Chemosphere 49 (2002) 413–420

www.elsevier.com/locate/chemosphere

Kinetics of heat-assisted persulfate oxidation of

methyl tert-butyl ether (MTBE)
Kun-Chang Huang *, Richard A. Couttenye, George E. Hoag
Environmental Research Institute, University of Connecticut, 270 Middle Turnpike, U-5210 Storrs, CT 06269, USA
Received 1 August 2001; received in revised form 10 June 2002; accepted 26 June 2002

Abstract

The kinetics of heat-assisted persulfate oxidation of methyl tert-butyl ether (MTBE) in aqueous solutions at various
pH, temperature, oxidant concentration and ionic strength levels was studied. The MTBE degradation was found to
follow a pseudo-first-order decay model. The pseudo-first-order rate constants of MTBE degradation by persulfate
(31.5 mM) at pH 7.0 and ionic strength 0.11 M are
0:13  104 , 0:48  104 , 2:4  104 and 5:8  104 s1 at 20, 30, 40
and 50 °C, respectively. Under the above reaction conditions, the reaction has an activation energy of 24:5  1:6 kcal/
mol and is influenced by temperature, oxidant concentration, pH and ionic strength. Raising the reaction temperature
and persulfate concentration may significantly accelerate the MTBE degradation. However, increasing both pH (over
the range of 2.5–11) and ionic strength (over the range of 0.11–0.53 M) will decrease the reaction rate. Reaction in-
termediates including tert-butyl formate, tert-butyl alcohol, acetone and methyl acetate were observed. These inter-
mediate compounds were also degraded by persulfate under the experimental conditions. Additionally, MTBE
degradation by persulfate in a groundwater was much slower than in phosphate-buffer solutions, most likely due to the
presence of bicarbonate ions (radical scavengers) in the groundwater.
Ó 2002 Elsevier Science Ltd. All rights reserved.

Keywords: MTBE; Methyl tert-butyl ether; TBF; TBA; Persulfate; Oxidation; Kinetics

1. Introduction and Diseases Registry (ATSDR) also states that drink-

Methyl tert-butyl ether (MTBE), a widely used gas-
oline additive, is found in numerous groundwater and
surface water reservoirs across the United States (Clarke
and Knowles, 1982; Leahy and Thompson, 1994). Due
to its unpleasant taste and odor even at low levels (e.g.,
15 lg/l) (Stocking et al., 2001), the US EPA established a
draft lifetime health advisory limit of 20–40 lg/l for
MTBE in drinking water in 1997 (US EPA, 1997) and
has classified MTBE as a possible human carcinogen
(Church et al., 1999). The Agency for Toxic Substances
*
Corresponding author. Tel.: +1-860-486-5893; fax: +1-860-
486-5488.
E-mail address: khuang@eri.uconn.edu (K.-C. Huang).
0045-6535/02/$ - see front matter Ó 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 4 5 - 6 5 3 5 ( 0 2 ) 0 0 3 3 0 - 2
ing or breathing MTBE may cause nausea, nose and
throat irritation and nervous system effects (ATSDR,
1996). Consequently, MTBE has imposed a serious risk
to public health and to the environment.
Because MTBE has a vapor pressure of 245 mm
Hg at 25 °C (US EPA, 1993), in the spilled gasoline
that migrates downward toward the groundwater table,
MTBE may evaporate and form a vapor phase plume in
soil. MTBE can transfer easily from gasoline to air and
then to water or from gasoline directly to water. With a
high aqueous solubility of 51.26 g/l at 25 °C (US EPA,
1993) and a relatively low Henry’s law constant of
0.0555 (dimensionless) at 25 °C (Bierwagen and Keller,
2001), MTBE in a gaseous plume tends to partition into
the aqueous phase, where it may migrate at the same
rate as groundwater. Although a study reported that the
414 K.-C. Huang et al. / Chemosphere 49 (2002) 413–420

microorganisms indigenous to stream-bed sediments (TOC) in water and wastewater (Clesceri et al., 1998).
under mixed aerobic and anaerobic conditions miner- Oxidation of large biological molecules such as proteins
alized MTBE by 73% and tert-butyl alcohol by 84% in and monoclonal antibodies proceeds rapidly while other
groundwater (Bradley et al., 1999), the biodegradation compounds such as n-butanol, 2-propanol, acetic acid,
of MTBE in the subsurface is usually very slow (Yeh acetonitrile, tartaric acid etc. are relatively more difficult
and Novak, 1994). An effective treatment process for the to oxidize (Clesceri et al., 1998). However, increasing the
remediation of MTBE contaminated soil and ground- temperature or oxidant concentration, which could in-
water is still being sought. crease the production rate of free radicals, can generally
In situ chemical oxidation is a promising remedial enhance the reaction rate.
alternative for the destruction of many contaminants in In aqueous solutions, the standard oxidation–reduc-
soil and groundwater. The oxidants, widely applied in tion potential (E0 ) for the half-cell persulfate reaction is
this process, include hydrogen peroxide (H2 O2 ), per-
2.01 V (Eq. (1)), which is comparable to ozone (E0 ¼
manganate (MnO 4 ) and ozone (O3 ). While MTBE is 2:07 V), hydrogen peroxide (E0 ¼ 1:78 V) and perman-
known to be persistent to aerobic and anaerobic bio- ganate (with E0 ¼ 1:70 V) (Eqs. (2)–(4)):
degradation (Suflita and Mormile, 1993), it can be
completely mineralized through advanced oxidation S2 O2  2
E0 ¼ 2:01 V
8 þ 2e ! 2SO4 ð1Þ
processes such as ozone oxidation (Leitner et al., 1994;
þ  0
Kang and Hoffman, 1998; Mitani et al., 2002), H2 O2 /UV O3ðgÞ þ 2H þ 2e ¼ O2ðgÞ þ H2 O E ¼ 2:07 V ð2Þ
oxidation (Wallington et al., 1988; Cater et al., 2000;
Chang and Young, 2000; Stefan et al., 2000) and photo- H2 O2 þ 2Hþ þ 2e ¼ 2H2 O E0 ¼ 1:78 V ð3Þ
catalytic TiO2 oxidation (Barreto et al., 1995). These MnO þ 
þ 4H þ 3e ! MnO2ðsÞ þ 2H2 O 0
E ¼ 1:70 V
4
early studies indicate that MTBE can be degraded with
radical-driven oxidation processes. ð4Þ
In this study, sodium persulfate (N2 S2 O8 ), a radical-
based oxidant, showed an excellent capability of degrad- However, persulfate oxidation at atmospheric tempera-
ing MTBE. The kinetics and several factors influencing ture is usually not effective. Persulfate is commonly used
the MTBE degradation with heat-assisted persulfate with UV light or under high temperature in order to
oxidation were studied. Kinetic parameters including initiate/enhance its radical oxidation mechanisms. Sul-
reaction order, rate constant and activation energy (Ea ) fate radicals, formed from photolytic or heat decom-
were determined, and the influences of temperature ðT Þ, position of persulfate (Eq. (5)), may initiate a series of
oxidant concentration, pH and ionic strength ðIÞ on radical chain reactions (Eqs. (5)–(16)) (Berlin, 1986),
MTBE degradation were investigated. Additionally, re- where organic compounds (i.e., M in Eq. (6)) are usually
action intermediates and their fate under heated per- degraded.
sulfate solutions were partially studied.
heat=hv
S2 O2 ƒ! 2SO
8 ƒƒƒ 4 ð5Þ
2. Persulfate chemistry S2 O2
þM! þM 2SO 
8 4

The reactions of persulfate (also known as peroxy- ðM represents organic compoundsÞ ð6Þ
disulfate and peroxodisulfate) ions with various organic SO þ H2 O ! HO þ 
HSO ð7Þ
4 4
and inorganic compounds have been extensively studied
(Kolthoff and Miller, 1951; House, 1962; Berlin, 1986). SO
4

þ M ! M þ products ð8Þ
Persulfate oxidation is generally conducted under heat-, HO þ M ! M þ products ð9Þ
photo- or metal-catalyzed conditions because the oxi-
dation rates can be greatly accelerated. High reactive 
M þ S2 O2
8 ! SO
4 þ products ð10Þ
species such as sulfate radicals (SO 4 ) and hydroxyl
radicals (HO ) are generated as a result of photolysis or SO 
4 þ HO ! chain termination

heat decomposition of persulfate ions in aqueous phases ði:e:; terminating free radicalsÞ ð11Þ
(Wilmarth and Haim, 1962; Nosov, 1966; Dogliotti and
SO 
4 þ M ! chain termination ð12Þ
Hayon, 1967; Hayon and McGarvey, 1967; Berlin, 1986;
Tanner and Osman, 1987). Persulfate ions in the pres- 2SO
4 ! chain termination ð13Þ
ence of ultraviolet (UV) light, heat and some metal  
catalysts are able to oxidize many organic substances HO þ M ! chain termination ð14Þ
into carbon dioxide. This has led to the use of the UV- 
2HO ! chain termination ð15Þ
persulfate or heated-persulfate oxidation as a standard
method for the determination of total organic carbon 2M ! chain termination ð16Þ
 K.-C. Huang et al. / Chemosphere 49 (2002) 413–420
Since hydrogen peroxide and ozone have relatively
short lifetimes in the subsurface as compared to
persulfate, persulfate is studied as a potential alternate
oxidant for the remediation of groundwater contami-
nated with MTBE and other toxic organic contami-
nants.
3. Experimental protocols
3.1. Materials
Sodium persulfate (98þ% purity, ACROS, New
Jersey) solutions of the required concentrations (i.e.,
7,
15, 32, 43 and 52 mM) were prepared with phosphate-
buffered solutions. The buffer solution used to control
pH was prepared by mixing 1.85 g/l of dibasic sodium
phosphate (Na2 HPO4 , Reagent grade, Fisher Scientific,
Pittsburgh, PA) and 1.52 g/l of monobasic sodium
phosphate (NaH2 PO4 , Reagent grade, J.T. Baker, Phil-
lipsburg, NJ) in deionized water (DI). The pH values of
all solutions were adjusted, whenever required, with 1 N
sodium hydroxide (NaOH, ACS grade, Fisher Scientific)
and perchloric acid (HClO4 , 69–72%, J.T. Baker). So-
dium chloride (NaCl, 99.9% purity, Enzyme grade,
Fisher Scientific) was used as ionic strength adjuster to
the phosphate-buffered solutions. MTBE and four de-
tected purgeable intermediates including tert-butyl for-
mate (TBF), tert-butyl alcohol (TBA), methyl acetate
(MA) and acetone (all with 99% purity, Chem Service,
West Chester, PA) were used as received. Other ana-
lytical reagents including ferrous ammonium sulfate
(FAS, Fe(NH4 )2 (SO4 )2 6H2 O, 99.4% purity, J.T. Baker),
ammonium thiocyanate (NH4 SCN, 99þ% purity, Fisher
Scientific) and sulfuric acid (H2 SO4 , 98% purity, Fisher
Scientific) were prepared with DI water to desired con-
centrations for persulfate measurements.
3.2. Analysis
Analysis of MTBE and four observed intermedi-
ates was conducted using a gas chromatograph (HP
GC-5890 Series II) equipped with a capillary column
ð0:25 mm  60 m, J&W DB-624), a flame ionization
detector, a purge & trap (Tekmar ALS 2016) and a
concentrator (Tekmar LSC 2000). Initial column tem-
perature preceding sample injection was at 35 °C. This
temperature was held for 2 min after the sample injec-
tion to the gas chromatograph. Subsequently, a pro-
grammed temperature increase of 20 °C/min until 180 °C
was made. With a column head pressure of 138 kPa, the
retention times (RT) of acetone, MA, TBA, MTBE,
TBF were 4.3, 4.8, 5.2, 5.4, and 7.3 min, respectively.
Calibration curves of five levels (i.e.,
5, 10, 30, 50 and
100 lg/l) and fluorobenzene (used as the internal stan-
415
dard with RT 8.2 min) were used for quantification in
the analysis.
In order to determine the Na2 S2 O8 concentration,
samples were prepared by placing 0.1 ml of reaction
solution in a 20-ml test tube. Subsequently, 0.9 ml of DI
water, 10 ml of 2.5 N H2 SO4 solution and 0.1 ml of 0.4
N FAS solution were added. The contents were mixed
and allowed to react for 40 min. Finally, 0.2 ml of 0.6 N
NH4 SCN solution was added, and the absorbance was
read with a spectrophotometer (Milton Roy Spectronic
601) at a wavelength of 450 nm. The calibration curve
established by the procedures described above using
Na2 S2 O8 solutions ranging from 50 to 1800 mg/l showed
a high linear correlation coefficient (i.e., R2 ¼ 0:993).
3.3. Kinetics experiments
The kinetics of heat-assisted persulfate oxidation of
MTBE was studied under pseudo-order conditions by
using a Na2 S2 O8 solution of 31.5 mM (
8 g/l, a con-
centration to be used in a pilot test) to oxidize MTBE in
the level of parts per million. Because TBF, TBA, MA,
acetone and other intermediate products compete with
MTBE for the oxidants formed in persulfate chemistry,
experiments were designed to collect, generally, at least
five data points within 1–2 half-lives of MTBE degra-
dation in order to minimize the influence of side reac-
tions. The fate of the four observed intermediates by
persulfate oxidation was also examined.
The experimental system and procedures developed
by Huang et al. (2001) were used in this study because
headspace free, isothermal and completely mixed con-
ditions can be achieved during the reactions. In general,
during the experiments the temperature controlled by an
incubator varied within 1 °C from its set point, and the
pH value maintained by phosphate buffers decreased by
up to 0.7. A typical experiment involved adding 93.5 ml
of phosphate-buffer solution and a Teflon-coated stir
bar into a 100-ml gas-tight glass syringe reactor (Fig. 1).
The plunger was inserted to displace the air in the sy-
ringe, and the reactor was then left overnight inside an
incubator set at a desired temperature. Subsequently, the
reactor was injected with 0.65 ll of MTBE (99%) and
the solution was mixed for 60 min on a magnetic stirrer
that was placed inside the incubator. A volume of 100 ll
was then collected from the reactor to determine the
initial MTBE concentration ([MTBE]0 ). Following the
initial sampling, the reaction was initiated by injecting
5 ml of Na2 S2 O8 solutions (7, 15, 32, 43 or 52 mM).
Samples were collected (generally more than six sample
points within a half-life time of MTBE) during the run
to monitor the MTBE degradation with time. Other
parameters such as pH and Na2 S2 O8 concentration were
measured in the beginning and at the end of the experi-
ment so that the change in total volume in the reactor
was less than 5%.
416 K.-C. Huang et al. / Chemosphere 49 (2002) 413–420

Fig. 1. The experimental system used for the kinetics study of heat-assisted persulfate oxidation of MTBE.

4. Results and discussion

4.1. Kinetic parameters

A typical degradation of MTBE with heat-assisted

persulfate oxidation against time is presented in Figs. 2
and 3, which show the MTBE degradation at different
temperatures and oxidant concentrations, respectively.
The results in Figs. 2 and 3 indicate that MTBE was
rapidly degraded with persulfate under the experimental
conditions. For example, in a persulfate solution of 31.5
mM (8 g/l) at 40 °C, the half-life of MTBE was less than
an hour (Table 1). In addition, it is evident from Figs. 2
and 3 that the reaction was significantly influenced
by temperature and oxidant concentration. The higher

Fig. 3. Effect of Na2 S2 O8 concentration on MTBE degradation

rate. Inset: plot of ln k1 vs ln [Na2 S2 O8 ]0 . [MTBE]0
0.06 mM;
pH
7; Temp: ¼ 40 °C.

the temperature and oxidant concentration, the faster

Fig. 2. Effect of temperature on the oxidation of MTBE by
persulfate. Inset: plot of ln k1 vs 1=T for Ea estimation using the
Arrhenius equation. [MTBE]0
0.06 mM; [Na2 S2 O8 ]0
31.5
mM; pH
7; I
0:06 M.
MTBE degraded. The obtained MTBE degradation data
were well fit (i.e., R2
1:0) with a pseudo-first-order
decay model in all runs albeit numerous oxidizing spe-
cies (e.g., SO   2 
4 , HO , M , S2 O8 and HSO4 ) could exist
in the system and react with MTBE. This indicates that
an oxidizing species (mostly likely the SO
4 under the
experimental conditions) might dominate the reaction
with MTBE. The early study by Dogliotti and Hayon
(1967) reported that sulfate radicals dominated in
the photolysis of persulfate ions in neutral and acidic
aqueous solutions. However, in alkaline solutions of
pH > 8:5, sulfate radicals were converted to hydroxyl
 K.-C. Huang et al. / Chemosphere 49 (2002) 413–420 417

Table 1
Rate constants of heat-assisted persulfate (½Na2 S2 O8 0 ¼ 31 mM) oxidation of MTBE in phosphate-buffered solutions under various
temperatures
[MTBE]0 (mM) Temperature (°C) pHinitial pHfinal k1  104 (s1 ) Half-lifea (h)
0.06 20 6.9 6.8 0.13 14.81
0.07/0.06b 30 6.9/6.9 6.8/6.7 0.48/0.59 4.01/3.26
0.06/0.07b 40 6.8/6.9 6.3/6.8 2.14/2.3 0.9/0.84
0.06 50 6.9 6.6 5.8 0.25
a
½ITotal ¼ ½Ipersulfate þ ½Ibuffer ¼ 0:06 M þ 0:05 M ¼ 0:11 M.
b
The experiment was conducted in duplicate.

radicals (Eq. (7)). In consequence, the rate law for the
degradation of MTBE by heat-assisted persulfate oxi-
dation is expressed as Eq. (17):
d½MTBE=dt ¼ k1 ½MTBE ð17Þ
where k1 ¼ f (T, [Na2 S2 O8 ], pH and I) is the pseudo-
first-order rate constant that represent a combined rate
of MTBE degradation by all oxidizing agents (e.g., SO 4 ,
HO , M and S2 O2 8 ) formed in the persulfate chemistry.
Sulfate radicals are likely to dominate the reaction with
MTBE under the experimental conditions used in this
study.
The results of a series of kinetics experiments under
various conditions (i.e., different temperature, oxidant
concentration, pH and ionic strength) are presented in
Tables 1–4. Pseudo-first-order rate constants (k1 ) of
MTBE degradation under pH
7.0, I
0.11 M and
[Na2 S2 O8 ]0
31.5 mM at 20, 30, 40, and 50 °C were
0:13  104 , 0:48  104 , 2:4  104 and 5:8  104 s1 ,
respectively (Table 1). It appears that at 40 °C the
MTBE degradation rate by persulfate oxidation is ap-
proximately an order of magnitude lower than those by
TiO2 /UV oxidation (i.e., 1:2  103 (Barreto et al.,
1995)) and H2 O2 /UV oxidation (2:3  103 (Wagler and
Malley, 1994); 5:8  103 (Stefan et al., 2000)).
The activation energy was determined based on the
data shown in Table 1 using the Arrhenius equation
ðk ¼ A expEa =RT Þ. ‘‘A’’ is the frequency factor, Ea is the
activation energy, R is the universal gas constant and T
is the temperature. Inset in Fig. 2 is a plot of ln k1 vs
1=T . In the temperature range of 20–50 °C, the Ea for the
reaction of MTBE with persulfate is 24:5  1:6 (kcal/
mol), indicating that an increase from 30 to 40 °C will
enhance the degradation rate of MTBE by a factor of

3.7.
Five concentrations (i.e.,
7, 15, 32, 43 and 52 mM)
of persulfate solutions were employed in the experiments
to investigate the influence of the oxidant concentration
on MTBE degradation. It is evident from Fig. 3 that
MTBE degraded faster at higher concentrations of
Na2 S2 O8 solutions. Additionally, Inset of Fig. 3, where
ln k1 is plotted against ln [Na2 S2 O8 ]0 shows a slope of

1.3, indicating that the degradation rate of MTBE was
directly proportion to initial Na2 S2 O8 concentration.
The rate constants and experimental conditions of this
set of experiments are presented in Table 2.
4.2. Degradation of reaction intermediates
The heat-assisted persulfate oxidation of MTBE in-
volves complex chain reactions and reaction scheme.
Identification of intermediates by gas chromatography-
mass spectrometry revealed that TBF, TBA, MA and
acetone were formed during the reactions. A represen-
tative graph showing the fate of these four purgeable
intermediate products in a persulfate solution at 40 °C is
presented in Fig. 4. It is evident from Fig. 4 that the four
detected intermediate products were also degraded by
persulfate chemistry under the experimental conditions.
Notice that TBA and TBF were greatly produced in the
early stage of the reactions and rapidly degraded as the
reaction proceeded. Acetone reached the highest con-
centration among the intermediates, indicating that it is

Table 2
Rate constants of heat-assisted persulfate oxidation of MTBE in phosphate-buffered solutions at 40 °C under various oxidant con-
centrations
[MTBE]0 (mM) [Na2 S2 O8 ]0 (mM) pHinitial pHfinal k1  104 (s1 ) Half-lifea (h)
0.07 7.1 7.0 6.9 0.38 5.07
0.05 14.7 7.0 6.8 0.87 2.21
0.06/0.07b 31.5 6.8/6.9 6.3/6.8 2.14/2.3 0.9/0.84
0.08 43.3 6.9 6.8 3.04 0.63
0.07 52.1 6.9 6.9 3.74 0.51
a
½ITotal ¼ 0:07–0:15 M.
b
The experiment was conducted in duplicate.
418 K.-C. Huang et al. / Chemosphere 49 (2002) 413–420

Table 3
Rate constants of heat-assisted persulfate oxidation of MTBE in phosphate-buffered solutions at 40 °C and under different pH values
[MTBE]0 (mM) pHinitial pHfinal k  104 (s1 ) Half-lifea (h) R2
0.06 2.5 2.4 3.05 0.63 0.99
0.07 5.0 4.3 2.55 0.75 1.0
0.06/0.07b 6.8/6.9 6.3/6.8 2.14/2.3 0.9/0.84 0.99/0.99
0.06 8.7 8.4 2.02 0.95 0.99
0.06 11.0 10.9 1.75 1.1 0.99
a
½Na2 S2 O8 0 ¼ 31:5 mM; ½ITotal ¼ ½Ipersulfate þ ½Ibuffer ¼ 0:06 M þ 0:05 M ¼ 0:11 M.
b
The experimental was conducted in duplicate.

Table 4
Rate constants of heat-assisted persulfate oxidation of MTBE in phosphate-buffered solutions (pH
7) at 40 °C under various ionic
strengths
[MTBE]0 (mM) [NaCl]0 (M) Itotal a M k  104 s1 Half-lifeb (h) R2
0.08 0.001 0.11 2.94 0.65 0.99
0.08 0.083 0.20 2.24 0.86 0.99
0.08 0.216 0.33 1.72 1.12 0.99
0.08 0.416 0.53 1.48 1.30 0.99
a
½ITotal of solution ¼ ½Ipersulfate þ ½Isodium chloride þ ½Ibuffer .
b
½Na2 S2 O8 0 ¼ 31:5 mM.

Scheme 1. Scheme of MTBE degradation with advanced oxi-

dation processes (adapted after Barreto et al., 1995).

and H2 O2 /UV (Stefan et al., 2000)). This implies that in

Fig. 4. Oxidation of MTBE oxidation intermediates by
Na2 S2 O8 . ½MTBE0
0:06 mM; [Na2 S2 O8 ]0
31.5 mM; pH

7; Temp: ¼ 40 °C.
not only more persistent toward radical oxidation but
also a possible oxidation product of TBF and TBA.
Similar degradation patterns of the four detected inter-
mediates were observed in the H2 O2 /UV–MTBE study
by Stefan et al. (2000). Moreover, comparison of initial
MTBE of 0.06 mM (i.e., 0.31 mM carbon) with the sum
of carbon (i.e., 0.31 mM and 0.33 mM at the reaction
time of 1 and 2 h, respectively) of TBA, TBF, acetone
and residual MTBE shows an agreeable mass balance,
indicating that TBA, TBF and acetone are the major
reaction intermediates.
The four detected intermediates were observed in two
previous studies that degraded MTBE with advanced
oxidation processes (i.e., TiO2 /UV (Barreto et al., 1995)
persulfate systems the degradation of MTBE is likely
mainly due to its reaction with sulfate and hydroxyl
radicals. It is generally accepted that the first step of
MTBE degradation is the abstraction of an alpha-
hydrogen by a hydroxyl radical (and/or a sulfate radical
in persulfate systems) (Kormann et al., 1988; Barreto
et al., 1995) (Scheme 1). The first step leads to a carbon-
centered radical which then reacts with oxygen to form a
peroxyl radical that may undergo acid-catalyzed hy-
drolysis reaction to yield TBF. It was reported that in
various aqueous systems TBF hydrolysis yielded stoi-
chiometric amounts of TBA (Church et al., 1999). In
addition, acetone and MA in addition to aldehydes and
carboxylic acids have been found as intermediate prod-
ucts prior to MTBE mineralization with H2 O2 /UV oxi-
dation (Stefan et al., 2000).
4.3. Reaction rate dependence of pH and ionic strength
Degradation of MTBE by persulfate (31.5 mM)
under various pH values (i.e., 2.5, 5, 7, 9 and 11) and
 K.-C. Huang et al. / Chemosphere 49 (2002) 413–420
ionic strength (i.e., 0.11 to 0.53 M) is shown in Tables 3
and 4, respectively. The rate of MTBE degradation with
persulfate is pH dependent (Table 3). The reaction rate
decreased with the increase in the pH value. The result
was expected because sulfate radicals and hydroxyl
radicals decay rapidly due to the reaction with hydroxyl
ions (Hayon and McGarvey, 1967). It was also in
agreement with the finding of Xu et al. (1989) that under
alkaline solutions the carbon dioxide formed from
complete MTBE oxidation could lead to the formation
of bicarbonate and carbonate ions that may inhibit
MTBE oxidation.
Table 4 shows the results and conditions of ionic
strength effect experiments. It is evident from Table 4
that the ionic strength effect on the oxidation of MTBE
with persulfate chemistry is similar to the pH effect.
The higher the ionic strength in solutions, the slower
MTBE degraded, most likely due to the decrease in
activity of persulfate and other reactive species as a re-
sult of high ionic concentration (electronic interference).
Moreover, a previous study reported that increasing
ionic strength decreased the persulfate decomposition
rate in a 0.1 M HClO4 solution and had no impact in a
0.1 M NaOH solution (Kolthoff and Miller, 1951). It is
likely that the rate of persulfate decomposition was re-
duced with the increase in ionic strength in neutral
solutions, which consequently decreased the MTBE de-
gradation rate.
4.4. MTBE degradation in groundwater by persulfate
oxidation
Replicate experiments were conducted to investigate
the oxidation of MTBE with persulfate (31.5 mM) at
40 °C in a groundwater sample that was spiked with
MTBE. The groundwater had a pH value of 8.2, an
alkalinity level of 314 mg/l as CaCO3 and TOC content
of 2.3 mg/l. The experimental results as shown in Fig. 5
indicate that MTBE was degraded during the experi-
ments. However, the degradation rate of MTBE in the
groundwater was much slower than that in phosphate-
buffered solutions, most likely because the bicarbonate
ions in the groundwater acted as radical scavengers that
inhibited the MTBE-radical reactions (Wagler and
Malley, 1994). A similar phenomenon was reported by
Xu et al. (1989) that the oxidation of several aromatic
compounds by ozonation was inhibited by carbonate
ions (500 mg/l) that reacted as hydroxyl radical scav-
engers. In our study, pH values in two groundwater runs
dropped from 8.2 to 6.5 within 4 h and to 3.2 at 30 h of
reaction, and these two runs had lower ionic strength in
solutions as compared to the experiments conducted in
phosphate-buffered solutions. Therefore, both the pH
and ionic strength were not considered as the reasons to
the observed phenomena.
419
Fig. 5. MTBE degradation by heat-assisted persulfate oxida-
tion in groundwater and phosphate buffered solution. [MTBE]0

0.06 mM; [Na2 S2 O8 ]0
31.5 mM; Temp: ¼ 40 °C.
5. Conclusions
This study investigated the degradation of MTBE
with heat-assisted persulfate oxidation, an old oxidation
process but a novel treatment concept that might be
applied for soil and groundwater remediation. The
MTBE oxidation kinetics and its affecting factors
including temperature, oxidant concentration, pH and
ionic strength were particularly examined. The MTBE
degradation was found to follow pseudo-first-order
kinetics. The results indicate that the reaction is sig-
nificantly influenced by temperature, oxidant concen-
tration, pH and ionic strength. Raising the reaction
temperature or the oxidant persulfate concentration can
increase the MTBE degradation rate while increasing
both pH (over the range of 2.5–11) and ionic strength
(over the range of 0.11–0.53 M) will lead to the opposite.
Additionally, intermediates including TBF, TBA, ace-
tone and MA were found to readily degrade by persul-
fate under the experimental conditions.
Although persulfate oxidation appears to have same
limitations (e.g., alkalinity inhibition) as Fenton’s che-
mistry and ozonation, persulfate is much more stable in
the subsurface as compared to the other two oxidants.
Attributed to its high solubility and stability under
normal subsurface conditions, persulfate could more
effectively transfer to the contaminated zones and react
with the contaminants. It is suggested to further study
and develop the persulfate oxidation technology for the
remediation of contaminated soil and groundwater.
Acknowledgements
This study was funded by the University Industrial
Affiliation Program of the Environmental Research In-
stitute at the University of Connecticut (UConn).
The authors acknowledge Marianela Hoz De Vila and
420 K.-C. Huang et al. / Chemosphere 49 (2002) 413–420
Josanlet Villegas of UConn Chemistry Department for
their valuable technical assistance in this research.
References
Agency for Toxic Substances and Diseases Registry (ATSDR),
1996. Toxicological profiles for 1methyl tert-butyl ether. US
Department of Health and Human Services, Atlanta, GA.
Barreto, R.D., Gray, K.A., Anders, K., 1995. Photocatalytic
degradation of methyl tert-butyl ether in TiO2 slurries: a
proposed reaction scheme. Water Res. 29, 1243–1248.
Berlin, A.A., 1986. Kinetics of radical-chain decomposition of
persulfate in aqueous solutions of organic compounds.
Kinet. Catal. 27, 34–39.
Bierwagen, B.G., Keller, A.A., 2001. Measurement of Henry’s
Law constant for methyl tert-butyl ether (MTBE) using
solid-phase microextraction. Environ. Toxicol. Chem. 20,
1625–1629.
Bradley, P.M., Landmeyer, J.E., Chapelle, F.H., 1999. Aerobic
mineralization of MTBE and tert-butyl alcohol by stream-
bed sediment microorganisms. Environ. Sci. Technol. 33,
1877–1879.
Cater, S.R., Stefan, M.I., Bolton, J.R., Safarzadeh-Amiri, A.,
2000. UV/H2 O2 treatment of methyl tert-butyl ether in
contaminated waters. Environ. Sci. Technol. 34, 659–662.
Chang, P.B.L., Young, T.M., 2000. Kinetics of methyl tert-
butyl ether degradation and by-products formation during
UV/H2 O2 water treatment. Water Res. 34, 2233–2240.
Church, C.D., Pankow, J.F., Tratnyek, P.G., 1999. Hydrolysis
of tert-butyl formate: kinetics, products, and implications
for the environmental impact of methyl tert-butyl ether.
Environ. Toxicol. Chem. 18, 2789–2796.
Clarke, N., Knowles, G., 1982. High purity water using H2 O2
and UV radiation. Effluent and Water Treat. J. Sept., 335–
341.
Clesceri, L.S., Greenberg, A.E., Eaton, A.D., 1998. Standard
methods for the examination of water and wastewater, 20th
ed. APHA, AWWA, and WEF.
Dogliotti, L., Hayon, E., 1967. Flash photolysis of persulfate
ions in aqueous solutions. Study of the sulfate and ozonide
radical anions. J. Phys. Chem. 71, 2511–2516.
Hayon, E., McGarvey, J.J., 1967. Flash photolysis in the

vacuum ultraviolet region of SO2 2
4 , CO3 and OH ions in
aqueous solutions. J. Phys. Chem. 71, 1472–1477.
House, D.A., 1962. Kinetics and mechanism of oxidation by
peroxydisulfate. Chem. Rev. 62, 185–200.
Huang, K.C., Hoag, G.E., Chheda, P., Woody, B.A., Dobbs,
G.M., 2001. Kinetics study of oxidation of chlorinated
ethenes with permanganate. J. Hazard. Mater. 87, 155–169.
Kang, J.W., Hoffman, M.R., 1998. Kinetics and mechanism of
the sonollytic destruction of methyl tert-butyl ether by
ultrasonic irradiation in the presence of ozone. Environ. Sci.
Technol. 32, 3194–3199.
Kolthoff, I.M., Miller, J.K., 1951. The chemistry of persulfate:
I. The kinetics and mechanism of the decomposition of the
persulfate ion in aqueous medium. J. Am. Chem. Soc. 73,
3055–3059.
Kormann, C., Bahnemann, D.W., Hoffman, M.R., 1988.
Photocatalytic production of H2 O2 and organic peroxides
in aqueous suspensions of TiO2 , ZnO and desert sand.
Environ. Sci. Technol. 22, 798–806.
Leahy, P.P., Thompson, T.H., 1994. US Geological survey
national water quality assessment program. USGS Open-
File Report 94-70, USGS, Reston, VA.
Leitner, K.V.N., Papailhou, A.L., Croue, J.P., Peyrot, J., 1994.
Oxidation of methyl tert-butyl ether (MTBE) and ethyl tert-
butyl ether (ETBE) by ozone and combined ozone/hydrogen
peroxide. Ozone: Sci. Eng. 16, 41–54.
Mitani, M.M., Keller, A.A., Bunton, C.A., Rinker, R.G.,
Sandall, O.C., 2002. Kinetics and products of reactions of
MTBE with ozone and ozone/hydrogen peroxide in water.
J. Hazard. Mater. 89, 197–212.
Nosov, E.F., 1966. Rate constant determination in the decom-
position of potassium and ammonium peroxydisulfate in
aqueous solution. Russ. J. Phys. Chem. 40, 1571–1572.
Stefan, M.I., Mack, J., Bolton, J.R., 2000. Degradation
pathways during the treatment of methyl tert-butyl ether
by the UV/H2 O2 process. Environ. Sci. Technol. 34, 650–
658.
Stocking, A.J., Suffet, I.H., McGuire, M.J., Kavanaugh, M.C.,
2001. Implications of an MTBE odor study for setting
drinking water standards. Am. Water Works Assoc. 93, 95–
105.
Suflita, J.M., Mormile, M.R., 1993. Anaerobic biodegradation
of known and potential gasoline oxygenates in the terrestrial
subsurface. Environ. Sci. Technol. 27, 976–978.
Tanner, D.D., Osman, S.A.A., 1987. Oxidative decarbonation
on the mechanism of potassium persulfate promoted decar-
bonation reaction. J. Org. Chem. 52, 4689–4693.
US EPA, 1993. Technical Information Review. Methyl tert-
butyl ether. Office of Pollution Prevention and Toxics,
Washington, D.C.
US EPA, 1997. Consumer acceptability advice and health
effects analysis on methyl tert-butyl ether. Office of Water,
Washington, D.C.
Wallington, T.J., Dagaut, P., Liu, R., Kurylo, M.J., 1988. Gas-
phase reactions of hydroxyl radicals of hydroxyl radicals
with the fuel additives methyl tert-butyl ether and tert-butyl
alcohol over the temperature range 240–440 K. Environ.
Sci. Technol. 22, 842–844.
Wagler, J.L., Malley, J.P., 1994. The removal of methyl
tertiary-butyl ether from a model ground water using UV/
peroxide oxidation. Am. Water Works Assoc. 108, 236–261.
Wilmarth, W.K., Haim, A., 1962. Peroxide reaction mecha-
nisms. In: Edwards, J.O. (Ed.), Interscience, New York.
Xu, S.C., Zhou, H., Wei, X., Jun, L., 1989. The pH dependence
and effects of the oxidative products of some aromatics
compounds in ozonation under UV irradiation. Ozone Sci.
Eng. 11, 281–296.
Yeh, C.K., Novak, J.T., 1994. Anaerobic biodegradation of
gasoline oxygenates in soils. Water Environ. Res. 66, 744–
752.