:4.
I. n
.3: .
It. {l'oln‘
I.
. Int-[‘1
v..V :7...ch
3 11....
on. 2.
.v.

V '5‘ .
‘0 f U- ¢ I
6 .Ir innit.
nuufvfl'unilllo’is

smmrfihfivnrvia
. 1e . inn.

..c.rv

. A. Eh...” I
LI .3: :flfll‘
hp. din I ”a! .
I. I ‘(crI’CJ3viu J.
n ‘ _. v: .3
i4 .
‘uprr aria”... Juliet Lulfhl
ill. iii... a 3%.
vual’lpfiflflvl ‘ ho. 11:503.!!! ‘1
i. a (I! . . bi"?! .
at: r I" . 01.!!! v Elatilllfl’

P . .fiifcup...
r‘ it... 1-,‘1

It.
.flllv: 1:4!
gillfio’buf‘l‘fibz
g, .-
kl...) u: 'a; .l’db‘ni'u'rl
it! «tiling. v
115%....»t‘ ,5 .Ol c
lvéutlllvv.t.A-.Iu.lvrol.'
15.7. II. {I} ‘ r. Cit 1:0
1!... in...

._...flr«!¥-v. 1:. It. v-.
)v. {tat}. .1. is!
lit» .31....va

Olin-III.» .
C I .
lat-C(51)»! ill-II

 

 

mvensm LIBRARIES

l““‘9‘“‘ ‘9‘“9‘ ‘9“‘9 “‘ “‘9‘9“‘9“‘ 1““‘1‘ “ “‘

 

‘

 

 

 

 

.‘I

This is to certify that the

thesis entitled

Oxidation of 1,3,5-Trichlorobenzene Using Ad-
vanced Oxidation Processes

presented by
Michael James Galbraith

has been accepted towards fulfillment
of the requirements for

M. S. de ree in Environmental
g Engineering

 

 

W 4am, W712

«ajor professor

 

Date 31 March 1993

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY
Michigan State

 

University

 

PLACE IN RETURN BOX to remove this checkout from your record.

TO AVOID FINES return on or before date due.

 

 

‘2'

 

DATE DUE DATE DUE DATE DUE

 

 

Lg

 

'f' ‘95 ' 4U
=mé
'. «rt (Q‘s?)

 

 

 

 

 

yaHPE/

 

 

 

 

————.————l—

 

 

 

 

ll

 

 

 

 

 

 

 

 

 

 

 

 

fl?

 

 

 

 

 

 

 

 

MSU le An Affirmative ActiorVEquel Opportunity institution

CWMH. '

OXIDATION OF 1,3,5-TRICHLOROBENZENE USING ADVANCED OXIDATION
PROCESSES

By

Michael James Galbraith

A THESIS

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

MASTER OF SCIENCE

Department of Civil and Environmental Engineering

1992

ABSTRACT

OXIDATION OF 1,3,5-TRICHLOROBENZENE USING ADVANCED OXIDATION
PROCESSES

By

Michael James Galbraith

A flow injection analysis (FIA) method for measuring dilute concentrations of
hydrogen peroxide in the presence of ozone is presented. A peroxidase catalyzed
oxidation of N,N-diethyl-p-phenylenediamine method was incorporated into a FIA
system for use in advanced oxidation studies. Tracer studies determining mixing
efficiencies of two commonly used photochemical reactors proposed for use in this
research are also presented.

The potential for using advanced oxidation treatment processes to remove 1,3,5-
trichlorobenzene (TCB) from aqueous solutions was investigated. Ozone, ozone/UV,
ozone/peroxide, and ozone/UV/peroxide treatments were compared. The effect of pH,
humic acid, and bicarbonate was investigated. At ozone and peroxide dosages of 6
mg/l and 60 uM, respectively, 99.6% of the TCB was oxidized in ten minutes,
exceeding the removal efficiency obtained using the other processes. No advantage is
gained by supplementing ozone with peroxide or UV light above a pH of 10.5.
Significant trichlorobenzene removal was still achieved with humic acid and

bicarbonate concentrations of 2 mg/l and 2 mM, respectively.

ACKNOWLEDGEMENTS

I wish to extend my appreciation to Dr. Susan J. Masten, Dr. Simon Davies, and
Dr. Mackenzie Davis for serving on my committee and for their advice,
encouragement, and professional guidance.

I also wish to thank Min Min Shu for the many hours of assistance with laboratory
work and gas chromatography analysis.

Funding for this research was provided by the Office of Research and
Development, US. Environmental Protection Agency under Grant R815750 to the
Great Lakes and Mid-Atlantic Hazardous Substance Research Center, and also by the

College of Engineering, Michigan State University.

iii

TABLE OF CONTENTS

List of Tables ................................................ vi
List of Figures ............................................... viii
I. Introduction .............................................. 1
1.1 General ........................................... 1
1.2 Advanced Oxidation Chemistry ........................... 2
II. Determination of Hydrogen Peroxide in Aqueous Ozone Solutions Using
Flow Injection Analysis ................................... 9
2.1 Introduction ........................................ 9
2.2 Materials and Methods ................................. 11
2.3 Results ............................................ 14
2.4 Discussion ......................................... 15
2.5 Conclusions ........................................ 16

III. Mixing Characteristic Comparison of Two Commercial Photochemical

Reactors .............................................. 17
3.1 Introduction ........................................ 17
3.2 Materials and Methods ................................. 19
3.3 Results ............................................ 23
3.4 Discussion ......................................... 25
3.5 Conclusions ........................................ 26

IV. Oxidation of 1,3,5-Trichlorobenzene Using Advanced Oxidative Processes . . 27

4.1 Introduction ........................................ 27
4.2 Backround ......................................... 28
4.3 Materials and Methods ................................. 31
4.4 Results ........................................... 39
4.5 Discussion ........................................ 48
4.6 Conclusion . . . . . .. . . . . . '. .......................... 53
V. Conclusions .............................................. 55
5.1 Conclusions ........................................ 55
5.2 Future Research ..................................... 56

iv

TABLE OF CONTENTS (cont.)

List of References ............................................. 58
Appendix A. Indigo Blue Method for Sampling Aqueous Ozone ............ 61
Appendix B. Trichlorobenzene Tracer Study Determining Time to Reach Steady
State ................................................... 65
Appendix C. Trichlorobenzene, Ozone, and Peroxide Sampling Summary for
Each Experiment ........................................... 68
Expt. 1 Hydrogen peroxide concentration optimization ................ 68
Expt. 2 Aqueous ozone treatment at varying ph (low range) ............ 7O
Expt. 3 Aqueous ozone treatment at varying ph (high range) ........... 72
Expt. 4 Ozone/Peroxide treatment at varying pH .................... 74
Expt. 5 Ozone/UV treatment at varying pH ....................... 78
Expt. 6 Comparing treatment processes .......................... 8O
Expt. 7 Comparing treatment processes with humic acid .............. 83
Expt. 8 Comparing treatment processes with bicarbonate .............. 85

Appendix D. Headspace Sampler, Gas Chromatograph Operating
Parameters ............................................... 87

Table 2.1
Table 3.1
Table 4.1
Table 4.2
Table 4.3
Table 4.4
Table C]
Table C.2
Table C3
Table C.4
Table C.5
Table C.6
Table C.7
Table C.8

Table C.9

Table C.10
Table C.11
Table C.12
Table C.l3

Table C.14

LIST OF TABLES

F IA System Parameter Optimization ....................... 14
Batch Tracer Results for Reactor # 7863 .................... 25
Ozone, Uv and Peroxide Reaction Rate Results ................ 41
Pseudo First Order Rate Constants for pH Variation Study ........ 44
Rate Constant Comparison with Humic Acid and Bicarbonate ..... 47
Hydroxyl Radical Concentration Results ..................... 48
Exp. 1 GC Data Summary .............................. 68
Exp. 1 Ozone and Peroxide Sampling Data .................. 69
Exp. 2 GC Data Summary .............................. 70
Exp. 2 Ozone Sampling Data ............................ 71
Exp. 3 CC Data Summary .............................. 72
Exp. 3 Ozone Sampling Data ............................ 73
Exp. 4 GC data Summary .............................. 74
Exp. 4 Peroxide Sampling Data .......................... 76
Exp. 4 Ozone Sampling Data ........................... 77
Exp. 5 GC Data Summary .............................. 78
Exp. 5 Ozone Sampling Data ............................ 79
Exp. 6 GC Data Summary .............................. 80
Exp. 6 Ozone and Peroxide Sampling Data .................. 82
Exp. 7 GC Data Summary .............................. 83

vi

LIST OF TABLES (cont)

Table C.15 Exp. 8 GC Data Summary .............................. 85

Table D1 Headspace Sampler and GC Operating Parameters ............. 87

vii

Figure 1.1

Figure 1.2

LIST OF FIGURES

Reactions of Aqueous Ozone in "Pure Water" .................

Reactions of Aqueous Ozone in the Presence of

Micropollutant M ..............................................

Figure 2.]
Figure 2.2
Figure 3.1
Figure 3.2
Figure 3.3
Figure 3.4
Figure 4.1
Figure 4.2
Figure 4.3
Figure 4.4
Figure 4.5
Figure 4.6
Figure 4.7
Figure 4.8
Figure A.1

Figure 8.1

FIA Configuration ....................................
Hydrogen Peroxide Standard Curve for the FIA System ..........
Schematic of Photochemical Reactor # 7863 ..................
Schematic of Supermix Photochemical Reactor # 7868 ...........
Batch Tracer Analysis for Supermix Reactor # 7868 .............
Continuous Flow Tracer Analysis on Supermix Reactor # 7 868 .....
Experimental Configuration ..............................
Hydrogen Peroxide Optimization ..........................
Ozone Treatment with Varying pH .........................
Ozone/Peroxide Treatment with Varying pH ..................
Ozone/UV Treatment with Varying pH ......................
Effects of pH on Various Treatment Processes .................
Comparing Treatment Processes in the Presence of Humic Acid . . . .
Comparing Treatment Processes with Bicarbonate ..............
Indigo Blue/Ozone Calibration Curve .......................

Trichlorobenzene Tracer Study ...........................

viii

3

4

15

21

24

32

4O

42

42

43

43

45

45

63

67

CHAPTER I

INTRODUCTION

1.1 GENERAL

Growing concern over the quality of groundwater and surface water has led to the
development of numerous techniques to treat contaminated waters. Remediation
technique investigations are needed to develop effective treatment technologies. This
research investigates the potential for using advanced oxidation processes to remove
recalcitrant organics from contaminated aqueous wastes. Advanced oxidation
processes are defined as processes that generate highly reactive hydroxyl radicals
(Glaze et al., 1987). The efficiency of advanced oxidation processes using ozone,
ozone/UV, ozone/hydrogen peroxide, and ozone/UV/hydrogen peroxide to treat
aqueous wastes were evaluated in this study.

1,3,5-Trichlorobenzene was the chosen target compound. This compound has been
shown to resist biological treatment (Kirk et al., 1989). It is commonly produced in
the chemical industry as a pesticide manufacturing byproduct (Lamporski et al., 1980).

A rapid and simple flow injection analysis (FIA) technique to measure dilute
concentrations of hydrogen peroxide in the presence of ozone was developed for this
advanced oxidation study. This is discussed in Chapter 2.

A commercial photochemical reactor was modified to perform as a continuous
flow through stirred reactor (CFSTR) . Tracer studies were performed to ensure the

reactor could be modeled as a CFSTR. These studies are discussed in detail in

Chapter 3.

The advanced oxidation studies are discussed in Chapter 4. The optimal peroxide
concentration for the ozone/peroxide process was determined. Individual ozone, UV,
and peroxide oxidant rate constants with trichlorobenzene were determined. The effect
of pH on each treatment process was then investigated. Finally, the efficiencies of
each treatment process in the presence and absence of humic acid and bicarbonate
were compared.

1.2 ADVANCED OXIDATION CHEMISTRY

Ozone can react with aromatic compounds directly or indirectly. The direct reaction
is selective, favoring compounds with electron donating groups (i.e., OH and NHZ).
Compounds with electron withdrawing groups (i.e., NO2 and Cl), like
1,3,5-trichlorobenzene, are not easily oxidized by ozone. These compounds can,
however, be oxidized with ozone indirectly using OH radicals.

OH radicals are the products of ozone decomposition. Figure 1.1 illustrates the
reactions of aqueous ozone in pure water (Staehelin et al., 1985). Hydroxide ions
initiate ozone decomposition by reacting with ozone, producing superoxide anions
(0;). Ozone then reacts with the superoxide anion to form ozonide radical ions and
oxygen. The protonated ozonide ion rapidly decomposes to produce oxygen and
hydroxyl radicals. The hydroxyl radicals decompose ozone further, producing more
superoxide ions, which enter into the cyclic reaction.

Addition of solutes to the ozone reaction scheme complicates matters further.

Figure 1.2 shows the effect of different solutes on ozone decomposition (Staehelin et

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H+\9l
H0; 1 " .OH A 03

O2

Figm'e 1.1 Reactions of Aqueous Ozone in ”Pure Water"
(Adapted from Staehelin and Hoigné, 1985)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ho3 OH s
02
S oxid.
LEGEND
M - Micropollutant (can also act as an initiator, scavenger, or promoter)
M w- Oxidized form of micropollutant
P - Promoter
P arid. - Oxidized form of promoter
1 - Initiator

1* - Protonated form ofinitiator formedby elecnontransferreection
S - Scavenger
S.“ - Oxidizedfomofscavenger

 

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

 

al., 1985). Four types of reactive species can be identified. First are the
micropollutants (M) that may be oxidized either by ozone directly or by secondary
oxidants. Second are the initiators (I), such as hydroxide or hydroperoxide ions, that
react with ozone to produce superoxide ions. Third are the promoters, e.g., formic and
humic acids, that react with hydroxyl radicals to produce superoxide radicals. Last are
the scavengers, such as carbonate and bicarbonate ions, that consume hydroxyl radicals
without regenerating superoxide anions. The presence of these solutes effect the rate
of ozone decomposition and secondary oxidant generation, and thus effect the rate at
which the micropollutant will be degraded.

The initiators of ozone decomposition that were used in this study are hydroxide
ions, ultraviolet light, and hydrogen peroxide. Ozone decomposition rates can be
increased by increasing the hydroxide ion concentration. This is done by raising the
pH. Hydroxyl radicals are produced from this ozone decomposition. Above a certain
hydroxide ion concentration, however, the hydroxyl radicals produced are less
available to react with the target chemical. This is due, in part, to the increased
carbonate and bicarbonate concentrations found in open systems with elevated pH
values. Carbonate and bicarbonate ions scavenge hydroxyl radicals, as shown

(Staehelin et al., 1985):

area; 1100,;
l I + 0H ' u- l i + 0H “ 1.1
co,‘2 co;

This scavenging reduces the removal rate of the target chemical.

In water, hydrogen peroxide dissociates into hydroperoxide ions, as shown:
H202 + H20 .. Ho; + 1130* 19-

Hydroperoxide ions act as initiators of the ozone decomposition. They react readily
with ozone, producing superoxide anions (O;). The superoxide ions in turn react with

ozone to form hydroxyl radicals, as shown (Paillard et a1. 1988):

03 + 110; ~ 011' + 02‘ + 02 1-3
Hydroperoxide ions can act as hydroxyl radical traps at high concentrations because
the ozone/peroxide reaction products react readily with hydroxyl radicals, as shown

(Paillard et al., 1988):

011- + no; -. Ho,- + on- 1.4
011- + 0, ~ Ho, + 02 1-5
110,; + H02“ -. H202 + o, 1.6

0H- + 110, ~ H20 + 02 1-7

OH' + 0H - 11202 1.8

0H' + 0; ~ 02 + 011- 1-9
At elevated hydroperoxide ion concentrations, hydroxyl radicals are consumed by the
competing reactions (equations 1.4 - 1.9), reducing the target chemical removal rate.
At low hydroperoxide ion concentrations (i.e., below the experimentally determined
optimal concentration), hydroxyl radical production decreases, also reducing the target
chemical removal rate.
Ultraviolet light initiates ozone decomposition by producing hydrogen peroxide, as

shown (Peyton et al., 1987):

03 + hv + H20 -* 17202 1.10
The peroxide formed can then react with ozone to produce hydroxyl radicals

(equations 1.2 and 1.3). Peroxide can also undergo photolysis, producing hydroxyl

radicals, as shown (Baxendale et. al., 1957):

H202 + hv ~ 2(0H)- 1-11
Guittonneau et al. (1990), showed that ozone/UV systems had higher hydroxyl radical
generation rates compared to HZOZ/UV systems. Even though the quantum yield for
the formation of hydroxyl radicals (via peroxide photolysis) is high, the molar
absorptivity for peroxide is low compared to ozone (at wavelengths greater than 200
nm). This indicates the ozone reaction with hydrogen peroxide (formed by ozone

photolysis) is the primary mechanism for the generation of hydroxyl radicals in

8

ozone/UV systems until the ozone is depleted. Once the ozone is depleted, the

peroxide/UV reaction is the primary mechanism.

CHAPTER II
DETERMINATION OF HYDROGEN PEROXIDE IN AQUEOUS OZONE
SOLUTIONS USING FLOW INJECTION ANALYSIS

2.1 INTRODUCTION

Hydrogen peroxide can be used in water and wastewater treatment to augment the
production of OH radicals in advanced oxidation processes. Combining peroxide with
ozone and/or ultraviolet light leads to the generation of hydroxyl radicals. The
measurement of hydrogen peroxide at low concentrations is required to determine and
maintain Optimal peroxide concentrations in advanced oxidation processes. Hydrogen
peroxide is also generated in water treatment systems using ozone, ultraviolet light, or
a combination of both. Hydrogen peroxide measurement is required, in this instance,
to determine reaction pathways, to better understand the advanced oxidation process,
and to monitor process water for potentially high concentrations of hydrogen peroxide.

Several methods to measure hydrogen peroxide at low concentrations have been
documented in recent years (Baga et al., 1988; Kieber et al., 1986; Wagner et al.,
1984). These methods involve the use of manual titrations that are complicated and
time consuming. An automated method, as opposed to manual titration, would
facilitate the monitoring of hydrogen peroxide in continuously operating water
treatment systems using advanced oxidation processes. It would also simplify
sampling techniques in research conducted on advanced oxidation processes.

A colorimetric titration method can be automated using flow injection analysis

(FIA). Flow injection analysis is based on the injection of a liquid sample into a

10

moving, non segmented continuous carrier stream of a suitable liquid. The injected
sample forms a zone, which is then transported towards a spectrophotometer which
continuously records the absorbance as it changes as a result of the passage of the
sample material through the flow cell (Rfizicka et al., 1981). An automated flow
injection analysis system has certain advantages, including higher sampling frequencies
and greater reproducibility of the experimental procedure, which can lead to higher
precision (Karlberg, 1989).

Bader ct a1. (1988), describes a photometric method for the determination of low
concentrations of hydrogen peroxide by a peroxidase (POD) catalyzed oxidation of
N,N-diethyl-p-phenylenediamine (DPD). This technique accurately measures hydrogen
peroxide concentrations up to 100 uM. It has a lower detection limit of 6 nM, and is
not effected by the dissolved material present in different natural waters. This method
of measurement was incorporated into a flow injection analysis system for use in
advanced oxidation studies.

Hydrogen peroxide samples are first bubbled with nitrogen gas to purge the ozone
from solution. This purging technique has been used in other hydrogen peroxide
methods (Peyton et al., 1987). The sample is then injected into a continuous flowing
carrier stream of phosphate buffer solution, where it later mixes with DPD and POD
reagents. The peroxide in the sample oxidizes the POD. The oxidized POD in turn
oxidizes the DPD to the radical cation DPD'+ which forms a stable red color. This
sample slug passes through a continuous flow cell positioned in a spectrophotometer

where the DPD'+ is continuously measured at 551 nm. The sample slug produces an

11

absorbance peak whose height is directly proportional to the peroxide concentration.
This paper describes the FIA system setup that produced the best peak height

reproducibility and broadest measurement range.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\. POD
SYRINGE /
PUMP > DPD WASTE
PBRIS'I'ALTIC CARRIER 9 U REACTION
UMP STREAM e V our:
P - 3 i MIXING 7‘
<EXCESS I TEES I
SAMPLE I
: SAMPLE
mgfm > b. LOOP srachopnoro-
METER
m mmcnon
VALVE

Figure 2.1 FIA Configuration

2.2 MATERIALS AND METHODS
E1 9 S D . .

Figure 2.1 represents the experimental setup for measuring hydrogen peroxide with
an FIA system. POD and DPD flows were driven with a syringe pump (Razel
Scientific, Model A-E). A peristaltic pump (Cole Parmer-Master Flex, Model 7520-
35) was used for the pH 6.0 phosphate buffer carrier stream. Using a plastic syringe,

the sample was loaded into the sample loop of a Teflon" rotary valve (Reodyne, Type

12

50). This two-way valve has a load sample position and an inject sample position.
When in the load position, the sample loop is not connected to the carrier stream.
When the valve is switched to the inject sample position, the carrier stream is
redirected to flush the sample loop into the carrier stream, where it later mixes with
the POD and DPD flow streams. This flow path is represented with the dashed arrows
in Figure 2.1. Teflon® tubing (0.031 inch inside diameter) was used for all flows. A
one centimeter quartz continuous flow-through cell positioned in a UV/vis
spectrophotometer (Model UV-1201, Shimadzu Scientific Instruments, Inc., Columbia,
MD) was used to continuously record the absorbance.

Reagents

Buffer stock solution: 0.5 M NaHzPO4 was added to 0.5 M NazHPO4 until a pH of
6.0 was obtained.

Hydrogen peroxide standard solution: 30% hydrogen peroxide (Baker Analyzed)
was diluted to 0.01 M and standardized via direct UV absorption (6 = 40.0 M"cm‘l at
7t = 240 nm). Appropriate concentrations were obtained by diluting this solution.

DPD reagent 0.5 grams N,N-Diethyl-p-phenylenediamine sulfate salt (DPD)
(Sigma Chemical Co., St. Louis) was dissolved in 100 ml 0.1 N H2804.

POD reagent: 50 mg Type 1 peroxidase from horseradish (Sigma Chemical Co.,
St. Louis, 78 purpurogallin units/mg) was dissolved in 100 ml distilled water.
Preceding

FIA system optimization: FIA system parameters were varied to obtain broad

operating ranges, lower detection limits, and reproducible absorbance peak heights.

l3

Phosphate buffer, POD, and DPD flow rates and concentrations were optimized.
Reaction zone length and sample loop size were also optimized. These parameters
influence the reaction time in the tubing, the extent of reaction, and the extent of
controllable sample dispersion.

Hydrogen peroxide standard curve: The F IA system was operated for 30 minutes
before standard samples were injected to purge air from the lines and to allow the
system to reach steady state. The spectrophotometer was then autozeroed at a
wavelength of 551 nm. Blank samples were9injected into the FIA system and the peak
height was recorded. This value was subtracted off the peak heights for the standard
solutions. Hydrogen peroxide samples ranging from 1 to 500 uM were injected in
triplicate and absorbance peaks were recorded. The spectrophotometer was autozeroed
between standards, if necessary. Calibration curves could not be done using hydrogen
peroxide in the presence of ozone because ozone degrades hydrogen peroxide.

Sample preparation: Samples (not the standards) were immediately bubbled with
nitrogen gas in a test tube to purge all ozone before injection into the FIA system. If
the sample pH was lower than 6.0, a minimum of 5 minutes purging time was used.
If the sample pH was above 6.0, a two minute purging time was used. Tests showed
that no hydrogen peroxide was lost due to the bubbling of nitrogen gas. Before the
first sample was taken, a blank was injected into the F IA system. The blank peak
absorbance value was subtracted off the sample peak absorbance. Hydrogen peroxide

concentrations were determined by correlating this value to the standard curve.

14
2.3 RESULTS

FIA tem timiz ti n

Table 2.1 shows the results of the F IA system parameter Optimization.

Table 2.1 FIA System Parameter Optimization

Buffer flow rate 6 ml/min.
Buffer concentration 0.5 M (NaHzPO4 / NazHPO4)
POD flow rate 0.54 ml/min.
DPD flow rate 0.54 ml/min.
POD concentration 5 mg/l
DPD concentration 50 mg/l
Reaction zone length 15 feet
Sample loop size 0.1 ml
FIA rve

Figure 2.2 shows the standard curve for hydrogen peroxide using the FIA system.
Error bars represent 95% confidence intervals on triplicate samples. Beyond 500 nM,
absorbance readings deviate from linearity. From 0 to 500 nM, [H202] = 29.7*A,
where A is the peak absorbance (blank corrected), and [H202] is the concentration of
hydrogen peroxide in uM. The correlation coefficient for this linear regression line fit
was 0.999. The average standard deviation at the low end of the standard curve
(between 1 and 10 uM) was 8.9 x 10" absorbance units, which corresponds to a

lower detection limit of 26 nM (based on three times the standard deviation).

15

PEAK ABSORBANCE

 

 

O. . 1 I I . 1 . I L l a #1
0 100 200 300 400 500 600

HYDROGEN PEROXIDE CONCENTRATION (UM)

Figure 2.2 Hydrogen Peroxide Standard Curve for the F IA System

2.4 DISCUSSION

Hydrogen peroxide concentrations as high as 500 M were accurately determined
without dilution. Samples can be taken at a frequency of one sample every forty
seconds. The range of this FIA method is five times greater than the manual method
described by Bader et al. (1988). Total automation of this system could easily be
accomplished by adding an in-line nitrogen bubbler between the sampling point and
the sampling loop of the FIA system.

A concentrated phosphate bufi'er concentration Of 0.5 M was needed to Offset the
pH dependence of this analysis. The formation of DPD"' occurs optimally at a pH of
6.0. A phosphate buffer concentration of 0.05 M did not sufficiently buffer samples

with a pH of less than 4.0 . These samples produced no FIA peaks. As the phosphate

16

concentration in the buffer was raised to 0.5 M to increase the buffer intensity, the
absorbance peak for the blank increased. This can be attributed to a change in
refractive index for the different buffer solutions. A peroxide sample adjusted to pH
levels between 2 and 10 with phosphoric acid and sodium hydroxide yielded identical
results when injected into the FIA system with the buffer concentration at 0.5 M.
2.5 CONCLUSIONS

Reproducible results were obtained for peroxide concentrations between 1 and 500
uM. The pH was controlled using a 0.5 M phosphate buffer. The DPD/POD method
combined with flow injection analysis is a fast and accurate method for measuring

dilute hydrogen peroxide concentrations in the presence of ozone.

CHAPTER III

MIXING CHARACTERISTIC CONIPARISON OF TWO COMMERCIAL
PHOTOCHEMICAL REACTORS

3.1 INTRODUCTION

Photoreactors are used to study photochemical reactions of interest. Batch reactors
(Prat et al., 1988; Xu et al., 1988; Peyton et al., 1988) and continuous flow reactors
(Peyton et al., 1987; Khan et al., 1985) have been used in waste treatment studies.
Improper mixing could effect the results obtained in these studies. Despite the
importance of good mixing in these reactors, little information is available on the
mixing characteristics of commercial photochemical reactors. Simple mixing studies
are an important first step in modeling photochemical reactor systems.

In batch experiments, complete mixing is desirable. Improper mixing in batch
experiments could yield erroneous reaction rate results. The desired levels of mixing
in continuous flow reactors depend on the type of system needed. Extensive mixing
(i.e. with impeller blades), extensive recirculation, or a combination of both, help
satisfy the requirements of continuous flow through stirred reactors (CFSTR). Poor
mixing and/or recirculation in a continuous system help satisfy the requirements of
plug flow reactors. Continuous flow systems can also be modeled as a CF STR/plug
flow combination (Nauman et al., 1983).

Mixing efficiencies in continuous flow systems can be determined by tracer
analysis. A washout function (equation 3.1) can be generated by recording the reactor

effluent concentration over time after implementing a negative step concentration

17

18
change, as discussed by Nauman et al., (1983). The washout function, W(t), is defined

as;

W(t = 392’. 3.1

C(O)d

where C(t) is the reactor concentration after the step concentration change, and C(0) is
the initial reactor concentration. For completely stirred vessels, the theoretical washout

function is given by;

W(t) = exp (24] 3.2
‘Iou

where t is the time after the step concentration change, and thy, is the reactor
hydraulic retention time. Step changes in concentration can be easily achieved by
switching reactor input lines to dye stock solutions. By integrating the washout
function (equation 3.1), the actual reactor retention time can be determined (equation

3.3);

{up = f we”; 3.3
o

where tap is the experimental reactor fluid retention time. The reactor dead volume is
obtained by taking the difference between the hydraulic residence time and

experimental residence time, i.e.;

dead volume = admin) 3.4

where Q, is the reactor flow rate.

19

This paper describes the tracer studies performed on two commercial
photochemical reactors proposed for use with advanced oxidation research. A CF STR
system configuration was chosen to simplify sampling techniques and mathematical
modeling, and to simulate a continuously flowing water treatment system.

3.2 MATERIALS AND METHODS
Photochemical Reactor System Description

Figure 3.1 represents the photochemical reactor chosen for our initial studies. This
reactor (Ace Glass, Inc., Vineland, NJ, Model #7863) had a working volume of 500
ml. It was modified by adding a continuous flow outlet port, as shown. A quartz
immersion well (Ace Glass, Inc., # 7874-38) was used in the reactor, providing a
housing for the low pressure photochemical immersion lamp (ACE Glass, Inc., #
12128). Water was circulated through the jacket of the immersion well when the lamp
was on. A magnetic stirrer (VWR Scientific, Model # 200) was used to stir the
reactor. Herein, this reactor shall be referred to as Reactor A.

Figure 3.2 shows the second photochemical reactor used. This Supermix reactor
(Ace Glass, Inc., Model #7868), equipped with a 85 ml recirculation loop, had a
working volume of 250 ml. This reactor will be referred to as Reactor B. A quartz
immersion well was also used in this reactor. A glass impeller shaft (Ace Glass, Inc.,
# 8068-08) was used to circulate the fluid in the reactor. The shaft was rotated in a
water cooled glass bearing (Ace Glass, Inc., # 8040-10) using an electric motor (Ace
Glass, Inc., # 13583) connected to a shaft through a flexible drive cable (Ace Glass,

Inc., # 8081).

20

COOLING WATER

 

DYE INJECTION 1 ,
LINE ’

 

SAMPLE LINE T0
SPECTROPHOTOMETER

 

 

 

 

 

CONTINUOUS FLOW
OUTLET PORT

QUARTZ PHOTOCHEMICAL
IMMERSION WELL

\ INSUFFICIENT

MIXING HERE

 

 

 

 

STIR

 

 

 

 

 

 

Figure 3.1 Schematic of Photochemical Reactor # 7863
(Reactor A)

21

FLEXIBLE SHAFT
CABLE

  
  
 

SAMPLE LINE TO
SPECTROPHO’IOMETER

 

 

COOLING WATER
PORTS

 

WATER COOLED
GLASS BEARING

 

 

 

 

 

 

 

 

 

 

CONTINUOUS FLOW
/ OUTLET POKI‘

IMPELLER BLADE
SHAFT

 

 

RECIRCULATION
LOOP

 

 

 

 

 

IMPELLER BLADES

 

QUAKIZ
IMMERSION WELL

 

 

 

 

Figure 3.2 Schematic of Supermix Photochemical Reactor # 7868
(Reactor B)

22
Batch Tracer Analysis Procedure an Photochemical Reactgr A

Reactor A was filled with deionized water to the level of the outlet port. A
sample line (0.031 inch inside diameter Teflon® tubing) was positioned near the top of
the reactor. A peristaltic pump (Cole Parmer, # 7520) continuously withdrew reactor
fluid through this line into a one centimeter quartz continuous flow cell (Model #
178.710-QS, Hellma Cells, Inc., Forest Hills, NY. ) that was positioned in a UV/vis
spectrophotometer (Model UV-1201, Shimadzu Scientific Instruments, Inc. Columbia,
MD). An injection line (0.031 inch inside diameter Teflon® tubing) was positioned in
the bottom of the reactor. A magnetic stirrer rotated at the highest allowable speed to
mix the reactor. A slug of concentrated methylene blue dye solution (Sigma Chemical
Co., St. Louis) was injected into the bottom of the reactor. The absorbance of the
effluent from the reactor was monitored with the spectrophotometer at 665 nm. The
time for the liquid in the reactor to reach a steady state concentration was recorded.
This experiment was repeated with three different types of magnetic stir bars.

Batsh Tracer Analysis Prgcedurs an Photachemisal Rsastar B

The same procedure used with Reactor A was used with Reactor B except the
impeller blades provided mixing. Magnetic stir bars were not used. Different impeller
blade shaft rotation speeds were used do determine how rotation speed affected reactor
mixing. Shafi rotation speeds were determined with a tachometer.

Continagus Flaw Tracer Analysis Prgcedare an Rsagtgr B
Methylene blue dye solution with an absorbance of 0.40 (at 665 nm) was pumped

into the reactor just below the impeller blades with a peristaltic pump (Cole Parmer,

23
Model # 7520—35) at a rate of 50 ml/minute. This provided for a hydraulic residence

time of 5 minutes. Tubing positioned at the outlet port continuously sampled the
reactor effluent. A peristaltic pump, continuous flow cell, and spectrophotometer was
used for sampling, as before mentioned. The methylene blue input line was switched
to clear deionized water after the reactor was filled and the sample line absorbance
leveled to 0.40. Absorbance versus time was continuously monitored until the effluent
absorbance was 0.00. The washout function was plotted and analyzed from these data
points.

3.3 RESULTS

Table 3.1 shows the results of the batch tracer study done on Reactor A. The
fastest time to achieve complete mixing was 8 minutes. Complete mixing was
achieved when the reactor effluent absorbance leveled out to a constant value (within
1%).

Figure 3.3 shows the results of the batch tracer study on the Reactor B. The
higher the shaft rotation speed, the better the mixing. At a rotation speed of 550
RPM, complete mixing was achieved in 90 seconds.

Figure 3.4 shows the results of the continuous flow tracer study with Reactor B.
The washout data was plotted against the theoretical stirred tank response and
theoretical plug flow response. A curve was fit to the experimental data points. By
integrating this curve (equation 3.3), a reactor residence time was calculated to be 4.94

minutes. This indicates that the dead volumein the reactor is 3.1 ml (equation 3.4).

TIME REQUIRED FOR COMPLETE MIXING

(SECONDS)

=C/Co

W(I)

rd

IJ

CM
CO!
IrrrIIrrrII

N U1 \1
U" 0 U‘

C)
(D

24

cu \J
(3 L”

I.)
L”

IlllIllllllllllllllllllllllllllllll

 

 

C)

I I
300 350 400 450 500 550 600

SHAFT ROTATION SPEED (rpm)

to
L”
(D

Figure 3.3 Batch Tracer Analysis for Supermix Reactor # 7868

I ACTUAL WASHOUT DATA
THEORETICAL STIRRED TANK RESPONSE
—e— THEORETICAL PLUG FLOW RESPONSE

 

 

0.8-*

0.6-

0.4—

0.2-

 

 

0.0 1 f‘ .1-
I‘rl'l'l'l'fi—filwfll'l

e 200 400 600 800 1000 1200 1400 1600 1800
TIME AFTER STEP CONCENTRATION CHANCE (SECS)

 

 

Figure 54 Continuous Flow Tracer Analysis on Supermix Reactor #7868

25

Table 3.1 Batch Tracer Results for Reactor A

 

 

Stir bar used Time for complete mixing
Octagonal stir bar (2") 10 minutes
Star head stir bar (1-3/8") 8.5 minutes
Star head stir bar (1-3/4") 8.0 minutes

 

3.4 DISCUSSION

Batch tracer results for Reactor A revealed this reactor was poorly mixed. Visual
observations showed the majority of the mixing occurred at the bottom of the reactor.
Very little mixing occurred in the gap between the quartz immersion well and the
inner wall of the reactor. The stir bars could not create sufficient turbulence to
effectively mix the fluid in this gap, which was only 0.63 inches wide. Since this
reactor was poorly mixed, no continuous flow tracer studies were performed.
Continuous flow studies with this reactor would require modeling it as a CFSTR and
plug flow reactor in series.

Batch tracer studies for the Reactor B proved this reactor was more efficiently
mixed. Since only 90 seconds was required for complete mixing in the batch mode, a
reactor retention time as low as five minutes (approximately three batch mixing
retention times) would assure complete mixing in a continuous flow configuration.

For this reason, a five minute retention time was chosen for the continuous flow tracer

26
study. Figure 3.4 shows that, as predicted, Reactor B behaved as a CF STR at a

retention time of five minutes. It follows that any reactor retention time greater than
five minutes can be modeled as a CFSTR. The results indicate 3.1 ml of dead space
exist, which is probably located in the three sampling port extensions at the top of the
reactor.
3.5 CONCLUSIONS

Reactor A did not provide adequate mixing. Modeling this reactor requires an
analysis of a CFSTR and plug flow reactor in series. Reactor B was sufficiently
mixed to support CFSTR modeling criteria at reactor retention times greater than or
equal to 5 minutes. Any future photochemical studies using reactors similar to Reactor
A should test the mixing efficiency prior to any in depth study. Any past
photochemical studies that used reactors Similar to Reactor A should be checked to

make sure the results were valid.

CHAPTER IV

OXIDATION OF 1,3,5 TRICHLOROBENZENE USING ADVANCED
OXIDATIVE PROCESSES

4.1 INTRODUCTION

Many studies have been done on advanced oxidation treatment processes using
ozone (Langlais et al., 1991), ozone/UV (Paillard et al., 1987), ozone/peroxide (Glaze
et al., 1989), and ozone/peroxide/UV (EPA, 1990). These techniques center around
the production of hydroxyl radicals, which are highly reactive, non selective oxidants.
Aqueous ozone itself is a powerful, but selective, oxidant. Recalcitrant compounds
that do not readily react with ozone ofien can be degraded by hydroxyl radicals.
Advanced oxidation processes generate hydroxyl radicals by initiating formation of
superoxide ions from ozone decomposition. These ions react with aqueous ozone to
form hydroxyl radicals (Glaze et al., 1987).

The ability of advanced oxidation processes to degrade recalcitrant organics is
effected, among others, by oxidant dose, UV light intensity, pH, retention time, the
nature the organic micropollutant (Guittonneau et al., 1990), and the concentration of
scavenging, promoting, or initiating solutes in the wastestrearn (Staehelin et al., 1985).
Contaminated natural waters and industrial waste streams may vary in pH and contain
significant quantities of these solutes. If advanced oxidation processes are used to
remediate these contaminated waters, the effect these variables have on target
compound removal efficiencies Should be investigated.

1,3,5-Trichlorobenzene, a recalcitrant chlorinated organic, was the chosen target

27

28

compound for these studies. This compound has been shown to resist biological
treatment (Kirk et al., 1989). It is commonly produced in the chemical industry as a
pesticide manufacturing byproduct (Lamporski et al., 1980).
Experiments were conducted to determine:
1. The optimal hydrogen peroxide concentration for trichlorobenzene removal
2. UV, peroxide, and ozone rate constants for trichlorobenzene
3. Effects of pH on advanced oxidation treatment processes
4. A comparison of the efficiencies of advanced oxidation treatment processes
in the presence and absence of humic acid and bicarbonate
4.2 BACKROUND
Hydrogen peroxide in water dissociates into hydroperoxide ions. These ions initiate
ozone decomposition, producing superoxide ions and hydroxyl radicals. At certain
concentrations, the hydroperoxide ions can act as hydroxyl radical scavengers (Paillard
et al., 1988). The first stage of this study determined the Optimal peroxide
concentration for degrading trichlorobenzene in the presence of ozone.
The overall rate of reaction for these advanced oxidation processes can be

formulated as:

11%?! = 1509103117“) 40,, [011mm] awe [men] 4,90911120211'101

where k is the reaction rate constant for the particular oxidant. [TCB], [03], [OH],
and [H202] are the reactor trichlorobenzene, ozone, hydroxyl radical, and peroxide

concentrations, respectively. I is the light intensity (irradiance) and (D is the quantum

29

efficiency. Equation 4.1 assumes the individual reaction rates are first order with
respect to the trichlorobenzene concentration (Glaze et al., 1980).

In the second stage of this study, the individual reaction rate constants for UV
light, hydrogen peroxide, and ozone with trichlorobenzene were determined by

simplifying and integrating equation 1:

999 reef 9999 res, 999 TCB,
k _ __ TCB, _ _ TCB, _ _ TCB, 4.2
03 [0,] e ’ ”202 [H202] e ’ W” [IO] 6

where TCBi and TCBf are the steady state reactor trichlorobenzene concentrations
before and after treatment, and 0 is the reactor retention time. Ozone and peroxide
concentrations represent steady state reactor concentrations after the treatment was
initiated.

Hydroxide ions initiate ozone decomposition by reacting with ozone to form
superoxide ions and hydroxyl radicals. In the third stage of this study, the effect of
pH on the trichlorobenzene removal rates with ozone treatment, ozone/UV treatment,
and ozone/peroxide treatment were determined. Pseudo first order oxidation rate

constants were calculated and compared at different pH values by simplifying equation

4.1 to:

w = kmmxrdammcaj 4.3

If we assume that the oxidant concentration [oxidant] is constant over the course of the

experiment (i.e., at steady state), the oxidant concentration and rate constant can be

30

combined, simplifying the equation further:

jjflgfl = k, [TCB] 4.4

Integrating equation 4.4 yields a pseudo first order rate constant, describing the

reaction rate of the overall advanced oxidation process, as shown:

TCB, 4.5

In the fourth stage of this study, the reaction rates for each treatment process
with and without the presence of humic acid and bicarbonate were determined.
Dissolved humic substances can act as initiators, promoters, or inhibitors of ozone
decomposition (Staehelin et al., 1985). Humic acid concentrations of 0.0, 2.0, and
10.0 mg/l were used. Bicarbonate and carbonate ions, common in natural water, act as
ozone decomposition inhibitors, scavenging hydroxyl radicals. Bicarbonate
concentrations of 0.0, 2.0, and 10.0 mM were used. Reaction rate constants (equation
4.5) were determined for ozone treatment, ozone/UV treatment, ozone/peroxide
treatment and ozonerV/peroxide treatment.

Hydroxyl radical concentrations for each treatment process not subjected to humic
acid and bicarbonate were estimated by assuming a hydroxyl radical reaction rate with
trichlorobenzene. The peroxide, photolysis and ozone rate constants from equation
4.2, along with the measured steady state concentrations of ozone, peroxide, and

trichlorobenzene, were substituted into equation 4.1, solving for the term k0,,[OH].

31

This calculation is shown in detail in the results section.
4.3 MATERIALS AND METHODS
Gmsral System Config_u_ratign

Figure 4.1 shows the experimental apparatus. A continuous flow system was used
to eliminate sampling time constraints and to better simulate configurations likely to be
used in actual treatment facilities. Gaseous ozone was produced using an ozone
generator (Polymetrics Corp., San Hose, CA, Model T-408) and bubbled into an ozone
contactor. The aqueous ozone solution was pumped into the reactor as opposed to
bubbling ozone gas directly into the reactor. This provided a continuous input ozone
flux. Ozone mass transfer corrections into reactor solutions due to varying
concentration gradients across the ozone liquid-gas interface are avoided with this
technique. Another advantage this method Offers is its ability to treat volatile
contaminants with low solubilities. Legube et al. (1983), reported that chlorobenzene
removal in ozonation bubble columns was more a consequence of stripping action than
oxidation reaction. Using this method, corrections due to volatilization of the
contaminant would not be required.
Phgtochemical Reactgr

A detailed schematic of the photochemical reactor is provided in Figure 3.2. This
reactor has a working volume of 250 m1 and is equipped with a recirculation chamber
that houses an impeller blade that provides continuous mixing. A quartz immersion
well houses a low pressure immersion lamp with a 254 nm principal wavelength

output. Piston pumps (Model RHSY, Fluid Metering Inc., Oyster Bay, NY) were used

32

OZONE GAS

A h

 

 

 

 

 

 

MOISTURE

 

 

REACTOR
PEROXIDE AND
OZONE SAMPLES

 

 

 

REACTOR
EFFLUENT

@—

OXYGEN OZONE WATER PHOIOCHEMICAL m
TANK GENERATOR COOLER REACTOR

 

 

_.oo

 

 

 

   

 

 

CONTAMINANI'

Figure 4.1 Experimental Configuration

33

to pump ozone and trichlorobenzene solutions into the reactor. Teflon® tubing was
used for all input and sample lines. Theinput lines were positioned just below the
impeller blades to ensure immediate mixing of the influent streams.
ngns Gas Generation aad Solution Prspasatign

Ozone gas (approximately 3%) was produced using an ozone generator
(Polymetrics Corp., San Hose, CA, Model T-408) that was fed pure dried oxygen.
The generator was cooled using water from a refrigerated circulating bath maintained
at 10 °C. Aqueous ozone was made by continuously bubbling ozone gas into a three
liter round bottom Pyrex flask containing deionized water that had been acidified to a
pH of 2.0 using phosphoric acid. This solution was pumped into the reactor with
piston pumps (Model RHSY, Fluid Metering Inc., Oyster Bay, NY) while a peristaltic
pump (Cole Parmer-Master Flex, Model 7520-35) continuously replenished the buffer
solution to the ozone contactor, maintaining a constant level. The contactor was
stirred to ensure a consistent effluent ozone concentration.
Wasts Strsasa Baffsr Prsparatign

The waste stream was buffered with phosphate so that a pH of 7.0 i 0.2 was
maintained in the reactor. after the ozone contactor buffer stream and waste stream
were pumped in at equal flowrates. This was necessary because the ozone contactor
solution was acidified to a pH of 2.0 with phosphoric acid.
1,3,5-f1frichlorgbgams Stggk Salatign Prspgatign

Trichlorobenzene (Aldrich Chemical Co., Milwaukee, WI) stock solutions were

prepared in 20 liter glass containers. Solid trichlorobenzene crystals were ground

34

using a mortar/pestle and added to the 20 liter vessel containing the waste stream
buffer. Quantities in excess of the solubility of trichlorobenzene in water had to be
added to obtain stock concentrations greater than 3 ppm. Stock solutions were mixed
for two days and allowed to settle another day before being decanted into three
separate 6 liter Pyrex flasks.

H i A i r ti n

Aldrich Humic Acid (sodium salt) was dissolved in deionized water (acidified to a
pH of 3.0) and passed through glass fiber filters. Stock solutions were stored in the
dark.

Analytjsal Methods

Analysis of inlet stream for ozone - The aqueous ozone in the inlet stream was
continuously monitored with a flow cell positioned in a UV/vis spectrophotometer
(Model UV-1201, Shimadzu Scientific Instruments, Inc., Columbia, MD). The ozone
contactor solution was continuously drawn through the flow cell with a hand operated
vacuum pump. An extinction coefficient of 3000 M"cm" was used to convert
absorbance units into concentration.

Reactor ozone analysis - Reactor ozone concentrations could not be measured
directly because of the possible presence of interfering compounds absorbing light at
258 nm. The indigo method (Bader et al., 1982) was used instead. Ozone samples
were withdrawn from the reactor with a hand operated vacuum pump and delivered
into 125 ml vacuum flasks containing known amounts of indigo blue solution. Ozone

concentrations were determined by correlating the decrease in absorbance of an indigo

35

solution to the ozone concentration (see Appendix A).

Reactor peroxide analysis - Reactor peroxide concentrations were determined by
flow injection analysis (F IA) using the peroxidase N,N-diethyl-p-phenylenediamine
(DPD) method (Bader et al., 1988). The FIA technique is discussed in Chapter 11.
Samples were withdrawn from the reactor and immediately bubbled with nitrogen gas
to purge the ozone before injection into the flow injection analysis system.

Trichlorobenzene sampling - Trichlorobenzene samples were collected at the
reactor effluent port. The effluent was allowed to drain into glass vials that contained
0.125 ml of 0.09 M sodium nitrite solution. The sodium nitrite quenched the ozone to
stop any further reaction. Aliquots of 10 ml were pipetted from these vials into
separate vials suitable for gas chromatography analysis. Five trichlorobenzene samples
were taken from the effluent for each steady state.

Headspace analysis of trichlorobenzene - 1,3,5-Trichlorobenzene was measured
using a gas chromatograph (Perkin-Elmer Autosystem, Norwalk, CT) equipped with a
flame ionization detector (FID) and a silica glass capillary column (Perkin-Elmer,
Model 624). The carrier gas was helium. A five point calibration curve was
performed for each experiment. Details of the gas chromatograph and headspace
sampler operating parameters are provided in Appendix D.

UV Light Intensity - The quartz immersion well and photochemical lamp were
cleaned with acetone before each experiment to maintain consistent UV light
transmission into the reactor and to remove any adsorbed trichlorobenzene. The light

intensity was measured with a potassium ferrioxalate actinometer (Murov, 1973) and

36

assumed to remain constant throughout the study.
Expsrimental Prgcsdiae

General - The reactor was operated with a hydraulic retention of 10 minutes by
pumping ozone contactor buffer and trichlorobenzene solutions in at 12.6 ml/min for
each experiment. The impeller blades that provided mixing were rotated at 500 RPM,
as determined by a tachometer. A trichlorobenzene tracer study determined the reactor
system required one hour (six retention times) to reach steady state after a step
trichlorobenzene input change (see Appendix B).

Each experiment was started by pmnping trichlorobenzene and ozone contactor
solution (not ozonated) into the reactor. Effluent samples were taken one hour later to
determine the initial steady state trichlorobenzene concentration. This initial
trichlorobenzene steady state concentration was kept within the range of 1.17 ppm to
1.42 ppm for each experiment. The effluent concentration for the subsequent
treatment processes were compared to this concentration to determine trichlorobenzene
removal efficiencies.

The reactor was then subjected to a series of treatments using aqueous ozone,
peroxide, UV light, pH adjusters, humic acid, or bicarbonate, depending on the process
being studied. Oxygen gas was bubbled into the ozone contactor when aqueous ozone
treatment was not being used. Aqueous ozone treatment was started by ozonating the
ozone contactor buffer solution to obtain a concentration of 12 mg/l for each
experiment, unless stated otherwise. This was easily accomplished by turning on the

power to the ozone generator (since the oxygen was already flowing through the

37

generator and into the contactor). The ozone contactor required 15 minutes to reach a
steady state ozone concentration after power was given to the generator. Effluent
trichlorobenzene samples were therefore taken 75 minutes after the ozone generator
was turned on.

Procedure to optimize hydrogen peroxide concentration- The contactor was
ozonated after initial trichlorobenzene steady state was achieved, subjecting the waste
stream to aqueous ozone. Effluent trichlorobenzene and aqueous ozone samples were
taken one hour later, after which different peroxide solutions were successively
pumped into the reactor using a syringe pump (Razel Scientific, Model A-E). In this
way, reactor peroxide concentrations ranging from 6 M to 6000 11M were achieved.
Effluent trichlorobenzene, ozone and peroxide were measured one hour after each new
peroxide input was initiated. The inlet ozone concentration for this experiment was
7.7 i 0.2 mg/l. Reactor pH was maintained at 7.0 i 0.2.

Determination of ozone, peroxide and UV rate constants- Individual rate constants
were determined by treating trichlorobenzene with ozone, peroxide, or UV light. The
reactor solution was subjected to these treatments after initial trichlorobenzene steady
state concentration was reached. Reactor pH was maintained at 7.0 i 0.2 for the
peroxide and UV studies. The ozone reaction rate constant was calculated with the
data obtained in the ozone treatment at varying pH study. The results from the
reactor pH of 2.24 was used because hydroxyl radical formation due to ozone
decomposition is minimal at low pH values.

pH variation studies - In each experiment, after the trichlorobenzene concentration

38

reached steady state, the solution in the reactor was subjected to either ozone,
ozone/UV, or ozone/peroxide treatment. Sodium hydroxide or phosphoric acid were
then pumped into the reactor with a variable Speed syringe pump (Razel Scientific,
Model A-99) at different rates to obtain the desired reactor pH. For each pH
condition, concentrations of trichlorobenzene, ozone, and peroxide (for peroxide
studies) were sampled from the effluent. The peroxide concentration determined by
the peroxide optimization study was used in the ozone/peroxide treatment.

Humic acid, bicarbonate comparison studies - The effect of humic acid and
bicarbonate on the efficiency of trichlorobenzene oxidation was assessed using four
advanced oxidation treatment techniques (ozone, ozone/UV, ozone/peroxide, and
ozone/peroxide/UV). This study consisted of three separate experiments. In the first
experiment, the waste stream was successively treated with the four treatment methods
without humic acid or bicarbonate. In the second experiment, the waste stream was
also successively treated with the four treatment processes. However, the reactor
solution first contained 2 mg/l humic acid for the four treatments. Once these studies
were finished, the humic acid concentration in solution was increased to 10 mg/l. In
the third experiment, the waste stream was successively treated with the four treatment
processes . The reactor solution first contained 2 mM sodium bicarbonate. Once
these studies were completed, the sodium bicarbonate concentration in solution was
increased to 10 mM. Humic acid and sodium bicarbonate solutions were pumped into
the reactor with a variable speed syringe pump (Razel Scientific, Model A-99) to

obtain the desired reactor concentrations. Reactor pH was maintained at 7.0 :1: 0.2 for

39

these studies.
4.4 RESULTS
H r 11 Per xid nc ntration timi ti n

The optimum removal of trichlorobenzene was obtained when a peroxide dosage
of 60 uM was used, as shown in Figure 4.2. The concentration of the reactor ozone
influent was 7.7 d: 0.2 mg/l. 97.8% of the trichlorobenzene was degraded in the
reactor (td = 10 minutes). The residual ozone and peroxide concentrations were 0.5 i
0.3 mg/l and 44.5 i 0.5 nM, respectively. Detailed results of ozone, peroxide, and
trichlorobenzene sampling for all experiments are provided in Appendix C.
Pagassiam Farrioxalats Actingmstsr Rssalts

Using a potassium ferrioxalate actinometer, the light intensity of the lamp emitted
to the aqueous solution was determined to be 1.21 watts. Specifications of the low
pressure immersion lamp list its output power to be 3.15 watts, indicating a 38% UV
absorption efficiency into the reactor solution.
UV, Psrgxids, aad ngns Rats ansflt Dstsrmiaatigg

The ozone reaction rate constant was calculated with the data obtained in the
ozone treatment at varying pH study (at a pH of 2.24). Results indicated that 49% of
the trichlorobenzene was removed photolytically without the presence of ozone, 22%
was removed with peroxide as the sole oxidant (with a reactor peroxide concentration
of 60 uM), and 45% was removed with ozone as the sole oxidant (with a reactor
ozone concentration of 4.9 mg/l). Table 4.1 lists the individual reaction rate constants

calculated from equation 4.2. The errors associated with the rate constants were

‘2'. TRICHLOROBENZENE REMAINING

 

40

   

 

[H202]: H202]: [P1202]= [HZOZ]=3 EH202]=
O .6rnM mM

1
.006mM _ OBmM

HYDROGEN PEROXIDE CONCENTRATION

Figure 4.2 Hydrogen Potomac Optimization

41

determined by incorporating the standard errors from the replicate samples into a

propagation of error analysis.

Table 4.1 Ozone, UV and Peroxide Reaction Rate Constants

 

 

 

 

 

Process pH k Units

Ozone 7.0 i 0.2 9.6 i 3.0 M" sec”1
Photolysis 7.0 :1: 0.2 8.9 :1: 3.6 x 10 4 watt"sec'1
Peroxide 7.0 :1: 0.2 6.6 i 1.8 M'1 sec"

 

 

 

 

pH Effest an Trsatmsnt Prggsss Rssalts
The effect of pH on the ozone, ozone/peroxide, and ozone/UV treatment

processes can be shown by plotting the percent trichlorobenzene remaining in the
reactor effluent against reactor pH for each treatment process (see Figures 4.3 through
4.5). The data presented in these three figures is combined into Figure 4.6. Table 4.2
lists the pseudo-first order reaction rate constants for each treatment pH as calculated
from equation 4.5. Standard errors represent 95% confidence intervals using the five
replicate trichlorobenzene measurements.
H i A i and Bic bonate C m arison Res lts

The results of studies in which the effect of humic acid and bicarbonate was
compared are illustrated by plotting the percent trichlorobenzene remaining in the

reactor effluent against the type of treatment process used (see Figures 4.7 and 4.8).

 

42

. . . » . l . r 4 I 1 . I .1
. . , . .. . 2, .. . .. . . a f / . .. .7 5. .. 14....2. . a... A... .. .I. .n .. ./. if. . 6. am”
. . e 5. 9... 7..., 5, x/ .3”, A... .. .. 4,1... .. .7. .. 7.474 :47. ,y.>.c._.x?, y , 5 . ,.
_...,~;..y.r...t7,..n.v,nfi..,).~.wm...4.$?,........#.> 39:41:11.; 7....v . 7....77; I .

.2
a a

 

, 1. 7f; .7; I g .. 7 , .I .., 7. , w

a.“

3 2.. 2 a.

AoO\ov ..
02.223”. aszUmoaofoaTm m _ .R

O 5 O 5 O 5 O 5 O .3 .J
.3 .. 4 . .:

 

REACTOR pH

Figure 4.3 Ozone Treatment With Varying pH

 

13

12

11

IO

 

 

and..—...E_q..._..a._..a._.au.—_...__~.a_ e.

O
4

5 O 5 O 5 O 5
.3 .4. 7. a. I .1

AoO\oL
oz_z_<2um “232880393 Ind. N

\J

REACTOR pH

Figure 4.4 Ozone/Peroxide Treatment with Varying pH

7.

I .15 - TRICHLOROBENZENE REMAINING

2 1,3,5 - TRICHLOROBENZENE REMAINING

(C/C0)
£3
I

U!

 

lllllllll

(C/C0)
is

I)

43

 

 

 

ITI 111 fIIIT
4 7

8 9 10 1 1 12
REACTOR OH

(I! ""‘

I
5

(H—

Figure 4.5 Ozone/UV Treatment With Varying pH

-4-- OZONE / PEROXIDE
-I- OZONE ONLY
—t- OZONE / uv

 

13

 

k
s.
tn —

REACTOR oH

F'iqure 4.6 Effects of pH on Various Treatment Processes

44

Table 4.2 Pseudo First Order Rate Constants for pH Variation Study

Treatment Process

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ozone 2.24 0.059 d: .007
Ozone 2.92 0.053 :I: .007
Ozone 5.25 0.160 :t .010
Ozone 6.21 0.218 1 .017
Ozone 7.17 0.242 :I: .009
Ozone 7.90 0.261 d: .017
Ozone 9.48 0.248 i .013
Ozone 10.55 0.213 t .014
Ozone 11.53 0.207 :I: .016
Ozone/Peroxide 2.35 0.113 :I: .021
Ozone/Peroxide 4.94 0.291 :I: .021
Ozone/Peroxide 6.96 0.367 :1: .016
Ozone/Peroxide 8.53 0.366 :I: .019
Ozone/Peroxide 10.10 0.253 :I: .039
Ozone/Peroxide 11.05 0.199 :I: .012
Ozone/Peroxide 11.50 0.151 :I: .014
Ozone/UV 2.32 0.320 :I: .013
Ozone/UV 4.75 0.343 i .012
Ozone/UV 7.22 0.325 :1: .016
Ozone/UV 8.95 0.303 :I: .016
Ozone/UV 12.01 0.187 i .013

 

 

45

N
.U‘
0

 

 

 

 

‘i : NO HUMIC ACID
q E HUMIC ACID = 2 mg/l
q = 1 I
0 ac o— E @ HUMIC ACID 0 mg/I
g h .
Z -I
‘2‘ 1
m d
a: .
g 15.0...
N
E: 2 E
a: .
0" 10.0—
A -I
I
g .T
93 .
lI.j .1
,1: 5.0-
“ -1
0.0—

 

 

 

 

 

 

 

OZONE OZONE/UV OZONE/H202 OZONE/HZOZ/Uv
TYPE or TREATMENT PROCESS

Figure 4 7 Comparing Treatment Processes in the Presence of Humic Acid

 

 

 

90.01
E

80.0— __
o 3 E NO BICARBONATE
g 700 : I: [BICARBONATE] = 2 mM
:2: ' 3 C [BICARBONATE] = 10 mM
LL) ‘
°‘ eon-a
w d
2 at
a I
ég 50.0:
o\ «
g8 40 0:
3:7 -I
a.) :
55 30.01
J. :
:3- 20.01 '.
N I f I I

10.0—I 7

2 ||
0.0 -1

 

OZONE OZONE/UV OZONE/H202 OZONE/HZOZ/UV
TREATMENT PROCESS TYPE

Ficy-He 4 8 Comocrinq Treatment Processes with Bicarbonate

46

Table 4.3 lists the pseudo-first order rate constants determined from equation 4.5 for
each process along with the treatment process concentration of humic acid or
bicarbonate used. k'n, is the relative pseudo-first order rate constant, i.e. the rate
constant normalized to the rate constant with no humic acid or bicarbonate for that
same process. The errors associated with the rate constants were determined by
incorporating the standard errors from the replicate samples into a propagation of error
analysis.

Equation 4.1 is integrated and solved for the term k0,,[OH]:

TCBf
-9 $403] + k MUG] + ngolezozl ) ‘ ”Ta, 4.6

 

kou [OH] = 0

The results from the studies using no humic acid or bicarbonate were used to calculate
k0H[OH] for the four treatment processes. Rate constants from Table 4.1 and results
for measured ozone, peroxide, and trichlorobenzene concentrations were substituted
into equation 4.6. The UV intensity was assumed to remain constant at 1.21 watts for
processes using ultraviolet light. Calculated k0,,[OH] values for each treatment process
are shown in Table 4.4. Estimated hydroxyl radical concentrations were calculated
assuming that the rate constant for the reaction of trichlorobenzene with OH radicals
was 2.5 x 109 M"sec". A structure-fimction relationship using vinyl chloride,
dichloroehtylene (DCE), and trichloroethylene (TCB) was used to estimate this value.
The reaction rate for chlorobenzene (Farhataziz et al., 1987) was multiplied by the

ratio of the rate constants of vinyl chloride and TCB (Farhataziz et al., 1987) to obtain

47

Table 4.3 Rate Constant Comparison with Humic Acid and Bicarbonate

Treatment

Process

  

Humic
Acid Cone.
(mg/1)

   

Bicarbonate
Cone. (mM)

     

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ozone 0 0 0.270 i 0.028 1 0
Ozone 2 0 0.341 t 0.111 1.26 -26.2
Ozone 10 0 0.225 :I: 0.021 0.83 16.8
Ozone/UV 0 0.389 i 0.024 1 0
Ozone/UV 0 0.299 i 0.012 0.77 23.3
Ozone/UV 10 0 0.157 i 0.015 0.40 59.6
Ozone/I-IZO2 0 0 0.555 t 0.032 1 0
Ozone/HZO2 0 0.375 :1: 0.018 0.68 32.4
Ozone/H202 10 0 0.213 :I: 0.017 0.38 61.6
Ozone/UV/HZO2 0 0.415 :1: 0.060 1 0
Ozone/UV/HZO2 0 0.316 1 0.041 0.76 23.9
Ozone/UV/HZO2 10 0 0.202 i 0.014 0.49 51.4
Ozone 0 0 0.270 :I: 0.028 1 0
Ozone 0 2 0.067 :I: 0.011 0.25 75.0
Ozone 0 10 0.017 :I: 0.011 0.06 93.5
Ozone/UV 0 0 0.389 1 0.024 1 0
Ozone/UV 0 0.226 :I: 0.026 0.58 41.9
Ozone/UV 0 10 0.155 :I: 0.011 0.40 60.1
Ozone/HZO2 0 0.555 :I: 0.032 1 0
Ozone/11202 0 0.263 :I: 0.011 0.47 52.6
Ozone/I-IZO2 0 10 0.180 :I: 0.015 0.32 67.5
Ozone/UV/HlOl 0 0.415 :1: 0.060 1 0
Ozone/UV/I-IZO2 0 0.272 t 0.022 0.65 34.5
Ozone/UV/HZO2 0 10 0.187 i 0.015 0.45 54.9

 

 

48

this estimate. The errors associated with the rate constants were determined by
incorporating the standard errors from the replicate samples into a propagation of error

analysis.

Table 4.4 Hydroxyl Radical Concentration Results

 

 

 

 

Treatment Process kOH[OH]
(Sec") (Moles/liter)
Ozone 7.0 i- 0.2 4.10 :1: 0.6 x 10'3 1.64 i 0.2 x 10’”
Ozone/UV 7.0 i 0.2 5.35 :1: 0.8 x 10'3 2.14 :1: 0.3 x 10'12
Ozone/Peroxide 7.0 i 0.2 9.04 :1: 0.6 x 10'3 3.60 i 0.2 x 10'12
Ozone/Peroxide/UV 7.0 i- 0.2 5.70 :1: 1.6 x 10‘3 2.28 :t .6 x 10'12

 

 

 

 

 

 

4.5 DISCUSSION

The hydrogen peroxide optimization results agree with work done by Paillard et al.
(1988). They reported optimal trichloroethane oxidation occurred at initial peroxide
concentrations of 60-70 11M. The optimal peroxide concentration in our study of 60
11M was determined for only one reactor ozone concentration. This occurred at reactor
ozone and peroxide concentrations of 0.5 mg/l and 44.5 11M, respectively, and at an
initial input ozone concentration of 3.9 mg/l (half of the concentration of the inlet
stream). Paillard et al. also reported that in open systems, optimal conditions were

obtained when 0.5 moles of peroxide were consumed per mole of ozone introduced.

49

Based on the above concentrations, our optimal peroxide concentration occurred when
0.2 moles of peroxide were consumed per mole of ozone introduced. Additional
studies could be done to determine the optimal ratio of ozone and peroxide introduced
to the reactor, as was done by Paillard et al. (1988).

The optimal pH for ozone treatment occurred at 7.90. The pseudo first order rate
constant steadily decreased above and below this optimal pH value. A 77% decrease
in the pseudo first order rate constant occurred when the pH was lowered from 7.9 to
2.2. A 21% decrease occurred when the pH was increased from 7.9 to 11.53. These
results differ slightly from work done by Masten, (1992), who reported the removal
efficiency for l-chloropentane (CPA) with straight ozone treatment increased as the pH
was raised from 2 to 10.5 . CPA is not reactive with ozone directly, much like
trichlorobenzene. Masten's studies were done in a closed batch type system, unlike
our continuous open system. This could explain why the results differed at high pH
values. Hydroxyl radical scavenging by carbonate ions is the likely cause for the
reduced trichlorobenzene removal efficiency in open systems at elevated pH values.
When the pH exceeds 10.3, the carbonate ion is the dominant carbonate species. The
carbonate ion scavenges hydroxyl radicals 20 times faster than the bicarbonate ion
(Glaze et al., 1987). If carbon dioxide is prevented from entering the system at
elevated pH values, a higher removal efficiency may be obtained. Reaction rates at
lower pH values are reduced because ozone is stable at low pH values. Fewer
hydroxyl radicals are formed at stable ozone conditions. This may not be the case for

all chemicals, however. If a compound is reactive with ozone, the removal rate may

50

increase with decreasing pH values (Masten et al., 1992).

The optimal pH for the ozone/peroxide treatment occurred between a pH of 7.0
and 8.5. The pseudo first order rate constant also steadily decreased above and below
this optimal pH value. A 69% decrease in the pseudo first order rate constant
occurred when the pH was lowered from 7.0 to 2.3. A 59% decrease occurred when
the pH was increased from 7.9 to 11.53. The decrease in removal efficiency as the pH
was increased above 7.9 was probably due to carbonate scavenging. Ozone/peroxide
treatment reaction rates are low at reduced pH's because. fewer hydroperoxide ions are

present, as shown:

111,02 + H20 .. 110; + 1130* pK = 11.6 4-7

Below a pH of 11.6, any decrease in pH will result in a decrease in hydroperoxide ion
concentration. Hydroperoxide ions react with ozone faster than peroxide (Langlais et
al., 1991). Decreased hydroperoxide concentrations result in decreased hydroxyl
radical concentrations, causing lower reaction rates.

The optimal pH for the ozone/UV treatment occurred at a pH of 4.7. The reaction
rate for this treatment process did not significantly decrease at lower pH values. The
efficiency of ozone/UV treatment was only reduced at elevated pH values (above 9.0).
A 7% decrease in the pseudo first order rate constant occurred when the pH was
lowered from 4.7 to 2.3. A 45% decrease occurred when the pH was increased from
4.7 to 12.0. The decreased reaction rate at elevated pH values was probably caused

by carbonate scavenging. Ozone/UV treatment is not effected at low pH values

51

because the UV light intensity is independent of reactor pH, unlike the hydroperoxide
concentration. Even though ozone does not degrade at low pH values, the UV light is
strong enough to degrade almost all the ozone, producing similar amounts of hydroxyl
radicals.

From Figure 4.6, it is shown that above a pH of 10.5, no advantage is gained by
supplementing ozone with peroxide or ultraviolet light. As the pH is increased above
10.5, the carbonate scavenging causes the advanced oxidation treatment techniques to
become less efficient. Additional efforts to promote hydroxyl radical formation would
be a waste of time and resources.

Reaction rates for the ozone/UV, ozone/peroxide, and ozone/UV/peroxide
treatments with humic acid decreased with increasing humic acid concentrations.
Humic acid did not seem to enhance radical chain reactions with these advanced
oxidation processes, as has been observed for straight ozone treatment (Masten, 1991,
Staehelin et al., 1985). Significant trichlorobenzene removal was obtained with 2 mg/l
humic acid. Typical humic acid concentrations for groundwater range from 003-010
mg carbon per liter. Lake and river water range from 0.5 to 4.0 mg carbon per liter
(Thurman, 1986). These concentrations can be converted to mg/l by assuming humic
acid is 30-50% organic carbon.

The ozone treatment reaction rate increased when 2 mg/l humic acid was used. A
two tailed t test determined the removal efficiencies for ozone treatment with 0.0 and 2
mg/l humic acid were statistically different based on the standard deviations of the five

replicate samples taken for each process. The reaction rate increased 26% and

52

decreased 16.8% when humic acid concentrations of 2 mg/l and 10 mg/l were used,
respectively. This agrees with the work by Masten and Staehelin, who reported that
commercial humic acid may propagate radical chain reactions. Ozone treatment was
the only process that showed this radical chain reaction propagation. This is probably
because the steady state ozone concentrations for the other treatments were very low
(in the range of 0.2 to 0.4 mg/l). The steady state ozone concentration for the straight
ozone treatment was about 2.0 mg/l. The ozone was readily available to the humic
acid in the ozone treatment process. This allowed sufficient time for the two
compounds to react and propagate radical chain reactions. The other advance
oxidation processes initiated ozone decomposition faster than humic acid, so no
increase in reaction rate was observed when small amounts of humic acid were
present.

As the reaction rates for all the processes studied decreased with increasing
bicarbonate concentration, it is apparent that hydroxyl radicals play a significant role in
the oxidation of trichlorobenzene using advanced oxidation processes. The ozone
treatment reaction rate decreased 93.5% when the bicarbonate concentration was
increased from 0 to 10 mM. Significant removal rates were still obtained for the
ozone/UV, ozone/peroxide, and ozone/peroxide/UV treatment methods with a
bicarbonate concentration of 2 mM. Bicarbonate concentrations in natural water vary.
Paillard et al., (1987), reported specific ground water, river water, and pond water
bicarbonate concentrations of 1.3 mM, 2.35 mM, and .2 mM respectively.

Table 4.4 shows the hydroxyl radical concentration calculated from equation 4.6 to

53

be very small. However, Glaze et al. (1989), reports that advanced oxidation treatment
of organic substrates are practical with steady state hydroxyl radical concentrations in
the range of 10'‘0 - 10"2 M. The ozone/peroxide treatment process had the highest
estimated hydroxyl radical concentration, whereas the straight ozone treatment had the
lowest.

The calculated reaction rate constants and hydroxyl radical concentrations indicate
the ozone/hydrogen peroxide treatment was the most efficient advanced oxidation
process for degrading 1,3,5-trichlorobenzene under the given experimental conditions.
Glaze et al. (1987), reports that the ozone/peroxide process yields more hydroxyl
radicals, is relatively cost effective, and is easier to adapt to current water treatment
designs. Ozone/UV may be more effective in removing organics if they significantly
undergo direct photolysis in the UV range.

4.6 CONCLUSIONS

Reaction rates for each treatment process are highest around a pH of 7.0.
Ozone/UV treatment reaction rates are pH independent in the range of 2 to 7. At
elevated pH the reaction rates decrease significantly. Compared to ozone/UV
treatment, the reaction rates for trichlorobenzene oxidation decreased more
dramatically in the ozone and ozone/peroxide treatment systems as the reactor pH
deviated from the optimal pH of 7.0 . No advantage is gained by supplementing
ozone with peroxide or ultraviolet light above a pH of 10.5.

Humic acid acts as both a hydroxyl radical promoter and scavenger. Ozone

reaction rates can increase in the presence low humic acid concentrations. Otherwise,

54

reaction rates decrease with increasing humic acid and bicarbonate concentrations.
Significant trichlorobenzene removals were still obtained with humic acid and
bicarbonate concentrations of 2 mg/l and 2 mM, respectively.

Based on an indirect determination of hydroxyl radical concentration, the
ozone/peroxide treatment was found to generate more hydroxyl radicals, making it the
chosen treatment process to remove dissolved 1,3,5-trichlorobenzene from waste

streams. The optimal input peroxide concentration for this process is 60 11M.

CHAPTER V

CONCLUSIONS

5.1 CONCLUSIONS

Ace Glass supermix photochemical reactor # 7868 (Reactor A) proved to be
sufficiently mixed, supporting continuous flow through stirred reactor (CFSTR) criteria
for reactor retention times greater than or equal to five minutes. Ace Glass
photochemical reactor # 7863 (Reactor B) could only support CFSTR modeling criteria
in series with a plug flow reactor. Mixing studies proved to be an essential first step
in continuous flow reactor studies. The reactor first chosen looked as though it could
easily satisfy CFSTR modeling criteria. Tracer studies proved otherwise and we were
forced to modify the system so it could be accurately modeled.

The method developed for measuring dilute concentrations of hydrogen peroxide
in the presence of ozone using the peroxidase catalyzed oxidation of N,N- diethyl-p-
phenylene diamine proved to be fast, accurate, and reproducible. Hydrogen peroxide
samples were accurately measured in the range from 1 to 500 nM, with a lower
detection limit of 26 nM. This method worked well in the advanced oxidation studies.
Total automation could easily be achieved by incorporating an in line nitrogen bubbler.
This would allow for continuous peroxide monitoring and eliminate the need to
manually sample the reactor effluent.

Advanced oxidation studies indicate ozone, ozone/UV, and ozone/peroxide

treatment processes produce optimal oxidizing environments around a pH of 7.0.

55

56

Ozone/UV treatment reaction rates are not effected until elevated pH levels are
reached, where reaction rates decrease significantly. Ozone and ozone/peroxide
reaction rates decrease at pH levels above and below 7.0. No advantage is gained by
supplementing ozone with peroxide or ultraviolet light above a pH of 10.5.

Humic acid acts as both a hydroxyl radical promoter and scavenger. Ozone
reaction rates can increase in the presence low humic acid concentrations. Otherwise,
reaction rates decrease with increasing humic and bicarbonate concentrations.
Significant trichlorobenzene removal with humic acid and bicarbonate concentrations
of 2 mg/l and 2 mM, respectively, is achievable.

The ozone/peroxide treatment process was more efficient for the oxidation of
1,3,5-trichlorobenzene when compared to the ozone, ozone/UV, and
Ozone/UV/peroxide treatment processes. The optimal input peroxide concentration for
this process was found to be 60 11M.

5.2 FUTURE RESEARCH

Based on the conclusions drawn from this research, the following areas should be
researched:

1) Hydroxyl radical concentrations must be determined to accurately determine
the reaction rate constant with trichlorobenzene. Hydroxyl radical probe analysis
should be reviewed and implemented for this purpose.

2) Treatment process rate constants should be verified by repeating experiments at
different retention times.

3) Other recalcitrant compounds should be studied to support the conclusions

57

of this research.

4) Reactor peroxide and ozone sampling techniques should be simplified and fully
automated. More treatment processes could then be investigated in a given
experiment.

5) Product identification studies should be included in future research.

LIST OF REFERENCES

LIST OF REFERENCES

Bader H., Sturzenegger V., and Hoigné J. (1988) Photometric Method for the
Determination of Low Concentrations of Hydrogen Peroxide by the Peroxidase
Catalyzed Oxidation of N,N-diethyl-p-phenylenediamine (DPD). Wat. Res. 22, 1109-
1115.

Bader H. and Hoigné J. (1982) Determination of Ozone in Water by the Indigo
Method; A Submitted Standard Method. Ozone Sci. Eng. 4, 169-176.

Baga A., Johnson G., Nazhat N., and Saadalla-Nazhat R. (1988) A Simple
Spectrophotometn'c Determination of Hydrogen Peroxide at Low Concentrations in
Aqueous Solutions. Anal. Chem. Acta. 204, 349-353.

Baxendale J .H. and Wilson J .A. (1957) The Photolysis of Hydrogen Peroxide at High
Light Intensities. Trans. Farad. Soc. 53, 344-356.

EPA (1990) SITE Program Applications Analysis Report, ULTROX International
Ultraviolet Radiation/Oxidation Technology, EPA-540/A5-89-012, US. Environmental
Protection Agency, Washington, DC..

Farhataziz T. and Ross AB. (1977) Selective Specific Rates of Reactions of Transients
in Water and Aqueous Solutions. Natl. Stand. Ref. Data Ser. (U. S. Natl. Bur. Stand.)
59.

Glaze W.H. and Kang J.W. (1989) Advanced Oxidation Processes. Description of a
Kinetic Model for the Oxidation of Hazardous Materials in Aqueous Media with

Ozone and Hydrogen Peroxide in a Semibatch Reactor. Ind. Eng. Chem. Res. 28,
1573-1580.

Glaze W.H., Kang J.W., and Chapin DH. (1987) The Chemistry of Water Treatment
Processes Involving Ozone, Hydrogen Peroxide and Ultraviolet Radiation. Ozone Sci.
Eng. 9, 335-352.

Glaze W.H., Peyton G.R., Huang F.Y., Burleson J.L., and Jones RC. (1980) Oxidation

of Water Supply Refractory Species by Ozone with Ultraviolet Radiation. EPA-600/2-
80-110. US. Environmental Protection Agency, Washington, DC.

58

59

Guittonneau S., De Latt J., Duguet J .P., Bonnel C., and Doré M. (1990) Oxidation of
Parachloronitrobenzene in Dilute Aqueous Solution by O3 + UV and H202 + UV : A
Comparative Study. Ozone Sci. Eng. 12, 73-94.

Karlberg B. and Pacey G. (1989) Flow Injection Analysis, a Practical Guide. Elsevier,
New York.

Khan S.R., Huang CR, and Bozelli J.W. (1985) Oxidation of 2-Chlorophenol Using
Ozone and Ultraviolet Radiation. Environ. Frog. 4, 229-238.

Kieber R. and Helz G. (1986) Two Method Verification of Hydrogen Peroxide
Determinations in Natural Waters. Anal. Chem. 58, 2312-2315. TF1, ,1 9 (my... 1: L. ) ,

Kirk P.W.W., Rogers HR, and Lester J .N. (1989) The Fate of Chlorobenzenes and
Permethrins During Anaerobic Sewage Sludge Digestion. Chemosphere. 18, 1771-
1784.

Lamporski L.L., Langhorst M.L., Nesterick T.J., and Cutie S. (1980) J. Assoc. Oflic.
Anal. Chem. 63, 27-32.

Langlais B., Reckhow D.A., and Brink DR. (1991) Ozone in Water Treatment:
Application and Engineering. Lewis Publishers, Chelsea, MI.

Legube B., Guyon S., Sugimitsu H., and Doré M. (1983) Ozonation of Some Aromatic
Compounds in Aqueous Solution : Styrene, Benzaldehyde, Naphthalene,
Diethylphthalate, Ethyl and Chlor Benzenes. Ozone Sci. Eng. 5, 151-170.

Masten SJ. (1991) Ozonation of VOC's in the Presence of Humic Acids and Soils.
Ozone Sci. Eng. 13, 287-313.

Masten SJ. and Hoigné J. (1992) Comparison of Ozone and Hydroxyl Radical-Induced
Oxidation of Chlorinated Hydrocarbons in Water. Ozone Sci. Eng. 14, 197-214.

Murov S.L. (1973) Handbook of Photochemistry, M. Dekker, New York.
Nauman EB. and Buffham BA. (1983) Mixing in Continuous Flow systems. Wiley,
New York.

Paillard H., Brunet R., and Doré M. (1988) Optimal Conditions for Applying an
Ozone-Hydrogen Peroxide Oxidizing System. Wat. Res. 22, 91-103.

6O

Paillard H., Brunet R., and Doré M. (1987) Application of Oxidation by a Combined
Ozone/Ultraviolet Radiation System to the Treatment of Natural Water. Ozone Sci.
Eng. 9, 391-418.

Peyton G. and Glaze W. (1988) Destruction of Pollutants in Water with Ozone in
Combination with Ultraviolet Radiation. Environ. Sci. Technol. 22, 761-767.

Peyton G. and Glaze W. (1987) Mechanism of Photolytic Ozonation. Photochemistry
of Environmental Aquatic Systems. ACS Symposium Series 327, 76-87.

Prat C., Vicente M., and Esplugas S. (1988) Treatment of Bleaching Waters in the
Paper Industry by Hydrogen Peroxide and Ultraviolet Radiation. Wat. Res. 22, 663-
668.

Rfizicka J. and Hansen EH. (1981) Flow Injection Analysis. Wiley, New York.

Staehelin J. and Hoigné J. (1985) Decomposition of Ozone in Water in the Presence of
Organic Solutes Acting as Promoters and Inhibitors of Radical Chain Reactions.
Environ. Sci. Technol. 19, 1206-1213. M H J g
Thurman EM. (1986) Organic Geochemistry of Natural Waters. Martinus Nijhoff/Dr.
W. Junk Publishers, Boston.

Wagner R. and Ruck W. (1984) Die Bestimmung von Wasserstoffperoxid und anderen
Peroxyverbindungen. Z. Wass. Abwass. Forsch. 17, 262-267.

Xu Y.M., Ménassa P., and Langford C. (1988) Photodecomposition of Several
Chloroaromatics Using a Commercial Prototype Reactor. Chemosphere. 17, 1971-
1976.

APPENDICES

APPENDIX A

INDIGO BLUE METHOD FOR SAMPLING AQUEOUS OZONE

Appendix A. Indigo Blue Method For Sampling Aqueous Ozone

Intr ti n

Reactor ozone concentrations are measured to gain insight on what chemical
reactions are taking place in solution and are needed to calculate the rate constants for
advanced oxidation process. An indigo blue method for ozone determination similar to
the one proposed by Bader (as before referenced) is used. Aqueous ozone quickly and
stoichiometrically decolorizes indigo trisulfonate so that the change in absorbance of
an indigo blue dye solution can be correlated to an aqueous ozone concentration with
the aid of a calibration curve.

Calibration Curve Procedgge

An aqueous ozone solution of 1.2 mg/l was generated by bubbling ozone gas into
an ozone contactor filled with deionized water acidified to a pH of 2.0 with phosphoric
acid. This solution concentration was determined by direct UV measurement at 260
nm with a molar absorptivity of 3000 M"cm".

Indigo blue stock solution with an absorbance of 1.00 at 600 nm was prepared in
deionized water acidified to a pH of 2.0 with phosphoric acid. 100 ml of this solution
was pipetted into 150 ml vacuum flasks that contained mini stir bars. Samples were
withdrawn from the ozone contactor into the flasks by creating a vacuum in the flask
with a hand operated vacuum pump. Teflon tubing, pierced through rubber stoppers,
carried the ozonated sample from the contactor to the flask. The flasks were

positioned on a magnetic stirrer to provide adeQuate mixing while sampling. The

61

62

flasks were weighed before and after sampling to determine the sample volume. The
absorbance of the indigo blue solution was taken after sampling. An ozone calibration
curve was generated from this data by calculating the change in absorbance due to the
dilution effect of the sample and subtracting this from the total change in absorbance
in the indigo solution. The difference equals the change in absorbance due to ozone
consumption. This value was plotted against the mg of ozone consumed to generate
the indigo blue ozone calibration curve.
Ratr zon arnl Proce r

Reactor ozone samples were withdrawn from the reactor as described in the
ozone/indigo blue calibration method. The sample line was positioned below the
reactor effluent port. The flasks were weighed before and afier sampling to determine
the sample volume, and the change in absorbance of the indigo blue solution due to
ozone consumption was compared to the calibration curve to obtain the weight of
ozone in the reactor sample. This weight was divided by the volume of ozone sample
to determine the reactor ozone concentration. The ozone calibration curve was not
performed for each experiment due to time constraints. The slope of the ozone
calibration curve was assumed to stay constant.
33311::

Figure A.1 shows the ozone calibration curve using the indigo blue method. A
linear relationship exists between the change in absorbance due to ozone consumption
and amount of ozone added to the indigo solution. The results indicate a 0.0327

decrease in absorbance for every 10 11g of ozone added to the indigo blue solution.

CHANGE IN ABSORBANCE 0F INDIGO BLUE SOLUTION
(OZONE CONSUMPTION ONLY)

lJlllllllllIllllll

 

63

Y-3.275 X
CORRELATION SOEFFICIENT =- .999

 

0.00

I I I I 1 I I I r ' 1 I

0.01 0.02 0.03 0.04 0.05 0.06
AMOUNT or OZONE ADDED (mg)
Figure A.1 Indigo Blue/Ozone Calibration Curve

64

Di i n

The indigo blue method is good for measuring reactor ozone concentrations, but it
is time consuming and requires many manual operations. A more efficient way to
measure reactor ozone concentrations would be to set up an automated flow injection
analysis system that would continuously monitor the ozone by injecting reactor
samples into a flowing stream of indigo solution while monitoring the absorbance on a
spectrophotometer. This would eliminate weighing the flasks before and afier
sampling, and also eliminate pipetting 100 ml of indigo solution for each reactor

sample taken.

APPENDIX B

TRICHLOROBENZENE TRACER STUDY DETERMINING TIIWE TO REACH

STEADY STATE

Appendix B. Trichlorobenzene Tracer Study Determining Time to Reach Steady

State

Intr tion
Time for trichlorobenzene to reach steady state in the reactor was determined.
This experiment supplements the methylene blue tracer analyses. A ten minute reactor

retention time was used. Results from the dye tracer studies indicated 30 minutes

A“ 11‘”-

would be required for the reactor to reach steady state. The extent to which TCB was

degraded by ultraviolet light and peroxide alone, without ozone, was also determined.

 

M

The photochemical reactor was initially filled with TCB free deionized water.
The reactor effluent was sampled to determine if TCB was desorbing from the surfaces
in the reactor. TCB solution and deionized water were then pumped in, each at a rate
of 12.6 ml/min. Reactor effluent samples were taken in triplicate every 15 minutes for
the first hour, and then every 30 minutes afler that, for a total of 5 hours. After five
hours, the UV light was turned on and triplicate samples were taken after one hour.
The UV light was then turned off and peroxide was pumped into the reactor to obtain
a concentration of 60 1.1M. Samples were again taken after 60 minutes to determine if

TCB reacted with hydrogen peroxide alone. The reactor solution was kept at a pH of

7.0 i 0.2.

65

66

Ram

Figure B.1 indicates the reactor needs 60 minutes (or 6 retention times) to reach a
steady state concentration. The UV light alone reduced the reactor TCB concentration
from 1371 ppb to 694 ppb (49.4% removal). Hydrogen peroxide reduced the reactor
TCB concentration from 1316 ppb to 1048 ppb (20.3% removal).
Di 8 i n

The reactor took twice the amount of time originally predicted by the dye tracer
study. Adsorption of TCB to the tubing or reactor walls or even losses to the air may
account for this. Six reactor retention times should be allowed to reach steady state.
Figure 8.1 also shows the reactor concentration does not fluctuate very much when
steady state is achieved. This indicates the pumps and mixing motor used in the these
studies are capable of maintaining steady state and complete mixing conditions. TCB
can be significantly degraded by UV light. Hydrogen peroxide alone was shown to
remove a small portion of TCB. This removal is questioned, because the peroxide was
pumped into the reactor right after the UV light was turned off. Significant amounts
of ozone may have been generated in solution (while the UV light was on) which may

have reacted with the peroxide to slow the return of TCB back to steady state.

1.3.5 - TRICHLOROBENZENE CONCENTRATION
(ppb)

1 600
1

1400-4

‘

1200-1

.1

1000-

-

800 —

q

600 -—

d

400 -

q

200 -

 

67

 

0

30

60 90 120

0% I I I I I I IiIiI I I I I I I I I I I I I I I

150 180 210 240 270 300 330 360

TIME AFTER INITIAL TCB INPUT INTO REACTOR

Figure 8.1

(MINUTES)

Trichlorobenzene Tracer Study

APPENDIX C

TRICHLOROBENZENE, OZONE AND PEROXIDE SAMPLING SUMMARY

FOR EACH EXPERIMENT

Ul-bQN-A

DONG

10

.1112
std:
at“
at“
corn. 8.8.
MRI can .
STD. m. .
SID. m.

museum
mucosa.
moat...
8mm.
ism.
01.8mm.

68

Exp. 1 Hydrogen Peroxide Concentration Optimization

971.32
26.814
33.289

126.9
3.4238
4.2506
13.065
0.3914

BMW“ WIOS

11
12
13
14
_ 15

avaoaua.
mom.

mm.
96m-
$8133“-

457.6
9.5872
11.902
47.111

1.096

”MIMI WIO3

1.6
17
18
19
20

mam.
510.061.

mm.
36m-
some“.

60.049
2.6975
3.3489
6.1822
0.3084

H202:.06mfl V003

21
22
23
24
25

11mm.
STEM.-

mm.
95m.
9587038!-

21.793
2.6816
3.3291
2.2437
0.3066

112028.008lnfl WIO3

26
27
28
29
30

avaoaua.
$10.81!.-
5mm.
arm.
$8199!!-

«a

. I"

'pr

47.294
1.8799
2.3339
4.8691
0.2149

2244.8

Table 0.1 Exp. 1 GO Data Summary

came
(99“)

37.41
96.47
436.7
1381

3006

935.4
993.9
978.4
996.9
952.1

127.8
120.9
129.5
128.3
128

441.2

463.8

463
463.1

456.9

61.2
55.27
61.41
61.69
60.67

20.26
23.32
20.61
25.71
19.07

46.87
45.43
47.78
46.12
50.27
54.2
2295
2195

cm

2961705
8921665
38248128

122249952
269395712

92422176
99179664
95446008
98296776
92251872

12440268
12113525
12865593
12710747
12197710

42471692
42555864
42032472
40053808
36278880

5660363
5233329
5647918
5765367
5421097

1814124
2136911
1917696
1993744
1700024

4191256
4168146
4212184
4274664
4267442
6023783

206543600
191794144

49922312
58326904
55231944
55846024
56527916

46712716
47177468
46120144
46615956
45809892

46004872
47371992
46969488
46828292
45065344

45510072
43377206
42918116
40888912
37542720

43727584
44764976
43477904
44183264
42241016

42328408
43329008
43990656
36668112
42136792

42277600
43373260
41677960
43821128
40131440
52547812
42549424
41315664

C.M.A

0.06
0.15
0.69
2.19
4.77

1.98
2.1
2.07
2.11
2.01

0.27
0.26
0.27
0.27
0.27

0.93
0.98
0.98
0.98
0.97

0.13
0.12
0.13
0.13
0.13

0.04
0.05
0.04
0.05
0.04

0900
L....._.

0.11
0.11
4.85
4.64

1.“

626.9
626.9
626.9
626.9
626.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9
469.9
469.9
469.9

acne:
(STD)

31.04
77.59
387.8
1163
3094

0.01.0.

0.05
0.12
0.62
1.85
4.94

C.M.A

0.059
0.153
0.693
2.189
4.766

69
Exp. 1 (cont.)

Table C.2 Exp.1 Ozone and Peroxide Sampling Data

WWW-IT salesman Stout same we mu mu mu m.
a mm mm sounm vouue Assam. 11me male mug me
m m Assam. (ml) DUEToer room mm can.
Dlumoumo 81%: mm 1119/1
WIN
1 201.88 225.15 1.06 23.3 0.85 0.2095 0.2 0.0095 0.12
2 201.17 217.97 1.06 16.8 0.89 0.1685 0.1524 0.01611 0.29
3 201.39 214.72 1.06 13.3 0.93 0.1335 0.1246 0.00888 0.20
4 200.93 218.02 1.06 17.1 0.89 0.17 0.1546 0.01536 0.27
5 204.93 224.66 1.06 19.7 0.87 0.185 0.1746 0.01041 0.16
6 200.97 218.19 1.06 17.2 0.88 0.175 0.1556 0.01936 0.34
7 201.26 215.98 1.06 14.7 0.89 0.1665 0.1359 0.03055 0.63
8 201.6 219.54 1.06 17.9 0.87 0.1855 0.1612 0.02434 0.41
9 202.23 221.34 1.06 19.1 0.86 0.1975 0.17 0.027510.44
10 201.77 218.97 1.06 17.2 0.8 0.2605 0.1555 0.10501 1.86
11 201.04 212.51 1.06 11.5 0.88 0.177 0.109 0.067981.81
12 203.88 222.66 1.06 18.8 0.8 0.2585 0.1675 0.090991.48
PEROXIDE SAMPUNG DATA
H202 AVBVGE PM
saw. mm can
I (MMBLAM) uM/l
1-1 0.209007 6751
1-2 0.209673 6772
2-1 0.197673 638
2-2 0.206673 668
3-1 0.128673 41.6
3-2 0.147113 47.5
4-1 0.016007 5.17
4-2 0.013673 4.42
SUMMARY
W mm mm STD NLET
m more one am CED!
can can can. (mg/I) cave.
(111‘) 11202 (1.1M) 11202 (mg/I) (mail)
6000 6761.628 0.21 +/- 0.20843 7.68
600 653.0146 0.26 +/- 0.22789 7.68
60 44.53919 0.5 +/- 0.29755 7.68
6 4.793282 1.72 +/- 0.516 7.68

MAUN‘

IDQNO)

11
12
13
14
15

16
17
18
19
20

21
22
23
24
25

26
27
28
29
30

70

Exp. 2 Aqueous Ozone Treatment at Varying pH (Low Range)

915mm.

1427.7
56.132
69.685

91137.17
126.49
4.6559
5.7802
8.8599
0.3621
91136.21
160.94
14.849
18.435
11.273
1.155
P113535
288.26
14.225
17.66
20.191
1.1064
Ph82.,2
837.13
13.422
16.662
58.635
1.0439
P118224
789.1
16.013
19.88
55.271
1.2455

Table 0.3 Exp. 2 (30 Data Summary

GOOD
(99b)

32.141
85.594
362.8
1083.7
3100.4

1383.1
1486.2
1398.6
1491.1
1379.5.

133.28
127.64
125.95
120.38
125.21

163.79
174.45
135.57
167.37
163.52

276.31
275.81
287.13
310.63
291.42

845.85
831.77
836.6
818.22
853.19

793.13
812.26
791.07

769.84.

779.19

cm

1975906
5866974
23022056
71390440

234932816

143878512
143355744
143718304
147467424
138833216

13571924
12622064
12785809
12097206
12279910

16634004
17529516
13813850
16402914
16255591

26761748
27052184
30714152
32293076
29993116

85001144
82042640
82158448
80522792
84005648

78284424
76242672
75634920
73045480
73450624

1m

39485368
44025408
40757844
42313000
48669388

50087680
46444244
49478256
47617308
48456772

49030068
47613416
48877568
48386956
47220532

48897928
48383016
49060500
47188112
47866048

'46633156

47224792
51504544
50054676
49554764

48385184
47492168
47284520
47383880
47407364

47524352
45194464
46035092
45685368
45387168

C.M.A

0.05
0.13
0.56
1.69
4.83

. 2.87
3.09
2.9
3.1
2.87

0.28
0.27
0.26
0.25
0.26

0.34
0.36
0.28
0.35
0.34

0.57
0.57
0.6
0.65
0.61

1.76
1.73
1.74
1.7
1.77

1.65
1.69
1.64
1.6
1.62

626.9 31

626.9 77.6
626.9 388
626.9 1163
626.9 3094

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

0.05
0.12
0.62
1.85
4.94

. . C.NI.A

0.05
0.133
0.565
1.687
4.827

71

Exp. 2 (cont)

Table 0.4 Exp. 2 Ozone Sampling Data

ME MEG-IT ”Rm STCXX SAME SAME MN
mm AFTERSAMlE

d—Ld-‘d-fi
upwméoomummawmd

NW

202.08
201.89
203.09
200.25
204.77
204.87
200.81
205.96
201.45
199.84
205.08
204.41
201.71
200.54
200.42

mm
mane

0GVQM§WN¢

10

can
("'0")
1.78109
1.71515
1.72403
2.17402
2.51667
2.2719
4.72882
3.05078
3.04073
4.54138
4.33755
4.60412
4.6898
5.34344
4.67891

mm

215.96
216.34
218.46
222.32
220.65
222.95
213.05
224.32
219.15
222.7
227.1
223.6
218.91
212.31
222.12

AM
Fm
manna
("10")
1.74009
2.32086
3.60678

4.49435

4.90405

saunau voume m mace
Mason (ml) custom
011mm
mm»
0.973 13.88 0.758 0.2155
0.973 14.45 0.753 0.22
0.973 15.37 0.74 0.2335
0.973 22.07 0.609 0.364
0.973 15.88 0.683 0.29
0.973 18.08 0.663 0.31
0.973 12.24 0.64 0.333
0.973 18.36 0.603 0.3705
0.973 17.7 0.616 0.3573
0.973 22.86 0.385 0.588
0.973 22.02 0.423 0.55
0.973 19.19 0.47 0.503
0.973 17.2 0.514 0.459
0.973 11.77 0.624 0.349
0.973 21.7 0.402 0.5715
SUMMARY
$10 1111 NET
am one
can
(mail)
11.92
+/- 0.089 7.17 11.92
11.92
11.92
+/- 0.437 6.21 11.92
11.92
11.92
+/- 2.408 5.25 11.92
11.92
11.92
+/- 0.345 2.92 11.92
11.92
11.92
+/- 0.943 2.24 11.92

11.92

0.1186
0.1228
0.1296
0.1759
0.1333
0.149
0.1061
0.1509
0.1463
0.181
0.1756
0.1567
0.1428
0.1025
0.1735

“BTW
m
("‘0")

0.149
0.149
0.149
0.149
0.149
0.149
0.149
0.149
0.149
0.149
0.149
0.149
0.149
0.149
0.149

MN
MOE
ram
0013mm

0.09691
0.09715
0.10387
0.18808
0.15666
0.16102
0.22689
0.21957
0.21098
0.40696
0.37441
0.34634
0.3162
0.24654
0.39801

 

macaw-

(DQNIQ

11
12
13
14
15

16
17
18
19
20

21
22
23
24
25

26
27
28
29
30

Exp. 3 Aqueous Ozone Treatment at Varying pH (High Range)

6142
at“
at“
I148
U“

can.“
matte.-
amid-v-

“.m-

“out,
mm.-
child-Uda-
ulterior.
91mm.
16‘“
a..."
mute.-
chad-Iden-
“.m-
91mm.
*“m-
meaty
mule.-
chad-Iden-
“tenor.
96mm.
16‘“-
.0..."
more;
amt.-
“.m-
19mm.
96“.!"1-
”out,
wet-gem-
smou-
910.00!!-
Swen-ring.

96.1mm-

1570.9
64.45
80.013

le7.1.
106.91
4.9684
6.1681
6.8057
0.3512
PHI7.'
115.72
1cx77
13.371
7.3661
0.7613
pit-18.68
187.39
13.73
17.045
11.929
0.9705
{11189.48
132.21
8.6582
10.749
8.4162
0.612
phi-11.83
197.73
17.006
21.112
12.586
1.2021

72

Table 0.5 Exp. 3 GO Data Summary

can:
(996)

33.495
76.975
387.33
1221.5
3072.1

1662
1527.2
1611.2
1548.8
1505.5

110.29
105.43
106.62
112.59
99.644

122.9
128.36
106.35
103.07

117.9

164.16
196.89
198.36
188.74
188.81

121.85
132.59
125.44
142.35
138.83

200.28
188.18
215.87
210.42
173.88

cm

2803429
6179142
33576728

104313592
269110688

166005696
144497520
152393984
152831728
146513824

10452030
9979315

10363486

10553177
9642318

11473997
12361594
10170830
9858496
10898346

15660004
18585460
19993348
17817678
17379028

11557052
12067652
13055726
13115812
13372054

18444356
17659184
19442464
19167136
16309344

1m

50046320
47999204
51834216
51062748
52379096

44770176
4241 1552
42394960
44230572
43622784

42478824
42426620
43569956
42013664
43374620

41846668
43165236
42866916
42873256
41435036

42759412
42310392
45180056
42314136
41257232

42512932
40796608
46650692
41298796
43174544

41279348
42063180
40370128
40829792
42043590

C.M.A

0.06
0.13
0.65
2.04
5.14

3.71
3.41
3.59
3.46
3.36

0.25
0.24
0.24
0.25
0.22

0.27
0.29
0.24
0.23
0.26

0.37
0.44
0.44
0.42
0.42

0.27
0.3
0.28
0.32
0.31

0.45
0.42
0.48
0.47
0.39

1m

626.9
626.9
626.9
626.9
626.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

com:

31.04
77.59
387.8
1163
3094

0.64.6.

0.05
0.12
0.62
1.85
4.94

CMA

0.06
0.13
0.65
2.04
5.14

Exp. 3 (cont)

73

Table 0.6 Exp. 3 Ozone Sampling Data

SAME WWII-IT WWII-IT
9mm AFTERSAIPLE

CDQNOIO'I'hQN-A

10

9126110111

205.4
203.28
201.15
199.75

199.9
201.19
202.73

202.8
200.23
204.19
202.71
203.95

204.3
202.84
200.37

mm
mm:

@mNGMhQN-‘t

am.
("10")

1.685612
1.702461
1.370858
1.049029
0.993051
1.256037
0.183866
0.150589
0.173218
0.835945
0.858618
0.891389
-0.03577
-0.02298
-0.0252

Mm

224.26
221.94
222.6
214.71
222.32
217.84
217.84
218.03
213.55
219.94
221.2
218.57
217.74
226.54
222.18

AM

manta
(mall)

1.58631

1.09937

0.16922

0.86198

-0.02798

5m SAME SAME MN

sownau
m

0.995
0.995
0.995
0.995
0.995
0.995
0.995
0.995
0.995
0.995
0.995
0.995
0.995
0.995
0.995

ME

(""1

18.86
18.66
21.45
14.96
22.42
16.65
15.11
15.23
13.32
15.75
18.49
14.62
13.44
23.7
21.81

0.71
0.71
0.7
0.8
0.73
0.77
0.85
0.85
0.87
0.81
0.78
0.82
0.88
0.81
0.82

SUMMARY

+/-

+/-

+/-

+/-

+/-

810
m

0.463

0.343

0.042

0.069

0.017

#4

7.16

7.9

10.6

9.48

mats
DLETOWTH
Diumauam
mm
0.2825
0.281
0.291
0.191
0.2695
0.224
0.1415
0.1405
0.126
0.187
0.2175
0.178
0.116
0.1885
0.176

11%

("19")

12
12
12
12
12
12
12
12
12
12
12
12
12
12
12

0.157881
0.15647
0.175733
0.129482
0.182224
0.142021
0.130609
0.13151
0.116956
0.135389
0.155267
0.126914
0.117884
0.190635
0.178154

mm:
m
IMO/II

0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15
0.15

0.12462
0.12453
0.11527
0.06152
0.08728
0.08198
0.01089
0.00899
0.00904
0.05161
0.06223
0.05109
-0.0019
-0.0021

-0.0022

(neuron-

(DONG

11
12
13
14
15

16
17
18
19
20

8101
6142
ma
at“
at“

CWT. 8.8.

was-
stall-i dev-

01‘.“-

ozone/11202
ever-.- one.-
ew dov-
no. error a
19 m
11. ed enor-

ozonelhzoz
m m.-
1w dev-

etd. error .

18 ed enu-
ozone/11202
ever-go m.-
ew du-

Steiner-

Exp. 4 Ozone/Peroxide Treatment at Varying pH

1425.6
72.92
90.527

[11186.96
36.452
2.9306
3.6382
2.557
0.2283

pH=11.5
316.2
20.239
25.126
22.18
1.5765

P118..53
36.675
3.8865
4.8249
2.5727
0.3027

74

Table 0.7 Exp. 4 60 Data Summary

coon
(P95)

31.827
80.39
381.25
1185
2612.8

23.717
27.119

1492.9
1384.4
1356.3
1378.9
1515.3

41.237
36.578
33.415
35.036
35.992

285.56
341.71
318.49
313.14
322.09

39.826
41.53
34.851
35.085
32.085

cm

2858768
7002482
34451760

1 10240800
248349344

1855054
2079382

155769776
141603632
142690208
146444112
151491296

4385030
3915893
3566565
3707843
3783736

29608656
35700320
34067220
31625920
32167694

3625998
4389139
3490700
3617496
3354322

11111511

52477424
50890064
52794188
54349472
55531640

45695532
44795960

45202396
45993804
46077108
46513488
43784256

46571176
46885568
46745648
46349448
46041744

45409760
45756248
46846428
44232408
43740088

39874172
46286276
43865984
45156520
45786324

C.M.A

0.054
0.138
0.653
2.028
4.472

0.041
0.046

3.409
3.161
3.097
3.148
3.46

0.094
0.084
0.076
0.08
0.082

0.652
0.78
0.727
0.715
0.735

0.091
0.095
0.08
0.08
0.073

1005

626.9
626.9
626.9
626.9
626.9

626.9
626.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

31
77.6
388
1163

3094

3094
3094

001.0.

0.05
0.12
0.62
1.85
4.94

4.94
4.94

C.M.A

0.05
0.14
0.65
2.03
4.47

0.04
0.05

21
22
23
24
25

26
27
28
29
30

31
32
33
34
35

36
37
38
39

marine.-
IMM-
«error.
19mm-
indemn-

ozone/h2o2
everep aorta.-
nanauo dev-
etd. error .
11 m

168derror1-

ozone/11292
everep eerie.-
eterldero dev-
etd. error .
'1. reuniting-

$etderror-

ozone/11202
average cone.-
umu dev-
etd. error a
as may...

$811th

194.7091
9.052153
11.23793
13.65839
0.705112

9" 810.1
113.896
34.51431
42.8483
7.989542
2.68847

91184.94
77.67928
9.15239
11.36237
5.449021
0.712919

[11182.35
460.9127
31.04475
49.39219
32.33196
4.000744

pumwmnmm

75

Exp. 4 (cont)

cm
(POD)

196.351
180.429
194.877
196.287
205.601

104.927
85.5047
174.015
100.867
104.166

77.7992
74.5431
81.3044
89.863
64.8867

414.6
486.726
503.711
438.614

Table 0.? (cont)

cm

20399144
17402340
19570358
20560388
20835650

9996954
8436150
15345929
9840492
10576532

8070696
7394539
7619730
8460548
6360756

40781144
51227184
52323868
45616660

45500016
42241128
43981612
45874728
44382784

41726492
43210352
38622436
42726864
44468408

45432776
43444720
41044888
41233576
42932552

43078836
46094552
45493756
45548508

C.AII.A

0.45
0.41
0.44
0.45
0.47

0.24
0.2
0.4

0.23

0.24

0.18
0.17
0.19
0.21
0.15

0.95
1.11
1.15

IN

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9

 

blank
std1

std2
std3
std4

r
if

0
“N‘GM‘QNdQN-‘QN-‘QM-‘QN-‘QNJ

QGQNVNOO’O’MUIOI-hthOQNNN-‘d-fi

mamas
Pam
A83.
0.029
0.0443
0.12
0.301
0.473
0.0267
0.287
0.285
0.282
0.152
0.145
0.152
0.077
0.072
0.071
0.109
0.109
0.11
0.071
0.07
0.077
0.069
0.068
0.068
0.236
0.235
0.235
0.253
0.254
0.252

Exp. 4 (cont)

Table 0.8 Exp. 4 Peroxide Sampling Data

PEAK-
BLANK

0
0.0153
0.091
0.272
0.444
0
0.2603
0.2583
0.2553
0.1253
0.1183
0.1253
0.0503
0.0453
0.0443
0.0823
0.0823
0.0833
0.0443
0.0433
0.0503
0.0423
0.0413
0.0413
0.2093
0.2083
0.2083
0.2263
0.2273
0.2253

PMSTAMJARD

can
(UM/II
9.3548
14.301
38.71
97.097
152.58
,0
72.923
72.362
71.522
35.107
33.147
35.107
14.099
12.698
12.418
23.063
23.063
23.343
12.418
12.138
14.099
11.858
11.578
11.578
58.637
58.357
58.357
63.399
63.679
63.119

can

25
75
125

70

AM
can
(uM/l)

72.27
84.45
13.07
23.16
12.89
11.67
58.45

63.4

+/-

+/-

+l-

+/-

+/-

+/-

+/- _

6111 error
on

triplicate

1.7466

2.8048

2.231

0.4007

2.6275

0.4007

0.4007

0.694

CDONO’OI#0N-‘

-0.1

0.02
1.27
3.91

201.43
203.46
206.03
200.82
204.28
204.57
203.75
205.07
201.72
200.3
200.63
201.58
205.39
201.61
199.98
201.37
204.47
206.86

77

Exp. 4 (cont)

Table C.9 Exp. 4 Ozone Sampling Data

Am
SAME
Hm

218.42
219.48
220.6
215.7
222.11
221.84
219.94
223.06
217.74
220.97
215.12
214.48
223.03
216.09
218.64
211.71
218.24
220.75

0.97
0.97
0.97
0.97
0.97
0.97
0.97
0.97
0.97
0.97
0.97
0.97
0.97
0.97
0.97
0.97
0.97
0.97

1"“)

16.99
16.02
14.57
14.88
17.83
17.27
16.19
17.99
16.02
20.67
14.49
12.9
17.64
14.48
18.66
10.34
13.77
13.89

0.839
0.845
0.852
0.846
0.828
0.829
0.843
0.826
0.842
0.809
0.852
0.865
0.756
0.786
0.748
0.742
0.692
0.677

m

0.136
0.129
0.122
0.128
0.147
0.146
0.132
0.148
0.133
0.165
0.123
0.11
0.219
0.188
0.226
0.232
0.283
0.298

moo SME SAME moo mean
ST!!! W Am m m

m
OW

0.1415
0.1345
0.1239
0.1262
0.1474
0.1434
0.1357
0.1485
0.1345
0.1668
0.1233
0.1113
0.1461
0.1232
0.1532
0.0913
0.1179
0.1188

A”.
m
our
one
moron
-0.00595
-0.00549
-0.00186
0.00184
-0.00089
0.00206
-0.00422
-0.00051
-0.00199
-0.00184
-0.00077
-0.00179
0.07245
0.0648
0.07283
0.14073
0.16461
0.17871

OZONE, PEROXIDE SAMPUNG SUMMARY

STD TREAT!“ mm m

8.8. - NO OZONE

+/-
+/-
.+/-
+I-
+/-
+/-
+/-

- 3311120:
0.0955 3311-120:
0.0883 3311120:
0.0325 3311120:
0.0748 3311120:
0.2194 3311120:
0.6273 3311120:

#1

6.96
6.96
11.5
11.05
10.1
8.53
4.94
2.35

more
can.
(uM/I)

72.27
34.45
13.07
12.89
11.67
23.16
58.45
63.4

+/-
+/-
+/-
+/-
+/-
+/-
+/-
+/-

STD
m

1.7466
2.8048
2.231
2.6275
0.4007
0.4007
0.4007
0.694

NITIAI.
m
m

m

(uMn)

72.2689
72.2689
72.2689
72.2689
72.2689
72.2689
72.2689
72.2689

211

("10")

-0.1069
-0.1046
-0.0391
0.0378
-0.0152
0.0365
-0.0795
-0.0086
-0.0379
-0.0272
-0.0162
-0.0424
1.2541

1.3665
1.1918
4.1557
3.6502
3.9286

01.5me

90‘101

H
O

12
13
14
15

16
17
18
19
20

21
22
23
24
25

26
27
28
29
30

Exp. 5

Table 0.10 Exp. 5 GO Data Summary

(m1

mm 4HL88
91112 83.89
std: 416.5
“114 1326
6165 3093
00418.8.
1187
1303
1341
1347
1225
uvlozone pH=7.22
mucous.- 49.71 51.57
stooev- 3.598 43.45
swam. 4.467 50.93
11mm.- 3.882 52.43
115105118. 0.312 50.19
uvlozone pH=2.32
madam. 52.44 53.59
srooev. 2.563 48.64
5105m- 3.182 54.11
11m. 4.095 54.83
xeroem- 0.222 51.03
uvlozone P11:4.75
macaw... 41.67 39.35
STDDEV- 1.509 43.57
moan. 1.873 44.78
14m- 3.254 42.02
xsroem- 0.131 41.64
uvlozone P11=8.95
moonwa- 61.62 64.27
81009!- 4.266 66.15
sroem- 5.296 55.21
11m.- 4.811 62.61
xerosm- 0.37 59.85
uv/ozone P11=12.01
“600190.- 197.8 201.1
smoev- 10.3 202.9
smear- 12.78 208
stratum. 15.45 196.1
0.893 181.1

1281
71.45
88.7

avooouc.
STDDEV-
SIDE"!-

$870851-

3680543
6559281
34373628

1 10545328
267536528

104919264
116555576
126073256
123069312
109682648

4721910
4009085
4802362
4609044
4691511

5080594
4553978
5216417
5151349
4801756

3688140
4120523
4199960
4091923
3977700

6206876
6300970
5232202
5962949
5767900

19191324
18907244
19794552
18754024
16317085

78

MFEA

55017168
50117284
52898776
53431396
55437808

42476596
42970780
45166336
43912520
43018324

43999880
44339272
45305436
42242276
4491 7012

45553908
44990016
46324140
45140388
45210028

45030732
45443396
48303756
46787360
45900012

46404764
45768332
45539536
45765100
46309772

45851728
44779924
45717096
45949360
43297504

CMA

0.07
0.13
0.65
2.07
4.83

2.47
2.71
2.79
2.8
2.55

0.11
0.09
0.11
0.11
0.1

0.11
0.1
0.11
0.11
0.11

0.08
0.09
0.09
0.09
0.09

0.13
0.14
0.11
0.13
0.12

0.42
0.42
0.43
0.41
0.38

1001::

627
627
627
627
627

470
470
470
470
470

470
470
470
470
470

470
470
470
470
470

470
470
470
470
470

470
470
470
470
470

470
470
470
470
470

Ozone/UV Treatment at Varying pH

cacao
(STD)
(we)
31 .04
77.59
387.8
1163
3094

0.0/LC.

0.05
0.12
0.62
1.85
4.94

CNIA

0.07
0.13
0.65
2.07
4.83

-flmtm 7m'A-n‘. Ir '

79

Exp. 5 (cont.)

Table 0.11 Exp. 5 Ozone Sampling Data

mmmmmflm SAME
I mmmmmm

SAME
m

MN OMEN CHAMEN
AM” AM” more

mama nammr Mama m0 memumaromumw rmmue
mummum emu muurmm
muuwmn

1 206.34 220.48 1.01 14.1 0.875 0.1305 0.125 0.006

2 206.14 227.4 1.01 21.3 0.808 0.197 0.176 0.0208.

3 205.61 223.16 1.01 17.5 0.847 0.158 0.15 0.008

4 202.69 225.31 1.01 22.6 0.808 0.197 0.185 0.0116

5 201.34 217.8 1.01 16.5 0.851 0.1545 0.142 0.0125

6 200.48 220.13 1.01 19.7 0.828 0.177 0.165 0.0119

7 206.47 230 1.01 23.5 0.795 0.2105 0.191 0.0191

8 200.87 220.41 1.01 19.5 0.824 0.181 0.164 0.0167

9 201.06 216.7 1.01 15.6 0.854 0.151 0.136 0.0151

10 201.99 218.38 1.01 16.4 0.854 0.151 0.142 0.0095

11 202.16 224.71 1.01 22.6 0.808 0.1975 0.185 0.0126

12 204.29 222.9 1.01 18.6 0.837 0.1685 0.158 0.0108

13 201.22 217.01 1.01 115.8 0.861 0.144 0.137 0.007

14 203.78 223.81 1.01 20 0.834 0.171 0.168 0.0033

15 203.35 227.76 1.01 24.4 0.804 0.201 0.197 0.0038

SUMVIARY

ans Rama near mo 51 run rummue

suns omm= KB sun an: Hanan:

W ”3. mm m (1119/1)

1mm" 1mm» lmwu

1 0.12951 12.24 0.153

2 0.2987 0.18888 +/- 0.24 7.22 12.24 0.153

3 0.13841 12.24 0.153

4 0.15666 12.24 0.153

5 0.23109 0.19115 +/- 0.09 2.32. 12.24 0.153

6 0.18569 ' 12.24 0.153

7 0.24744 12.24 0.153

8 0.26132 0.2677 +/- 0.06 4.75 12.24 0.153

9 0.29434 12.24 0.153

10 0.17654 12.24 0.153

11 0.17025 0.17475 +/- 0.01 8.95 12.24 0.153

12 0.17744 12.24 0.153

13 0.13441 12.24 0.153

14 0.05016 0.07742 +/- 0.12 12.01 12.24 0.153

15 0.0477 12.24 0.153

-L‘ ‘ I. J.H.elu-Om

”I01?

fi‘W" '

(”#0704

“DONG,

11
12
1.3
14
1s

16
17
18
19
20

STDi
"62
std:
"d4
std!

counts.

Mia-

93‘“.

“can"

“in.
Munr-
Stalin-

'buiomr-

1571.55
83.0977
103.163

pH-7.29
105.524
18.2376
22.6413
6.71464

1.28864»

131-129.71
117.869
7.754
9.62631
7.50017
0.54789

PH-7.29
32.1061
4.35266
5.40367
2.04296
0.30755

80

Exp. 6 Comparing Treatment Processes

Table 0.12 Exp. 6 (50 Data Summary

corn:
("5)

36.121
74.685
375.28
1166.8
2838.7

1434
1599.8
1557.8
1624.5
1641.6

96.061
124.11
89.36
91.587
126.5

127.38
116.53
124.24
111.57
109.63
31.323
35.194

32.996
34.804
36.516
30.96
25.254
25.868
30.398

cm

3420922
6564998
36136324

113503296
269742400

162462432.
167773120
173031872
174159760
177728736

11728412
15975902
13154551
10824900
13761370

16194422
15912401
16013914
13753391
14313340
3099191

3413571

7204697
7202104
6969824
6083813
6053545
2681779
2811325

um

54065468
50180996
54969404
55531168
54246144

47456980
44026632
46605196
45019572
45471376

39501324
44076388
46771044
37788984
37391028

43142128
45460984
43511404
40639560
42868880
42341656
41507584

50437560
48971212
46136312
44058184
48524524
44365784
39578076

CMA

0.06
0.13
0.66
2.04
4.97

3.42
3.81
3.71
3.87
3.91

0.3
0.36
0.28
0.29
0.37

0.38
0.35
0.37
0.34
0.33
0.07
0.08

0.14
0.15
0.15
0.14
0.12
0.06
0.07

mac

626.9
626.9
626.9
626.9
626.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9
469.9
469.9

cone
(STD)

31.04
77.59
387.8
1163
3094

0.04.6.

0.05
0.12
0.62
1.85
4.94

CARA

0.06
0.13
0.66
2.04
4.97

21
22
23
24
25

26
27
28
29
30

OZONE/11202
m cone.-
W dov-
ud. error .
% m u
% 01¢ oner-
hm
bunk

OWZOZIUV
aveng- com.-
um dev-
fld. error -
% I'm .
% Itd CW-
018*
b“

pH=7.29
6.093288
4.402661
5.465749
0.387725
0.311086

pH-7.29
24.77086
9.682773
12.02082
1.576207
0.684171

81

Exp. 6 (cont)

Table 0.12 (cont)

cm
(DPD)

8.5063
6.2842
10.205
-1.2785
6.7494
32.423
31.806

37.677
12.553
28.107
27.229
18.287
31.333
31.254

CAFE

4470761
4262166
4546636
3436239
4082511
3459986
3315819

6161216
4749537
6265203
5806387
5493002
3138506
3199328

47099612
47500428
45976296
47687988
44953684
45667124
44613624

38228592
46355112
45136408
42458508
47411208
42865228
43806480

(LA/LA

0.0949
0.0897
0.0989
0.0721
0.0908
0.0758
0.0743

0.1612
0.1025
0.1388
0.1368
0.1159
0.0732
0.073

lm

469.9
469.9
469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9
469.9
469.9

d-t-J—t—‘A
ml‘mn‘ommummhoma

moonrpm-‘A—‘i’
omaumawma

(mm
2.1

1.2
0.2
0.4
0.2

0.0313 0

0.043.
0.1236

0.323

0.0117
0.0923
0.2917

0.0287 0

0.033
0.034
0.031
0.108
0.108
0.109
0.152
0.155
0.155

+/-
+/-
+/-
+/-
+/-

0.0043
0.0053
0.0023
0.0793
0.0793
0.0803
0.1233
0.1263
0.1263

82

Exp. 6 (cont.)

Exp. 6 Ozone and Peroxide Sampling Data

mmmm
mmmm

0.998
0.998
0.998
0.998
0.998
0.998
0.998

"0.998

0.998
0.998
0.998
0.998
0.998
0.998
0.998

(""1

18.22
18.51
18.32
15.12
15.92
15.89
15.71
14.13
20.64
14.79
11.77
19.64
21.13
16.03
12.41

0.716
0.712
0.715
0.81
0.8
0.797
0.85
0.883
0.816
0.849
0.876
0.814
0.797
0.854
0.882

0.283
0.286
0.283
0.188
0.198
0.202
0.148
0.136
0.183
0.149
0.123
0.185
0.202
0.145
0.117

penoxlos smpuuo DATA

81mm

monsoon

("W0
10.1
13.87
39.87
104.2
0
1.123
1.383
0.605
20.57
20.57
20.83
31.98
32.76
32.76

(0WD

5
25
75

("W0

1.037

20.66

32.5

+/-

m
m
m
oumou

0.154
0.156
0.155
0.131
0.137
0.137
0.135
0.124
0.171
0.129
0.105
0.164
0.174
0.138
0.11

1m.
(0W!)

0.981

0.371

1.113

OZONE PEROXIDE W SUMMARY

STD WWW man

an
("'9")

0.0193
0.1217
0.1112
0.1689

one

one

O3IUV
03114202

[*1

7.29
9.71
7.29
7.29

0.3669 oamzozmv 7.29

mm
are.
(0M)

1.037
20.66
32.5

+/-
+/-
+/-
+/-
+/-

510
am
(0W!)

0.981
0.371
1.113

Anon
own:
no:
on:
m
0.129
0.13
0.128
0.057
0.061
0.065
0.013
0.012
0.012
0.02
0.017
0.021
0.027

0.007
0.006

mm
mus
com.
(“I“)
60
60
60
60
60

iii

("‘9")

2.157
2.147
2.141
1.15
1.169
1.243
0.243
0.258
0.174
0.421
0.452
0.321
0.396
0.126
0.156

H3

("'9")

12
12
12
12

(”#QN‘

COGVO’

11
12
13

15

16

18
19
20

83

Exp. 7 Comparing Treatment Processes with Humic Acid

.1112
ml:
at“
ltd!

swea-

«terror-

1333.5
84.355
104.72

44.173
23.003
34.77
3.3126
2.3323

67.368
2.2779
2.8279
5.0521
0.1897

31.251
2.5328
3.1444
2.3436
0.2109

Table 0.14 Exp. 7 60 Data Summary

coal;
(ppb)

39.363
92.702
400.01
1157.3

2769.5 232965952

1381.6
1433.2
1242.6
1258.8
1384.2
11.769
1.4559

91437.3
13.133
73.486
61.164

15.21

57.87
7.2923
9.2053

"187.3
68.037
64.55
67.367
66.2
70.685
9.4344
6.2567

PH87.3
28.429
32.218
33.933

28.69
32.985
6.6341
10.368

CAFEA

2819107
6526089
30478360
94282104

1m

45622920
44846276
48538140
51896936
53586776

C.M.A

0.06
0.15
0.63
1.82
4.35

HUIIO ACID . 2 mgll
144214640 49848084
151057184 50330852
141609040 54423720
132482648 50258736
143092864 49365924

1 184800
150069

48075492
49222576

2.89
3
2.6
2.64
2.9
0.02
0

NUIIO ACID 3 2 mgll

2118617
6647356
7054005
2270413
6728184
668956
886975

47318136
38838072
48529872
46217684
48594264
43807360
46013840

0.04
0.17
0.15
0.05
0.14
0.02
0.02

HUIIO M210 3 2 mall

7477529
6370517
7260369
8008159
7803032
850444
609971

47057224
42022052
46098008
51647220
47450268
43047200
46555956

0.16
0.15
0.16
0.16
0.16
0.02
0.01

HUIIO A0“) I 2 mall

3582973
4143686
4248655
3620123
4262954
709304
1060299

46331880
48596724
47813524
46484224
49070468
51057920
48838432

0.08
0.09
0.09
0.08
0.09
0.01
0.02

1“”

626.9
626.9
626.9
626.9
626.9

469.9
469.9
469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9
469.9
469.9

com:
(81' D)

31 .04
77.59
387.8
1163
3094

0.64.0.

0.05
0.12
0.62
1.85
4.94

QMA

0.06
0.15
0.63
1.82
4.35

21
22
23
24
25

26
27
28
29
30

21
22
23
24
25

21
22
23
24
25

21
22
23
24
25

84
Exp. 7 (cont.)

Table 0.14 (cont.)

man come W wen C.NLA
(ppb)
ozoNEIuvm2o2 p11.7.3 Home ACID . 2 ma
mm.- 56.7862 50.497 6317741 50125824 0.126
Itmddw- 18.0166 48.419 5713147 46950148 0.122
Itd.orvor- 22.367 88.761 10461345 50742816 0.206
11mm.- 4.25853 45.249 5969778 51889320 0.115
11mm. 1.50032 51.005 6193522 48728968 0.127
m 9.1781 980092 50995164 0.019
m 10.205 1030888 48241524 0.021
ozoueonu pu.7.3 HUIIIC ACID .10 mull
m 00110.- 140.926 164.08 17498500 49773716 0.352
u-mou- 14.9242 146.58 13424540 42630264 0.315
310. error . 18.5278 132.46 15207836 53294792 0.285
Swen-h.- 10.5684 135.47 15140865 51912556 0.292
fibulerror- 1.2428 126.04 13212265 48591980 0.272
bill: 3.8059 484537 60797672 0.008
ozonauv pu.7.3 Home A610 .10 man
mm— 277.233 283.93 32337684 52315840 0.618
«mucu- 16.8659 259.73 30489600 53732060 0.567
«m- 20.9384 277.8 31926832 52746720 0.605
as m. 20.7904 263.19 29393454 51146320 0.575
swm. 1.40449 301.51 33964532 51860480 0.655
at“ 12.601 1294990 49078484 0.026
m 9.8951 1033884 49895888 0.021
02005192110an pit-7.3 HUIIIC note .10 mall
mun-m.- 157.982 168.83 18782474 50810816 0.37
«may. 11.4927 158.79 17560190 50371680 0.349
«.m. 14.2678 146.63 17819278 55140868 0.323
11mm.- 11.8475 169.68 19245540 51814520 0.371
assum- 0.95704 145.97 16484694 51228856 0.322
m 9.14 996714 52076068 0.019
m 6.2431 649048 49646612 0.013
ozouamvmzo2 p11.7.3 HUIIIC ACID .10 lug/1
mm- 177.229 190.92 20566518 49073728 0.419
«mm- 8.39357 176.2 19280496 49656744 0.388
«m- 10.4203 171.04 19571656 51851904 0.377
sw... 13.2909 178.11 19508136 49731156 0.392
11mm. 0.69897 169.87 19213408 51234228 0.375
m 11.731 1157082 47101260 0.025
blank 6.6964 661858 47199688 0.014
.183 269.93 26260676 46459264 0.565

w

469.93
469.93
469.93
469.93
469.93
469.93
469.93

469.93
469.93
469.93
469.93
469.93
469.93

469.93
469.93
469.93
469.93
469.93
469.93
469.93

469.93
469.93
469.93
469.93
469.93
469.93
469.93

469.93
469.93
469.93
469.93
469.93
469.93
469.93
469.93

85

Exp. 8 Comparing Treatment Processes with Bicarbonate

STDI
0182
It“
61114
a“

CONT. 8.8. dllmod 50%

0115de

munc-
M“.
storm.

1172.7
697.72
866J9
1172.7

CONT. 8.3. not dllutod

DONG

11
12
13
14
15

16
17
18
19
20

man.-
wow-

OZONEIUV
means.-
we.»
enorm-
16mm.
xflm

1361.7
41.195
51.142

91130.05
693.81
36.861
45.762
50.951
3.0059

"1:15.05
141.81
25.572
31.747
10.414
2.0853

Table C.15 Exp. 8 60 Data Summary

coat.
(M)

29.08
30.93
675.8
1153
2054

CAI-EA

3454106
3744100
82966792

was

47383152
48275280
48968528

140832768 48708168
247679104 48091680

Summnihuhannuanl

1304
832.1
1436
1522
769.6

70625616
53045168
84183080
82024656
48997500

47304016
54329832
51423508
47371368
53967216

Sodlum Blcorbonoto 3 2111'

167588624 53732128
166028192 52073440
160524528 51137856

1365
1396
1374
1382
1291
59.71

681.6
711.7
634.8
715.3
725.7
47.85

47.87
80.6
427.4
1220

180.1
145.8
108.9
139.3
135.1
25.85

161859488

51266416

152491312 51561952

6593537

77066328
85579480
72622656
80213872
76837104
4593655

3432444
4208664
32773476
91141744

17519176
17891422
14970963
17020896
17756600
2800593

50433284

Sodium Bicarbonate - 2m!

48256436
51458636
48589928
48007264
45365796
43848736

45677012
33263044
48846764
47608712

Sodium Bicarbonate 8 2m!

38855416
47615424
50745808
47085136
50401228
49478300

QMA

0.07
0.08
1.69
2.89
5.15

1.49
0.98
1.64
1.73
0.91

3.12
3.19
3.14
3.16
2.96
0.13

0.45
0.38
0.3
0.36
0.35
0.06

LCDG

489.8
489.8
489.8
489.8
489.8

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9
469.9

626.9
626.9
626.9
626.9

469.9
469.9
469.9
469.9
469.9
469.9

coax:
(STD)

31.2
70.3
586

1171
1952

31
77.6
388
1163

0.01.6.

0.06
0.14
1.2
2.39
3.99

0.05
0.12
0.62
1.85

QMA

0.07
0.08
1.69
2.89
5.15

0.08
0.13
0.67
1.91

21
22
23
24
25

26
27
28
29
30

21
22
23
24
25

21
22
23
24
‘ 25

21
22
23
24
25

21
22
23
24
25

21
22
23

Note - peaks for the jug samples exceeded the limits for the FD det.

OZONE/11202
average cone.-
aIandaN dev-

and. CW -
1; remaining -
SS etd enor-
Nari

OZONE/UVIH202

average cone.-
mo dew

816.."-

average obno-
m dew
atd. error .
as remaira‘ng .
16 aid error-
OZONE/N202
avorage cone.-
38W dev-
oid. error .
18 remaining .
18 etd error-

OZONEIUWH202

average cone.-
etandard dev-
atd. error -
as ammo .
16 aid error-
TCB STOCK
average oonc...
standud dev-

etd. error .

pH.6.85
98.014
5.8703
7.2878
7.1978
0.4787

pHI6.85
89.934
13.771
17.096
6.6044
1.1229

[DH-6.85
1143.5
70.064
86.982
16.028
5.7135
pH36.85
288.37
16.141
20.039
78.823
1.3163
[Hi-6.85
224.27
19.334
24.002
83.53
1.5766
pHI6.85
209.72
19.934
24.747
84.599
1.6255

2019.8
127.54
158.34

86

Exp. 8 (cont.)

Table C.15 (cont.)

coax:
(nob)

Sodium Bicarbonate - ZmIl
99.493

88.889
104.58

100.63

96.47

26.81 6
Sodium Bicarbonate a ZmIl

84.371
93.087
74.232
111.38
86.597
25.155

cm

13561902
12928769
14218605
13702667
13897192
2588852

11969428
12079587
10126761
10923202
11748724
2549045

UREA

49041572
51036812
49424472
49106504
51486196
44093084

49915224
46661128
46538964
36540364
48018728
46281324

Sodlum Bicarbonate - 10 mil
1253.3 124047208 44314564
1109.4 127153544 51187012

1125
1067.8
1161.8

119543504 47472476
125136896 52293080
132555880 51006308

Sodium Bicarbonate - 10 ml

314.48
292.79
280.34
273.72
280.5

35014336
34724216
34724280
33617652
35261884

47086780
49881352
51915488
51373436
52691040

Sodium Bicarbonate - 10 mIl

239.95
209.89
245.37
226.24
199.91

30200098
26603388
30732644
27762180
24355208

52029212
51694344
51885828
50438792
49424748

Sodium Bicarbonate - 10 mll

223.29
238.15
198.02
194.06
195.06

26514764
29407086
25091760
23996722
24365780

48744956
51009540
51352224
49997236
50535080

1974.9 221560272 51265736
1967.7 220506640 51207400
2192.3 227322240 47383376

C.NI.A

0.276539
0.253322
0.287683
0.27904

0.269921
0.058713

0.239795
0.258879
0.217597
0.298935
0.24467

0.055077

2.799242
2.484098
2.518165
2.392992
2.598813

0.743613
0.696136
0.668862
0.654378
0.66922

0.580445
0.514629
0.592313
0.550413
0.492774

0.543949
0.576502
0.488621
0.479961
0.482156

4.3218

4.306148
4.79751

100w:

469.9
469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9
469.9
469.9

469.9
469.9
469.9

APPENDIX D

HEADSPACE SANIPLER AND GAS CHROMATOGRAPH

OPERATING PARAMETERS

Appendix D. Headspace Sampler and Gas Chromatograph Operating Parameters

1,3,5-Trichlorobenzene was measured using a gas chromatograph (Perkin-Elmer
Autosystem, Norwalk, CT) equipped with a flame ionization detector (FID) and a
silica glass capillary column (Perkin-Elmer, Model 624). The carrier gas was helium.
Table D.l summarizes the operating parameters of the headspace sampler and gas

chromatographer used for the trichlorobenzene analysis.

Table D.1 Headspace Sampler and GC Operating Parameters

Flame Ionization Detector

Temperature 250 °C
Injector

Temperature 200 °C

Flow 8.0 ml/minute

GC Conditions

Oven
Equilibrium Time 1.0 minute
Temperature 90 °C
Hold Time 15 minutes
Carrier Gas Helium (high purity)
Pressure 20 psi
Headspace Conditions
Bath Temperature 90 °C
Transfer Line and
Needle Temperature 100 °C
Carrier Gas Pressure 20 psi
Equilibrium Time 90 minutes

87

HICH

11111111111111i“