A systematic study of the degradation of dimethyl phthalate using a high-frequency ultrasonic process

Article abstract of DOI:10.1016/j.ultsonch.2012.11.005

A comprehensive study of the sonochemical degradation of dimethyl phthalate (DMP) was carried out using high-frequency ultrasonic processes. The effects of various operating parameters were investigated, including ultrasonic frequency, power density, initial DMP concentration, solution pH and the presence of hydrogen peroxide. In general, a frequency of 400 kHz was the optimum for achieving the highest DMP degradation rate. The degradation rate was directly proportional to the power density and inversely related to the initial DMP concentration. It was interesting to find that faster removal rate was observed under weakly acidic condition, while hydrolysis effect dominated in extreme-basic condition. The addition of hydrogen peroxide can increase the radical generation to some extent. Furthermore, both hydroxylation of the aromatic ring and oxidation of the aliphatic chain appear to be the major mechanism of DMP degradation by sonolysis based on LC/ESI-MS analysis. Among the principle reaction intermediates identified, tri- and tetra-hydroxylated derivatives of DMP, as well as hydroxylated monomethyl phthalates and hydroxylated phthalic acid were reported for the first time in this study. Reaction pathways for DMP sonolysis are proposed based on the detected intermediates.

Full text of DOI:10.1016/j.ultsonch.2012.11.005

Ultrasonics Sonochemistry 20 (2013) 892–899
Contents lists available at SciVerse ScienceDirect
Ultrasonics Sonochemistry
journal homepage: www.elsevier.com/locate/ultson
A systematic study of the degradation of dimethyl phthalate using a
high-frequency ultrasonic process
b
L.J. Xu a, W. Chu a, , Nigel Graham
⇑
a Department of Civil and Structural Engineering, Research Centre for Urban Environmental Technology and Management, The Hong Kong Polytechnic University,
Hung Hom, Kowloon, Hong Kong
b Department of Civil and Environmental Engineering, Imperial College London, South Kensington Campus, London SW7 2AZ, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 October 2012
Received in revised form 8 November 2012
Accepted 8 November 2012
Available online 23 November 2012
A comprehensive study of the sonochemical degradation of dimethyl phthalate (DMP) was carried out
using high-frequency ultrasonic processes. The effects of various operating parameters were investigated,
including ultrasonic frequency, power density, initial DMP concentration, solution pH and the presence of
hydrogen peroxide. In general, a frequency of 400 kHz was the optimum for achieving the highest DMP
degradation rate. The degradation rate was directly proportional to the power density and inversely
related to the initial DMP concentration. It was interesting to find that faster removal rate was observed
under weakly acidic condition, while hydrolysis effect dominated in extreme-basic condition. The addi-
tion of hydrogen peroxide can increase the radical generation to some extent. Furthermore, both hydrox-
ylation of the aromatic ring and oxidation of the aliphatic chain appear to be the major mechanism of
DMP degradation by sonolysis based on LC/ESI-MS analysis. Among the principle reaction intermediates
identified, tri- and tetra-hydroxylated derivatives of DMP, as well as hydroxylated monomethyl phtha-
lates and hydroxylated phthalic acid were reported for the first time in this study. Reaction pathways
for DMP sonolysis are proposed based on the detected intermediates.
Keywords:
Dimethyl phthalate
Ultrasound
Sonochemical
Hydrogen peroxide
Intermediates
Advanced oxidation process
Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction
lapse of bubbles in the liquid [6,7]. The final collapse of the bubbles
produces extremely high temperatures (>5000 °C) and pressures
Å
Phthalate esters (PAEs) are widely used as plasticizers in indus-
trial processes where they impart flexibility in various resins [1].
Some phthalates are suspected to be endocrine disrupting com-
pounds (EDCs) by mimicking estrogens, and therefore the possible
leaching of phthalates from plastics into the aqueous environment
may lead to negative effects on organisms [2]. Many studies have
been carried out trying to degrade PAEs, although most of them
were conducted using biological methods, and monomethyl
phthalate (MMP) and phthalic acid (PA) were found to be the most
frequent end-products [3–5].
Compared to biological methods, advanced oxidation processes
(AOPs) have attracted greater interest in degrading refractory
organics, by virtue of their high efficiency and the avoidance of fur-
ther treatment of biological sludge. The application of ultrasonic
irradiation, as an alternative to the more established AOPs (e.g.
O3/H2O2, O3/UV, H2O2/UV), has the advantage that it is potentially
simpler, safer, cleaner, requires less chemical dosing, and has been
widely investigated. The propagation of ultrasound through a li-
quid induces both physical and chemical processes by acoustic
cavitation: the formation, growth and adiabatically implosive col-
(>100 MPa), cleaving water molecules into OH and HÅ radicals in-
side the bubble and in the liquid shell surrounding the bubble
[8]. These radicals can combine mutually or with peripheral water
Å
molecules to form other radical species (e.g. HO2 ) and H2O2, which
in turn can oxidize organic substrates causing their degradation or
even mineralization [9,10].
Although many studies have been performed using ultrasound
to degrade different organic compounds, the information concern-
ing the sono-degradation of PAEs is scarce, especially concerning
the possible intermediates and products, which is useful to evalu-
ate the process and/or find other proper processes to combine with
it. Besides, many studies have focused on improving the efficiency
of organic compound degradation by incorporating other AOPs
with ultrasound [11–13], which is adverse for scrutinizing the de-
tailed mechanism of sonolysis due to the faster reaction rate of the
combined processes.
In this study, dimethyl phthalate (DMP), the simplest member
of the group of PAEs, was selected as the model compound to sys-
tematically examine its sono-degradation. DMP is one of the most
common PAEs in the aqueous environment. It demonstrates a rel-
atively high water solubility (ca. 4000 mg/L at 25 °C) and is more
hydrophilic (logKow = 1.60) compared to other PAEs. It is consid-
ered to have negligible volatility (with a Henry’s law constant of
⇑
Corresponding author. Tel.: +852 2766 6075; fax: +852 2334 6389.
E-mail address: cewchu@polyu.edu.hk (W. Chu).
1350-4177/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ultsonch.2012.11.005

L.J. Xu et al. / Ultrasonics Sonochemistry 20 (2013) 892–899
893
1.22 ꢀ 10ꢁ7 atm m3/mol) [1,14]. Owing to its environmental signif-
2.3. Ultrasonic system
icance, intensive efforts have been made to develop methods to re-
move DMP from waters and wastewaters. Many studies have
reported the degradation of DMP using diverse processes, such as
photocatalysis with TiO2 suspensions [15], ozonation with high sil-
ica zeolites and UV radiation [16], photochemical degradation by
Fenton reagent [17], etc. Most of these methods are viable only
in the presence of catalysts or other chemical inducers, giving rise
to both chemical consumption and further chemical contamina-
tion. The sonolysis of DMP utilizing solely a high frequency ultra-
sonic system (HFUS) has the advantage in this respect, but no
information concerning DMP sonolysis has so far been reported.
In view of the above, the performance of HFUS with DMP as the
selected compound was investigated in this study. The effects of
several parameters on the efficiency of the system were evaluated,
these being the ultrasonic frequency, ultrasonic power, DMP con-
centration, initial solution pH, and the effect of the presence of
H2O2. In addition, the main intermediates of DMP sono-degrada-
tion were identified and possible degradation pathways are also
proposed.
Sonolytic experiments were conducted in a cubic stainless steel
jacketed reactor with an effective volume of 1.5 L (tailor-made by
Ning Bo Scientz Biotechnology Co., China) (see Fig. 1). Three differ-
ent frequencies of 400, 800, and 1200 kHz were provided respec-
tively by three independent basins with an identical reactor size.
Electrical power was transduced into ultrasonic vibration by six
piezoelectric ceramics stuck uniformly in two rows (i.e. in a
2 ꢀ 3 format) on the back of the basin bottom. The nominal input
power was adjustable manually via the control panel. A convolute
cooling finger was submerged in the tested water to keep the solu-
tion temperature constant at 28 2 °C (not applicable in the calo-
rimetric experiment). A volume of exactly 250 mL DMP solution
was used throughout this study, except for the tests relating to dif-
ferent water volumes. The initial pH of DMP solution was main-
tained at 6.5 0.2 unless stated otherwise (e.g. for the tests at
different initial pH levels). Aliquots of DMP solution were with-
drawn from the reactor at predetermined time intervals for further
measurement. The total reduction in solution volume was kept be-
low 5% of the initial volume. All the selected tests were duplicated
with an observed deviation of less than 5%.
2. Materials and methods
2.1. Chemicals
3. Results and discussion
DMP (99.6%) and H2O2 (35% by weight) were purchased from
Sigma Aldrich Inc., USA. All chemicals were of analytical standard
and used without further purification. Solutions of 0.1 M sulfuric
acid and 0.1 M sodium hydroxide were used for pH adjustment.
Deionized-distilled water prepared from a Millipore Waters Milli-
Q water purification system was used exclusively in this study.
3.1. Ultrasonic frequency optimization
The DMP degradation performance for different ultrasonic fre-
quencies (400, 800, and 1200 kHz) and powers is shown in Fig. 2.
The results suggested that the DMP sono-degradation follows
pseudo first-order kinetics. The greatest degradation performance
was obtained from the 400 kHz process using its maximum input
power (available) of 120 W. A reaction rate constant, kA, of
6.7 ꢀ 10ꢁ3 minꢁ1 was observed and 87.2% of DMP was removed
in 300 min. At the same frequency, the degradation rate decreased
substantially when the input power reduced only moderately to
100 W, indicating that the power supplied to the ultrasound sys-
tem was also a major factor affecting the compound degradation
efficiency. Analogously, a similar phenomenon was observed for
the 800 kHz process, in which an increase in the nominal input
power from 120 W to its maximum of 160 W produced a corre-
sponding increase in the DMP removal performance. However,
the performance at 800 kHz was inferior to that of the 400 kHz/
120 W situation, and the magnitude of the kA of the 800 kHz pro-
cess was less than one third of the 400 kHz process with the same
input power (120 W). The worst performance was evident from the
1200 kHz process at its maximum nominal input power of 60 W,
where there was no measurable DMP degradation within 30 min.
2.2. Analytical method
The DMP concentration was determined by high performance
liquid chromatography (HPLC) (Waters), comprising a HPLC pump
(Waters 515), a UV detector (Waters 2489), an auto sampler
(Waters 717), and a Brava C18-BDS column (5 lm particle size,
250 ꢀ 4.6 mm). The mobile phase of acetonitrile/water (60/40,
vol/vol) was employed at a flow rate of 1.0 mL/min. The injection
volume was 10 lL and the concentration of DMP was detected at
230 nm. The concentration of H2O2 was determined by measuring
the absorbance of titanium peroxide ðTiO22þÞ using a spectropho-
tometer at 405 nm [18]. The ultrasonic power dissipated into the
liquid was calculated using a calorimetric method [19]. In the calo-
rimetric experiment, the temperature of the solution was mea-
sured by a digital thermometer and recorded at a 15 s interval
for the initial 10 min.
The identification of intermediates was performed by a Thermo
Quest Finnigan LCQ Duo LC/MS system equipped with a PDA-UV
detector and a mass spectrometer with an electrospray ionization
source and a quadrupole ion-trap mass analyzer. Positive mode
was applied to detect DMP, while negative mode was chosen to
qualitatively detect the intermediates. The mobile phase was a
mixture of 100% methanol (A) and 0.1% formic acid (B) with a lin-
early gradient flow from 10%A:90%B to 90%A:10%B within 55 min.
From 55 to 60 min, the ratio turned back to the initial composition.
The [DMP]0 for mass analysis was 0.25 mM. The intermediates
without commercial standard were quantified by their nominal
concentrations using the same response factor of DMP. This
approximation is reasonable on the basis that most of the UV
absorbance may be derived from the resonance structure of the
ring [20]. The total organic carbon (TOC) was analyzed by Shima-
dzu TOC-5000A analyzer.
stainless steel container
US controller
cooling water
cooling finger
outflow
cooling water
inflow
piezoelectric ceramics
holder
Fig. 1. Schematic diagram of the ultrasonic apparatus.

894
L.J. Xu et al. / Ultrasonics Sonochemistry 20 (2013) 892–899
0.0
-0.4
-0.8
-1.2
-1.6
-2.0
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
400kHz_100W
400kHz_120W
800kHz_120W
800kHz_160W
1200kHz_60W
125mL
250mL
500mL
750mL
-2.4
-2.8
-3.2
0
60
120
180
240
300
0
60
120
180
240
300
Time (min)
Time (min)
Fig. 2. Temporal variation of DMP concentration with ultrasonic frequency and
power ([DMP]0 = 0.1 mM; 250 mL solution volume).
Fig. 3a. Temporal variation of DMP concentration with solution volume
([DMP]0 = 0.1 mM; 400 kHz; 120 W).
The observation from these tests that the most effective fre-
quency for degrading DMP was 400 kHz, agrees with the results re-
ported in other studies [9]. The ultrasonic frequency plays an
essential role in determining the efficacy of the process by influ-
encing the cavitation performance. The number of acoustic cycles
and bubble collapses increases as the frequency increases, whereas
the bubbles formed at high-frequency with a correspondingly
smaller explosion diameter, release less energy than the low-fre-
quency bubbles for one single pulsation [21,22]. Thus, a higher fre-
quency may compensate for the lower energy release per single
bubble explosion by occurring more frequently. However, an above
optimal frequency only leads to a small energy release for a single
bubble, which could not be accumulated sufficiently by the repeat-
ing frequency. The optimal frequency is determined by the com-
prehensive performance of energy release (bubble sizes and
repeating frequency) and is likely to vary for different target com-
pounds. From the results of these initial tests, the most effective
frequency was found to be 400 kHz and this was used in all further
tests in this study.
et al. also reported that the calorimetrically determined ultrasonic
power measurements were independent of the water volume
[25]. Hence, it was concluded that the variation in solution volume
only changed the power density; the power density can be deduced
from Eq. (1) by incorporating the water density, as follows:
Power density ðW=mLÞ ¼ CpÅ
qÅdT=dt:
ð2Þ
Since Cp and density (mass/volume) are constants, a higher
q
power density (via less solution volume) leads to a more rapid
temperature rise, as observed experimentally.
The relationship between DMP decay rate and power density is
more evident in Fig. 3(b), where the power density and kA were in-
versely related to solution volume. An approximately linear rela-
tionship of kA as a function of power density was indicated from
the results (inset of Fig. 3(b)), confirming the positive role of power
density on DMP degradation. Hwang et al. also investigated the
dependence of the reaction rate constant on the power density
by varying the input power to obtain different power densities,
and the linear relationship was also proposed [26]. This positive
role of power density can be explained by the extent of the cavita-
3.2. Effect of power density
In order to clarify the power effect on sonolysis efficiency, the
DMP degradation performance was investigated by varying the
solution volume with a constant nominal input power of 120 W.
From Fig. 3(a), a significant decrease of the degradation rate was
observed with the increase of solution volume, indicating the
importance of power density (the power applied per unit volume
of irradiated liquid [23]) on DMP sonolysis.
The actual power dissipated into the water can influence the
amplitude of the ultrasonic wave, causing the bubbles to oscillate
more violently, and thereby giving rise to an increased production
of radicals and a greater pyrolysis performance [24]. The ultrasonic
power actually entering the solution can be determined approxi-
mately by measuring the initial rate of solution temperature rise,
dT/dt, and by Eq. (1):
0.07
10
8
10
8
400kHz
800kHz
0.06
0.05
0.04
0.03
0.02
0.01
0.00
6
4
6
2
0
0.00 0.02 0.04 0.06
Power density (W/mL)
4
2
Power ðWÞ ¼ ðdT=dtÞÅCpÅM
ð1Þ
0
where Cp is the specific heat of water (4200 J/kg °C), and M is the
mass of water (kg). It was observed that the initial rate of solution
temperature rise was inversely related to the solution volume, but
the total power absorbed was almost the same for different vol-
umes, accounting for 6.3–8.2% of the nominal input power. Kimura
125
250
375
500
625
750
Volume (mL)
Fig. 3b. Variation of DMP degradation rate and power density with solution volume
([DMP]0 = 0.1 mM; 400 kHz; 120 W).

L.J. Xu et al. / Ultrasonics Sonochemistry 20 (2013) 892–899
895
tion activity, whereby the number and/or density of the cavitating
bubbles increases with increasing power density [27].
oxide (reactions R4, R6 and R10), which accumulates linearly in
solution during the period of ultrasound irradiation (data not
shown) and which also acts as a hydroxyl radical scavenger
(R11; kOH = 2.7 ꢀ 107 Mꢁ1 sꢁ1 [28]):
Comparative tests were conducted with the 800 kHz process
and a similar linear trend was observed but with a much lower gra-
dient (inset of Fig. 3(b)). Thus, the 400 kHz process with a higher
gradient displayed a better DMP degradation capability compared
to that of the 800 kHz system with the same input power. Hence, in
general, the slope can well characterize the sonochemical property
among different ultrasonic systems and the linear correlation be-
tween kA and power density potentially might become a useful tool
in predicting the sonochemical process performance if the charac-
terization constant (slope) is predetermined.
H2O ! HÅ þ ÅOH
HÅ þ ÅOH ! H2O
2HÅ ! H2
ðR1Þ
ðR2Þ
ðR3Þ
ðR4Þ
ðR5Þ
ðR6Þ
ðR7Þ
ðR8Þ
ðR9Þ
ðR10Þ
ðR11Þ
2ÅOH ! H2O2
ÅH þ O2 ! HO2Å
HO2Å þ HO2Å ! H2O2 þ O2
2ÅOH ! OÅ þ H2O
2OÅ ! O2
3.3. Effect of initial DMP concentration
Previous studies have shown that the first-order degradation
rate of organic compounds by ultrasound is inversely related to
the initial compound concentration (e.g. [11]) and similar results
were found in this study where the initial DMP concentration
was varied from 0.01 to 0.15 mM (Fig. 4). The reason for this
behavior has not been fully explained in previous studies, and
one hypothesis is proposed here.
In view of the low volatility and hydrophilic nature of DMP, the
principal mechanism of DMP degradation in the HFUS is most
likely its oxidation by free radicals both at the cavitation bubble
interface and in the aqueous phase, rather than by pyrolysis inside
the bubble. This was confirmed by conducting tests in the presence
of 1.0 M MeOH, an effective radical (ÅOH) scavenger (kOH = 9.7 -
ꢀ 108 Mꢁ1 sꢁ1 [28]), in the DMP solution, and no measurable
reduction in the DMP concentration was evident for 300 min. Thus,
the degradation of DMP by hydroxyl radicals can be written as
follows:
OÅ þ 2HÅ ! H2O
OÅ þ H2O ! H2O2
ÅOH þ H2O2 ! H2O þ HO2Å
Å
Thus, the net rate of change of OH radicals in solution can be
expressed as follows:
y
w
ꢁd½ÅOHꢂ=dt ¼ kg ꢁ kOH½ÅOHꢂ ½DMPꢂ ꢁ kPer½ÅOHꢂ ½H2O2ꢂ
z
ꢁ kInt½ÅOHꢂ ½Intꢂ
ð4Þ
Å
where kg is the zero-order rate constant for OH radical generation;
[DMP], [H2O2] and [Int] are respectively the time-dependent con-
centrations of DMP, hydrogen peroxide, and DMP reaction interme-
diates; and kOH, kPer and kInt are the corresponding reaction rate
constants. Since under steady state conditions, d[ÅOH]/dt = 0, and
initially (t ꢃ 0 s) [H2O2] and [Int] can be assumed to be zero, and
[DMP] = [DMP]0 (initial DMP concentration), Eq. (4) can be simpli-
fied to give the [ÅOH] concentration as follows:
x
ꢁd½DMPꢂ=dt ¼ kDMP½ÅOHꢂ ½DMPꢂ
ð3Þ
The rate constant kDMP has been reported to be second-order
(x = 1) with a value of 2.67 ꢀ 109 Mꢁ1 sꢁ1 [29]. During sonication
Å
at a constant power input, it can be assumed that the rate of OH
Å
generation is constant, and that the OH concentration in solution
during the reaction maintains a steady state, whereby the genera-
tion of ÅOH radicals equals their consumption by reaction with the
target compound and scavengers [9]. The main reactions in pure
water induced by ultrasonic irradiation are summarized as follows
(R1–11) [9,10]. A prominent product of sonication is hydrogen per-
1=y
½ÅOHꢂ ¼ ðkg=kOH½DMPꢂ0Þ
ð5Þ
Combining Eq. (3) (assuming x = 1) and Eq. (5), gives the follow-
ing expression for the degradation of DMP:
1=y
ꢁd½DMPꢂ=dt ¼ kDMPðkg=kOH½DMPꢂ0Þ ½DMPꢂ
ð6Þ
Eq. (6) represents the first order degradation kinetics for DMP
where the pseudo first-order rate constant (kA) is equal to kDMP
(kg/kOH [DMP]0)1/y. The results shown in Fig. 4 are consistent with
the inverse dependency of kA on [DMP]0 as indicated by Eq. (6),
which in general terms can be written as:
C0=0.01mM
C0=0.05mM
C0=0.03mM
C0=0.10mM
C0=0.15mM
0.0
-1.0
-2.0
-3.0
-4.0
kA ¼ A½DMPꢂꢁ0 B
where A = kDMP (kg/kOH)
1/y, and B = 1/y.
ð7Þ
-4.2
-4.5
-4.8
-5.1
-5.4
The results obtained experimentally (Fig. 4) were in close agree-
ment with Eq. (7) (r2 = 0.98), with the value of B determined to be
0.39 (y = 2.6).
For the subsequent tests, unless otherwise stated, an initial
DMP concentration of 0.05 mM was selected in order to shorten
the reaction time.
-4.5 -4.0 -3.5-3.0 -2.5 -2.0
Ln C0(mM)
3.4. Effects of initial solution pH on DMP sono-degradation
0
60
120
180
240
300
The solution pH is generally an important factor influencing the
efficiency of the oxidation processes, and the influence of initial pH
on the sono-degradation of DMP was studied in the range of pH
Time (min)
Fig. 4. Effect of initial DMP concentration on DMP degradation (400 kHz; 120 W).

896
L.J. Xu et al. / Ultrasonics Sonochemistry 20 (2013) 892–899
Å
2–11; the results are presented in Fig. 5. It can be seen that the rate
constant was not substantially affected by pH, but generally de-
creased (by 15%) with increasing pH for the pH range of 4–9.
Complementary tests were carried out without ultrasound
(only stirring) to determine the direct effect of hydrolysis of DMP
at the different pH conditions. No significant hydrolysis was de-
tected for all the pH conditions except for pH 11. Hydrolysis was
evident at pH 11 corresponding to a DMP degradation of 49.4% in
240 min, which when compared to the overall degradation of
85.3% in 240 min with ultrasonic irradiation, indicated the exten-
sive role of DMP hydrolysis in strongly alkaline conditions. Yim
et al. have also reported the significant hydrolysis at pH 12 when
they examined the sonolysis of another PAEs member, diethyl
phthalate (DEP). In addition, the monoalkyl phthalate, which was
regarded as the hydrolysis product of DEP, also showed a faster
ultrasonic degradation efficiency under alkaline conditions [30].
Since the hydrolysis of DMP also followed pseudo first-order kinet-
ics (data not shown), the net degradation rate excluding hydrolysis
is also shown in Fig. 5. By excluding hydrolysis effects, it is clear
that the rate of DMP degradation by sonication decreased system-
atically, and to an increasingly greater extent, between pH 5 and
11. Since DMP is a non-dissociating compound, the most likely
reason for this degradation with increasing pH is the effect of
increasing hydroxyl radical scavenging by hydroxide ions
(kOH = 1.3 ꢀ 1010 Mꢁ1 sꢁ1 [28]), and the decreasing oxidation
solution, in which OH are generated directly from H2O2 sonolysis
[10,31]:
H2O2 ! 2ÅOH
ðR12Þ
Å
and where OH radicals are scavenged by H2O2, as described previ-
ously. Given the complexity of these related mechanisms, it is ex-
pected that the net effect of the presence of H2O2 on DMP
degradation will depend on the prevailing reaction conditions. A
previous study by the authors with a group of iodinated compounds
[23] showed that the extent of compound degradation by ultra-
sound was significantly enhanced by the addition of H2O2 to the
solution, but that the enhancement was significantly dependent
on the compound concentration (i.e. much less at higher
concentration).
In this study, the impact of H2O2 on DMP degradation was con-
sidered by the application of different H2O2 concentrations (0–
10 mM) at the start of a sonication reaction lasting 180 min. Prior
to these tests it was established that direct oxidation of DMP by
H2O2 alone was not significant (data not shown). The results of
addition of hydrogen peroxide in the presence of ultrasound are
shown in Fig. 6, where in general H2O2 had a promoting effect on
DMP degradation, which increased with the initial H2O2 concentra-
tion. As observed in the previous study mentioned above [23], the
promoting effect of H2O2 was dependent on [DMP]0, where the
presence of moderate concentrations of H2O2 (<5 mM) significantly
enhanced the compound degradation rate of the 0.02 mM DMP
solution, but gave very little enhancement in the case of the
0.05 mM DMP solution. However, at the greatest H2O2 concentra-
tion (10 mM) the degradation rates for the two DMP concentra-
tions were relatively close. It is believed that these effects reflect
Å
potential of OH with increasing pH [17].
The lower DMP degradation rate at pH 2 compared to the rates
at pH 4 and 5 is believed to be the effect of hydroxyl radical scav-
enging by hydrogen sulphate ions (kOH ꢃ 1 ꢀ 106 Mꢁ1 sꢁ1 [28]),
arising from the use of sulphuric acid to lower the pH in these tests
(pKa of HSO4 ꢃ 1.9). At pH 4 and 5 the concentration of added sul-
phate is very low and therefore the extent of scavenging will be
considerably less than at pH 2.
Å
the relative concentrations (molar ratio) of OH radicals and DMP
in the solutions, and the relative contribution that H2O2 makes to
ÅOH radical generation or scavenging. The accumulation of H2O2
during the sonolysis process within 3 h was less than 0.2 mM in
a 0.05 mM DMP solution (data not shown). Based on the result of
Fig. 5, under this low range of [H2O2], the role of H2O2 is mainly
a radical source rather than a radical scavenger.
3.5. Effect of addition of hydrogen peroxide on DMP decomposition
As discussed previously, hydrogen peroxide is generated during
ultrasound irradiation via reactions R4, R6 and R10. The influence
of H2O2 on the sonochemical degradation of a target compound
is expected to be determined by its impact on the concentration
of ÅOH in solution, and the direct, or synergistic (with ÅOH), reaction
between H2O2 and the compound. In the former case, the presence
of H2O2 can have both a positive and negative effect on [ÅOH] in
3.6. Identification of reaction intermediates from DMP sonolysis
The ultrasonic degradation of DMP is mainly attributed to hy-
droxyl radical oxidation (see Section 3.3). However, the sonolytic
process is slower compared to other ÅOH generating AOPs (e.g. Fen-
12
10
8
8
0.0
7
-0.4
-0.8
-1.2
-1.6
-2.0
6
5
2
4
6
8
10 12
6
pH
pH=2.02
pH=4.01
pH=5.06
pH=7.04
pH=9.03
pH=11.02
4
2
0
[
DMP 0
]
=0.02mM
=0.05mM
[DMP 0
]
0
60
120
180
240
0
2
4
6
8
10
[H2O2]0 (mM)
Time (min)
Fig. 5. Effect of initial solution pH on DMP degradation ([DMP]0 = 0.05 mM;
Fig. 6. Influence of hydrogen peroxide dose on the DMP degradation rate (400 kHz;
400 kHz; 120 W).
120 W).

L.J. Xu et al. / Ultrasonics Sonochemistry 20 (2013) 892–899
897
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1
2
3
4
5
6
7
8
9
10
DMP
mass balance
of benzene ring
0
2
4
6
8
10
12
14
Time (h)
Fig. 7. The concentration evolution of DMP and the detected sonolysis intermediates of DMP (400 kHz; 120 W; C0 = 0.25 mM).
DMP
O
C
OCH3
OCH3
CH3OH
C
O
1
4
6
HO
CO2CH3
CO2CH3
3
CO2CH3
5
HO
CO2CH3
COOH
COOH
(HO)2
CO2CH3
CO2CH3
CH3OH
8
COOH
COOH
9
(HO)2
CO2CH3
COOH
(HO)3
CO2CH3
CO2CH3
10
HO
COOH
COOH
OH
7
HO
HO
CO2CH3
CO2CH3
OH
Further products
Fig. 8. Degradation pathways of sonochemical degradation of DMP (Dashed arrows indicate reactions that require more than one step).
ton, UV/H2O2, etc.) due to its distinct heterogeneous water–bubble
scavenging in sonication process is correlated to the hydrophobic-
ity of the solutes rather than the specific reactivity towards ÅOH in
homogeneous solution [32]. Psillakis et al. have proved that the
Å
interface which can repel hydrophilic compounds from the OH
Å
concentrated area. It has been reported that the efficiency of OH

898
L.J. Xu et al. / Ultrasonics Sonochemistry 20 (2013) 892–899
PAEs with higher hydrophobicity demonstrated greater sonolytic
degradation efficiency [33]. Therefore, reaction intermediates with
more hydrophilic characteristics might have difficulties in compet-
sound power density, and decreased moderately with increasing
pH in the range typically found in the environment (pH 5–9). High-
er initial concentration results in a lower degradation rate. Both the
hydroxylation of the benzene ring and the oxidation the aliphatic
chain were found to be the dominating mechanisms of DMP sonol-
ysis. In addition, the HFUS was found to be effective in degrading
more hydrophobic DMP, but less efficient in degrading its more
hydrophilic intermediates. All these findings may cast some light
on understanding the mechanism of the sonolytic processes and
finding proper process to enhance the efficiency.
Å
ing OH with their mother compound, DMP. The accumulation of
these intermediates not only confirm this hypothesis, but makes
them easier to be detected. This explains why many new interme-
diates were reported in this study for the first time, because they
were easily escaped from the detection in other AOPs.
The information of the identified intermediates is summarized
(see Table 1 in Supporting information), and the evolution profile
of the main intermediates is shown in Fig. 7. For some intermedi-
ates the addition of one formate anion, [M+45]ꢁ, was observed due
to the use of 0.1% formic acid as the mobile phase [34]. All the
intermediates detected in this study had shorter retention times
than DMP, implying a higher polarity than their mother compound.
The hydroxyl groups attached to the benzene ring readily partici-
Acknowledgements
The authors are grateful for the financial support of a research
grant from the Research Grants Council of the Hong Kong Special
Administrative Region, China (Project No. PolyU 5146/10E).
pate in the p–p conjugation effect, and therefore the electron den-
sity was increased. In addition, the carboxylic group was also
beneficial for electron withdrawing, all resulting in a higher polar-
ity of these intermediates. Many studies have reported hydroxyl-
Appendix A. Supplementary data
Å
ation in different AOPs involving the mechanism of OH radicals
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ultsonch.2012.
11.005.
[35–37], and hydroxylated intermediates were also found in sono-
chemical reactions [8]. However, in the case of DMP, only mono-
and bi-hydroxylated phthalates were reported by using other AOPs
[35,36]. The tri- and tetra-hydroxylated derivatives of DMP, the hy-
droxy-MMP, and hydroxy-PA are the reported for the first time in
this study. This is partly because LC/MS is more capable of detect-
ing hydrophilic intermediates than that of GC/MS. Another reason,
as mentioned above, is the slower generation of hydroxal radicals
on the hydrophobic surface of bubbles resulting in the accumula-
tion of the more hydrophilic intermediates in the solution.
As can be seen from Fig. 7, compounds 1–4 were more abundant
during the early period, but declined later after reaching a peak at
6–8 h. Compound 2 with evident LC signal hardly responded to MS
detection (see Fig. 1 in SI). With further degradation of these pri-
mary intermediates, secondary derivatives appeared and accumu-
lated gradually (compounds 5–10). In addition, the dimethyl
hydroxy-phthalate with more hydroxylated groups increased
slower during the reaction, indicating that the hydroxylation
potentially proceeded step by step. Hence, the degradation path-
way of DMP sonolysis was proposed in Fig. 8. The degradation of
DMP involved the hydroxylation of the aromatic ring, the oxidation
of the aliphatic chain, or both. It should be noted that the interme-
diates of MMP and PA have been reported to be much less toxic
than DMP [38].
References
[1] C.A. Staples, D.R. Peterson, T.F. Parkerton, W.J. Adams, The environmental fate
of phthalate esters: a literature review, Chemosphere 35 (1997) 667–749.
[2] S. Takeuchi, M. Iida, S. Kobayashi, K. Jin, T. Matsuda, H. Kojima, Differential
effects of phthalate esters on transcriptional activities via human estrogen
receptors alpha and beta, and androgen receptor, Toxicology 210 (2005) 223–
233.
[3] D.R. Shelton, S.A. Boyd, J.M. Tiedje, Anaerobic biodegradation of phthalic-acid
esters in sludge, Environ. Sci. Technol. 18 (1984) 93–97.
[4] D. Vega, J. Bastide, Dimethylphthalate hydrolysis by specific microbial esterase,
Chemosphere 51 (2003) 663–668.
[5] D.W. Liang, T. Zhang, H.H.P. Fang, Anaerobic degradation of dimethyl phthalate
in wastewater in a UASB reactor, Water Res. 41 (2007) 2879–2884.
[6] M. Chiha, S. Merouani, O. Hamdaoui, S. Baup, N. Gondrexon, C. Petrier,
Modeling of ultrasonic degradation of non-volatile organic compounds by
Langmuir-type kinetics, Ultrason. Sonochem. 17 (2010) 773–782.
[7] W.P. Xie, Y. Qin, D.M. Liang, D. Song, D.W. He, Degradation of m-xylene
solution using ultrasonic irradiation, Ultrason. Sonochem. 18 (2011) 1077–
1081.
[8] M. Inoue, Y. Masuda, F. Okada, A. Sakurai, I. Takahashi, M. Sakakibara,
Degradation of bisphenol
(2008) 1379–1386.
A using sonochernical reactions, Water Res. 42
[9] M. Inoue, F. Okada, A. Sakurai, M. Sakakibara, A new development of dyestuffs
degradation system using ultrasound, Ultrason. Sonochem. 13 (2006) 313–
320.
According to the mass balance as shown in Fig. 7 and 57% of
detectable benzene ring structures were opened when 96% DMP
was removed. However, no evident TOC removal was observed
during this period (data not shown). This suggests the likely forma-
tion and accumulation of many aliphatic organics with higher
hydrophilicity, accounting for the intact of TOC. If mineralization
is one of the treatment objectives, other proper AOPs should be
combined with ultrasound, such as UV irradiation (although DMP
is inert toward direct photolysis [39]) to improve the decomposi-
tion of the more hydrophilic intermediates. Na et al. have proved
that the mineralization of DEP was improved significantly when
ultrasound was combined with UV irradiation, which might be also
applicable for DMP degradation [40,41].
[10] Y.G. Adewuyi, Sonochemistry: environmental science and engineering
applications, Ind. Eng. Chem. Res. 40 (2001) 4681–4715.
[11] L.K. Weavers, N. Malmstadt, M.R. Hoffmann, Kinetics and mechanism of
pentachlorophenol degradation by sonication, ozonation, and sonolytic
ozonation, Environ. Sci. Technol. 34 (2000) 1280–1285.
[12] J. Peller, O. Wiest, P.V. Kamat, Synergy of combining sonolysis and
photocatalysis in the degradation and mineralization of chlorinated aromatic
compounds, Environ. Sci. Technol. 37 (2003) 1926–1932.
[13] Y. Segura, R. Molina, F. Martinez, J.A. Melero, Integrated heterogeneous sono-
photo Fenton processes for the degradation of phenolic aqueous solutions,
Ultrason. Sonochem. 16 (2009) 417–424.
[14] H.S.D.B. (HSDB), <http://toxnet.nlm.nih.gov>, (2007).
[15] E.D. Du, Y.X. Zhang, L. Zheng, Photocatalytic degradation of dimethyl phthalate
in aqueous TiO(2) suspension:
a modified Langmuir-Hinshelwood model,
React. Kinet. Catal. Lett. 97 (2009) 83–90.
[16] Y.H. Chen, N.C. Shang, D.C. Hsieh, Decomposition of dimethyl phthalate in an
aqueous solution by ozonation with high silica zeolites and UV radiation, J.
Hazard. Mater. 157 (2008) 260–268.
[17] X.K. Zhao, G.P. Yang, Y.J. Wang, X.C. Gao, Photochemical degradation of
dimethyl phthalate by Fenton reagent, J. Photochem. Photobiol. A-Chem. 161
(2004) 215–220.
4. Conclusions
[18] D.W. O’Sullivan, M. Tyree, The kinetics of complex formation between Ti(IV)
and hydrogen peroxide, Int. J. Chem. Kinet. 39 (2007) 457–461.
[19] T.J. Mason, J.P. Lorimer, D.M. Bates, Quantifying sonochemistry – casting some
light on a black art, Ultrasonics 30 (1992) 40–42.
[20] C.D. Adams, S.J. Randtke, Ozonation by-products of atrazine in synthetic and
natural-waters, Environ. Sci. Technol. 26 (1992) 2218–2227.
In this study, DMP can be degraded by HFUS via reacting with
ÅOH radicals, and the degradation follows pseudo first-order kinet-
ics. The optimal performance corresponded to a frequency of
400 kHz. The degradation rate increased directly with the ultra-

L.J. Xu et al. / Ultrasonics Sonochemistry 20 (2013) 892–899
899
[21] C. Petrier, A. Francony, Ultrasonic waste-water treatment: incidence of
ultrasonic frequency on the rate of phenol and carbon tetrachloride
degradation, Ultrason. Sonochem. 4 (1997) 295–300.
[31] Y. Ku, Y.H. Tu, C.M. Ma, Effect of hydrogen peroxide on the decomposition of
monochlorophenols by sonolysis in aqueous solution, Water Res. 39 (2005)
1093–1098.
[22] A. Henglein, Sonochemistry – historical developments and modern aspects,
Ultrasonics 25 (1987) 6–16.
[32] A. Henglein, C. Kormann, Scavenging of oh radicals produced in the sonolysis
of water, Int. J. Radiat. Biol. 48 (1985) 251–258.
[23] B. Ning, N.J.D. Graham, P.D. Lickiss,
A
comparison of ultrasound-based
[33] E. Psillakis, D. Mantzavinos, N. Kalogerakis, Monitoring the sonochemical
degradation of phthalate esters in water using solid-phase microextraction,
Chemosphere 54 (2004) 849–857.
advanced oxidation processes for the removal of X-ray contrast media,
Water Sci. Technol. 60 (2009) 2383–2390.
[24] S.F. Xiong, Z.L. Yin, Z.F. Yuan, W.B. Yana, W.Y. Yang, J.J. Liu, F. Zhang, Dual-
frequency (20/40 kHz) ultrasonic assisted photocatalysis for degradation of
methylene blue effluent: synergistic effect and kinetic study, Ultrason.
Sonochem. 19 (2012) 756–761.
[25] T. Kimura, T. Sakamoto, J.M. Leveque, H. Sohmiya, M. Fujita, S. Ikeda, T. Ando,
Standardization of ultrasonic power for sonochemical reaction, Ultrason.
Sonochem. 3 (1996) S157–S161.
[26] A. Hwang, S. Na, J. Ha, J. Khim, Degradation of diethyl phthalate by sono-fenton
process and its dependence on the power density, Jpn. J. Appl. Phys. 50 (2011).
[27] M. Sivakumar, A. Gedanken, Insights into the sonochemical decomposition of
Fe(CO)(5): theoretical and experimental understanding of the role of molar
concentration and power density on the reaction yield, Ultrason. Sonochem.
11 (2004) 373–378.
[28] G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, Critical-review of rate
constants for reactions of hydrated electrons, hydrogen-atoms and hydroxyl
radicals (ÅOH/ÅOꢁ) in aqueous-solution, J. Phys. Chem. Ref. Data 17 (1988) 513–
886.
[34] C.Y. Quang, G. Kocan, D. Tang, D.M. Fast, S.M. Michael, Development of an LC–
ESI-MS–MS method with formate anion attachment for detecting neutral
molecules in rat plasma, Am. Lab. 38 (2006) 26.
[35] B.L. Yuan, X.Z. Li, N. Graham, Reaction pathways of dimethyl phthalate
degradation in TiO2-UV–O2 and TiO2-UV–Fe(VI) systems, Chemosphere 72
(2008) 197–204.
[36] K.S. Tay, N.A. Rahman, M.R. Bin Abas, Fenton degradation of dialkylphthalates:
products and mechanism, Environ. Chem. Lett. 9 (2011) 539–546.
[37] Y.F. Rao, W. Chu, Reaction mechanism of linuron degradation in TiO2
suspension under visible light irradiation with the assistance of H2O2,
Environ. Sci. Technol. 43 (2009) 6183–6189.
[38] S. Jonsson, A. Baun, Toxicity of mono- and diesters of o-phthalic esters to a
crustacean, a green alga, and a bacterium, Environ. Toxicol. Chem. 22 (2003)
3037–3043.
[39] Y.H. Chen, L.L. Chen, N.C. Shang, Photocatalytic degradation of dimethyl
phthalate in an aqueous solution with Pt-doped TiO(2)-coated magnetic
PMMA microspheres, J. Hazard. Mater. 172 (2009) 20–29.
[29] G. Wen, J. Ma, Z.Q. Liu, L. Zhao, Ozonation kinetics for the degradation of
phthalate esters in water and the reduction of toxicity in the process of O-3/
H2O2, J. Hazard. Mater. 195 (2011) 371–377.
[40] S. Na, Y.G. Ahn, M. Cui, J. Khim, Significant diethyl phthalate (DEP) degradation
by combined advanced oxidation process in aqueous solution, J. Environ.
Manage. 101 (2012) 104–110.
[30] B. Yim, Y. Nagata, Y. Maeda, Sonolytic degradation of phthalic acid esters in
aqueous solutions. Acceleration of hydrolysis by sonochemical action, J. Phys.
Chem. A 106 (2002) 104–107.
[41] S. Na, C. Jinhua, M. Cui, J. Khim, Sonophotolytic diethyl phthalate (DEP)
degradation with UVC or VUV irradiation, Ultrason. Sonochem. 19 (2012)
1094–1098.