a“.

2.).

_.
.5

.~!v.}crdllol ‘

,._,._.‘.£§

We”: a.
.r... ,:K

.. 4.."

12.7 ‘r_x.s..13i u:

.. » v
, wtummm. 1».
V , . V 6&4},
3.”, .5.»
u r V . 4 ‘ .
.ﬁmmﬁﬁwn . . . z
3%? . . .

:uﬁub ‘ ,
a;

if!!!
:34 c6:
. I}. In...
3...
1.
id“:

‘ v tux-VI

 

in;

 

LIBRARY
Michiga. grate l
University

 

This is to certify that the
dissertation entitled

Degradation of polycyclic aromatic hydrocarbons (PAHs)
in aqueous media using alternating current

presented by

Emmanuel Samman Pepprah

has been accepted towards fulfillment
of the requirements for the

Ph.D. degree in Environmental Engineering

 

 

 

 

Major W5 Signature

l2 DH 26237”

Date

 

MSU is an affin'native-action, equal-opportunity employer

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 pJCIRC/DateDueindd-pt

DEGRADATION OF POLYCYCLIC AROMATIC HYDROCARBONS
(PAHs) IN AQUEOUS MEDIA USING ALTERNATING CURRENT

By

Emmanuel Samman Pepprah

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

2007

ABSTRACT

DEGRADATION OF POLYCYCLIC AROMATIC HYDROCARBONS
(PAHS) IN AQUEOUS MEDIA USING ALTERNATING CURRENT

By

Emmanuel S. Pepprah

Abiotic treatment techniques such as ozonation, hydrogen peroxide oxidation, UV
irradiation, chlorination electrochemical oxidation, sonochemical oxidation, etc. have
been extensively investigated and applied for the treatment of water and wastewater
containing organic compounds such as polycyclic aromatic hydrocarbons (PAHS). In
contaminated soils, biodegradation and surfactants are often employed as a treatment
technique to degrade or “wash” PAHs in situ. Published electrochemical studies have
focused on the use of direct current (DC) to degrade organics in solution. However,
relatively few studied have explored the potential of using alternating current (AC) to
degrade organics in solution. In this dissertation, AC is explored for the degradation of
PAHs in aqueous solutions. The effect of AC waveform shape, current density and
frequency, and supporting electrolyte on the rate of degradation of PAHs was
investigated to determine optimum conditions required for potential pilot-scale
implementation of the proposed treatment technique. Degradation rates and byproducts
formed were compared when AC, DC and ultrasound delivered at the same output power
were separately applied to spiked aqueous solutions containing PAHs. Degradation
byproducts and mechanisms were identiﬁed. Naphthalene, phenanthrene and pyrene were

used as the model PAH compounds.

The key ﬁndings of this study are: (1) Similar to DC, AC can degrade PAHs in
solution; (2) DC was more efﬁcient in degrading PAHs than AC at an equivalent root
mean square (rms) current density. However, potential ﬁeld-scale application using AC
would be cheaper for capital cost because no rectiﬁcation is required; (3) Degradation
rates increased with the increase in current density and decreased with increased AC
frequency; (4) For square, sine, and triangular AC waveforms explored, square wave AC
signal produced the fastest degradation at an equivalent peak current density; (5) Both
DC and ultrasound, at an equivalent output power exhibited comparable degradation rates
for naphthalene whereas, AC at 60 Hz showed relatively low degradation rate at the same
output power; (6) Similar to electrochemical degradation of PAHs using DC reported in
published studies, hydroxyl radicals were identiﬁed as the key oxidizing agents in both
sonochemical and electrochemical degradation using AC in this study; and (7) Both
diffusion and migration processes were responsible for the rate of mass transfer of PAHs
to the reaction sites (surface of electrodes or in the vicinity of the electrodes) where the
PAHs were degraded or converted by the hydroxyl radicals. However, migration was the
dominant mass transfer phenomenon responsible for the degradation of PAHs in the AC

and DC electrochemical cells.

Keywords: Electrokinetics, ground-water contamination, organic chemicals, PAH,
naphthalene, salicylic acid, phenanthrene, pyrene, alternating current (AC), direct current
(DC), AC frequency, AC waveform, electrolysis, electrochemical degradation,

sonochemical degradation, and hydroxyl radical.

DEDICATION

“In all things, give thanks to the Lord.” I therefore thank the Almighty God for giving me
the strength and wisdom to complete this work. I dedicate this dissertation to my

wonderful wife, Kesewa Pepprah, son, Michael Pepprah, and daughter, Chelsea Pepprah.

iv

ACKNOWLEDGEMENT

I am very grateful to my lovely wife, Kesewa Pepprah, my son, Michael Pepprah and
daughter, Chelsea Pepprah for their prayers, support and patience.

My sincere gratitude goes to my academic advisor, Dr. Milind V. Khire for his
guidance and support during my stay as a Ph.D student at Michigan State University.

I am indebted to my guidance committee members, especially Dr. Dan Jones for his
assistance in running mass spectrometry tests and identiﬁcation of degradation
byproducts. I also thank Dr. Hui Li and Dr. Volodymyr Tarabara for their useful
comments and suggestions.

I extend my utmost thanks to the Department of Civil and Enviromnental
Engineering, Michigan State University for partially funding my Ph.D program by
offering me teaching assistantship.

I am grateful to Mr. Joseph Leykam at the RTSF Macromolecular Core Facility for
his technical support during the analysis of samples, and Mr. James C. Brenton for his
technical assistance related to the fabrication of the test cells. I am also thankful to Prof.
Satish Udpa and Mr. Michael Shiu C. Chan for their help with the selection of the
function generator, oscilloscope, and the ampliﬁers.

This project has been funded by the US. National Science Foundation (NSF) Grant

No. BBS-0402772. However, this dissertation has not been reviewed by NSF.

TABLE OF CONTENTS

LIST OF TABLES .............................................................................................................. X
LIST OF FIGURES ......................................................................................................... XII
KEY OF ABBREVIATIONS ........................................................................................ XVI
KEY OF SYMBOLS .................................................................................................... XVII
INTRODUCTION ............................................................................................................... 1
BACKGROUND ON PAHS ............................................................................................ l
EXISTING REMEDIATION TECHNIQUES ................................................................ 3
ELECTROCHEMICAL DEGRADTION OF ORGANICS ............................................ 3
Type of Electrical Current ....................................................................................... 3

Target Organic Compounds ..................................................................................... 5
Electrodes ................................................................................................................. 7
Supporting Electrolyte ............................................................................................. 8
Analytical Techniques ............................................................................................. 9
DEGRADTION OF ORGANICS USING ULTRASOUND ........................................... 9
Ultrasound Equipment ............................................................................................. 9

Target Compounds ................................................................................................. 10
OBJECTIVES AND METHODOLOGY ...................................................................... 12
DISSERTATION ORGANIZATION ........................................................................... 14

PAPER NO. 1: ELECTROREMEDIATION OF NAPHTHALENE IN AQUEOUS

SOLUTION USING ALTERNATING AND DIRECT CURRENTS .............................. 16
ABSTRACT .................................................................................................................. 16
INTRODUCTION ......................................................................................................... 1 7
BACKGROUND ........................................................................................................... 18

DC Electrolysis ...................................................................................................... 18
AC Electrolysis ...................................................................................................... 20
OBJECTIVES ............................................................................................................... 21
MATERIALS AND EXPERIMENTAL METHODOLOGY ....................................... 21
Reaction Chamber .................................................................................................. 23
Electrical Equipment .............................................................................................. 24
Spiked Solution Preparation .................................................................................. 24
Supporting Electrolyte ........................................................................................... 25
Testing Procedure .................................................................................................. 27
Sampling ................................................................................................................ 29
Experimental Parameters Monitored ..................................................................... 30

vi

RESULTS AND DISCUSSION .................................................................................... 30

Supporting Electrolyte: NaCl ................................................................................. 31
Supporting Electrolyte: NaZSO4 ............................................................................. 38
DC Tests ................................................................................................................. 38
AC Tests ................................................................................................................. 40
Intermediate Byproducts ........................................................................................ 42
Temperature ........................................................................................................... 47
pH ........................................................................................................................... 47
Redox Potential and Dissolved Oxygen ...................................... - .......................... 49
Electrical Conductivity .......................................................................................... 50
Supporting Electrolyte: NaZSO4 ............................................................................. 50
SUMMARY AND CONCLUSIONS ............................................................................ 50

PAPER NO. 2: DEGRADATION OF PHENANTHRENE AND PYRENE IN

SOLUTION USING ALTERNATING CURRENT ......................................................... 53
ABSTRACT .................................................................................................................. 53
INTRODUCTION ......................................................................................................... 54
OBJECTIVES ............................................................................................................... 55
MATERIALS AND EXPERIMENTAL METHODOLOGY ....................................... 56

Reaction Chamber .................................................................................................. 58
Electrical Equipment .............................................................................................. 59
Spiked Solution Preparation .................................................................................. 59
Testing Procedure .................................................................................................. 61
Monitored Experimental Parameters ..................................................................... 63
RESULTS AND DISCUSSION .................................................................................... 64
Electrochemical Degradation of Phenanthrene in Solution ................................... 66
Effect of Current Density ....................................................................................... 66
Oxidizing Agents ................................................................................................... 68
Hydroxyl Radical Test ........................................................................................... 69
Degradation Kinetics ............................................................................................. 71
Effect of Type of Supporting Electrode ................................................................. 72
Effect of AC Frequency ......................................................................................... 74
Effect of AC Waveform ......................................................................................... 78
Electrode Fouling ................................................................................................... 80
Electrochemical Degradation of Pyrene in Solution .............................................. 80
Effect of Current Density ....................................................................................... 80
Effect of Type of Supporting Electrode ................................................................. 83
Effect of AC Frequency ......................................................................................... 84
Effect of AC Waveform ......................................................................................... 84
Modeled versus Measured Concentrations of Phenanthrene and Pyrene .............. 85
Electrode Mass Decay ............................................................................................ 93
Intermediate Byproducts ........................................................................................ 93
SUMMARY AND CONCLUSIONS ............................................................................ 97

vii

 

PAPER NO. 3: EFFECT OF ALTERNATING CURRENT WAVEFORM AND
FREQUENCY ON THE DEGRADATION OF NAPHTHALENE IN AQUEOUS

SOLUTION ........................................................................................................................ 99
ABSTRACT .................................................................................................................. 99
INTRODUCTION ....................................................................................................... 100
MATERIALS AND EXPERIMENTAL METHODOLOGY ..................................... 104

Reaction Chamber ................................................................................................ 106
Electrical Equipment ............................................................................................ 106
Spiked Solution Preparation ................................................................................ 107
Testing Procedure ................................................................................................ 108
Experimental Parameters Monitored ................................................................... 110
RESULTS AND DISCUSSION .................................................................................. I 10
Effect of AC Waveform ....................................................................................... l 12
Effect of AC Frequency ....................................................................................... 115
Hydroxyl Radical Probe Study ............................................................................ 122
Hydroxyl Radical Scavenger Study ..................................................................... 125
Modeled versus Measured Concentrations of Naphthalene ................................. 127
Effect of Convection Mass Flux .......................................................................... 133
Intermediate Byproducts ...................................................................................... 135
Reaction Mechanism and Pathway ...................................................................... 136
Divided Cell Setup ............................................................................................... 139
Measured Initial and Final Key Test Parameters ................................................. 142
SUMMARY AND CONCLUSIONS .......................................................................... 148

PAPER NO. 4: A COMPARATIVE STUDY OF SONOCHEMICAL AND
ELECTROCHEMICAL DEGRADATION OF NAPHTHALENE TN AQUEOUS

SOLUTION ...................................................................................................................... 15]
ABSTRACT ................................................................................................................ 151
INTRODUCTION ....................................................................................................... 1 52
MATERIALS AND EXPERIMENTAL METHODOLOGY ..................................... 155

Reaction Chambers .............................................................................................. 158
Electrical Equipment ............................................................................................ 158
Ultrasound Equipment ......................................................................................... 159
Spiked Solution Preparation ................................................................................ 159
Testing Procedure ................................................................................................ 160
Experimental Parameters Monitored ................................................................... 161
RESULTS AND DISCUSSION .................................................................................. 162
Effect of Power (AC, DC and Ultrasound) .......................................................... 163
Similarities between Sonochemical and Electrochemical Degradation ............... 165
Oxidizing Agents ................................................................................................. 165
Degradation Kinetics ........................................................................................... 169
Degradation Byproducts ...................................................................................... 169
Heat Generation ................................................................................................... 170
Mechanism for OH Radical Generation .............................................................. 17]
Reaction Site ........................................................................................................ 172

viii

 

Reaction Mechanism and Pathway ...................................................................... 173
Modeled versus Measured Concentrations of Naphthalene ................................. 179
SUMMARY AND CONCLUSIONS .......................................................................... 190

PAPER NO. 5: DEGRADATION OF NAPHTHALENE IN THE AQUEOUS PHASE

OF SPIKED SATURATED OTTAWA SAND .............................................................. 192
ABSTRACT ................................................................................................................ 192
INTRODUCTION ....................................................................................................... 193
MATERIALS AND EXPERIMENTAL METHODOLOGY ..................................... 194

Reaction Chamber ................................................................................................ 196
Electrical Equipment ............................................................................................ 196
Spiked Sand ......................................................................................................... 196
Testing Procedure ................................................................................................ 197
RESULTS AND DISCUSSION .................................................................................. 198
Comparison of Degradation Rates ....................................................................... 198
Comparison of Degradation Rates in Spiked Sand Using AC and DC ............... 202

Effect of Current Density ..................................................................................... 206

Effect of AC Frequency ....................................................................................... 207
Experimental Parameters Monitored ................................................................... 208
SUMMARY AND CONCLUSIONS .......................................................................... 209
SUMMARY AND CONCLUSIONS .............................................................................. 210
REFERENCES ................................................................................................................ 213

ix

LIST OF TABLES

PAPER NO. 1: ELECTROREMEDIATION OF NAPHTHALENE IN AQUEOUS
SOLUTION USING ALTERNATING AND DIRECT CURRENTS

Table 1-1: Experimental variables ........................................................... 25
Table 1-2: Measured initial and ﬁnal values of temperature, pH, standard redox
potential, dissolved oxygen concentration, and electrical conductivity of
test cell and control cell solutions ............................................................ 45

PAPER No.2: DEGRADATION OF PHENANTHRENE AND PYRENE IN
SOLUTION USING ALTERNATING CURRENT

Table 2-1: Summary of Experimental Parameters ......................................... 64
Table 2-2: Measured initial and ﬁnal values of temperature, pH, standard redox
potential, and electrical conductivity of test cell and control cell
solutions ........................................................................... 95

PAPER No.3: EFFECT OF ALTERNATING CURRENT WAVEFORM AND
FREQUENCY ON THE DEGRADATION OF NAPHTHALENE

IN AQUEOUS SOLUTION
Table 3—1: Summary of Experimental Parameters ....................................... 111
Table 3-2: Summary of Experimental Results ........................................... 118
Table 3-3: Summary of Measured Initial and Final Key Test Parameters ............ I43

PAPER NO. 4: A COMPARATIVE STUDY OF SONOCHEMICAL AND
ELECTROCHEMICAL DEGRADATION OF NAPHTHALENE

IN AQUEOUS SOLUTION
Table 4-1: Experimental Variables ......................................................... 162
Table 4-2: Summary of Measured Parameters ........................................... 188

PAPER NO. 5: DEGRADATION OF NAPHTHALENE IN THE AQUEOUS
PHASE OF SPIKED SATURATED OTTAWA SAND

Table 5-1: Summary of Experimental Parameters ....................................... 198

Table 5-2: Measured initial and ﬁnal values of temperature, pH, standard redox
potential, and electrical conductivity of test cells ........................... 204

xi

LIST OF FIGURES

PAPER NO. 1: ELECTROREMEDIATION OF NAPHTHALENE IN AQUEOUS

Figure 1-1:

Figure 1-2:

Figure 1-3:

Figure 1-4:

SOLUTION USING ALTERNATING AND DIRECT CURRENTS
Schematic of the experimental setup .......................................... 22

Change in concentration of naphthalene in solution for AC and DC tests
with NaCl as supporting electrolyte ............................................ 31

Change in concentration of naphthalene in solution for DC tests with
Na2S04 as supporting electrolyte .............................................. 39

Change in concentration of naphthalene in solution for AC tests with
Na2804 as supporting electrolyte .............................................. 41

PAPER NO. 2: DEGRADATION OF PHENANTHRENE AND PYRENE IN

Figure 2-1:
Figure 2-2:
Figure 2-3:

Figure 2-4:

Figure 2-5:

Figure 2-6:

Figure 2-7:

Figure 2-8:

Figure 2-9:

Figure 2-10:

Figure 2-11:

SOLUTION USING ALTERNATING CURRENT

Schematic of the experimental setup .......................................... 57
Change in concentration of phenanthrene in solution ........................ 66
Change in concentration of salicylic acid ..................................... 70

Change in concentration of phenanthrene with time using NaCl or
NaZSO4 as supporting electrolyte .............................................. 73

Effect of AC frequency on degradation rates of phenanthrene ............. 75

Temperature-time proﬁle of test solutions containing only DI water and
30/70 Acetonitrile/DI water ..................................................... 76

Effect of AC waveform shape on degradation rates of phenanthrene. . . ..79

Concentration of pyrene versus time for various current densities... ......81

Change in concentration of pyrene with time using NaCl or NaZSO4 as

supporting electrolyte ............................................................ 83
Effect of AC frequency on degradation rates of pyrene ...................... 84
Effect of AC waveform shape on degradation rates of pyrene ............. 85

xii

Figure 2-12:

Figure 2-13:

 

Simulated concentrations versus test data for phenanthrene tests using
DC .................................................................................. 88

Simulated concentrations versus test data for pyrene tests using DC. 89

PAPER No.3: EFFECT OF ALTERNATING CURRENT WAVEFORM AND

Figure 3-1(a):

Figure 3-1(b):

Figure 3-2:
Figure 3-3:
Figure 3-4:
Figure 3-5:

Figure 3-6:

Figure 3-7:

Figure 3-8:

Figure 3-9:

Figure 3-10:

Figure 3-11:

Figure 3-12:

FREQUENCY ON THE DEGRADATION OF NAPHTHALENE
IN AQUEOUS SOLUTION

Plan View of Hypothesized Funnel-and-Gate Electrochemical Reactive

Barrier ............................................................................ 101
Cross-section of Hypothesized F unnel-and-Gate Electrochemical Reactive
Barrier ............................................................................ 102
Schematic of the experimental setup .......................................... 105
Effect of AC waveform on the degradation rate of naphthalene .......... 1 13

Square; Sine; and Triangular AC signal having the same peak current..l 14
Effect of AC frequency on the degradation rate of naphthalene .......... 1 16

Effect of AC frequency on the pseudo-ﬁrst-order degradation rate
constants .......................................................................... 121

Effect of AC frequency on the rate of degradation of salicylic acid
(hydroxyl radical probe) ........................................................ 124

Reduced rates of degradation of naphthalene in solutions in the presence
of varying concentrations tertiary butanol (hydroxyl radical
scavenger) ........................................................................ l 26

Simulated concentrations vs test results for naphthalene DC tests........ 129

Effect of stirring on degradation rate of salicylic acid at different AC
frequencies ....................................................................... 1 34

Simpliﬁed electrochemical degradation pathway of naphthalene in aqueous
solution ........................................................................... 137

Change in concentration of naphthalene with time in a divided cell setup
using AC and DC. .............................................................. 140

xiii

Figure 3-13: Change in concentration of salicylic acid with time in a divided cell setup
using AC and DC ............................................................... 142

PAPER NO. 4: A COMPARATIVE STUDY OF SONOCHEMICAL AND
ELECTROCHEMICAL DEGRADATION OF NAPHTHALENE

IN AQUEOUS SOLUTION
Figure 4-1: Schematic of experimental setup for sonochemical degradation of
naphthalene in solution ......................................................... 156
Figure 4-2: Schematic of experimental setup for electrochemical degradation of
naphthalene in solution ............................................................ 157
Figure 4-3: Concentration-time proﬁle for both sonochemical and electrical
degradation of naphthalene in solution at 30 W and 60 W output
power .............................................................................. 163
Figure 4—4: Concentration-time proﬁle for both sonochemical and electrical
degradation of salicylic acid (OH radical probe) in solution at 30 W and
60 W output power ............................................................ 167
Figure 4—5: Pseudo-ﬁrst-order degradation rate constants with respect to naphthalene
and salicylic acid versus power ................................................ 168
Figure 4-6: Simpliﬁed sonochemical and electrochemical degradation pathway of
naphthalene in solution with OH radicals as the primary oxidizing
agent ............................................................................... 175
Figure 4-7: Simpliﬁed alternative electrochemical degradation pathway of
naphthalene in solution to naphthol with radical cation precursor ........ 178
Figure 4-8: Simulated concentrations versus test data for naphthalene tests using
DC ................................................................................. 182
Figure 4-9: Simulated concentrations versus test data for salicylic acid tests using
DC ................................................................................. 184

PAPER No.5: AQUEOUS PHASE DEGRADATION OF NAPHTHALENE IN
SPIKED SATURATED OTTAWA SAND

Figure 5-1: Schematic of the experimental setup .......................................... 195
Figure 5-2: Comparison of electrochemical degradation of naphthalene in spiked
solution and sand using AC ................................................... 199

xiv

Figure 5-3:

Figure 5-4:

Figure 5-5:

Figure 5-6:

Comparison of electrochemical degradation of naphthalene in spiked
solution and sand using DC ................................................... 201

Comparison of electrochemical degradation of naphthalene in spiked sand
using AC and DC ............................................................... 202

Effect of current density on electrochemical degradation of naphthalene in
spiked sand using AC .......................................................... 207

Effect of AC frequency on electrochemical degradation of naphthalene in
spiked sand using AC .......................................................... 208

XV

KEY OF ABBREVIATIONS

AC = Alternating Current

API = Atmospheric pressure ionization

APCI = Atmospheric Pressure Chemical Ionization
ATSDR = Agency for Toxic Substance and Disease Registry
DC = Direct Current

D1 = de-ionized

D0 = Dissolved Oxygen

ESI = Electrospray Ionization

FERB = F unnel-and-gate Electrochemical Redox Barrier
HLB = Hydrophilic Lipophilic Balance

HPLC = High Performance Liquid Chromatography

LC = Liquid Chromatography

LCR = Inductance-Capacitance-Resistance

MCL = Maximum Contaminant Level

MDL = Method Detection Limit

MGP = Manufactured Gas Plant

MS = Mass Spectrometry

OH' = Hydroxyl Radical

PAH = Polycyclic Aromatic Hydrocarbon

RMS = Root Mean Square

USEPA = United States Environmental Protection Agency

UV = Ultraviolet

xvi

KEY OF SYMBOLS

C = Concentration at time (t).

CO = Initial Concentration
D = Diffusion coefﬁcient
E = Electrical potential

f = AC frequency

F = Faraday’s constant

j = current density

J = Mass ﬂux

k = decay constant
lp = Peak Current

Irms = Root Mean Square Current
R = Molar gas constant
T= Temperature in Kelvin

v = Linear velocity of solution

VI) = Peak Voltage

Vrms = Root Mean Square Voltage
x = Distance in the x direction

2 = charge on a reacting species

Z = Impedance

xvii

INTRODUCTION

BACKGROUND ON PAHS

Polycyclic Aromatic Hydrocarbons (PAHS) are a group of aromatic compounds
consisting of two or more fused benzene rings. PAHS have been contaminants of
concerns (COCS) over the past few decades because their toxicity and tendency to
contaminate air, water, and soil. They are formed by incomplete combustion of coal, oil,
or gas and organic matter such as grilled meat or tobacco. PAH contamination in the
environment originates from sources such as volcanoes, forest ﬁres, burning coal,
automobile exhaust, discharges from industrial or wastewater treatment plants, or
evaporation into air from contaminated soil or water.

PAHS occur in the atmosphere in both the solid phase (as particulates) and in the
vapor phase. Three-ring PAH compounds are found in the atmosphere primarily in the
gaseous phase, whereas, ﬁve- and six-ring PAHS are found mainly in the solid phase;
four-ring PAH compounds are found in both phases (ASTDR 1995). Basu and Saxena
(1978a) reported concentrations of selected PAHS in surface waters used as drinking
water sources in four US. cities (Huntington, West Virginia; Buffalo, New York; and
Pittsburgh and Philadelphia, Pennsylvania). Total concentrations of PAHs ranged from
4.7 ng/L in Buffalo to 600 ng/L in Pittsburgh. Data summarized by Sorrel et al. (1980)
indicate low levels of PAHS in ﬁnished drinking waters of the United States. Reported
maximum concentrations for total PAHS (based on measurement of 15 PAHS) in the
drinking water of 10 cities ranged from 4 to 24 ng/L; concentrations in untreated water
ranged from 6 to 125 ng/L. The relatively low concentrations of PAHs in ﬁnished

drinking water were attributed to efﬁcient water treatment processes as well low water

 

:cncerrtratio
tithe Xatlo
the range of
this et al.
ZéPP-Jli ('4
:tmcentrati

1:11 l'plt-cr...

.“ . -
‘II‘TOXI-m v

.‘IK‘L\

 

solubility of PAHS. Shiraishi et al. (1985) found PAHS in tap water at concentrations
ranging from 0.1 to 1.0 ng/L, primarily as chlorinated derivatives of naphthalene,
phenanthrene, ﬂuorene, and ﬂuoranthene. Basu and Saxena (1978b) reported total PAH
concentrations in groundwater from three sites in Illinois, Indiana, and Ohio to be in the
range of 3 to 20 ng/L. PAHs have been detected in urban runoff generally at
concentrations much higher than those reported for surface water. Data collected as part
of the Nationwide Urban Runoff Program indicate concentrations of individual PAHs in
the range of 300 to 10,000 ng/L, with the concentrations of most PAHS above 1,000 ng/L
(Cole et al. 1984). Industrial efﬂuents also have elevated PAH levels. Morselli and
Zappoli (1988) reported elevated PAH levels in reﬁnery waste waters, with
concentrations for most PAHS in the range of 400 ng/L (benzo[b]ﬂuoranthene) to 16,000
ng/L (phenanthrene). Staples et al. (1985) reported median concentrations in sediment of
approximately 500 ug/kg (dry weight) for 15 PAHS.

Human exposure to PAHs usually occurs by breathing air contaminated by wild
ﬁres or coal tar, or by eating foods that have been grilled. PAHS have been found in at
least 600 of the 1,430 National Priorities List (NPL) sites identiﬁed by the United States
Environmental Protection Agency (USEPA) (ASTDR 1995). PAHS are classiﬁed as
potential carcinogens. Examples of PAHS include naphthalene (two benzene rings),
phenanthrene (three benzene rings), anthracene (three benzene rings), pyrene (four

benzene rings), benzo[a]pyrene (ﬁve benzene rings).

EXISTINI

hale]
. Bittl'Cj

COIIIJ,’

 

 

l.“ L
l~,,
kicLN 'tll’cn'
~d"!
J P (I; .
n.“ [U :\

 

EXISTING REMEDIATION TECHNIQUES

Several potential remediation technologies for cleaning contaminated water and soils
have been investigated at the laboratory scale and some of them have been applied in
pilot-scale or ﬁeld-scale settings. These technologies are summarized as follows.

0 Granular activated carbon (GAC) adsorption techniques involve transferring
contaminants from a liquid phase to a solid phase by sorption. In this process, no
reactions take place and hence there is no degradation of the CDC;

0 Permeable reactive barriers (PRBs) with zero valent iron, limestone, compost, etc.
have been used to treat contaminated groundwater;

o Bioremediation involves the use of microorganisms to degrade organic
contaminants;

0 Advanced oxidation treatments such as ozonation, hydrogen peroxide oxidation,
UV irradiation, photochemical reactions, and F enton oxidation or a combination of
some of the above involves the generation of highly oxidizing OH radicals (OH')
to degrade contaminants in solution; and

o Membrane processes such as microﬁltration, nanoﬁltration, and reverse osmosis
are used to remove particles in water that are too small for conventional ﬁlters to

remove.

ELECTROCHEMICAL DEGRADTION OF ORGANICS

Type of Electrical Current
Electrochemical methods typically involve electric current across electrodes in aqueous

media to generate redox reactions. Numerous studies (e.g. Alshawabkeh and Sarahney

 

"""Vﬁm 1,,-
EK‘I-uluii 1‘1”

 

 

 

 

2005; Li et al. 2005; Goel et al. 2003; Stock and Bunce 2002; Saracco et al. 2000;
Comninellis 1994) and applications (Kreysa G., Heitz E. 1986) on electrochemical
degradation methods have focused on the use of direct current (DC) to degrade organic
contaminants in solution. However, only a few studies (Chin and Cheng 1985; Nakamura
et al. 2005) have explored the use of alternating current (AC) to degrade organic
contaminants in solution. Reported applied currents have ranged between as low as 1 mA

(Alshawabkeh and Sarahney 2005) and as high as 10,000 mA (Saracco et al. 2003)

. . . 2 2 . .
corresponding to current densmes of 0.075 mA/cm and 50 mA/cm . Current densrtles

applied in electroremediation of soils are commonly in the range 0.025 — 5 A/m2 (Acar et

al. 1991; Hamed et al. 1991).

Majority of work on electroremediation has been conducted on laboratory scale,
although some pilot—scale tests have been conducted (Gibbs 1966; Acar and
Alshawabkeh 1996; Ho et al. 1997). Other examples of ﬁeld trials are outlined in a
review by Will (1995). A number of problems not encountered in the laboratory arise at
actual sites, e.g. the presence of large objects (> 10 cm) buried in the soil (Lageman et al.
1989; Lagemen 1993). Both conducting materials, such as metal objects, and resistive
substances, such as concrete, will disrupt the normal current path (Page and Page 2002).
Before ﬁeld trails can be performed, site investigations are necessary to locate such
objects and to provide information on the nature and distribution of contaminants and
characteristics of the soil (Lindgren et al. 1998).

For molecular substances, electro-osmosis is the predominant form of transport
under an electric ﬁeld, although partial dissociation into ions of some organic molecules

will result in electromigration as well (Page and Page 2002). Lageman et al. (1990) has

 

 

demonstraz;
,~ V'II.
leasltic. R

ﬁner-"- -.r
Lklutdii 01nd: -

t; Iv --- 0‘
tpp..tatlor.

Current [I

 

demonstrated that the removal of a range of aromatic organic compounds from soils was
feasible. Reddy and Saichek (2004) conducted laboratory investigation to determine
contaminant mass removal from low permeability clay, by using periodic voltage

application and surfactants.

Current Efﬁciency

Stock and Bunce (2002) applied 5, 10, 20, 50, 100 and 200 mA currents to study the
dechlorination efﬁciency of atrazine and reported that when applied currents 2 20 mA led
to Z 93% loss of initial atrazine within 30 minutes. The current efﬁciency decreased as
applied current increased and was greatest (16%) at 10 mA. Yusem and Pitauro (1992)
also reported a decrease in current efﬁciency with increased applied current in

electrocatalytic hydrogenation of soybean. Stock and Bunce (2002) attributed this trend

to the fact that a greater proportion of adsorbed hydrogen (Hads) was lost to unproductive

reaction of hydrogen gas evolution instead of participating in the dechlorination reaction
of atrazine. At high current densities, the rate of electrochemical generation of adsorbed
hydrogen becomes faster than the rate at which the substrate can migrate to the electrode

surface and become adsorbed.

Target Organic Compounds

Rajeshwar et al. (1994) and Rodgers et al. (1999) have demonstrated that electrochemical
degradation can be used to treat waste water containing toxic and non-biodegradable
organic compounds. Among others, organic compounds that have been used in published

electrochemical degradation studies include:

 

 

' Air.

 

 

"\_‘

37".”. I? ‘ CA
111 ha“).-

9‘. . J.

\I\ irt‘ut"? .

o Naphthalene, with concentrations ranging between 8.3 mg/L to 35 mg/L
(Alshawabkeh and Sarahney 2005; Goel et al. 2003);

o Phenol, with concentrations ranging between 100 mg/L to 9,400 mg/L (Li et al.
2005; Chin and Cheng 1985);

o Coumaric acid, with concentrations ranging between 10 mg/L to 400 mg/L
(Saracco et al. 2003);

o Atrazine, with concentrations ranging between 4 mg/L to 83 mg/L (Stock and
Bunce 2002);

o 2,4-dichlorophenoxyacetic acid, with concentrations ranging between 55 mg/L to
332 mg/L (Yasman et al. 2004);

o 2,4-dichlorophenol, with concentrations ranging between 57 mg/L to 244 mg/L
(Yasman et al. 2004);

o Salicylic acid at a concentration of 1,000 mg/L (Marselli et al. 2003);

o 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) at a concentration of 1,130 mg/L
(Marselli et al. 2003); and

o Trichlorobenzene, at a concentration of 119 ug/L (Nakamura et al. 2005).

Degradation By-products

Sonochemical degradation of organics in solution generally occurs through two distinct
degradation pathways: oxidation by hydroxyl radicals and pyrolytic decomposition
(Thompson and Doraiswamy 1999). In electrochemical oxidation, organic compounds in
aqueous solutions can be oxidized on an anode (anodic oxidation) by indirect oxygen
atom transfer and direct electron transfer (Rodgers et al. 1999; Kirk et al. 1985). During

the indirect oxygen atom transfer, oxygen radicals, especially the hydroxyl radicals

 

generated '
ai 3001; 5
readil} re;
cause the r.

The

'C‘IIGLIIIDI‘. I"

...A‘n a:
Mutt-em l

v {I‘D
M.‘
“av-Ii:
“MIR”.
:.

 

generated from water electrolysis, help in the oxidation of organic substances (Iniesta er
al. 2001; Simond et al. 1997; Comninellis 1994; Lee et al. 1981). The hydroxyl radicals
readily react with the organic molecules adsorbed on or in the vicinity of the anode to
cause the reaction presented below in Eq. 3 (Simond et al. 1997).

The formation of quinones has been reported by researchers who also observed
aromatic quinones during electrolytic oxidation of aromatic hydrocarbons (Brillas et al.
2000; Oturan 2000; Panizza et al. 2000). GC/MS analysis by Goel et al. (2003) indicated
the presence of 1,4-naphthoquinone when naphthalene solutions were electrolyzed and
concluded that naphthalene degradation could be due to direct anodic oxidation,
oxidation by an intermediate (e.g. hydroxyl radicals), or a combination of both. With
sufﬁcient OH' radicals in the reaction medium, ring cleavage may occur leading to the
formation of carboxylic acids. Li et al. (2005) reported that benzoquinone formed from
the oxidation of phenol could be degraded with ring breakage to various carboxylic acids.
One of the proposed mechanisms involves the adsorption of benzoquinone onto the anode
surface, which gives up an electron, and the carbon that is double-bonded with oxygen is
attacked by a neighboring hydroxyl radical. When this process is repeated at the para
position, the ring could be opened and benzoquinone would be broken down into small

organic molecules such as carboxylic acids (Houk er al. 1998; Lund and Baizer 1991).

Electrodes
Various types of electrodes that have been used in electrochemical degradation studies
including:

0 Titanium core with mixed metal coating with dimensions 10.2 cm by 1.3 cm by

0.12 cm (Alshawabkeh and Sarahney 2005);

Stainless steel cathode (6.35 cm by 3.18 cm by 0.07 cm) and titanium anode with
mixed metal oxide anode (6.35 cm by 3.18 cm by 0.07 cm) separated by a

distance of 1.4 cm (Goel et al. 2003);

Ti/SnOz-Sb, Ti/Rqu, and Pt anodes with dimensions 3 cm by 2 cm by 0.15 cm

and stainless steel plates of the same dimensions were used as cathodes. The

separation between the electrodes was 0.8 cm (Li et al. 2005);

Boron-doped diamond and lead oxide anodes, 20 cm2 area was exposed to the

electrolyte (Panniza and Cerisola 2003);

Platinized titanium anode and stainless steel plate as cathode, the electrode
surface area was 200 cm2 (Saracco er al. 2000);
Ti-Ru-Sn-SbOz, Ti/Pt and carbon electrodes (Panizza et al. 2000);

Reticulated vitreous carbon (RVC) (Stock and Bunce 2002); and

Ti/SnOz-Sb and Ti/IrOz anodes with 4 cm2 surface area and the cathode was a

platinum spiral enclosed in a 10 ml porous porcelain pot (Comninellis 1994).

Supporting Electrolyte

Salts are normally added to reacting media as supporting electrolyte to increase the

electrical conductivity in order to reduce the voltage required to produce a given current

value. Some supporting electrolytes that have been used in published electrochemical

studies include:

Sodium chloride, 50 ml of 0.2 M NaCl was used (Alshawabkeh and Sarahney

2005); 0 up to 10 g/L (Panizza er al. 2000);

Analytical
.traluiea

Identih d;

degradstér
lHPLC l. 1:
lIC .\l S l.
Carbon
pr‘r

\ l ("phi

Crating»q

 

 

0 Sodium sulfate (NaZSO4), 1.42 g/L (Goel et al. 2003); 35.5 g/L (Li et al. 2005);

2.84 g/L (Saracco er al. 2000); 21.3 g/L (Stock and Bunce 2002);

0 Sodium nitrate (NaNO3) (Alshawabkeh and Sarahney 2005);

o Lithium chloride (LiCl), 6.36 g/L (Stock and Bunce 2002);

0 Sodium hydroxide (NaOH), 0.8 g/L (Nakamura et al. 2005); and

o 1 M sulfuric acid (H2SO4) (Stock and Bunce 2002; Chin and Cheng 1985).

Analytical Techniques

Analytical techniques that have been used to quantify concentrations and to qualitatively
identify degradation by products of organic contaminants in sonochemical and electrical
degradation studies include among others: High Performance Liquid Chromatography
(HPLC), Gas Chromatography (GC), Liquid Chromatography / Mass Spectrometry
(LC/MS), Gas Chromatography / Mass Spectrometry (MS), LC/MS/MS, Total Organic
Carbon (TOC) Analyzer, Chemical Oxygen Demand COD) Analyzer,
Spectrophotometer, Electron Spin Resonance (ESR), Micellar Electrokinetic Capillary

Chromatography (MECC), and Solid-Phase Micro-Extraction (SPME).

DEGRADTION OF ORGAN ICS USING ULTRASOUND

Ultrasound Equipment
Various types of ultrasound equipment have been used in sonochemical degradation of

organics. These include:

 

Target 11

0193m-

among .
. .-
' '.
O I ~
' l

 

An Ultrason 250 (LabPlant Ltd., UK) operating at a frequency of 80 Hz, and UP
400S (Dr Hielscher GmbH, Germany) operating at a frequency of 24 Hz.
Ultrasound energy was delivered through hom-type probes (Psillakis et a] 2004).
An IKA Labortechnik IKASONIC U50 control unit operating at a frequency of 30
Hz. Ultrasound energy was delivered via titanium probes (Little er al. 2002).

An ultrasonic transducer UES 1.5 — 660 pulsar (Ultrasonic Energy Systems Inc.)
operating at 640 — 690 Hz frequency (Kim et al. 2002).

Three ultrasound systems, a vibrating plate reactor (VPR, 358 Hz), a near ﬁeld

acoustical processor (NAP, 16 & 20 Hz), and radial tube resonator (RTR, 20 Hz).
Ultrasound energy was delivered through transducers with areas 25 cm2, 1262
cm2, and 376 cm2, for VPR, NAP, and RTR respectively (Hung et al. 2002).

A Misonics sonicator operating at a frequency of 20 kHz. The ultrasound energy

was delivered through a hom-type titanium probe (Goskonda er al. 2002).

Target Compounds

Organic compounds that have been used in sonochemical degradation studies include,

among others:

Methyl tert-Butyl Ether (MTBE) at a concentration of 37 mg/L (Hung et al.
2002);

2,4-dinitrophenol at a concentration of 20 mg/L (Guo et al. 2005);

Phenanthrene, at a concentration of 1.25 mg/L (Wheat and Tumeo 1997);

Biphenyl at concentrations ranging between 0.6 mg/L (Little et al. 2002) and 1.25

mg/L (Wheat and Tumeo 1997);

10

 

o All 16 PAHS listed by USEPA at total initial concentration of 480 ug/L (Psillakis

et al. 2003); and
0 Organic pollutants in landﬁll leachate, with initial COD of 4770 mg/L (Wang et

al. 2005).

11

 

 

OBIEC'

The ke}
OI PAH:

L. . l-
n .2 'W .
r‘lnanrlsu

h:

.' n -
3.11:0»

v\4t—us

on esred

1
.3 -~. W ,.'
bICLIrL‘GC\

oxid;:1._~.n .

M

1:16 of Lie;

(1‘

nerg} 1;“

 

 

 

OBJECTIVES AND METHODOLOGY

The key objectives of this dissertation were to: (l) evaluate electrochemical degradation
of PAHS in spiked aqueous solutions using an alternating current (AC). Naphthalene,
phenanthrene and pyrene were used as model PAH compounds; (2) evaluate
electrochemical degradation of PAHS in saturated spiked Ottawa sand using AC; (3)
investigate the effect of current density, AC waveform shape and frequency on the
degradation rates of naphthalene; (4) compare the efﬁciency of DC and an equivalent AC
on degradation rates of PAHs; (5) prove that anodic oxidation of PAHS occurred on both
electrodes when AC electrolysis was used unlike DC electrolysis in which anodic
oxidation of naphthalene occurred on only one electrode (the anode); and (6) compare the
rate of degradation of PAHS in aqueous solution using ultrasound energy and electrical
energy (AC and DC) delivered at the same output power.

Laboratory scale tests were conducted in order to achieve the objectives listed
above. The test cell setup consisted of a 1 liter Pyrex glass beaker with a Teﬂon cap.
Tests were conducted in batch modes in undivided cell as well as divided cell setup with
l-liter solution in each compartment. These materials of the cell were selected to
eliminate or reduce sorption of the contaminants onto the walls of the reaction vessel and
the cap. Untreated Titanium plates were used as electrodes for this study. A 0.1 Hz to 2
MHz Sweep Function Generator was used to generate the AC signals. The power supply
consisted of a bipolar operational power supply/ampliﬁer capable of producing 200 V
AC/DC and 1 A AC/DC output or a maximum electrical power of 200 W. Voltage and

current measurements were obtained with the aid of an oscilloscope. The ultrasound

l2

 

 

the colle.

solutions

CO m DD (m \:
salmon-5 (

concentra:

 

 

him} 50‘.

IIICIQaSQ III

equipment used was a Sonics VCX 750 that operated in a continuous mode at a ﬁxed
frequency of 20 kHz.

Aqueous solutions, spiked with naphthalene were prepared by adding naphthalene
(98% purity) crystals to deionized (DI) water in a capped Pyrex bottle and allowing the
mixture to stand. The excess suspended naphthalene crystals were then ﬁltered off and
the collected ﬁltrate was used for the tests. The average concentration of naphthalene
solutions used in this study was 20 mg/L (~0.15 mM) so that adequate amounts of
naphthalene would be available in solution for quantiﬁcation of concentrations and
identiﬁcation of degradation byproducts. Due to relatively low solubility of phenanthrene
and pyrene in water, a cosolvent (acetonitrile) was used to dissolve phenanthrene and
pyrene into solution at concentrations greater than the aqueous solubility of the
compounds. 10 mg or 5 mg/L of phenanthrene or pyrene crystals were added to one liter
solutions of acetonitrile (30%) and deionized (DI) water (70%) (v/v) to achieve target
concentrations of 10 mg/L or 5 mg/L, respectively. The proportion of acetonitrile was
just enough to ensure that the PAH with the lower solubility, pyrene, remained in the
binary solution without precipitating out to form crystals in the solution. In order to

increase the electrical conductivity of the electrolyte, NaCl and anhydrous sodium sulfate

(Na2SO4) were added as the supporting electrolytes in separate sets of experiments. 0.5 g

of sodium chloride (NaCl) or sodium sulfate (Na2S04) crystals were added to 1 liter test

solution to produce 500 mg/L NaCl or Na2S04 solution.

Prior to passing AC through the cells, initial samples were taken for high

performance liquid chromatography (HPLC) analyses. During the tests, samples were

13

 

 

taken for

samples '

malt ses :
cells etc-e

at the 53.":

 

 

 

 

 

DISSERI

 

“thihalgr

I

taken for HPLC analyses at elapsed time intervals of 1, 2, 4, 8, 24, and 48 hours. The
samples were analyzed directly without extracting the PAH from the aqueous solution.
Samples were also collected for Liquid Chromatography / Mass Spectrometry (LC/MS)
analyses to determine the degradation products formed. Control cells, similar to the test
cells except without any current passed through them were set up, sampled, and analyzed

at the same time intervals as the test cells.

DISSERTATION ORGANIZATION

This dissertation has been has been organized into ﬁve sections. The ﬁrst paper compares
the degradation of naphthalene in spiked aqueous solution using AC and DC, as well as
the effect of current density and the supporting electrolyte on the rates of degradation for
naphthalene. The second paper investigates the electrochemical degradation of
phenanthrene and pyrene in 30% acetonitrile and 70% de-ionized (D1). The third paper
addresses the effect of AC waveform and frequency on the degradation rates of
naphthalene in spiked aqueous solution. Hydroxyl radical probes and scavengers were
used to demonstrate that hydroxyl radicals were the primary oxidizing agents in both
electrochemical and sonochemical degradation of organics. A divided cell setup was used
to prove that anodic oxidation of PAHS occurs on both electrodes in AC electrolysis,
unlike DC electrolysis in which anodic oxidation occurs on only one electrode (the
anode). Effect of current density and AC frequency is evaluated in this paper using
experimental data as well as modeling the degradation of the PAHS using l-D diffusion
and migration equation. The fourth paper presents a comparative study of the degradation

of naphthalene in spiked aqueous solution using electrical (AC and DC) energy and

14

 

ultra.» ;I'

533.1111 ti

 

 

 

ultrasound energy. The ﬁfth paper presents the degradation of naphthalene in spiked

saturated Ottawa sand using AC and DC.

15

 

PAPE
AQI

 

ABSTR.

fete It'd}
[be deg:
of current
1563110 e
sulfate l ‘
if-Pilltdut'

’1 LQL,‘: '
nhp‘.u‘d .a

D)“ c '
urhl.ld}\
Q“ ‘

6““ oi

 

 

PAPER NO. 1: ELECTROREMEDIATION OF NAPHTHALENE IN
AQUEOUS SOLUTION USING ALTERNATING AND DIRECT
CURRENTS

ABSTRACT

The key objectives of this study were to evaluate the use of alternating current (AC) for
the degradation of naphthalene in spiked aqueous solutions and to investigate the effect
of current density on the degradation rates of naphthalene. Direct current (DC) was also

used to compare the rates of degradation. Sodium chloride (NaCl) and anhydrous sodium

sulfate (Na2804) were used as the supporting electrolytes. Degradation rates and

byproducts formed were investigated when DC and AC were separately passed through
naphthalene solutions. A square wave AC, having a frequency of 0.1 was used.

Naphthalene solutions having an initial concentration of about 20 mg/L (~0.15 mM) were

subjected to an AC peak. current density of 6 mA/cm2 and DC density of 6 mA/cmz,

using NaCl as the supporting electrolyte. About 70% reduction in concentration of the
naphthalene in solution was observed after a period of 48 hrs when DC was applied. The

effect of current density on the electrochemical degradation rate of naphthalene in

aqueous solution was also investigated at current densities of 1 mA/cm2 (AC; DC), 3
mA/cm2 (AC; DC), and 6 mA/cm2 (AC; DC) using NaZSO4 as the supporting electrolyte.

AC peak current densities of 1, 3, and 6 mA/cm2 resulted in overall conversions of 77%,

87% and 95%, respectively, of naphthalene in solution and the corresponding values for

DC application were 95% for all current densities except the initial degradation rates

16

 

 

 

“ere h:

h}dt0X}I

INTROI

Pohc}el

across Ult

‘lli‘dmllieg

 

 

 

were higher for higher DC densities. Based on the degradation products formed,

hydroxylation is believed to be the key mechanism for the degradation of naphthalene.

INTRODUCTION

PolycyClic aromatic hydrocarbons (PAHS) are a group of ubiquitous and nonionic organic
chemicals that are of great environmental concern due to their toxicity and suspected
carcinogenicity (Fernandez et al. 2000). PAHS are a class of organic compounds that
consist of two or more fused benzene rings and/or pentacyclic molecules that are
arranged in various chemical conﬁgurations (Bamforth and Singleton 2005). According
to Lee et al. (1981), PAHS are the largest class of chemical carcinogens. Clar (1964) and
Harvey (1991) have reported in detail about the evidence of PAH carcinogenicity in
animals. In general, the greater the number of benzene rings, the higher the toxicity of the
PAH (Cemiglia 1992). The United States Environmental Protection Agency (USEPA)
estimates that 3,000 to 5,000 former manufactured gas plant (MGP) sites are located
across the United States and many of these contaminated sites contain substantial
quantities of PAHS (USEPA 2000). The USEPA maximum contaminant level (MCL) for
benzo[a]pyrene (PAHS) is 0.2 ug/L (epa.gov). PAHS have been found in some drinking
water supplies in the United States and background levels of PAHS in drinking water
range from 4 ng/L to 24 ng/L (ATSDR 1995).

Remediation of contaminated soil and groundwater is an important issue that is of
concern among environmentalists and hydrogeologists. In-situ and ex-situ treatment
techniques may be used to remediate contaminated soil and groundwater. Several

approaches including biodegradation and advanced oxidation processes have been used to

17

 

 

JIV ‘7 P
d\ if» v
-

3Q‘JCI
meﬁl

medr

II\(I

Ihed
t not
condu

serum.

DC El

ﬂecng

If.

'/

 

 

degrade PAHS in solution. Electrochemical degradation of organic contaminants in

aqueous solution is an emerging remediation technique. Existing electrochemical redox
methods depend on passing a direct current (DC) through electrodes in an aqueous

medium.

BACKGROUND

The degradation of organics such as PAHS, phenol, benzene etc. has been investigated by
various researchers using DC electrolysis. However, relatively few studies have been
conducted where AC electrolysis was explored for the degradation of organics in

solutions.

DC Electrolysis

Alshawabkeh and Sarahney (2005) investigated the effect of DC density on
electrochemically enhanced transformation of naphthalene in solution. A divided cell
setup, connected by a glass tube was used. Each compartment was a 550 m1 glass bottle
with a Naﬁon membrane in the connecting glass tube physically separated the
electrolytes in the two compartments. The electrodes were made up of titanium mesh
with mixed metal oxide coating. The electrodes were 10.2-cm-long by 5.0-cm-wide by
0.12-cm-thick. 50 m1 NaCl solution (0.2 M) was used as supporting electrolyte to
increase the electrical conductivity of the electrolyte. Three DC densities: 1 mA/L, 9
mA/L, and 18 mA/L were used to explore naphthalene transformation at two initial
concentrations (13 mg/L and 25 mg/L). Both concentration levels depicted increase in

transformation rates with increase in the current density. The transformation rate of

naphthalene was enhanced by electro-generation of C12(g) from C1. in solution. The pH of

18

 

the timely
potential 2
mA I CUT

l'eur b} pr

GO:
Initial co?

Sill-.1) \\ er

Stainless ~

 

the anolyte dropped from its initial value of about 5.5 to 1.9 and the corresponding redox
potential increased from about 500 mV to 1380 mV at the end of the test period when 9
mA/L current density was used. Transformation of naphthalene at the anolyte produced
four byproducts as indicated by relatively small peaks on the chromatogram.

Goel et al. (2003) explored the degradation of naphthalene in solution (having an
initial concentration of about 9.8 i 1.5 mg/L using DC. The reactor vessels used for this
study were 500 ml amber-colored bottles undivided cells. The electrodes consisted of a

stainless steel cathode and a 6.35-cm-Iong by 3.18-cm-wide by 0.15-cm-thick titanium
mesh anode with mixed metal oxide coating. 1.42 g/L sodium sulfate (Na2S04) (0.01 M)
was used to increase the conductivity of the electrolyte. The solutions were electrolyzed

at current densities ranging from 1.2 mA/cm2 to 9.9 mA/cmz. In this study, naphthalene

degradation rates were insensitive to current density and independent of the system pH.
1,4-naphthoquinone was identiﬁed as a byproduct of the electrolysis.
Li et al. (2005) conducted laboratory experiments on the kinetics and pathway of

electrochemical degradation of phenol having initial concentrations equal to 100 and 490

mg/L using three materials for the anode: Ti/SnOz-Sb, Ti/Rqu, and Pt at a current

density of 20 mA/cmz. The tests were conducted in batch electrolysis using 100 ml glass

beakers and the electrodes were 3-cm-long by 2-cm-wide by 0.15-cm-thick. 0.25 M

NaZSO4 was used as the supporting electrolyte. The fastest degradation occurred on

Ti/SnOz-Sb followed by Pt followed by the Ti/Rqu anode. Nevertheless, complete

19

 

‘ “- "w‘ 54

 

phenol r

electroch-

AC Elect
Chntanc
and the

reference

acid. A

EICCIIOI} \
used as .;
manila:
ﬁfquencg
hi'dIOQLL

.\'.:L-
an lmIiaI

lfv AC

 

phenol removal was achieved by all the anodes. Intermediate byproducts of the

electrochemical degradation of naphthalene included benzoquinone and organic acids.

AC Electrolysis
Chin and Cheng (1985) superimposed a constant alternating voltage onto a DC potential
and the resulting composite potential wave was applied between the working and

reference electrodes. The electrolyte consisted of 0.01 to 0.1 M phenol in l M sulfuric

acid. A platinum foil having 12.5 cm2 area was used as the anode for batch cell

electrolysis of phenol. A 1 liter glass jar immersed in a constant temperature both was
used as an undivided cell. The rate of conversion of phenol increased with increasing
magnitude of the alternating voltage and decreased with increasing alternating voltage
frequency. Phenol was oxidized to benzoquinone and eventually reduced to
hydroquinone.

Nakamura et al. (2005) investigated the decomposition of trichlorobenzene having
an initial concentration of 0.66 pM by AC electrolysis in aqueous solution. A 30-kHz,
15V AC potential was applied to the working electrode, counter electrode, and ground
electrode. Each electrode was a titanium plate coated with platinum and was 17.5-cm-
long by 3.5-cm-wide by 0.1-cm-thick. The reactor was an undivided cell, about 19 cm
long and 9.5 cm in diameter with a cap. The supporting electrolytes were NaOH and
NaCl (20 mM). Trichlorobenzene was degraded to intermediate products which included
2-ethyl-1-hexanol when NaOH was used as the supporting electrolyte and to

trichloromethane when NaCl was used as the supporting electrolyte.

20

 

0le 7'

T115 It: s
can be
and I:
Xaphzl .‘
50111;“) ,j
3903 l t .3
similar I.
naphth. i,

PAHs.

HATEQ

The e\ x‘

deal-In" .

c.

 

OBJECTIVES

The key objectives of this study were to: (1) demonstrate that alternating current (AC)
can be used for electrochemical degradation of naphthalene in a spiked aqueous solution;
and (2) investigate the effect of current density on the degradation rates of naphthalene.
Naphthalene, which contains two fused benzene rings, was selected as a model PAH
compound because in published studies (Alshawabkeh and Sarahney 2005); Goel et al.
2003) naphthalene has been used. Also, naphthalene has chemical and physical properties
similar to other PAHS, except that it is more soluble in water, a property that makes

naphthalene more available in aqueous solution and more exposed to reactivity than other

PAHS.

MATERIALS AND EXPERIMENTAL METHODOLOGY

The experimental setup is illustrated in Figure 1-1 and the various components are

detailed as follows:

21

 

L «urn—n-

 

 
 

.:n?:_.l. L.u.w/~:‘ .. 'K -

 

\

~ a--\ \---

Il-UI

35m .mEoEtoqxo 2: Lo ochEonom z; .2sz

98 BUGOUV douﬂom 98 “mob douﬁom

 

 

 

 

 

 

 

 

 

 

 

 

30oz?“ Socoﬁﬁnmaz I I I I I I. I I, £5ng Socoﬁﬁsmaz
DER/l I I IL m A
.wmwwmm odd H- w ewwmmmmm
/ I- --mw...H.I o o, IIIIHIIIU AHIWHM
Eudn— HOuNxX/i. IluWHIIwmmmm O NO ..II WI ”U 0. AI IHIHMIIIIMIH Ili HUMNUQ mmdﬁw Hug: a
WItIHm- oIIo o A .
l I I- _ _ - I I nIIIlI I. owobooﬁo
m8 coach. I I l I -l ,, . /: ESESE.

tom mdzmﬁam I I III I I

 

5&5 830m U< > m:
a .
a l I I II I UtIIILWM NH. l- l .IiloIl I lit ,l_.,,4I,.I.I:I
I; I I I II .wamlt II III. . . .
.s...\lll.\,v W
Houauocow III I I ll l

Emma How/d. UQ \ Homage“ “£09m

acted?»

22

 

 

Reactior
l‘ne test

COIl$l>lct

a I; '_
C‘nuIilcti

Used as L.

bt‘I‘.\‘ecn . I

“ 33mm.-
3Ilt1 DC
tamed 0
We Use:

u a ObSE-h

Ln
”r3

77‘
(“D
Q.
a
,.

 

Reaction Chamber

The test cells (reaction chambers), which were undivided cells operating in batch modes,
consisted of one liter Pyrex glass beakers with a snugly ﬁt Teflon cap. These materials
were selected to eliminate or reduce sorption of naphthalene and byproducts onto the
walls of the reaction vessel and cap. The test solution was not mixed during the
application of the current. Ultra corrosion-resistant pure high strength titanium sheets,
Grade 2 (98.9%, purity) and Grade 5 (95%, purity) were used as electrodes for this study.

Alshawabkeh and Sarahney (2005), Goel et al. (2003), Li et al. (2005), and Comninellis

(1994) used titanium as a base metal and coated it with mixed metal oxides (Rqu, IrOz,

and SnOz). The titanium plates used in this study were not coated. Uncoated titanium is

cheaper than gold, platinum, or titanium coated electrodes and hence was selected for this
project. Titanium offers more resistance to corrosion in a wide range of aggressive
conditions and can therefore be used for potential ﬁeld application. The titanium sheets
used as electrodes in this project were 12.0-cm-long by 5.0-cm-wide and the distance

between them was maintained equal to 8.0 cm. The contact area of the solution with each

electrode was 55 cm2 at the beginning of the test and decreased to approximately 53 cm2

as samples were collected for analyses. The thickness of the electrodes used for the AC
and DC application was 1.7 mm for Grade 2 electrodes. However, few DC tests were
carried out with thicker (~ 3.2 mm) Grade 5 electrodes (Table 1-1). Thicker electrodes
were used for the DC application because substantial loss of electrode mass of the anode
was observed when thinner electrodes were tested. The volume of the aqueous solutions

spiked with naphthalene in the test and control cells was one liter.

23

 

 

Electric
An AC

tenerat:
.-\C it It."
to the p

from rte;

Ule OUR-f

link \t'i

:‘V'I:
eﬁul‘rajcf

  
  

The p0“ ,\

 

PIOKIUC in

Electrical Equipment

An AC function generator/counter capable of delivering 20 VIP-[7 (AC) was used to

generate a square wave AC. An AC square wave was selected because unlike a sinusoidal
AC where during a single cycle, the voltage gradually increases from negative maximum
to the positive maximum and back again, an AC square wave function shifts suddenly
from negative to positive, stays there for half a cycle, and then jumps to full negative and
stays there for half a cycle. Therefore, in one half cycle of the AC wave, the alternating

current instantaneously ﬂows in one direction and then ﬂows in the opposite direction in

the other half cycle. The root mean square voltage (Vrms) of a square wave is equal to its

peak voltage (VP) unlike a sine wave whose Vrms is equal to 0.707Vp. Hence, an

equivalent square wave current delivers more electrical power for the same peak voltage.
The power supply consisted of a bipolar operational power supply/ampliﬁer capable of

producing 200 V AC and l A AC output or a maximum electrical power of 200 W.

Spiked Solution Preparation

Aqueous solutions, spiked with naphthalene were prepared by adding naphthalene (98%
purity) crystals to deionized (DI) water in a capped Pyrex bottle and allowing the mixture
to stand. The greater the retention time, the greater the amount of naphthalene that
dissolves into solution, until its aqueous solubility of 32.2 mg/L (Schwarzenbach et al.
2003), is reached when there is no more increase in the concentration of naphthalene in
solution. The excess suspended naphthalene crystals were then ﬁltered off and the

collected ﬁltrate was used for the tests. The average concentration of naphthalene

24

Support

  
  
 
 

In order

sndi um

 
   

experts?

 

solutions used in this study was 20 mg/L (~0.15 mM) so that adequate amounts of
naphthalene would be available in solution for quantiﬁcation of concentrations and

identiﬁcation of degradation byproducts.

Supporting Electrolyte

In order to increase the electrical conductivity of the electrolyte, NaCl and anhydrous
sodium sulfate (NaZSO4) were used as supporting electrolytes in separate sets of

experiments (Table 1-1).

Table 1-1: Experimental variables.

 

 

 

 

 

 

 

 

 

Supporting Applied Electrode Grade, Current Density Total
Electrolyte Current (%Purity); and Tests
(mA) Thickness (mm)
mA mA/cm2 mA/cm3
NaCl 330 (AC) Grade 2 (98.9%); 330 6 0.33 2
1.7 mm
NaCl 330 (DC) Grade 5 (95%); 330 6 0.33 2
3.2 mm
NaCl 0 Grade 2 (98.9%); 0 0 0 2
(Control) 1-7 mm
NazSO4 55 (AC) Grade 2 (98.9%); 55 1 0_055 2
1.7 mm
Na2$O4 55 (DC) Grade 2 (98.9%); 55 1 0,055 2
1.7 mm
Na2$O4 165 (AC) Grade 2 (98.9%); 165 3 0.165 2
1.7 mm
N32504 165 (DC) Grade 2 (98.9%); 165 3 0.165 2
1.7 mm

 

 

 

 

 

 

 

25

 

 

naphtha...

€631,331“.

 

 

Table 1-1 Cont’d

 

 

 

NaZSO4 330 (AC) Grade 2 (98.9%); 330 6 0,33 2
1.7 mm

Na2804 330 (DC) Grade 2 (98.9%); 330 6 0,33 2
1.7 mm

N82304 0 Grade 2 (98.9%); 0 0 0 2
(Control) 1-7 mm

 

 

 

 

 

Total Number
of Tests: 20

 

 

 

Notes:
1. The frequency of the AC for all the tests was 0.1 Hz.

2. The volume of the test solution was 1 liter.

Alshawabkeh and Sarahney (2005) used NaCl in electrochemical degradation of

naphthalene in aqueous solution and showed that electro generation of C12(g) enhances

the degradation of naphthalene in solution. For experiments with NaCl as supporting
electrolyte, 0.5 g of sodium chloride crystals were dissolved in a one-liter solution. The
resultant concentration of NaCl was approximately 500 mg/L (~8.5 mM). This
concentration which is relatively low was selected in order to achieve the target current
densities that could be delivered by the power supply available for the tests in this study.

Saracco et al. (2000) performed tests where electrolyte salt concentrations (e.g. NaCl,
NaZSO4, H2S04) ranged from 0.2 — 2 M and reported that higher salt concentrations

reduce the electrical resistance between the electrodes and hence will reduce the

operating costs. However, greater salt concentration would result in greater post-

26

 

 

 

iff'dilllt"

m x ‘
upft‘n’f

a: 5(in p5

enteritis

 

treatment costs and hence would be less favorable. The only exception might be

represented by discharging the treated efﬂuent into the sea (Saracco et al. 2000).

For experiments with Na2SO4 as the supporting electrolyte, a concentration of

500 mg/L (~3.S mM) was used. Goel et al. (2003) used sodium sulfate in electrochemical
degradation of naphthalene in aqueous solution using DC. This concentration was
selected because the USEPA has proposed a maximum contaminant level goal of sulfate

at 500 mg/L (~3.5 mM) and hence the proposed electrochemical degradation technique is
potentially environmentally friendly. Unlike when NaCl was used, using Na2SO4 as the

supporting electrolyte did not result in any signiﬁcant mass loss of the electrodes, and the
test solutions did not form any precipitates as a result of corrosion of the electrode
material, particularly the anode in the DC tests. Also no chlorinated by products were
formed as was the case when NaCl was used. Chlorinated by products may be more toxic

than the parent compound.

Testing Procedure

Prior to passing current through the solution in the cells, initial samples were collected
for analyzing naphthalene concentration using High Performance Liquid
Chromatography (HPLC). Liquid Chromatography/Mass Spectrometry (LC/MS)
analyses were also conducted on test cell samples as well as control cell samples. An AC
square wave, having a frequency of 0.1 Hz was applied to the AC cell test cells. This AC
frequency was selected because Chin and Cheng (1985) conducted a study on AC
electrolysis using phenol and reported that the rate of conversion of phenol decreased

with increasing alternating voltage frequency. Hence, we used the lowest AC frequency

27

pro-tint;
density

also int:

at these

Cllfl'ffll i.

tltcttt‘ll}

was can

 

alerage ,;
in tentat-
$135365. I

Gilliam Ct

 

produced by the function generator that was available for this study. The effect of AC
density on the rate of electrochemical degradation of naphthalene in aqueous solution was

also investigated by maintaining the frequency constant at 0.1 Hz. Tests were performed

at these three AC densities: 1 mA/cm2 (55 mA peak current), 3 mA/cm2 (165 mA peak
current), and 6 mA/cm2 (330 mA peak current) using NaZSO4 as the supporting

electrolyte. An additional test for AC density equal to 6 mA/cm2 (165 mA peak current)

was carried out using NaCl as the supporting electrolyte.

The AC peak current density was obtained by dividing the peak current by the
average area of each electrode in contact with the solution. The level of the test solutions
in contact with the electrode decreased as samples were collected for measurements and
analyses. Therefore the reported current density is average value of the initial and ﬁnal

current densities.

A test at DC density equal to 6 mA/cm2 (165 mA, peak current) was carried out

using NaCl as the supporting electrolyte. However, the anode lost a signiﬁcant amount of
mass (~ 6%) within 24 hours of the application of the DC and formation of precipitates in
the solution was observed. However, during an equivalent AC test, the mass loss of both

electrodes was negligible (< 1%) and the mass loss for DC as well as AC tests was

negligible when NaZSO4 was used as the supporting electrolyte. Hence, more tests at

various current densities were carried out with NaZSO4 as the supporting electrolyte as

summarized in Table 1-1.

28

 

 

Sump“
Rephen

sdqun

 

Sampling
Replicate tests were performed to evaluate the change in concentration of naphthalene in
solution with time. A summary of tests carried out is presented in Table 1-1. For each
sampling round, two samples were analyzed with HPLC to quantify concentrations. The
error bars in Figures 1-2, 1-3, and 1-4 represent the maximum and minimum values of the
normalized concentration (C/Co) for the initial test and the replicate test recorded for the
total four samples.

Samples were collected for HPLC analysis at time intervals equal to 0, 1, 2, 4, 8,
24, and 48 hrs. The reaction chamber was not mixed. Samples were collected while the
electrical current was applied and the test was not paused during sampling. The analysis of the
samples to determine the concentration of naphthalene in aqueous solution was consistent
with US EPA Methods 550, 610 and 8310. Samples were analyzed directly without
extracting the PAH from the aqueous solution. This procedure minimizes potential
interferences from reagents, solvents and glassware required for the extraction method
(Alshawabkeh and Sarahney, 2005). HPLC analyses were performed using a tunable UV
detector (detection at 254 nm), a Supelco Supelcosil LC-PAH, 250 mm x 4.6 mm x 5 pm
column, an automated gradient controller, and deionized water/acetonitrile mobile phase.
The ﬂow rate of the mobile phase was 1.5 mL/min and the column was ﬂushed with 1:99
(v/v) acetonitrile/water for 5 minutes followed by 60:40 (v/v) acetonitrile/water for 30
minutes and by 100:0 (v/v) acetonitrile/water for 15 minutes. Standard solutions were

prepared to bracket the expected concentration level of the analytes. Calibration curves

. . 2 . .
With an average obtained R value of 0.998 were used to determme the concentration of

naphthalene in the aqueous solution as the test progressed.

29

 

 

other Ct,
touting-if
IKC‘ICI cg
lube. It
lllt Spit:
decim-

Siiitd .s

 

A control cell, similar to the test cells except with no current ﬂowing through it
was set up, sampled and analyzed at the same time intervals as the test cells. The test
cells and the control cells were placed in the same water bath in order to distribute the
heat generated as a result of Joule heating in the test cell. This also allowed the control
cell to attain approximately the same temperature as the test cells. Replicate control cells

were placed in the temperature chamber where the temperature of the chamber was ﬁxed

0 . . .
at 1 C higher than the maxrmum temperature measured In the test cell.

Experimental Parameters Monitored

Apart from monitoring the concentration of naphthalene in aqueous solution over time,
other experimental parameters monitored included pH, temperature, redox potential,
conductivity, and dissolved oxygen of the solution. A portable dissolved oxygen/pH
meter equipped with a pH electrode, an oxidation-reduction probe and dissolved oxygen
probe, was used to measure pH, redox potential and dissolved oxygen, respectively, of
the spiked solutions over time. An electrical conductivity meter was used to measure the
electrical conductivity and a thermometer was used to measure the temperature of the

spiked solutions in the control and test cells

RESULTS AND DISCUSSION

Table 1-1 outlines the experimental program.

30

 

Suppn

IHUIC

f.

or

AC an;

electrol

IE
I

o

if
I

0

i4

nm
OU\U .cozctaccocoo 6222:9321. Devin-.502

3..
0

tire I

. am as

Supporting Electrolyte: NaCl
Figure 1-2 shows the average concentration of naphthalene in solution over time in the

AC and DC test cells as well as in the control cell when NaCl was used as the supporting

electrolyte for AC peak density and DC density was equal to 6 mA/cmz.

 

 

 

 

 

1-2 I I I I l
O l‘ : : : : : '
2 - -
U - L .
:- 1 ............................... Tl ................................................ -
.2 - -
“
g l- . ﬂ . ‘ : .
5 ' g 3 up : . Control -
o 0.8 I . , .............. ............ ............ u
c h . I . . I
O _ .
U . . .n._ . .
g L: 3 3 ; 3 E L
2 0.6 """""" """"""""" 3 """"""" ‘j
a b n . o . . I
4: I d
H
S I. ' t : t t .
e 04 p5,. ............ - ........... a. ............. = .............. ' ........... -2 ............ ..
z ' .i E ? E 3 2
'8 _. . - 6mAlcm (DC)
£- .c =20mgIL :
G 02 _ O .............. ............
E ' _ Supporting Electrolyte: NaCI ; g g
2 _ AC: Square Wave, f = 0.1 Hz 5 g g

'l I I ‘-\,ﬁ_'
0 6mAlcm2(AC)

 

0 10 20 30 40 50 60
Elapsed Time [hrs]

Figure 1-2: Change in concentration of naphthalene in solution for AC and DC tests with
NaCl as supporting electrolyte.

31

 

Thee
ICC}

ofIC

mgll
daeae
chronis
degrads

Ihﬂhod

 

 

The error bars represent the maximum and minimum values of the concentration ratio

(C/Co) for the initial test (t = 0), and for samples collected during the 48-hour application

ofAC or DC.

By electrolyzing naphthalene solution having an initial concentration of about 20

mg/L (~0.15 mM) with an AC peak current density of 6 mA/cmz, no naphthalene was

detected in the spiked aqueous solution in the test cell after a period of 48 hrs. The
chromatograms produced by the HPLC indicated several peaks corresponding to the

degradation products, most of which had lower molecular weight than naphthalene. The

method detection limit (MDL) for naphthalene was 59 pg/L. A plot of ln[C/Co] where C

is the concentration of naphthalene in solution at time, t and C0 is the initial (t = 0)

concentration of naphthalene in solution, versus time depicted a pseudo ﬁrst-order

degradation kinetics (R2 = 0.995) of naphthalene in solution. The observed pseudo ﬁrst-

order rate constant (kobs) was 1.21 X 10.1 hr-l, and the corresponding half-life was 5.7

hrs.
Figure 1-2 also shows the average concentration of naphthalene in solution over

time in the DC test cell. By electrolyzing naphthalene solution having an initial

concentration of about 20 mg/L (~0.15 mM) with a direct current density of 6 mA/cmz,
about 70% reduction in concentration of the naphthalene in solution was observed after a
period of 48 hrs. The chromatograms produced by the HPLC indicated several peaks

corresponding to the degradation products similar to those observed for the AC tests.

Unlike the AC test, a plot of ln[C/Co] versus time of the DC test did not indicate a good

32

 

 

dhnens

Want-

 

correlation for a pseudo ﬁrst-order degradation kinetics (R2 = 0.865) under our

experimental conditions.

Figure 1-2 shows that faster degradation rate was observed in the AC test than the
DC test. This shows that using AC for electrochemical degradation of naphthalene in our
system seemed to be more efﬁcient than using DC. Furthermore, because the surface of
electrodes was not prepared by the deposition of a layer of metal oxide to produce
dimensionally stable electrodes, about 6% loss in the mass of the anode in the DC test
was observed at the end of the test period. Initial mass of the electrode was approximately
100 g. However, during the AC test, loss of mass of the electrodes was insigniﬁcant and
could not be measured with a weighing scale having an accuracy of 0.01 g. The loss in
the mass of the anode in the DC test was visible from the pitting of the electrode.
Perhaps, the pitting of the anode resulted in fouling the electrode surface and hence the
observed reduction in degradation rate of naphthalene as compared to the DC test cell.
Electrode pitting did not occur for the AC tests.

Decrease in naphthalene concentration in the control cell over time was observed
as shown in Figure 1-2. However, the loss of naphthalene from solution in the control cell
is signiﬁcantly less compared to the decrease observed for the AC and DC test cells. Goel
et al. (2003) and Alshawabkeh and Sarahney (2005) also reported naphthalene loss in
their control cells over time and attributed it to factors such as volatilization, potential
sorption to cell components etc. The chromatograms produced by the HPLC for the
control cell did not show any peaks except the naphthalene peak. Thus, no degradation

products were formed indicating that no reactions occurred in the control cell.

33

 

I

I‘

T .
l“ 21
"

that e'i.

oxidatj

 

PlIlSU-n

CIC‘CITQL

Otidiziz‘.

 

 

Comninellis (1994), Brillas et al. (2000), and Panizza et al. (2000) have reported
that electrochemical systems generate hydroxyl radicals (OH') at the anode due to the

oxidation of water:
a + —
H,O—>0HadS+H +e (1)

Among oxygen radicals such as 02", OH., ”02., and R00 that may be generated

electrochemically in solution, the hydroxyl radical (OH) is the most reactive (Oturan and

Pinson 1995). Brillas et a1. (2000) also reported that the primary oxidizing agent in an

electrochemical degradation study was the hydroxyl radical, although secondary

oxidizing agents such as H202 and H02. radicals could have been formed. Nakamura et

al. (2005) proposed a mechanism based on selective redox reactions with various
radicals generated by AC electrolysis that allows both oxidation and reduction at the
same reaction site between adjacent electrodes. Based on the analyses of intermediate
products of the degradation process, Nakamura et al. (2005) reported that the mechanism

involved selective redox reactions with short-lived and highly reactive radicals: hydrogen

atoms (Ho) and hydroxyl radicals (OHO) that are electrochemically generated on the

electrodes as presented in Equations 2 and 3.

H20+e"—>H°+0H- (2)

0H7—>0H'+e_ (3)

Formation of hydroxylated degradation products in this study indicated that the primary

reactive species in the test cells might have been hydroxyl radicals (OH)

34

 

 

 

AC \t‘i
l'Cl CI);
naphtl‘...
fift‘lTut‘
ntédath

PJSSI‘I;

 

Chen et al. (2003) reported that under AC voltages, the electrodes cannot be
classiﬁed as ﬁxed anode or cathode. One electrode acts as the cathode in a half cycle and
as the anode in the other half cycle and therefore both cathodic electrochemical reactions
and anodic electrochemical reactions under DC voltages occur on each electrode under
AC voltages. Thus, in this study, by applying an AC, which resulted in constant polarity
reversal of the electrodes at a rate dependent upon the AC frequency, anodic oxidation of
naphthalene in solution occurred on both electrodes. A majority of published work on
electrochemical degradation has focused on the use of DC. This relies mainly on anodic
oxidation to degrade non-polar organic compounds and hence the cathode acts as a
“passive” electrode in the oxidation process. However, when an AC is used, the constant
polarity reversal ensures that both electrodes participate in the oxidation process,
depending on their polarity in a given AC cycle. During the one half of a cycle when one
of the electrodes has a positive electrical potential, it instantaneously acts as the anode
while the other electrode, at a negative electrical potential, acts as the cathode. In the
subsequent half cycle, polarities are reversed and the electrodes acquire electrical
potentials opposite to what they had in the previous cycle and this continues throughout
the period of application of the AC.

Besides direct anodic oxidation, organic pollutants can also be degraded by
indirect electrolyses, generating in-situ chemical reactants to convert them to less harmful
products (Panizza and Cericola 2003). The presence of chloride ions may result in the
formation of chlorine gas instead of oxygen gas at the anode (Alshawabkeh and Sarahney
2005). The mechanism of electro generation of chlorine from chloride ions in solution

has been presented by Panizza et al. (2000) as shown in Equations 4 to 6.

35

 

’-

i “F". ""1 '

The In;

enhan.

 

2Cl‘ —> Cl2 + 26— (+1.3583 V) (4)

Cl,+H,0—>HOCI+H++CZ‘ (5)

HOCI —> H+ + 0C!" (6)

The hypochlorous acid (HOCI) formed (Equation 5) acts as an oxidizing agent, thus

enhancing the overall degradation. Electro-generated C12(g) has been used in the

treatment of landﬁll leachate and textile efﬂuent (Chiang et al. 1995; Naumczyk er al.
1996), and in both cases, complete removal of ammonia and accompanying COD
reduction was achieved. Therefore, we believe that the degradation of naphthalene could
have been due to anodic oxidation (on the instantaneous anodes for AC tests) which was
enhanced by the electro generation of chlorine gas in solution when NaCl was used as the
supporting electrolyte.

The evolution of gases at both electrodes was observed as the electrolysis of the
solution using AC progressed. Nakamura et al. (2005) also reported evolution of gases in
the solution during the test when AC electrolysis was carried out at room temperature
with a square wave having a frequency equal to 30 kHz and 15 V amplitude using
electrodes made up of titanium coated with platinum. Chen et al. (2003) reported that
both hydrogen and oxygen were deposited on the silver electrodes of the capacitors under
AC voltages. In the half cycle, when one electrode is at a positive electrical potential
(instantaneous anode), oxygen gas is produced at that electrode and hydrogen gas is

produced at the other electrode (instantaneous cathode) according to Equation 7:

36

 

. v-J
. 't'-*7 u
l

 

 

gas at '.
supper“
Visible
toned. .
In the l

terms.

at the j

 

”100 -+ 02 + 4H t + 46’ ....... E0 = +1.229V(Anode)
4H 20 + 4e" —> 2H2 + 40H "...E" = -0.827V(Cathode) (7)

The presence of chloride ions in the solutions might also have generated chlorine

gas at the instantaneous anode according to Equation 4. In the DC test with NaCl as the

supporting electrolyte, evolution of gas was observed on the cathode (probably H2). No

visible gas bubbles were observed at the anode. Instead, the electrode material started
corroding as soon as the DC was applied to the test cell. The test solution turned cloudy

as the test progressed, accompanied by pitting and ﬂaking of the anode material, and

formation of precipitates in the solution. It is possible that C12(g) instead of 02(g) evolved

at the anode because Cl- were preferentially oxidized at the anode (Equation 4). This

might have resulted in the formation of titanium chloride complexes that precipitated in

the test solution, which made the solution cloudy as the test progressed. It is also likely
that instead of the oxidation of either water or C1- ions at the anode, the titanium anode

acted as a reactive anode and hence dissolved into the test solution according to Equation

8 or Equation 9,

. .2 —
Tl —> T1 + + 26 (+1.63 V) (8)

. .3+ _
T1 —9 Tl + 3e (+1.37 V) (9)
This also, could have formed titanium complexes that precipitated out of the test solution.
These competing reactions at the anode decreased the rate of anodic oxidation of

naphthalene. However, in the tests involving the application of AC to the test cells,

37

 

 

 

contin.
q cle -.
anode .
other L

reduet.

fiultlt;

radical.
deem;
SUESQ‘QL;
the dg:

“WE et‘

 

 

55300.11

ﬂ ..
Apt-n5,“

continuous evolution of gases was observed on both electrodes. Thus, during the half

cycle of the AC, when one electrode acquires a positive electrical potential (instantaneous

anode), water molecules are oxidized and 02(g) is evolved, and H2(g) is evolved at the

other electrode, with a negative electrical potential (instantaneous cathode) by the
reduction of water molecules. In the subsequent half cycle of the AC, the electrodes
acquire electrical potentials opposite to what they had in the previous cycle and the
oxidation and reduction of water molecules continued at the instantaneous anode and

cathode, respectively. These cycles continue throughout the application of the AC. The

evolution of 02(g) on both electrodes might have resulted in the formation organic

radicals and subsequently organic hydroperoxides which were relatively unstable. The
decomposition of such intermediates led to molecular breakdown and formation of
subsequent intermediates with lower carbon numbers (Comninellis 1994). This enhanced
the degradation rate of naphthalene in the AC test cells and hence AC appeared to be
more effective in degrading naphthalene as compared to DC when NaCl was used as the
supporting electrolyte. An advantage of AC over DC is that AC is readily available and
cheaper in terms of capital cost because no rectiﬁcation is required to convert it to DC.
High wattage DC power supplies are relatively expensive and hence would be relatively

expensive for full-scale remediation.

Supporting Electrolyte: NaZSO4

DC Tests
Figure 1-3 shows the average concentration of naphthalene in aqueous solution when the

current densities of the applied DC were 1 mA/cmz, 3 mA/cmz, and 6 mA/cmz.

38

 

 

9 P 9
h 6) on

Normalized Naphthalene Concentration, CICo
O
to

 

 

 

 

 

 

 

 

 

Elapsed time [hrs]

Figure 1-3: Change in concentration of naphthalene in solution for DC tests with Na2S04
as supporting electrolyte.

For each sampling round, two samples were analyzed with HPLC to quantify naphthalene
concentration. Figure 1-3 presents the average concentration reported from these
duplicate samples and two additional duplicate samples analyzed for a replicate test cell.

The error bars represent the maximum and minimum values of the concentration ratio

(C/Co) for the initial test (t = 0), and for samples collected during the 48-hour application

of DC. When 55 mA current (current density ~ 1 mA/cmz) was passed through the

naphthalene test solution, about 95% reduction in the concentration of naphthalene was

observed within 24 hrs. Similarly, by passing 165 mA and 330 mA currents (current

39

densities ~ 3 and 6 mA/cmz) through the naphthalene test solution, about 95% reduction

in the concentration of naphthalene in the aqueous solution was observed within 24 hrs.
However, the initial rate of degradation was greatest for the highest current density.
Thus, the higher the applied DC current density, the faster was the degradation rate of
naphthalene in solution. Alshawabkeh and Sarahney (2005) have reported similar

increase in naphthalene degradation rates with increasing applied current density.

A plot of ln[C/CO] versus time for the initial degradation rate (up to 8 hrs)

depicted a pseudo-ﬁrst-order degradation kinetics for naphthalene in solution under our

experimental conditions. The observed pseudo-ﬁrst-order degradation rate constants were

1.39><10'l hr"(R2 = 0.9274), 2.24><10'1 hr"(le2 = 0.9997), and 3.49><10'1 hr"(le2 =

0.9953) for DC densities of l mA/cmz, 3 mA/cmz, and 6 mA/cmz, respectively. The

corresponding half-lives were 5 hrs, 3 hrs, and 2 hrs, respectively. During the 48-hr test
period, about 20% reduction in the concentration of naphthalene in the control cell was
observed. This reduction in the concentration may have been due to volatilization, and
potential sorption to cell components. However, unlike the HPLC chromatogram for the

test cells, the control cell did not show any peaks corresponding to potential byproducts.

AC Tests

Figure 1-4 shows the average concentration of naphthalene in aqueous solution when the

peak current densities of the applied DC were 1 mA/cmz, 3 mA/cmz, and 6 mA/cmz.

40

 

flu I. 01 q

OUxU .COSFJCOUCOU magniﬁcathﬁz UCN:Q§OZ

 

d;

‘

“fitter
'1.
I m

D
s

 

 

 

 

I I I I I l I r Tj ! I I I I I fI I I I I T I I II I I I I
O 1 V._.'- ------- : -------------- r -------------- -------------- : ------------ -
2 b a: : : 1 I .
U h‘. : .
a ' 'l I I "
C - ; C .
'3 I \E : ; ' '
E 08 L..:.. g: ....... . ................................... F .......... ..'Control..--
0 ' ‘ a C =20mgIL *
C _ O .J
O V : : :
a 0.6 - -------------- f- -‘ -------------- §~Supportlng Electrolyte: Nast4 -
5 ' ' § AC: Square Wave,f=0.1 Hz
To ‘ - : :
5 b V k 1
E 0.4 .— ............................................................................. <-
cu . =3 2 . ..
Z 1mAIcm :
'D ' : f '
o _ . . . .
.5 ' 7 3
'5' 02 .. ................................................. N-
E f’ 7 : 3mNc'." \EBS '
é .. ' : .
2 I
h . . _ ﬂi.: I
o I I I I I IA I I I I J I I I I I l6 [PM-CIPI I I Iil I—I4I
0 10 20 30 40 50 60

Elapsed time [hrs]

Figure 1-4: Change in concentration of naphthalene in solution for AC tests with NazSO4
as supporting electrolyte.

The overall rats of degradation of naphthalene during the 48-hr test duration for these
three current densities were approximately 77%, 87%, and 95%, respectively. Thus, the
greater the magnitude of applied AC (or alternating voltage), the faster was the rate of
degradation of naphthalene. Chin and Cheng (1985) have reported a similar trend when
they evaluated degradation of phenol at various applied alternating voltages. Kappana et

al. (1952) used AC electrolysis for the oxidation of glucose to gluconate and showed that

the efficiency increased with increasing AC density. A plot of ln[C/C0] versus time

depicted a pseudo ﬁrst-order degradation kinetics (R2 = 0.995) of naphthalene in solution

41

 

 

lé‘nﬂil.

when “

lnterm

 

under our experimental conditions. The observed pseudo first-order degradation rate

constants (kobs) were 3.37 X 10.2 hr-l, 4.29 X 10'2 hr-l, 6.84 X 10.2 hr.1 for applied AC

densities of 1 mA/cmz, 3 mA/cmz, and 6 mA/cmz, respectively, and the corresponding

half-lives were 21 hrs, 16 hrs, and 10 hrs.
Thus, DC degraded naphthalene at a faster rate than AC for a given current

density as depicted by the pseudo-ﬁrst-order degradation rate constants and half-lives,

when Na2804 was used as the supporting electrolyte.

Intermediate Byproducts

Liquid Chromatography with mass spectrometry (LC/MS) analysis of degradation
products was conducted for the AC and DC test cells containing NaCl as the supporting
electrolyte. LC/MS was carried out in these ionization modes: +/- atmospheric pressure
chemical ionization mode (APCI), and +/- electro-spray ionization mode (ESI). Oasis
Hydrophilic Lipophilic Balance (HLB) extraction cartridges were used to extract
degradation byproducts from the sample solutions and acetonitrile and DI water were
used to elute them from the cartridges. Acetonitrile and DI water were used as blanks in
the analysis to eliminate observed peaks of chemicals identiﬁed by MS that originated
from these solvents and the LC column. LC/MS analysis was also conducted on the
control samples. LC/MS analyses were performed using a Waters Quattro Micro API
mass spectrometer, a Thermo Hypersil — Keystone BetaBasic - 18, 150 mm x 1.0 mm x 5
pm column, an automated gradient controller, and deionized water (DD/acetonitrile
mobile phase. The ﬂow rate of the mobile phase was maintained at 0.15 mL/min and the

column was ﬂushed with 10:90 (v/v) acetonitrile/DI water for 10 minutes followed by

42

 

60.41}

“amt

.lC

benzer

 

 

7.1!» ‘

lu‘n‘FHi

60:40 (v/v) acetonitrile/ DI water for 30 minutes, and by 100:0 (v/v) acetonitrile/ DI
water for 10 minutes.

LC/MS analyses indicated that the key degradation products formed during the
AC tests were dihydroxynaphthalene, naphthoquinone, chloronaphthol,
benzenedicarboxylic acid, and malic acid. For the DC tests, the electrochemical
degradation products were dihydroxynaphthalene, naphthoquinone, benzenedicarboxylic
acid, and malic acid. The exact isomers formed were not determined. As the tests
progressed, the color change of the test solutions to light yellow might have been due to
the formation of naphthoquinone in the solutions. Formation of quinones has been
reported due the hydroxylation process in oxidizing media (Oturan 2000). The formation
of naphthoquinone is consistent with the ﬁndings by Brillas et al. (2000), Panizza er al.
(2000), and Saracco et al. (2000) who also have reported the formation of aromatic
quinones during electrolytic oxidation of aromatic hydrocarbons.

GC/MS analysis by Goel et al.(2003) reported 1,4-naphthoquinone as the reaction
byproduct when naphthalene solutions were electrolyzed and concluded that naphthalene
degradation is due to direct anodic oxidation, oxidation by an intermediate (e.g. hydroxyl
radicals), or a combination of both. Goel et al. (2003) also reported that chlorination
experiments at pH 4 formed monochlorinated (l-chloro and 2-chloronaphthalene) and
chlorinated naphthoquinones. HPLC analysis by Alshawabkeh and Sarahney (2005)
indicated that transformation of naphthalene at the anolyte produced four compounds
with small peaks that had retention time of 12.0 — 12.5 min. Gas phase sampling was
beyond the scope of this study. Hence, any volatile byproducts which may have escaped

from the solution were not detected or analyzed.

43

Table 1-2 summarizes the measured initial and ﬁnal values of temperature, pH,
standard redox potential, dissolved oxygen concentration, and electrical conductivity of

the test solutions. These parameters are discussed as follows.

 

44

 

7:: ...S......::uu::u 2.41.2.1: 7.4.2.5212u 4:22.12; 23.34.. 3.2.52.3: .:2 ..a.::...._.2::.:..2 2:22: :5: 3:... 21.2.2: \c...:7.>..<< ”at; $th

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Q: wmw E 3“ E E. 2 _ 3 EN N5 3: 69 m 4892

w; :w 3 3. 3. am 3. Q 3: m2 we 63 m .892
2: mg 4.0 8.. m2 3:. N: 3 5: we 3 Ge _ vomsz

wmw 3w 3. 9m 03. N: S. we we of mm 63 _ 532
mm: cm: as m.» ohm com m6 m6 ﬂo— m.o_ o o Caz
com: ow: m.m ms ow mum m.: m6 oém ﬁe omm Ge 0 5oz
omﬂ ow: 9w cs 30 omm 2: v.0 N.mm Wm. omm AU<V o .UwZ
35,..— EEE .25.— EEE .55...— .aEE FE...— REE 33m .33.: 325 £525

3.3% ARREV 95v AUOV ova—c.535—
bréosuneo :ecahaoogo E553; In 9:522.th Ema—:5 23....an wire—mam
33.58:— :owzxo 328m:— xoeou FEE—«5

 

 

 

 

 

 

 

28:38 :8 35:8 use :8 “no“ mo 33:26:00 185020
98 dough—50:8 :owzxo 32036 .Eccﬁom xococ @3933 in 658382 mo mos—S, 3:5 can REE 3.5302 ”m; 033.

45

 

\v. :t.. V M11\ .u\.\~...\

 

 

 

 

 
 

NE _.o 83 U< 00:09“ 05 .«0 hugger. 2F ”802

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

O; :m E 3 m? gm 3 no oi Ni o o 532
32 5w 3 3 E w; 0.: «.0 EN 2: 9mm Ge 0 «032
5 wow v.0 M2 93 N: 2 .m 0.0 3%. 02 0mm 6.3 0 «032

 

03:00 NA 050,—.

 

46

 

 

 

Tunp
The 7.;
fheic5

“BIC?

“iﬁnn

 

 

 

pH

The h]
”363531.
he D<

€\ectrg

Prodm

Temperature
The test cells and control cells (ﬁrst set) were placed in a water bath. Heat generated in
the test cells due to Joule heating by the electrical current was distributed through the

water bath to the control cell and hence both cells attained approximately constant
temperature. However, the temperature of the test cells was up to 6 c)C higher than the
control cell temperature at the highest current density (Table 1-2). The temperature

within all test cells ranged between 19 °C and 25 c’C throughout the testing period.

Hence, a duplicate control cell was placed in the temperature chamber at 26 oC.

pH

The initial pH of all test cell solutions ranged from 6.3 to 6.8. The ﬁnal pH which was
measured 48 hours after the application of AC ranged from 4.2 to 5.2 and 11.2 to 11.5 for
the DC cells. The only exception was the AC test cell with NaCl as the supporting

electrolyte which had a ﬁnal pH of 10.7. The oxidation of water molecules at the anode

+ . . . . .
produced H ions in solution while the reduction of water molecules at the cathode
produced OH' ions in solution according to Equation 7. Also, formation of byproducts

havmg acrdlc properties IS another possrble source of H ions In solution. Therefore at

the time of measurement of the ﬁnal pH values in the AC tests, the oxidation of water
. . . . +
molecules was more prominent than its reduction and hence the generation of more H

ions decreased the pH of the solutions.

47

 

 

 

clean

of the

 

Continuous bubbling of gases was observed on both electrodes in all the AC test

cell solutions throughout the testing period. During the 6 mA/cm2 AC test with NaCl as

the supporting electrolyte, it is likely that water reduction was dominant at the time of
measurement. In all the DC tests, the ﬁnal measured pH of the solutions ranged between
11.2 and 11.5, indicating that water reduction was more prominent. No visible bubbling
of gas was observed at the anode during the DC test with NaCl as the supporting
electrolyte. Instead, the anode corroded throughout the test, which resulted in a mass loss
of the anode at the end of the test. Therefore instead of water oxidation, the anode also
acted as a reactive electrode and lost mass according to Equations 8 and 9.

The presence of chloride ions in solution could have resulted in the generation of

C12(g) (Equation 4) at the anode, which was not visible to the unaided eye. However,
continuous bubbling of gas was observed at the cathode throughout the duration of the

test. The gas evolved at the cathode was probably H2(g) from the reduction of water
molecules (Equation 7), accompanied by the generation of OH- ions in solution, resulting

in increased pH of the test solution. No measurable loss of mass was recorded for the
cathode. The pH of the control cell solutions remained approximately constant at 6.5
throughout the testing period.

Li et al. (2005) reported a drop in pH from 5.3 to below 4.0 when DC was used to

electrolyze phenol solution in a cell with 0.25 M NaZSO4 as electrolyte using platinum
(Pt) and Ti/Rqu anodes. A drop in pH from 5.3 to about 3.5 aﬁer 5 hrs, which

approached 7.0 after 18 hrs using Ti/SnOz-Sb anode was also reported by the authors. Li

48

 

eIaC

 

and;

:- ~. ‘
repel

deer

Ikani

Rt‘dli‘

Theit

gene:
”Hie.

bubh'

,wa

 

et al. (2005) concluded that the drop in pH was apparently caused by the formation of
acidic substances from the phenol degradation. Alshawabkeh and Sarahney (2005)
reported a drop in pH from about 5.8 to 1.9 in the anolyte when DC was used to
electrolyze naphthalene in aqueous solution containing 50 ml of 0.2 M NaCl using

titanium electrodes with mixed metal oxide coating.

Redox Potential and Dissolved Oxygen

The measured ﬁnal standard redox potential values for all the AC test cell solutions were
generally higher than the initial values indicating that the solutions became more
oxidizing as the tests progressed. Therefore it is likely that the oxidation of water

molecules was dominant in all AC test cell solutions and hence most of the observed gas

bubbles could have been due to the evolution of 02(g), and to some degree H2(g) and /or

C12(g) (for NaCl as the supporting electrolyte). 1n the DC test cell solutions, the

measured ﬁnal standard redox potential values were signiﬁcantly lower than the initial

values and hence reduction of water molecules was more prominent. Therefore the

observed gas bubbles on the cathode likely contained H2(g). No visible gas bubbles were

observed at the anode. The control cell, however, did not register any signiﬁcant change
in redox potential.

The measured ﬁnal dissolved oxygen concentration in all test cell solutions were

generally higher than the initial values except in the 6 mA/cm2 AC cell with NaCl as the

supporting electrolyte, which recorded a drop from 7.5 mg/L to 3.5 mg/L (Table 1-2).

49

 

in. q IEIi

 

 

Electrical Conductivity

Supporting Electrolyte: NaCl

For the AC test cells solutions containing NaCl as the supporting electrolyte, the ﬁnal
conductivity of the solutions was greater than the initial value. Slight loss in the electrode
mass was also observed for the AC tests. Due to the mass loss, the metallic electrode
could have leached ions (probably titanium ions) into the solution, which increased the
electrical conductivity of the solution. In the DC tests, signiﬁcant increase in the
electrical conductivity of the solutions was recorded at the end of the tests. This was
likely due to the continuous leaching of titanium ions into solution from the signiﬁcant

loss of mass (~ 6%) and pitting of the electrode material that was observed.

Supporting Electrolyte: Na 2504

The ﬁnal measured electrical conductivity of all AC test cell solutions containing

NaZSO4 as the supporting electrolyte was approximately the same as the initial values.

No loss of mass of electrode was recorded in these tests and hence potential electrode
corrosion followed by leaching into the solution did not occur. Also, it is likely that the
byproducts formed did not have ion dissociation properties that would have signiﬁcantly

increased the conductivity of the solution.

SUMMARY AND CONCLUSIONS

The key objectives of this study were to evaluate the use of AC for the degradation of
naphthalene in a spiked aqueous solution and to investigate the effect of current density
on the degradation rates of naphthalene. Direct current (DC) was also used to compare

the rates of degradation. A square wave AC, having a frequency of 0.1 Hz and current

50

 

 

diﬂrf

as the

 

densities equal to 1, 3, and 6 mA/cm2 were used. NaCl and anhydrous NaQSO4 were used

as the supporting electrolytes. The key conclusions of this study are as follows.

Both AC and DC are capable of degrading naphthalene in aqueous solution into
hydroxylated byproducts;

DC, at a given current density, degraded naphthalene at a rate faster than AC
having a frequency equal to 0.1 Hz, with an equivalent current density as the DC
density. However, for NaCl as the supporting electrolyte, the presence of chloride
ions in solution resulted in the corrosion of the anode, which fouled the anode
surface and hence resulted in a degradation rate for DC test cell that was less then
that for the AC test cell at the same current density.

The higher the DC or AC peak density, the higher the degradation rate of
naphthalene in solution.

The AC test cell solutions generally became acidic (oxidizing environment) at the
end of the 48-hr testing period whereas the DC test cells solutions became
alkaline (reducing environment).

An advantage of using an AC instead DC for electrochemical degradation is that
an AC is readily available from the power grid and no rectiﬁcation is required.
Also, in AC electrolysis, the corrosion of reactive electrode material in this study
was signiﬁcantly less compared to the loss of mass of anode observed for DC
electrolysis, resulting in pitting of the electrode. However, the efﬁciency of DC in
degrading naphthalene in solution was higher than an AC, especially when the
electrodes were stable did not signiﬁcantly dissolve into solution. Hence, the

treatment time for an AC having 0.1 Hz frequency would be longer than an

51

equivalent DC. The treatment time may be further impacted if a greater
frequency, such as 60 Hz, which is available from the power grid is used for the
AC. Additional lab-scale as well pilot-scale experiments are needed to evaluate

the effect of frequency and the shape of the AC signal.

52

 

 

the

CDng;

 

 

 

PAPER NO. 2: DEGRADATION OF PHENANTHRENE AND
PYRENE IN SOLUTION USING ALTERNATING CURRENT

ABSTRACT

The objective of this study was to use an alternating current (AC) to investigate the
degradation of phenanthrene and pyrene in spiked aqueous solutions. Degradation rates and
byproducts formed were investigated when a square wave AC was applied to solutions

containing pyrene and phenanthrene. Phenanthrene and pyrene solutions were subjected to

AC densities equal to 3.0 mA/cmz, 10.2 mA/cm2 and 18.5 mA/cmz. About 65%, 75% and

90% reductions in the initial concentrations of phenanthrene were observed within 48 hrs of
the application of the current densities, respectively. Similarly, about 90% reduction in the

concentrations of pyrene solutions was observed within 48 hrs of application for 3.0

mA/cmz, 10.2 mA/cm2 and 18.5 mA/cm2 AC densities. However, the rate of initial

degradation was faster for higher AC densities. Increase in the AC frequency resulted in a
decrease in the degradation rate of both phenanthrene and pyrene in solution. A square-wave

AC produced a higher degradation rate than a sine wave supplied at an rms (root mean
square) current density of 10.2 mA/cm2 and frequency of 60 Hz in the degradation of
phenanthrene. Using sodium chloride (NaCl) as the supporting electrolyte resulted in

higher degradation rates than using sodium sulfate (Na2804). Formation of 2,3-
dihydroxybenzoic acid and 2,5-dihydroxybenzoic acid from the degradation of salicylic
acid (a hydroxyl radical probe) was consistent with the ﬁndings in literature which

indicated that hydroxyl radicals were the primary oxidizing agent. The Nenrst-Planck

equation was used to model the rate of mass transfer to the electrodes, where reactions

53

 

 

\k CTL’

1")”
it ¥¥
»

other

INTI

P01}.

01 tax

 

131'):

 

the 1.

Cl". t
u'nk.‘

 

were expected to occur. The simulated concentration-time proﬁles were in reasonable
agreemement with the observed concentrations. The primary degradation products, among

others, included phenanthrenequinone and pyrenequinone.

INTRODUCTION

Polycyclic aromatic hydrocarbons (PAHS) are a class of organic compounds that consist
of two or more fused benzene rings and/or pentacyclic molecules that are arranged in
various chemical conﬁgurations (Bamforth and Singleton 2005). Of environmental
concern are primarily the PAHS ranging in size from naphthalene (two rings) to coronene
(seven rings) (Johnson et al. 2005). They are highly recalcitrant molecules that can
persist in the environment due to their hydrophobicity and low water solubility (Cemiglia
1992) as well as the presence of dense cloud of n-electrons on both sides of the ring
structures, making them resistant to nucleophilic attack (Johnson et al. 2005). They
originate from two main sources: these are natural (biogenic and geochemical) and
anthropogenic (Mueller et al. 1996). The PAHs present in the atmosphere are derived
principally from the combustion of fossil fuels in heat and power generation, refuse
burning and coke ovens (Bjorseth and Ramdahl 1985). These sources together contribute
more than 50% of the nationwide (US) emission of benzo[a]pyrene, a hydrocarbon that is
widely employed as a standard for PAH emissions (Bjorseth and Ramdahl 1985). Vehicle
emissions are another major source of PAH contamination, particularly in the urban areas
of industrialized countries, contributing as much as 35% to the total PAH emissions in
the USA (Grimmer 1983). Natural sources, such as forest ﬁre and volcanic activity, also

contribute to the overall burden, but anthropogenic sources are generally acknowledged

54

 

 

10'

[a

it‘ll

to be the most important source of PAHS in atmospheric pollution (Grimmer 1983). The
United States Environmental Protection Agency (USEPA) estimates that 3,000 to 5,000
former manufactured gas plant (MGP) sites are located across the United States and
many of these contaminated sites contain substantial quantities of PAHS (USEPA 2000).
Several approaches including biodegradation and advanced oxidation processes
have been used to degrade PAHS in solution. Electrochemistry, a link between physical
chemistry and electronic science, has a potential for the development of new advanced
methods for water puriﬁcation (Grimm et al. 1998). Existing electrochemical redox
methods depend on passing a direct current (DC) across electrodes in an aqueous
medium. Direct electrochemical oxidation at the anode (anodic oxidation) may result in
the breakdown of organic contaminants, depending on the electrode type and the
electrochemical redox potential of the contaminant (Alshawabkeh and Sarahney 2005).
Pepprah and Khire (2008) have reported degradation of naphthalene in aqueous solution

by subjecting it to an AC and comparing it to the use of DC.

OBJECTIVES

The key objective of this study was to use an alternating current (AC) to investigate the
degradation of phenanthrene and pyrene in spiked aqueous solutions. Effect of current
density, frequency, and waveform of the AC was also evaluated. AC was used for this
study because other researchers (Chin and Cheng 1985; Fedkiw and Chao 1985;
Nakamura et al. 2005; Pepprah and Khire 2008) have demonstrated that AC is capable of
degrading other organic compounds in solution. Phenanthrene was selected due to its

structural similarities to many of the higher order PAHS currently recognized as being

55

 

ham

PAH

laor

 

 

 

Sol:

ties.

 

hazardous to health (Little et al. 2002). In addition, phenanthrene is the most common
PAH found at most contaminated Manufactured Gas Plants (MGP) sites (Ko et al. 1998;
Laor et al. 1998). Pyrene was selected due to its similarities to the higher order PAHS.
Square wave AC having frequencies of 0.1 Hz, 1 Hz, and 60 Hz were used in this
study. 0.1 Hz frequency is the lowest frequency available on most commonly used
function generators including the one used in this study and AC is supplied at 60 Hz from
the power grid. The effect of current density, AC frequency and waveform, and type of
supporting electrolyte on the degradation rates of the phenanthrene and pyrene in
acetonitrile/DI water binary solution were investigated in this study. Comparison of the
application of DC and AC signals has been reported by Pepprah and Khire (2008).
Pepprah and Khire (2008) reported a faster rate of degradation of naphthalene in aqueous
solution for DC when an equivalent density AC was used. However, the signiﬁcant
electrode mass loss (~7.5%) was observed during a 48-hour test when DC was used with
NaCl as the electrolyte, whereas, the mass loss during an equivalent AC test was
negligible. In addition, AC is readily available from the power grid for potentially
implementing the technology at the ﬁeld-scale where the power demand may be

relatively high.

MATERIALS AND EXPERIMENTAL METHODOLOGY

The experimental setup is illustrated in Figure 2-1 and the setup components are

described below.

56

....._..:._....~

 

.u-I-I-n-

958 35.5.35 05 mo ocaﬁonom ”TN 833 1..

98 “web cousﬁom E
,1 1, admin Ho 3853525

 

Till-1111

 

 

 

m8 coach. 11. .1- -1- 1,. __
,, . / 1- 060583 82¢qu

11 1.6-, 1, at ..... 1 1 1.6 .
. . . O.

7 ,
. . w

. 1 .1 livlwl41
1 .111 131 119....-. 1 111.6 .

 

 

, , .11

a

Houauocow 111111.11. . -
couucsnH Emma 830m UQ \ “bummed uﬁomﬁm

57

 

 

Rear

The '

alli

elimi

 

CDT} 1

53.7?

 

Reaction Chamber

The test setup used for the electrochemical degradation of PAHS in solution consisted of
a 1 liter Pyrex glass beaker with a Teﬂon cap. These materials of the cell were selected to
eliminate or reduce sorption of the contaminants onto the walls of the reaction vessel and
the cap. Titanium plates were used as electrodes for this study. Other researchers

(Alshawabkeh and Sarahney 2005; Li et al. 2005; Goel et a1. 2003) used titanium as a

base metal and coated it with mixed metal oxides such as Rqu, IrOz, and SnOz. The

titanium plates used in this study were not coated. Uncoated titanium is cheaper than
commonly used electrodes such as gold, platinum, or titanium coated with mixed metals.
Uncoated titanium electrodes were selected for this project due to budget constraints.
Titanium offers resistance to corrosion in a wide range of aggressive conditions and
hence has the potential for ﬁeld scale application. In addition, the total biocompatibility
of titanium assures, safer use in humane bone and tissue replacement, harmlessness to
terrestrial and marine ﬂora and fauna and non-interference with microbiological
processes (Data sheet No. 19, Titanium Information Group). Electrodes made up of
graphite and nickel show poor current efﬁciency in organic degradation (Rodgers et al.
1999). In addition, graphite has relatively high impedance to AC and hence power
transfer to the sample is not as efﬁcient. Furthermore, titanium anodes are more stable
than nickel, lead, zinc or mercury (Data sheet No. 19, Titanium Information Group). The
titanium sheets used as electrodes in this project were 12.0-cm-long by 5.0-cm—wide by

1.72-mm-thick and the distance between them was maintained equal to 8.0 cm. The

. . 2 2
contact area of the solution With the electrode was 55 cm but reduced to 53 cm as

samples were collected for analyses during the experiment.

58

 

 

 

Elect
A [1 I

AC.

AC 1.
for}.

and

C017?

313:

SIM;

Pil:

 

Electrical Equipment

A 0.1 Hz to 2 MHz Sweep Function Generator was used to generate a square wave form
AC. A square waveform AC was selected because the voltage shiﬁs suddenly from
negative to positive, stays there for half a cycle, and then jumps to full negative and stays

there for the rest of the half cycle. This sequence is repeated during the entire period

when the square wave AC is applied. Therefore, the root mean square voltage (Vrms) of a
square wave is equal to its peak voltage (VP), unlike a sine wave where Vrms is equal to

0.707 Vp. Thus, during a cycle for given voltage and current amplitudes, a square wave

AC delivers greater power compared to sinusoidal or other AC signals. The power supply
consisted of a bipolar operational power supply/ampliﬁer capable of producing 200 V AC

and 1 A AC output or a maximum electrical power of 200 W.

Spiked Solution Preparation

Due to relatively low solubility of phenanthrene and pyrene in water, a cosolvent was
used to get appreciable amounts (i.e., greater than the aqueous solubility of the
compounds) of phenanthrene and pyrene into solution so that adequate amounts would be
available in solution for quantiﬁcation of concentrations and identiﬁcation of degradation
byproducts. The aqueous solubility of phenanthrene is 1.1 mg/L and for pyrene it is 0.14
mg/L (Schwarzenbach et al. 2003). Acetonitrile was selected as the cosolvent because
Ottinger et al. (1999) used acetonitrile as a cosolvent in electrochemical degradation
studies of a model PAH compound, benzo[a]pyrene. 10 mg or 5 mg of phenanthrene or
pyrene crystals were added to one liter solutions of acetonitrile (30%) and deionized (DI)

water (70%) (v/v) to achieve target concentrations of 10 mg/L or 5 mg/L, respectively.

59

 

 

 

Th

sol

L011

, I
)le

Uliilii

 

 

The proportion of acetonitrile was just enough to ensure that the PAH with the lower
solubility, pyrene, remained in the binary solution without precipitating out to form

crystals in the solution. This was followed by the addition of 0.5 g of sodium chloride
(NaCl) or sodium sulfate (NaZSO4) crystals to the solutions to increase the electrical

conductivity of the electrolyte during the electrochemical tests. Sodium chloride was
selected as a supporting electrolyte because apart from increasing the conductivity of the
solution, Alshawabkeh and Sarahney (2005) and Panizza and Cerisola (2003) have
demonstrated that electrogeneration of chlorine gas from chloride ions in solution results
in the formation of hypochlorous acid (HOCl), which enhances the oxidation of organic
compounds in solution. Also sodium sulfate was selected as a supporting electrolyte

because previous studies (Pepprah and Khire 2007) demonstrated that the loss of mass of

titanium electrodes during DC or AC electrolysis when Na2S04 was used was negligible

unlike NaCl, which resulted in signiﬁcant mass loss. Goel et al. (2003) also used sodium
sulfate in electrochemical degradation of naphthalene in aqueous solution using DC.

Saracco et al. (2000) performed tests where electrolyte salt concentrations (e.g.

NaCl, Na2SO4, H2804) ranged from 0.2 to 2 M and reported that higher salt

concentrations reduce the electrical resistance between the electrodes and hence will
reduce the operating costs. However, greater salt concentration would result in greater
post-treatment costs and hence would be less favorable. The only exception might be

represented by discharging the treated efﬂuent into the sea (Saracco et al. 2000). Hence,

we added NaCl or N32804 that yielded a relatively low concentration equal to 0.5 g/L,

which was just adequate to achieve the target highest AC density that could be delivered

60

 

(ID

(I!

by the power supply available for the tests in this study. After NaCl or Na2SO4 was

dissolved completely, the solution was ﬁltered and the collected ﬁltrate was used for all

analyses and tests.

Testing Procedure

Prior to the commencement of the passage of an AC or DC through the cells, initial
samples were collected for high performance liquid chromatography (HPLC) analyses.
Replicate tests were conducted for a majority of the tests. The electrical impedance (Z) of
each test cell was measured using an LCR Meter. The average initial impedance of the
test solutions was 100 Q and its impedance at the end of the test period was 120 Q.
However, the ﬁnal impedance varied depending on the test conditions. For a majority of
the tests, 0.1 Hz AC frequency was selected because Chin and Cheng (1985) reported that
the rate of conversion of phenol increased with increasing magnitude of the applied
alternating voltage superimposed on a DC and decreased with increasing alternating
voltage frequency. Hence, we used the lowest AC frequency produced by the function
generator that we had available for the study. Additional tests were carried out at 1 Hz
and 60 Hz AC frequency to explore the effect of AC frequency on degradation rates. A
sine wave AC delivered at 60 Hz was used to compare the effect of AC waveforms
(square and sine) on degradation rates. 60 Hz was selected because AC is supplied as a
sine wave at 60 Hz from the power grid in the United States. Also, although our power
supply device could supply a maximum voltage of 200 V, we were limited by its
maximum output current of 1 A. In addition to the current density and frequency, the
effect of the initial concentration of phenanthrene and pyrene on degradation rates was

also explored.

61

 

l‘
M

 

10.3

both

HPL

anal;
18 i
den;

mL

al‘ev

Phe

C01

 

 

The tests were performed at three AC densities of about 3 mA/cm2 (165 mA),

10.2 mA/cm2 (550 mA peak current), and 18.5 mA/cm2 (1,000 mA peak current) for

both phenanthrene and pyrene solutions. During the test, samples were collected for
HPLC analysis at elapsed time intervals of 0, 1, 2, 4, 8, 24, and 48 hours. The samples
were analyzed directly without extracting the PAH from the aqueous solution. HPLC
analyses were performed using a tunable UV detector (detection at 254 nm), a Waters C-
18 PAH, 250 mm x 4.6 mm x 5 pm column, an automated gradient controller, and
deionized water (DD/acetonitrile mobile phase. The ﬂow rate of the mobile phase was 1.2
mL/min and the column was ﬂushed with 10:90 (v/v) acetonitrile/DI water for 5 minutes
followed by 60:40 (v/v) acetonitrile/ DI water for 10 minutes, and by 100:0 (v/v)
acetonitrile/ DI water for 5 minutes. This gradient was used for the analysis of
phenanthrene and pyrene in order to obtain an appropriate separation of the degradation
product peaks on the chromatogram during the tests. Standard solutions were prepared to

bracket the expected concentration level of the analytes. Calibration curves with an

. 2 . .
average obtained R value of 0.998 were used to determme the concentration of

phenanthrene and pyrene in the solutions as the test progressed. Samples were also
collected for LC/MS scans to determine the degradation products formed.

Control cells, similar to the test cells except without any current passed through
them were set up, sampled, and analyzed at the same time intervals as the test cells. The
test cells were placed in a relatively large volume water bath maintained at the room
temperature to reduce the rate of temperature rise of the test solutions as a result of Joule

heating due to the current ﬂowing in the test cell. The control cells were placed in a heat

62

 

ﬂit“

man

 

chamber. The maximum temperature of the solution in the test cells reached 46 C)C. In

order to mimic or exceed the temperatures in the test cells, all control cells were placed in

a temperature chamber at a constant temperature equal to 50 oC.

LC/MS analysis of degradation products was conducted in these ionization
modes: +/- atmospheric pressure chemical ionization mode (APCI), and +/- electrospray
ionization mode (ESI). Oasis Hydrophilic Lipophilic Balance (HLB) extraction cartridges
were used to extract degradation byproducts from the sample solutions and acetonitrile
and DI water were used to elute them from the cartridges. Hence, both acetonitrile and DI
water were used as blanks in the analysis to eliminate observed peaks of chemicals
identiﬁed by the MS that originated from these solvents and the LC column. Also,
LC/MS analysis was conducted on the control samples. LC/MS analyses were performed
using Waters Quattro Micro API mass spectrometer, a Thermo Hypersil — Keystone
BetaBasic - 18, 150 mm x 1.0 mm x 5 pm column, an automated gradient controller, and
deionized water (DD/acetonitrile mobile phase. The ﬂow rate of the mobile phase was
maintained at 0.15 mL/min and the column was ﬂushed with 10:90 (v/v) acetonitrile/DI
water for 10 minutes followed by 60:40 (v/v) acetonitrile/ DI water for 30 minutes, and

by 100:0 (v/v) acetonitrile/ DI water for 10 minutes.

Monitored Experimental Parameters

Apart from measuring the concentration of phenanthrene and pyrene in solution over
time, other experimental parameters monitored included initial and ﬁnal values of pH,
temperature, and electrical conductivity. A dissolved oxygen/pH meter, equipped with a

Platinum Series pH Electrode, was used to measure pH of the spiked solution. A portable

63

conductivity meter was used to measure the electrical conductivity and a microprocessor
thermometer was used to measure the temperature of the test cell and control cell

solutions.

RESULTS AND DISCUSSION

Table 2-1 outlines the list of parameters explored in the experimental program.

Table 2-1: Summary of Experimental Parameters

 

 

 

 

 

 

 

 

 

Compound Initial Supporting AC AC RMS AC No.
Concentr- Electrolyte Frequency Waveform Density of
ation “121 [mA/cm’] tests
Co
[mg/Ll

Phenanthrene l 0 NaCl 0. 1 Square 3 .0 2
Phenanthrene l 0 NaCl 0. 1 Square 10.2 2
Phenanthrene l 0 NaCl 0. 1 Square 18.5 1
Phenanthrene 10 Na2$O4 0. 1 Square 10.2 1
Phenanthrene 1 0 Na2S04 1 Square 10.2 2
Phenanthrene 1 0 NaZSO4 60 Square 10.2 1
Phenanthrene 10 Na2S04 6O Sine 10.2 1
Phenanthrene 1 0 NaZSO4 0 DC 3 .0 l

 

 

 

 

 

 

 

64

 

Table 2-1 Cont’d

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

tests

 

Phenanthrene 10 Na2S04 0 DC 6.0 1
Phenanthrene 1 0 Na2S04 0 DC 10.2 1
Phenanthrene 1 0 NaCl Control - - -
Pyrene 5 NaCl 0. 1 Square 3 .0 2
Pyrene 5 NaCl 0. 1 Square 10.2 2
Pyrene 5 NaCl 0. 1 Square 18.5 2
Pyrene 5 Na2S04 0. 1 Square 10.2 1
Pyrene 5 Na2SO4 1 Square 10.2 1
Pyrene 5 NaZSO4 60 Square 10.2 2
Pyrene 5 NazSO4 6O Sine 10.2 1
Pyrene 5 NaZSO4 0 DC 3.0 1
Pyrene 5 Na2SO4 0 DC 6.0 1
Pyrene 5 Na2S04 0 DC 10.2 1
Pyrene 5 NaCl Control - - 1
Salicylic Acid 100 NaCl 0.1 Square 10.2 1
Total
no.0f 29

 

Electrochemical Degradation of Phenanthrene in Solution

Eﬂect of Current Density

Figure 2-2 presents the average concentration of phenanthrene (C0 = 10 mg/L) collected
from the test cells subjected to AC densities equal to 3.0 mA/cmz, 10.2 mA/cm2 and 18.5

2
mA/cm .

 

[111 'jiiiiliiiiljriiriiI'l'...
0 v I a a a

 

 

 

 

 

O f ' . : : : : : .1
.1. _ . : k . i - m‘
8 1 - I '2 """""""" ; """ 1...... """ ; """ ; ﬁr: Control
.. ' ‘1 3 3 E ' ' '
5 ' El _ _ c =10 mglL -
:1: . . u . ; ; ° .
g 0 8 __g,,h __________ g ____________ g _______________ g, _ _ Supporting Electrolyte: INIaCI____d
g ' _ ‘ : i ' AC: Square Wave,f='0.1 Hz 4
l . .

2 - an ' ' ,
O _ .1
0

0 ............................................ q
1: 1
2
c d
H

c i
2 2n
.2 1- . i I i 2 i 1 m

a. _ E 3 5 k=3.98x10' hr' 3 1
'U f 3': i 3 : :
ﬁ 02 '__§ ............... g ........... 2, ________ I ...... s .............. \Emzwwm2
E '5 5 ‘ :: “1224111621161 ..Eg 2
Z . , _ , ' , 18.5 mAlcm
o l I I I I l I #1 J I I I I I I I I I I I I I I J L I I I I
0 10 20 30 40 50 60

Elapsed time [hrs]

Figure 2-2: Change in concentration of phenanthrene with time

By passing AC having a peak current of 1 A, equivalent to an average AC density of 18.5

2 . . . .
mA/cm through the phenanthrene solution, about 90% reduction in the concentration of

66

phenanthrene in the solution was observed within 48 hrs. The AC density was obtained
by dividing the peak current by the average cross sectional area of the test solution in
contact with the electrodes. The level of the test solution in contact with the electrode

decreased as samples were collected for measurements and analysis. Therefore an initial

current density of 18.2 mA/cm2 was applied to the solution and after the 48 hr-testing
period; the current density reached about 18.9 mA/cm2 resulting in an average value of
about 18.5 mA/cmz. By passing AC having a peak current of 0.165 A and 0.550 A,

equivalent to an average AC density of 3.0 mA/cm2 and 10.2 mA/cm2 through the

phenanthrene solution, about 65% and 75%, respectively, reduction in the concentration
of phenanthrene in the solution was observed within 48 hrs. Thus, increase in current
density resulted in an increase in the degradation rate of phenanthrene in solution. This is
consistent with the trend observed by Alshawabkeh and Sarahney (2005), who reported
an increase in degradation rate of naphthalene as current densities were increased.

A majority of published work on electrochemical degradation has focused on the
use of DC. This relies mainly on anodic oxidation to degrade non-polar organic
compounds and hence the cathode acts as a “passive” electrode in the oxidation process.
Pepprah and Khire (2008) have reported that when an AC is used, the constant polarity
reversal ensures that both electrodes participate in the oxidation process, depending on
their polarity in a given AC cycle. During the one half of a cycle when one of the
electrodes has a positive electrical potential, it instantaneously acts as the anode while the
other electrode, at a negative electrical potential, acts as the cathode. In the subsequent

half cycle, polarities are reversed and the electrodes acquire electrical potentials opposite

67

to what they had in the previous cycle and this continues throughout the period of

application of the AC.

Oxidizing Agents

Organic compounds in aqueous solutions can be oxidized on an anode (anodic oxidation)
by direct electron transfer and indirect oxygen atom transfer (Polcaro et al. 1999; Iniesta
et al. 2001; Rodgers et al. 1999; Chiang et al. 1995; Kirk et al. 1985). In the direct
electron transfer process, organics are adsorbed on the anode surface and give up

electrons to the anode (Li et al. 2005) as presented in Equation 1

(Organic )

ads

— ne‘ —> product (1)

During the indirect oxygen atom transfer, oxygen radicals, especially the hydroxyl
radicals generated from water electrolysis, help in the oxidation of organic substances
(Iniesta et al. 2001; Simond et al. 1997; Comninellis 1994; Lee et al. 1981). The
formation of hydroxyl radicals on the anode surface can be expressed by Equation 2,

where M stands for the anode (Simond et al. 1997; Comninellis 1994)

M+H20—>M[OH']+H++e‘ (2)

The hydroxyl radicals readily react with the organic molecules adsorbed on or in the

vicinity of the anode to cause the reaction presented below in Equation 3 (Li et al. 2005)

M [0H '] + Organic —> M + product (3)

Meanwhile, the hydroxyl radicals react with each other to form molecular oxygen to
complete the electrolysis of water molecules (Simond et al. 1997; Comninellis 1994) as

shown in Equation 4

68

2M[0H.]—)2M+02+2H++26_ (4)

Nakamura et a1. (2005) proposed a mechanism based on selective redox reactions
with various radicals generated by AC electrolysis that allowed both oxidation and
reduction at the same reaction site between adjacent electrodes. Based on the analysis of
intermediate products of the degradation process, it was revealed that the mechanism

involved selective redox reactions with short-lived and highly reactive radicals: hydrogen
atoms (H.) and hydroxyl radicals (OH.) that are electrochemically generated on the
electrodes. Based on the hydroxylated degradation products formed, it is likely that OH

radicals produced in our test solutions were mainly responsible for the degradation of

phenanthrene and pyrene.

H ydroxyl Radical Test

OH. attack the benzene rings of aromatic compounds to produce hydroxylated

compounds. Various OH radical probes and techniques have been used in previous

studies to detect the generation of OH. in the reaction media. These include salicylic acid

(Li et al. 2003; Marselli et al. 2003; Oturan 2000; Jen et al. 1998; Scheck and Frimmel
1995), 4-hydroxybenzoic acid (Anderson et al. 1987), benzoic acid (Oturan and Pinson
1995), para-chlorobenzoic acid (pCBA) (Elovitz and von Guten 1999; Haag and Yao
1992), 2,2-diphenyl-l-picrylhydrazyl (DPPH) (Sehgal et al. 1982), 5,5-dimethyl-l-
pyrroline-N-oxide (DMPO) and electron spin resonance (ESR) (Marselli et al. 2003), p-
nitrosodimethylaniline (RNO) and spectrophotometer measurements (Tanaka et al.

2004), and RNO / ESR (Comninellis 1994). Since ESR methods require expensive and

69

 

0.: CV. BUN: rﬂtgcz

 

 

 

sophisticated equipments; they were not used in this study due to budget constraints.

Trapping of OH. using salicylic acid, and HPLC measurements were selected as hydroxyl

radical probe technique for this study. Since salicylic acid had a greater solubility in the
binary solution, 100 mg/L was used for the electrochemical cell so that enough
degradation products would be produced for identiﬁcation purposes. Figure 2-3 shows a

decrease in concentration of salicylic with time.

 

 

I l I l I
I ' l I l
- ' ‘ ' : : 1

Normalized Salicylic Acid Concentration, CICo

 

 

 

- C =100 mgIL 5 3 1
° : :
0'2 "Supporting Electrolyte: NaCl """""""""""""" ‘
e AC: Square Wave,f=0.‘l Hz,]=10.2 mA/cm2 -
011111
0 10 20 30 40 50 60
Elapsed Time [hrs]

Figure 2-3: Change in concentration of salicylic acid

70

About 70% reduction in concentration of salicylic acid was achieved in 48 hrs
accompanied by the formation of degradation products such as 2,3-dihydroxybenzoic
acid and 2,5-dihydroxybenzoic acid. The formation of these degradation products is
consistent with numerous studies (e.g., Marselli et al. 2003) that hydroxyl radicals were

involved in the electrochemical degradation process.

Degradation Kinetics
The initial degradation rate of phenanthrene in solution could be described as pseudo

ﬁrst-order decay as shown in Equation 5:

d C
—EJ]'= “klcl (5)

where C is the concentration of phenanthrene and k is the pseudo ﬁrst-order degradation

rate constant. A plot of ln[C/Co] versus time was used to estimate k. The k obtained from

Figure 2, for the initial 8 hrs of application of AC density of 18.5 mA/cm2 was 1.22 x 10'

l hf] (R2 = 0.99) and the corresponding half-life was 5.7 hrs. Similarly, the degradation

rate at an AC density of 10.2 mA/cm2 exhibited initial rapid degradation until after 8 hrs

when the degradation rate decreased substantially. About 75% reduction in the
concentration .of phenanthrene in solution was observed during the 48-hour AC

application. The initial degradation rate for this AC density was characterized by an

observed pseudo ﬁrst-order degradation rate constant (k) of 8.98 X 10'2 hr.l(R2 = 0.92)

and the corresponding half-life was 7.7 hrs. The degradation rate at an AC density of 3

mA/cm2 was 1.84 X 10.2 hr.1 (R2 = 0.99) and the corresponding half-life was 34.7 hrs.

71

The phenanthrene degradation rate was faster when a higher AC density (or alternating
voltage) was used. This ﬁnding is consistent with the previous ﬁndings (Chin and Cheng
1985) whereby, the rate of conversion of phenol increased with increasing magnitude of
alternating voltage superimposed on a DC. During the testing period of 48 hours, as
shown in Figure 2-2, no signiﬁcant reduction in phenanthrene concentration was

observed in the control cell.

Effect of Type of Supporting Electrode

Besides direct anodic oxidation, organic pollutants can also be degraded by indirect
electrolyses, generating in situ chemical reactants to convert them to less harmful
products (Panizza and Cerisola 2003). The presence of chloride ions may result in the
formation of chlorine gas instead of oxygen gas at the anode (Alshawabkeh and Sarahney
2005). The mechanism of electrogeneration of chlorine from chloride ions in solution has

been presented by Panizza et al. (2000) as shown in Equations 6 to 8.

2Cl— —> C12 + 29— (+1.3583 V) (6)
C12+H20——>H0C1+H++Cl‘ (7)

HOCI —') H+ +0Cl— (8)

The hypochlorous acid (HOCI) that is formed (Equation 7) acts as an oxidizing agent,
thus enhancing the overall degradation. It has been used in the treatment of landﬁll
leachate and textile efﬂuent (Chiang et al. 1995; Naumczyk et al. 1996), and in both

cases, complete removal of ammonia and accompanying COD reduction was achieved.

On the other hand, it was expected that using Na2804 as supporting electrolyte would not

72

 

1} \

generate additional oxidizing agents in solution that would enhance overall degradation
rates. Figure 2-4 shows a plot the concentration time proﬁle of two tests conducted under

similar experimental conditions of 10 mg/L phenanthrene initial concentration and 10.2

2 . . .
mA/cm current densny, except that NaCl was used as the supporting electrolyte in one

test and Na2804 in the other test.

 

 

 

 

 

7 I T I I II 1 I I I I T I I I ‘l I I r I l I I I I 1' I I I I

8 5.2! .— .r : a 2 z = l
a 1 9 F317 """"""" : """""""" : """ : """ a: 'Control"""1

a 'V- i I i l . I
c . .
*5 " C ' : : : : -
.3 0.3 -.[H ........ ............ .............. ..... . ........ ............... ............ 4
0 l' ' ' - 1 "
o l .
c ' . E E '
8 - g g . Supporting Electrolyte: .1
g 0.6 r --------- .4 ------ - -------- g --------------- g ------------ '- -------------- g- ------------ d
2 F : : : Na 80 "
s - .
g - : : . : : : ~
5 0.4 - ----------- 4— ------------- ----------- ------------ -
.c - - . - - 1 a
Q. _ ; d
'8 i . . - 3 -
N ' ' ‘ E 0 ENaCI
a; 0.2 -‘c =10mglL -------------- : ------ J
E 1' o - i I
‘z’ AC: Square Wave,f=0.1 Hz,j=10.2 mivcm2 _ 2

o I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
0 10 20 30 40 50 60

Elapsed Time [hrs]

Figure 2-4: Concentration of phenanthrene using NaCl or NaZSO4 as supporting
electrolyte

The faster degradation rate exhibited by the test cell containing NaCl conﬁrms that

degradation rates were most likely enhanced by the formation of HOCl from the anodic

oxidation of C1' to produce chlorine gas in solution. Therefore we believe that the

73

 

[11

.v.
n\..

 

 

degradation of phenanthrene in our system could have been due to direct anodic

oxidation which was enhanced by the electrogeneration of chlorine gas in solution.

Effect of AC Frequency

Square wave AC having a peak AC density of 10.2 mA/cm2 was used to investigate the

effect of AC frequency (f) on degradation rates of phenanthrene in the test cells. Figure 2-

5 shows that applying a square wave peak AC density of 10.2 mA/cm2 at frequencies 0.1,
. -2 -l 2 -2 -l 2
l, and 60 Hz resulted in k equal to 1.3 x 10 hr (R = 0.95), 0.5 X 10 hr (R = 0.99),
-2 -l 2 . . 2 .
0.6 x 10 hr (R = 0.99), respectively. A DC (f: 0) densny of 10.2 mA/cm resulted in

k equal to 12.4 x 10‘2 in" (R2 = 0.96).

74

 

IIIIIIIIj'IIIU'TIII'II—ITIIIII

 

 

Normalized Phenanthrene Concentration, CICo

 

 

 

t c =10 mglL ' I 5 B To Hz (no)
0 : :
0.2 *- --------- ------------ «-
T Supporting Electrolyte: Na2 SO . . .
1- .
- DC, AC (Square Wave), 1 = 10. 2 mAlcm2 3 E .
o n I I L I I I n I n I n n l n n 1 l I I 1 n n I 4 I I 1
0 10 20 30 40 50 60

Elapsed Time [hrs]

Figure 2-5: Effect of AC frequency on degradation rates of phenanthrene

These results indicate that the rate of degradation of phenanthrene in aqueous solution
generally decreased with increasing AC frequency. This ﬁnding is consistent with the
trend observed by Chin and Cheng (1985) that the rate of conversion of phenol decreased
with increasing alternating voltage frequency superimposed on a DC. The voltage
supplied by our electrical test equipment maxed out aﬁer 8 hrs of application of the DC,
thus resulting in decreasing amount of current supplied. This slowed down the
degradation rate after 8 hrs. Figure 2-6 shows the temperature-time proﬁle of a typical

test solution containing 30% Acetonitrile/70% DI water as solvent with sodium sulfate as

75

the supporting electrolyte and another typical test solution containing 100% DI water as

the solvent with sodium sulfate as the supporting electrolyte.

45
qt: 40
C
.9
§
8 35
‘65
d)
I-
h
0
e 30
a
E
0
D.
g 25

20

 

I r I I

 

.........................

30% AcetonitrleI70% DI Water-57-

  
  

 

....................................

100°/. Diwater,_,,,,,_,,,,,§ ...............

.................................................................................

Supporting Electrolyte: NaZSO4

DC: j = 6 mAlcm2

J

I_I #I I
20

I4 I I I I I I I I I I4 I I

10 15

I I I

5
Elapsed Time [hrs]

I I I J

-

I I I L

on:
I
q
q

 

Figure 2-6: Temperature-time proﬁle of test solutions containing only DI water and
30/70 Acetonitrile/DI water

The presence of acetonitrile in the test solution increases the resistance of the test

solution. Greater resistance results in greater voltage applied by the ampliﬁer to maintain

the given current which results in greater temperature of the solution. Greater temperature

further increases the resistance of the solution. In constant current mode, the ampliﬁer

automatically adjusted the voltage to compensate for the increase in resistance of the tests

solution in order to maintain the set current through the test solutions. Therefore after the

voltage maxed out at 200 V, the supplied current decreased over time leading the applied

76

current decreased for the DC tests (Figure 2-5). However, still DC resulted in greater rate

of degradation than any AC at an equivalent rms current.

Continuous bubbling of gases (probably 02 and H2) in the test solution was

observed as the tests progressed, however, the rate of bubbling of the gas decreased as the
frequency of the applied AC was increased until no gas bubbles were visible at the
electrodes at an AC frequency of 60 Hz. Stock and Bunce (2002) reported that
electrocatalytic hydrogenation of hydrogenolysis of atrazine (Equation 9) was in

competition with hydrogen evolution, (Equation 10):

(R —X)adsM+2HadSM a (R—H)adsM+HX (9)

HadSM + HadSM —> H2 + 2M (10)

According to Equations 4 & 10, if the formation of OH radicals and hydrogen radicals are
precursors for the generation of oxygen gas and hydrogen gas at the anode and cathode,
respectively, then it is likely that the rate of gas bubbling at the electrode gives an
indication of the rate of formation of these radicals at the electrodes. The higher the rate
of the observed bubbling of the gas(es), the greater the rate of hydroxyl or hydrogen
radical formation at the electrodes. Hence, based on the visual observation of gas
bubbles, it is likely that increase in the AC frequency resulted in decrease in the rate of
formation of hydroxyl radicals at the electrodes. This impacted the rate of oxidization of
phenanthrene at the instantaneous anodes.

Therefore it is probable that either reduced rate of generation of hydroxyl radicals
with increased AC frequency, or back-and-fonh movement of the reacting molecules that

slowed down the rate of movement of reacting molecules to the reaction site (the

77

 

 

:1:

inc

electrode), or a combination of both factors resulted in decreased degradation rate with
increased AC frequency. It is also possible that competing anodic and cathodic reactions
at the same reactions sites (both electrodes) as a result of the polarity reversal of the

applied AC slowed down the oxidation of the phenanthrene molecules.

Effect of A C Waveform

Since AC is supplied as a sign wave at 60 Hz in the United States, the effect of the AC
waveform on degradation rates was explored using a 60 Hz square wave and a 60 Hz sine
wave. Figure 2-7 shows that a square—wave AC resulted in a greater degradation rate than

the sine wave AC supplied at an equivalent rms (root mean square) current density equal

to 10.2 mA/cm2 and frequency of 60 Hz.

78

 

II 'rIII'IITI'rIIfrIIFI'TIIr

. i ' L1. ' ' .

 

 

 

 

Normalized Phenanthrene Concentration, CICo

 

 

 

1; .
L ECO 8 10 mglL 3 .
_ ESupportlng Electrolyte: Na2804 :
* EAC: f = 60 Hz,l = 10.2 mAIcm2 (rms) .
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I u
0.6
0 10 20 30 40 50 so

Elapsed Time [hrs]

Figure 2-7: Effect of AC waveform shape on degradation rates of phenanthrene

During a single cycle in a sinusoidal AC wave, the voltage or current gradually increases
from negative maximum to the positive maximum and back again. For an AC square
wave function, the voltage or current shiﬁs suddenly from negative to positive, stays
there for half a cycle, and then jumps to full negative and stays there for half a cycle.
Therefore, in a cycle of the AC wave, the alternating current instantaneously ﬂows in one
direction and then ﬂows in the opposite direction in the other half cycle. Therefore the
full peak current for a square wave is felt at any given period in the cycle, unlike the sine

wave that sweeps from zero through lower current values until the peak current value is

79

attained and then sweeps back to zero. Therefore, the square wave was more efﬁcient in

degrading phenanthrene than an equivalent sine wave.

Electrode Fouling

The degradation of phenanthrene in solution was characterized by an initial relatively
rapid degradation within 0 to 8 hrs of the application of AC and a slower rate from 8 to
48 hrs. The decrease in degradation rate may be due to, among other factors, such as
reduced mass of reacting molecules available for reaction, and electrode fouling over
time as the new titanium sheet lost its initial metallic sheen (light gray color) to a dark
gray color as the test progressed. The electrode fouling was conﬁrmed by utilizing
electrodes, which had been used in a previous test to electrolyze replicate phenanthrene
test solution. It resulted in decreased degradation rate as compared to the new electrodes.
By polishing the electrodes that had been used in previous test with abrasives such that
their metallic sheen was restored, degradation rates using the polished electrodes were

approximately the same as using new electrodes for the tests.

Electrochemical Degradation of Pyrene in Solution
Effect of Current Density

Figure 2-8 presents the average concentration of pyrene in the test cells at AC densities

equal to 3.0 mA/cmz, 10.2 mA/cmz, 18.5 mA/cmz, and for the control cell.

80

 

 

  

 

 

 

1 ."‘---m_'. B H ‘7 . ”3 lcént'roi f
o 3 I E c =5mgIL ‘1

U l : : ; 0
a | : : 2 Supporting Electrolyte: NaCl ‘
- i‘ 3 3 3 .
5 0.8 '. 2 _2 _1 ------ AC: Square Wave,f =0.1 Hz «a
'13 . 3mm". (k=4.52x10 hr ) 5 3 E -
b .3 - ; 3 3 ‘ .
§ _ - 2 9 2 a .
g 0.6 —-{ --------- § ------------ -------------- ; --------------- -------------- g ------------- -
U ' E f : .,
g b i ' i 10.2 mAlcm2 I '
(D l- : . : -
u . . .2 .1 .
5 0.4 km .......... (k=12.44x10 hr )...--; ............. _
‘D ,_ : : : : ‘
0 _ .
.!
7° ' : a s : s '
g 0.2 - -------------- --------- . -- ; -------------- ----------- ------------ -
Z - . : E E .

. 18.5mAIcm2(k=17.28x10'2hr'1)i 0 g ,

o I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
0 10 20 30 4O 50 60
Elapsed Time [hrs]

Figure 2-8: Concentration of pyrene versus time for various current densities

For each sampling round, two samples were analyzed using HPLC to quantify
concentrations. Figure 2-8 presents the average concentration reported from these two
samples and two additional samples from a replicate test cell. The error bars represent the
maximum and minimum values recorded for the total four samples. For all three current
densities, 90 % reduction in concentration of pyrene in the solution (initial target
concentration = 5 mg/L) was observed within 48 hrs. Similar to the phenanthrene
solution, the initial degradation rate of phenanthrene in solution could also be described

as pseudo ﬁrst-order decay similar to that for phenanthrene presented in Equation 1.

81

Although initial degradation rates were faster for higher current densities, the lowest
current density applied, ultimately resulted in the same percentage degradation after 48
hrs of testing. Hence in terms of savings in electrical energy, it would be more

economical in running tests at low current densities.

By comparing the time-concentration proﬁles of the 18.5 mA/cm2 and 10.2

2 . . . . . .
mA/cm current densmes, It can be observed that although these current densmes differ

by almost a factor of two, their corresponding degradation rates constants differ

signiﬁcantly by less than a factor of two. However, by comparing the time-concentration

proﬁles of the 10.2 mA/cm2 and 3 mA/cm2 current densities, these current densities

differ by about a factor of three and their corresponding degradation rate constants differ
approximately by a factor of three. Therefore it can be inferred that beyond a certain
current density, the rate of movements of the reactants to the reaction sites (the
electrodes) becomes the rate limiting step because excess oxidizing agents are produced
at the electrodes to potentially oxidize the reactants. This is consistent with previous
studies by Stock and Bunce (2002) who applied currents 5, 10, 20, 50, 100 and 200 mA
on the dechlorination efﬁciency of atrazine and reported that applied currents 2 20 mA
led to Z 93% loss of initial atrazine within 30 minutes. The current efﬁciency decreased
as applied current increased and was greatest (16%) at 10 mA. Yusem and Pitauro (1992)
also reported a decrease in current efﬁciency with increased applied current in

electrocatalytic hydrogenation of soyabean. Stock and Bunce attributed this trend to the

fact that a greater proportion of “ads was lost to unproductive reaction of hydrogen gas

evolution instead of participating in the dechlorination reaction of atrazine. Apart from

82

this explanation, we believe that a greater proportion of the applied current in our system
was dissipated as heat as a result of Joule heating, which was evident by increased rate of
temperature rise of our test solutions with increased current density. As shown in Figure
2-8, the control cell solution did not indicate any signiﬁcant reduction in concentration

over the entire testing period.

Effect of Type of Supporting Electrode

Using NaCl as the supporting electrolyte resulted in a higher degradation rate of pyrene

than using NaZSO4 as the supporting electrolyte as shown in Figure 2-9. This is

consistent with the results obtained for the tests involving phenanthrene.

 

 

 

 

 

I I I I I I I I I I I I I I I I I I I T 1 r 1 I T T I r I I

' n; - Us; : 3 : : {:1 Control ‘
O 1 ...... ......... ............... ; .............. j ............... ; ............ a-
U - .g 5 C = 5 mgIL -
a . : ' : ° .
c“ .- 1 I 3 Supporting Electrolyte: Na SO , NaCl .
O : : 2 4
-.-, 0.8 --. ---------- o -. -------------- 2----
g ’ AC: Square Wave, f = 0.1 Hz, j = 10.2 mAlcm ‘
g ' ' g : : : : t
g * : s 3 . 2 s z '
0 (L6 ...3 ............. .............. ..... .. .. i ............. --------------- ............ .4
U . : : : : ' .
o O
5 - Nazso .
" ' : 3 f E 3 E "
$ 0.4 ......... ...... ‘ ..: .............. ............. : .............. : ............... ............ .-
1, _ - , ; Voltage maxed out on amplifier; q
a _ ‘ (current intensity dropped) ‘
= ___;_) I
w " : 3 : f 3 3 ‘
E 0.2 --.- ------------- r .. ------------- r ------------ CI
0 . : : : : . : ..
z _ ‘ NaCl

i- 3 E E E i E .

o l I I L I I I I I I I I I I I I I I I I I I I I j l I I I L
0 10 20 3O 40 50 60
Elapsed Time [hrs]

Figure 2-9: Change in concentration of pyrene with time using NaCl or NaZSO4 as
supporting electrolyte

83

Effect of A C Frequency
Similar to the phenanthrene test cells, increasing AC frequency resulted in decreased rate

of degradation of pyrene in solution as shown in Figure 2-10.

 

 

 

 

 

 

I I I I I ! I I I I I I if I I I I I I r I. I I I I l I I I I
' m - a? : a : : {1 Control -
1 77" -------- : -------------- : ------- - ------- : -------------- : --------------- : ------------ -
° . -=:-' V i : : r : J
2 : L“ \r
C i I - ' l l I
.g 0.8 bit ------ =3 ----------- ------- -------------- --------------- ----------- «1
N I- - . - V . . 1
3g . E ‘ 3 E ' i 3 ,
L . . .
8 b : : : u : ' 60 Hz -
C ...5 ............. .5 ...... . ....... f ..... i ............. 3 ........... 5 ........ ...-_
8 0'6 . 3 E E E ' V 1 Hz
at _ " 0.1 Hz .
C . . . . .
9 - a E = s s z s -
E 0.4 -3 .............. g ...... . ..... : ........ - ...... ...... 3 ............... Z ............ .-
.8 .. I i I i .1
a”. - C = 5 mgIL I
a - o , . '
E 0'2 "Supporting Electrolyte: Nazso4 ; """" """""""" i """ ' """ '
2 _ E 0 Hz (DC)
. DC, AC: Square Wave, j = 10.2 mA/cm2 . .
o I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
0 10 20 30 40 50 60
Elapsed Time [hrs]

Figure 2-10: Effect of AC frequency on degradation rates of pyrene

Eﬂect of AC Waveform
AC square wave produced a faster degradation rate than an equivalent AC sine wave as

shown in Figure 2-11. This is also consistent with trends observed for phenanthrene.

84

 

     

 

 

 

 

j: I I I r ! I I I I ! I I I I l I I I I I I I j r l I I I j
‘3‘. 5 j j [J Control "
o 1 ------------------------------------- . -------------- : --------------- : ------------ «-
2 d
U ' ‘ . . . 1
g" 5 ; 2 s 3 s ‘
'g 0.8 5 ---------- g --------------- 3 -------------- g --------------- 3 ------------ -
b ~ . . . . d
c u
a:
g : : z : ' f .
o 0-5 i ‘i 3' I """"""" " “Sine """" "'
U : : : : : .
g _ Square ‘
d) . . ; . . .
g ' : t : i : : ‘
a. 0.4 --3 --------------- g ------------- ,3 -------------- g ------------- 3 --------------- g ------------ -
3 . I I Z I .
L3 ' C = 5 mglL { 3 '
II I- O i 2 u
E 0.2 -- Supporting Electrolyte: Na SO ---- E -------------- E- ' - - - -' ‘‘‘‘‘ -
g . 2 4 s s .
: AC: Square Wave, f= 0.1 Hz, j = 10.2 mAIcm2
o I I n n I I I I n I I I I I n I I I I 1 I I I I I I 1 I 1 n_
0 1O 20 30 40 50 60

Elapsed Time [hrs]

Figure 2-1 1: Effect of AC waveform shape on degradation rates of pyrene

Modeled versus Measured Concentrations of Phenanthrene and Pyrene

It is generally believed that electrochemical reactions occur mainly on the surface or in
the vicinity of the electrodes. Therefore, the rate of reaction of the organic reactants in the
test cell depends on the rate of generation of OH radicals at the electrodes as the
oxidizing agents, as well as the rate at which the reacting molecules move to the
electrodes for subsequent oxidation by the OH radicals. Mass transfer in solution occurs
by diffusion, migration and convection. Diffusion and migration result from a gradient in
electrochemical potential and convection results from imbalance of forces on the solution

when the solution is mixed or stirred (Bard and Faulkner 2001). Assuming a linear mass

85

transfer, and linear electric ﬁeld, the mass ﬂux to the electrode can be expressed by the

Nernst-Planck Equation (Equation 1 l):

6C1x)— Z: DC-Al—E + Cv(x)

 

J(x)=—D x (11)

where J is the mass ﬂux, D is the diffusion coefﬁcient of the reacting specie in solution,
C is the instantaneous concentration of the reactants in solution at time t, x is the distance
in the x direction or along the direction of the electrical current, 2 is the net charge on the
reacting species, F is F araday’s constant, R is molar gas constant, T is temperature (K),
AE/l is the gradient arising from the change in potential AE over a distance I, and v is the
linear velocity of the solution. Because the test cell solutions were not stirred, the third

term in Eq. 6 can be neglected and hence Equation 1 1 becomes:

6C(x) zF DC AE

—_ — (12)

J = —D
(x) 6x RT 1

 

For uncharged reactant, the second term in Equation 12 can be neglected to yield F ick ’s

ﬁrst law of diffusion as follows:

6C(x)
J(x) = —D—6;c— (13)

Equation 13 was applied to model the rate of mass transfer of Phenanthrene or pyrene to
the electrodes in the DC setup. The following assumptions were made when applying

Equation 12 (or Equation 13) to simulate the Phenanthrene or pyrene concentrations:
1. Reactants (Phenanthrene or pyrene) are instantaneously converted to products
when they come into contact with the electrode. Hence, C(t) at the surface of

electrode is equal to zero for t Z O. This establishes a concentration gradient that

86

drives reactants by diffusion to the surface of the electrodes and this condition is
considered as the maximum concentration gradient that may be established in a
given time step;

2. Steady-state difﬁlsion conditions exist in a given time step which was assumed
equal to 1 hr;

3. Linear mass transfer exists in the given time step of 1 hr; and

4. Linear electric ﬁeld exists in the given time step of 1 hr; and

Equation 13 was applied with C of phenanthrene equal to CO (~ 10 mg/L); or with C of

pyrene equal to C0 (~ 5 mg/L) for t = 0 at the cathode and C = 0 for t Z 0 at the anode.

Assuming a time step equal to 1 hr, the C of phenanthrene or pyrene at cathode was
estimated by subtracting the mass of phenanthrene or pyrene that moved to the anode for
degradation during the previous time step. This computation was carried out for the test

duration of 24 hours and the modeled normalized concentrations were compared (Figure

2-1 1 & 2-12) to the test results for DC tests carried out at j = 3 and 6 mA/cmz.
Schwarzenbach et al. (2003) reported diffusion coefﬁcients of organic solutes equal

-5 2 -1 -6 2 -1
to about 3.0 x 10 cm s for small molecules to 5.0 x 10 cm s for those of molar
mass near 300 g/mol. Based on this data, the diffusion coefﬁcient of phenanthrene was
. -6 2 -1 -6 2 -l . .
estimated as 8.0 X 10 cm s and that of pyrene as 7.0 X 10 cm s from the d1ffusnon

coefﬁcients vs. molar mass chart presented by Schwarzenbach et al. (2003). Equation 13
was applied to simulate only diffusion driven mass ﬂux of phenanthrene or pyrene to the
anode for oxidation. The simulated normalized concentrations are presented in Figures 2-

12&2-13.

87

 

 

 

 

 

 

   
   
   

 

 

 

 

! l I I I ! I I I I I T ﬂ I I

8 fDit‘fusion Only” 4 “““ ‘
B ' Co = 10 mgIL ‘
c“ i . ‘
:2 . Supporting Electrolyte. Nazso4 -
2 “”‘bﬁ """"" DC ' “
E 1 \ \ 1 d
o 1 x 5 ' .
0 I \ . :
g 'l'. ‘ \ "
o _ __________________________________ ...... 5‘; _______________ ._
g - ' . ...... 3 \ ‘ d

.- 1 l .m ‘ . T
g - Model: Diffusion Only . -
2 0-4 ‘ .. _ . . . . . 2 ; """""" I“
3 ~ Model: DIfoSlOl‘i + Migration (j = 3 mA/cm ) ........ .
a, . ---- .
3 . """"" Model: Diffusion + Migration (j = 6 mA/cmz) .
L3 0.2 --------------- -
g . Test Data: j = 3 mA/cm2 .
L- ..
5 I Test Data: j = 6 mA/cm2 ,

o 1 1 1
0 5 10 15 20 25

Elapsed Time [hrs]

Figure 2-12: Simulated concentrations versus test data for phenanthrene tests using DC

88

 

 

 

 

 

 

 

""l""j"‘rl"f'l""
1 .Difl‘uslonml‘; : : ~|
o ’ ‘o. i t — Model: Diffusion Only
8 P.‘o..
- a. g __ _ . . . ._ 2
g— 0.8 _.\\l Model. fous:on+MgratIon(j-6mA/cm)
'5 t ‘5‘“ ......... . . . . 2
g . ; Model: fousuon + Mgratlonq = 3 anm)
é ’ :\ a... . , 2
C 06 _ ............... ..\ ......... 9.6:. TestDataJ=6rnNan
8 - i \'
g \ g I TastDatazj=3anm2
a? 0.4 F- ................................ \. ............. 9.6%.. .............. ................ ..
'o ' ’- \ ------ j '
.g l \\ ...... . 1
E c asmelLl 3 E ........ ‘
E 02 _.... .................................. \ .\ ............. :.........T:s'-
O SuppoﬂingEleotmlyteMaSO i “ E T
Z : 2 4 \:‘ .
: DC ; 3 j ~“
0 1 1 1 1 I 1 1 14I 1 1 1 1 I 1 1 #1 I #1 1 1
0 5 10 15 20 25
HWTII‘I‘BH’I’S]

Figure 2-13: Simulated concentrations versus test data for pyrene tests using DC

The simulated degradation rate as a result of phenanthrene or pyrene migration to anode

only due to diffusion is orders of

experiments run at DC densities equa

supporting electrolyte. When both diffusion and migration are used for simulating the

mass ﬂux of phenanthrene or pyren

normalized concentrations are presented in Figures 2-11 & 2-12 when the net charge, 2,

magnitude less than that observed during the

l to 3 mA/cm2 or 6 mA/cm2 with Na2S04 as the

e to the anode by using Eq. 12, the simulated

was assumed equal to -0.04 and 0.1 l for phenanthrene and pyrene respectively.

89

 

For neutral molecules such as phenanthrene or pyrene, 2 should be equal to zero.
However, it is hypothesized that the presence of anions in the test solution (from the
supporting electrolyte) imparted an effective negative charge on the phenanthrene or
pyrene molecules and rendered them partially negatively charged molecules. These
anions dragged the phenanthrene or pyrene molecules along as they migrated to the
positively charged electrode (anode), where anodic oxidation resulted in the degradation
of the phenanthrene or pyrene molecules. Therefore, the phenanthrene or pyrene
molecules moved in the solution as a result of migration simulated by the second term in
Equation 12 as well as diffusion. Hence, the rate of migration of naphthalene molecules
towards the anode is enhanced by assigning a negative value to z for phenanthrene or
pyrene.

By assigning the phenanthrene and pyrene molecule an effective partial negative

charge of -0.04 and -0.11 respectively, the simulated normalized concentrations versus

time data ﬁts well with that for the experimental data when Na2804 was used as the

supporting electrolyte for DC densities equal to 3 mA/cm2 and 6 mA/cmz. The rate of

anodic oxidation of phenanthrene or pyrene depends not only on the rate of migration of
reacting molecules to the reaction site (the anode or instantaneous anode) but also on the
rate of production of OH radicals at the anode (or instantaneous anode). Hence, as the
current density is decreased, it may decrease the rate of production of OH radicals at the
anode and may become the rate limiting step. When Equation 12 is applied, the rate
limiting step assumed is the rate of mass transfer to the anode (by diffusion and
migration). Hence, Equation 12 would overestimate the decrease in the concentration (or

underestimate the concentration at a given time, t) of the reactant when current density is

90

 

low enough not to degrade all phenanthrene or pyrene mass diffusing and migrating to

the anode (or instantaneous anode). At higher current densities, the predictions from

Equation 12 would be more accurate. It is beyond the scope of this study to estimate the

rate of hydroxyl radical production as a function of current density and other variables.

Hence, it is not possible to calculate analytically at what current densities the assumptions

are valid when Equation 12 is applied.

While Equation 12 is for DC and cannot be applied for an AC, it can be used to

explain the reduced rate of degradation when AC was applied and as the frequency of the

AC was increased from 0.1 to 60 Hz (Figures 2-5 & 2-9). Proposed reasons for the

decrease in degradation rates with increased AC frequency are as follows:

1.

By assuming that the phenanthrene molecules had an effective partial negative
charge of say 2 = -0.04 under our experimental conditions based on the acceptable
ﬁt for DC data, increasing the frequency of the AC resulted in back-and-forth
movement of the reacting molecules, and hence the net mass ﬂux of the molecules
toward the electrodes was reduced. If the net distance traveled by the
phenanthrene molecules to get to an instantaneous anode, L, in Equation 12, is
varied, until the modeled values of concentrations fit with the experimental data
for f equal to 0.1 Hz, le, and 60 Hz the values of L obtained are 30 cm, 100
cm, and 10 cm, respectively, compared to the electrode spacing of 8 cm. For
pyrene, AC frequencies of 0.1 Hz. 1 Hz and 60 Hz respectively resulted in L
values of . . . ..cm, 100 cm, and 110 cm, respectively. Assuming degradation rate at
the electrodes are the same for each frequency, it can be hypothesized that

increase in the AC frequency results in taking longer duration for a phenanthrene

91

i on". L.<-~ 3... . — .

or pyrene molecule(s) to reach the instantaneous anode. This may be due to the
back and forth movement of the molecule as a result of the reversal in the

direction of the current proportional to the frequency of the AC.

. Continuous bubbling of gases at both electrodes (probably 02 and H2) in the test

solution was observed as the tests progressed. However, the rate of bubbling of
the gas decreased as the frequency of the applied AC was increased until no gas
bubbles were visible at an AC frequency of 60 Hz. Simond et al. (1997) and
Comninellis (1994) reported that the hydroxyl radicals, formed on an anode react
with each other to form oxygen gas at the anode to complete the electrolysis of
water molecules. Stock and Bunce (2002) reported that hydrogen radicals formed
on a cathode react with each other to produce hydrogen gas at the cathode. If the
formation of OH radicals and hydrogen radicals are precursors for the generation
of oxygen gas and hydrogen gas at the anode and cathode, respectively, then it is
likely that the rate of gas bubbling at the electrode gives an indication of the rate
of formation of these radicals at the electrodes. The higher the rate of the
observed bubbling of the gas(es), the greater the rate of hydroxyl or hydrogen
radical formation at the electrodes. Hence, based on the visual observation of gas
bubbles, it is likely that increase in the AC frequency resulted in decrease in the
rate of formation of hydroxyl radicals at the electrodes. This impacted the rate of
oxidization of phenanthrene and pyrene at the instantaneous anodes.

. Competing anodic reactions (oxidation) and cathodic reactions (reduction) at the
same reaction sites (on the surface of the electrodes) may have been responsible

for the decrease in the oxidation rate of both phenanthrene (Figure 2-5) and

92

pyrene (Figure 2-9) as the AC frequency was increased. However, in DC
electrolysis, oxidation reactions occurred at the anode only and reduction
reactions occurred at the cathode only and hence there was no competing anodic
oxidation and cathodic reduction at the same reaction site. Therefore DC was

more effective in degrading phenanthrene or pyrene in the test cells.

Electrode Mass Decay

The mass of the electrodes used in the test cells were measured before and after the end
of the tests. Less than 1% loss in the mass of the electrodes was observed during the 48-
hr test period for all tests. The formation of 1ight brown ﬂaky particles that sunk to the
bottom of the solution in the test cells was observed at the end of the testing period when
NaCl was used as the supporting electrolyte. Aﬁer the tests, the solutions were decanted
and the residue was soaked in acetonitrile for 24 hrs and then sonicated for 30 minutes.
HPLC analysis of the extract showed about 3 pg of phenanthrene in the residue from the
phenanthrene tests and about 23 pg of pyrene in the residue from the pyrene tests. These
results indicate that the residue formed did not signiﬁcantly sorb either phenanthrene or

pyrene from of the solutions. The dry mass of the residue was about 4 g.

Intermediate Byproducts

LC/MS analyses indicated molecular masses that suggest the formation of degradation
products of phenanthrene as three isomers of phenanthrol, dihydroxyphenanthrene,
phenanthrenequinone, and naphthalenedicarboxylic acid. The primary degradation
products formed during the pyrene tests were pyrenequinone, pyrenol, dihydropyrenol,

pyrenedihydrodiol, and chloro-pyrenol. The exact isomers formed were not determined.

93

These degradation byproducts formed are relatively more soluble than their parent
compound (phenanthrene or pyrene), hence it is expected that they would be more bio-
available in solution for subsequent degradation by microorganisms. Chlorinated by
product such as chloro-pyrenol may be more toxic than the parent compound pyrene. The
formation of the quinones is consistent with those found by other researchers who also
observed aromatic quinones during electrolytic oxidation of aromatic hydrocarbons
(Brillas et al. 2000; Oturan 2000; Panizza et al. 2000). GC/MS analysis by Goel et al.
(2003) indicated the presence of 1,4-naphthoquinone when naphthalene solutions were
electrolyzed and concluded that naphthalene degradation could be due to direct anodic
oxidation, oxidation by an intermediate (e.g. hydroxyl radicals), or a combination of both.
LC/MS analysis by Pepprah and Khire (2008) of naphthalene in aqueous solution
electrolyzed with a square-wave AC indicated the formation of dihydroxynaphthalene,
naphthoquinone, chloronaphthol, benzenedicarboxylic acid, and malic acid as the
degradation products. The color of the solution during the tests changed to light yellow
after about 4 hrs of electrolyzing the solution with AC. This yellow color may be due to
the formation of quinones in solution. Gas phase sampling was beyond the scope of this
study. Hence, any volatile byproducts which may have formed and escaped from the
solution were not detected or analyzed.

Table 2-2 summarizes the measured initial and ﬁnal values of temperature, pH,
standard redox potential, dissolved oxygen concentration, and electrical conductivity of

the test solutions.

94

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

N2: who may NS msm m . _ N NE o 0Q «Ommmz 2 82555::
w: _ $0 mé m6 adm :N o6 O on «Ommmz S 0:255:05
BE. Po mé Em 5.3 N? o.m o DD e.Ommaz 2 0:855:25
:3 who ..v we Ndm m. _ N No. cc 3% ..OmSZ E 0:255:05
emu who can we Nam». QE Nd— oe Beam ..OmEZ o. 0:255:05:
mum 5mm >6 m6 ode NA: NA: _ Ems—om ..OmSZ 3 0:85:32:
m .8 mg «rev o6 QNM ”2 2: _.o 295m ..OmSZ E 0:255:25
Mme wmw _.w m6 Q3 52 3: ..o 296m 6:2 2 o:o£.:a:o:m
m3 mum ms 90 adm 3: NA: ..o 983% 5:2 2 0:255:02:
mmw 5w :6 me o. .N odm 9m _.o 83$ 5:2 2 0:25:22:
3::— EEE 35...— EEE =25— .33:—
—m1_ F5925. FE 5.5%
33325.30 .00. 36:0: 5:259...— o>a>> 3%.:— ﬂicks—m
$3.58.”.— In 9.39.2.th 2.9.2.0 O< 04.. U w:_tona=m ESQ—:30

 

 

 

 

 

 

 

 

 

80:28 :8 75:8 98 :8 52 mo 33:26:00 18.5020 98 .2: 92382.83 mo mos—g 35.: :5. EEE 35:32 ..N.N 2an

95

 

 

 

 

 

 

 

 

 

 

 

 

Va 5 no we as. :2 o c 3586 Q2 m 82$
3 m3 3 3 3m 9: 2: o u: .832 m 2.23
o: _ was 3 3 3m :2 3 o u: v0.3.2 m 055
m: as 2. 3 :2 0.2 3. o u: .832 m 823
8a 96 .3. 3 :2 «.3 2: 8 saw .832 m 223
was was 3 3 3m 2m 2: 8 235 .832 m 223
a... :8 nm 5 EM EN 2: _ swam 632 m 82$
23 as S. a :2 we .2: S 23$ .832 m 085
as :w 3 we was. 2: 2: S seam 62 m 055
80 :w 3 3 SM 2: 2: 3 22am 52 m 0:05
m3 5 as 3 EN 2:. 9m 3 «saw Caz m 255
m5 5 S we 92. :2 o o 638 Q2 2 sausages:

 

 

 

 

 

 

 

 

 

 

 

 

PEOU N-N 03:...

 

96

SUMMARY AND CONCLUSIONS

The objective of this study was to use an alternating current (AC) to investigate the

degradation of phenanthrene and pyrene in spiked aqueous solutions. Phenanthrene and

pyrene solutions having an initial spiked concentration of 10 mg/L or 5 mg/L,

respectively, were electrolyzed at three AC densities equal to 3.0 mA/cmz, 10.2 mA/cm2

and 18.5 mA/cmz. A square wave AC having a frequency equal to 0.1 Hz was used in a

majority of the experiments. Other frequencies explored were 1 Hz and 60Hz, and

degradation rates using 60 Hz sine wave was compared to 60 Hz square wave. The key

conclusions are as follows.

1.

2.

AC is able to degrade phenanthrene and pyrene in aqueous solutions;

Greater the current density, faster the initial rate of degradation;

. Migration contributed signiﬁcantly to the movement of reacting molecules to the

reaction sites (the electrodes) as compared to difﬁJsion;
Increasing the AC frequency resulted in decreased degradation rates;

A square wave AC produced a higher degradation rate than a sine wave AC

supplied at equal rms current density of 10.2 mA/cm2 and frequency of 60 Hz in

the degradation of phenanthrene;

Electrode fouling was one of the reasons for the reduction in the rate of
degradation as the test progressed;

The degradation rate of both phenanthrene and pyrene in solution could be

described as pseudo ﬁrst-order decay; and

97

8.

10.

Formation of 2,3-Dihydroxybenzoic acid and 2,5-Dihydroxybenzoic from the
oxidation of salicylic acid is consistent with reported ﬁndings that OH radicals
were the primary oxidizing agents in electrochemical degradation; 9
The rate of anodic oxidation of phenanthrene and pyrene depends on the rate of
production of OH radicals at the anode (or instantaneous anode) as oxidizing
agents as well as the rate of migration of reacting molecules to the reaction site
(the anode or instantaneous anode). At higher current densities, not only more OH
radicals are produced at the anode but also the rate of migration of the reacting
molecules to the anode (or instantaneous anode) increases during a DC
application. However, when AC is used, the change in the direction of the current
decreases the rate of migration of the reactant to the instantaneous anode. Hence,
as the frequency of the applied AC increases, the rate of degradation decreases;
and

Both phenanthrene and pyrene molecules in solution exhibited an “effective”

partial negative charge due to the presence of the supporting electrolyte.

98

PAPER NO. 3: EFFECT OF ALTERNATING CURRENT
WAVEFORM AND FREQUENCY ON THE DEGRADATION OF
NAPHTHALENE IN AQUEOUS SOLUTION

ABSTRACT

The key objectives of this study were to: (1) use alternating current (AC) to investigate
the effect of AC waveform and frequency on the rate of degradation of naphthalene in
aqueous solution; (2) explore the mechanism of degradation of naphthalene when
subjected to AC and DC; and (3) prove that anodic oxidation of naphthalene occurred on
both electrodes during AC electrolysis, unlike direct current (DC) electrolysis in which

anodic oxidation of naphthalene occurred on only one electrode (the anode). The results

indicated that at 3 mA/cm2 peak AC density, the rate of degradation of naphthalene in

aqueous solution having ~ 20 mg/L (0.15 mM) initial concentration, decreased with
increasing AC frequency. For square, sine, and triangular AC signals explored, square
wave AC signal resulted in the fastest degradation rate of naphthalene. Hydroxyl radicals
were identiﬁed as the primary oxidizing agent in the solution. Hydroxylated naphthalene
byproducts were detected in both compartments of a divided cell setup indicating that
anodic oxidation occurred within both compartments in AC electrolysis, unlike DC
electrolysis in which hydroxylated byproducts were detected in only the anolyte but not
in the catholyte. The Nenrst-Planck equation was used to model the rate of mass transfer
to the electrodes where reactions were expected to occur. A comparison of the simulated
concentration-time proﬁles with the test results indicate that both diffusion and migration
processes were responsible for the rate of mass transfer of naphthalene to the reaction

sites (surface of electrodes or in the vicinity of the electrodes) where naphthalene was

99

degraded or converted by the hydroxyl radicals. However, migration was the dominant

phenomenon for mass transfer to the electrodes in the AC and DC electrochemical cells.

INTRODUCTION

The United States Environmental Protection Agency (USEPA) estimates that 3,000 to
5,000 former manufactured gas plant (MGP) sites are located across the United States
and many of these contaminated sites contain substantial quantities of PAHS (USEPA
2000). Remediation of contaminated soil and groundwater is an important issue that is of
concern among environmentalists and hydrogeologists. In-situ and ex-situ treatment
techniques may be used to remediate contaminated soil and groundwater. Conventional
remediation techniques include pump-and-treat, incineration, long term storage, and soil
washing. Other approaches include biodegradation, and advanced oxidation processes.
Alshawabkeh and Sarahney (2005) proposed the concept of Electrochemical Redox
Barrier (ERB), where contaminants could be potentially degraded by placing electrodes
in the soil to intercept the contaminant plume and applying DC to generate the redox
conditions required to degrade the contaminants as the groundwater ﬂows through this
reactive zone. Pepprah and Khire (2008) reported that AC is capable of degrading PAHS
such as naphthalene, phenanthrene, and pyrene in solution resulting in the production of
intermediate byproducts such as quinones and diols. A funnel-and-gate (Birke et al. 2003;
Gupta et al. 1998)) electrochemical redox barrier (FERB) (Figures 3-1 a & 1b) may be

used for in-situ treatment of contaminated plumes.

100

Scam ozuomom 3280208805 030-3?355m uoﬂmoﬁoﬁ: .«o 26$ .85 ”Ag—-m oSmE

 

 

5302:. 30:
. 3333586

/

$32823 J.1.....i-__u._.._.JJ..._-...-,..w

38:... 52:3: ...- J7, w-_.,_...._._a_.J
8.22.8 ,- _.

i I’-
I,

!/

combow l
9:23 53:35 08828.; l. .

.225". ll

 

Saw ,

2.5.: x _, .

EaEEScoo \.- (....
.. \- ...,
... econ

- -.----. 350$

O: ON COSONQL
8.59.8585

 

 

>>m_> z<.._n_

101

cot—am 0363M 3380:2585 oumwéconccsm voﬂmoﬁogm mo 5:08-880 “3:4” oSmE

 

 

:o_uoo..-._u >3: 3532320

.i ill ill ll

econ guano. Raga—.8535
_

 

 

 

 

 

 

 

 

 

 

 

lull @333
... V/ocou 350m
0M0 two ”En—ha W, \-
3536505 603558530 ... «suspeﬂcﬁu;
o a L \\ o o . ....
atom -¥\\ \ woo /\\.\
.. . \ ﬂ . \lx- 3233‘
-\. . .- lil- .
\ - .- i lam. NM
@580 52:85 noumuoton. ......»- _ ..., «CON $395395
88.8% 38% 52:3: 8888.. O 8,28

 

 

< - < 20:.0mm mwOm—U

102

The FERB is made up of low hydraulic conductivity cutoff walls, e.g. slurry walls or
sheet piles for the funnel with a gate(s) that contain(s) in-situ reaction zones. Ground
water primarily ﬂows through high conductivity gates. In this approach, the electrodes
can be placed in a casing in the gate region containing the groundwater contaminated
with PAHS in the subsurface. Furthermore, AC and/or DC are applied to degrade PAHS
present in the groundwater. To ensure complete removal of PAHS, several reactive
barriers may be constructed in series, down gradient of the contaminant plume to
intercept the plume as it travels through these reactive zones. AC electrolysis may also be
used as an ex-situ treatment technique whereby contaminated groundwater is pumped out
of an aquifer for subsequent above surface treatment in an electrochemical reactor.
Pepprah and Khire (2008) have shown that naphthalene in aqueous solution can be
degraded by AC electrolysis. They have also presented the effect of current density for a
ﬁxed frequency, and type of supporting electrolyte on the rate of degradation of
naphthalene. The key objectives of this study were to:
I. investigate the effect of AC waveform and frequency on the rate of degradation
of naphthalene in solution, which is a model PAH compound;
2. prove that electrochemically generated hydroxyl radicals (OH') were
responsible for the degradation of naphthalene; and
3. demonstrate that anodic oxidation of naphthalene occurred on both electrodes
when AC electrolysis was used, while for DC electrolysis, anodic oxidation of
naphthalene occurred on only one electrode (the anode).
Pepprah and Khire (2008) have reported that the rate of degradation of

naphthalene in aqueous solution in a laboratory scale experiment, increased with

103

increasing peak AC density. In addition, they also reported that the rate of degradation

when NaCl was used as the supporting electrolyte was greater than when Na2804 was

used as the supporting electrolyte.

Naphthalene, which contains two fused benzene rings, was selected as a model
PAH compound because Alshawabkeh and Sarahney (2005), Goel et al. (2003), and
others have also used naphthalene to investigate electrochemical degradation in aqueous
solution. Also, naphthalene has chemical and physical properties similar to other PAHS,
except that it is more soluble in water, a property that makes naphthalene more available
in aqueous solution and more exposed to reactivity than others PAHS (Alshawabkeh and

Sarahney 2005).

MATERIALS AND EXPERIMENTAL METHODOLOGY

The experimental setup is illustrated in Figure 3-2 and the setup components are

described below.

104

9:8 Econ—toga 05 .«o osmﬁocom ”Wm oSwE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

93 Bob
l ll com—30m 250:?“
L E ocoﬁﬁnmuz
A
mm- mum. t a t mﬂwmm ....
lilymlﬁiwﬂwwaﬂlmmw OQ mHiHHHHHHNHHWH
:23 noon? . w.» -..,MM mo. ammunw - . l ..8“;on awn—w .85 H
l l .. ... mum- u w
\...\\\x\ 7 ill i. i. [J l...l.«lzl
. Quebec—o Esaaamh.
m8 couch.
59: atom o< > m:
O
_
l i l0. q ,j. 0% 1} 1.... .l
t l t -lb-..w.tt... :1 - - l - o . o o
uouuuocow .. .

 

 

 

cove—Sm >395 33cm 09 \ Ema—mam 839mm

105

Reaction Chamber

The test cell setup consisted of a 1 liter Pyrex glass beaker with a Teﬂon cap. These
materials of the cell were selected to eliminate or reduce sorption of the contaminants
onto the walls of the reaction vessel and the cap. Titanium plates (99% purity) were used
as electrodes for this study. Other researchers (Alshawabkeh and Sarahney 2005; Li et al.

2005; Goel et al. 2003) used titanium as a base metal and coated it with mixed metal

oxides such as Rqu, IrOz, and SnOz. The titanium plates used in this study were not

coated. Uncoated titanium is cheaper than commonly used electrodes such as platinum, or
titanium coated electrodes. Hence it was selected for this project. Also, titanium offers
resistance to corrosion in a wide range of aggressive conditions (Titanium and the
Environment 2002) and thus, could be used for potential ﬁeld scale application. The
titanium sheets used as electrodes in this study were 12.0-cm-long by 5.0-cm-wide by

l.72-mm-thick and the distance between them was maintained equal to 8.0 cm. The

. . . 2 2
active contact area of the solution With the electrode ranged from 53 cm to 55 cm as

samples were collected for analyses.

Electrical Equipment

A 2 MHz Sweep Function Generator capable of generating AC signals having frequency
ranging from 0.1 Hz to 2 MHz was used. The power supply consisted of a bipolar
operational power supply/ampliﬁer capable of producing 200 V AC/DC and 1 A AC/DC
output or a maximum electrical power of 200 W. Voltage and current measurements

were obtained with the aid of an oscilloscope.

106

Spiked Solution Preparation
Aqueous solutions, spiked with naphthalene were prepared by adding naphthalene (98%,
purity) crystals to one liter of deionized (DI) water in capped Pyrex bottles and allowing

the suspensions to stand for more than three days. This was followed by the addition of

0.5 g of anhydrous sodium sulfate (NaZSO4) to each suspension, to increase its electrical

conductivity. The excess suspended naphthalene crystals were then ﬁltered off and the
collected ﬁltrate was used for the tests. Sodium sulfate was selected as the supporting
electrolyte because previous studies (Goel er al. 2003) used sodium sulfate in

electrochemical degradation of naphthalene in aqueous solution. Also Pepprah and Khire
(2008) showed less degeneration of the electrode surface when Na2804 was used
compared to when NaCl was used as the supporting electrolyte. The resultant
concentration of NaZSO4 was approximately 500 mg/L (~3.5 mM). This concentration

was selected to meet the USEPA has proposed a maximum contaminant level goal of

sulfate as 500 mg/L (~3.5 mM). Saracco et al. (2000) performed tests where electrolyte

salt (e. g. NaCl, NaZSO4) concentrations ranged from 0.2 — 2 M and concluded that higher
salt concentrations reduce the electrical resistance between the electrodes and hence will
reduce the operating costs. However, greater salt concentration would result in greater

post-treatment costs and hence would be less favorable. The only exception might be

represented by discharging the treated efﬂuent into the sea (Saracco et al. 2000).

107

Testing Procedure

Tests were conducted in a batch mode in l-liter undivided cells as well as in a divided
cell setup with l-liter solution in each (anode and cathode) compartment. Prior to passing
current through the cells, initial samples were taken for high performance liquid
chromatography (HPLC) analyses. Replicate tests were conducted for majority of the
tests. The effect of AC waveform (square, sine and triangular) on the degradation rate of
naphthalene in solution was explored at a frequency of 60 Hz. This frequency was
selected because AC from the power grid is supplied at 60 Hz frequency in the United
States. The effect of AC frequency on the degradation rate of naphthalene was
investigated at frequencies 0.1 Hz, 1 Hz, 10 Hz, 100 Hz, and 1,000 Hz using a square
wave signal. A square wave signal was selected because our test results indicated that for
a given peak current, the square wave was the most efﬁcient among the three wave
signals for the electrochemical degradation of naphthalene in the test solutions. Pepprah

and Khire (2008) reported that although the initial degradation rate of naphthalene at 6

mA/cm2 peak AC density was faster, both 3 mA/cm2 (165 mA, peak current) and 6

mA/cm2 (330 mA, peak current) AC densities ultimately resulted in approximately the

same ultimate degradation aﬁer 48 hours of AC electrolysis. Therefore running these

tests at a lower AC density resulted in savings in electrical energy expended for the same

. . . . . 2
ultimate conversron of naphthalene. An additional reason for the selection of 3 mA/cm
. 2 .
was because the temperature of the test solutions electrolyzed at 3 mA/cm remained
approximately constant at the room temperature (~19 0C) throughout the test whereas for

2 . . o
the 6 mA/cm current denSIty, the temperature increased to about 25 C, so that loss of

108

naphthalene from the test solutions by volatilization as a result of increased temperature
is minimized by running the tests at a lower current density. The peak AC density was

obtained by dividing the peak current by the average active cross sectional area of the

electrodes in contact with the solution. Tests were conducted using 3 mA/cm2 square
wave at 0.1 Hz with salicylic acid as a hydroxyl radical (OH) probe, as well as tertiary

butanol as OH. scavenger to explore the degradation mechanism of naphthalene.

During the tests, samples were collected for HPLC analyses at time intervals of O,
1, 2, 4, 8, 24, and 48 hours. Samples were analyzed directly without extracting the
naphthalene from the aqueous solution. This procedure minimized potential interferences
from reagents, solvents, and glassware required for the extraction method (Alshawabkeh
and Sarahney 2005). HPLC analyses were performed using a diode array detector
(detection at 255 nm), a Waters C-18 PAH, 250 mm x 4.6 mm x 5 pm column, an
automated gradient controller, and deionized water (DI) / acetonitrile mobile phase. The
flow rate of the mobile phase was 1.2 mL/min and the column was ﬂushed with 10:90
(v/v) acetonitrile / DI water for 5 minutes followed by 60:40 (v/v) acetonitrile / DI water
for 10 minutes, and by 100:0 (v/v) acetonitrile / DI water for 5 minutes. This gradient
was used for the analysis in order to obtain appropriate separation of the degradation
product peaks on the chromatograms during the tests. Standard solutions were prepared

to bracket the expected concentration level of the analytes. Calibration curves with an

2 . . .
average R value of 0.998 were used to determme the concentration of naphthalene 1n the

solutions as the tests progressed. Samples were also collected for Liquid

Chromatography/Mass Spectrometry (LC/MS) analyses to determine the degradation

109

products formed. Control cells, similar to the test cells, except without any current passed
through them were set up, sampled, and analyzed at the same time intervals as the test
cells. The test cells were placed in a relatively large volume water bath maintained at the
room temperature to reduce the rate of temperature rise of the test solutions as a result of

Joule heating.

Experimental Parameters Monitored

Apart from monitoring the concentration of naphthalene in aqueous solution over time,
other experimental parameters monitored included the initial and ﬁnal values of pH,
temperature, redox potential, electrical conductivity, and dissolved oxygen concentration
of the solutions. A portable dissolved oxygen/pH meter, equipped with a pH electrode, an
oxidation-reduction probe and dissolved oxygen probe, was used to measure pH, redox
potential and dissolved oxygen, respectively, of the test solutions. An electrical
conductivity meter was used to measure the electrical conductivity and a thermometer
was used to measure the temperature of the spiked solution in the control and test cells

throughout the testing period.

RESULTS AND DISCUSSION

Table 3-1 outlines the list of parameters explored and varied in the experimental program.

110

Table 3-1: Summary of Experimental Parameters

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Compound Cell AC AC Peak AC Density No.
Frequency Waveform 2 A/L of
[sz mA/cm m tests
Naphthalene Undivided 60 Square 3 1 65 2
Naphthalene Undivided 60 Sine 3 165 2
Naphthalene Undivided 6O Triangular 3 165 2
Naphthalene Undivided 0 DC 3 165 2
Naphthalene Undivided 0. 1 Square 3 1 65 2
Naphthalene Undivided 1 Square 3 165 2
Naphthalene Undivided 1 0 Square 3 165 2
Naphthalene Undivided 100 Square 3 165 2
Naphthalene Undivided 1000 Square 3 165 2
Naphthalene Undivided - - 0 O 2
(Control) (Control)
Naphthalene Divided O. 1 Square 3 82.5 1
Naphthalene Divided 0 DC 3 165 l
Naphthalene Divided - - O 0 1
(Control) (Control)
Naphthalene Undivided O. 1 Square 3 165 3
/tert-butanol
Salicylic Undivided 0.1 Square 3 165 1
acid
Salicylic Undivided 10 Square 3 165 1
acid
Salicylic Undivided 1000 Square 3 165 1
acid
Salicylic Undivided 0 DC 3 165 1
acid
Salicylic Divided 0.1 AC 3 165 1
acid

 

 

111

 

Table 3-1 Cont’d

 

 

 

 

 

 

 

 

 

 

 

Salicylic Divided DC 3 165 1
acid
Salicylic Undivided - 0 0 1
acid (Control) (Control)
Salicylic Divided - O O 1
acid (Control) (Control)
Total
Number 38
of Tests
Notes:

1. The volume of the test solution in an undivided cell was 1 liter and it was 1 liter
for each compartment of the divided cell having total volume equal to 2 liter.
2. All tests were performed by passing 165 mA current (peak AC or DC).

3. Current density [mA/cmz] = current [mA] / cross sectional area of electrodes
[cmzl
4. Current density [mA/L] = current [mA] / volume of solution [L]

Effect of AC Waveform

Figure 3-3 illustrates the effect of the shape of AC wave form on the degradation rate of

naphthalene in the test cell solution.

112

 

't'j'IIIITUI'TIITT'l'rliﬁﬁrthU—Iﬁr

 

 

Normalized Naphthalene Concentration, ClCo

 

 

 

l , .
F § co=2o mgIL
0-2 C’E’Supporting Electrolyte: Nazso4 """ """" """"""" "'l
i ' ‘ l
l gAc: Square Wave,f=60 Hz,j=3mA/cm2
oH....1....1....L...Lt..mi....
0 10 20 30 40 50 60
Elapsed Time [hrs]

Figure 3-3: Effect of AC waveform on the degradation rate of naphthalene

The error bars in Figure 3-3 represent the maximum and minimum values of the

normalized concentration (C/Co) for the initial test and the replicate test recorded for the

total four samples. By passing square, sine and triangular wave currents through

naphthalene test solutions (~ 20 mg/L (0.15 mM) initial concentration) in separate tests,

the square wave produced the highest degradation rate (k ~ 1.77 x 10'2 hr'l), followed by

the sine wave (k ~ 1.34 X 10.2 hr-l), and then the triangular wave (k ~ 1.16 X 10‘2 hr'l).

Thus, for a given peak current, the square wave was the most efﬁcient among the three

waveforms. The primary reason for this ﬁnding is that the rate of degradation is a

113

function of the root mean square (rms) current. The rms current, Irms is the DC equivalent

for the AC signal. For a DC signal, the rms current is the same as the peak current. Figure

3-4 illustrates a schematic of these three wave forms and the resulting rms values are:

Irms(Sine) = 0.707110 (1)
I m (Square) 2 I p (2)

Inns (Triangular) = 0.5771p (3)

Ip

l ,(g , L 1,. _.

I p (Square) = I p (Sine) = I p (Triangular)

Figure 3-4: Square; Sine; and Triangular AC signal having the same peak current

The greater the magnitude of current passed through the solution for a given period, the
greater the degradation rate and hence the observed trend as shown in Figure 3-4.
By applying an AC, which resulted in constant polarity reversal of the electrodes

at a rate dependent upon the AC frequency, anodic oxidation of naphthalene in solution

114

occurred on both electrodes. Previous studies on electrochemical degradation have
focused on the use of DC where the cathode acts as a “passive” electrode in the oxidation
process. However, when an AC is used, the constant polarity reversal creates anodic
oxidation reactions on both electrodes alternately, depending on their polarity in a given
AC cycle. During one half of a cycle when one of the electrodes has a positive electrical
potential (instantaneous anode), it instantaneously acts as the anode while the other
electrode, at a negative electrical potential (instantaneous cathode), instantaneously acts
as the cathode. In the subsequent half cycle, polarities are reversed and the electrodes
acquire electrical potentials opposite to what they had in the previous cycle and this

continues throughout the period of application of the AC.

Effect of AC Frequency

Square wave AC having a peak AC density of 3 mA/cm2 and frequency varying from 0.1

Hz to 1,000 Hz were used to investigate the effect of AC frequency (f) on degradation
rates of naphthalene in the test cells. Figure 3-5 presents the effect of AC frequency on

the degradation of naphthalene.

115

=2 IL
Co 0mg

Supporting Electrolyte: Nazso4

_'AC: Square Wave, 1 = 3 mA/cm2 (AC or DC)
..............,.

 

- a I I I I I I I ! I I I I

 

 

Normalized Naphthalene Concentration, CICo

 

 

 

 

_ z0.1Hz
o'1....1L...1....i....i..j1.°'.*=.(°9).'
0 10 20 30 40 50 60

Elapsed Time [hrs]

Figure 3-5: Effect of AC frequency on the degradation rate of naphthalene

The overall degradation of naphthalene during the 48-hr period ranged from 41% for the

highest frequency (f = 1,000 Hz) to 87% for the lowest frequency (f = 0.1 Hz) for a ﬁxed

1p equal to 3 mA/cmz. A DC 0" = 0) density of 3 mA/cm2 resulted in 96% degradation

after 48 hrs of electrolysis. These results indicate that the rate of degradation of

naphthalene in aqueous solution decreased with increasing AC frequency. This ﬁnding is

116

consistent with the trend observed by Chin and Cheng (1985) that the rate of conversion

of phenol decreased with increasing alternating voltage frequency superimposed on a DC.

2 . . .
For the 3 mA/cm current densrty used 1n the tests, at frequencres less than 60 Hz,

continuous bubbling of gases at the electrodes may have also contributed to the reduction
in concentration of naphthalene in the test solution as a result of sparging. Determining
the component of mass loss due to sparging was beyond the scope of this study.
However, at frequencies of 60 Hz or more, no gas bubbles were observed with unaided
eye.

The degradation of naphthalene in all the test solutions depicted pseudo-ﬁrst-

order degradation kinetics. A plot of ln[C/C0] versus time yielded the observed pseudo-

ﬁrst-order degradation rate constants (k) and corresponding computed half-lives (t1 /2)

which are reported in Table 3-2.

117

 

 

 

 

 

 

 

 

 

 

 

 

mm 332 8.8 2: 82 m 228 82 8228: 8.22282
8 822 2.: 2.2 2 m 22cm 2: 8228: 822282
3 282 2.8 2:. E m 822m 2 8228: 822282
2 882 2.8 88 2: m 88m _ 8228: 822282
8 882 2.2 8.2. 82 m 22cm 3 8228: 822282
8 282 2.2. w _ .2 82 m on o 8222: 822282
.8 882 2.8 e _ .2 82 m 2885 8 8222: 822282
a 8.22 2.: 3.2 2 m 85 8 8228: 822282
2 282 2 .8 E2 .2 m 22cm 8 8228: 822282
.1...

2: av 23.? mg: 3)::— _ 592:: F:—

=e=acﬁuon 1.2.5:— x v N 528263 55:32,.—
.x. N: a: 2 225 u< 28.. 3. u< :6 8.825

 

 

 

 

 

 

 

 

 

238% 3582225 mo baegm ”NA 2%...

118

 

 

 

 

 

 

 

 

 

 

mm mad ....2 82 32 m 82% 2 8228: 28 2222
2 8.82 8.2 82 2: m 825 3 2.228: 28 2222
222 22822
2 88¢ :12 32 2 m 088m 3 82225 88822282
922
misses
2 282 2.2 28 a: m 22$ S 8228: 88822282
222
2.2225 2.8 6822
5 882 8.2 m2 2 m 22$ 3 82822282
@2288
.8 223 8.8 22 mg on o 2222 822282
aéoﬁc
8 882 mg 2.8. n2 m on o 8222
3:08
ow 3.3 2.2 mom as m 82$ S 8222 822282
2 =08
ow 2.82 2.2 3m as m 22$ S 8222
28:59 98:5an
- - - - o o - - 8228: 822282

 

 

 

 

 

 

 

 

 

 

228 3 222.

 

119

N

DVq

.GQ .8 U< 283 E250 <8 m3 wﬁmmmn 3 32:8th 225 $93 =<

.2232on «0 258508 05 mm m

.2 mo 0.5.23 wcmccoamoboo 2: 2 NE
28:28 :8 28 of E 0:225:93 mo 2:22.00 82 cots—2232 2220-55 £88K 323,20 2 2
A :3 cows—cm #23 05 :23 88:00 E 328820 mo 83 3:268 380 \ F25 €8.25 .I. H 802:: 2:252 3250

N

N

ANMVIA

”$on

 

 

 

 

 

 

$2288

8 - - - ms m on o 8222 28 2222
958$

8 82¢ 22 2 2 E m on o 8222
22.22

G 882 2.2. 82 ms 8 u< S 8222 28 2282
328

.8 E82 8: 8.8 m2 m u< 3 2.222

8 $3 $8 8.2 E m on o 8228: 28 222.2

N - - - ms m 225 82 B228: 28 222.2

 

 

 

 

 

 

 

 

 

 

228 ~-m 22¢

 

120

For a square wave with peak AC density of 3 mA/cmz, an empirical relationship

between the applied AC frequency (f) and the observed pseudo first-order degradation
rate constant (k) of naphthalene in aqueous solution at ﬁve AC frequencies (0.1 Hz, 1 Hz,

10 Hz, 100 Hz, and 1,000 Hz), was derived from Figure 3-6 as shown in Equation 4

k = —O.00810gf + 0.033 ............. R2 = 0.9593 (4,

 

0.045 T r T
l — y =§0.0329 - 0.00804Iog(x) R= 0.97043

   

0.025

0.020 2" Co a 20 mglL .

Supporting Electrolyte: Na 80
0.015 t... 2 ‘

 

AC: Square Wave, 1 = 3 mAlcmz

0.010 ‘ ‘
0.1 1 10 100 1000

Pseudo-First-Order Degradation Rate Constant [hr'1]

 

 

AC Frequency [Hz]

Figure 3-6: Effect of AC frequency on the pseudo-ﬁrst-order degradation rate constants

Decrease in naphthalene concentration in the control cell over time was observed as

shown in Figure 3-5. However, the loss of naphthalene from solution in the control cell is

121

significantly less compared to the decrease observed for the AC and DC test cells. Goel et
al. (2003) and Alshawabkeh and Sarahney (2005) also reported naphthalene loss in their
control cells over time and attributed it to factors such as volatilization, potential sorption

to cell components etc.

Hydroxyl Radical Probe Study
Brillas et al. (2000) reported that electrochemical systems generate hydroxyl radicals at

the anode due to the oxidation of water:
0 + _
H2090Hads+H +6 (5)

Nakamura et al. (2005) proposed a mechanism based on selective redox reactions with
various radicals generated by AC electrolysis that allowed both oxidation and reduction
at the same reaction site between adjacent electrodes. Based on the analysis of
intermediate products of the degradation process, it was revealed that the mechanism

involved selective redox reactions with short-lived and highly reactive radicals: hydrogen

atoms (H.) and OH. that are electrochemically generated at the electrodes (Nakamura et

al. 2005).

OH radicals attack the benzene rings of aromatic compounds to produce

hydroxylated compounds. Various OH. radical probes and techniques have been used in

published studies to detect the generation of OH' in reaction media. The probes and
methods include salicylic acid (Li et a1. 2003; Marselli et al. 2003; Oturan 2000; Jen er
al. 1998; Scheck and Frimmel 1995), 4-hydroxybenzoic acid (Anderson et al. 1987),
benzoic acid (Oturan and Pinson 1995), para-chlorobenzoic acid (pCBA) (Elovitz and

von Guten 1999; Haag and Yao 1992), 2,2-diphenyl-l-picrylhydrazyl (DPPH) (Sehgal et

122

al. 1982), 5,5-dimethyl-l-pyrroline-N-oxide (DMPO) and electron spin resonance (ESR)
(Marselli et a1. 2003), p-nitrosodimethylaniline (RNO) and spectrophotometer
measurements (Tanaka et al. 2004), and RNO / ESR (Comninellis 1994). Because ESR

methods require expensive and sophisticated equipment, they were not used in this study

due to budget constraints. Trapping of OH. using salicylic acid, and HPLC measurements

were selected as hydroxyl radical probe technique for this study. Jen et al. (1998)

reported that the oxidation efﬁciency of OH. is usually limited by the rates of OH.

generation and hence, increasing the OH. generation and keeping it at a high level are the

main factors that control the degradation efﬁciency of aqueous organic substances. Haag

and Yao (1992) also reported that in contrast to most other oxidants, reactions of OH.

with organics containing OH or C-C multiple bonds generally proceed with rate

constants approaching diffusion-controlled limit (1010 M.1 3.1) and therefore oxidation

rates are usually limited by the rate of OH' generation and competition by other OH.

scavengers in solution rather than by inherent reactivity with the oxidant. Figure 3-7

shows the change in concentration of salicylic acid solutions [~ 20 mg/L (0.14 mM)

mltlal concentration] SUbJCCth to square wave Signals havmg 3 mA/cm peak current

density at frequencies equal to 0.1 Hz, 1 Hz, 10 Hz, 100 Hz and 1,000 Hz respectively,

and DC (f: 0) at 3 mA/cm2 current density.

123

Co = 20 mgIL

Supporting Electrolyte: Nazso4

AC: Square Wave, j = 3 mAIem2

 

 

 

 

 

 

 

/Control

1 I I I I l U I I I I I I 'ﬁ 1. Vi I 1 ! U I I I ! I ITV
0 RE}? ﬁ_ E _ ----- ‘ r -- =
o ’ 2 - l f 100H '
- : = z
,8 l '= s t s ; -
E 0.8 --~ ----- -------- , ----- -------- -- ------------ ----------- -
E ' ' 1 1 ' V rum: 1
3 _ -
g l- : : : ' : 1
o 0.6 ----------- ------- --------- -
E _ 1 . ‘ 3 zf—1Hz +
o
< » l -
.2 " : t . : : ‘
5' 0.4 - """""" ------------- """""""""" """""""""" """"" .‘?f=0.1 Hz'-
% b I I I I I u
m P I
'8 .. : : : : : .
'% o_2 . ............. 2. ........ .............. 2‘ ............ ............ ..
E . . q
z I- . f I

o I I I I ‘ I I I I ' I J Lgl I I I I l I I I !=-o.(D-c)l

0 5 10 15 20 25 30

Elapsed Time [hrs]

Figure 3-7: Effect of AC frequency on the rate of degradation of salicylic acid (hydroxyl
radical probe)

The salicylic acid in solution was hydroxylated to two derivatives, 2,3-dihydroxybenzoic
acid and 2,5-dihydroxybenzoic acid which were detected by retention time and UV
absorption spectral matching with authentic samples. The lowest AC frequency (0.1 Hz)

produced the fastest degradation rate for AC electrolysis, and the 1,000 Hz produced the

slowest degradation rate as depicted in Figure 3-7. The DC density, 3 mA/cm2 (f = 0)

124

showed the fastest degradation rate than all the AC tests. The control cell did not show

any measurable reduction in concentration.

Hydroxyl Radical Scavenger Study

Tertiary butanol was used as OH. scavengers to further indicate that OH. were

responsible for the oxidation of naphthalene in the electrochemical cells. Tertiary butanol

is widely used as an OH. scavenger in radiation chemistry and biochemistry (Wand
1995). Li et al. (2003) used tertiary butanol as an OH. scavenger and reported that since

tertiary butanol reacts preferentially with 011., the amount of OH. in the reactor decreased
and hence the rate of aniline degradation decreased. Figure 3-8 shows the rate of

degradation of naphthalene in the presence of the varying concentrations of the OH.

scavengers.

125

 

r'IIIII.IUIII'IITT'rIIIfr'It'I

L 37 ‘ . Co 8 20 mglL .
p : —i a I O I
L g .- \\ 3 5 Supporting Electrolyte. NaZSO4
o 8 ' '\ 5 5 2 '
' b ' “ “; """"""" I ' "AC: Square Wave, f = 0.1 Hz, 1 . 3 mAIcm "
' r 3 .

t-Butanol as OH Radical Scavenger -

 

Normalized Naphthalene Concentration, CICo

 

 

 

: 'l

- Ii

': : . : : : J

0,4 .3 .............. z .............. 1.5mM .......... = ............... s ............ -
.5 1 E V ' ﬂ5mM ' .

0'2 :- """ ’ """"" § """""" """"""" 50"“ 00.15mm‘"
a s a i ‘3: .

.2 3 3 s a s .

o I I I4 I I I I I I I I I I I I I I I I l I I I I I I I I I

0 1O 20 30 40 50 60

Elapsed Time [hrs]

Figure 3-8: Concentration of naphthalene in solution versus time in the presence of
varying concentrations tertiary butanol (hydroxyl radical scavenger)

The scavenger reduced the degradation rate of naphthalene as its concentration was

sequentially increased in the test cells.

The hydroxylation of the OH. probe (salicylic acid), and the suppression of the

degradation of naphthalene in the test cells in the presence of the OH. scavenger, is

consistent with numerous studies that OH radicals were the primary oxidizing agent in

the electrochemical cells for AC as well as DC tests.

126

Modeled versus Measured Concentrations of Naphthalene

It is generally believed that electrochemical reactions occur on the surface or in the
vicinity of the electrodes. Therefore, the rate of reaction of the organic reactants in the
test cell depends on the rate of generation of OH radicals at the electrodes as the
oxidizing agents, as well as the rate at which the reacting molecules move to the
electrodes for subsequent oxidation by the OH radicals. Mass transfer in solution occurs
by diffusion, migration and convection. Diffusion and migration result from a gradient in
electrochemical potential and convection results ﬁom imbalance of forces on the solution
when the solution is mixed or stirred (Bard and Faulkner 2001). Assuming a linear mass
transfer, and linear electric field, the mass ﬂux to the electrode can be expressed by the

Nernst-Planck Equation (Equation 6):

6C (x) zF

AE
J(x )=—D——— ax —EDCT+CV()C) (6)

where J is the mass ﬂux, D is the diffusion coefficient of the reacting specie in solution,
C is the instantaneous concentration of the reactants in solution at time t, x is the distance
in the x direction or along the direction of the electrical current, 2 is the net charge on the
reacting species, F is Faraday’s constant, R is molar gas constant, T is temperature (K),
AE/l is the gradient arising from the change in potential AE over a distance I, and v is the
linear velocity of the solution. Because the test cell solutions were not stirred, the third
term in Equation 6 can be neglected. Hence, Equation 6 becomes:

J(x)=—D———— ———DC

8C (x) 2F 3%
6x RT 1

(7)

127

For uncharged reactant, the second term in Equation 7 can be neglected to yield Fick’s

ﬁrst law as follows:

J(x) = —D§g(x—)
6x

(8)
Equation 7 was applied to model the rate of mass transfer of naphthalene to the electrodes
in the DC setup. The following assumptions were made when applying Equation 7 (or
Equation 8) to simulate the naphthalene concentrations:

1. Reactants (naphthalene) are instantaneously converted to products when they
come into contact with the electrode. Hence, C(t) at the surface of electrode is
equal to zero for t 2 0. This establishes a concentration gradient that drives
reactants by diffusion to the surface of the electrodes and this condition is
considered as the maximum concentration gradient that may be established in a
given time step;

2. Steady-state diffusion conditions exist in a given time step which was assumed
equal to 1 hr;

3. Linear mass transfer exists in the given time step of 1 hr; and

4. Linear electric field exists in the given time step of 1 hr.

Equation 7 was applied with C of naphthalene equal to CO (~ 20 mg/L) for t = 0 at the

cathode and C = O for t 2 0 at the anode. Assuming a time step equal to 1 hr, the C of
naphthalene at cathode was estimated by subtracting the mass of naphthalene that moved
to the anode for degradation during the previous time step. This computation was carried

out for the test duration of 48 hours and the modeled normalized concentrations were

compared (Figure 3-9) to the test results for DC tests carried out at j = 3 and 6 mA/cmz.

128

 

 

 

 

 

 

 

 

 

 

l l i I I
o 1 '- 5 """"""""""""""""" ""
2 P .’ ' ' ' , u
o. - 3‘. Model: Diffusion Only 4
C .- ‘= II
o : ----- . e o o 0
'ﬁ 0.8 ‘1‘- ‘‘‘‘‘‘‘‘‘ Model: DIfoSIOn + Migration (j = 3 mA/cmz) "1
5 b :i‘ 1
c : ' ......... . . . . .
3 * g 3'. Model: DifoSion + Migration (J = 6 mA/cmz) i
C F E I -i
8 0-6 '" ...i """" . Test Data: 3 mA/cm2 "‘
2 - ~ 5'. .
:4; ' i_ '. - Test Data: 6 mA/cm2 '
.5 * : g | .
E 0.4 ---S----'g---| --------------------------------------------------------------------------- -
g h E i"! '1
z _ , '. I. .
3 . . I . . . . , .
.5 : "o. “ : : : :
‘6 0.2 'J """" 0. """" ‘ """"""""" """""""" "
E - ' "-. 3a? : : : -
3 “.- ‘ts. l
0 l w... ‘--._t . l l ._
0 10 20 3O 40 50

Elapsed Time [hrs]

Figure 3-9: Simulated concentrations versus test data for naphthalene tests using DC

Schwarzenbach er al. (2003) reported diffusion coefﬁcients of organic solutes equal

-5 2 -1 -6 2 -1
to about 3.0 X 10 cm s for small molecules to 5.0 X 10 cm s for those of molar
mass near 300 g/mol. Based on this data, the diffusion coefﬁcient of naphthalene was
estimated as 9.0 X 10'6 cmzs'l from the diffusion coefﬁcients vs. molar mass chart

presented by Schwarzenbach et al. (2003). Equation 8 was applied to simulate only
diffusion driven mass ﬂux of naphthalene to the anode for oxidation. The simulated

normalized concentration is presented in Figure 3-9. The simulated degradation rate as a

129

result of naphthalene migration to anode only due to diffusion is orders of magnitude less

. . . . ' 2
than that observed during the experiments run at DC densrties equal to 3 mA/cm or 6

mA/cm2 with NaZSO4 as the supporting electrolyte. When both diffusion and migration

are used for simulating the mass ﬂux of naphthalene to the anode by using Equation 7,
the simulated normalized concentrations are presented in Figure 3-9 when the net charge,
2, was assumed equal to -0.35 for naphthalene.

For neutral molecules such as naphthalene, 2 should be equal to zero. However, it is
hypothesized that the presence of anions in the test solution (from the supporting
electrolyte) imparts an effective negative charge on the naphthalene molecules. These
anions dragged the naphthalene molecules along as they migrated to the positively
charged electrode (anode), where anodic oxidation resulted in the degradation of the
naphthalene molecules. Therefore, the naphthalene molecules moved in the solution as a
result of migration simulated by the second term in Equation 7. Hence, the rate of
migration of naphthalene molecules towards the anode is enhanced by assigning a
negative value to z for naphthalene.

By assigning the naphthalene molecule an effective partial negative charge of -

0.35, the simulated normalized concentrations versus time data ﬁts well with that for the

experimental data when Na2$O4 was used as the supporting electrolyte for DC densities

equal to 3 mA/cm2 and 6 mA/cmz. The rate of anodic oxidation of naphthalene (or

salicylic acid) depends not only on the rate of migration of reacting molecules to the
reaction site (the anode or instantaneous anode) but also on the rate of production of OH

radicals at the anode (or instantaneous anode). Hence, as the current density is decreased,

130

it may decrease the rate of production of OH radicals at the anode and may become the
rate limiting step. When Equation 7 is applied (Figure 3-9), the rate limiting step assumed
is the rate of mass transfer to the anode (by diffusion and migration). Hence, Equation 7
would overestimate the decrease in the concentration (or underestimate the concentration
at a given time, t) of the reactant when current density is low enough to not degrade all
naphthalene mass diffusing and migrating to the anode (or instantaneous anode). At
higher current densities, the predictions from Equation 7 would be more accurate. It is
beyond the scope of this study to estimate the rate of hydroxyl radical production as a
function of current density and other variables. Hence, it is not possible to calculate
analytically at what current densities the assumptions are valid when Equation 7 is
apphed.

While Equation 7 is for DC and cannot be applied for an AC, it can be used to
explain the reduced rate of degradation when AC was applied and as the frequency of the
AC was increased from 0.1 to 1,000 Hz (Figure 3-7). Proposed reasons for the decrease
in degradation rates with increased AC frequency are as follows:

1. By assuming that the naphthalene molecules had an effective partial negative
charge of say 2 = —0.35 under our experimental conditions based on the acceptable
ﬁt for DC data, increasing the frequency of the AC resulted in back-and-forth
movement of the reacting molecules, and hence the net mass flux of the molecules
toward the electrodes was reduced. If the net distance traveled by the naphthalene
molecules to get to an instantaneous anode, L, in Equation 7, is varied, until the
modeled values of concentrations ﬁt with the experimental data for f equal to 0.1

Hz, 1H2, 10 Hz, 100 Hz, and 1,000 Hz the values of L obtained are 14 cm, 17

131

cm, 25 cm, 27 cm, and 45 cm, respectively, compared to the electrode spacing of
8 cm. Assuming degradation rate at the electrodes are the same for each
frequency, it can be hypothesized that increase in the AC frequency results in
taking longer duration for a naphthalene molecule(s) to reach the instantaneous
anode. This may be due to the back and forth movement of the molecule due to

the reversal in the direction of the current proportional to the frequency of the AC.

. Continuous bubbling of gases at both electrodes (probably 02 and H) in the test

solution was observed as the tests progressed. However, the rate of bubbling of
the gas decreased as the frequency of the applied AC was increased until no gas
bubbles were visible at an AC frequency of 100 Hz or more. Simond et al. (1997)
and Comninellis (1994) reported that the hydroxyl radicals, formed on an anode
react with each other to form oxygen gas at the anode to complete the electrolysis
of water molecules. Stock and Bunce (2002) reported that hydrogen radicals
formed on a cathode react with each other to produce hydrogen gas at the cathode.
If the formation of OH radicals and hydrogen radicals are for the generation of
oxygen gas and hydrogen gas at the anode and cathode, respectively, then it is
likely that the rate of gas bubbling at the electrode gives an indication of the rate
of formation of these radicals at the electrodes. The higher the rate of the
observed bubbling of the gas(es), the greater the rate of hydroxyl or hydrogen
radical formation at the electrodes. Hence, based on the visual observation of gas
bubbles, it is likely that increase in the AC frequency resulted in decrease in the
rate of formation of hydroxyl radicals at the electrodes. This impacted the rate of

oxidization of naphthalene at the instantaneous anodes.

132

3. Competing anodic reactions (oxidation) and cathodic reactions (reduction) at the
same reaction sites (on the surface of the electrodes) may have been responsible
for the decrease in the oxidation rate of both naphthalene (Figure 3-5) and
salicylic acid (Figure 3-7) as the AC frequency was increased. However, in DC
electrolysis, oxidation reactions occurred at the anode only and reduction
reactions occurred at the cathode only and hence there was no competing anodic
oxidation and cathodic reduction at the same reaction site. Therefore DC was

more effective in degrading naphthalene or salicylic acid in the test cells.

Effect of Convection Mass Flux
According to Equation 6, stirring the test solutions increases the mass ﬂux towards the
electrodes by introducing a convection mass ﬂux component. Additional DC test and AC

tests at frequencies of 1 Hz, 10 Hz and 100 Hz were conducted as shown in Figure 3-10.

133

Co I: 20 mgIL

Supporting Electrolyte: NaZSO4

AC: Square Wave, ] = 3 mAIcm2

 

 

  
  
  

DC / Control (stirred, unstlrred)

" .-‘ I I 'I. l I r I r I I i f ij I—ff! I I I I I fT I
o 1 31.4.2; ‘§M='“-:-=‘ 5: ~————-—-——3 f-1000 Hz(unstirred)
Q ' ' , ' _
a. 3 : 3 3 Ir - 100 Hz (unstlrred)

3 E I 3 E ".
.2? 0.8 .......... '. .......... _;.f=100Hz(s'tirred)
.3 g P g 3 3 >2 gf-io Hz(ustirred)
: . . :
0 06 , .. ...; .............. ,....... ..', ............
g ' Z 3 F :f = 1 Hz (unstlrred)
'8 ' ' :
<
,2 f : : f
E, 0.4 -------- ------------- g ------------- :- ---------- 7 -;-i=o.1 Hz (unstirred)
% ' f
(I) : . I
'8 . . ‘ ; gf= 10 Hz(stlrred)
.! 0.2 ...... .. .... ' ............. .t ............
g - a . 2 a
o b i 5
z . i \ =i= 1 Hz (stirred)
o M n I 1 1 _‘

 

 

 

o 5 10 15 2° 25 3°
f = 0 (DC) (unstlrred)
Elapsed Time [hrs] f = 0 (cc) (stirred)

Figure 3-10: Effect of stirring on degradation rate of salicylic acid at different AC
ﬁ'equencies

In the AC tests conducted at 1 Hz and 10 Hz frequencies, stirring signiﬁcantly increased

degradation rate. At 100 Hz frequency, stirring did not result in any signiﬁcant increase

in degradation rate. For degradation to occur, the reacting species have to move to the

electrodes, and at the electrodes, OH radicals have to be produced to degrade the reacting

species in the vicinity or on the surface of both electrodes. Back-and-forth movement of

the salicylic acid molecules in solution subjected to AC slowed down the net ﬂux towards

134

the electrodes. Stirring the solution increased the mass ﬂux towards the electrodes. At 1
Hz and 10 Hz AC frequencies, enough OH radicals were generated at the electrodes and
hence degradation rates were enhanced. However, at 100 Hz frequency, although stirring
increased the mass ﬂux towards the electrodes, less OH radicals were produced such that
the rate of production of OH radicals was the rate limiting step. Stirring the DC test
solution did not signiﬁcantly increase the degradation rate as compared to the test cell in
which the solution was not stirred. In the DC tests, OH radicals were produced at only
one of the electrodes (the anode) whiles the cathode acted as a “passive” electrode in the
oxidation of salicylic acid. Because only one electrode participated in the degradation, the
degradation rate in the DC test was not as high as in the AC tests at 1 Hz and 10 Hz

frequencies.

Intermediate Byproducts

LC/MS analyses indicated molecular masses that suggest the formation of degradation
products such as naphthol, dihydroxynaphthalene, trihydroxynaphthalene,
dihydrotrihydroxynaphthalene, naphthoquinone, hydroxynaphthoquinone, and
benzenedicarboxylic acid. The exact isomers formed were not determined. These
degradation byproducts formed are relatively more soluble than the parent compound
(naphthalene), hence it is expected that they would be more bio-available in solution for
subsequent degradation by microorganisms. The change in color of the test solutions to
light yellow as the tests progressed may have been due to the formation of
naphthoquinone in the solutions. Formation of quinones is a common observation of the
hydroxylation process in oxidizing media (Oturan 2000). The formation of

naphthoquinone is consistent with those found by other researchers who also observed

135

aromatic quinones during electrolytic oxidation of aromatic hydrocarbons (Brillas et al.
2000; Panizza et al. 2000; Saracco et al. 2000). GC/MS analysis by Goel et al.(2003)
indicated the presence of at least one reaction product, 1,4-naphthoquinone, when
naphthalene solutions were electrolyzed and concluded that naphthalene degradation
could be due to direct anodic oxidation, oxidation by an intermediate (e.g. hydroxyl

radicals), or a combination of both.

Reaction Mechanism and Pathway

OH. react with organic pollutants in two different ways:

1. by electrophilic addition to a double bond or an aromatic system; and
2. by abstraction of a hydrogen atom from a carbon atom (Schwarzenbach et al.
2003)
Based on the identified degradation products, a simpliﬁed degradation pathway for
the electrochemical degradation of naphthalene in aqueous solution is proposed as

illustrated in Figure 3-1 1.

136

20:28 30033 E 02035220: .20 @353 202208200 32802005020 02.2385 ”2% 052E

Eon
0222323302352

..wd
1000.}63

2.0.2..22023052559625: o=o_aa.=na=>uoucb=._h

30 IO

2 ....20 mm ..wo

,. Em
IO $0

22020

0
. ..wd 0:022:3522323232
0 20
.:O
0
.. ill. .. , ..wd 020250052252
.20m
ON
2020333.. H
cowohex: ”.2
one—aauaaaagegﬁn ...—.3222 one—«ﬁanaz
mo
1 l ..Wo . ll - ..w . i 1-.
.20 . .20 . _ a .20
IO :0

137

The exact isomers formed were not determined. Hence the structures shown are just one

example of the possible isomers of each byproduct. Initial hydroxylation of naphthalene

by an OH. radical resulted in the formation of naphthol. Further hydroxylation resulted in

the formation of dihydroxynaphthalene (naphthalenediol), which was further oxidized by
hydroxylation to trihydroxynaphthalene or by hydrogen abstraction to naphthoquinone. It
is possible that addition of hydrogen atoms to naphthoquinone may have reduced it back
to dihydroxynaphthalene, which is similar to a reaction step reported in a previous study
(Chin and Cheng 1985). Chin and Cheng (1985) reported that in an undivided cell,
benzoquinone produced at the anode was subsequently reduced to hydroquinone at the
cathode. Therefore, it is possible that in our setup, the naphthoquinone, produced by the
oxidation of naphthalene on an instantaneous anode, was subsequently reduced to

dihydroxynaphthalene on an instantaneous cathode as shown in Equation 9.
+ .—
Addition of hydrogen atoms to the trihydroxynaphthalene resulted in the formation of

dihydrotrihydroxynaphthalene. Further hydroxylation of the naphthoquinone formed

resulted in the formation of hydroxynaphthoquinone. With sufficient OH. radicals in the

reaction medium, ring cleavage in naphthoquinone may have occurred leading to the
formation of benzenedicarboxylic acid. Li et al. (2005) reported that benzoquinone
formed from the oxidation of phenol could be degraded with ring breakage to various
carboxylic acids. One of the proposed mechanisms involves the adsorption of

benzoquinone onto the anode surface, which gives up an electron, and the carbon that is

138

double-bonded with oxygen is attacked by a neighboring hydroxyl radical. When this
process is repeated at the para position, the ring could be opened and benzoquinone
would be broken down into small organic molecules such as carboxylic acids (Houk et al.
1998; Lund and Baizer 1991). The present experimental ﬁnding is consistent with the
proposed mechanism because both naphthoquinone and benzendicarboxylic acid (from
probable ring cleavage of naphthoquinone) were identiﬁed as intermediate byproducts.

The possible mineralization of naphthalene was not investigated in this study.

Divided Cell Setup

In order to prove that anodic oxidation occurred on both electrodes in AC electrolysis,
unlike DC electrolysis in which anodic oxidation occurred on only one electrode (the
anode), further tests were conducted in a divided cell setup. The anode compartment was
separated from the cathode compartment by means of a Naﬁon membrane (N-117) at the
interface of two 30 mm diameter glass tubes connecting one-liter glass beakers. In studies
(Alshawabkeh and Sarahney 2005; Marselli et al. 2003) Naﬁon membrane has been used
as a proton-permeable membrane at a cell junction that physically separates the
electrolytes in the two compartments.

Figure 3-12 shows the change in concentration of naphthalene with time for both

AC (3 mA/cmz, 0.1 Hz) and DC (3 mA/cmz) divided cell setup as well as a control

divided cell setup.

I39

Co = 20 mgIL_

 

l I T I I ! I I I I I I I

 

q

: l2 : ' : :
'17 \ : : : : _: i
.2 --s 3 = a 7 2 .
0.6 ...:-.-- """ """"""" j

1 ' :ﬁi‘ .............. ...... Supporting Electrolyte: N82304-

I "I ‘1- ' : I
”I" I i 2

- . “Fl ; 3 AC: Square Wave, r = 0.1 i-iz, ] a 3 mAlcm i

' Va. : . s s s t

0.8 --s~ ll . ------- g ------------ ‘= -------- g -------------- §~--------gIcenaoi-~-+

 

 

 

  

Normalized Naphthalene Concentration, CICo

 

 

 

_ 5 DC Catholyte
L . : ‘ : : 5
04 H. ............. .............. 5 ............
. 22 i I ' AC:
. . g g g . a Compartment1
02 -.3 ........ .-.71 .............. ........... 3 .............. 5 ........ “I I
' is 2 s s s " 9°mP‘“‘“°“‘2
. i r 3 3
o l I 1 1 1 l l a I LI I n n 1 I 1 I 1 I I I n I IDCIAPonlytAej
0 1O 20 30 4O 50 60

Elapsed Time [hrs]

Figure 3-12: Change in concentration of naphthalene with time in a divided cell setup
using AC and DC.

Both compartments in the AC test exhibited approximately the same reduction in
concentration over time but the anolyte in the DC divided cell setup depicted much faster
rate of decrease in the concentration than the catholyte. LC/MS analysis of samples
electrolyzed with 3 mA/cm2 square wave AC at 0.1 Hz frequency indicated the presence
of hydroxylated byproducts including dihydroxynaphthalene, naphthoquinone, in both
compartments. However, no hydroxylated byproducts were detected in the cathode
compartment (catholyte) of the divided cell DC setup. Hydroxylated byproducts were

detected in the anode compartment (anolyte). The reduction in concentration of

140

naphthalene in the catholyte may have been due to potential sorption on the reactor
components and sparging due to continuous bubbling of hydrogen gas from the reduction
of water molecules. Goel et al. (2003) showed some loss of naphthalene in a control cell
experiment over a 9 hr period when nitrogen gas was bubbled through a naphthalene
solution at a rate of 1.6 L/day corresponding to a passed current of 100 mA. Also, using
salicylic acid as the test solution resulted in the formation of 2,3-dihydroxybenzoic acid
and 2,5-dihydroxybenzoic acid in both compartments of the AC divided cell setup,
accompanied by reduction in concentration in both compartments. However, these
byproducts were only detected in the anolyte in the DC divided cell setup but not in the
catholyte, in which approximately 10% loss in concentration was attributed to sparging
by H2 formed from the reduction of water molecules in the catholyte. Since salicylic acid
is relatively more soluble in water as compared to naphthalene and hence a lower Henry’s
constant, loss of mass in the catholyte of the salicylic acid test solution was signiﬁcantly
less than that of the naphthalene test solution. Hence it is likely that a greater proportion
of the reduction in concentration of naphthalene in the catholyte was mainly due to
sparging. Possible reduction of naphthalene in the catholyte could have also contributed
to the reduction in concentration.

Figure 3-13 shows the concentration change of salicylic acid in solution that

depicted the same trend as the naphthalene test solutions.

141

co . 20 mgIL

Supporting Electrolyte: NaZSO‘

AC: Square Wave, f- 0.1 Hz, 1 = 3 i'riAIcm2

 

 

 

I I I f I I I I fir r I I I I I I I I ! I I I ‘—
_L_. E} E S . Control Cellw-Ei‘
"‘ I ' ﬁZ—e—oc Catholyte—_{Z] ‘
33 7 : : -
0.8 _ ........ ........ . ....... ................. ..... ......... ............... -
r — : : : - -
b == 1

 

Normalized Salicylic Acid Concentration, ClCo

 

 

 

 

0 10 20 30 40 50
Elapsed Time [hrs]

Figure 3-13: Change in concentration of salicylic acid with time in a divided cell setup
using AC and DC.

Thus, it was concluded that anodic oxidation, mediated by OH. attack occurred on both

electrodes in AC electrolysis but on only the anode in DC electrolysis. The experimental

results are summarized in Table 3-2.

Measured Initial and Final Key Test Parameters

Measured initial and ﬁnal values of key experimental parameters are shown in Table 3-3.

142

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

02...:me
0.0 Qm 2.0V 0:a wé 0.0 0.: m6. m N: 000; 00002003 0000202002
0.8029
0.0 w.m 20v 0; 0.0 0.0 0: 2.02 m N: 00— 0003023 000—05232
0.80di
m0 WM 2.0 mam 0.0 0.0 0.3 Na— m N: A: 00030:: 0002052202
300003
m.m N0 000 «.3. 0.0 0.0 m.w_ N02 m NI _ 0003003 002052202
3.8029
0.0 0.0 0? 00m 0.0 0.0 Qw— ~.0_ m NE _.0 0005003 022052002
_.m 5m E.— 20 0.: 00 5mm N02 m OD 0005003 002902202
$32500
m0 m.m 0N0 :v 0.0 0.0 w.0_ 0.3 m N: 00 00002003 0202052002
4250
0.0 N0 002V 0mm 50 w.0 2.: 0.2 m NE 00 0005003 0002052002
Aogomv
0.2. 1m 000 o; 0.v 0.0 Qw— m.w_ m N: 00 00030.5 002052202
.05.”— ?E: 022.2 .002: .052 .002. .052 .00.:—
SEE
00005000000 _>E_ _~E0\<E_
0&sz 30:000.— .Uo. 5500:
0020005 20002 0.50005 :0 0030000802. D< 0.00m U< :00 0:009:00

 

 

 

 

 

 

 

 

0000080002 30,—. 20v— EEm 0:0 REE 00000002 .00 waﬁsm ”Wm 030,—.

143

 

28

 

 

 

 

 

 

 

 

 

 

 

5w 0.w 0mm :30 0.0 0.0 0.0: 0.0m £00500 0 - 00030:: 030:8
060
0.0 9w :mm 000 0:: 0.0 5mm N.:~ :05: m 0 00030:: 0:30:0m
0:00:09 0:00
Nam 0.x 5: a? 0d 0.0 mm: o.:~ N: 000.: 00030:: 0:30:0m
3004355 0:00
:0 0.0 050 03. «:0 0.0 2.: mdm m N: c: 00030:: 030:8
@05me 0:00
2. w.0 N00 00: m: 0.0 :0: mdm m N: :0 00030:: 030:3
:26 m. 0
3:00:09 3:83-00:
0.0 :0 mm: Ev :0 :20 0.0: Wow m NE :0 00030:: 0:205:002
:25 2.0:
30:08 3:05:00:
0.0 0.m 3‘0 000 :m.v m0 Nd: 0.0m m N: :0 00030:: 0:205:002
@5209
E 0.: m m? mi 0.0 20 EN 69 m - 0022:
0 :o:< 0:205:32
m.m Em N00 30 0.: 0.0 :20». N.:N :09 m - 00035
3:30 :0 =00:
Nw 0.m m0: mmv 0:: m0 0.:N :0: m N: :0 00035
as: m: 2 :00: 2225002
0.0 5m mm: mmv EN m0 0.:N :.0: m N: :0 00035
0.0 0.0 mg 00m m0 0.0 0.0: No: 22:80: 0 - 00030:: 05:05:00.7:

 

 

 

 

 

 

 

 

 

 

 

 

03:00 m-m 050:.

 

 

144

.FEo: 00000020 :0 0000 0800000 00000 \ 7:: 0:00.50 n m

N

5025: 0:800 0:056 .N
.65 8 :003 0:00.50 <8 00: $7.00: :3 008808: 0:03 0000: =< .:

 

 

 

 

 

 

“meaoz
3:000

ms 9w :00 3.0 0.0 0.0 9mm 5. : N 20.5500: 0 . 000015 0:00 0:30:0m
2 :00:
0.0 0.0 :8 ms. 3. 3. 0.00 EN 0208000 - 8035
0505000

2 3 a- 0? m0: 3. 30 08 Gem o 0023: 22.2.0.3
05093
:2 0.0 $0 :5 n: 3 0.00 0.00 Gem 0 8220
@008 :0 :00:

0.0 0.0 E E. 0.2 3 0.3 EN m 0:3 822: 232.5%
€00.me 2 :00:
2 0.0 an E mm 0.0 0.00 SN m 0:3 823:

 

 

 

 

 

 

 

 

 

 

 

 

380 3. 2%:

 

145

The temperature of the test solutions ranged between approximately 17 OC and 22 OC.

The heat that might have been generated in the test solutions as a result of Joule heating
was absorbed by the water bath so no drastic increase in temperature was observed.

The measured initial pH of the naphthalene AC test solutions generally ranged
between 6.5 and 6.8 while the ﬁnal pH values were between 4.1 and 5.0, indicating that
the AC test solutions became acidic at the end of the 48-hr testing period. In a divided
cell set-up where electrolytes are separated to prevent mixing, but the ﬂow of charge is
permitted, pH and redox conditions may develop (Alshawabkeh and Sarahney 2005). At
the anode, protons and oxygen gas are produced (oxidizing, acidic conditions develop),
and at the cathode, hydroxyl anions and hydrogen gas are produced (reducing, alkaline

conditions develop) as presented in Equations 12 and 13.

2H20 —> 02 + 4H+ + 48. ........ E0 = +1.229V(An0de)(10)

4H,0 + 46' —> 2H, + 40H‘...E° = —O.827V(Cath0de)(11)

Other electrolysis products may result depending on the availability of ions and
their electrochemical redox potential. Although some secondary reactions might be
favored at the cathode because of their lower electrochemical potential, water reduction
is dominant (Alshawabkeh and Sarahney 2005). Therefore, it is likely that the oxidation
of water molecules was dominant at the time of measurement, resulting in the generation
of more H+ ions in the solutions and hence the test solutions became more acidic at the
end of the testing period. The type of supporting electrolyte, degradation byproducts

formed, among other factors may also inﬂuence the complex acid/base chemistry of the

146

reaction media. The formation of byproducts such us benzenedicarboxylic acid, with
acidic properties could have also made the ﬁnal pH of the test solution acidic. The initial
and ﬁnal pH values of the salicylic acid AC test solutions however, remained
approximately unchanged, between 4.2 and 4.5. The DC test cells used in electrolyzing
the naphthalene solutions recorded initial pH of 6.5 and ﬁnal pH of 11.5 and that of the
salicylic acid solution were respectively 4.4 and 11.4; indicating that the DC test
solutions became alkaline at the end of the 48-hr testing period.

The measured ﬁnal standard redox potential of the test solutions was generally
higher than the corresponding initial values. The measured dissolved oxygen
concentration was also depicted a similar trend, indicating that the AC test solutions
became more oxidizing at the end of the test period. However the DC test cell solutions
became more reducing at the end of the test period. Continuous bubbling of gases was

observed on both electrodes in test cells throughout the testing period when solutions

were electrolyzed at 3 mA/cm2 current density (at 0.1 Hz and 1 Hz AC frequencies). The

rate of evolution of gases at the electrodes was observed to decrease with increasing AC

frequency. At higher frequency (over 100 Hz) no gas bubbles were observed with

unaided eye. The observed gas bubbles could have been due to the evolution of 02(g)

and/or Hz(g). During the AC half cycle, when one electrode is at a positive electrical

potential (instantaneous anode), oxygen gas is produced at that electrode (Equation 12)

and hydrogen gas is produced at the other electrode (instantaneous cathode) according to

Equation 13. Therefore it is likely that the generation of 02(g) in the test solutions was

more dominant at the end of the test period.

147

The measured ﬁnal electrical conductivity of the test solutions remained

approximately the same as the initial values, ranging between 810 ,uS and 860 p8.

SUMMARY AND CONCLUSIONS

The key objectives of this study were to: (1) use alternating current (AC) to investigate
the effect of AC waveform and frequency on the rate of degradation of naphthalene in
aqueous solution; (2) explore the mechanism of degradation of naphthalene when
subjected to AC and DC; and (3) prove that anodic oxidation of naphthalene occurred on
both electrodes (instantaneous anodes) when AC electrolysis was used, unlike direct
current (DC) electrolysis in which anodic oxidation of naphthalene occurred on only one
electrode (the anode).

Three AC wave forms (square, sine and triangular) supplied at a frequency of 60
Hz were used to investigate the efﬁciency of the AC wave forms in degrading

naphthalene in aqueous solution. Square wave AC having with peak current density of 3

mA/cm2 and frequency varying between 0.1 Hz and 1,000 Hz was used to investigate the

effect of AC frequency on degradation rates of naphthalene. A divided cell setup was
used to isolate the degradation byproducts formed at both electrodes to identify the key
degradation mechanisms and differences in AC and DC electrolyses. The Nenrst-Planck
equation was used to model the rate of mass transfer to the electrodes, where reactions
were expected to occur, and the simulated concentration-time proﬁles were compared
with test results. The key conclusions of this study are as follows:

1. For a ﬁxed peak current density and frequency of applied AC, the square wave

AC was most efﬁcient for the degradation of naphthalene in solution. The next

148

most efﬁcient wave form was sine wave followed by the triangular wave form.
The key reason for this, for a given peak current, the total electrical power
(product or rms current to rms voltage) applied by a square wave AC is greater
than sinusoidal wave form followed by a triangular wave form.

. For a ﬁxed peak current density, increase in AC frequency resulted in a decrease
in the rate of degradation of naphthalene. The key reason for this decrease is the
rate of migration of naphthalene to the electrodes was reduced due to the rate of
change in the direction of the current which is a ﬁinction of the frequency of the
wave form. In addition, increase in the AC frequency may have suppressed the

rate of production of hydroxyl radicals.

. The hydroxylation of the OH. probe (salicylic acid) to form 2,3-dihydroxybenzoic

acid and 2,5-dihydroxybenzoic acid, and the suppression of the degradation of
naphthalene in the test cells in the presence of an OH' scavenger, t-butanol, is
consistent with numerous studies that OH radicals were the primary oxidizing
agent in the electrochemical cells for AC as well as DC tests.

. The reduction in concentration of naphthalene and salicylic acid in both
compartments of a divided cell setup as well as the formation of hydroxylated by-
products in both compartments is an indication that anodic oxidation occurred on
both electrodes in AC electrolysis. In DC electrolysis, hydroxylated by-products
were detected in only the anolyte. In DC application, there was rapid reduction in
concentration in the anolyte (over 90% degradation in 24 hrs of testing) where

naphthalene and salicylic acid were used as the reactants. The rate of reduction in

149

concentration in the catholyte of the salicylic acid and naphthalene tests, during
the 48 hr testing period was relatively small.

The rate of anodic oxidation of naphthalene and salicylic depends on the rate of
production of OH radicals at the anode (or instantaneous anode) as oxidizing
agents as well as the rate of migration of reacting molecules to the reaction site
(the anode or instantaneous anode). At higher current densities, not only more
OH radicals are produced at the anode but also the rate of migration of the
reacting molecules to the anode (or instantaneous anode) increases during a DC
application. However, when AC is used, the change in the direction of the current
decreases the rate of migration of the reactant to the instantaneous anode. Hence,
as the frequency of the applied AC increases, the rate of degradation decreases.
Both diffusion and migration processes were responsible for the rate of mass
transfer of naphthalene to the reaction sites (surface of electrodes or in the
vicinity of the electrodes) where naphthalene was degraded or converted by the
hydroxyl radicals. However, migration was the dominant mass transfer
phenomenon responsible for the degradation of naphthalene in the AC and DC

electrochemical cells.

150

PAPER NO. 4: A COMPARATIVE STUDY OF SONOCHEMICAL
AND ELECTROCHEMICAL DEGRADATION OF NAPHTHALENE
IN AQUEOUS SOLUTION

ABSTRACT

The objective of this study was to compare the rate of degradation of naphthalene, a
model polycyclic aromatic hydrocarbon (PAH) in aqueous solution was subjected to
ultrasound energy and electrical energy (AC and DC) delivered at the same out put
powers (30 W and 60 W). Under the experimental conditions, DC and ultrasound
exhibited approximately equal degradation rates at 60 W output power. However, at 30
W, rate of degradation for DC was slightly higher. AC was least efﬁcient at both output
powers. As the output power was increased from 30 W to 60 W for AC, DC as well as
ultrasound, it resulted in faster degradation of naphthalene. The energy applied in each
test exhibited pseudo-ﬁrst-order degradation kinetics for the decrease in the concentration
of naphthalene in solution. The degradation byproducts identiﬁed in both electrical (AC
and DC) and sonochemical degradation were identical, which suggested that it is highly
likely that the primary oxidizing agent was the same. Formation of 2,3-dihydroxybenzoic
acid and 2,5-dihydroxybenzoic acid from both sonochemical and electrical degradation of
salicylic acid (a hydroxyl radical probe) was consistent with the ﬁndings in literature
which indicated that hydroxyl radicals were the primary oxidizing agent. The Nenrst-
Planck equation was used to model the rate of mass transfer to the electrodes in the AC
and DC tests, where reactions were expected to occur, and the simulated concentration-
time proﬁles were compared with test results. Both diffusion and migration processes

were responsible for the rate of mass transfer of naphthalene to the reaction sites (surface

151

of electrodes or in the vicinity of the electrodes) where naphthalene was degraded or
converted by the hydroxyl radicals. However, migration was the dominant mass transfer
phenomenon responsible for the degradation of naphthalene in the AC and DC

electrochemical cell.

INTRODUCTION

PAHS constitute an extraordinarily large and diverse class of organic molecules. The
major sources of PAHS are crude oil, coal and oil shale. Coal tar and petroleum residues
produced in the reﬁning process contain high percentages of PAHS, representing a wide
range of molecular sizes and structural types (Harvey 1997). According to Lee et al.
(1981), PAHS are the largest class of chemical carcinogens, and both Clar (1964) and
Harvey (1991) also have presented reported in detail about the evidence of PAH
carcinogenicity in animals. When ingested, PAHS are rapidly absorbed into the
gastrointestinal tract due to their high lipid solubility (Cemiglia 1984). The United States
Environmental Protection Agency (USEPA) estimates that 3,000 to 5,000 former
manufactured gas plant (MGP) sites are located across the United States and many of
these contaminated sites contain substantial quantities of PAHS (USEPA 2000).

Adewuyi (2001) reported in a review that sonolysis is capable of degrading several
target compounds including, amongst others, phenol, chlorophenols, nitrophenols,
polychlorinated biphenyls, chloroaromatics, pesticides, dyes, CFCs, PAHS, surfactants,
present in relatively dilute solutions typically in micro- to milli-mollar range. Frequencies
above 16 kHz, above the threshold of human hearing, are classed as ultrasound. In

ultrasound applications, the utilized frequencies usually begin at 20 kHz and extend to

152

500 MHz (Thompson and Doraiswamy 1999). The lower frequency range is utilized for
many studies into sonoluminescence, synthesis, and degradation. This range is also
utilized in many ultrasound baths which tend to operate between 30 and 40 kHz (Little et
al. 2002). Ultrasonic waves pass through any substance having elastic properties.
Molecules oscillate in the direction of the wave in a compression cycle followed by a
rarefaction cycle. In ﬂuids, ultrasound causes high-energy acoustic cavitation, that is, the
formation of microscopic vapor bubbles in the low pressure (rareﬁed) part of the
ultrasonic wave when the low pressure overcome the surface tension of the ﬂuid. The
bubbles collapse or implode in the compression part of the wave creating very minute
high-energy movements of the solvent that result in localized high shear forces (Meegoda
and Veerawat 2002). The implosive collapse of the bubbles in the liquid can cause
localized pressures of the order of 1,000 atm and temperature as high as 5,000 K
generated over a few microseconds time interval. Their lifespan of these bubbles is of
approximately 2 us and size is approximately 200 um make direct measurements difﬁcult
(Suslick and Flint 1987). Sonolysis of environmental pollutants constitute a remediation
method of increasing promise (Goskonda et al. 2002). Chemical changes attributable to
the effect of ultrasound are largely theorized as being a result of reactions within the
resultant cavitation bubbles, observed within the sound wave (Suslick and Flint 1987).
Much of degradation chemistry is accepted as depending on the formation of free radicals
within these bubbles (Little et al. 2002).

On the other hand, the electrochemical degradation of organics such as PAHS,
phenol, benzene etc. has been investigated by various researchers using DC electrolysis.

However, relatively few studies have been conducted where AC electrolysis was

153

explored for the degradation of organics in solutions. Comninellis (1994), Brillas et al.

(2000), and Panizza et al. (2000) have reported that electrochemical systems generate

hydroxyl radicals (OHO) at the anode due to the oxidation of water molecules. Among
oxygen radicals such as 02", OH., H020, and R00. that may be generated

electrochemically in solution, the OH. is the most reactive (Oturan and Pinson 1995).

Nakamura et al. (2005) proposed a mechanism based on selective redox reactions with
various radicals generated by AC electrolysis that allows both oxidation and reduction at
the same reaction site between adjacent electrodes. Based on the analysis of intermediate
products of the degradation process, Nakamura et al. (2005) reported that the mechanism

involved selective redox reactions with short-lived and highly reactive radicals: hydrogen

atoms (H0) and hydroxyl radicals (OH) that are electrochemically generated on the

electrodes. Therefore both sonochemical and electrochemical degradation of organics in
solution involve the generation of free radicals; the OH' being the most reactive of the
radicals. However, these two methods have not been compared to ﬁnd out which method
is more efﬁcient for degrading the target contaminant.

Therefore, the objective of this study is to compare the rate of degradation of a
model PAH — naphthalene, in aqueous solution using ultrasound energy and electrical
energy (AC and DC) delivered at the same output power. The ultrasound equipment used
for this study was capable of delivering a maximum ultrasound power at 60 W to our test
solutions. Hence tests were conducted at 60 W and half of the maximum output power,
i.e. 30 W, respectively. Naphthalene, which contains two fused benzene rings, was

selected as a model PAH compound because in published studies Alshawabkeh and

154

Sarahney (2005) and Goel et al. (2003) have also used naphthalene in electrochemical
and sonochemical studies. Also, naphthalene has chemical and physical properties similar
to other PAHS, except that it is more soluble in water, a property that makes naphthalene
more available in aqueous solution and more exposed to reactivity than other PAHS

(Alshawabkeh and Sarahney 2005).

MATERIALS AND EXPERIMENTAL METHODOLOGY

The experimental setup for the sonochemical tests is presented in Figure 4—1 whereas the

electrochemical setup is presented in Figure 4-2.

155

coca—00 E 000—05090: mo 5000890 0028000280 00.0 @3000 0000080308 .00 0008000m ”:0 2&5

0030300 0050590
:0 82060002

 

., mummmm..mw m. m o mwdmmmw
53 0000?- -. i .... WA... .0 .WW 1 1-00.0000 000% 0000-0

 

 

 

 

   

0009:“ 05500.35
00005.30 £054 ..

156

:00200 :_ 0:205:90: .00 :000000w00 003800000020 00.0 0300 350500900 .00 0008005 ”N40 0.5%»

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

900 000.5
, l -1- cows—00 0500:?“
. 5 80150002
mm - WWW
mo. amrpll- I 1 0001000 002w 000m:
.. o .. exhaub-
..., . .. 1---..0..00.=00_0 83:30.01
59: 003000 U< > m:
- i -, - - - -1 M... ii .i W ‘. ,,.,. ...
0980:0w fl.
>395 0032* DO \ 009350 0.0—0&5

0:000:05

157

Reaction Chambers

The reaction chamber consisted of a 1 liter Pyrex glass beaker with a Teﬂon cap for
sonochemical as well as electrochemical tests. These materials of the cell were selected to
eliminate or reduce sorption of the contaminants onto the walls of the reaction vessel and
the cap. Titanium plates were used as electrodes in electrochemical tests for this study.

Alshawabkeh and Sarahney (2005), Li et al. (2005) and Goel et al. (2003) have used

titanium as a base metal and coated it with mixed metal oxides such as Rqu, IrOz, and

SnOz. The titanium plates used in this study were not coated. Uncoated titanium being

cheaper than commonly used electrodes such as platinum, or titanium coated electrodes,
was selected for this project. Also, titanium offers resistance to corrosion in a wide range
of aggressive conditions and hence, could be used for potential ﬁeld scale application.
The titanium sheets used as electrodes in this study were 12.0-cm-long by 5.0-cm-wide

by 1.72-mm-thick and the distance between them was maintained at about 8.0 cm. The

. . 2 2
contact area of the solution With the electrode ranged from 53 cm and 55 cm as samples

were collected for analyses. The reaction vessel for the sonochemical tests was a l-liter

glass beaker with a Teﬂon cap.

Electrical Equipment

A 0.1 Hz to 2 MHz Sweep Function Generator was used to generate the AC signals. The
power supply consisted of a bipolar operational power supply/ampliﬁer capable of
producing 200 V AC/DC and 1 A AC/DC output or a maximum electrical power of 200

W. Voltage and current measurements were obtained with the aid an oscilloscope.

158

Ultrasound Equipment

The ultrasound equipment used was a Sonics VCX 750 that was operated in a continuous
mode at a ﬁxed frequency of 20 kHz. The ultrasound was delivered through a one inch
diameter titanium alloy probe and the ultrasound output amplitude ranged from 1% to
100%. Although the maximum output of this model was 750 W, the power it delivered
depended on the viscosity of the liquid being processed. The higher the viscosity, the
greater the power delivered by the ultrasound equipment. By setting the amplitude to
100%, the maximum power delivered to our test solutions was 60 W. Hence all tests were

conducted at 30 W and 60 W ultrasound power output.

Spiked Solution Preparation

Aqueous solutions, spiked with naphthalene were prepared by adding naphthalene (98%,
purity) crystals to one liter of deionized (DI) water in a capped Pyrex bottles and
allowing the suspensions to stand for about three days. This was followed by the addition
of 0.5 g of anhydrous sodium sulfate to each suspension, to increase its electrical
conductivity. Sodium sulfate was selected as a supporting electrolyte because Goel et al.
(2003) and Pepprah and Khire (2008) have used sodium sulfate in electrochemical
degradation of naphthalene in aqueous solution. The resultant concentration of Na2SO4
was approximately 500 mg/L (~3.5 mM). This concentration was selected in order to
maintain it at USEPA’s maximum contaminant level (MCL) for sulfate, which is equal to

500 mg/L (~3.5 mM). Saracco et al. (2000) performed tests where electrolyte salt (e.g.

NaCl, NaZSO4) concentrations ranged from 0.2 — 2 M and concluded that higher salt

concentrations reduce the electrical resistance between the electrodes and hence will

reduce the operating costs. However, greater salt concentration would result in greater

159

post-treatment costs and hence would be less favorable. The only exception might be
represented by discharging the treated efﬂuent into the sea (Saracco et al. 2000). The
excess suspended naphthalene crystals were then ﬁltered off and the collected ﬁltrate was
for the tests. Although no supporting electrolyte was required for the sonochemical test
solutions, in order to simulate similar reaction media in all tests, the test solutions used

for the sonochemical tests also contained 500 mg/L sodium sulfate.

Testing Procedure

Electrochemical tests were conducted in a batch mode in l-liter undivided cells. Prior to
the commencement of the passage of AC through the cells, initial samples were collected
for high performance liquid chromatography (HPLC) analyses. Replicate tests were
conducted for some of the tests. The effect of AC on the degradation rate of naphthalene
in solution was explored using a sine wave at a frequency of 60 Hz. This frequency was
selected because AC is supplied at 60 Hz frequency in the United States. The AC power
was obtained by the product of the root-mean-square (rms) current and the rms voltage.
During the tests, samples were collected for HPLC analyses at elapsed time intervals
equal to 0, 1, 2, 4, 6, and 8 hours. Sonochemical tests were also conducted in a batch
mode in l-liter glass beakers and sampled at the same frequency as the electrochemical
tests. The samples were analyzed directly without extracting the naphthalene from the
aqueous solution. HPLC analyses were performed using a diode array detector (detection
at 255 nm), a Waters C-18 PAH, 250 mm x 4.6 mm x 5 pm column, an automated
gradient controller, and deionized water (DD/acetonitrile mobile phase. The ﬂow rate of
the mobile phase was 1.2 mL/min and the column was ﬂushed with 10:90 (v/v)

acetonitrile/DI water for 5 minutes followed by 60:40 (v/v) acetonitrile/ DI water for 10

160

minutes, and by 100:0 (v/v) acetonitrile/ DI water for 5 minutes. This gradient was used
for the analysis in order to obtain appropriate separation of the degradation product peaks
on the chromatograms during the tests. Standard solutions were prepared to bracket the

expected concentration level of the analytes. Calibration curves with an average obtained

2 . . .
R value of 0.998 were used to determine the concentration of naphthalene in the

solutions as the tests progressed. Samples were also collected for Liquid Chromatography
/ Mass Spectrometry (LC/MS) analyses to determine the degradation products formed.
Control cells, similar to the test cells except without any current or ultrasound passed
through them, were set up, sampled, and analyzed at the same time intervals as the test
cells. The test cells were placed in a relatively large volume water bath to reduce the rate

of temperature rise of the test solutions.

Experimental Parameters Monitored

Apart from monitoring the concentration of naphthalene in aqueous solution over time,
other experimental parameters monitored included the initial and ﬁnal values of pH,
temperature, redox potential, electrical conductivity, and dissolved oxygen concentration
of the solutions. A portable dissolved oxygen/pH meter, equipped with a pH electrode, an
oxidation-reduction probe and dissolved oxygen probe, was used to measure pH, redox
potential and dissolved oxygen, respectively, of the test solutions. An electrical
conductivity meter was used to measure the electrical conductivity and a thermometer
was used to measure the temperature of the spiked solution in the control and test cells

throughout the testing period.

161

RESULTS AND DISCUSSION

Table 4-1 outlines the list of parameters explored and varied in the experimental program.

Table 4-1: Experimental Variables

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Compound Energy Power Output Frequency No. of
[W] tests
Naphthalene Electrical (AC) 60 60 Hz 1
Naphthalene Electrical (AC) 30 60 Hz 1
Salicylic Acid Electrical (AC) 60 60 Hz 1
Salicylic Acid Electrical (AC) 30 60 Hz 1
Naphthalene Electrical (DC) 60 0 1
Naphthalene Electrical (DC) 30 O 1
Salicylic Acid Electrical (DC) 60 O l
Salicylic Acid Electrical (DC) 30 0 1
Naphthalene Ultrasound 60 20 kHz 2
Naphthalene Ultrasound 30 20 kHz 2
Salicylic Acid Ultrasound 6O 20 kHz 1
Salicylic Acid Ultrasound 30 20 kHz l
Naphthalene - Control - 1
Naphthalene - Control - 1
Salicylic Acid - Control - 1
Total No. 17
of tests

 

 

 

 

 

162

Figure 4-3 shows the change in concentration (~ 20 mg/L, initial concentration) of
naphthalene in the test cells run at 30 W and 60 W ultrasound, DC and AC power, as well

as the control cell.

 

 

I I l I I I I I I I I I I I I I I I
1"‘\.‘ """""""" 5L“ """""""" "'1
b \% . Y + ﬁQControl Cell ..
0.8 '- ------ i! ---------- -------- """"""':"AC(30W)"'
- g g ; 3AC(GOW) -
0.6 I— --------------- >1 ---------------------------------------------------------------- d

o

.5

rr
'3
.1

i ,, s i
' E E US (30 W) ‘
’ co - 20 mgIL S ' '

Normalized Naphthalene Concentration, CICo

 

 

 

0.2 -------- ~ ---------------- ' DC (30 W) -
’ Su ortln Electrol 9: Na 80 5 ‘
j, AC: Sine Wave, f8 60 Hz ‘ 3 US (60 W) .
o 1 I n I 1 n n l n 1 1 ‘ l 1 n . l n n
0 2 4 6 8 10
Elapsed Time [hrs]

Figure 4-3: Concentration-time proﬁle for both sonochemical and electrical degradation
of naphthalene in solution at 30 W and 60 W output power

Effect of Power (AC, DC and Ultrasound)

The error bars in Figure 4-3 represent the maximum and minimum values of the

normalized concentration (C/CO). The DC test and the ultrasound test cells exhibited

similar degradation rates at 60 W and 30 W power output. However, the AC test cell

163

showed signiﬁcantly lower rate of degradation at 60 Hz frequency. At 30 W output
power, DC showed slightly higher rate of degradation. Therefore, for a given current
density, DC was more efﬁcient in degrading naphthalene in solution compared to an
equivalent AC at 60 Hz. At lower frequency, the rate of degradation of AC increases
(Paper No. 3). It is due to factors such as competing anodic and cathodic reactions at the
same reaction site (the electrodes) in AC electrolysis unlike DC electrolysis where anodic
oxidation occurs only on the anode. The key reason for DC showing greater degradation
rate than AC is the constant reversal in the direction of the current when AC is applied
decreases the rate of migration of the PAH to the reaction sites which are both electrodes
(instantaneous anodes) in an AC cell. Because the degradation of the PAH only occurs at
the electrodes, the decrease in the rate of migration of the PAH to electrode decreases the
rate of degradation. As the frequency of the applied AC increases, the rate of migration
further decreases because the rate of change of direction of the current ﬂow increases and
hence taking longer time for the PAH to reach one of the electrodes.

Another possibility for the higher degradation rate for DC compared to AC is that
DC produces hydroxyl radicals at a greater rate at an equivalent current. However, it was
not possible to quantify the rate of production of hydroxyl radicals.

Figure 4-3 also illustrates that the higher the output power of DC, AC or
ultrasound, the greater the degradation rate of naphthalene in solution. This is consistent
with other studies (Psillakis et al. 2004; Park et al. 2000) that reported increased rate of
conversion of PAHs with increased ultrasound output power. As ultrasound power
increases, the number of collapsing cavitational bubbles increases, resulting in enhanced

degradation rates. Increasing electrical power output results in increased electrical current

164

density and hence increased rate of degradation of target organics in solution. This is
consistent with ﬁndings by Alshawabkeh and Sarahney (2005) who reported increased
rate of degradation of naphthalene in aqueous solution with increased applied current

denshy.

Similarities between Sonochemical and Electrochemical Degradation

Oxidizing Agents

Sonolysis has been shown to produce reactive species including H202, H02, Hi, and

OH., which potentially can start radical chain reactions throughout the sonolyzed media

(Lorimer and Mason 1987). Riesz et al. (1990 and Sehgal et al. (1982) have veriﬁed the
presence of hydroxyl radicals by techniques such as electron spin resonance and 2,2-

Diphenyl-l-picrylhydrazyl. Brillas et al. (2000) reported that the primary oxidizing agent

in an electrochemical degradation study was the hydroxyl radical (OH.), although

secondary oxidizing agents such as H202 and H02. radicals could have been formed.
Hydroxyl radicals attack the benzene rings of aromatic compounds to produce
hydroxylated compounds. Various OH radical probes and techniques have been used in
many studies to detect the generation of OH. in the reaction media. These include
salicylic acid (Li et al. 2003; Marselli et al. 2003; Jen et al. 1998; Scheck and Frimmel
1995), 4-hydroxybenzoic acid (Anderson et al. 1987), benzoic acid (Oturan and Pinson
1995), para-chlorobenzoic acid (pCBA) (Elovitz and von Guten 1999; Haag and Yao
1992), 2,2-diphenyl-1-picrylhydrazyl (DPPH) (Sehgal et al. 1982), 5,5-dimethyl-l-

pyrroline-N-oxide (DMPO) and electron spin resonance (ESR) (Marselli et al. 2003), p-

165

nitrosodimethylaniline (RNO) and spectrophotometer measurements (Tanaka et a1.
2004), and RNO / ESR (Comninellis 1994). Since ESR methods require expensive and

sophisticated equipments; they were not used in this study due to budget constraints.

Trapping of OH. using salicylic acid, and HPLC measurements were selected as hydroxyl

radical probe technique for this study.

Jen et a1. (1998) reported that the oxidation efﬁciency of OH. is usually limited by

the rates of OH. generation and hence, increasing the OH. generation and keeping it at a

high level are the main factors that control the degradation efﬁciency of aqueous organic

substances. Haag and Yao (1992) also reported that in contrast to most other oxidants,

reactions of OH. with organics containing OH or C-C multiple bonds generally proceed
. . . . . . 10 -l -1

With rate constants approaching d1ffusron-controlled limit (10 M s ) and therefore

oxidation rates are usually limited by the rate of OH. generation and competition by other

OH. scavengers in solution rather than by inherent reactivity with the oxidant. This

implies that, the higher the rate of generation of OH. in solution, the faster the

hydroxylation rate of the OH radical probe and hence the faster the rate of decrease of the
concentration of the probe (salicylic acid) in solution. However, in an electrochemical
setup, the rate of diffusion and migration of the PAH also is an important factor

controlling the rate of degradation. Figure 4-4 shows a plot of the rate of degradation of

the hydroxyl radical probe, salicylic acid (CO ~ 20 mg/L) using both ultrasound and

166

electrical energy (AC and DC) at 30 W and 60 W power output. Degradation rates

increased with increased power as observed in the naphthalene test cells.

 

 

 

 

 

I j I I I I I l I I I I f I I I I 1 I
(:3 1 RE“ 5? ﬂ? “ <E>Control """ "
0 ‘\.\ .
- I - I
.5 A 3 ‘ AC (30 W)
‘3 0.8 - ------------- -------- ---------- '-AC(60W)--
g g f us (30 W)
8 2 3 us (60W)
3 s z z
0 0,5 _ .............. .‘ .............. 3 .................................................. -
2 ~ ' 1
0 .
< 5
.2 E
.3 0.4 — ------------------------------- ---------------------------------------------------- «-
— A
a: .
w .
u i I I ' u
3 _ c -20m IL. I 5 3
E 0.2 - o 9 --------- ---------------- 3 -- ------------- 0
Supporting Electrolyte: Na 80 2 3
§ ' c s 2 4 , A DC (30W)
- A : ° W ,f=60 H : ;
o l :ne 1 avle 1 n z: I l 1 1 I 1 n 1 Dc (60 w)
0 2 4 6 8 10
Elapsed Time [hrs]

Figure 4-4: Concentration-time proﬁle for both sonochemical and electrical degradation
of salicylic acid (OH radical probe) in solution at 30 W and 60 W output power

Figure 4-5 shows the pseudo-ﬁrst-order degradation rate constants for both

sonochemical and electrical degradation for applied power of 30 W and 60 W.

167

0.5

 

l 17 _|
C = 20 mglL (Naphthalene, Salicylic Acid)

° : DC
Supporting Electrolyte: NaZSO4 1

0-4 -~ * AC: Sine Wave, f= so Hz 3 --------------- E ----

Degradation Rate Constant, k [hr-1]

 

 

 

30 60 3o 60
Naphthalene Salicylic Acid

Power [W]

Figure 4-5: Pseudo-ﬁrst-order degradation rate constants for naphthalene and salicylic
acid

The reduced rate of sonochemical degradation of salicylic acid compared with
naphthalene may be due to the fact that degradation occurred mainly in the bulk solution
by OH radical attack because salicylic acid, being partially polar preferred to stay in
solution unlike non-polar naphthalene that migrated into cavitational bubbles and hence
was degraded by pyrolysis as well. Also, in general, reaction rates involving OH radicals
increase with increasing molecular size based upon a rule of additivity (Westerhoff et al.
1999). Therefore naphthalene having eight hydrogen atoms as potential sites that can be

substituted by OH radicals tend to exhibit higher degradation rate than salicylic acid with

168

four potential sites. This suggests that the main degradation mechanism in sonochemical
degradation was mediated by OH radicals. However, electrochemical degradation rates of
salicylic acid were generally higher compared with naphthalene, suggesting that an
alternative degradation mechanism may have contributed to the degradation. Also the rate

of movement of the reacting molecules in the electrochemical test cell to the reaction

Sites (anodes) was enhanced by the partial dissoc1atlon of salicylic acrd in solution to H

and negatively charged salicylate ion. Salicylate ions, being negatively charged, were
attracted by the positively charged electrode (anode) and hence, the rate of migration to
the reaction site was enhanced unlike naphthalene molecules which have no net charge in

solution. Due to this reason, DC showed the highest degradation rates.

Degradation Kinetics

Park et al. (2000) reported that the degradation of PAHS by ultrasonic irradiation in
solution was ﬁrst-order with respect to the PAHS. Zhaobing et al. (2005) reported that
ultrasonic degradation of 2,4-dinitrophenol ﬁtted a pseudo—ﬁrst-order degradation

kinetics. The degradation of naphthalene as well as salicylic acid in all the test solutions

in this study also depicted pseudo-ﬁrst-order degradation kinetics. A plot of ln[C/C0]

versus time yielded the observed pseudo ﬁrst-order degradation rate constants (k)

reported in Figure 4-5.

Degradation Byproducts
Because both sonochemical and electrical degradation of organics in solution involve the
utilization of hydroxyl radicals as the primary oxidizing agent, similar degradation

products were observed in all the tests. LC/MS analyses indicated molecular masses that

169

suggest the formation of degradation products such as naphthol, dihydroxynaphthalene,
trihydroxynaphthalene, dihydrotrihydroxynaphthalene, naphthoquinone,
hydroxynaphthoquinone, and benzenedicarboxylic acid. The exact isomers formed were
not determined. These degradation byproducts formed are relatively more soluble than
the parent compound (naphthalene), hence it is expected that they would be more bio-
available in solution for subsequent degradation by microorganisms. The change in color
of the test solutions to light yellow as the tests (sonochemical and electrochemical)
progressed may have been due to the formation of naphthoquinone in the solutions.
Formation of quinones is a common observation of the hydroxylation process in
oxidizing media (Oturan 2000). The formation of naphthoquinone is consistent with those
found by other researchers who also observed aromatic quinones during electrolytic
oxidation of aromatic hydrocarbons (Brillas et al. 2000; Panizza et al. 2000; Saracco et
al. 2000). GC/MS analysis by Goel et al. (2003) indicated the presence of at least one
reaction product, 1,4-naphthoquinone when naphthalene solutions were electrolyzed and
concluded that naphthalene degradation could be due to direct anodic oxidation,
oxidation by an intermediate (e. g. hydroxyl radicals), or a combination of both.
Formation of 2,3-dihydroxybenzoic acid and 2,5-dihydroxybenzoic acid as
electrochemical as well as sonochemical degradation products of salicylic acid is
consistent with numerous studies indicating that hydroxyl radicals were the primary

oxidizing agent.

Heat Generation
Both sonochemical and electrical degradation tests resulted in heat generation. Hence all

test solutions were placed in a water bath to maintain the temperature of the test solutions

170

between 19 oC and 25 °C. One major drawback of using ultrasound or electrical energy

for potential ﬁeld scale application is the loss of part of the applied energy as heat.
However, no additives are required for the oxidation process unlike chemical methods

where additives such as hydrogen peroxide, ozone or ferrous ion are commonly used.

Mechanism for 0H Radical Generation

Ultrasound causes high-energy acoustic cavitation, which consists of formation of
microscopic vapor bubbles in the low pressure (rareﬁed) part of the ultrasonic wave when
the low pressure overcome the surface tension of the ﬂuid. The bubbles collapse or
implode in the compression part of the wave creating very minute high-energy
movements of the solvent that result in localized high shear forces (Meegoda and
Veerawat 2002). The implosive collapse of the bubbles in water can cause localized
pressures of the order of 975 bars and temperature as high as 4200 K generated over a
few microseconds time interval (Lorimer and Mason 1987). Temperatures of over 5,000
K have been reported in the ultrasonic cavitation of organic and polymeric liquids (Flint
and Suslick 1991). Under these conditions, water vapor undergoes a thermal dissociation
to yield extremely reactive hydrogen and hydroxyl radicals (H' and OH') (Hung et al.

2002).

H2 0 Thermal _ Dissociation _ 4200K _ 975 _ bar > H 0 + 0H 0 (1)

 

On the other hand, Comninellis (1994), Brillas et al. (2000), and Panizza et al.
(2000) have reported that DC electrochemical systems generate hydroxyl radicals at the

anode due to the oxidation of water:

171

 

Anodic _ Oxidation o + _

AC electrochemical systems also generate hydroxyl radicals (Paper No. 3) by anodic
oxidation of water molecules at an instantaneous anode. During one half of a cycle when
one of the electrodes has a positive electrical potential (instantaneous anode), it
instantaneously acts as the anode while the other electrode, at a negative electrical
potential (instantaneous cathode), instantaneously acts as the cathode. In the subsequent
half cycle, polarities are reversed and the electrodes acquire electrical potentials opposite
to what they had in the previous cycle and this continues throughout the period of

application of the AC.

Reaction Site
It is generally believed that there are three potential sites for chemical reactions in
ultrasonically irradiated aqueous solutions: (1) the bubble itself; (2) the interface between
the bubble, and the surrounding liquid; and (3) the bulk solution. Compounds may
degrade directly via pyrolytic reactions occurring inside the bubble and/or at the
interfacial region or indirectly via radical reactions occurring at the interface and/or the
solution bulk (Thompson and Doraiswamy 1999). For electrochemical systems, it is
generally believed that the reaction site in electrochemical processes is the surface of the
electrodes. Chen (2004), Feng and Li (2003), and Stucki et al. (1991) have reported that
the effectiveness of electrochemical wastewater treatment depends on the nature of the
anodes used in the process.

Therefore, in sonochemical degradation, hydroxyl radicals generated in the solution

attack target organic contaminants in solution near as well at a distance from the

172

ultrasound probe. In electrical degradation, adsorbed hydroxyl radicals generated on the
surface of the electrodes (anode) are responsible for the degradation of organics in
solution. As a result, the type of electrode material used and the rate of migration of the
chemical specie to the electrode affect the degradation rates and degradation mechanism.
Gattrell and Kirk (1993 and Comninellis and Pulgrin (1991) have reported that using
platinum as electrodes resulted in the formation of products such as maleic acid and

oxalic acid from the anodic degradation of phenol. However, complete mineralization of

phenol to C02 was achieved by using Pb02 as an anode (F eng and Li 2003; Iniesta et al.

200]; Schumann and Grundler 1998).

Reaction Mechanism and Pathway

Sonochemical degradation of organics in solution generally occurs through two distinct
degradation pathways: oxidation by hydroxyl radicals and pyrolytic decomposition
(Thompson and Doraiswamy 1999). In electrochemical oxidation, organic compounds in
aqueous solutions can be oxidized on an anode (anodic oxidation) by indirect oxygen
atom transfer and direct electron transfer (Rodgers et al. 1999; Kirk et al. 1985). During
the indirect oxygen atom transfer, oxygen radicals, especially the hydroxyl radicals
generated from water electrolysis, help in the oxidation of organic substances (Iniesta et
al. 2001; Simond et al. 1997; Comninellis 1994; Lee et a1. 1981). The hydroxyl radicals
readily react with the organic molecules adsorbed on or in the vicinity of the anode to

cause the reaction presented in Equation 3 (Simond et al. 1997).

Anode[0H ' Lds + Organic —> Anode + Pr oduct (3)

173

Meanwhile, the hydroxyl radicals react with each other to form molecular oxygen to
complete the electrolysis of water molecules (Li et al. 2005; Comninellis 1994) as shown

in Equation 4.

2Anode[0H']ads —-> 2Anode + 02 + 2H+ + 2e‘ (4)

Hydroxyl radicals react with organic pollutants in two possible ways presented as
follows:
1. by electrOphilic addition to a double bond or an aromatic system, and
2. by abstraction of a hydrogen atom from a carbon atom (Schwarzenbach et al.
2003)
Based on the identiﬁed degradation products, a simpliﬁed degradation pathway for the
electrochemical degradation of naphthalene in aqueous solution is proposed as illustrated

in Figure 4-6.

174

 

“coma wEEExo beﬁts

2: mm £888 :0 5:5 5:28 E ecu—«Egan: mo .3368 core—omuwoc 3282—08820 can 33.55088 comm—95m no-..» Rama

...8

Eigenuazgnonaom

:ooo/ei/

:ooo.zw

o5.2—E.—aehneezntasehﬁn cue—nausnaaaxe—iﬂuh
:0 IO

.om

I .
:

,...w.o
:o :0

IO :0

,wo

“:0

O
a H Limo eggs—5523.333:
0 IO
.10
O
r. .1 I fwd 2.25:... as
.ION L . a:— Z
04
:ocombmnej W

comp—PA: W . .I
_ v.
one—«5.33.38635
:0

>

s\Vt

. .:o.

:0 . IO

......ﬁaz- . ,

I .. _ Swazi—«z

\i
\\
\\

- hwy—.3.—

..ocaeauuoc 952:8?
..8 6 95!: com

I,

/:

 

175

The exact isomers formed were not determined. Hence, the presented molecular
structures are an example of the possible isomers of each byproduct. Initial
hydroxylation of naphthalene by an OH radical resulted in the formation of naphthol.
Further hydroxylation resulted in the formation of dihydroxynaphthalene
(naphthalenediol), which was further oxidized by hydroxylation to
trihydroxynaphthalene, or by hydrogen abstraction to naphthoquinone. It is possible that
addition of hydrogen atoms to naphthoquinone may have reduced it back to
dihydroxynaphthalene, which is similar to a reaction step reported by previous studies
(Chin and Chen 1985). Chin and Cheng (1985) reported that in an undivided cell,
benzoquinone produced at the anode was subsequently reduced to hydroquinone at the
cathode. Therefore, it is possible that in our setup, the naphthoquinone, produced by the
oxidation of naphthalene on an instantaneous anode, was subsequently reduced to
dihydroxynaphthalene on a cathode (or instantaneous cathode) as shown in Equation 5.

C10H602 + 2H+ + 2e" <=> C10H802 (5)
Therefore naphthoquinone and dihydroxynaphthalene were probably in redox
equilibrium. Addition of hydrogen atoms to the trihydroxynaphthalene resulted in the
formation of dihydrotrihydroxynaphthalene. Further hydroxylation of the naphthoquinone
formed, resulted in the formation of hydroxynaphthoquinone. With sufﬁcient hydroxyl
radicals in the reaction medium, ring cleavage in naphthoquinone may have occurred
leading to the formation of benzenedicarboxylic acid. Li et al. (2005) reported that
benzoquinone formed from the oxidation of phenol could be degraded with ring breakage
to various carboxylic acids. One of the proposed mechanisms involves the adsorption of

benzoquinone onto the anode surface, which gives up an electron, and the carbon that is

176

 

double-bonded with oxygen is attacked by a neighboring hydroxyl radical. When this
process is repeated at the para position, the ring could be opened and benzoquinone
would be broken down into small organic molecules such as carboxylic acids
(Comninellis 1994; Lee et al. 1981). The present experimental ﬁnding is consistent with
the proposed mechanism because both naphthoquinone and benzendicarboxylic acid
(from probable ring cleavage of naphthoquinone) were identiﬁed as intermediate
byproducts. The possible mineralization of naphthalene was not investigated in this
study.

In the direct electron transfer process, organics are adsorbed on the anode surface

and give up electrons to the anode as presented in Equation 6.
[OrganchdS — e —) [Organzc _ radzcal _ cation] (6)

Naphthalene may undergo direct anodic oxidation to form a naphthalene radical cation
that can act as a precursor for the formation of hydroxylated species such as naphthol as

illustrated in Figure 4-7.

177

 

.uOmSooE
5:8 18:33 53» 35:93 8 cots—Om E ecu—Stine m0 @353 cosmuﬁwew 3282—25020 gum—:33 onEEG ”We 2:me

.we see—a2
EC
, h
,. .wo
:8.
ON:
. .w.» 820 .83 Bonsai
838.20 . 8:83.
€22 oaosao

625 625

Soﬁe—Eel

178

The naphthalene radical cation may also undergo cathodic reduction in AC electrolysis,
when electrode polarities are reversed from positive to negative, and reform back to
naphthalene. This probably accounted for the slow reaction kinetics encountered in the
AC test cells (Figures 4-3 & 4-4). Also, continuous bubbling of gases was observed on
the electrodes in the DC test cells but no continuous bubbling of gases was observed in
the AC test cell except few bubbles that were stagnant on the electrode surface. Therefore
according to Equation 4, if the formation of hydroxyl radicals are precursors for
formation of molecular oxygen to complete the electrolysis of water molecules, then it is
also possible that less hydroxyl radicals were produced at the electrodes as a result of
polarity reversal in the AC test cells unlike the DC test cells in which continuous
production of OH radicals at the anode efﬁciently produced more for faster rate of

oxidation of naphthalene or salicylic acid.

Modeled versus Measured Concentrations of Naphthalene

It is generally believed that electrochemical reactions occur mainly on the surface or in
the vicinity of the electrodes. Therefore, the rate of reaction of the organic reactants in the
test cell depends on the rate of generation of OH radicals at the electrodes as the
oxidizing agents, as well as the rate at which the reacting molecules move to the
electrodes for subsequent oxidation by the OH radicals. Mass transfer in solution occurs
by diffusion, migration and convection. Diffusion and migration result from a gradient in
electrochemical potential and convection results from imbalance of forces on the solution
when the solution is mixed or stirred (Bard and Faulkner 2001). Assuming a linear mass
transfer, and linear electric ﬁeld, the mass flux to the electrode can be expressed by the

Nernst—Planck Equation (Equation 7):

179

 

6C“) —5€-Dci‘§- + CV00 (7)
6x RT 1

J(x)=—Dm
where J is the mass ﬂux, D is the diffusion coefﬁcient of the reacting specie in solution,
C is the instantaneous concentration of the reactants in solution at time t, x is the distance
in the x direction or along the direction of the electrical current, 2 is the net charge on the
reacting species, F is Faraday’s constant, R is molar gas constant, T is temperature (K),

AE/l is the gradient arising from the change in potential AE over a distance I, and v is the

linear velocity of the solution. Because the test cell solutions were not stirred, the third

 

term in Equation 7 can be neglected and hence Equation 7 becomes:

J(x) =—D—— 6C(x)— ZF —DC-A—E— (8)
6x RT 1

For uncharged reactant, the second term in Equation 8 can be neglected to yield Fick’s

ﬁrst law as follows:

8C (x)

J(x) = —D 6x

(9)

Equation 9 was applied to model the rate of mass transfer of naphthalene to the electrodes

in the DC setup. The following assumptions were made when applying Equation 7 (or
Equation 8) to simulate the naphthalene concentrations:

1. Reactants (naphthalene) are instantaneously converted to products when they

come into contact with the electrode. Hence, C(t) at the surface of electrode is

equal to zero for t 2 O. This establishes a concentration gradient that drives

reactants by diffusion to the surface of the electrodes and this condition is

180

F
l.

considered as the maximum concentration gradient that may be established in a
given time step;

2. Steady-state diffusion conditions exist in a given time step which was assumed
equal to 1 hr;

3. Linear mass transfer exists in the given time step of 1 hr; and

4. Linear electric ﬁeld exists in the given time step of 1 hr.

Equation 9 was applied with C of naphthalene equal to CO (~ 20 mg/L) for t = O at the

cathode and C = 0 for t 2 O at the anode. Assuming a time step equal to 0.1 hr, the C of
naphthalene at cathode was estimated by subtracting the mass of naphthalene that moved
to the anode for degradation during the previous time step. This computation was carried
out for the test duration of 8 hours and the modeled normalized concentrations were
compared (Figure 4-8) to the test results for DC tests carried out at constant power output

of 30 W and 60 W respectively.

181

 

i

 

 

 

 

 

 

 

 

 

 

 

l l l i ’
o 1 t j : g ...............
2 ~t‘ . ' : 3 1
U __ 'o.‘ . ;
5 - ‘3“ E Model : Diffusion Only
'3 .. ...... . 0' ..... ' .....................
g 0'8 _ a.“ . ----- Model : Diffusion + Migration (30 W)
. \i
§ . a. F ‘ . ------- - Model : Diffusion + Migration (60 W)
,. 0. : :

g 06 __ ______________ .l“‘ O TestDataz30W
3 ~ ' '-. ‘s‘ Q I Test Data : 60 W
2 st,
.3 ' z i ‘-. 2 2 ‘
c 0.4 ..-.. ......... y ........... 5..-... ....... § ........ a ....... - ......... j. ............... ..
Q .‘ .0 ‘ ‘. ;
g h ". ~‘2.
.8 _ ' 0...... ..‘. ,
.8 f 0.... ..~: -
i 0.2 _ .............. f ................ . . . .. ........... i‘m .......... . .............. -
E h I ' ...-."O. . 't
2 .

o l l l l

0 2 4 6 8 10
Elapsed Time [hrs]

Figure 4-8: Simulated concentrations versus test data for naphthalene tests using DC

Schwarzenbach et al. (2003) reported diffusion coefﬁcients of organic solutes equal to

about 3.0 x 10'5 cmzs-1 for small molecules to 5.0 x 10'6 cmzs-1 for those of molar mass

near 300 g/mol. Based on this data, the diffusion coefﬁcient of naphthalene was

. -6 2 -1 . . .
estimated as 9.0 X 10 cm s from the d1ffusron coefﬁcrents vs. molar mass chart

presented by Schwarzenbach et al. (2003). Equation 9 was applied to simulate only
diffusion driven mass ﬂux of naphthalene to the anode for oxidation. The simulated
normalized concentration is presented in Figure 4-8. The simulated degradation rate as a
result of naphthalene migration to anode only due to diffusion is orders of magnitude less

than that observed during the experiments run at constant DC power output of 30 W and

182

60 W with NaZSO4 as the supporting electrolyte. When both diffusion and migration are

used for simulating the mass ﬂux of naphthalene to the anode by using Equation 8, the
simulated normalized concentrations are presented in Figure 4-8 when the net charge, 2,
was assumed equal to -O.2 for naphthalene.

For neutral molecules such as naphthalene, 2 should be equal to zero. However, it is
hypothesized that the presence of anions in the test solution (from the supporting
electrolyte) impart an effective negative charge on the naphthalene molecules. These
anions dragged the naphthalene molecules along as they migrated to the positively
charged electrode (anode), where anodic oxidation resulted in the degradation of the
naphthalene molecules. Therefore, the naphthalene molecules moved in the solution as a
result of migration simulated by the second term in Equation 7. The rate of migration of
naphthalene molecules towards the anode is enhanced by assigning a negative value to z
for naphthalene. By assigning the naphthalene molecule an effective partial negative

charge of -O.2, the simulated normalized concentrations versus time data ﬁts well with

that for the experimental data when NaZSO4 was used as the supporting electrolyte for

DC power output of 30 W and 60 W.
The simulated degradation rate as a result of salicylic acid migration to anode only

due to diffusion is orders of magnitude less than that observed during the experiments run

at constant DC power output of 30 W and 60 W with Na2SO4 as the supporting

electrolyte. When both diffusion and migration are used for simulating the mass ﬂux of

naphthalene to the anode by using Equation 8, the simulated normalized concentrations

183

 

are presented in Figure 4-9 when the net charge, 2, was assumed equal to -0.35 for

salicylic.

 

 

 

 

 

 

 

 

 

 

 

 

I I I I
8 1 i. Diffusion Only . --------------- 1
5. it.
5 . "is. . § Model : Diffusion Only
g 0'8 -"..“ """ """ 7 """"""" ' ' " ----- Model : Diffusion + Migration (30 W)
2:; - '-.. s‘ Model : Diffusion + Migration (60 W)
° ’ .'t ‘o E 0 Test Data : 30 w
o 0.6 _ ......... .'i. ..... ‘
'3 .. l ‘\ 5 I Test Data : 60 W
< .- '0'. § 2 _ .
'9 ’ '-‘ ‘\ 5 3 E -
E”. 0.4 _ _______________ In... ...... “\‘ﬂ... ....... ....... ............. ............... -
.3 " ' "a E .‘s ' '
g 02 ---... "...‘ni ................. ............... _.
g i- '5‘. ‘.... «-
‘z’ I f ' "--....I "o
0 l l 1 .‘...-"~-
0 2 4 6 8 10
Elapsed Time [hrs]

Figure 4-9: Simulated concentrations versus test data for salicylic acid tests using DC

Salicylic acrd dissoc1ates partially in solution for form the salicylate anion and H , hence

it was expected that its apparent effective charge in the test solution would be greater
than naphthalene which is a neutral molecule as depicted by the model.

The rate of anodic oxidation of naphthalene (or salicylic acid) depends not only on
the rate of migration of reacting molecules to the reaction site (the anode or instantaneous
anode) but also on the rate of production of OH radicals at the anode (or instantaneous

anode). Hence, as the current density is decreased, it may decrease the rate of production

184

h”
A
I
ﬁll...”

of OH radicals at the anode and may become the rate limiting step. When Equation 8 is
applied, the rate limiting step assumed is the rate of mass transfer to the anode (by
diffusion and migration). Hence, Equation 8 would overestimate the decrease in the
concentration at a given time, t of the reactant when current density is low enough not to
degrade all naphthalene mass diffusing and migrating to the anode (or instantaneous
anode). At higher current densities, the predictions from Equation 8 would be more
accurate. It is beyond the scope of this study to estimate the rate of hydroxyl radical
production as a function of current density and other variables. Hence, it is not possible to
calculate analytically at what current densities the assumptions are valid when Equation 8
isapphed.

While Equation 8 is for DC and cannot be applied for an AC, it can be used to
explain the reduced rate of degradation when AC was applied. Proposed reasons for the
decrease in degradation rates using AC are as follows:

1. By assuming that the naphthalene molecules had an effective partial negative
charge of say 2 = -O.2 under our experimental conditions based on the acceptable
fit for DC data, increasing the frequency of the AC resulted in back-and-forth
movement of the reacting molecules, and hence the net mass flux of the molecules
toward the electrodes was reduced. If the net distance traveled by the naphthalene
molecules to get to an instantaneous anode, L, in Equation 8, is varied, until the
modeled values of concentrations ﬁt with the experimental data the values of L
obtained is 40 cm compared to the electrode spacing of 8 cm. Assuming
degradation rate at the electrodes are the same for each frequency, it can be

hypothesized that AC frequency results in taking longer duration for a

185

 

 

naphthalene molecule(s) to reach the instantaneous anode. This may be due to the
back and forth movement of the molecule due to the reversal in the direction of

the current.

. Continuous bubbling of gases at both electrodes (probably 02 and H2) in the test

solution was observed as the tests progressed. However, no gas bubbles were
visible at an AC frequency of 60 Hz. Simond et al. (1997) and Comninellis (1994)
reported that the hydroxyl radicals, formed on an anode react with each other to
form oxygen gas at the anode to complete the electrolysis of water molecules.
Stock and Bunce (2002) reported that hydrogen radicals formed on a cathode
react with each other to produce hydrogen gas at the cathode. If the formation of
OH radicals and hydrogen radicals are precursors for the generation of oxygen gas
and hydrogen gas at the anode and cathode, respectively, then it is likely that the
rate of gas bubbling at the electrode gives an indication of the rate of formation of
these radicals at the electrodes. The higher the rate of the observed bubbling of
the gas(es), the greater the rate of hydroxyl or hydrogen radical formation at the
electrodes. Hence, based on the visual observation of gas bubbles, it is likely that
the AC frequency resulted in decrease in the rate of formation of hydroxyl
radicals at the electrodes. This impacted the rate of oxidization of naphthalene at
the instantaneous anodes.

Competing anodic reactions (oxidation) and cathodic reactions (reduction) at the
same reaction sites (on the surface of the electrodes) may have been responsible
for the decrease in the oxidation rate of both naphthalene (Figure 4-5) and

salicylic acid (Figure 4-7) as the AC frequency was increased. However, in DC

186

electrolysis, oxidation reactions occurred at the anode only and reduction
reactions occurred at the cathode only and hence there was no competing anodic
oxidation and cathodic reduction at the same reaction site. Therefore DC was
more effective in degrading naphthalene or salicylic acid in the test cells.

Table 4-2 summarizes the measured initial and ﬁnal values of the experimental

parameters.

187

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

hum _ 5 v6 w.m mmv 3v We we 9mm odm om 3:895 one—mﬁaamz
omw wow 3V Qm m? Re me we cém 9cm co 95032: ecu—mﬁnqaz
and Eo<
omw cmw as vw awm one Nd— a.m 1mm v.2 on 23385 0:523
Ave 23.
ova 8w m. 2 as 5mm New 3: w. m 3am 9.2 cc Bamboo—m anthem—mm
09
35 RR 0.5 w.m mum omv v.2 ﬁe w. _ N 5.2 cm Rogue—m ecu—«ﬁnaaz
09
>3 3w ”2 w.m emm wmv Nd— wé aém 93. cc .8585 ecu—«ﬁnamz
GS 23.
wow oww 5m Ow 3v w; w.m Qm 3am odm cm .8585 2302mm
GS 22
New 3w WE Wm 5m wvv QM o6. 38 0.8 cc 305er 332—mm
6.3
m5 wow o6 w.m a? m3 N6 no mean 0.2 cm Row—SSE ace—aﬁsamz
G3
omw com 06 w.m mam m3 9m >6 wdm @6— oo Borneo—m. ocoﬁﬁnmaz
33m 35.: .35— EEE 35m :52: :35 .33.: 35h BE:—
EEQJ SEED 9.5
DEB—5:3 cough-.359 33:83 cow .3—
..BEooE now»: 32:35 .83.. 6.3—=55 I: 9:533:55. .33.:— mwuoﬁm 65595.0

 

 

$20883 @23on mo meE:m ”we 2an

188

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8

53 com 5w 9w 5mm wmm of ad 0.2 odm .8280 - 2“?me

m _ w _ 5 m6 w.m a3. wmv >6 5o 92 ad_ 35:00 - one—mﬁnmaz
8

N3 0mm 06 1w vow aov QM o6 QMN odm cm 250235 omwwowwm
8

aow cow NH Wm w?» m: QM 9v 9mm odm co vasoé—D 0_Wmo“~am

 

3:8 3 22¢

 

 

9
OO
1

SUMMARY AND CONCLUSIONS

The key objective of this study was to compare the rate of degradation of PAHs in

aqueous solution using ultrasound energy and electrical energy (AC and DC) delivered at

the same output power. Tests were conducted at 60 W and 30 W output powers.

Naphthalene, which contains two fused benzene rings, was selected as a model PAH

compound. Salicylic acid was used as a hydroxyl radical probe. The key conclusions of

this study are as follows.

Both ultrasound and electrical (AC and DC) energies are capable of oxidizing
naphthalene as well as salicylic acid in solution.

DC, at a given power output, degraded naphthalene at a rate faster than a sine
wave AC having a frequency equal to 60 Hz, at an equivalent output power.
However, both DC and ultrasound exhibited comparable degradation rates for
naphthalene but DC exhibited a faster degradation rate than ultrasound when
salicylic acid was used. This was attributed to the formation of negatively charged
salicylate ion in solution which was attracted by the positively charged electrode
(anode) and hence the rate of migration of the reacting molecules to the reaction
site (surface of anode) was enhanced. AC showed the least degradation rate
possibly due to decrease in the migration of the PAH to the electrode due to the
constant reversal of the direction of the current. It is also possible that the rate of
hydroxyl radical formation for AC was less than an equivalent DC.

The higher the output power, the greater the degradation rate of naphthalene or
salicylic acid in solution which is due to a greater rate of formation of hydroxyl

radicals

190

 

DC, AC and ultrasound, all depicted pseudo-ﬁrst-order degradation kinetics with
respect to naphthalene or salicylic acid in solution.

Formation of 2,3-dihydroxybenzoic acid and 2,5-dihydroxybenzoic from the
oxidation of salicylic acid is consistent with reported ﬁndings that hydroxyl
radicals were the primary oxidizing agents in both sonochemical and
electrochemical degradation.

The rate of anodic oxidation of naphthalene depends on the rate of production of
OH radicals at the anode (or instantaneous anode) as oxidizing agents as well as
the rate of migration of reacting molecules to the reaction site (the anode or
instantaneous anode). At higher power output, not only more OH radicals are
produced at the anode but also the rate of migration of the reacting molecules to
the anode (or instantaneous anode) increases during a DC application. However,
when AC is used, the change in the direction of the current decreases the rate of
migration of the reactant to the instantaneous anode.

Naphthalene molecules in solution exhibited an “effective” partial negative
charge due to the presence of the supporting electrolyte subjected to electrical

gradients.

191

 

PAPER NO. 5: DEGRADATION OF NAPHTHALENE IN THE
AQUEOUS PHASE OF SPIKED SATURATED OTTAWA SAND

ABSTRACT

The objectives of this study were to assess degradation of naphthalene in the aqueous
phase of spiked Ottawa sand and to explore the effect of AC frequency and current
density on degradation rates. AC having sine wave shape and frequencies equal to 0.1 Hz

and 60 Hz as well as square wave AC at 0.1 Hz were used for the tests. The peak AC

. . 2 2 2 .
current densrtres were 3 mA/cm , 6 mA/cm , and 10 mA/cm . Direct current (DC) was

. . . 2 .
applied in a separate test at current densrty equal to 3 mA/cm . The key conclusrons of

this study are: (1) both DC and sine and square AC can degrade naphthalene or PAHs in
the aqueous phase of spiked sand; (2) the rates of degradation of naphthalene in the
aqueous phase when sand was also present was less than the degradation rate at an
equivalent current density when only the aqueous solution (without the sand) was
subjected to electrochemical degradation; (3) increase in the peak density of AC resulted
in an increase in the degradation rates due to greater rate of formation of the hydroxyl
radicals which are primarily responsible for the PAH degradation; and (4) an increase in
the AC frequency resulted in a decrease in the degradation rate due to slower rates of
migration of the PAH to the electrodes due to periodic reversal in the direction of the

current.

192

INTRODUCTION

Much of the practical work on electroremediation on soils has been undertaken in the
laboratory on spiked synthetic soils (Page and Page 2002). Direct current (DC) is mostly

used in studies involving electroremediation of soils. Current densities applied have been

in the range 0.025 to 5 A/m2 (Acar et al. 1991; Hamed et al. 1991). Initially,

electroremediation was applied to remove salts or alkalis from soils (Puri and Anand
1936; Gibbs 1966; Vadyinina 1968). Its application to the removal of heavy metals was
first attempted in the early 19803 (Hamnett 1980; Segall er al. 1980; Agard 1981).
Successful transport of a large number of heavy metals through soils by electric ﬁelds
with subsequent removal onto an electrode or into the solution surrounding the electrode
(catholyte or anolyte) has since then been demonstrated in several laboratory studies
(Page and Page 2002). The metals are usually present as cations and are therefore
transported mainly by electromigration toward the cathode. However, certain elements,
e.g. Cr and As, may be present as oxy-anions and are therefore transported to the anode
(Banerjee 1987; Lindgren er a1. 1994; Pamakcu, and Wittle 1994). Reddy and Saichek
(2004) conducted laboratory investigation to determine contaminant mass removal from
low permeability clay, by using periodic DC voltage application and surfactants. The
results showed that considerable comtaminant removal can be achieved by employing a
voltage gradient of 2.0 V (DC)/cm, along with a periodic mode of voltage application.
PAHs have been found in at least 600 of the 1,430 National Priorities List (NPL)
sites identiﬁed by the United States Environmental Protection Agency (USEPA)
(ASTDR 1995) and are classiﬁed as potential carcinogens. For molecular substances,

electro—osmosis is the predominant form of transport under an electric ﬁeld, although

193

partial dissociation into ions of some organic molecules will result in electromigration as
well (Page and Page 2002). Lageman et al. (1990) has demonstrated that the removal of a
range of aromatic organic compounds from soils was feasible.

Most studies on the electroremediation of organic compounds have focused on
using DC to transport them by electromigration and/or electroosmosis with subsequent
removal onto electrodes or into solutions surrounding the electrodes (catholyte or
anolyte). The objective of this study, was to use alternating current (AC) to assess the
degradation of a PAH (naphthalene) in a clean poorly graded sand spiked with spiked
naphthalene solution Naphthalene was used as the model PAH compound because it has
relatively high solubility in water ( ~ 30 mg/L) as compared to other higher order PAHS.
Clean Ottawa sand was selected as the porous media because it would not signiﬁcantly

sorb naphthalene from the spiked solution used for saturating the sand.

MATERIALS AND EXPERIMENTAL METHODOLOGY

The experimental setup is illustrated in Figure 5—1 and the setup components are

described below.

194

 

.958 EcoEtoqxo 2.: mo ogﬁosom ”Tm oSwE

93 50.5
9:8 .5 Gown—cm

£803?“ woo—Em
E one—«5.35 Z

 

 

 

3&qu mad—w aux—A

 

 

 

 

$55020 EEG—SE.

 

39: :38 3. > m:
o o
_

 

 

 

 

. i r. 7 A .-- ..9 . 1| I.i I. .r.,,
I . -11 l l .-...igﬂ : 11.1- ’11. - 1 O O

nouuuocow Kt.»
cowocsm >395 uoBom 0Q \ avg—95w aa—ommm

 

 

 

 

195

Reaction Chamber

The test cell setup consisted of a 1 liter Pyrex glass beaker with a Teﬂon cap. These
materials of the cell were selected to eliminate or reduce sorption of the contaminants
onto the walls of the reaction vessel and the cap. Titanium plates were used as electrodes
in electrochemical tests of this study. Alshawabkeh and Sarahney (2005), Li et al. (2005),

and Goel er al. (2003) have used titanium as a base metal and coated it with mixed metal
oxides such as Rqu, erz, and SnOz. The titanium plates used in this study were not

coated. Uncoated titanium being cheaper than commonly used electrodes such as
platinum, or titanium coated electrodes, was selected for this project. Also, titanium
offers resistance to corrosion in a wide range of aggressive conditions. The titanium
sheets used as electrodes in this study were 12.0-cm-long by 5.0-cm-wide by 1.72-mm-
thick and the distance between them was maintained equal to 8.0 cm. The purity of the

titanium was about 99%.

Electrical Equipment

A 0.1 Hz to 2 MHz Sweep Function Generator was used to generate the AC signals. The
power supply consisted of a bipolar operational power supply/ampliﬁer capable of
producing up to 200 V AC/DC and 1 A AC/DC output or a maximum electrical power of
200 W. Voltage and current measurements were obtained with the aid of an oscilloscope

with accessories.

Spiked Sand
Aqueous solutions, spiked with naphthalene were prepared by adding naphthalene (98%,

purity) crystals to one-liter of deionized (DI) water in a capped Pyrex bottle and allowing

196

the suspension to stand for up to three days. This was followed by the addition of 0.5 g of

anhydrous sodium sulfate (Na2SO4) to each suspension, to increase its electrical

conductivity. Sodium sulfate was selected as the supporting electrolyte because all

aqueous solution tests without sand were carried out using NaZSO4 as the supporting

electrolyte. The target concentration of N32804 was approximately 500 mg/L (~3.5

mM). This concentration was selected to meet the USEPA’s maximum contaminant level

goal for sulfate equal to 500 mg/L (~3.S mM). The resulting spiked aqueous solution was

added to 1,000 cm3 of dry Ottawa sand placed in a l-liter glass beaker until the sand was

fully saturated. It was assumed that saturation was attained when a thin ﬁlm of the spiked

solution formed on the surface of the sand in the glass beaker.

Testing Procedure

Tests were conducted in batch mode in l-liter undivided cells. Prior to applying AC to
the cells, initial samples were collected for high performance liquid chromatography
(HPLC) analyses For each sampling event, three aqueous samples were collected, one
from the middle of the test cell and one each close to the surface of the electrodes, with
the aid of a glass syringe and needle. During the tests, samples were collected for HPLC
analyses at elapsed time intervals of O, 1, 2, 4, 8, 24, and 48 hours. The samples were
analyzed directly without extracting the naphthalene from the aqueous solution. HPLC
analyses were performed using a diode array detector (detection at 255 nm), a Waters C-
18 PAH, 250 mm x 4.6 mm x 5 pm column, an automated gradient controller, and
deionized water (DD/acetonitrile mobile phase. The ﬂow rate of the mobile phase was 1.2

mL/min and the column was ﬂushed with 10:90 (v/v) acetonitrile/DI water for 5 minutes

197

followed by 60:40 (v/v) acetonitrile / DI water for 10 minutes and by 100:0 (v/v)
acetonitrile / DI water for 5 minutes. This gradient was used for the analysis in order to
obtain appropriate separation of the degradation product peaks on the chromatograms

during the tests. Standard solutions were prepared to bracket the expected concentration

level of the analytes. Calibration curves with an average obtained R2 value of 0.998 were

used to determine the concentration of naphthalene in the solutions as the tests

progressed.

RESULTS AND DISCUSSION

Comparison of Degradation Rates
Table 5-1 shows the experimental variables.

Table 5-1: Experimental Variables

 

 

 

 

 

 

 

 

 

 

Medium AC Wave AC Peak Current AC No. of
Form Density Frequency Tests
[mA/cm’] [Hz]

Solution Square 3 0.1 2
Sand Square 3 0.1 1
Sand Sine 6 0.1 1
Sand Sine 6 60 1
Sand Sine 10 60 1
Sand DC 3 0 1
Sand Control 0 0 1

Solution Control 0 O 2

Total No. of 11
Test

 

 

 

 

 

198

Figure 5-2 shows a comparison of the degradation of naphthalene in spiked

aqueous solution and with and without Ottawa sand using square wave AC. The peak

current density and frequency were equal to 3 mA/cm2 and 0.1 Hz, respectively.

Normalized Naphthalene Concentration, CICo

1.4

1.2

0.2

 

 
 
  

 

 

...........................................................

. u : : : i '
'5: =' : : : : :IControl (Send)

 

control (Solution)

....................................................................

center of Cell (Send)

 

co - 1e - 2o mglL ‘
b Supporting Electrolyte: Na2 SO iiiiii .............. 3 a 5'99”” (39“)
4 i ' '3 Center of Cell (Solutlon)

AC: Square Wave, f I 0.1 Hz

1 I I I I | I I I I I I I I I I L I I I l I I I I l I I I I ,

 

0 10 20 30 40 50 60
Elapsed Time [hrs]

Figure 5-2: Comparison of electrochemical degradation of naphthalene in spiked solution
with and without sand using AC.

The control tests were carried out without the passage of AC or DC while the rest

of the experimental setup was similar to other tests where AC or DC was passed. In

addition, the control cell for sand was placed in a static temperature chamber at 44 °C to

simulate a conservative scenario of heating. When AC or DC was applied, while the

199

water bath kept the temperature rise to a minimum, it reached 32.9 °C when DC was
applied for the cell containing sand. For other tests, the temperature was less than 21 °C.
DC test produced higher temperature because the DC resistance of the system was greater
than the impedance when AC was applied. Hence, greater electrical power was utilized in
the DC cell for a ﬁxed current.

Aqueous phase samples collected close to the surface of the electrodes generally
exhibited faster degradation rate than samples collected from the center of the cell when
sand was present. This result is consistent with the general belief that electrochemical
reactions occur on the surface or in the vicinity of the electrodes. Samples collected in the
vicinity of both electrodes showed signiﬁcant degradation. Pepprah and Khire (2008)
reported that in AC electrolysis, constant polarity reversal of the electrodes ensures that
both electrodes participate in the anodic oxidation process. By applying an AC, which
resulted in constant polarity reversal of the electrodes at a rate dependent upon the AC
frequency, anodic oxidation of naphthalene in solution occurred on both electrodes.
Previous studies on electrochemical degradation have focused on the use of DC where
the cathode acts as a “passive” electrode in the oxidation process. However, when an AC
is used, the constant polarity reversal creates anodic oxidation reactions on both
electrodes alternately, depending on their polarity in a given AC cycle. During half cycle
when one of the electrodes has a positive electrical potential, it instantaneously acts as the
anode (instantaneous anode) while the other electrode, at a negative electrical potential
acts instantaneously acts as the cathode (instantaneous cathode). In the subsequent half

cycle, polarities are reversed and the electrodes acquire electrical potentials opposite to

200

what they had in the previous cycle and this continues throughout the period of
application of the AC.

As shown in Figure 5—2, the control cell samples did not indicate signiﬁcant
reduction in concentration during the 48-hr testing period.

Figure 5-3 shows a comparison of the normalized naphthalene concentration

samples with and without sand subjected to DC having current density equal to 3

 

   
   
 
 
 

 

2
mA/cm .

1.4 p . I I fr 1 j T I I ! I I ! I I I .

1.2 ........... ........ .............. .......... ............ ..

_ 1 '0 '1 : : ' : : .

_ 3' f f 5 Control (Sand) 3 .

1 ~ . ------- 3 ------------- a ----- e --------- 2 ------------- 3 ------------- g ------------ j

1‘ 3 3 Control (Solution) g .

-= s 2 2 s““‘<92 -

0-3 """ E """" ' """ E """""""" 3 """"""""" 5 """""""" 5 """" '

00318-20mgIL :

0.6 """"';‘”"“““3"supponrngauoctroryto:Nazso; .

 

 

j = 3 mAIcmz (DC). ..

Normalized Naphthalene Concentration, CICo

 

 

 

 

0.4
"'2 i i """""" """"""" '2
I n 4
0
0 1o 20 so 40 50 so

Elapsed Time [hrs]

Figure 5-3: Comparison of electrochemical degradation of naphthalene in spiked solution
and sand using DC

Aqueous phase samples collected in the vicinity of the anode of the cell with sand

exhibited faster degradation of naphthalene, comparable to samples collected from the

201

middle of the test cell in the aqueous solution tests. Samples collected in the vicinity of

the cathode did not showed less degradation because in DC electrolysis, anodic oxidation

occurs only on one electrode (the anode). Hence, unlike in an AC test, the cathode in DC

test does not participate in the oxidation of the organics in the aqueous phase.

Comparison of Degradation Rates in Spiked Sand Using AC and DC

Figure 5-4 shows the normalized concentration of naphthalene for cells containing the

sand where AC and DC having current densities equal to 3 mA/cm2 were passed.

Normalized Naphthalene Concentration, CICo

1.4

1.2

0.8

0.6

0.4

0.2

C0818-20mglL

 

 

 

 

I . 1 . I t s . . l . . . u I . Supporting Electrolyte: NaZSO4-I-

- : I I AC: Square Wave,f-0.1 Hz.
- : i‘ O : -
--; ”a I - ; 1:3 mAIcm (ACIDC) -
- f9- ; . 3 i 2 :
: b: s .
'4“) _______________________________________________________ e,Control__,__
‘ i s: ‘ -
. i: ‘k . a
f , ...................................................................... ‘
. AC Center 4
_. ........................................................... -: ............ .
P :: ~
“E ----------- - Cathode (Sand) ---------------------------------------------- ‘
' Center (Sand) ' '
P ' : : : : : '
b-Anode (Sand); ----------- j -------------- g ------------- f ---------- 0 Ac Electrodes‘
. i l l l l i l l l l i l .

 

 

 

 

0 10 20

 

30 40 50 60

Elapsed Time [hrs]

Figure 5-4: Comparison of electrochemical degradation of naphthalene in spiked sand
using AC and DC

202

 

Samples collected from the vicinity of both the electrodes in the AC tests exhibited
degradation. Piowever, samples collected from the vicinity of the anode showed much
faster degradation than those collected in the vicinity of the cathode in the DC test. Based
on the observed concentration proﬁle, it is likely that anodic oxidation occurred only at
the anode, and as naphthalene was being oxidized, more molecules moved from the
center of the cell as well as in the vicinity of the cathode, towards the anode where they
were subsequently oxidized. Table 5-2 shows that samples collected at the end of the test
indicated acidic/oxidizing conditions at the anode and alkaline/reducing conditions at the

cathode.

203

 

 

 

 

 

 

 

 

 

 

 

 

EQ
CQ Sm CQ m8 2:
a $2 2ng _ 2:3 Q Q
Q .32 Q 3% Q E Q an». Q 3 Q E q: 2N o m on 28
am: 82 am: a8 am: .2
2m: Sum :5 3m :8 <2 Q Q
Q 2: Q o3 Q 20 Q as Q 3 Q 3 3m 92 8 2 05m 28
ammo? am: 5. am: 3
:8 N3 :5 S... :6 ma Q Q
Q as Q 8w Q 3m Q 8m Q 3 Q 3 3a 0.2 8 e 25 25m

am: 89. a Q So am: 3
cm: as: :8 so 2.“: 3 Q Q
Q 8% Q g Q 2m Q 2v Q :2 Q as a: 3m 3 a 05m 2%

am: no

:8 we Q Q
Q 8E Q 5 Q a: Q a? Q n: Q we mam SN 3 m 28% 2%

g 6%
Q as Q :w Q 3. Q can Q 3 Q 3 3: N2 3 m 22cm 8:28
:5... 3:5 :5... 3:5 as... 3...: .25. 3:..—
$52....
2.53 9.5 5: E25 as...
5332655 32.82— AUoV 55:92,..— Eutau 26>?
.8395 5.x: 2.65% E. 2322.5... o< v.3.— o< u< £5.52

 

 

 

 

 

 

 

 

2.8 38 .8 53:26:00 185020 van .3358“. x82 Envy—9m in £38358 no 823 :25 can REE @8332 &.m 2an

204

”6.3 N 03500—0 no .3503 0.: E 380:3 0388 n N m
”Gad _ 03500.0 :0 $503 0.: E 380:8 0.93m n _ m
AUDV 03:80 05 :o .3503 0.: E 380:3 038% u HQ

”:00 80“ 05 no 8280 05 So: 380:3 038% n U
AUQV 0:28 «o .9503 2: E 380:3 2ng n <

 

 

80qu
Q Q

Qm; Q:w Q3. 6;? Q; Q; 0.2 3: 3:8 8:28
Q Q

Q 22 Q Sm Q Em Q a? Q A; Q 2 <3. 98 380 2.8

 

 

 

 

 

 

 

 

 

 

 

 

380 3 22¢

 

205

This observation is consistent with a review presented by (Page and Page (2002). Page

and Page (2002) reported that in DC electrolysis of moist soils, the predominant process

. . . . . +
is the electrolysrs of water molecules, Wthh results in the formation of H and oxygen at

the anode and OH- and hydrogen at the cathode. Therefore under the condition of

hindered migration in soils, the immediate result is a local change in pH of the pore
solution so that acidity increases with time at the anode and alkalinity increases at the

cathodes.

Effect of Current Density

Because AC from the power grid is supplied as a sine wave at 60 Hz frequency in the
United States, tests were conducted using sine wave at 60 Hz frequency to explore the
potential for pilot-scale application. If AC available from the power grid can be used for
this application, it would be cheaper for capital cost compared to DC. DC requires power
supplies which are relatively expensive for power demand required to run pilot-scale or

field-scale electrochemical tests. Figure 5-5 illustrates normalized concentration versus

time proﬁle for tests conducted using 60 Hz AC sine wave at 6 mA/cm2 and 10 mA/cm2

peak current densities.

206

 

14 [IIWﬁIIITI'IIII'IIIIIIIII'IIII
O

Irv...‘llvv-v-<—--v<u§104‘-—4<liiv4<: uuuuuuuuuuuu q

-'Control ----- -

............................................. .4

 

...........................................

Electrodes (6 mAlcm )
. I

    

a Center of Cell (6 mAIcm )

0.4 -.E.._.. ........... ' .............. :L ............ 5........—.....;. .......... .
C =18mgIL : 1 :
30 3 ' ' '

 

2
Center of Cell (10 mAIcm )
~ I

02 "Supporting Etectrotyte: $52304 """ """"""" """"""

Normalized Naphthalene Concentration, CIC

2
Electrodes (10 mAlcm )

EAC: Sine dee, f= 60 Hz : : : I
l I I I I L I I I I I I L I I I I I I I . I I I I l I I I g

0 10 20 3O 40 50 60
Elapsed Time [hrs]

 

Figure 5-5: Effect of current density on electrochemical degradation of naphthalene in
spiked sand using AC

The concentration of naphthalene decreased near the electrodes and away from the
electrodes as the tests progressed. Increasing the current density from 6 mA/cm2 to 10
mA/cm2 resulted in an increase in the degradation rate. This is similar to the trend
reported by Pepprah and Khire (2008) that the rate of degradation of naphthalene in

spiked aqueous solution increased with increased current density for both AC and DC.

Effect of AC Frequency

Figure 5-6 illustrates normalized concentration versus time proﬁle of tests conducted

using 6 mA/cm2 peak AC density at 0.1 Hz and 60 Hz frequency.

207

 

104 I I1 I I I I 1 r]. T I I I l I I I I ‘1 roII I I I I

 

Locatlon (frequency):

 

Electrodes (60 Hz)
\7 Center of Cell (0.1 Hz)

g ‘ Center of Cell (60 Hz)
0.4 h ------------------------------ - --------------- 3 -------------- . ---------- : ------------ ...

co = 18 mg/L g =3 Electrodes (0.1 Hz)

 

0,2 —-SUPporting Electrolyte: NaZSO4

Normalized Naphthalene Concentration, 0100

AC: Sine Wave,j= '6mA/cm2

o IllLlllllllJlJl‘llJJ'ljlllllJl

0 10 20 30 40 50 60
Elapsed Time [hrs]

 

 

 

Figure 5-6: Effect of AC frequency on electrochemical degradation of naphthalene in
spiked sand using AC

Degradation rate of naphthalene for the 0.1 Hz frequency test was greater than that at 60
Hz frequency. This is consistent with the ﬁndings by Chen and Chin (1985) who reported
that degradation rate of phenol in solution decreased with increased AC frequency

superimposed on a DC.

Experimental Parameters Monitored
Apart from monitoring the aqueous phase concentration of naphthalene in the spiked sand

over time, other experimental parameters monitored included the initial and ﬁnal values

208

of pH, temperature, redox potential, and electrical conductivity. A portable dissolved
oxygen/pH meter, equipped with a pH electrode, and oxidation-reduction probe, was used
to measure pH and redox potential respectively, of the test solutions. An electrical
conductivity meter was used to measure the electrical conductivity and a thermometer
was used to measure the temperature of the spiked solution in the control and test cells

throughout the testing period. Table 5-2 shows the various measurements.

SUMMARY AND CONCLUSIONS

AC and DC were separately applied to spiked naphthalene solutions with and
without Ottawa sand to assess degradation. Potential application of this technology would
be for cleanup of PAH contaminated ground water in sandy or rocky aquifers. Square
wave AC having frequency equal to 0.1 Hz was used. The current density for both AC
and DC was 3 mA/cmz. The key ﬁndings of this study are: (1) both DC and square wave
AC degraded naphthalene in the aqueous phase of the sand; (2) at an equivalent current
density, the rate of degradation of naphthalene in the aqueous phase when sand was
present was less than when only the aqueous solution were subjected to AC or DC; and

(3) rate of degradation for AC was less than that for DC at an equivalent current density.

209

SUMMARY AND CONCLUSIONS

The key objectives of this research dissertation were to: (1) demonstrate that alternating
current (AC) can be used for electrochemical degradation of PAHS (naphthalene,
phenanthrene and pyrene) in spiked aqueous solutions; (2) investigate the effect of AC
characteristics such as shape of the waveform, current density, and frequency on the
degradation rates when only the aqueous solution of the PAH versus the spiked solution
added to a clean coarse sand were subjected to AC electrolysis; (3) compare the rate of
degradation for naphthalene (a model PAH), in aqueous solution using ultrasound energy
and electrical energy (AC and DC) delivered at the same output power, and (4) identify
degradation mechanisms and pathways. Laboratory-scale tests were conducted in order to
achieve the objectives listed above.
The key ﬁndings of this study are as follows.

1. Similar to DC, AC also can degrade PAHS in free standing solutions or when
these solutions are mixed in sandy media;

2. DC at a given current density, was more efﬁcient in degrading PAHS than an
equivalent AC. However, it is expected that pilot-scale or ﬁeld-scale
application using AC would be potentially cheaper than DC for capital cost
when AC from the power grid is used because no rectiﬁcation is required;

3. Rate of degradation increases when the current density is increased;

4. Rate of degradation decreases when the frequency of the AC is increased. DC
has the lowest frequency (f = 0) and achieves the highest rate of degradation

for an equivalent input power;

210

10.

11.

For square, sine, and triangular AC waveforms explored, square wave AC
signal produced the fastest degradation rate for a given current density of the
AC. Thus, the rate of degradation is a function of the rms input power;
Degradation kinetics observed in the tests could be ﬁtted using pseudo-ﬁrst-
order decay model all PAHS tested;

Mass loss of electrodes during AC electrolysis was signiﬁcantly less than
during DC electrolysis for an equivalent power input;

Oxidation of PAHS occurred on both electrodes in AC electrolysis whereas, it
occurred on only one electrode (the anode) in DC electrolysis. This indicates
that in AC electrolysis, each of the electrodes acted as instantaneous anodes
alternately during the half cycle of the AC where the oxidation took place;
Both DC and ultrasound, at an equivalent power output, exhibited comparable
degradation rates of naphthalene in solution unlike AC at 60 Hz frequency,
which produced signiﬁcantly less degradation rate at the same power output.
This indicates that, compared to DC, while there will be saving in the capital
cost when AC from the power grid is used, the treatment time would be longer
when AC is used.

Formation of 2,3—dihydroxybenzoic acid and 2,5-dihydroxybenzoic from the
oxidation of salicylic acid (a hydroxyl radical probe) is consistent with
reported ﬁndings that OH radicals were the primary oxidizing agents in both
sonochemical and electrochemical degradation; and

Both diffusion and migration processes were responsible for the rate of mass

transfer of PAHS to the reaction sites (surface of electrodes or in the vicinity

211

of the electrodes). However migration was the dominant mass transfer

process.

212

REFERENCES

Acar, Y. B., Alshawabkeh, A. N. (1996). Electrokinetic remediation. 1. Pilot-scale tests
with Pb-spiked kaolinite. J. Geotech. Eng. 122(3), 173-185.

Acar, Y. B., Gale R. J., Putnam, G. A. (1991). Acid/base distributions in electrokinetic
soil processing. Transp. Res. Rec. 1288, 23-34.

Adewuyi, Y. G. (2001). Sonochemistry: Environmental science and engineering
applications, Ind. Eng. Chem. Res., 40, 4681-4715.

Agard, J. B. R. (1981). A study of electro-reclamation and its application to the removal
of toxic metals from contaminated soils. MSc. thesis, Manchester Univ.,
Manchester, England.

Agency for Toxic Substances and Disease Registry (ATSDR) (1995). Toxicological
proﬁle for polycyclic aromatic hydrocarbons. Atlanta, GA: US. Department of
Health and Human Services, Public Health Service.

Alshawabkeh A. N., Sarahney (2005). Effect of current density on enhanced
transformation of naphthalene. Environ. Sci. Technol. 39, 5837-5843.

Anderson, R. F., Patel, K. B., Stratford, M. R. L. (1987). Radical spectra and product
distribution following electrophilic attack by the OH radical on 4-hydroxy-benzoic
acid and subsequent oxidation. J. Chem. Soc. Faraday Trans., 83, 3177.

Bamforth, S. M. and Singleton, l. (2005). Review: Bioremediation of polycyclic aromatic
hydrocarbons: Current knowledge and future directions. J. Chem. Technol .
Biotechnol., 80(7), 723-736.

Banerjee, S. (1987). Electro—decontamination of chrome-contaminated soils. Proc., 13th
Annual Research Symp.: Land Disposal, Remedial Action, Incineration &
Treatment of Hazardous Waste, EPA, Cincinatti, 193-200; Rep. No. EPA/600/9-
87/0115, US. Environmental Protection Agency, Cincinatti.

Bard, A. J., Faulkner, L. R. (2001), Electrochemical methods: fundamentals and
applications, John Wiley & Sons Inc., New York.

Basu D. K., Saxena J. (1978a). Monitoring of polynuclear aromatic hydrocarbons in
water: 11. Extraction and recovery of six representative compounds with

polyurethane foams. Environ. Sci. Technol. 12, 791-795.

Basu D. K., Saxena J. (1978b). Polynuclear aromatic hydrocarbons in selected U.S.
drinking waters and their raw water sources. Environ. Sci. Technol. 12, 795-798.

213

Birke, V., Burmeier, H., Rosenau, D. (2003). Design, Construction, and Operation of
Tailored Permeable Reactive Barriers. Pract. Periodical of Haz., Toxic, and
Radioactive Waste Mgmt., 7 (4), 264-280.

Bjorseth, A., Ramdahl, T. (1985). Handbook of Polycyclic Aromatic Hydrocarbons. Vol.
2, eds. A. Bjorseth & T. Ramdahl, Marcel Dekker: New York, pp. 1-20.

Brillas, B., Calpe, J. C., and Casado, J. (2000). Mineralization of 2,4-D by advanced
electrochemical oxidation processes. Water Research, 34(8), 2253-2262.

Buxton, G. V., Greenstock, C. L., Helman, W. P., Ross, A. (1988). Critical review of rate
constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals
(OH/O") in aqueous solution. J. Phys. Chem. Ref. Data, 17, 513-886.

Cerniglia, C. E. (1984). Microbial degradation of polycyclic aromatic hydrocarbons, Adv
Appl Microbiol. 30, 351-368.

Cemiglia, C. E. (1992). Biodegradation of polycyclic aromatic hydrocarbons.
Biodegradation, 3(2-3), 351-368.

Chen, G. H. (2004). Electrochemical technologies in wastewater treatment, Sep. Purif.
Technol., 38, 11-41.

Chen. W. P., Wang, Y., Wang, X. X., Wang, J., and Chan, H. L. W. (2003). Water-
induced DC and AC degradation in TiOz-based ceramic capacitors. Materials
Chemistry and Physics, 82(3), 520-524.

Chiang, L. C., Chang, J. B., and Wen, T. C. (1995). Indirect oxidation effect in
electrochemical oxidation treatment of landﬁll leachate. Water, Research 29(2),
671-678.

Chin, D-T, and Cheng, C. Y (1985). Oxidation of phenol with AC Electrolysis. J.
Electrochem. Soc., 132 (1 1), 2605-2611.

Clar, E. (1964). Polycyclic hydrocarbons. Academic Press, New York.
Cole R. H., Frederick R. E., Nealy R. P., et al. (1984). Preliminary ﬁndings of the priority
pollutant monitoring project of the nationwide urban runoff program journal. J.

Water Pollut. Control Fed 56, 898-908.

Comninellis, C., De Battisti, A. (1996). Electrocatalysis in anodic oxidation of organics
with simultaneous oxygen evolution. J. Chim. Phys. 93, 673-679.

Comninellis, Ch. (1994). Electrocatalysis in electrochemical conversion/combustion of
organic pollutants for waste water treatment. Electrochim. Acta 39, 1857-1862.

214

Comninellis, Ch., Pulgrin, C. (1991). Anodic-oxidation of phenol for wastewater
treatment, J. Appl. Electrochem. 21, 703-708.

Elovitz, M. 8., von Guten, U. (1999). Hydroxyl radical/ozone ratios during ozonation
processes 1, Ozone Sci. Eng., 16, 113-120.

Fedkiw, P. S., Chao, J. C. (1985). electivity modiﬁcation in electroorganic reactions by
periodic current control: Electroreduction of nitrobenzene.” AIChE Journal, 31(9),
1578-1580.

F eng, Y. J., Li, X. Y. (2003). Electrocatalytic oxidation of phenol on several metal-oxide
electrodes in aqueous solution, Water Res. 37, 2399-2407.

Fernandez, P., Vilanova, R. M., Martinez, C., Appleby, P., and Grimalt, J. O. (2000). The
historical record of atmospheric pyrolytic pollution over Europe registered in the
sedimentary PAH from remote mountain lakes. Environ. Sci. Tech. 34(10), 1906 -

1913.

Flint, E. B., Suslick, K. S. (1991). The temperature of cavitation, Science 253, 1397-
1399.

Gattrell, M., D.W. Kirk, D. W. (1993). A study of the oxidation of phenol on platinum
and preoxidized platinum surfaces, J. Electrochem. Soc. 140, 1534-1540.

Gibb, H. J. (1966). Research on electroremediation of saline-alkali soils. Trans. Am. Soc.
Agricultral Eng. 9(2), 164-169.

Goel, R. K., Flora, J. R. V., Fen'y, J. (2003). Mechanisms of naphthalene removal during
electrolytic aeration. Water Res. 37, 891-901.

Goskonda, S., Catallo, W. J. Junk, T. (2002). Sonochemical degradation of aromatic
organic pollutants. Waste Management 22, 351-356.

Grimm, J ., Bessarabov, D., Sanderson, R. (1998). Review of electro-assisted methods for
water puriﬁcation. Desalination 115, 285-294.

Grimmer, G. (1983). Environmental Carcinogens: Polycyclic Aromatic Hydrocarbons.
CRC Press: Boca Raton, Fla.

Guo, Z., Zheng, Z., Zheng, S., Hu, W., Feng, R. (2005). Effect of various sono-oxidation
parameters on the removal of aqueous 2,4-dinitrophenol. Ultrasonics
Sonochemistry 12, 461-465.

Gupta, N., Sass, B. M., Gavaskar, A.R.; Sminchak, J.R.; Fox, T.C.; Snyder, F.A.;

O‘Dwyer, D.; Reeter, C. (1998). Hydraulic Evaluation of a Permeable Barrier
Using Tracer Tests, Velocity Measurements, and Modeling. Designing and

215

Applying Treatment Technologies: Remediation of Chlorinated and Recalcitrant
Compounds. Battelle Press, Columbus, OH. 157-162, 1998.

Haag, W. R., Yao, C. C. D. (1992). Rate constants for reaction of hydroxyl radicals with
several drinking water contaminants, Environ. Sci. Technol., 26(5), 1005-1013.

Hamed, J., Acar, Y. B., Gale, R. J. (1991). Pb(II) removal from kaolinite by
electrokinetics. J. Geotech. Eng. 1 17(2), 241-271.

Hamnett, R. (1980). A study of the process involved in electoreclamation of
contaminated soils. MSc. thesis, Manchester Univ., Manchester, England.

Harvey, R. G. (1997). Polycyclic Aromatic Hydrocarbons, Wiley-VCH, Inc., New York.

Harvey, R.G. (1991). Polycyclic aromatic hydrocarbons; Chemistry and carcinogenicity.
Cambridge University Press, New York.

Ho, 8. V., Athmer, C. J., Sheridan, P. W., Shapiro, A. P. (1997). Scale-up aspects of
Lasagna (TM) process for in situ soil decontamination. J. Haz. Mat. 55, 39-60.

Houk, L. L.,‘ Johnson, S. K., Feng, J. R., Houk, R. S., Johnson, D. C. (1998).
Electrochemical incineration of benzoquinone in aqueous media using quaternary

metal oxide electrode in the absence of a soluble supporting electrolyte. J. Appl.
Electrochem. 28(11), 1167-1177.

Hung, H.-M., Kang, J .-W., Hofﬁnann, M. R. (2002). The sonolytic destruction of methyl
tert-butyl ether present in contaminated groundwater. Water Environ. Res. 74, 545-
556.

lniesta, J ., Gonzalez-Garcia, J ., Exposito, E., Montiel, V., Aldaz, A. (2001). Inﬂuence of
chloride ion on the electrochemical degradation of phenol in alkaline medium using
bismuth doped and pure PbOz anodes, Water Res. 35, 3291-3300.

Iniesta, J., Michaud, P. A., Panizza, M., Cerisola, G., Aldaz, A., Comninellis, Ch.
(2001). Application of synthetic boron-doped diamond electrode in electro-
oxidation and in electro-analysis, Electrochim. Acta 46 (2001) 3573-3578.

Jen, J .-F., Leu, M.-F., T.C. Yang, T. C. (1998). Determination of hydroxyl radicals in an
advanced oxidation process with salicylic acid trapping and liquid chromatography,
J. Chromatogr. A, 796(2), 283-288.

Johnson, A. R., Wick, L. Y., Harms, H. (2005). Principles of microbial PAH-degradation
in soil. Environmental Pollution 133: 71-84.

Kappana, A. N. and Joshi, K. M. (1952). Part I. Oxidation of glucose to gluconic acid.
Survey of techniques. J. Ind, Chem, Soc., 50, 331-338.

216

Kim, D. K., O’Shea, K. E., Cooper, W. J. (2002). Degradation of MTBE and related
gasoline oxygenates in aqueous media by ultrasonic irradiation. J. Environ. Engr.
128, 806-812.

Kirk, D. W., Shariﬁan H., Foulkes, F. R. (1985). Anodic oxidation of aniline for waste
water treatment, J. Appl. Electrochem. 15, 285-292.

Ko, S., Schlautman, M. A., Carraway, E. R. (1998), Effects of solution chemistry on the

partitioning of phenanthrene to sorbed surfactants. Environmental Science
Technology, 32(22), 3542-3548.

Kreysa, G., Heitz, E. (1986). Principles of Electrochemical Engineering. VCH,
Weinheim, New York.

Lageman, R. (1993). Electro-reclamation (applications in the Netherlands). Environ. Sci.
Technol. 27(13), 2648-2650.

Lageman, R., Pool, W., Sefﬁnga, G. A. (1990). Electro-reclamation: State-of-the-art and
future developments. Contaminated Soil ’90, F. Arendt, M. Hinsenveld and W. J.
van den Brink, eds., Kluwer Academic, Dordrecht, The Netherlands, 1071-1078.

Lagemen, R., Pool, W., Sefﬁnga, G. (1989). Electro-reclamation: Theory and practice.
Chem. Ind. 18, 585-590.

Laor, Y., Farmer, W. J., Aochi, Y., Strom, RF. (1998), Phenanthrene binding and
sorption to dissolved and mineral-associated humic acid. Water Research, 32(6)
1923-1931.

Lee, M. L., Novotny, M. V., and Battle, K. D (1981). Analytical chemistry of polycyclic
aromatic compounds. Academic Press, New York.

Li X-Y., Cui, Y-H., Feng, Y-J., Xie, Z-M., Gu, J-D. (2005). Reaction pathways and
mechanisms of electrochemical degradation of phenol on different electrodes.
Water Res. 39, 1972-1981.

Li, Y., Wang, F., Zhou, G., Ni, Y. (2003). Aniline degradation by electrolytic oxidation,
Chemosphere 53(10), 1229-1234.

Lindgren, E. R., Mattson, E. D., Kozak, M. W. (1994). Electrokinetic remediation of
unsaturated soils. ACS Symp. Ser. 554, 33-50.

Little, C., Hepher, M. J., El-Sharif, M. (2002). The sono-degradation of phenanthrene in
an aqueous environment. Ultrasonics 40, 667-674.

217

Lorimer, J. P., Mason, T. J. (1987). Sonochemistry part 1 — the physical aspects, Chem.
Soc. Rev. 16, 239.

Lund, H., Baizer, M. M. (1991). Organic Chemistry. Marcel Dekker, New York, p. 616.

Marcelli, B.,Garcia-Gomez, J ., Michaud, P.-A., Rodrigo, M. A., Comninellis, Ch. (2003).
Electrogeneration of hydroxyl radicals on boron-doped diamond electrodes. J.
Electrochem. Soc. 150, D79-D83.

Meegoda, J. N., Veerawat, K. (2002). Ultrasound to decontaminate organics in dredged
sediments, Soil and Sediment Contamination 11(1), 91-116.

Morselli L, Zappoli S. (1988). PAH determination in samples of environmental interest.
Sci. Total Environ. 73, 257-266.

Mueller, J. G., Cemiglia, C. E., Pritchard, P. H. (1996). Bioremediation of environments
contaminated by polycyclic aromatic hydrocarbons, in Bioremediation: Principles
and Applications, ed by Crawford R. L. and Crawford, D. L. Cambridge University
Press, Idaho, 125-194.

Nakamura, A., Hirano, K., and lji, M. (2005). Decomposition of trichlorobenzene with
different radicals generated by alternating current electrolysis in aqueous solution.
Chemistry Letters, 34 (6), 802-803.

Naumczyk, J., Szpyrkowicz, L., and Zilio-Grandi, F. (1996). Electrochemical treatment
of textile wastewater. Wat. Sci. Tech., 34 (11), 17-24.

Ottinger, S. E., Mayura, K, Lemke, S. L., McKenzie, K. S., Wang, N. Kubena, L. F .,
Phillips, T. D. (1999). Utilization of electrochemically generated ozone in the
degradation and detoxiﬁcation of benzo[a]pyrene. Journal of Toxicology and
Environmental Health Part A, 57(8), 565-583.

Oturan, M. A. (2000). An ecologically effective water treatment technique using
electrochemically generated hydroxyl radicals for in situ destruction of organic
pollutants: Application to herbicide 2,4-D. J. Appl. Electrochem. 30(4), 475-482.

Oturan, M. A. and Pinson, J. (1995). Hydroxylation by electrochemically generated OH'
radicals. Mono- and polyhydroxylation of benzoic acid: Products and isomers’
distribution. J. Phys. Chem, 99, 13948-13954.

Page, M. M., Page, C. L. (2002). Electroremediation of contaminated soils. J. Environ.
Engr. 128(3), 208-219.

Pamakcu, S., Wittle, J. K. (1994). Electrokinetically enhanced in situ soil

decontamination. Remediation of Hazardous Waste Contaminated Soils, D. L.
Wise, and D. J. Trantolo, eds., Marcel Dekker, New York, 245-298.

218

Panizza, M., Cerisola, G. (2003). A comparative study of direct and indirect
electrochemical oxidation of polyaromatic compounds. Annali di Chimica 93, 977-
984.

Panizza, M., Bocca, C., Cerisola, G. (2000). Electrochemical treatment of wastewater
containing polyaromatic organic pollutants. Water Res. 34, 2601-2605.

Park, J. K., Hong, S. W., Chang, W. S. (2000). Degradation of polycyclic aromatic
hydrocarbons by ultrasonic irradiation, Environ. Technol. 21, 1317-1323.

Pepprah, E., Khire, M. V. (2008). Electroremediation of naphthalene in aqueous solution
using alternating and direct currents. J. Environ. Engr., ASCE, 134-1 (in press).

Polcaro, A. M., Palmas, S., Renoldi F., Mascia, M. (1999). On the performance of
Ti/Sn02 and Ti/Pb02 anodes in electrochemical degradation of 2-chlorophenolfor
wastewater treatment, J. Appl. Electrochem. 29, 147-151.

Psillakis, E., Telekom, A., Maintains, D., Nikolopoulos, E., Kalogerakis, N. (2003). J.
Environ. Mont. 5, 135-140.

Puri, A. N., Anand, B. (1936). Reclamation of alkali soils by electrodialysis. Soil Sci. 42,
23-27.

Rajeshwar, 1K., Ibanez, J. G., Swain, G. M. (1994). J. Appl. Electrochem 24, 1007.

Reddy, K. R., Saichek, R. E. (2004). Enhanced electrokinetic removal of phenanthrene
from clay soil by periodic electric potential application. J. Environ. Sci. & Health
A39(5), 1189-1212.

Riesz, P., Kondo, T., Khrishna, M. C. (1990). Sonochemistry of volatile and non-volatile

solutes in aqueous solutions: e.p.r. and spin trapping studies, Ultrasonics 28, 295—
303.

Rodgers, J. D., Jedral, W. J., N]. Bunce, N. J. (1999). Electrochemical oxidation of
chlorinated phenols, Environ. Sci. Technol. 33, 1453-1457.

Saracco G., Solarino, L., Aigotti, R., Specchia, V., Maja, M. (2000). Electrochemical

oxidation of organic pollutants at low electrolyte concentrations. Electrochim. Acta
46, 373-380.

Schwarzenbach, R. P., Gschwend, P. M., and Imboden, D. M. (2003). Environmental
Organic Chemistry. John Wiley & Sons, Inc., New Jersey.

Scheck, C. K., Frimmel, F. H. (1995). Degradation of phenol and salicylic acid by
ultraviolet radiation/hydrogen peroxide/oxygen, Water Res., 29 (10), 2346-2352.

219

 

Schumann, U., Grundler, P. (1998). Electrochemical degradation of organic substrate at
PbOz anodes: monitoring by continuous C02 measurements, Water Res. 32, 2835-
2842.

Segall, B. A., O’Bannon, C. E., Matthias, J. A. (1980). Electroosmosis chemistry and
water quality. J. Geotech. Eng. 106(10), 1148-1152.

Sehgal, C., Yu, T. J., Sutherland, R. G., Verrall, R. E. (1982). Use of 2,2-Diphenyl-l-
picrylhydrazyl to investigate the chemical behavior of free radicals induced by
ultrasonic cavitation, J. Phys. Chem. 86, 2982-2986.

Shiraishi H, Pilkington N. H., Otsuki A., et al. (1985). Occurrence of chlorinated
polynuclear aromatichydrocarbons in tap water. Environ. Sci. Technol. 19, 585-590.

Simond, 0., Schaller V., Comninellis, C. (1997). Theoretical model for the anodic
oxidation of organics on metal electrodes, Electrochim. Acta 42, 2009-2012.

Sorrel R. K., Brass H. J., Reding R. (1980). A review of occurrences and treatment of
polynuclear aromatic hydrocarbons. EPA-600/D-8l-066.

Staples C. A., Werner A. F., Hoogheem T. J., et a1. (1985). Assessment of priority
pollutant concentrations in the United States using STORET database. Environ.
Toxicol. Chem. 4, 131-142.

Stock N. L., Bunce, N. J. (2002). Electrocatalytic dechlorination of atrazine. Can. J.
Chem. 80, 200-2006.

Stucki, S., Kotz, R., Career, B., W. Suter, W. (1991). Electrochemical wastewater
treatment using high overvoltage anodes II: anode performance and application, J.
Appl. Electrochem. 21, 99-104.

Suslick, K. S., Flint, E. B. (1987). Sonoluminescence from non-aqueous liquids, Nature
330, 553-555.

Tanaka, F., Feng, C., Sugiura, N., Maekawa, T. (2004). p-nitrosodimethylaniline (RNO)-
based evaluation of enhanced oxidative potential during electrochemical treatment
of high-salinity wastewater, J. Environ. Sci. & Health, A39 (3), 773-786.

Thompson, L.H., Doraiswamy, L. K. (1999). Sonochemistry: science and engineering,
Ind. Eng. Chem. Rev. 38, 1215-1249.

Titanium and the Environment (Issue 1 April 2002). Data Sheet No. 19. Titanium
Information Group.

USEPA <http://www.epa.gov/safewater/mcl.html> (05/16/2006)

220

 

USEPA (2000). A Resource for MGP Site Characterization and Remediation. EPA/542—
R-OO-005.

Vadyunina, A. F. (1968). Meliorative effect on direct electric current on leaching of
solonchakous solonetz. Proc. Trans. of the 9th Int. Congress of Soil Science,
Adelaide, 455-463.

Wang, S., Huang, B., Lu, X., Liao, L. (2005). Sonochemical decomposition of organic
pollutants in landﬁll leachate. Fresenius Environmental Bulletin 14(5), 727-731.

Wang, S. L. (1995). The study and application of electronic pulse irradiation facility.
Doctoral dissertation, University of Science and Technology of China, Hefei.

Westerhoff, P., Aiken, G., Amy, G., Debroux, J. (1999). Relationships between the
structure of natural organic matter and its reactivity towards ozone and hydroxyl
radicals, Wat. Res. 33(10), 2265-2276.

Will, F. (1995). Removing toxic substances from the soil using electro-chemistry. Chem.
Ind., 15th May, 376-379.

Yasman, Y., Bulatov, V., Gridin, V. V., Agur, Sabina, Galil, N., Armon, R., Schechter
(2004). A new sono-electrochemical method for enhanced detoxiﬁcation of

hydrophilic chloroorganic pollutants in water. Ultrasonics Sonochemistry 11, 365-
372.

Yusem, G. J ., Pintauro, P. N. (1992). The Electrocatalytic Hydrogenation of Soybean Oil,
.1. Am. Oil Chem. Soc. 69, 399—404.

Zhaobing, G., Zheng, Z., Shourong, Z., Wenyong, H., Feng, R. (2005). Effect of various

sono-oxidation parameters on the removal of aqueous 2,4-dinitrophenol.
Ultrasonics Sonochemistry, 12, 461-465.

221