.‘. ‘L‘ ._..

Maw-u

 

w Tuféfi‘ < .

r- ;"
341's

.4

”WE-“>46;

‘-

-:£-::: «a»: 1%"

«a!

0

.§
3‘s
2““

 

"gang :5” ' ‘ “~ "Lb: ‘5"
'.' i. g3 3"!) § 935 -
\fiaah'vklm‘imfix‘mg‘a‘z:fllfifi‘ -

2* .{HQ' in‘ifi ’x .
dtt'QéLLHefd"
~' a?

. #2353455.“ ' ‘ .‘
‘%.,spamw‘€.. "
0" .-z_',‘ "fa-1:? ‘

X
‘ v.
"3.3%. fl?!
'wdi «$.13
A Fifi}: , .
¥ me:- M ~ ,
i) e 3" 4 > . ' ,’ ’3‘“ ‘1"
V “a, w ‘ Q; -: ‘
- 8": é‘ofi ‘1’ ' , “t ivy?!"
,A‘,‘ ¢ '. ‘ f" . z. " ' ki‘fifikggi
T) d‘ 7'1" x.” s. A ‘4‘ - pl 1'"
¥ 4;; U _, ‘ uh!‘ .

,w’f . '
i

1
(A

s.
3%

..= “f
f. ‘ . I
2i?! 3i

i

g} 53:21. .
‘2' '63 .sm‘ '
7 “+19% if

all.

3‘- I'
1-4?
a;

i" .,.
“ tin-3v
:grizfitfifi
in??? ..

§-
‘1
I ;?.§5h i

in
fry ‘
M

3 r" 9’

“Q45
iflfi“?
5‘: J 1‘
r .

'19

r
5
A
l

 

I
(0004
J’ 7 y; 8' H J

This is to certify that the
. thesisentitled

TREATMENT OF METHYL TERT-BUTYL ETHER IN METEA
SOIL USING FENTON’S REAGENT

presented by
SHEBA GOKLANY

has been accepted towards fulfillment
of the requirements for the

Master of degree in Environmental Engineering
Science

 

 

Aflmflfl W

/ Major Prof or’s Signature
(yluflxf :30]. 200 ¢

Date

MSU is an Affinnative Action/Equal Opportunity Institution

 

.- m-—.—.—-—-

 

 

_UBRARY
M'Chigan State
University

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/DateDuepryp. 1 5

. ____ _..__._—-. ._ _ _

 

TREATMENT OF METHYL T ER T -BUTYL ETHER IN METEA SOIL USING
FENTON’S REAGENT

By

Sheba Goklany

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

2004

ABSTRACT

TREATMENT OF METHYL T ER T -BUTYL ETHER IN METEA SOIL USING
FENTON’S REAGENT

By

Sheba Goklany

The degradation of methyl tert-butyl ether (MTBE) in Metea soil using hydrogen
peroxide, with and without the addition of Fe”, was studied. Greater than 95% of MTBE
was degraded when the concentration of hydrogen peroxide added was 30,000 mg/L or
greater, under conditions where Fe2+ was not added to the reaction mixture. The average
initial MTBE concentration in the soil was 15.9 mg/L, the initial pH was between 2 and
3, and the reaction time was two hours. When Fe2+ was added to the soil, greater than
99% removal of MTBE was obtained when the concentration of Fe” added was between
4 and 20 mM. The average initial concentration of MTBE in the soil in this case was
15.6 mg/L, the H202 concentration was 15 mM, the initial pH was between 2 and 3, and

the reaction time was 120 minutes.

Although MTBE was almost completely degraded using this technology, the formation of
higher and lower molecular weight byproducts was observed when no F 62+ was added to
the system. The presence of higher molecular weight compounds was possibly due to
dimer formation. The lower molecular weight byproducts were tentatively identified as
propionaldehyde and acetone. When Fe2+ was added to the system, only lower molecular
weight byproducts were formed, which were tentatively identified as formaldehyde,

propionaldehyde, and acetone.

_‘_—B c an. .iq. ninth.“ thirty, . . .u.—‘ 1‘. I I

} . i. \é' 'v‘lifEi.."1I.-Iu. Ogit‘l’}»

To Ashok, Bharati, Sharmila, Sunil and Pradeep

iii

ACKNOWLEDGEMENTS

I would like to thank my advisor, Dr. Susan Masten, who helped and guided me through
the entire course of this thesis. This work would not have been possible without her
support. It has been a very good learning experience for me, and I am thankful to Dr.

Masten for this wonderful opportunity to help me learn and grow under her guidance.

1 am grateful to Dr. Simon Davies and Dr. Roger Wallace, who provided me with

valuable suggestions and excellent advice.

I would also like to thank Mr. Yan Pan (MSU) for helping me in the lab with the

equipment and my experiments. His help is truly appreciated.

My family and friends have always supported me and I would like to express my

appreciation for helping and motivating me all the way.

iv

TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

CHAPTER 1 — INTRODUCTION

1.1

Background

1.2 Oxygenate Requirements of the Clean Air Act
1.3 Physical and Chemical Properties of MTBE
1.4 Fate and Transport of MTBE in Soil
1.5 Regulatory Standards
1.6 Toxicity Effects
1.7 MTBE in the Michigan Environment
1.8 Remediation Strategies for MTBE
CHAPTER 2 — REACTION MECHANISM AND BYPRODUCTS
2.1 Chemistry of Hydrogen Peroxide Reactions
2.2 Degradation Products of MTBE by F enton’s Reagent
CHAPTER 3 — MATERIALS AND METHODS
3.1 Materials
3.2 Analytical Methods
CHAPTER 4 — TREATMENT OF MTBE USING FENTON’S REAGENT (I)
4. 1 Introduction
4.2 Materials and Methods
4.2.1 Materials
4.2.2 Experimental Method
4.3 Results and Discussion
4.3.1 Effect of H202
4.3.2 Effect of Fe3+ on MTBE Degradation in Soil
4.3.3 pH Monitoring for Soils Treated with Hydrogen Peroxide
4.4 Formation of Byproducts
4.5 Volatilization Losses
CHAPTER 5 — TREATMENT OF MTBE USING F ENTON’S REAGENT (II)
5. 1 Introduction
5.2 Materials and Methods
5.2.1 Materials
5.2.2 Experimental Method
5.3 Results and Discussion

5.3.1 Effect of Fe2+ on Degradation of MTBE in Soil

vii

viii

OO\)O\UIJ>UJN-*v-‘

5.3.2 Effect of FeZVHzoz Ratio 55

5.3.3 Two Stages of MTBE Degradation 57

5.3.4 pH Monitoring for Soils Treated with Fenton’s Reagent 58

5.4 Formation of Byproducts 60
CHAPTER 6 — CONCLUSIONS AND SCOPE FOR FUTURE WORK 65

REFERENCES 68

vi

Table 1-1.

Table 1-2.

Table 2-1.

Table 2-2.

Table 3-1.

Table 4-1.

Table 4-2.

Table 4-3.

Table 4-4.

Table 4-5.

Table 4-6.

Table 4-7

Table 5-1.

Table 5-2.

Table 5-3.

Table 5-4.

Table 5-5.

Table 5-6.

LIST OF TABLES

Physical and chemical properties of MTBE
Guidelines for MTBE in drinking water in various states

Degradation of MTBE in soil microcosms receiving 2 uL of H202
(Y eh and Novak, 1995)

MTBE degradation rates in soil microcosms at depths of 1.2 m and
2.1 m (Y eh and Novak, 1995)

Soil characteristics of Metea soil
Trial 1. Average initial concentration of MTBE = 429.75 mg/L
Trial 2. Average initial concentration of MTBE = 340.25 mg/L

Trial 3. Average initial concentration of MTBE = 419 mg/L

Variation of pH with hydrogen peroxide concentration for Trials 1,2,

and 3

Percent conversion of MTBE to propionaldehyde

Percent conversion of MTBE to acetone

Total conversion

Variation of MTBE concentration and percent oxidation with Fe2+
concentration. Initial concentration of MTBE = 388.75

mg MTBE/kg soil

Variation of pH with concentration of Fe2+

Percent conversion of MTBE to formaldehyde

Percent conversion of MTBE to propionaldehyde

Percent conversion of MTBE to acetone

Total conversion

vii

21

22

26

34

34

35

38

45

45

46

54

58

62

63

63

63

Figure 2-1.

Figure 2-2.

Figure 2-3.

Figure 4-1.

Figure 4-2.

Figure 4-3

Figure 4-4.

Figure 4-5.

Figure 4-6.

Figure 5-1.

Figure 5-2.

Figure 5-3.

LIST OF FIGURES

Mechanism of MTBE degradation by 'OH radical (Stefan et al.,
2000; Xu et al., 2004)

MTBE degradation products obtained from Route A (Stefan et al.,
2000)

MTBE degradation products obtained from Route B (Stefan et al.,
2000)

Effect of hydrogen peroxide concentration on percent oxidation of
MTBE, initial pH = 2-3. Initial concentration of MTBE was
between 299.25 and 429.75 mg MTBE/kg soil

Initial [H202] = 20,000 mg/L (500,000 mg/kg soil), pH = 2-3,
no Fe2+ added. Chromatogram shows presence of lower molecular
weight compounds and dimers

Initial [H202] = 15,000 mg/L (375,000 mg/kg soil), pH = 2-3,
no F e2+ added. Only formation of lower molecular weight compounds
was observed

Initial [H202] = 10,000 mg/L (250,000 mg/kg soil), pH = 2-3,
no Fe2+ added

Initial [H202] = 8000 mg/L (200,000 mg/kg soil), pH = 2-3,
no Fe2+ added

Initial [H202] = 5000 mg/L (125,000 mg/kg soil), pH = 2-3,
no Fe2+ added

Effect of Fe2+ concentration on percent oxidation of MTBE.

Initial concentration of MTBE was approximately

388.75 mg MTBE/kg soil, [H202]o = 510 mg/L (12,750 mg/kg soil),
pH = 2-3

[Fe2+] = 4 mM (5600 mg/kg soil), pH = 2-3, [MTBE]0 = 388.75
mg/kg soil, [H202] = 510 mg/L (12,750 mg/kg soil)

[Fe2+] = 6 mM (8400 mg/kg soil), pH = 2-3, [MTBE]O = 388.75
mg/kg soil, [H202] = 510 mg/L (12,750 mg/kg soil)

viii

l6

17

18

33

40

41

42

43

53

60

61

CHAPTER 1

INTRODUCTION

1.1 Background

Methyl tert-butyl ether (MTBE) is a synthetic compound, which is manufactured by the
chemical reaction between methanol and isobutylene. MTBE is added to gasoline with
the intent of lowering air pollution emissions by making the fuel burn cleaner. It has been
used in the United States since 1979 as a replacement for lead as an octane enhancer in
concentrations as high as 8% by volume in mid- and high-grade gasoline. It has also been
used as a fuel oxygenate at higher concentrations (11 to 15% by volume) because it
increases the combustion efficiency of gasoline by reducing the amount of harmful
emissions, such as carbon monoxide and ozone, which are direct or indirect products of

incomplete fuel combustion (U SEPA MTBE Fact Sheet # 2, 1998; Jacobs et al., 2001).

However, the use of MTBE as a fuel additive has created one of the most widespread
environmental problems of the late twentieth century. MTBE is highly soluble in water,
and has been found in groundwater at over 250,000 contaminated sites throughout the US

as of 2000 (Jacobs et al., 2001).

Sources of MTBE in soil and groundwater arise from leaking underground gasoline
storage tanks, pipelines, gasoline spills, run-off, and precipitation. The major source of

MTBE contamination originates from leaking underground storage tanks (U STs).

1.2 Oxygenate Requirements of the Clean Air Act

The Clean Air Act Amendments of 1990 (CAA) requires the use of oxygenated gasoline
in areas with high levels of air pollution. Although the CAA does not specify MTBE,

most refiners chose to use MTBE due to its blending properties and for economical

reasons (U SEPA, MTBE, Gasoline, 2003).

The two oxygenated gasoline programs are:

1. Winter Oxyfuel Program: This program was implemented in 1992, and requires that
the gasoline sold in cities that have elevated levels of carbon monoxide during cold
months contain 2.7% oxygen by weight. The primary oxygenate used in this program

is ethanol (USEPA, MTBE, Gasoline, 2003).

2. Year-round Reformulated Gasoline Program: This program was implemented in
1995, and requires the use of reformulated gasoline (RFG) year-round in cities that
have high levels of ground-level ozone. RFG is oxygenated gasoline that contains a
minimum of 2% oxygen by weight and is specially blended to have fewer polluting
compounds than conventional gasoline. As of 2003, 30% of the gasoline sold in the
United States was RFG, out of which about 87% contained MTBE (USEPA, MTBE,

Gasoline, 2003).

1.3 Physical and Chemical Properties of MTBE
MTBE is an -ether with a chemical formula of C5H120 and the following structure
(Howard, 1993):
CH3
H3C— 0 — C —CH3

CH3

The physical and chemical properties of MTBE are summarized in the following table:

Table 1-1. Physical and chemical properties of MTBE

 

 

 

 

 

 

 

 

 

 

 

 

 

CAS Registry Number " 634-04-4
Boiling Point (°C) a 55.2
Melting Point (°C) a -109

Molecular Mass (g/mole) a 88.15
Water Solubility (mg/L) at 25 °C a 51,000
Vapor Pressure (mm Hg) at 249
25 °C a
Henry’s Law Constant b 0.022
Log Octanol/Water Partition Coefficient “ 1.24
Diffusivity in air (cmz/s) ° 0.0792
Diffusivity in water (cmz/s) ° 9.41 x105

 

3 Howard, 1993; b Kinner, 2001; c MT DEQ, 2003

At standard temperature (25°C) and pressure (760 mm Hg), MTBE is a colorless and

flammable liquid (Jacobs et al. 2001).

This aliphatic ether is highly soluble in water (51,000 mg/l) due to its hydrophilic nature.

MTBE also has a low octanol-water partition coefficient (log Kow = 1.24).

1.4 Fate and Transport of MTBE in Soil

The physical and chemical characteristics of MTBE determine its fate and transport in the
environment. When MTBE is spilled on the surface of the soil, it volatilizes readily. Due
to it high water solubility, MTBE does not readily sorb to soil or rocks. In the case of a
petroleum release in soil, MTBE and BTEX are co-dissolved at first. However, since
MTBE does not sorb to soil in aquifer systems, it moves readily with the groundwater,
which results in considerable separation of these hydrocarbons (Douglas et al., 2002).
The mobility of a contaminant in groundwater is expressed as a ratio of groundwater
velocity to the contarninant’s velocity and is termed as Retardation Factor. For MTBE,
the retardation factor is approximately equal to 1, indicating that MTBE would move at
approximately the same velocity as the groundwater in which it is dissolved (U SEPA

MTBE Fact Sheet # 2, 1998; Kinner, 2001).

Once MTBE volatilizes into the air, it may undergo degradation in the presence of UV
light (Kinner, 2001). It has been found that MTBE is unstable in the atmosphere.
Atmospheric water droplets (rain, clouds, and fog) contain hydrogen peroxide and iron
ions, which, in the presence of UV radiation, react to form °0H, a strong oxidizing agent
(Guillard et a1, 2003). MTBE can also be removed fiom the atmosphere by its dissolution

into rainwater. After the rainwater falls to the ground surface, MTBE will infiltrate into

the ground or flow over impervious surfaces such as roadways, potentially contaminating

the ground water or being discharged into surface water (Kinner, 2001).

1.5 Regulatory Standards

The United States Environmental Protection Agency (U SEPA) has not set a national
drinking water standard for MTBE, although some states have defined their own limits.
MTBE is on the Candidate Contaminant List and USEPA is considering setting
regulatory standards for it. The potential health effects and occurrence of MTBE are
under study by the USEPA, other government organizations, and academic institutions

(U SEPA, MTBE, Drinking Water, 2003).

The US. Environmental Protection Agency (EPA) Office of Water has issued an
Advisory on methyl tert-butyl ether in drinking water. This advisory states that keeping
MTBE concentrations in the range of 20 to 40 ug/L of water or below will avert
unpleasant taste and odor effects in drinking water. Since these concentrations are
approximately 20,000 to 100,000 times lower than the range of exposure levels that
caused cancer and non-cancer effects in rodents, this taste and odor threshold would also
protect consumers from potential health effects (U SEPA, Drinking Water Advisory,

1997).

MTBE is regulated drinking water contaminant in California with a primary maximum

contaminant level (MCL) of 13 ug/L, effective May 2000. The public health goal for

MTBE in California is also 13 pg/L. It also has a secondary MCL of 5 ug/L, which

addresses taste and odor thresholds.

Several other states have set guidelines for MTBE in drinking water supplies. These

guidelines have been summarized Table 1-2.

Table 1-2. Guidelines for MTBE in drinking water in various states

 

 

 

 

 

 

 

 

 

 

 

 

 

State Guidelines (pg/L) Basis
Arizona 3 35 Health-based guideline
California a 5 Secondary MCL
13 Proposed Primary MCL
Connecticut a 100 Guideline
Florida a 50 Drinking water standard
Massachusetts 3 70 Proposed guideline
Michigan b 240 Health-based guideline
40 Drinking water guideline
New Hampshire 3 100 Guideline
New Jersey 3 70 Health-based MCL
New York a 50 MCL
Rhode Island a 50 Guideline
Vermont 8 40 Drinking Water Standard

 

 

 

 

‘ Douglas et a1, 2002; b MDA, 2002

1.6 Toxicity Effects

Since the introduction of MTBE to oxygenated fuel in the United States, acute health
system complaints such as headaches, nausea, dizziness, and breathing difficulties, have
been reported nationwide. However, existing studies of acute health risks of MTBE do

not support these claims (Jacobs et al., 2001).

No studies have been conducted on MTBE’s carcinogenic effects on humans, but studies
indicate that MTBE is carcinogenic to rats and mice. Research animals exposed by
inhalation to high concentrations of MTBE developed cancerous or non-cancerous
effects. USEPA’s Office of Water has concluded that there does not exist sufficient
available data to estimate the potential health risks of MTBE at low exposure levels in
drinking water. However, the current data supports the conclusion that MTBE is a

potential human carcinogen at high doses (U SEPA, MTBE, Drinking Water, 2003).

However, little or no data exists for the reproductive, developmental, genotoxic, and
carcinogenic effects of MTBE on humans. More studies are needed to determine the

cancer and non-cancerous effects of MTBE on humans (Jacobs et al, 2001).

1.7 MTBE in the Michigan Environment

The presence of fuel oxygenates is tested in each official sample obtained under the
State’s Motor Fuels Quality Program. The Michigan Department of Agriculture obtains
approximately 1800 samples in response to complaints and routine investigation and
monitoring. Oxygenates, such as MTBE, that are present in excess of 1% in fuel are

recorded on the test report of the sample (MDA, 2002).

In Fiscal Years 1994 and 1995, 1% or higher concentrations of MTBE were present in
40% of all samples. This percentage dropped through subsequent years until it reached
13% in Fiscal Year 1998. This drop was attributed to economic factors, cost and

availability of MT BE. In 2000, 13% of all samples contained MTBE, with an average

concentration of 3%. Based on Michigan’s 4.8 billion-gallon consumption of gasoline,
this amounted to 4,320,000 gallons of MTBE. In 2001, 17% of all samples tested

contained MTBE, with an average concentration of 4.75% (MDA, 2002).

In Michigan, the health-based clean up level for MTBE for residential use is 240 ppb
(MDA, 2002). Water sources contaminated with MTBE become unsuitable for human
consumption due to its turpentine-like taste and odor (Jacobs, 2001). The Michigan
Department of Environmental Quality (MDEQ) has established an aesthetic drinking
water concentration, protective of taste and odor thresholds, of 40 ppb for MTBE. Since
the health-based criterion is much higher than the aesthetic-criteria, a person should thus
taste or smell MTBE before it poses a health risk. Several of Michigan’s lakes and rivers
have been analyzed for the presence of MTBE by the MDEQ Surface Water Quality

Division. MTBE has not been detected in any of these samples (MDA, 2002).

The total number of Underground Storage Tanks reported by the Waste and Hazardous
Materials Division, MDEQ, as of December 31, 2003 was equal to 86,525. The number
of facilities out of these was reported to be equal to 25,705. The number of confirmed
releases with MTBE groundwater impacts greater than or equal to 40 ppb, have been

reported to be 678 as of April 2003 (Storage Tank Information Database, 2004, MDEQ).

1.8 Remediation Strategies for MTBE

Treatment technologies for MTBE remediation in soil include excavating the soil, soil

vapor extraction, Fenton’s reagent and Advanced Oxidation Processes (AOPs).

Excavation is reasonably successful to depths of about 20 to 30 feet, the practical limit
equipment such as backhoes and excavators can reach. After the soil is excavated, on-site
treatment may be achieved using low temperature thermal desorption. The drawbacks of
this technique are high transportation costs if the MTBE contaminated soil has to be

transported off-site for treatment and major site disruption (Jacobs et al., 2001).

Soil vapor extraction, an in-situ technology, removes volatile contaminants fiom the soil
in the unsaturated zone above the groundwater by drawing out the contaminant vapors
with a vacuum applied to the subsurface (USEPA MTBE Fact Sheet #2, 1998; Jacobs et
al., 2001). When excavation together with soil vapor extraction was applied to a
contaminated site located at North Tahoe P.U.D. Maintenance (National Avenue), Lake
Tahoe, CA, it was observed that the MTBE concentration dropped form 40 ug/L to non-

detectable (MTBE Treatment Profiles, 2003).

Biodegradation methods for soil treatment are not recommended for treating MTBE
because it is considered recalcitrant to biodegradation USEPA MTBE Fact Sheet #2,
1998; Jacobs et al., 2001). In experiments conducted by Yeh and Novak (1995), it was
observed that MTBE was not biodegraded in over 100 days. Due to the high solubility
and mobility of MTBE in groundwater, it is recommended to treat MTBE at the source,

i.e., while it is still in the soil.

Using activated carbon is not considered a good alternative for the treatment of MTBE

since its high solubility in water limits adsorption onto activated carbon (Anderson,

2000). In studies conducted by Anderson (2000), he observed that only 52% removal of
MTBE was obtained for an initial solution concentration of 100 ug/L. The low affinity of
MTBE towards granular activated carbon makes this process undesirable and expensive.

Advanced Oxidation Processes (AOPs) provide a promising alternative technique for the
treatment of MTBE. These processes depend on the generation of the highly reactive
hydroxyl radical that reacts with most organic compounds (Jacobs et al., 2001). The
relative oxidizing power of the hydroxyl radical from Fenton’s Reagent, which is a
combination of H202 and Fe“, is 2.06, with C12 being assigned a value of 1.0. The
oxidizing power of OH is only second to fluorine, which has a value of 2.23 (Jacobs et

al., 2001).

Fenton’s Reagent is one such Advanced Oxidation Process, where the 'OH radical is
produced as a result of reaction between H202 and Fe2+. Fenton’s Reagent has been
documented as an effective treatment technology for the oxidation of MTBE (Ray and

Selvakumar, 2000; Kealey et a1, 2001; Burbano et al., 2003; Xu et a1. 2004).

Ray and Selvakumar (2000) studied the removal of MTBE from water using Fenton’s
Reagent. They observed that although complete mineralization of MTBE was not
achieved, greater than 99% degradation of MTBE was accomplished after one hour
contact time using a H202: Fe2+ ratio of 1:1. The major byproducts obtained in their study
were tert-butyl alcohol, tert-butyl formate, and acetone. Small quantities of isobutene
were also detected. The pH of the mixture dropped form about 5 to 3 after one-hour

contact time, which may have occurred due to the formation of organic acids. Mass

10

balances for two tests, which gave MTBE removal percentages of approximately 99%
each at H202: Fe2+ ratios of 1:1 and 6:1 respectively, were conducted as input of MTBE
and output of products (calculated as carbon). Approximately 80% and 60% of the
product could not be accounted for with H202: Fe2+ ratios of 1:1 and 6:1 respectively,
which was attributed to the formation of some volatile products that escaped the reaction

system, or the generation of certain non-volatile organics.

Xu et al. (2004) investigated the remediation of MTBE using Fenton’s Reagent in
aqueous solution. They observed that, under optimum conditions of 15 mM H202, 2 mM
Fe2+, pH 2.8, and at room temperature, 99% degradation of lmM MTBE solution was
achieved in 120 minutes. They hypothesize that the degradation of MTBE followed a
two-stage reaction. In the first stage, MTBE degrades swiftly based on a Fe2+/H202
reaction. During the second stage, which was the Fey/H202 stage, the degradation of
MTBE is less rapid. The primary intermediates and byproducts identified in their study

were tert-butyl formate, tert-butyl alcohol, methyl acetate, and acetone.

In experiments conducted by Burbano et al. (2003), the effect of H202: Fe2+ molar ratio
on the extent of degradation of MTBE was studied in water at a pH of 3. They discovered
that 99% degradation of MTBE was observed when the Fe“: H202 ratio was 1:1. The
degradation of MTBE decreased when the Fe“: H202 ratio was changed to values greater
than or smaller than 1.0. Their results showed that for ratios of 1.0, 2.0, 5.0, and 10.0, the
maximum difference in percent MTBE transformation was approximately 8%. However,

experiments with lower ratios of 0.5:1 and 0.75:1 showed significant differences. The

11

MTBE transformation was only 63.6% when a molar ratio of 0.5:1 was used. The MTBE
degradation increased to 84.9% when the ratio of F e”: H202 was 0.75:1. They deduced
that an excess of H202 would promote the formation of hydroperoxyl radicals (1102'),
which would favor the consumption of 0H: (refer reaction 2-6), whereas a lower

concentration of H202 would limit the Fenton’s reaction.

The Terra Vac Corporation, (2002) utilized the OxyVac Technology on a site
contaminated with BTEX (Benzene, Toluene, Ethyl benzene, and Xylene) and MTBE
from a gasoline spill in Concord, Massachusetts. This technology is patented by the Terra
Vac Corporation, which utilizes in-situ chemical oxidation combined with soil vapor
extraction for the remediation of volatile and semi-volatile compounds. Fenton’s Reagent
was used to remediate the contamination at this site. The site dimensions were
approximately 200 it x 200 ft. The soil and groundwater were treated using Fenton’s
reagent based on results from a bench-scale study. The system consisted of 20 injection
points and 5 soil vapor extraction wells. 10% H202 and 5% FeSO4 solutions were
periodically injected. Eight applications, amounting to approximately 1600 gallons of
H202 solution, were made over a one-year period. Groundwater monitoring results fiom
two wells showed reductions in MTBE concentrations fi'om 964 ug/L to 64 pg/L (93%
reduction), and from 3400 ug/L to 61 jig/L (98% reduction), respectively. As of January
2002, only one monitoring well remained with an MTBE concentration greater than the

clean up criteria of 70 ug/L (MTBE Treatment Profiles, 2002)

12

The goal of the current study is to study the feasibility of the oxidation of MTBE in soil
using F enton’s reagent with an attempt to understand how pH, concentrations of H202
and Fe”, and the ratio of H202: Fe2+ would effect the degradation of MTBE. The
experiments conducted in this study have been designed specifically to control
volatilization losses that may occur from the MTBE or its byproducts that are formed
during the reactions, thus giving accurate information about treatment efficiencies. These
experiments would also prove valuable for the detection and quantification of the
byproducts formed, since volatilization losses were minimized. The experiments were
also planned in a manner such that they were completely safe to conduct in the lab. We
have also made an effort to characterize the possible byproducts of MTBE using

hydrogen peroxide and Fenton’s Reagent.

l3

CHAPTER 2

REACTION MECHANISM AND BYPRODUCTS

2.1 Chemistry of Hydrogen Peroxide Reactions

Hydrogen peroxide reacts with both F e2+ and Fe”. F enton’s reaction, which takes place
between Fe (II) and H202 (Barb et al., 1951; Walling, 1975), generates °OH, a very
reactive oxidant (Reaction 2-1) (Pignatello, 1992). This °OH may be scavenged by
reacting with another F e2+ (Reaction 2-2) or may react with an organic compound

(Pignatello, 1992; Xu et al., 2004).

2+ 3+ - . _
H202 + Fe I Fe + H0 + OH (2 1)
'OH + Fe2+ . Fe3+ + H0' (22)
'OH + Organic _> Products (2-3)

The chain radical mechanism for the reaction of F e3+ (aq) systems is as follows (Barb,

1951; Walling, 1975):

H202 + Fe3+ —> Fe2+ + H+ + H02 (24)
Fe2+ + 11202 ———> Fe3+ + HO’ + Ho- (25)
'0H + H202 —_—> H20 + HO2° , (2-6)
HO2° + Fe3+ _, HI + 02 + Fe2+ (2-7)
HO2° + Fe2+ ——> HO2' + Fe3+ (2-8)

14

Reactions (2-4) and (2-7) are complex and pH dependent, due to the following equilibria:

Fe“ + H202 H H+ +Fe00H2+ —> Fe2+ + HO2° (2-9)

In the presence of excess peroxide, the concentration of Fe2+ is much less than that of
Fe3+ since the rate of decomposition of hydrogen peroxide by ferric ion is slower than
that by ferrous ion (Barb, 1951). Reaction (2-6) is another mechanism for hydroxyl

radical scavenging (Pignatello, 1992).

2.2 Degradation Products of MTBE by Fenton’s Reagent

The two characteristic features of MTBE are its ether linkage and the tertiary carbon
atom. The initial attack by the 'OH radical can take place either at the methoxy group or
any of the three equivalent methyl groups leading to carbon-centered radicals (Stefan et

a1, 2000; Xu et al., 2004).

This has been illustrated in the following reactions (Stefan et a1, 2000; Xu et al., 2004):

(CH3)3C-OCH3 + ' OH —> (CH3)3C-OCH2° + H20 (2-10)

CH3OC(CH3)2CH3 + 'OH —> CH3OC(CH3)2CH2' '1' H20 (2-11)

The °OH radical is an electrophile, and the C-H bonds adjacent to the oxygen are
responsible for a pronounced stereoelectronic effect that produces high rates of H-atom

abstraction. As a result, the H atoms of the methoxy group are more likely to be attacked

15

by the OH radical (Stefan et a1, 2000; Xu et al., 2004). This has been depicted in Figure

2-1 as Route A.
‘im
'0H/02 H3C —0 — C— CH3 '0H/02
I
CH3
CH3 CH3
' I
2 '02CH20-C— CH3 2 CH3O-C-CH202'
l I
CH3 CH3
(CH3)3COCH2-O4-CH20C(CH3)3 CH30(CH3)2CCH2-04-CH2C(CH3)20CH3
Route A Route B

Figure 2-1. Mechanism of MTBE degradation by °0H radical (Stefan et al., 2000;

Xu et al., 2004)

The tetraoxides formed in Routes A and B are generated due to self-termination reactions
of the corresponding peroxyl radicals (Stefan et al., 2000; Xu et al., 2004). These
tetraoxides undergo further reactions to give several products as shown in figures 2-2 and

2-3.

16

(CH3)3COCH2-O4-CH2OC(CH3)3

l

HCHO
CH3COCH3
(H3C)3COCHO
(HsC)3C0H
(I'I3C)3C02C(CH3)3

(CH3)3COCH2OH

Figure 2-2. MTBE degradation products obtained from Route A (Stefan et al.,

2000)

17

CH3O(CH3)2CCH2-O4-CH2C(CH3)2OCH3

CH30(CH3)2CCHO
CH3O(CH3)2CCH20H
HCHO
CH3COOCH3
CH3COCH3
(HsC)3C0H

CH3O(CH3)2C0H

Figure 2-3. MTBE degradation products obtained from Route B (Stefan et al.,

2000)

In earlier studies conducted on MTBE, several different products and intermediates have
been identified. Experiments were conducted by Xu et al., (2004), for an initial MTBE
concentration of 1 mM, Fe2+ concentration of 2 mM, hydrogen peroxide concentration of
15 mM at a pH of 2.8 for a reaction time of 420 minutes. The intermediates and by-
products of MTBE using Fenton’s Reagent were identified as tert-butyl formate (TBF),
tert-butyl alcohol (TBA), methyl acetate (MA), and acetone. The reactions were carried

out in graduated and plugged test tubes at room temperature of 21 i 1 °C. No carbon

18

balance for the initial and final organic compounds was conducted. The addition of Fe2+
and pH adjustment of the reaction mixture was performed after MTBE was already added
to the test tubes. Additionally, the reaction mixture was not chilled before analysis was
made. The experiments were not controlled to prevent volatilization of MTBE or it s
byproducts due to the reasons listed above. It is thus possible that some byproducts may
have escaped due to volatilization losses since no other byproducts were detected during

the course of the experiment.

In experiments conducted by Stefan et al. (2000), the initial concentration of MTBE and
H202 were 0.92 and 18.26 mM, respectively. It was observed that the some of the
primary reaction products during treatment of MTBE using UV/ H202 process were tert-
butyl formate (TBF), tert-butyl alcohol (TBA), acetone, methyl acetate, 2-methoxy-2-
methyl propionaldehyde (MMP), and formaldehyde, which were obtained during the
reaction time of 80 minutes. These may be further oxidized to the corresponding
aldehydes and acids. A good carbon balance obtained during the treatment indicated that
most of the byproducts had been accurately detected and quantified. They also observed
that acetone reaches the highest concentration compared to other early intermediates,
suggesting that acetone is comparatively less reactive than other organic compounds
towards °0H radicals and it may be generated from some MTBE byproducts.
Formaldehyde reacted very rapidly with the hydroxyl radicals, however, its time profile
suggested that it was possibly generated fiom some of the MTBE byproducts. Other
intermediates, such as hydroxy-iso-butyraldehyde, hydroxyacetone, pyruvaldehyde, and

some organic acids such as hydroxy-iso-butyric, formic, pyruvic, acetic, and oxalic acid

19

were also detected during the reaction time of 80 minutes. Acetaldehyde, iso-
butyraldehyde, glycolic, and glyoxylic acids were detected at concentrations lower than
0.02 mM (Stefan et al., 2002). All organic compounds were eventually mineralized after
80 minutes in this study. As part of this study, the byproducts TBF and TBA were also
subjected to treatment using H202/UV. It was observed that when a 0.92 mM TBF
solution was treated with 10 mM H202, the major primary byproducts obtained were
acetone and formaldehyde. Other primary intermediates obtained were formic acid,
hydroxy-iso-butyraldehyde, and low levels of TBA. Further byproducts obtained were
hydroxyacetone, pyruvaldehyde, hydroxy-iso-butyric, pyruvic, acetic, and oxalic acids. a
good carbon balance was obtained during the reaction time of 80 minutes. 0.55 mM TBA
solution was treated with 10 mM H202 solution for 60 minutes, which resulted in
complete mineralization. The primary intermediates formed were acetone, formaldehyde,
and hydroxy-iso-butyraldehyde. Pyruvaldehyde, formic, hydroxy-iso-butyric, pyruvic,
acetic, and oxalic acids were formed as a result of further oxidation processes. It should
be noted that the pH of the reaction mixture was not adjusted to a value between 2 and 3,
as it had been in all other studies considered. This may have lead to different products

being formed or effected the rate of the reaction.

Yeh and Novak (1995) studied the effect of hydrogen peroxide on the degradation of
MTBE in soils and observed that MTBE was chemically oxidized to tertiary butyl
alcohol (TBA) and acetone when hydrogen peroxide was added to soil microcosms that
contained aquifer material from an organic rich site. The soil utilized for this study was

mainly silty loam and rich in organics. The concentrations of total iron and dissolved

20

oxygen in the groundwater at this site were 7.0 mg/L and 0.4 mg/L, respectively. The
aquifer material was obtained from depths of 0, 0.6, 1.2, 2.1, and 3 m. 2 uL of a 30%

H202 solution was added to the soil microcosms every 7 to 10 days.

The approximate values obtained for the initial and final concentrations of MTBE in soil

microcosms receiving 2p.L of 30% H202 each week have been summarized in table 2-1.

Table 2-1. Degradation of MTBE in soil microcosms receiving 2 ul of H202 (Y eh

 

 

 

and Novak, 1995)
Experimental Initial MTBE Final MTBE
Conditions Concentration (mg/l) Concentration (mgfl)
H202 Added 100 85
H202 + Nutrients Added 105 55

 

 

 

 

 

The degradation of MTBE increased when nutrients were added to the system has been
attributed to the impact of a biological phenomenon, rather than a chemical reaction (Y eh

and Novak, 1995).

In the work done by Yeh and Novak, 1995, it was observed that soils from 1.2 m and
2.1 m depths exhibited the highest rates of MTBE loss. The approximate values of MTBE
degradation rates obtained by Yeh and Novak (1995) at depths of 1.2 m and 2.1 m have

been presented in Table 2-2.

21

Table 2-2. MTBE degradation rates in soil microcosms at depths of 1.2 m and

2.1 m (Y eh and Novak, 1995)

 

 

 

 

Experimental Degradation Rate Degradation Rate
Conditions (mg/l/day°g) (mg/[Iday°g)
Depth = 1.2 m Depth = 2.1 m
H202 Added 0.025 0.07
H202 + Nutrients Added 0.08 0.08

 

 

 

 

The variation in the degradation of MTBE with depth may depend on the depth of the
groundwater table, which was approximately 1 m below the surface at this site. This
might have enhanced the movement of the hydroxyl radicals and increased the efficiency

of the process.

Yeh and Novak, 1995, conducted another set of experiments where the oxidation of
MTBE was carried out in aqueous solution without the addition of soil. The experiment
was conducted using an initial concentration of MTBE of approximately 80 mg/L, Fe2+
concentration of 1 mM, H202 volume of 2 pL and a pH value of 4.0. The concentration
of MTBE was reduced to approximately 30 mg/L. TBA, acetone, and several other
unidentified reaction products were obtained (Y eh and Novak, 1995). It is thus possible
that when the oxidation of MTBE was carried out in soil, no other byproducts were

obtained due to the oxidation of those byproducts by soil microorganisms.

It was also observed that when the initial concentration of Fe2+ was 1 mM, the volume of

30% hydrogen peroxide used was 2 uL, and the pH was 4.0, the MTBE oxidized was

22

0.45 mg and 0.5 mg without and with the addition of nutrients, respectively. Thus, the
oxidation of MTBE was not inhibited by the phosphate was present in solution. It was
hypothesized that phosphate undergoes reactions in solution such as precipitation,
complexation, and adsorption, which reduces the impact of phosphorous on H202 (Y eh

and Novak, 1995).

It was also hypothesized that biological degradation of TBA and acetone, generated
during the degradation of MTBE, may deplete the oxygen in the soil, thus creating an
environment that would cause iron reduction. The reduced iron would thus catalyze the
added hydrogen peroxide to generate hydroxyl radicals. This would lead to an aerobic-
anaerobic cycle (Yeh and Novak, 1995). Since MTBE is recalcitrant to biodegradation,
this mechanism could be used for remediation of MTBE in groundwater containing iron

(Y eh and Novak, 1995).

Ray and Selvakumar, (2000), studied the removal of MTBE from water using Fenton’s
Reagent. The major byproducts obtained in their study were tert-butyl alcohol, tert-butyl
formate, and acetone. Small quantities of isobutene, and a gas of molecular weight 56,
not soluble in water, were also detected but not quantified. The volume of gas produced
during this experiment was not quantified. Mass balances for two tests, which gave
MTBE removal percentages of approximately 99% each at H202: Fe2+ ratios of 1:1 and
6:1 respectively, were conducted as input of MTBE and output of products (calculated as
carbon). The initial MTBE concentrations in these tests were approximately 21.0 mg/L

and 17.4 mg/L, respectively, and the initial concentration of a 5% H202 solution used was

23

236 mg/L for both these tests. Approximately 80% and 60% of the product was
unaccounted for H202: Fe2+ ratios of 1:1 and 6:1 respectively, which was attributed to the
formation of some volatile products that escaped the reaction system, or the generation of
certain non-volatile organics. These experiments show that certain byproducts that were

formed were not identified and quantified during the experiments.

From the above review of several studies conducted on the treatment of MTBE using
Fenton’s reagent, it is obvious that better techniques and experimental designs are
required to ensure that volatilization losses of the parent compound and its byproducts are
minimized to efficiently identify and quantify the byproducts formed. This is very
essential if this technology is to be adopted for use in the field. It is also possible that the
byproducts depend on the ratio of concentration of MTBE and H202 used, the time of the
reaction, and the presence of other organic compounds, such as other contaminants or soil
organic matter, that may be present in the system. These conditions should be evaluated
and optimum conditions must be developed using bench-scale studies before Fenton’s

reagent or similar Advanced Oxidation processes are applied in the field.

In the current study, some of the byproducts obtained with and without the addition of
soluble Fe2+ to the soil system have been discussed in later chapters. However, additional
work is recommended for complete identification and quantification of all the byproducts

and intermediates formed during the reaction of MTBE with Fenton’s Reagent.

24

CHAPTER 3

MATERIALS AND METHODS

3.1 Materials

Chemicals. Methyl-tert butyl ether (99%) was obtained from Acros Organics. Methanol,

Baker analyzed ® I-IPLC solvent, was used for the preparation of the MTBE stock

solution. All dilutions were prepared using distilled deionized water (Ultra Pure Water
System, Technic Central Systems). MTBE stock solution (40,000 mg/L) was stored at

4°C for not more than a week.

Soil. Metea soil, which is a low organic content soil, was cleaned of debris, and the soil

was screened using a 20-mesh sieve. The sieved soil was then contaminated with MTBE
in the lab to achieve a concentration of 50 g MTBE/kg of soil. This soil was placed in a
bell jar, which was wrapped with aluminum foil to prevent the photolysis of MTBE, and

stored at 4 °C to slow down bacterial degradation and avoid volatilization of MTBE.

The soil characteristics of Metea soil are summarized in Table3-l.

25

Table 3-1. Soil characteristics of Metea soil ‘

 

 

 

 

 

 

 

 

 

 

 

Category Result Method b
pH 5.43 i 0.87 Water pH Measurement
Organic Matter 0.4 :l: 0.1 Loss on Ignition
(%)
Iron (ppm) T 2 0.1 N HCl Extraction
Lime Index 71 SMP Buffer Method
Neutralizing 13.1 Indicator Titrimetric Method
Value (%)
% Sand 78.9 Bouyoucos Hydrometer
% Silt 12.5 Bouyoucos Hydrometer
% Clay 8.6 i 0.71 Bouyoucos Hydrometer

 

 

 

8 Soil tests conducted at Soil and Plant Nutrient Laboratory, Michigan State University
b Recommended Chemical Soil Test Procedures for the North Central Region, SB 1001,

Revised January, 1998

3.2 Analytical Methods

MTBE was quantified using a Perkin Ehner Headspace Sampler HS 40 equipped with a
Perkin Elmer Auto System Gas Chromatograph (GC) and a photo ionization detector
(PID). The GC was equipped with a fused silica PE 624 glass capillary column (30 m x
0.53 mm). The oven temperature was programmed at 45 °C for the entire run. The
injector and detector temperatures were 200 °C each. The carrier gas flow rate was
19.0 mL/min. The sample temperature for the autosarnpler was 80 °C. The needle
temperature and transfer temperature were both 100 °C. The thermostat time was 25.0

minutes. The pressure was maintained at approximately 25 psi.

Assay and calibration samples were carried out in clear, flat bottom 20 ml headspace

vials, which were sealed with 20 mm crimp seal with Teflon® lined rubber septa

26

obtained from National Scientific Company, Duluth, GA. A calibration curve was
obtained by diluting the stock solution in double deionized water. The samples vials each
contained 5 ml aqueous solution. The calibration curve, which bracketed the
concentrations of the test samples, was obtained by running triplicate samples of MTBE
solutions at concentrations between 0 and 160 mg/L. The MTBE stock solution was kept
chilled in an ice bath throughout the experiment to avoid the volatilization of MTBE.
Chromatograms were recorded and the data was assessed using the Turbochrom

Navigator software (Perkin Elmer Corp).

The analysis of the byproducts was conducted using a Perkin Elmer Autosystem gas
chromatogram equipped with an ECD, an autosampler, and a DB-5 column (J & W
Scientific, Folsom, CA). The oven temperature was programmed to hold for 1 minute at
50°C. The temperature was then programmed to increase to 220°C at a rate of 4 °C/min,
and finally increase to 250°C at a rate of 20 °C/min, where it was held for 5 minutes. The
injector and detector temperatures were 180 °C and 300 °C, respectively. The carrier flow

rate was 1.0 mL/min.

Formaldehyde, propionaldehyde, and acetone were analyzed using USEPA Method 556

with the modification of using MTBE instead of hexane for the extraction (U SEPA,

Method 556).

27

CHAPTER 4

TREATMENT OF MTBE USING FENTON’S REAGENT (I)

4.1 Introduction

The use of F enton’s Reagent has been documented as an effective technology for the
oxidation of MTBE in contaminated soil and groundwater (Ray and Selvakumar, 2000;
Kealey et al., 2001; Burbano et al., 2003; Xu et al., 2004). This current study is an
attempt to investigate the oxidation of MTBE using F enton’s Reagent. Fenton’s reagent
utilizes the oxidation power of the hydroxyl radical, 'OH, which is generated as a result
of the reaction between Fe2+ and H202, to degrade organic contaminants. These radicals
are very unstable and have a high oxidation potential of 1.9 eV (Burbano et al., 2003).
They non-selectively attack organic compounds and initialize oxidation pathways to
eventually yield complete degradation pathways of the parent organics (Burbano et al.,
2003). The present work considers the degradation of MTBE with and without the

addition of soluble iron.

4.2 Materials and Methods

4.2.1 Materials
Hydrogen peroxide (purity, 30%) Baker analyzed ® ACS reagent (Stabilized) was used
after dilution. An aqueous solution of 5N solution of sodium thiosulfate (obtained from

Sigma) was prepared using distilled deionized water (Ultra Pure Water System, Technic

28

Central Systems). Reagent water, sterile filtered, was obtained from Sigma-Aldrich Co.,
and was used for preparing 15% and 3% hydrogen peroxide. Sulfuric acid (with more
than 51% acid, Baker Analyzed ® ACS reagent) was used to acidify water and hydrogen
peroxide solutions to obtain an initial pH of 1.8 for the reactions. When 4.65 mL of the
pH 1.8 solutions was added to 0.2 grams of soil, the final pH was between 2 and 3. The
15% and 3% solutions of hydrogen peroxide were prepared before each reaction. The
solution pH was measured using Orion Model SA 720A pH meter. The initial and final

pH of the reaction mixture was measured using pH paper.

4.2.2 Experimental Method

To evaluate if the degradation of MTBE occurred in the presence of mineral iron, this set
of experiments was conducted without the addition of soluble iron. The oxidation
reaction of hydrogen peroxide was carried out in clear, flat bottom 20 mL headspace
vials. The pH of the initial reaction mixture was maintained between 2 and 3 for optimum
reaction conditions (Pignatello, 1992; Burbano et al., 2003; Xu et al., 2004) by using
acidified water and hydrogen peroxide. These headspace vials, containing appropriate
volumes of acidified water, were weighed on an analytical balance. An aliquot (0.2
grams) of the contaminated soil, which had been stored at 4 °C, was added to these vials
using a spatula. The soil was introduced into the vials gently to reduce agitation, which
might drive off volatile compounds, and the vials were then sealed immediately using a
Mininert 2—way valve. Only 0.2 g of soil was chosen due to the exothermic nature of the

reaction.

29

The reaction was initiated by adding acidified hydrogen peroxide, which was introduced
into these vials via a syringe that could pass through the auxiliary septum of the valve.
The final H202 concentrations in the reaction mixtures were between 60 and 40,000 mg/L
(i.e. 1500 mg/kg soil and 1,000,000 mg/kg soil, respectively). 15% H202 was used for
concentrations from 40,000 mg/L (1,000,000 mg/kg) to 15,000 mg/L (375,000 mg/kg).
3% H202 was used for concentrations from 10,000 mg/L (250,000 mg/kg) to 60 mg/L
(1500 mg/kg). Two solutions with different concentrations of hydrogen peroxide were
used so that the volume of H202 added was easily measurable. The volumes of acidified
water and the hydrogen peroxide were adjusted so that the final volume of the reaction
mixture in the headspace vials was 5 mL. The auxiliary septum of the 2-way valve was
immediately sealed after the hydrogen peroxide was introduced. The samples were then
mixed for 120 minutes using a Mistral Multi Mixer, Melrose Park, ILL. Rapid bubbling
was seen for approximately 10 minutes after the addition of the hydrogen peroxide. As a
result, 120 minutes was considered a sufficiently long time for the reaction to take place.
After 120 minutes, the reaction was terminated by adding 200 uL of 5N sodium
thiosulfate solution (Xu et al., 2004). To get the correct volume of sodium thiosulfate,
aliquots of sodium thiosulfate were added gradually until the reaction ceased (as noted by
a termination of bubble formation). These vials were then chilled at 4°C for 60 minutes to
prevent volatilization losses before analysis. The 2-way valves were then replaced with
20 mm crimp seal with Teflon lined rubber septa, and these samples were then analyzed
using the Perkin Elmer Headspace Sampler HS 40 equipped with a Perkin Elmer Auto

System Gas Chromatograph (GC) and a photo ionization detector (PID). Chromatograms

30

were read and the data was interpreted using the Turbochrom Navigator software (Perkin

Elmer Corp.).

This experiment was conducted three times in triplicate. During the first trial, 15% H202
was used throughout the experiment and initial results from 60 mg/L (1500 mg/kg soil) to
10,000 mg/L (250000 mg/kg soil) were inaccurate due to experimental error. This series
of experiments were then repeated using 15% and 3% hydrogen peroxide. The third set
was conducted with a longer run time than the earlier sets to look at the presence of
byproducts. The data from the complete second and third sets and half of the first set,
from hydrogen peroxide concentrations 10,000 mg/L (250,000 mg/kg soil) to 40,000
mg/L (1,000,000 mg/kg soil) are shown in this chapter. The complete data from the first
set is shown in the Appendix. The mean values for each experimental trial are presented

in the figure together with vertical error bars indicating the standard deviations.

4.3 Results and Discussion

The initial pH of the reaction mixture was maintained between 2 and 3 (Pignatello, 1992;
Burbano et al., 2003; Xu et al., 2004). Xu et al. (2004) found that at pH values greater
than 4, the decomposition rate of MTBE decreases due to a reduction in the concentration
of free iron species in solution. This could be due to the formation of Fe (H) complexes,
which would hinder further reaction between Fe2+ and H202 (Xu et al., 2004).
Additionally, at pH greater than 3, retardation of the reaction between MTBE and
hydrogen peroxide may occur due to the precipitation of iron oxides such as F e203°nH20,

and Fe(OH)3 (Pignatello, 1992; AMRClearinghouse, 2004). At low pH, the reaction

31

between MTBE and hydrogen peroxide is hindered since at high concentration of H”,
equilibrium would shift towards the left, thus, inhibiting the formation of Fe2+ as is seen

in reaction (2-9) (Pignatello, 1992).

Fe3+ + H202 <———> H+ + Fe00H2+ —> Fe2+ + H02 (29)

Additionally, at very low pH, i.e. at very high concentration of hydrogen ion, W becomes
a major hydroxyl radical scavenger. This has been shown in reaction 4-1 (Tang and

Huang, 1996). The electrons for this reaction may be obtained from ferrous ions.

OH + H” +e' —> H2O (4-1)

For this reaction, k = 7 x 109 M'ls'1

4.3.1 Effect of H202
The only iron present was in the form of mineral iron in the soil, and that amounted to 2
ppm. It was observed that the removal percentage of MTBE increased with an increase in

the concentration of the hydrogen peroxide added. This is shown in Figure 4-1.

32

120 -

 

100 ~
80 ~
8
‘3 +Trial 1
.3 6O — +Tria12
3 Trial 3
Q

 

 

 

I I I I

0 10000 20000 30000 40000 50000
H202 Concentration (mg/L)

Figure 4-1: Effect of hydrogen peroxide concentration on percent oxidation of

MTBE, initial pH = 2-3. Initial concentration of MTBE was between 299.25 and

429.75 mg MTBE/kg soil

The results of Trial 1, 2, and 3 have been shown in Tables 4-1, 4-2, and 4-3. These depict

the percent oxidation of MTBE and concentration of MTBE as the concentration of

hydrogen peroxide were varied.

33

Table 4-1.

Trial 1. Average initial concentration of MTBE = 429.75 mg/L

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H202 H202 % Oxidation :1:

Concentration Concentration Standard Concentration :1: Standard
(mg/L) (mflg soil) Deviation Deviation (mg MTBE/kgsoil)
40000 1000000 99.1 :t 0.38 3 18.13 i 1.57
35000 375000 95.4 :t 1.35 33.45 i 5.63
30000 750000 93.9 i 2.78 39.57 i 11.55
25000 625000 89.1 i 2.82 59.66 i 11.73
20000 500000 91.7 i 3.57 48.73 :i: 14.84

. 15000 375000 81.3 :t 4.99 91.79 i 20.72
10000 250000 67.1 i 9.32 150.82 i 38.72
Table 4-2. Trial 2. Average initial concentration of MTBE = 340.25 mg/L

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H202 H202 % oxidation :I:

Concentration Concentration Standard Concentration :1: Standard
(mg/L) (mm soil) Deviation Deviation (HE MTBE/kg soil)
40000 1000000 2 99.3 i 0.00 ND
35000 875000 97.8 i 2.16 21.55 :1: 4.79
30000 750000 95.4 i 1.69 28.71 i 3.39
25000 625000 91.5 i 3.10 41.75 1 10.21
20000 500000 87.1 i 3.28 55.19 i 6.10
15000 375000 77.5 :t 4.12 87.76 :t 20.21
10000 250000 65.0 i 4.34 130.07 1 38.26

8000 200000 39.1 i 6.19 211.88 i 39.95
5000 125000 21.5 i 3.43 268.87 1: 46.10
1200 30000 13.61 i 11.78 290.47 :l: 20.51
120 3000 9.0 i 6.64 308.41 i 47.09
60 1500 6.5 :35] 319.40i71.16

 

 

 

 

ND =Not Detected

34

 

 

Table 4-3. Trial 3. Average initial concentration of MTBE = 419 mg/L

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

H202 H202 % Oxidation :1: Concentration :1:
Concentration Concentration Standard Standard Deviation
(mg/L) (mg_/l<_g soil) Deviation (mgMTBE/kg soil)
40000 1000000 2 99.3 d: 0.00 ND
35000 875000 98.8 i 0.29 19.13 :1: 1.17
30000 750000 97.8 :i: 0.19 23.15 i 0.78
25000 625000 95.8 i 0.82 31.28 i 3.30
20000 500000 95.1 i 0.55 34.14 :1: 2.23
15000 375000 91.1 i 1.29 50.29 i 5.22
10000 250000 79.3 :1: 2.36 98.00 i 9.57
8000 200000 34.1 :t 4.83 280.88 1 19.54
5000 125000 18.3 i 6.47 344.75 :t 26.18
1200 30000 11.1 i 2.98 373.74 :1: 12.07
60 1500 7.8 d: 6.88 387.03 i 27.84

 

ND =Not Detected

As can be see from Figure 4-1, the percent removal of MTBE increases with increasing
hydrogen peroxide concentration, with almost complete oxidation of MTBE at an H202
concentration of 40,000 mg/L (1,000,000 mg/kg soil). The results obtained from all
experimental runs are very similar. This would mean that at high concentrations of
hydrogen peroxide, nearly complete degradation of MTBE is possible if the initial pH is
between 2 and 3, and under similar MTBE: H202 ratios and soil type used. This can be
explained because when hydrogen peroxide is in excess, due to the rapid oxidation of
Fe2+, Fenton’s reaction often written as ‘Fe2+ + H202’, is more accurately represented as

Fey/H202. This is often missed in literature. The main oxidant in the reaction is believed

to be °OH (Pignatello, 1992).

35

 

Pignatello (1992) investigated the degradation of the herbicides 2,4-
dichlorophenoxyacetic acid (2,4-D) and 2,4,5-t1ichlorophenoxyacetic acid (2,4,5-T). He
discovered that conversion of the organic contaminant to CO2 was independent of the Fe
oxidation state. In his study, conducted at a pH 2.7-2.8, the concentration of 2, 4-D
decreased from 0.1 mM to ahnost 0.0 mM in less than 20 minutes when the Fe3+
concentration used was 1.0 mM and that of hydrogen peroxide was 10 mM. The
degradation of 2,4,5-T at a pH of 2.8 was slower, which was possibly due to other
competing reactions (Pignatello, 1992). He also discovered that 2,4-D ring mineralization
increased from 45 to 69% when the hydrogen peroxide concentration was increased from
10 to 500 mM. This trend is consistent with the present study, where it was also observed
that the degradation of MTBE increased with increase in hydrogen peroxide

concentration.

In the present study, it was observed that approximately 95% or greater degradation of
MTBE was obtained when the concentration of H202 was 30,000 mg/L (750,000 mg/kg
soil) or higher. Additionally, the percent degradation of MTBE was approximately 92%
when the H202 concentration was approximately 588 mM (19,992 mg/L or 499,800
mg/kg soil). It can thus be inferred that the nature and the initial concentration of the
contaminant, presence of other organic compounds that may be present on the site play a
significant role in deciding the hydrogen peroxide dose for nearly complete oxidation of

the contaminant.

36

4.3.2 Effect of Fe3+ on MTBE Degradation in Soil
In the presence of Fe2+ and F e”, hydrogen peroxide forms highly reactive hydroxyl
radicals (OH), and possibly other reactive species (Pignatello, 1992) (Refer Reactions 2-

1 to 2-9).

From the experiments conducted in this study, it can be concluded that in the presence of
excess peroxide, nearly complete oxidation of MTBE can be achieved using Fey/H202.
In the presence of excess peroxide, the concentration of Fe” is much smaller than that of
Fe3+ since the rate of decomposition of hydrogen peroxide by ferric ion is slower than

that by ferrous ion (Barb, 1951) (Refer Reactions 2-1 and 2-4).

In studies conducted by Pignatello with hydrogen peroxide, he observed that under initial
conditions of [2,4-D] = 0.1mM, [Fe] = 1mM, [H202] = 10mM, pH = 2.7, the percent 14C
in solution for ring-”C-2,4-D vs. was ahnost identical for both Fe3+ and Fe“ except
during the first few minutes. He concluded that under conditions of excess peroxide,

there seemed to be little or no advantage of using Fe2+ over Fe”.

4.3.3 pH Monitoring for Soils Treated with Hydrogen Peroxide

The pH of the reaction mixture for each of the samples was measured before and after the
reaction. As mentioned before, the initial pH of the reaction mixture for each sample was
maintained between 2 and 3. It was observed that the final pH of the reaction mixture was
between 1 and 2 for samples where the hydrogen peroxide added was between

40,000 mg/L (1,000,000 mg/kg soil) and 10,000 mg/L (250,000 mg/kg soil). The pH

37

increased gradually as the amount of hydrogen peroxide added decreased. The final pH of
the reaction mixture for a hydrogen peroxide concentration of 60 mg/L (1500 mg/kg soil)
was between 7 and 8. Please refer the Figure 4-4 for pH vs. concentration of H202 table.
The controls contained 0.2g of contaminated soil and acidified water, and their initial pH
was also maintained between 2 and 3. The final pH of the controls, to which hydrogen
peroxide was not added, was around 7. Table 4-4 depicts variation of pH with hydrogen

peroxide concentration.

Table 4-4. Variation of pH with hydrogen peroxide concentration for Trials 1, 2,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

and 3
H202 Concentration (mg/L) H202 Concentration (mglkgfsoil) pH

40000 1000000 Between 1-2
35000 875000 Between 1-2
30000 750000 Between 1-2
25000 625000 Between 1-2
20000 500000 Between 1-2
1 5000 375000 Between 1-2
10000 250000 Between 1-2
8000 200000 Between 2-3
5000 125000 Between 3-4
1200 30000 Between 5-6
120 3000 Between 7-8
60 1500 Between 7-8

 

 

 

 

 

From this data, it can be deduced that in the presence of excess H202, the pH of the
reaction mixture decreases due to the release of protons as is seen fi'om the Reaction (2-

4).

H202 + Fe3+ —* Fe” + H” + H02 (24)

38

When the concentration of the hydrogen peroxide added is low, the pH of the soil
increases. This can be explained based on the fact that the soil used in this study is sandy
soil (78.9%), which has a pH of 5.43 :l: 0.87. Sandy soil is a poor buffer, and hence does
not resist changes in pH. The fact that the final pH of the controls was approximately 7

further supports this theory.

The slightly higher pH of the soil treated with low concentrations of hydrogen peroxide
than the controls could be due to the generation of HO' during reaction (2-1).

H202 + Fe” Fe” + H0' + -OH (21)

——>

4.4 Formation of Byproducts

As shown in Figure 4-2, it was observed that at the initial higher concentrations of
hydrogen peroxide, higher molecular weight byproducts were formed. It is believed that
these byproducts may be dimers formed due to the presence of free radicals such as
'C(CH3)3. The formation of these byproducts did not occur at hydrogen peroxide

concentrations lower than 15,000 mg/L (375,000 mg/kg soil).

It was also observed that two lower molecular weight byproducts were formed at nearly
all concentrations of hydrogen peroxide added. Based on the retention times of the
compounds, these compounds were tentatively identified as propionaldehyde and

acetone. The concentration of these byproducts increased with the decrease in the

39

concentration of hydrogen peroxide, and this leads to the conclusion that the reaction is

less complete at lower concentrations.

Shown below are several Chromatograms with varying concentrations of hydrogen
peroxide. No formation of dimers at lower hydrogen peroxide concentrations was

observed.

5.60
5.50
5.40
5.30

5.20

5.10 L
5.00 I

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0

 

 

 

Response (mV)

 

Time (minutes)

Figure 4-2. Initial [H202] = 20,000 mg/L (500,000 mg/kg soil), pH = 2-3, no Fe2+
added. Chromatogram shows presence of lower molecular weight compounds and

dimers

40

5.50

5.40

5.30

5.20

Response (mV)

5.10

5.00 U L _

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0

 

 

 

 

 

Time (minutes)

Figure 4-3. Initial [H202] = 15,000 mg/L (375,000 mg/kg soil), pH = 2-3, no Fe”

added. Only formation of lower molecular weight compounds was observed

41

6.0

5.8

5.6

5.4

Response (mV)

5.2

 

 

 

5.0

 

 

itirlrrrnlrLrllllllluntlnrr111 11111111111111111114411111111111

 

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0

Time (Minutes)

Figure 4-4. Initial [H202] = 10,000 mg/L (250,000 mg/kg soil), pH = 2-3, no Fe2+

added

42

8.5

8.0

7.5

7.0

6.5

Response (mV)

6.0

5.5

5.0

Figure 4-5.

added

 

 

 

 

 

UL _

 

.ILIIlllIllllIllJlIIIILLIIIIILLILIIIIIIIILJ

 

 

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0

Time (minutes)

Initial [H202] = 8000 mg/L (200,000 mg/kg soil), pH = 2-3, no Fe2+

43

8.0

7.5

 

7.0

6.5

6.0

Response (mV)

5.5

 

 

.Jblrc

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0

5.0

 

 

[JJIIIIIIIIJLJIIIIIlllIlLllIllllIl

 

Time (minutes)

Figure 4-6. Initial [H202] = 5000 mg/L (125,000 mg/kg soil), pH = 2-3, no Fe2+

added

The concentration of the lower molecular weight byproducts decreased when the

hydrogen peroxide concentration reached 5000 mg/L (125,000 mg/kg soil). This may

have occurred because hydrogen peroxide becomes the limiting reagent at these lower

concentrations.

Tables 4-5 and 4-6 depict the percent conversion of MTBE to propionaldehyde and
acetone on a molar basis. However, as can be seen from Table 4-7, the total conversion

does not add to 100%. Thus, a more complete study of the byproducts needs to be done to

identify and quantify all of the byproducts and intermediates.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 4-5. Percent conversion of MTBE to propionaldehyde
H202 H202 Initial MTBE
Concentration Concentration concentration Propionaldehyde % Conversion to
(mgL) (mm soil) (mmoles) formed (mmoles) propionaldehyde
8000 200000 8.82x10“ 1.65x10'5 1.9
10000 250000 8.82x10‘4 1.41x10'5 1.6
Table 4-6. Percent conversion of MTBE to acetone
H202 H202 Initial MTBE
Concentration Concentration concentration Acetone formed % Conversion to
(mg/L) (mflg soil) (mmoles) (mmoles) acetone
8000 200000 8.82x10'4 1.79x10'5 2.0
10000 250000 8.82x104 ND NA

 

 

 

 

 

 

ND = Not detected, NA = Not applicable

45

 

 

Table 4-7. Total conversion

 

 

 

 

 

 

 

 

 

H202 H202 Initial MTBE
Concentration Concentration concentration Total conversion to
(mg/L) (mg/3g soil) (mmoles) byproducts (%)
8000 200000 8.821110“ 3.9
10000 250000 8.82x104 1.6

 

4.5 Volatilization Losses

In this current study, Metea soil had initially been contaminated with MTBE at a
concentration of 50,000mg/kg in December in the lab. This concentration dropped to
approximately 1280 mg MTBE/kg soil in April. The soil finally leveled off to a
concentration of approximately 416 mg MTBE/kg soil. These losses have been attributed

to volatilization. This proves that MTBE is very volatile in the environment.

In a study conducted by Callender and Davis, they calculated the Henry’s constant to be
0.032 at 25°C, indicating that MTBE was more volatile than previously reported, which

reported a value of 0.022 at 25°C (Callender and Davis, 2001).

Due to the volatility of MTBE, experiments must be designed so as ensure that losses due
to volatilization of MTBE are minimized during the experiments. Only then can one
make correct judgments about degradation of MTBE using a particular treatment

technology.

46

It can be concluded that although the oxidation of MTBE is ahnost complete at high
concentrations of hydrogen peroxide, there are byproducts that are formed that must be
identified before hydrogen peroxide should be used in the field for the treatment of
MTBE. Additionally, these byproducts must be analyzed for their toxicity, and their fate
and transport in the environment before this technology is recommended for use in the

field for the remediation of MTBE.

47

CHAPTER 5

TREATMENT OF MTBE USING FENTON’S REAGENT (II)

5.1 Introduction

As was seen in the previous chapter, in the presence of mineral iron and at optimum pH
conditions, MTBE was effectively degraded under conditions of excess hydrogen
peroxide. Further experiments were conducted by adding Fenton’s Reagent, i.e., Fe2+ and
hydrogen peroxide solution to the MTBE contaminated soil. This chapter discusses the

results obtained fi'om these experiments.

5.2 Materials and Methods

5.2.1 Materials

Hydrogen peroxide (purity, 30%) Baker analyzed ® ACS reagent (Stabilized) diluted to
3% using reagent water, sterile filtered, obtained from Sigma-Aldrich Co., was used for
the reaction. Aqueous solutions of 0.1M ferrous ammonium sulfate (obtained from
Sigma) and 5N solution of sodium thiosulfate (obtained fi‘om Sigma) were prepared using
double deionized water (Ultra Pure Water System, Technic Central Systems). Sulfuric
acid (> 51% acid, Baker Analyzed ® ACS reagent) was used to acidify water, hydrogen
peroxide, and ferrous ammonium sulfate solutions to obtain an initial pH of 1.8 for all the
solutions. When 4.65 ml of the pH 1.8 solutions was added to 0.2 grams of soil, a final

pH of 2-3 resulted. The 3% solution of hydrogen peroxide was prepared fresh before each

48

reaction. The solution pH was measured using Orion Model SA 720A pH meter. The

initial and final pH of the soil slurry was measured using pH paper.

5.2.2 Experimental Method

To evaluate if the degradation of MTBE occurred in the presence of soluble ferrous iron
catalyzed hydrogen peroxide, this set of experiments was conducted by adding freshly
prepared ferrous ammonium sulfate solution, its final concentration between 0.1 mM
(140 mg/kg soil) and 50 mM (70,000 mg/kg soil) in the reaction mixture. The oxidation
reaction with Fenton’s Reagent was carried out in clear, flat bottom 20 ml headspace
vials. The pH of the initial reaction mixture was maintained between 2 and 3 for optimum
reaction conditions (Pignatello, 1992; Burbano et al., 2003; Xu et al., 2004) by using
acidified water, ferrous ammonium sulfate, and hydrogen peroxide. An aliquot (0.2
grams) of the contaminated soil, which had been stored at 4°C, was added to these
headspace vials, which already contained appropriate volumes of acidified water and
ferrous ammonium sulfate solution, using a spatula. The soil was introduced into the vials
gently to reduce agitation, which might drive off volatile compounds, and the vials were
then sealed immediately using a Mininert 2-way valve. Only 0.2 g of soil was chosen due

to the exothermic nature of the reaction.

The reaction was initiated by adding acidified hydrogen peroxide, which was introduced
into these vials via a syringe that could pass through the auxiliary septum of the valve.
The concentration of Fe2+ in the reaction mixture was varied from 0.1 mM (140 mg/kg

soil) to 50 mM (70,000 mg/kg soil). The initial concentration of hydrogen peroxide was

49

kept constant at 510 mg/L (12,750 mg/kg soil) for all samples. 3% H202 was used for all
reactions. The volumes of acidified water, ferrous ammonium sulfate, and the hydrogen
peroxide were adjusted so that the final volume of the reaction mixture in the headspace
' vials was 5 ml. The 2-way valve was immediately sealed after the introduction of the
hydrogen peroxide. The samples were mixed for 120 minutes using a Mistral Multi
Mixer, Melrose Park, ILL. Rapid bubbling was seen for approximately 10 minutes after
the addition of the hydrogen peroxide. As a result, 120 minutes was considered a
sufficiently long time for the reaction to take place. After 120 minutes, the residual
hydrogen peroxide was quenched using 200 uL of 5N sodium thiosulfate solution (Xu et
al., 2004). This volume of sodium thiosulfate was arrived at after several trials of
gradually increasing volumes until the reaction finally ceased (as noted by a termination
of bubble formation). These vials were then chilled at 4°C for 60 minutes to prevent
volatilization losses before analysis. The 2-way valves were then replaced with 20 mm
crimp seal with Teflon lined rubber septa, and these samples were then analyzed using
the Perkin Elmer Headspace Sampler HS 40 equipped with a Perkin Elmer Auto System
Gas Chromatograph (GC) and a photo ionization detector (PID). Chromatograms were
read and the data was interpreted using the Turbochrom Navigator software (Perkin

Elmer Corp.).
This experiment was conducted in triplicate. The mean values have been presented in the

figure together with vertical error bars indicating the standard deviations of the three

experiments.

50

5.3 Results and Discussion

The initial pH of the reaction mixture was maintained between 2 and 3 to ensure
optimum conditions (Pignatello, 1992; Burbano et al., 2003; Xu et al., 2004). At pH
values greater than 4, the decomposition rate of MTBE decreases due to the reduction in
the free iron species in solution. This could be due to the formation of Fe (1]) complexes,
which would hinder fin‘ther reaction between Fe2+ and H202 (Xu et al., 2004).
Additionally, at pH values greater than 3, retardation of the reaction between MTBE and
hydrogen peroxide may occur due to the precipitation of iron oxides such as F e203'nH2O,
and Fe(OH)3 (Pignatello, 1992; AMRClearinghouse, 2004). At low pH, the reaction
between MTBE and hydrogen peroxide is hindered since at high concentration of H”,
equilibrium would shift towards the left, thus inhibiting the formation of Fe2+ as is seen in

the Reaction (2-9) (Pignatello, 1992).

Fe” + H202 <——> H+ +Fc00H” ——> Fe” + H02 (29)

5.3.1 Effect of Fe2+ on Degradation of MTBE in Soil

The relative oxidizing power of hydroxyl radical from Fenton’s Reagent is 2.06, while
that of hydrogen peroxide is 1.31 (C12 = 1) (Jacobs et al., 2001). As a result, hydroxyl
radicals generated fi'om Fenton’s Reagent can non-selectively attack organic compounds
and initialize oxidation pathways to eventually yield complete degradation pathways of

the parent organics (Burbano et al., 2003).

51

Fe2+ can catalyze the production of OH fi‘om H202 (Reaction 2-4) (Xu et al., 2004; Barb
etal,1951)

H202 + Fe” Fe” + H0“ + on (2-1)

——>

Since the concentration of Fe” influences the production of OH, it has a significant

effect on the degradation of MTBE.

In this experiment, the initial concentration of hydrogen peroxide was 510 mg/L (12,750
mg/kg soil), while the initial concentration of Fe2+ was varied between 0.1 and 50 mM
(i.e., 140 mg/kg soil and 70,000 mg/kg soil, respectively). It was observed that the
percent oxidation of MTBE increased abruptly as the concentration of Fe” in the reaction
mixture was increased. The MTBE concentration was below limits of detection for
F e2+/1~I2O2 molar ratios between 0.27 and 1.3. As shown in Figure 5-1, the degradation of
MTBE using Fenton’s Reagent gradually decreased as the Fe2+/H2O2 ratio was increased

beyond 1.3.

In studies conducted by Xu et al., 2004, the reactions were carried out at an initial MTBE
concentration of 1 mM, hydrogen peroxide concentration of 10 mM and a pH of 2.8. It
was observed that the percent of MTBE removal increased as the concentration of Fe2+
was increased. The MTBE removal peaked at a Fe2+ concentration of 2 mM, where the

percent MTBE degradation was approximately 94%. The percent degradation started to

52

decrease beyond this Fe2+ concentration. This agrees with the results obtained in the

current study.

120.00 1
100.00 -
80.00 -

60.00 ~

% Oxidation

40.00 -

20.00 :-

 

 

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

Fe“ Concentration (mM)

Figure 5-1. Effect of Fe2+ concentration on percent oxidation of MTBE. Initial
concentration of MTBE was approximately 388.75 mg MTBE/kg soil,

[H202].) = 510 mg/L (12,750 mg/kg soil), pH = 2-3.

53

Under similar experimental conditions of pH and initial MTBE concentration, when no
Fe2+ was added to the reaction mixture, only approximately 15% of MTBE was degraded
with a hydrogen peroxide concentration of 510 mg/L (12750 mg/kg soil) as was seen in
the experiments conducted without the addition of Fe“. It can be concluded that the

addition of Fe2+ to hydrogen peroxide can greatly enhance the oxidation of MTBE.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 5-1. Variation of MTBE concentration and percent oxidation with Fe2+
concentration. Average initial concentration of MTBE = 388.75 mg
MTBE/kg soil.
MTBE
H202 Fe2+ Fe2+ Concentration
Concentration Concentration Concentration % MTBE (mg MTBE/kg
(mg/L) (mM) (mg/kg soil) Oxidation soil)
510 0.1 140 14.4 i 6.83 334.74 i 25.57
510 0.5 700 56.2 i 5.09 178.21 at 19.07
510 l 1400 70.2 :t 2.32 125.80 i 8.69
510 2 2800 90.0 i 3.08 51.80 :t 11.54
510 4 5600 2 99.3 :1: 0.00 ND
510 6 8400 2 99.3 i 0.00 ND
510 8 11200 2 99.3 i 0.00 ND
510 10 14000 2 99.3 i 0.00 ND
510 15 21000 2 99.3 :i: 0.00 ND
510 20 28000 2 99.3 i 0.00 ND
510 30 42000 97.4 i 0.13 23.97 :1: 0.50
510 40 56000 84.6 i 2.25 71.81 i 8.43
510 50 70000 66.2 i 3.55 140.79 :t 13.29

 

 

 

 

 

 

ND = Not detected

Table 5-1 depicts the percent oxidation and concentration of MTBE as Fe2+ concentration

is varied from 0.1 mM (140 mg/kg soil) to 50 mM (70,000 mg/kg soil). The H202

concentration was 510 mg/L (12,750 mg/kg soil) for all samples.

54

 

In studies conducted by Pignatello (1992), complete transformation of 0.1 mM 2,4-
dichlorophenoxyacetic acid (2,4-D) was complete in less than 1 minute when the
concentrations of Fe2+ and hydrogen peroxide were greater than 1 mM. It was also
observed that when Fe2+ and H202 were present at lower concentrations of 0.25 mM
each, the reaction was over in less than 2 minutes and resulted in 76-88% transformation
of the contaminant. This would suggest that one or both of the reactants could have been
limiting. In the present study, the percent destruction of MTBE was increased from 70%
to 90% when the Fe2+ concentration was increased from 1 mM (1400 mg/kg soil) to 2
mM (2800 mg/kg soil). Since the MTBE concentration was not detectable when the Fe2+
concentration was increased to 4 mM (5600 mg/kg soil), it is possible that Fe2+ was the

limiting reagent at lower concentrations.

The concentration of F e2+ and hydrogen peroxide to cause complete destruction of the
parent compound seems to depend on its nature and initial concentration, other
competing organic compounds that may be present, and presence of metals, which may
lead to additional reactions such as complexation and precipitation. It is thus imperative

that a bench-scale study be conducted on a site before this technology as actually applied.

5.3.2 Effect of Fez+lH2O2 Ratio
As discussed above, the degradation of MTBE increased when the concentration of F e2+
in the reaction mixture was increased. This may be due to the fact that when the

concentration of Fe” in the reaction mixture is low, Fe2+ is the limiting reagent in the

55

reaction between H202 and Fe2+ (Equation 2-1), which limits the amount of OH

produced for the reaction.

H202+Fe” _. Fe”+H0‘+ '0H (21)

At higher concentrations of Fez: i.e., when the Fey/H202 ratio was greater than 1.3, a
decrease in the degradation of MTBE was observed. Although more 'OH could be
generated due to a higher concentration of Fe2+ present, °OH may have been scavenged
by another Fe2+ due to the reaction between Fe2+ and the hydroxyl radical (Xu et al.,

20041

 

~0H + Fe” > Fe” + H0' (51)

Additionally, if a large amount of OH is generated as a result of excess Fe“, the
oxidation of MTBE may be suppressed due to the reaction of OH with H202, or by the
combination of two OH groups to form H202 as is depicted in the reactions 5-2 and 5-3

(Xu et a1, 2004).

'0H + H202 ——§ H2O + HO2° (5-2)

OH + OH ———> H202 (5-3)

56

5.3.3 Two Stages of MTBE Degradation

It is believed that the degradation of MTBE using Fenton’s Reagent is a two-stage
reaction (Xu et al., 2004). During the first stage of the reaction, Fe2+ reacts rapidly with
H202 to produce a large amount of OH. The generated 'OH reacts with MTBE, which,
as a result, is degraded very quickly during this first stage. This is known as the FezI/
H202 stage. During this first stage, the F e3+ produced can react with H202 to generate

Fe2+ and HO2° as seen in Reactions 2-4 through 2-8 (Xu et al., 2004)

The rate constant for the reaction between H202 and Fe2+ is 53 Lmole'ls'1 (Barb et al.,
1950), while that of the reaction between H202 and Fe” is 0.02 Lmole’ls'l (Sun et al.,
1993). It can be deduced that, unless the concentration of Fe” is approximately three
orders of magnitude greater than that of Fey, the reaction between Fe2+ and H202 is
much faster than that between F e3+ and H202, resulting in a greater rate of 'OH formation
and hence rapid degradation of the organic contaminant during the first stage of the
reaction. The ratio of Fe3+:Fe2+ would depend on the pH and the redox potential of the

system.

The second stage, which is the Fey/H202 stage, is much slower (Barb et al., 1951; Xu et
al., 2004) due to the fact that the decomposition of hydrogen peroxide by ferric ions is
much slower than that by ferrous ions (Barb et al., 1951). This would also agree with the
findings from the previous chapter, where only approximately 15% oxidation of MTBE
was observed when no Fe2+ was added to the reaction mixture, other reaction conditions

remaining unchanged.

57

 

5.3.4 pH Monitoring for Soils Treated with Fenton’s Reagent

The pH of the reaction mixture for each of the samples was measured before and after the

reaction. As mentioned before, the initial pH of the reaction mixture for each sample was

maintained between 2 and 3. It was observed that the final pH of the reaction mixture was

between 7 and 8 for samples where the concentration of Fe2+ added was between 0.1 mM

(140 mg/kg soil) and 10 mM (14,000 mg/kg soil). The pH decreased gradle as the

concentration of Fe2+ was increased. The final pH of the reaction mixture for a Fe2+

concentration of 50 mM (70,000 mg/kg soil) was between 4 and 5. Please refer Table 5-2

for pH vs. concentration of Fe2+ table. The controls contained 0.2g of contaminated soil

and acidified water, and their initial pH was maintained between 2 and 3. The final pH of

the controls, which did not have any Fenton’s Reagent added to them, was approximately

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

7.
Table 5-2. Variation of pH with concentration of Fe“
H202 H202 Fe” Fe”
Concentration Concentration Concentration Concentration
(mg/L) (mflg soil) (mM) (mgfig soil) Final pH

510 12750 0.1 140 Between 7-8
510 12750 0.5 700 Between 7-8
510 12750 1 1400 Between 7-8
510 12750 2 2800 Between 7-8
510 12750 4 5600 Between 7-8
510 12750 6 8400 Between 7-8
510 127 50 8 1 1200 Between 7-8
510 12750 10 14000 Between 7-8
510 12750 15 21000 Between 5-6
510 12750 20 28000 Between 5-6
510 12750 30 42000 Between 4-5
510 12750 40 56000 Between 4-5
510 12750 50 70000 Between 4-5

 

 

 

 

 

 

58

 

When the concentration of the hydroxyl radicals is low, the pH of the soil increases. This
can be explained based on the fact that the soil used in this study is sandy soil (78.9%),
which has a pH of 5.43 i 0.87. Sandy soil is a poor buffer, and hence does not resist
changes in pH. Another factor that could lead to an increase in the pH at low

concentrations of Fe2+ is the release of HO" when Fe2+ reacts with hydrogen peroxide

(Reaction 2-1).

H202 + Fe” ——> Fe” + H0 + ~0H (2-1)

The fact that the pH of the controls was approximately 7 further supports this theory.

In the presence of excess Fe“, a large amount of Fe3+ is produced as is seen in reaction

2-1. The generated F e3+ can react with H202 to form H+, which would lower the pH of the

system (Reaction 2-4).

H202 + F63+ _—"' F62+ + H+ + HOz' (2‘4)

Pignatello, 1992, also observed that within seconds of addition of peroxide, the initial pH

of 4.5 and 6.0 dropped to 3.2-3.4, which may have been a result of release of protons

from hydrolysis of the Fe3+ product.

59

5.4 Formation of Byproducts

It was observed that when Fe2+ was added to the reaction system, there was no formation
of higher molecular weight compounds such as those seen without the addition of Fe“. It
was also seen that the byproducts formed when Fe2+ was added were different fi'om the
case when it was not. This has been shown in the Figures 5-2 and 5-3. The byproducts
based on the retention time were tentatively identified as formaldehyde, propionaldehyde

and acetone.

5.10

5.08

5.06

5.04

 

 

5.02

 

 

5.00

Response (mV)

4.98

4.96

 

 

4.94

 

IIIIIIIIIIIIIIIIerTITTTTIIIIIIIIIIIIII

0.5 1.0 1.5 2.0 2.5 3.0 3.5

Time (minutes)

Figure 5-2. [Fe2+] = 4 mM (5600 mg/kg soil), pH = 2-3, [MTBE]o = 388.75 mg/kg

soil, [H202] = 510 mg/L (12,750 mg/kg soil)

60

It can be seen that at times around 1.7 minutes, the baseline does not come down, as it

should. This suggests the presence of a compound that elutes slowly from the column.

5.08 {I

5.06

5.04
5.02

 

5.00

4.98

 

4.96

Response (mV)

 

4.94

 

IIIIIIIIIIIIIIIIIIIIIIII‘FIIIIIIIIIIIIIT

0.5 1.0 1.5 2.0 2.5 3.0 3.5

Time (minutes)

Figure 53. [Fe”] = 6 mM (8400 mg/kg soil), pH = 2-3, [MTBE]o = 388.75 mg/kg

soil, [H202] = 510 mg/L (12,750 mg/kg soil)

61

Due to the different byproducts obtained in cases where Fe2+ was and was not added to
the reaction mechanism, it is reasonable to assume that different reactions occur during
both cases. This could be due to the formation of different intermediates or could be
different stages of oxidation since the rate of the reaction between the contaminant and

the hydroxyl radical is faster in the presence of Fe”.

Tables 5-3, 5-4, and 5-5 depict the percent conversion of MTBE to formaldehyde,
propionaldehyde, and acetone on a molar basis. However, as can be seen from Table 5-6,
the total conversion does not add to 100%. Thus, a more complete study of the
byproducts needs to be done to identify and quantify all of the byproducts and

intermediates.

Table 5-3. Percent conversion of MTBE to formaldehyde

 

 

 

 

 

 

 

 

 

 

Fe” Fe” Initial MTBE
Concentration Concentration Concentration Formaldehyde % Conversion to
LmM) Msoil) (mmoles) formed (mmoles) formaldehyde
4 5600 8.82x10'4 6.33x10'5 7.2
15 21000 8.82x10'4 4.3lx104 48.9
30 42000 8.82x10“ 6.93x10'S 7.8

 

 

62

Table 5-4. Percent conversion of MTBE to propionaldehyde

 

 

 

 

 

 

 

 

 

 

Fe” Fe” Initial MTBE
Concentration Concentration concentration Propionaldehyde % Conversion to
(mM) (mm soil) (mmoles) formed (mmoles) propionaldehyde
4 5600 8.82x104 2.52x10'5 2.8
15 21000 8.82x10" 8.58x10‘6 0.9
30 42000 8.82x104 1.07x10'5 1.2

 

Table 5-5. Percent conversion of MTBE to acetone

 

 

 

 

 

 

 

 

 

 

 

Fe” Fe” Initial MTBE
Concentration Concentration concentration Acetone formed % Conversion to
(mM) (mm soil) (mmoles) (mmoles) acetone
4 5600 8.821110“ 5.521110“5 0.6
15 21000 8.821110“ 4.171105 4.7
30 42000 8.821110“ 1.681105 1.9

 

 

Table 5-6. Total conversion

 

 

 

 

 

 

 

 

 

Initial MTBE
Fe2+ Concentration Fe2+ Concentration concentration Total conversion to
(mM) (mg/3g soil) (mmoles) byproducts (%)
4 5600 8.821110“ 10.6
15 21000 8.82x10“ 54.5
30 42000 8.82x10“1 10.9

 

 

63

However, due to the presence of byproducts, it is imperative that these compounds are
identified and a complete understanding of their toxicity effects, environmental fate and
transport is made before this treatment is applied to contaminated sites. It is also
important to conduct a bench-scale study before actually applying this technology to any
contaminated site due to variation among different sites due to factors such as organic

content, type of soil, presence of other contaminants, and presence of metals.

64

CHAPTER 6

CONCLUSIONS AND SCOPE FOR FUTURE WORK

Form this study, it was seen that the effective degradation of MTBE using Fenton’s
Reagent is possible, with and without the addition of Fe”. MTBE was almost completely
oxidized under conditions of excess peroxide when no Fe2+ was added to the reaction
mixture. When Fe2+ was added, ahnost complete oxidation of MTBE could be achieved

even at low concentrations of hydrogen peroxide.

In the presence of mineral iron, at pH values between 2 and 3, and reaction time of 120
minutes, MTBE was effectively degraded under conditions of excess hydrogen peroxide.
Greater than 95% of MTBE destruction was achieved when the hydrogen peroxide added
was equal to or greater than 30,000 mg/L (750,000 mg/kg soil). The initial MTBE

concentration in the soil was between 299.25 and 429.75 mg MTBE/kg soil.

Although the degradation of MTBE was almost complete under these experimental
conditions and soil type, higher molecular weight byproducts were formed at high
concentrations of hydrogen peroxide. In addition, it was also observed that lower
molecular weight byproducts were formed at nearly all concentrations of hydrogen
peroxide added. Based on the retention times of the compounds, the lower molecular

weight compounds were tentatively identified as propionaldehyde and acetone.

65

When F e2+ was added to the reaction mixture, nearly complete degradation of MTBE was
obtained when the F e2+ concentration was between 4 and 20 mM (i.e., 5600 and 28,000
mg/kg soil). The reaction had been conducted at pH values between 2 and 3, initial
hydrogen peroxide concentration of 5 10 mg/L (12,750 mg/kg soil), and a reaction time of
120 minutes. It was also observed that there was no formation of higher molecular weight
compounds as was seen in the case where no Fe2+ was added to the reaction mixture.
Based on the retention times of the compounds, these byproducts were tentatively

identified as formaldehyde, propionaldehyde, and acetone.

It is recommended that additional work be done for complete identification and
quantification of all the byproducts. and intermediates formed during the reaction of
MTBE with Fenton’s Reagent (with and without the addition of Fe“). It is imperative
that a thorough analysis of these byproducts, which would include their toxicity, fate and
transport in the environment, be conducted before this technology is recommended for

use at a contaminated site.

Since nearly complete degradation of MTBE is possible in soil without the addition of
external iron, this technology could prove very beneficial provided the byproducts

formed are less toxic than the parent compound.
Additionally, the formation of intermediates and byproducts of the reaction should be

evaluated at different times so as to develop a complete reaction mechanism for MTBE

degradation.

66

The degradation of MTBE using similar reaction conditions as those employed in this
study, but different soil types can also be conducted, which would help us understand the

effect of different parameters on reaction rates and the byproduct formation.

It is recommended that for a contaminated site, a bench-scale experiment be conducted so
as to optimize reaction conditions such as pH, Fe2+ and H202 dose, and reaction time,
since these parameters will vary with each site. Additionally, it is imperative that
effective mixing of all reagents is ensured at the site so that the contaminants can react
with the hydroxyl radicals generated. The concentration of the hydrogen peroxide should

be sufficiently low so that it does not cause any explosive hazard at the site.

67

REFERENCES

AMRClearinghouse.org, 2004. Abandoned Mine Reclamation Clearinghouse,
http://www.amrclearinghouse.org/Sub/AMDbasics/LZFerrous-FerricIron.htm

Anderson, M. A., 2000. Removal of MTBE and Other Organic Contaminants from Water
by Sorption to High Silica Zeolites.Environ. Sci. Technol. 34 (4), 725-727.

Barb, W.G., Baxendale, J.H., George, P., Hargrave, KR, 1950. Reactions of Ferrous and
Ferric Ions with Hydrogen Peroxide. Part I-The Ferrous Ion Reaction. Trans. Faraday
Soc. 47, 462-500.

Barb, W.G., Baxendale, J.H., George, P., Hargrave, KR, 1951. Reactions of Ferrous and
Ferric Ions with Hydrogen Peroxide. Part H-The Ferric Ion Reaction. Trans. Faraday Soc.
47, 591-616.

Burbano, A., Dionysiou, D., Suidan, M., Richardson, T., 2003. Chemical Destruction of
MTBE using Fenton’s Reagent: Effect of Ferrous Iron/ Hydrogen Peroxide ratio. Water
Sci. Technol. 47 (9), 165-171.

Callender, T., Davis, LC, 2001. Environmental Behavior of Methyl tert-Butyl Ether: A
study of Henry’s Law Constant and the Dispersion of MTBE through River Bottom Sand
and Soil. Proceedings of the 2001 Conference on Environmental Research. May 21-24,
2001, Manhattan, KS. http://www.engg.ksu.edu/HSRC/01Proceed/docs/ l 37.pdf

Douglas, R.M., John, 03., 2002. MTBE in Groundwater: Physical Properties and
Regulatory Outlook. Gradient: Trends in Environmental Science and Technology Hydro
Geo Chem Inc. Tucson, AZ. http://www.hgcinc.com/fall99 news/mtbehtml.

Guillard, C., Charton, N., Pichat, P., 2003. Degradation Mechanism of t-Butyl Methyl
Ether (MTBE) in Atmospheric Droplets. Chemosphere. 53, 469-477.

Howard, RH, 1993. Handbook of Environmental Fate and Exposure Data for Organic
Chemicals. Volume IV. Solvents 2. 71-76. Lewis Publishers, Chelsea, Michigan.

Jacobs, J., Guertin J ., and Herron C., 2001. MTBE: Effects on Soil and Groundwater
Resources. Lewis Publishers. Boca Raton, Florida.

Kealey, J ., Roth, R., Pezzullo, J ., 2001. In-Situ Remediation of MTBE using OxyVac
Technology, Terra Vac Corporation. The Annual International Conference on Soils,
Sediments and Water, October 22-24, 2001. University of Massachusetts, Amherst, MA.

Kinner, N. E., 2001. Fate, Transport, and Remediation of MTBE. Testimony before the

United States Senate Committee on Environment and Public Works, April 23, 2001.
Salem, NH.

68

Michigan Department of Agriculture, 2002. Motor Fuels Quality Program. MTBE
Report.

Montana Department of Environmental Quality, 2003. Final Guidance Document:
Montana Tier 1 Risk-Based Corrective Action Guidance Appendix E: Volatilization
Factors, http://www.deq.state.mt.us/rem/hwc/rbca/NewRBCA1 1-2003/AppE/Apn E-
01volfac l 0-03.pdf. October 2003

MTBE Treatment Profiles, 2002. In Situ Chemical Oxidation, Soil Vapor Extraction at
Site MA — 1, Concord, MA.
http://cluin.org/products/mtbe/usersearch/mtbe_details.cfm?SiteID=360

MTBE Treatment Profiles, 2003. Excavation, Soil Vapor Extraction at North Tahoe
P.U.D. Maintenance (National Avenue), Lake Tahoe, CA. ht_tp://clu-
in.org/products/mtbe/usersubmit/mtbe update details.cfin?SiteID=180.

MTBE: Drinking Water Regulations and Monitoring Results, 2004. California
Department of Health Services.
http://www.dhs.cggov/ps/ddwem/chemicals/MTBE/mtbeindex.htrn

Pignatello, J.J., 1992. Dark and Photoassisted Fe3+ - Catalyzed Degradation of
Chlorophenoxy Herbicides by Hydrogen Peroxide. Environ. Sci. Technol. 26, 944-951.

Ray, A., Selvakumar, A., 2000. Treatment of MTBE using Fenton’s Reagent. US.
Environmental Protection Agency. National Risk Management Research Laboratory.
Water Supply and Water Resource Division. Urban Watershed Management Branch.
Edison, New Jersey. EPA/600/JA-00/ 193.

http://www.epa. gov/ORD/NRMRL/Pubs/600J A03 1 17/600j a031 l7.pdf

Stefan, M. I, Mack, J., Bolton, J. R., 2000. Degradation Pathways during the Treatment
of Methyl tert-Butyl Ether by the UV/H2O2 Process. Environ. Sci. Technol. 650-65 8.

Storage Tank Information Database, 2004. Michigan Department of Environmental
Quality.

Sun, Y., Pignatello, J .J ., 1993. Photochemical Reactions Involved in the Total
Mineralization of 2,4-D by F $711202va . Environ. Sci. Technol. 304-310.

Tang, W.Z., Huang, CR, 1996. 2,4-Dichlorophenol Oxidation Kinetics by Fenton’s
Reagent. Environ. Techno]. 17,1371-1378.

United States Environmental Protection Agency, 1997. Drinking Water Advisory:
Consumer Acceptability Advice and Health Effects Analysis on Methyl Tertiary-Butyl
Ether (MTBE), EPA-822-F-97-009. December 1997.

http://www.epa.gov/waterscience/drinking/mtbefact.pdf

69

United States Environmental Protection Agency, 1998. MTBE Fact Sheet # 2.
Remediation of MTBE Contaminated Soil and Groundwater. EPA 510-F-97-015, January
1998, http://www.epa.gov/swerust1/mtbe/mtbefs2.pdf

US. Environmental Protection Agency, 2003. Methyl Tertiary Butyl Ether (MTBE).
Gasoline. September 2003. http:// www.cpa,gov/mtbe/gas.htm.

US. Environmental Protection Agency, 2003. Methyl Tertiary Butyl Ether (MTBE),
Drinking Water. September 2003. http://www.ep§gov/mtbe/water.htm.

US. Environmental Protection Agency, Method 556, Determination of Carbonyl
Compounds in Drinking Water by Pentafluorobenzylhydroxylarnine Derivatization and
Capillary Gas Chromatography with Electron Capture Detection. Revision 1.0. June
1998. http://www.epa.gov/safewater/methods/metS56.pdf

Walling, C., 1975. Fenton’s Reagent Revisited. Accounts of Chemical Research. 8, 125-
13].

Xu, X., Zhao, Z., Li, X., Gu, J., 2004. Chemical Oxidative Degradation of Methyl tert-
buty] ether in Aqueous Solution by Fenton’s Reagent/ Chemosphere. 55, 73-79.

Yeh, C. K., Novak, IT, 1995. The effect of hydrogen peroxide on the degradation of
methyl and ethyl tert-butyl ether in soils. Water Environ. Res. 67, 828-834.

70

             

N STATE U

293 0

A

3 1

 

 

 

iliiiillililll‘lliil
2504 531

5