 

 

 

 

 

Scan. (853) - (843, 857) least. 1.00)
100
325 .
R ‘ ,
e
1 801 p
a i
t 73 .
i ‘ .
v 50* 442 —
e
A u
b 40“ _
u l
n 4
d
a 20‘
n 1
C 4
e
- A LinL
U ' *‘ I f
100 200 300 400 500
M/Z
MASS SPECTRUM Data File: [100,1001X09209604 3-0CT-88 19:24

Sample: 1“ c-BuOH 20min l00-2-20-220-5-260-20-300-5
8? 13'02' 81 (P03.) GC 450.6c 8?: ml: 73.0000 Inc. 157.0286 Lv 0.00

 

 

 

 

 

 

 

 

 

 

 

Scan! (871) - (863, 874) (coat. 1.00)
100

1 7D .
a 442
‘ 2:5
1 807 -
a 4
t ‘ t
i ,
V 60¢ )-
e 1 l
A 325 r
b 40* -
u 1 r
n 1 207 '7‘ Y ‘ >
d 4 ~10.0 .

1 7
a 20‘ 1 0 "
n 1 ‘27 ,
C 4 >
2 3

° 1 1 01 3

- L 4.1 AimA L L

u . f , . 43 v v 1 v

100 200 300 400 500

H/Z

   

 

+
Me3Si07C COzSiMe3—l' O COzSiMe3
- 'OSCiMe3 -‘OSiMe3
117 90
442 325
COzSiMe3
- COQSiMej,
or
1 l7
325 207
zsl- CO 308181734
' H 90
COzSiMe3 C038iMe3
. C9
- CO - COzSiMe3 -CO
or —-——> ——> ——>
28 414 ,8
1 l7 “
CHO CHO
29-CHO
297 , 269
385 89l-°OSiMe3
C90H
, -m 0
‘5' 28
180

Figure B.12a Interpretation of the mass spectrum for the silylation product XI A,B.
Molecular ion weight = 442.

166

 

M<33SiO Q
IMez
0 C02 _% 383 Q
. 1+ Si migration 44 ‘CO 355
Me3SiO C COzSIMe3 W 427 ’ fl _%
15 -CO 399 44
CH. HO
/SiMe3
-‘OSiMe3 9C0 COZSiMe3 @C0 C02
442 89
0 0 0 0
——>
353
CH. HO CHO HO
-CO -
0 SiMe3 28 441 C03
0 9C0 C02

-'C051Me3 @OSiMe3

O @‘O
CH0

223 325 CH0 CH0

44 -C02 309

. GOSiMe3

. -‘OSiMe 3
‘—
74

 

280

. -'OSiMe3
‘—
73

180

 

Figure B.12b Interpretation of the mass spectrum for the silylation product XI A,B with
SiMe3 migration during GC/MS analysis. Molecular ion weight = 442.

167

+
CH. C0251Me37'

 

442 -‘COzSiMe3
COzSiMe3 fir

0.0

-'CO -°COzSiMe3 O Q 180
0r —77—4> 414 -———H>
“8 117 ® - COzSiMe3
-'H
m 0 0

 

 

297
C90
l79
OSiMe3
or -.CHO -COZSIMC3 QC 0
29 117
CHO
296
413
-'H -'CO
9 2
/OSiMe3 8 73l-‘Si Me3
385
OSiMe3 O
O O //
(30;! CRDH
323 223

Figure B.12c Interpretation of the mass spectrum for the silylation product XI A,B.
Molecular ion weight = 442.

168

+
CH0 COzSiMe3—l°

     

442
398 -‘OSiMe 3
- A zsl-CO
905110123
219
-'CO , -'Me .

CHO COzSiMe3

-‘OSiMe3 .

74 . 89 -‘OSiMe3

l ‘H 1%028iMe3 l ‘OS'M
- 1 e3

'V
OSIIVIC3 -A) 89:05“ 163
C9

. 191
@O 960 0:0

 

 

309

281 OH

- ' ~USiMe3 19?-
741 0119/1123 741 -'H

I ®OH

«—

' ‘

00H 0 0

207 235

Figure B.12d Interpretation of the mass spectrum for the silylation product XI A,B with
SiMe3 migration during GC/MS analysis. Molecular ion weight = 442.

169

Mass Spectrum of the silylation product XII

HASS SPECTRUM Data File: [100,1001X09209605 J-OCT-BB 19:89
Sample: I" c-BuOH 31min 100-2-20-220-5-260-20-300-S

RT 13'26' El (903.) CC 450.6c 8?: ml: 73.0000 Inc. 1083,6340 Lv 0.00
Scan. (897) - (891, 900) (acct. 1.00]

 

 

 

 

 

 

 

 

100
1 73 ,
R b
e
1 80'4 _
a i
t V
1 147 355 l
V 60‘ ”
e 0
A l
b 40" )-
U ‘ 323 t
n 1 5 S *
d .
a 20-4 )-
n 1
235 I
c 1 207 299 3 9 503 .
179 3 ° ‘71
O < 133 221 i L 510 )
if,rgl AI u. 2L 1 rrdo L LL
0 T V’ V ' r ' ff ' I ' ' rfi Ff ' ' f“ f f ' ' I ' '
100 200 300 400 500 600

H/Z

l7O

- 7
-'OCH 39 \

__, 323

29
501"I"I\{‘()SiMe3

(OSiMe3) 4‘3 0231Me3

   

Me3Si02C OzSiMe3

70 -CO
5) 28 - COZSiVle}
502

8:l"OSiIVI€3 ”7 28 - C0
_- - Me3SiOzC
C0281Me3 385

_ C ‘.OSIMC3

-‘H /4 \:)SiMe3
CHO COZSiMe3

353 352

-'COzSiMe3
N y -.H
118

 

“ 413
. -‘OSiMe3

0 CHO OzSiMe3

 

Me3Si02C

 

Figure B.13a Interpretation of the mass spectrum for the silylation product XII.
Molecular ion weight = 530.

I71

- +
CHO COzSiMe 7'

    

Q Si migration - C02
——> ——->
44
Me3Si02C COzSiMe3 Me3SiOzC COzSiMe3 Me3SiOz OgSiMe3

530

-DSiMe3
A I
(9
9

0'0

Me3Si02C COzSiMe3

 

397
Me3SiOzC OzSiMe3
74I'OOSIME’2 118 -'COzSiMe3
3H
fi3=SiMe2 (l)=SiMe3

” O
or ——>'CH3 515 'C02 Q Q Q Q
15 44 (9

. . Me3Si02C
Me3S102C COzS1Me3
368

73
’/1e3si ]l7l-‘COzSiMe3
251
72
(.9 72

Me3Si02C
295 I79

471

Figure B. 1 3b Interpretation of the mass spectrum for the silylation product XII with
SiMe3 migration during GC/MS analysis. Molecular ion weight = 530.

Mass Spectrum of the silylation product XIII

MASS SPECTRUM Data r119: x09169415 16-589-94 18:45
Sample: asrerpz DBSMS 100-2-20-300
RT 9'33- :1 (903.) cc 450.6: 3?: all 73.0000 Inc. 0.0119 Lv 0.00

Scan! (638)

 

 

 

 

 

 

 

too

1 73

4 17
R
o 4
1 80‘
J
t 4
1 4
v 60*
e 4
A 4
' 401
}. 6C2
n . 323
d
‘ 207
n 93
c .
o 57 05133 :1“ 281309 ]97 5 518

8 J 495 14
on LfilL‘wALthlllfiv ll‘JALJ 4+1 11L LLlr ' V A v
' * ‘fi ' I a
100 300 400 500 600

173

 

+ 33H
hde3SK32CI c132snwe3‘7- (3t ,/
'(‘OSiMe3-'H)
-OSiMe3
—>
177
Me3Si02C OzSiMe3 Me3Si02C COzSiMe3
441
618 ' C02 - C0251Me3
44
118 ‘l‘

COzSiMe3 Me3Si02C
397 323

- (OSiMe3 smlgg

 

323 235

——. O
44 0.0

 

Figure 8.14 Interpretation of the mass spectrum for the silylation product XIII.
Molecular ion weight = 618.

174

Mass Spectrum of the silylation product XIV

MASS SPECTRUM

D

ata File: x021294oa

Z‘DEC-94

22:

15

 

 

 

 

 

 

 

 

Sample: ISHINSOUL oas 100-2-20-220-5-260-20-300-5
RT 7'53- 21 (903.) cc 450.6c BP: n/z 147.0000 Int. 26.4542 Lv 0.00
Scan! (474) (47], 476) (coat. 1.00)

100

147 .
R 4 L
. y .
1 00‘ 411 P
a 1
t .
1
v 607 '
e i
A 4 439
3 40- 73 _
¢~ . 426 t
n 337 1 422 r ,
d 4 -io.o l
a 20+ 249 "
n 175 r
c ‘ 3 3 301
e . lhl 21: 263 309 »
U whowu . ., ‘. A?A.v {fl 444 L. . 4f; ‘ir, 2, £ . . r v
100 200 300 400 500

175

MezSiOCO CHO

- (’OSiMe3
-'H ) -'OS7iMe2
—-———->
88
3CH3
15

Me3Si COzSiMe3

- (:03 - SiMe3
-——> ———>
44 . 73

 

   

O
382 -CO
/8'
- OSiMe3 (2:0 COZSIIVIC3
\ -( OSiMe3
0337

Figure B. 1 5 Interpretation of the mass spectrum for the silylation product XIV.
Molecular ion weight = 426.

 

176

Mass Spectrum of the silylation product XV

”ASS SPECTRUM Data 711.: (100,100|X02129409 2-DEC-94 22:39
Salplo: [SHINSOUL REHEAT l00-2-20-220-5-260-20-300-5

RT 10'18' SI 4903.) GC 450.6c 39: II! 237.0000 Inc. 73.9358 Lv 0.00

 

 

 

 

 

 

 

 

 

 

 

Scan! (619 to 620) - (617, 621) (coat. 1.00)

100

237 350
R >
. b
1 00* "
a h
t ‘ ’
l 2‘ 6 3‘6 >
v 60‘ l r
o ‘ ’
295
A
3 3

‘0‘ 221 J 5 1-
u" r
n I
d '10.0 '
a I
n
c 2 1 »
- ’1’ 1 '

2 fl AL ‘ A“ fifiA .L . f -
100 200 300 400

H/Z

177

- OSiMe3

   

Figure B. 16 Interpretation of the mass spectrum for the silylation product XV.
Molecular ion weight = 336.

APPENDIX C

APPENDIX C

OZONATION PRODUCTS PURIFICATION

Benz[a]anthracene ozonated samples were purified using reverse phase medium
pressure column (C18 column) followed by thin layer chromatography (TLC). Reverse
phase medium pressure chromatography (Cl8 silica, 500 x 20 mm glass column, pressure
rated 150 psi, DyChrom, Santa Clara, CA) was used to separate the crude products. The
mobile phase was controlled by a low pressure pump (Model 81-M-2, Chemco,
Sunnyvale, CA) at a flow rate of 4 mL/min. The column was conditioned using the
starting solvent mixture. The sample was dissolved in the same solvent mixture and
loaded into the system. Two mobile phase gradients were used for the different ozonation
times. Gradients of 90/10 and 70/30 methanol/water to pure methanol were applied for 10
min and 20 min ozonation products, respectively. Afier fractionation of the crude
products on the CI8 column, TLC was employed to further separate some of the
fractionated samples. The silica gel plates (20 x 20 cm, 250nm Uniplate, Analtech,
Newark, DE) were eluted with hexane/ether mixture or a mixture of benzene/chloroform
with a minimal amount of methanol. TLC separation for the fractionated samples from
Cl8 column depended on the sample purity and quantity of the sample (minimum loading

for each plate < 15 mg).

Flow charts shown in Table Cl and C2 give the sequence of separation for 10

min and 20 min ozonation sample; each fraction was analysis by GC/MS.

178

179

793 $0.9— méém Yoém

 

 

 

 

 

 

 

 

 

 

_ 0 _ _
w— u bfimuogxom
3% 8a... 00... 9:
0am
_~m¢meaw-~+a _~mvmohmaé _~m4mohm-m+a
________ ________ :______
35:21
008240002880 23:5? 08.050.08.80
8% 85... 00.2.2 050240588800 3% Sam 0.: a:
2 a. a 83:50 8% 8% 0d a: a 2 a. E 83.80
_ - _ _ _
:0»: 6:0
5 E E E 2 E E a a
_ _ _ _ _ 0 _ _ _

 

acguaaaioué £0 0.0
83000 350 .3 080030 £2038 nOmumz P08 + 0388 S

 

 

gauging 28 m

0058800 88 3

 

 

 

03:58 2 .8.“ 2800 .23 ©3002 0383 8a 540 30c gram mo 030,—.

180

_N_m4.ndo_u_. we: N_m~v_m—£h_-m+v.:

a
_________

E: 4 00020000
32.. EB 8% 000 a:
30 a. I“. 200800

_lJ

TE V: M-; 0.; WE 01—"— 5-;

P _ _ _ _ _ _

8380000350006
30 dd
E mm m...— vm mm on um am am

_ _______L

8000038350020 £0 0.0
0.8000 0300 .3 08E0 £0028 nOmusz 0.08 + 0358 E0

 

 

 

888553480 2:. m
00000000 38 0N

 

 

 

00358 on 00.0 0:000 505 0000000 03800 .8.“ 5:0 300 500.30% NO 030...

18]

Typical GC chromatograms for the samples from the C,8 column and samples that had
undergone additional separation procedures are shown in Figure Cl and C2,
respectively. After the first separation step, most samples were not pure enough for
GC/MS analysis. As shown in Figure C.l, these samples still resulted in shoulders
between peaks, causing overlapping of the mass spectra of neighboring compounds.
Therefore, the molecular ion for the first separation compounds could not be identified
and further purification was required. After a second separation step, some of the
fractionated samples, such as F4+5-3 from 10 min ozonation as illustrated in Figure C.2,

was pure enough for GC/MS analysis.

Although the separation procedure used could purify (or concentrate) compound
V and X, this procedure failed because of several problems. First of all, the silica on the
TLC surface catalyzed the further oxidation of PAHs (or ozonation products) by any free
radicals present. Therefore, large number of compounds, e. g. ethers or esters, were
generated. When these compounds were produced, the similarities in their physical and
chemical properties resulted GC peaks that could not be separated, resulting in
overlapping mass spectra. Furthermore, these products (ethers or esters), as the major

peak shown in Figure C.2, were not produced by the ozone direct reaction pathway.

In conclusion, another separation method should be developed to improve the
purity of ozonation products. The purification using reverse phase medium pressure

column and TLC as described here can not be used for this purpose.

182

TIC Data File; (100,1501x05109614 10-HAY-96 16:47
Sample: BlalAn F3+3001 OBS 100-2-20-220-5-280-20-300-5

Scan. 200 to 1200(1200) R? 2'59” to l7°$8‘(17'58') EItPos.) Lv 0.00
Operator: 00

 

Hax 8889.22 R.T.
§ 10 15 mag. abund.
100 ‘ L ‘ , ‘

A

b

u

n 80‘

d

a

n

C 60*

e

40*

 

20‘

 

 

 

 

 

- - v r - v7 1 v . . r ' * r I ' ' ‘* 'l.0 8889.22

600 800 1000 1200
Scan

Figure C.1 GC chromatograms of sample F3 from 10 min ozonation sample followed
by separation on a C13 column.

 

 

 

 

 

 

rIC Data File: [100,1solx05179520 l7-HAY-95 23 12
Sample: B(a]An 94+s-3 oas too-z-zo—zzo-s—zao-zo-aoo—s
Scan! 200 to 1200(1200) RT 2'59" to l7'SB"(l7’SB“) EltPos.) Lv 0.00
Operator: 00
Max 11345.7 a.r.
S 10 15 mag. abund.

100 ‘ ‘ * - - * A
A TIC
D
U
n 80.1
a
a
n
c 60‘
e

40%

20*

- - ' r - - - v - - - . - - - v ' ' - '1.o 11845.7
20° 400 600 900 1000 1200

Scan

Figure C.2 GC chromatograms of sample F4+5-3 from 10 min ozonation sample
followed by a second separation.

APPENDIX D

 

APPENDIX D
MASS SPECTRA AND INTERPRETATION OF

SILYLATED BENZ[a]ANTHRACENE OZONATION PRODUCTS

In this appendix, the raw spectra obtained from GC/MS analysis is followed by
the interpretation of these spectra for benz[a]anthracene ozonation products. Products
were silylated by BSTFA + 1% TMCS and heated at 100°C for 60 min prior to GC/MS
analysis. The silylation reagent converts -OH or -COOH functional groups to -OSiMe3 or

-COOSiMe3.

The mass spectra from library search in NIST MS Search (ver 1.5, 1996) or
BenchTop/BPM Search (ver 3.10d, Wiley & Son, 6th ed., 1994) which were matched the
raw mass spectra from GC/MS analysis are also illustrated. The commercial compound,
2-carboxybenzaldehyde (Sigma-Aldrich Library of Rare Chemicals, Milwaukee, WI),
was prepared so as to compare its GC/MS spectra with that of compound XIV A/B. From
the GC chromatograph and mass spectra of 2-carboxybenzaldehyde before and after TMS
derivatization, the target compound found to exist in two isomeric forms which were

identical to those shown in compound XIV MB.

184

185

of, W.;"... 006’...

l (228) u A (244) m A (240) IV (262) v (260)
©
©©0 ©© ©© 0...... © ©
. {no V
' coon 330
V‘ (2") VII A.3 (276) "11092) [X A3 (276)
H © "° © ° 6) ° © ° C)
©©© ©©© @063 © © © G
on o 1:, a"
11 B (244) m n (240) CHO coon
x (258) X! A,B (26o) xu A,B (276)
C(m 0” @Em @E
. m @coon coon coon
XIII (16o) XIV A,B (150) XV (138) XVI (166)

Figure D.1 Products formed from the ozonolysis of benz[a]anthracene. Roman numbers
correspond to the ID assigned to each product. Molecular weights are given
in parentheses.

186

Mass Spectrum of benz[a]anthracene (I)

MASS SPECTRUM Data File: [100,1001110129604 Q-OCT-OO 10:56
Saano: BlalAn 1n CH3CN 4min 100-2-20-220-5-260-20-300-5

RT 12'52' 8! (903.) 6C 450.6: DP: ml: 228.0000 Int. 274.9452 Lv 0.00
Scan. (859)

 

 

 

 

 

100
228
R r
c F
l 804 D
' 4
t b
1
V 60‘ ‘
Q
,A
b ‘04 u-
U
II
d 4
. <4 :-
2° 1 4
n
c 1
1 * .. l‘ ‘R
- Lin J L A J14 A
01 f f V ' I .0 V v f I V V V f I fi' * 1 * Y fifi * i
50 100 150 . 200 250 300

"la

Mass spectrum of benz[a]anthracene obtained from NIST database (NIST#: 114976,
CAS#: 56-55-3)

100. 228

000

114
101

 

 

20 4O 6O 80 100 120 140 160 180 200 220
(M)Benz[a]anthracene

187

Mass Spectra of the silylation product II A,B

MASS SPECTRUM Data P110: [100,1001X10119609 3~OCT-88 21:02
Sanplo: BICIAn 1min 085 100-2-20-220-5-260-20-300-5

RT 14'00' El (903.) CC 450.6c 8P: MI! 316.0000 Int. 43.4585 Lv 0.00
Scan! (944)

 

 

 

 

 

 

 

 

 

 

100
316 l
R .
e 1
1 80‘ _
a .
t
1 73
V 60‘ .
e
A .
b L-
U
n
d
' 1-
n
c .
e
355 ’
. A J-Ari .' -A _. v
50 100 150 200 250 300 350 400
H/Z
MASS SPECTRUM Data File: (100,1001X10119611 J-OCT-BB 21:47
Sample: BIaIAn 4min 035 lOO-2-20-220-5-260-20-300-5
RT 15'49' EI (Pos.l GC 450.6c 8P: n/z 316.0000 Int. 154.0589 Lv 0.00
Scan. (1056)
100
316 .
R .
e . .
1 80‘ _
I ‘ r
t .
i .
V 60‘ ’
e 73 .
A 4 .
b 40‘ ’
u
n r
d < r
a r-
n .
C ,
e r

 

 

 

 

 

I88

01'

   

316

- Me3

——> 301

-15

——> 285

-31

__, 270

-46

-‘OSiMe3
- n

_, 226

-90

Figure D.2 Interpretation of the mass spectra for the silylation product 11 A,B.
Molecular ion weight = 316.

189

Mass Spectra of the silylation product 111 A,B

MASS spccraun Data P119: (100,1oolx10119610
Sample: alalhn Znin 035 100-2-20-220-5-260-20-300-5

RT 17'00' EI (Pos.) GC 450.6c 3?: II: 404.0000 Int.

Scan. (1135)

3-OCT-08 21:24

112.6099 Lv 0.00

 

 

 

 

 

 

100
. 404 p
R D
G
1 00‘ .
I ‘ r
1'. f
1 .
V 60‘ -
e s
73
A 1
b 40‘ ,
u r
n 3 6 ’
d D
8 20- .-
n ‘ 1 7
c 4 2 7 Zr 21‘ '
o 4
1 3 3 9
AL 1- J .10 ‘1‘... . 4 L L A
o . r 4, , 00.. . 0 04,, 0 a ,
100 200 300 400 500
"ll
MASS SPECTRUM Dltl Elle: (100,1001X10119610 3-OCT-00 21:24

Sample: Blalhn Znin DDS 100-2-20-220-5-260-20-300-5

R1 16'24' 3! (Pos.) 6C 450.60 3?: ml: 404.0000 Int.

Scan. (1095)

316.4309 Lv 0.00

 

 

 

 

 

 

100
. 404 .
a ‘ ’
o ‘ ’
1 80' .
. D
t 1
1 ‘ ’
v 60‘ P
e b
13
A b
b 4 o- ”
u 1
n 1 .
d D
' 20‘ ’
n . *
c D
e 1 7 3 9
2 5 3 5 1
0.4 ._1 g v w L v v #L “fit“ f4fi+2 i + Li Y L Lfi'Ljr ‘77 v v '7 V f

100 200 300

400 500

I90

01'

   

404
- 881-(‘08iMe3 - H)

316
or
315

Figure D.3 Interpretation of the mass spectra for the silylation product 111 A,B.
Molecular ion weight = 404.

191

Mass Spectrum of the silylation product IV

MASS SPECTRUM Data r110: [100.1001x09279606 27-SEP-96 13:46
Sanple: BIaIAn+03 0.25 2min 1n BSTFA 100-2-20-220-5-280-20
R? 8'48“ BI (Pos.) GC 450.6c BP: I/l 319.0000 Int. 578.5945 Lv 0.00

 

 

 

 

 

 

 

 

Scan! (588) ' (585. 590) (coat. 1.00]
100
319 )
3C4
R ‘ r
L
e L
1 80‘
a r
c ‘ ‘
I 1 ’
e ‘ ‘ P
A 1
b 40* "
\I 1 p
n < rfi—w - v vv‘Ti '
d . '10.0 *
. 20‘ b
0 1
c . 1J7 2 ‘ 274 r
° < 1 2 '2 a I 291
- - 14.41.11). JLLL AmAi# 4 AL I - - 0 - -
UT ' V '—' Tfii' * v I ‘ *ffiTf'i' ' Tfi'fi f' T ' 'fif' ' I
50 100 150 200 250 300 350 400

I92

 

Figure D.4 Interpretation of the mass spectrum for the silylation product IV.
Molecular ion weight = 334.

193

Mass Spectrum of the silylation product V

MASS SPECTRUM Data r11e- [100 IOOIX
. . 10119610 - - -
Sa-pJo: BlalAn Zntn OBS 100-2-20-220-5-260-20-300-5 3 OCT 38 Zl.2‘
R1 13'11' :1 (Po: ) CC 450 6c 39 m/
. . : 2 231.0000 Int. 999.1099
Scan. (881) - (875, 883) (coef. 1.00] L" 0.00

 

 

 

 

 

 

 

100
231
R
e
1 80“ >
. )-
t
i
V 60‘
e .
A 1 '
260
b ‘0‘ 2c2 '
u -
n
d
a 20‘ ’
n 101 h
c 1 5
e 88
76 l 116
U A A v 2'. V++AT ‘_ v _ ’
50 100 150 ‘ 200 250 300 350

Mass spectrum of 2-(2’~formyl)phenyl-3-naphthaldehyde obtained from BenchTOp/BPM
database (Serial #: 254673, CAS #: 132335-14-9)

Library Reference Spectra Page 1

Query:
NV: 260
Comp: c18h1202

Peaks: 231,70-100

Serial: 254673 CA5 Reg No: 132335-14-9 MW: 260 Form: C18 312 02
2-pheny1naphthalene-1,2'-dicarboxaldehyde
l-Naphthalenecarboxaldehyde, 2-(2-tormy1pheny1)-

 

231
202

115

A A l A A A A

260

 

 

I 176

I Y‘r‘r v V V v ' fjfij Y r 1’

 

 

I fif‘r 0 vi

2? ' - so' ' ' '7'5' . ' 100 ' 125 150 17s ' 200 225 250 27s

l94

Mass Spectra of the silylation product VI A,B

MASS.SPECTRUM Data F11e: [100,1001X10119610 3-0CT-88 21:24
Sample: BIaIAn 2min 085 100-2-20-220-5-260-20-300-5

RT 11°53“ 8! (Pos.) GC 450.6c 8P: m/z 332.0000 Int. 70.9747 Lv 0.00
Scan. (794)

 

 

 

 

 

 

 

100
332
R .
e
1 80‘ _
I
t ‘ r
1 1
v 60“ "
e
A 4
b 40‘ -
u
n
d .
a L
n
c
e >
50 100 150 200 250 300 350 400
H/z
MASS SPECTRUM Data Plle: [100,1001X10119610 3-OCT-88 21:24
Sample: BlaiAn 2min 085 100-2-20-220-5-260-20-300-5
RT 13°43“ :1 (Pos.) GC 450.6c 8?: m/z 332.0000 Int. 115.1673 Lv 0.00
Scan. (916)
100
1 332 .
a 1 .
e .
1 80‘ ’
a . .
t
i I
v 601 C
e .
243 1
A
b )-
u r
n r
d .
a _
n
c
e

 

 

 

 

 

 

I95

- °OSiMe3
- 89

 
 

\Cfo
-29

303 (OS'M H) 281- CO
- o 1 c3 -
332 k

1» ©©©

317

Figure D.5 Interpretation of the mass spectra for the silylation product VI A,B.
Molecular ion weight = 332.

196

Mass Spectra of the silylation product VII A,B

MASS SPECTRUM Data P110: [100.1001X10119610 3-0CT-00 21:24
Sample: 8(aIAn 2min 085 100-2-20-220-5-260-20-300-5

RT 13'31' EI (Poe.) GC 450.6c 8P: m/a 231.0000 Int. 139.3189 Lv 0.00
Scanl (903)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100
. 231 .
R
e 215 ’
l 80‘ '
a
t V
i
V 50‘ 2C2 »
e 73 r
A
b 40d »
u 319
n .
d .
a 20.. .-
2 101 1 7 348 >
‘3 176 2 1 ' 1
o A 0f - - . - r -
100 200 ' 300 400
M/z
MASS SPECTRUM Data Plle: [100.1001X10119610 3-0CT-88 21:24
Sample: BIaIAn 2min 085 100-2-20-220-5-260-20-300-5
RT 14'16' 81 (Poe.) 6C 450.6: 8?: m/z 231.0000 Inc. 123.2544 Lv 0.00
Scanl (953)
100
1 231 b
2 5
R
e .
1 80‘ ”
. .
t .
1 73 .
V 60‘ 2c2 ~
e .
r
A 4 319
b 40‘ 7
u b
n u
d .
. ,_
n a .
C ,
e
100 200 300 400

I97

©© . .
CO_O=S|M62
305 G
-CO
.43 I_ CH3 - -c02SiMe3 ©©. HO °CHO

231

 

O - °COzSiMe3 202
© 4;

CO;SiMe3

319

 

Figure D.6 Interpretation of the mass spectra for the silylation product VII A,B.
Molecular ion weight = 348.

198

Mass Spectrum of the silylation product VIII

nass saberaun

Data File:

[100.1201X1V119610

1
3-OCT-83 21:24

 

 

 

 

 

 

Sample: BlalAn 2min 085 100-2-20-220-5-260-20-300-5 .
RT 14'14' EI (Pos.) GC 450.6c 8?: m/z 319.0000 Int. 31.2527 Lv 0.00
Scan! (951) - (948. 954) (coef. 1.00)
100
319
R P
e D
1 80‘ ‘
a b
t
’1 b
V 60‘ “
e t
A
b ‘0‘ 147 c
U P
n L
d
3 20‘ -
n 436
c 2 3 1 .
1 9 377
e 7793 I 181 245. 303 L
A LL 1 I [ah I‘LiLJ 1 LI ”.1 _‘ 1‘ a L “L ‘_ a A llalL ‘ l 4?. .1" . u A A ll .1 J fiLL; .11 A A A. 1%: l
100 200 300 400 500

M/Z

 

I99

   

436 319

 

Figure D.7 Interpretation of the mass spectrum for the silylation product VIII.
Molecular ion weight = 436.

Mass Spectra of the silylation product IX A,B

- nass_svccrnun

Data File:

200

(100,1001X10119611

3-OCT-00 21:47

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample: 8(alAn 4mln 085 100-2-20~220-5-260-20-300-5
RT 15'28' 21 (Poe.) GC 450.6c 8P: m/z 421.0000 Int. 35.6202 Lv 0.00
Scan! (1033) - (1029. 1035) (coat. 1.00)
100
421
R D
e 1
1 80‘ “
. ‘ b
c D
1 ‘ D
v 60‘ b
e 3 .
b ‘0‘ I-
U i 2 ’
n 193 ’
d 6 ’
9 20~ *
n b
C 4
1 6
e ‘ 1.1 2 3 273 - 1
0 _
100 200 400 500
M/Z
uass spccrauu Data Plla: (100,1oolx10119610 3-ocr-ae 21:24
Sample: BlalAn 2min OBS 100-2-20-220-5-260-20-300-5
RT 16'13' 81 (Poa.) GC 450.6c 3P: m/a 147.0000 Int. 14.9352 Lv 0.00
Scan. (1083)
100
1 I7
R 1 .
Q P
l 80‘ b
a 1
t 73 348 ,
1 2:1 .
V 60-| :-
. J 3‘3 0
2(7 >
A
b ‘0‘ 2:8 1-
u 4
n 4
3 I
d 57 93 i
‘ 207 -
n 4
c < 1 1.2 414 .
g ‘ 2 6 4 7
0‘ ..
100 200 300 400

 

201

258 'fEEliEP
-27 .

K1118
-15 333 231

 

 

- 451- 'COOHI

/SiMe3

    

Figure D.8 Interpretation of the mass spectra for the silylation product IX A,B.
Molecular ion weight = 348.

 

202

Mass Spectrum of the silylation product X

MASS-SPECTRUM Data tlle: (100.1001x10119610 J-OCT-Oo 21:24
Sample: BtalAn 2nln DBS 100-2-20-220-5-260-20-300-5

RT 14°32” E! (Poa.) 6C £50.6c 3?: I]: 250.0000 Int. 929.5906 Lv 0.00
Scan. (971)

 

 

 

 

 

 

 

100
258
R
e .
1 80* .
a
t
i »
V 60‘ "
e t i
A .
b 40* v ..
u 202
n 2 0
d
l 20~ 101 L
n
c 00 *
. .5 J (3’ I
U ##Afialfi I 1 - v Y L'- YL fl v‘ v“ fig fiA fii v A,
50 100 150 200 250 300 350

Mass spectrum of benz[a]anthraquinone obtained from NIST database (N IST #: 101274,
CAS #: 2498-66-0)

 

 

 

 

 

100. 258
so- 202
230
88
' 129
so 75 .
115 150 174
0 III] I: {Ii-Irilil I l

50 7O 90 110 130 150 170 190 210 230 250
(M)Benz(A)anthracene-7,12-dione

 

203

Mass Spectra of the silylation product XI A,B

was! 3930730" Data Plle: (100,1201x10119611 J-ocr-sa 21::7
Sanpla: 3(alAn lain 035 100-2~20-220-5-260-20-300-5

3? 13'20- at (Poe.) cc 450.6c as: a/z 260.0000 Inc. 45.2911 Lv 0.00
Scan! (091)

 

 

 

 

 

 

 

 

 

100

2 o f

R ‘ ’

e .

1 so: .

a n

t 2.1

£ 1

v so~ .

e

A >

b 301 1(7 _

u 4 202

n b

a 1

I .-

n

C

G .

.3. tut—J.A. t.
300 350
n/z
mass specrnuu Data 911.: (100,1201x10119611 3-ocr-aa 21 47

Sample: 8(aIAn ¢nln Des 100-2-20-220-5-260-20-300-5

RT 13'45' EX (Poe.) GC 050.6c 39: ml: 231.0000 Int. 45.0081 Lv 0.00
Scan! (919)

 

 

 

 

 

 

100
: 2 1 y
R 210 .
o D
1 00‘ ~
a 1 b
t
1 ‘ I
v 60‘ -
O
A
b 40‘ 147 2'2 ~
ll
n 3 2 >
d 77
' 20* -
n 1-1 2 5 2 3 3
c 3 as l 51 0 3
175 .
0 .151 ‘3 ; . 2:1 2-9 3 7 L
d ‘“' 1 ' -
so 100 150 200 250 300 350

"It

204

   

-'CHO
'——>
- 29
260 23 l
-29 13430

Figure D.9 Interpretation of the mass spectra for the silylation product XI A,B.
Molecular ion weight = 260.

 

205

Mass Spectra of the silylation product XII A,B

nass,spacrauu

Data P110: [100,1001X10119611

Sample: BlalAn 4min 005 100-2-20-220-5-260-20-300-5

J-OCT-BG 21:47

 

 

 

 

RT 15'09' 81 (003.) 6C 450.6: DP: m/z 331.0000 Int. 24.3370 Lv 0.00
Scan. (1012) - (1000, 1015) (coat. 1.00)
100

3 3 1 p
R D
e
l 80‘ P
a
c b
i
V 60* 215 p
e 73 >
A r
b 404 316 F
u ‘ b
n b
d J
' 20‘ ”
n r
c 117
' 52 l 4 ’

A ‘ [fl 1711-1. .‘L a [filaikf‘ ‘1

 

 

 

100

MASS SPECTRUM

Data Plle:

200

300

(100,1001X10119610

Sample: Blalhn 2min 085 100-2-20-220-5-260-20-300-5

R? 16'00' El (903.)

 

400

3-OCT-08 21:24

 

 

 

 

 

 

 

GC 450.6: 3?: ml: 231.0000 Int. 53.1997 Lv 0.00
Scan! (1069)
100

231 n
a l L
. D
1 00‘ t
. b
t r
l l 1
V 60* 3‘ r-
e * 73 >
A .
b 40* -
u
n ‘ 3 3 >
d 2.2 303 3 r

2 8
‘ 20* »
n < v
C >
104
U
100 200 300 400

H/Z

206

 

231

 

 

303

 

Figure D.10 Interpretation of the mass spectra for the silylation product XII A,B.
Molecular ion weight = 348.

207

Mass Spectrum of the silylation product XIII

HAS: SPECTRUM Data Plle: (100,1201X10119611 3-007-80 21:47
Sample: Blalhn 4min 005 l00-2-20-220-5-260-20-300-5

R? 7'42' 31 (Poa.) GC 450.60 0?: ml: 73.0000 Int. 56.4437 Lv 0.00

Scan! (515)

 

 

 

 

 

 

100
73
R
9
1 80" "
‘ r
t
1
V 60‘ ”
e . 3C4 .
A
b ‘04 v-
u
n ,
d
4 20* -
fl 1 2 6
C
115 l ’1’ ’i‘ 2 9 ‘
°- r..m V'. .1] “l““lALI Lm-m- _ -‘A A-J. l Aa_‘_ - A. A1 -
T ' ' ' ' I ' ' ' ' I f Y ' r T r X fl ‘ j rAhfiL" ' rA' ' ' ‘r I V
50 100 150 200 250 300 350 400
Ml!

Mass spectrum of silylated dihydroxy-naphthalene obtained from NIST database (NIST
#: 40361, CAS #: 1032-28-6)

100~ L>S

 

 

 

 

50 .
\/'
/S
. \
289
45 201
O IIIIYIIIIIITITIITIII{TWAITIFITYILfAT—IIIIIIIIIILIIIIITTIIIIIIIAIMTITWIAITYIAII[11%1Tl(JILI[111m L

40 70 100 130 160 190 220 250 280
(M)Silane, [1 .5-naphthalenediylbis(oxvllbis[trimethyl-

 

208

Mass Spectra of the silylation product XIV A,B

"A33 SPECTRUH

Salp}a: BlalAn 10min OBS 100-2-20-220—5-260-20-300-5
5'01‘

I?

Data Plla: [100,1201X10119614 3-OCT-80 22:57

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

El (905.) 6C 450.6c DP: ml: 207.0000 Int. 75.2624 Lv 0.00
Scanl (402)
100
207 ,
a 1 .
e ‘ '
1 807 .
. P
t
1 ‘ '
" 50‘ 1:3 ~
0 r
A
b 40* ~
" 75
n
d .
I 20‘ 105 b
n 1 7 1 3 b
c .
170
‘ 5‘39 11 262 3‘1: r
0, I- v #1 ' JYILL v T VL... *fiL 4‘22 '.
50 100 150 200 250 300
H/Z
H533 SPECTRUM Data Illa: (100,1001X10119610 3-OCT-08 21:24
Sample: Blalhn 2mln 005 100-2-20-220-5-250-20-300-5
R? 5'40” 31 (Pos.) GC 450.6c DP: m/z 133.0000 Int. 19.5510 Lv 0.00
Scan. (300)
100
133 179 .
R J 2 7 ’
G- 1
1 00* -
1C5
a 75 .
c P
1 P
v 60* +
c P
A
b 40* ~
u < 1
fl
‘ 20‘51 13 '
fl ,
c 57 93 1 9 i 222 299 >
e ‘ . 1i!“ i HI' I: .1 I 253 231 314 1
031.. -. 1111.1 :1 l 1.1 . ".I | m. -11I Ill I 1 _
50 100 150 200 250 300

M/Z

209

GC chromatograph of 2-carboxybenzaldehyde in pure solvent

TIC Data Plle: (100,1001x09279601 27-SEP-96 11:42
sample: 2-C8A in CH2C12 085 100-2-20-220-5-260-20-300-5

Scan! 300 to 500(932) RT 4'29' to 7'29'(13'57‘) 81(906.) Lv 0.00

Operator: OQ

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Max 6972.56 3.3.
5 § 7 mag. abund.
100 ‘ * * * 1, ‘ 1 m ‘ ‘ ‘ ‘*
A 1 ‘1' I C
b
1.1
n 801
d 1
a 4
n 4
c 60*
Q
l
l
40*
20*
1
1
J .
0 - .
- - - . . - ~ - - . - - - - . . . . . ~1.0 8972.56
3°° 350 400 450 5
Scan
GC chromatograph of 2-carboxybenzaldehyde 1n srlylation reagent
21c Data 911.: (100,1001x09279602 27-snp-96 13:42
Sample: 2-caa 1n asrra oas 100-2-20-220-5-260-20-300-5
Scan. 300 to 500(500) RT 4'29' to 7'29'(7'29') 81(903.) Lv 0.00
Operator: OQ
uax 29762.9 3.2. b
7 mag. a nod.
110° 1 1 1,? - - - a 1.2 1A§11414321 - - - . - - 1 A
« 210
b
1| .
n 904
a 1
a 4
('1 1
c 60*
. 4
40*
20*
A _._.._~ L L «AA-win
. - - . , - 3+ . 3 . - - - - . - - - . ~1.o 29762.9

 

30° 350 400 450 500
Sean

210

Mass Spectrum of the silylation product XV

"A38 3920130» Data 011a: (100,1201x10119614 3-0CT-00 22:57
Samp}a: 0(a1An 10min 035 100-2-20-220-5-260-20-300-5
R? 5'54' 01 (Poe ) 0c 450 6c 09* ml: 73 000
. . . . 0 Int. 08.
3¢.n. ‘39.) 2278 Lv 0.00

 

 

100
73 267
a .
. 1
1 00*
. .
t
1
v 60* ’
o t
A
b 40* _
u
n
d
‘ F
n -
C
C
l A. L

 

 

 

 

 

H/Z

Mass spectrum of silylated 2-hydroxybenzoic acid obtained from NIST database (N IST
#: 64161, CAS #: 3789-85-3)

 

100. 73 l.
—i‘——
e O
\S/.
50 - /\
' 45
9 5 267
1
149 3209
O 3 1 “1.111 1 -1 1 111211 J 4.1
IIIIII[ITYIIIIIIITYYIITT—IIrI—TTIIITIIYIIIerI—TITITTIIUIIrYUTYl—V—mIIIIIIIIIIIIUIIITIAI

 

 

30 60 90 120 150 180 210 240 270
(M)Benzoic acid. 2-[(trimethylsilyl)oxyl-, trimethylsilyl ester

211

Mass Spectrum of the silylation product XVI

HA38-SPEC?RUH Data Plle. (100 1001x
. 10119610 - - ,
Sample: Blalhn 2mln 005 100-2-2o-22015-250-2o-300-5 3 OCT 80 21’2‘
RT 7'07' :1 (Po: 1 00 450 6c 02- m]: 73
' - - .0000 ,

 

 

 

 

100
R
,
e
l 80‘ ’
a , -
c
1
V 60‘
e 295
A
b 40*
U
n
a
a u
n ..
C .
e 310

 

 

 

 

H/Z

Mass spectrum of silylated 2-phthalic acid obtained from NIST database (NIST #: 71607,
CAS #: 2078-22-0)

100. 147
73

 

 

310
L. 1
IWIIIIIIIIIIITIIIIUIIU IIIAW
4O 70 100 130 160 190 220 250 280 310
(M)1,2-Benzenedicarboxylic acid, bis(trimethylsilyl) ester

 

 

 

 

 

”‘71111114711!@ltiiiwjflfiflfiflfl“