Ultrasound-assisted third-liquid phase-transfer catalyzed esterification of sodium salicylate in a continuous two-phase-flow reactor

Article abstract of DOI:10.1016/j.ultsonch.2009.06.004

The esterification of sodium salicylate to synthesize butyl salicylate by third-liquid phase-transfer catalysis under ultrasound irradiation was investigated in a continuous two-phase-flow reactor. The reactor was designed to keep the third-liquid phase in the middle part and to have the aqueous and organic phases flowing through it in countercurrent. Using tetra-n-butylphosphonium bromide to prepare the third-liquid phase for this esterification, the product yield in the organic outlet (toluene) at 70 °C was 49.7% in silent condition, showing the reaction promoted simply by countercurrent mixing of the aqueous and organic phases. In the conditions of space time at 168 min, stirring at 150 rpm and ultrasound irradiation (28 kHz, 300 W), the product yield was greatly enhanced to 78.2%. As prepared, above 90% of the added catalyst existed in the third-liquid phase, and after 4-h on stream for a large excess of n-butyl bromide to sodium salicylate, the fraction of catalyst retaining in the reactor was reduced to around 80%. The distributions of catalysts between phases before and after reaction were analyzed. A kinetic model was proposed to estimate the apparent rate constants, and the feasibility for third-liquid phase-transfer catalysis assisted by ultrasound irradiation in a continuous flow reactor was demonstrated.

Full text of DOI:10.1016/j.ultsonch.2009.06.004

Ultrasonics Sonochemistry 17 (2010) 239–245
Contents lists available at ScienceDirect
Ultrasonics Sonochemistry
journal homepage: www.elsevier.com/locate/ultsonch
Ultrasound-assisted third-liquid phase-transfer catalyzed esterification
of sodium salicylate in a continuous two-phase-flow reactor
*
Hung-Ming Yang , Guan-Yi Peng
Department of Chemical Engineering, National Chung Hsing University, 250 Kuo-kuang Road, Taichung 402, Taiwan, ROC
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 May 2009
Received in revised form 18 May 2009
Accepted 3 June 2009
Available online 7 June 2009
The esterification of sodium salicylate to synthesize butyl salicylate by third-liquid phase-transfer catal-
ysis under ultrasound irradiation was investigated in a continuous two-phase-flow reactor. The reactor
was designed to keep the third-liquid phase in the middle part and to have the aqueous and organic
phases flowing through it in countercurrent. Using tetra-n-butylphosphonium bromide to prepare the
third-liquid phase for this esterification, the product yield in the organic outlet (toluene) at 70 °C was
49.7% in silent condition, showing the reaction promoted simply by countercurrent mixing of the aque-
ous and organic phases. In the conditions of space time at 168 min, stirring at 150 rpm and ultrasound
irradiation (28 kHz, 300 W), the product yield was greatly enhanced to 78.2%. As prepared, above 90%
of the added catalyst existed in the third-liquid phase, and after 4-h on stream for a large excess of
n-butyl bromide to sodium salicylate, the fraction of catalyst retaining in the reactor was reduced to
around 80%. The distributions of catalysts between phases before and after reaction were analyzed. A
kinetic model was proposed to estimate the apparent rate constants, and the feasibility for third-liquid
Keywords:
Third-liquid phase
Phase-transfer catalysis
Ultrasound irradiation
Butyl salicylate
Two-phase-flow reactor
phase-transfer catalysis assisted by ultrasound irradiation in
demonstrated.
a continuous flow reactor was
Ó 2009 Elsevier B.V. All rights reserved.
1. Introduction
can be greatly promoted by TLPTC compared to liquid–liquid PTC.
For the benzoylation of sodium 4-acetylphenoxide in a TLPTC
system the reaction rate reported was much faster than that
conducted in a liquid–liquid PTC system by 25–28 times, and the
concentration of catalytic intermediate in the third-liquid phase
slightly decreased with increasing the reaction time [6,7].
Phase-transfer catalysis is an effective technology to conduct
the reactions between mutually insoluble reactants and has been
widely applied in organic synthesis, such as pharmaceuticals,
flavors, dyes, and perfumes [1]. According to which phases pres-
ent in the system, phase-transfer catalysis can be divided into
several categories, i.e. liquid–liquid, solid–liquid, gas–liquid,
solid–liquid–liquid, and tri-liquid types, among which the reac-
tion in tri-liquid phases is catalyzed by a highly concentrated
catalyst thin layer, called the ‘‘third-liquid” phase. The third-
liquid phase-transfer catalysis (TLPTC) is termed as a green
technology for that it can give a high product yield at mild con-
ditions without aprotic or halide-containing solvents and the
third-liquid phase is easily separated to be reused [2].
In preparing the third-liquid phase, the amount of inorganic salt
and the solvent polarity are two important factors, and the aque-
ous reactant is also important in certain conditions [3,4]. In
third-liquid catalyzed reactions the catalytic intermediate, which
is produced from the reaction of aqueous reactant and phase-
transfer catalyst, in the third-liquid phase plays a significant role
in conducting the intrinsic reaction [5]. Generally, the reaction rate
In recent years, the application of ultrasound irradiation in
organic synthesis has been broadly extended. It enhances the reac-
tion rates, yield and selectivity of product rather significantly [8].
When the ultrasonic waves propagate in a liquid solution, the
alternate compression and depression are produced to form cavi-
ties within which temperature and pressure can reach 5000 K
and several hundred bars. The reaction rate is thus enhanced. Re-
cently, the combination of ultrasound or microwave activation
with phase-transfer catalysis has attracted considerable interest
[9]. The effect of ultrasonic wave on the Cannizzaro reaction cata-
lyzed by a phase-transfer catalyst showed that an ultrasonic wave
of 20 kHz dramatically accelerated the reaction [10]. For the use of
ultrasound irradiation on the liquid–liquid phase-transfer cata-
lyzed epoxidation and dichlorocyclopropanation of 1,7-octadiene,
the reaction rates were greatly enhanced either in epoxidation by
using Aliquat 336/phosphortungstic acid or in dichlorocyclopropa-
nation by using benzyltriethylammonium chloride as the catalyst
[11,12]. The reaction of oximes and dichloromethane catalyzed
by ultrasound-assisted PTC gave 95% of product yield in 3.5 h,
* Corresponding author. Tel.: +886 4 22840510x609; fax: +886 4 22854734.
E-mail address: hmyang@dragon.nchu.edu.tw (H.-M. Yang).
1350-4177/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.ultsonch.2009.06.004

240
H.-M. Yang, G.-Y. Peng / Ultrasonics Sonochemistry 17 (2010) 239–245
but the time needed to get 92% in the silence condition was 16 h
[13]. For selective nitration of phenols the rate was enhanced with
dilute nitric acid (6 wt.%) and tetrabutylammonium bromide under
sonication [14].
Ultrasound irradiation together with liquid–liquid phase-trans-
fer catalysis has shown a significant improvement in the reaction
rate. In the past efforts, the application of ultrasound in continu-
ously operated third-liquid phase-transfer catalysis was not stud-
ied. By combining the highly efficient TLPTC with ultrasound, the
overall reaction rate would be enhanced. The aim of the present
study is to investigate the green synthesis of butyl salicylate by
an efficient process of ultrasound-assisted third-liquid phase-
transfer catalysis in a countercurrent two-phase-flow reactor. A
kinetic model was proposed to estimate the reaction rate con-
stants. The distributions of the catalyst between phases in the con-
tinuous operation mode were analyzed.
Fig. 1. Schematic flow diagram of the continuous TLPTC reaction system. T1:
feedstock of aqueous phase, T2: feedstock of organic phase, T3: make-up catalyst,
T4: storage tank of aqueous phase, T5: storage tank of organic phase, P1, P2: feeding
pump, R: reactor, U: ultrasound generator (bath type).
2. Experimental
The aqueous reactant sodium salicylate (ArCOONa, 99.5+%)
from SHOWA was used without further treatment. Tetra-n-butyl-
phosphonium bromide (TBPB or QBr in general), tetra-n-butylam-
monium bromide (TBAB), tetra-n-butylammonium iodide (TBAI),
Benzyltributylammonium bromide (BTBAB), n-butyl bromide
(RBr), the desired product n-butyl salicylate (ArCOOR), and other
reagents are all reagent-grade chemicals from SHOWA, ACROS,
TEDIA, Lancaster and TCI.
continuous process is shown in Fig. 1. Before starting a kinetic
run, the total mixture in preparing the third-liquid phase was first
introduced into the reactor, and the temperature was raised to the
set point. The continuous run was started by the procedures, add-
ing a definite amount of RBr into the reaction zone to meet the set-
ting concentration of RBr, pumping and balancing the aqueous and
organic streams between the inlet and outlet, and then activating
the ultrasound irradiation. The flow rates of the aqueous and or-
ganic streams were controlled at the desired values. The operation
at steady state would be achieved after the time on stream approx-
imately corresponded to the space time. The product yield in the
outlet and the variations of catalyst and catalytic intermediate
after reaction were analyzed. The organic and aqueous samples
were withdrawn in the sampling port of the process lines at each
chosen time. The amount of Q+ in the third-liquid phase was deter-
mined by the Volhard’s method. ArCOOR and ArCOOQ were ana-
lyzed by HPLC. The organic sample and the third-liquid sample
were diluted into 4 cm3 of acetonitrile and analyzed with internal
2.1. Formation and determination of the third-liquid phase
Typical preparation procedures for the third-liquid phase em-
ployed in continuous operation were as follows. An aqueous solu-
tion of ArCOONa at 0.25 M was prepared using 200 cm3 of
deionized water. Then 0.0125 mol of TBPB and 1 mol of NaBr were
introduced into 50 cm3 of the aforementioned aqueous solution,
which was subsequently mixed with 50 cm3 of toluene solvent.
After 15 min of shaking at 70 °C, the third-liquid phase that con-
tained catalytic intermediate ArCOOQ was formed. This third-li-
quid phase residing between the aqueous phase and toluene was
separated for analysis or for kinetic runs. To get ArCOOQ, the
third-liquid phase was mixed with 10 cm3 of deionized water, from
which ArCOOQ was extracted three times using 10 cm3 of chloro-
form. After separating chloroform, the gel form of ArCOOQ was
purified by drying and identified with NMR. The amount of tetra-
n-butylphosphonium cation (denoted as Q+) was determined by
the titration method in water–chloroform system using sodium
tetraphenyl borate (0.05 N of Na+TBPꢀ) as the titrant and bromo-
phenol blue (0.2% in ethanol) as the indicator. When the end point
is reached, the color of the chloroform layer will change from blue
to colorless.
method using HPLC with C-18 (5-lm) column and variable-wave-
length UV detector at 254 nm. The flow rate of eluent was 1.0 cm3/
min in a volume ratio of acetonitrile/water to be 70/30.
3. The kinetic model
For the third-liquid phase-transfer catalyzed esterification un-
der ultrasound irradiation (denoted by)))) symbol), the key step
is the formation of catalytic intermediate, which is produced from
the reaction of ArCOONa and catalyst QBr in the aqueous phase.
The reaction system is shown in the following:
2.2. The continuous two-phase-flow reactor system and kinetic
measurements
A continuous two-phase-flow reactor system equipped with
ultrasound irradiation was set up for third-liquid phase-transfer
catalysis. The reactor was designed in a way to keep the third-li-
quid phase in the middle part and to have the aqueous and organic
phases flowing through it in countercurrent. Two membranes
(hydrophobic and hydrophilic types) were installed at upside and
down side of the reaction zone to reduce the entrainment of the
third-liquid phase by the out-flowing streams. For simplicity, the
bath type of sonicator with variable ultrasonic frequencies and
powers (typically, 300 W by six ultrasonic transducers, each with
50 W/28 kHz) was selected. The schematic flow diagram for this
The reaction and mass-transfer steps involved in the above overall
reaction are:

H.-M. Yang, G.-Y. Peng / Ultrasonics Sonochemistry 17 (2010) 239–245
241
where Vaq is the volume of aqueous phase in the reactor.
(a) reaction of QBr and ArCOONa in the aqueous phase to form
ArCOOQ which is then distributed into the third-liquid
phase and the organic phase
Eq. (10) can be rearranged to give
aq
Vaq
vaq
C
ArCOONa;0
th
ArCOOQ
ꢀ 1 ¼
gkC
K
RBrda
¼ kappsaq
ð11Þ
aq
ArCOONa
ArCOONa ðaqÞ þ QBr ðaqÞ ꢀ ArCOOQ ðaqÞ þ NaBr ðaqÞ;
ArCOOQ ðaqÞ ꢀ ArCOOQ ðthÞ ꢀ ArCOOQ ðorgÞ
C
th
ArCOOQ
where
a
¼ CoRrBgr=Caq
,
saq ¼ Vaq=vaq and kapp
¼
gka
dC
KRBr
.
ArCOONa
The apparent reaction rate constant kapp is thus functions of
intrinsic reaction rate, ultrasonic efficiency, ratio of reactant,
third-liquid volume, distributions of reactants and concentration
of ArCOOQ in the third-liquid phase. At steady state, the product
ArCOOR in the organic outlet comes from the consumption of
ArCOONa in the aqueous stream. Thus, the conversion of ArCOONa
into ArCOOR in the flow system is
with the distribution relation
th
ArCOOQ
aq
C
KArCOOQ
¼
ð1Þ
C
ArCOOQ
(b) transfer of RBr from the organic phase into the third-liquid
phase,
ð2ÞRBr ðorgÞ ꢀ RBr ðthÞ
vaqCaAqrCOONa;0 ꢀ vaqCAaqrCOONa ¼ vorgCAorrgCOOR
ð12Þ
with the distribution relation
With the same volumetric flow rates of aqueous and organic phases
th
ðvaq ¼ vorgÞ and defining the product yield Ys in the organic outlet at
C
C
RBr
org
RBr
steady state, Ys ¼ CoArrgCOOR=CAaqrCOONa;0, Eq. (12) can be rearranged to
KRBr
¼
ð2Þ
give
(c) intrinsic reaction of ArCOOQ and RBr in the third-liquid
phase under ultrasound irradiation,
aq
C
C
ArCOONa
1 ꢀ Ys ¼
ð13Þ
aq
ArCOONa;0
ArCOOQ ðthÞ þ RBr ðthÞ ! ArCOOR ðthÞ þ QBr ðthÞ
with the reaction rate of ArCOOQ
By substituting Eq. (13) into (11) and rearranging, we can get
the performance equation at steady state as
th
th
ArCOOQ
ꢀr
¼
gkC
Cth
ð3Þ
ArCOOQ RBr
Ys
ð1 ꢀ YsÞ
kapp
¼
ð14Þ
aq
s
where k is the reaction rate constant and
g
represents the
efficiency factor of ultrasound irradiation on the reaction.
(d) transfer of the regenerated QBr from the third-liquid phase
into the aqueous phase,
The apparent rate constant kapp can be obtained from Eq. (14) and
used to describe the reaction system under ultrasonic irradiation.
4. Results and discussion
QBr ðthÞ ꢀ QBr ðaqÞ
with the distribution relation
4.1. Reproducibility and stability tests of two-phase countercurrent
flow reactor
aq
QBr
C
KQBr
¼
ð4Þ
th
QBr
C
The reproducibility and stability of the present continuous two-
phase-flow reaction system were performed. Prior to the reaction,
the third-liquid phase and equal volumes of water and toluene
were first introduced into the reaction zone. The reaction condi-
tions were set at 0.25 M of ArCOONa, 3.0 M of RBr, 1 mol of NaBr
and 0.0125 mol of catalyst TBPB with toluene as the organic sol-
vent, the flow rate (aqueous or organic phase) at 0.3 cm3/min (i.e.
168 min of space time), temperature at 70 °C, and 150 rpm of stir-
ring speed under ultrasound irradiation (28 kHz, 300 W). The
aqueous and organic streams were continuously fed into the reac-
tion system. The flow was controlled by regulating the volumetric
balance of streams between the inlet and outlet. As the steady state
was attained, the product yield in the organic outlet nearly main-
tained a constant. Fig. 2a shows the results of Runs 1, 2, and 3
for 4-h on stream. The average product yields in the organic outlet
at steady state were 78.2% for Run 1, 78.5% for Run 2, and 78.3% for
Run 3, demonstrating a good reproducibility for this continuous
operation. The stability of the reaction process was further tested
for 10-h on stream, and 80.1% of the product yield in the organic
outlet was observed, as shown in Fig. 2b. The 4-h test (78.2% of
yield) and 10-h test (80.1% of yield) revealed the steady perfor-
mance of the third-liquid phase catalyzed esterification. For the
two runs, the volume of the third-liquid phase was diminished
from 8.0 cm3 initially to 2.6 cm3 at 4 h and to 1.4 cm3 at 10 h. Since
the intrinsic reaction was conducted in the third-liquid phase by
the countercurrent two-phase flow, the catalyst would be distrib-
uted into the organic and aqueous phases and was entrained out
of the reactor. This caused the catalyst in the reaction zone gradu-
ally reduced. Further, when the use of RBr was in large excess with
respect to ArCOONa, the catalytic intermediate ArCOOQ in the
(e) transfer of the product ArCOOR from the third-liquid phase
into the organic phase with the distribution relation,
th
ArCOOR
org
C
C
KArCOOR
¼
ð5Þ
ArCOOR
Taking material balance for ArCOONa in the continuous flow
system at steady state, it gives
vaqCaAqrCOONa;0 ꢀ vaqCaAqrCOONa ¼ ðꢀrArCOONaÞVth
ð6Þ
where Vth is the volume of third-liquid phase, and vaq is the volu-
metric flow rate of aqueous phase. The rate of change of ArCOOQ
in the third-liquid phase is the net rate of formation and consump-
tion of ArCOOQ, i.e.
th
dC
ArCOOQ
¼ ðꢀrArCOONaÞ ꢀ ðꢀrArCOOQ
Þ
ð7Þ
dt
From the experimental results in analyzing ArCOOQ, the amount of
ArCOOQ in the third-liquid phase almost kept at constant (typically
around 0.084 mmol before and after 180 min), hence,
th
dC
ArCOOQ
ffi 0
ð8Þ
ð9Þ
dt
Substitution of Eqs. (7) and (8) into (6), we have
vaqCaAqrCOONa;0 ꢀ vaqCaAqrCOONa ¼ ðrArCOOQ ÞVth
Substituting Eqs. (2) and (3) into (9) with d ¼ Vth=Vaq gives
th
ArCOOQ
vaqCaAqrCOONa;0 ꢀ vaqCaAqrCOONa
¼
gkC
KRBrCoRrBgrdVaq
ð10Þ

242
H.-M. Yang, G.-Y. Peng / Ultrasonics Sonochemistry 17 (2010) 239–245
100
(a) 100
80
60
40
20
0
80
60
40
20
0
30
60
90
120 150 180 210 240 270
Time on stream (min)
0
(b)100
0
30
60
90
120
150 180 210 240 270
Time on stream (min)
80
Fig. 3. Effects of ultrasound and stirring speeds on reaction rates. Conditions: 0.25 M
of ArCOONa, 3.0 M of RBr, 1 mol of NaBr, 0.0125 mol of TBPB, toluene solvent, 70 °C,
flow rate (aq. or org. phase) at 0.3 cm3/min, 300 W of ultrasound; (h) 0 rpm 0 kHz;
(}) 0 rpm 28 kHz; (4)150 rpm 0 ; and (s) 150 rpm 28 kHz.
60
40
20
0
stirring and at silent condition, the product yield in the organic
outlet was 49.7% at steady state. With ultrasound irradiation, the
product yield increased to 58.2%, about a rise of 10%. For the case
of stirring at 150 rpm without ultrasound, the product yield was
65.1%, a little higher than that applying ultrasound alone. With
stirring and ultrasound irradiation, a synergic effect was attained
to get a product yield of 78.2%. The order of apparent reaction rates
kapp is 1ꢀ510 rpm 28 kHz (0.0215 minꢀ1) > 150 rpm 0 kHz
(0.0112 min ) > 0 rpm 28 kHz (0.00835 minꢀ1) > 0 rpm 0 kHz
(0.00593 minꢀ1), showing the third-liquid phase-transfer cata-
lyzed reaction significantly enhanced by ultrasound irradiation in
a flow system.
0
100
200
300
400
500
600
700
Time on stream (min)
Fig. 2. Performance of the continuous third-liquid catalyzed reaction: (a) repro-
ducibility test, (h) Run 1, (}) Run 2, (4) Run 3; (b) 10-h stability test. Conditions:
0.25 M of ArCOONa, 3.0 M of RBr, 1 mol of NaBr, 0.0125 mol of TBPB, toluene
solvent, 70 °C, flow rate (aq. or org. phase) at 0.3 cm3/min, 150 rpm, 28 kHz/300 W
of ultrasound.
4.3. Effects of power and frequency of ultrasound
third-liquid phase would be much consumed. The fraction of
QþðthirdÞ was lowered from 91.4% initially to 39.5% at 4 h and 22.7%
at 10 h. Within the reactor, the fraction of QþðorgÞ increased from
7.1%þinitially to 20.8% at 4 h and to 20.7% at 10 h, but the fraction
of QðaqÞ increased from 1.5% initially to 16.3% at 4 h and decreased
to 6.2% at 10 h. The total fraction of Q+ within the reactor was
76.6% at 4 h, and declined to 49.6% at 10 h on stream. However,
with the gradual reduction of the third-liquid volume, the concen-
tration of Q+ in the third-liquid phase still kept at a near constant
after the steady state reached and gave a steady productivity.
Besides, the catalyst entrained to the outlet streams can be easily
separated and fed back to the reaction system. The stability test
confirms the feasibility of ultrasound-assisted third-liquid phase-
transfer catalyzed reaction operated in a continuous two-phase-
flow mode.
Table 2 shows the effects of power and frequency of ultrasound
on the third-liquid catalyzed esterification. With stirring speed at
150 rpm, flow rate at 0.3 cm3/min and temperature at 70 °C under
300 W of ultrasound at different frequencies, the product yields at
steady state and apparent rate constant were 65.1% (silent,
kapp = 0.0112 minꢀ1), 78.2% (28 kHz, kapp = 0.0215 minꢀ1), 71.5%
(40 kHz, kapp = 0.0150 minꢀ1), 71.2% (50 kHz, kapp = 0.0148 minꢀ1
)
and 69.3% (80 kHz, kapp = 0.0135 minꢀ1), decreasing with the in-
crease of ultrasonic frequency. At 80 kHz of ultrasound irradiation,
the higher frequency resulted in a larger sonic resistance and loss
of energy during the sonic waves propagating through the med-
ium. The effect of power of ultrasound on the reaction was per-
formed at a frequency of 28 kHz. A high power of ultrasound
offers much energy into the medium to enhance the reaction. With
100 W of ultrasound, the product yield at steady state was 64.3%,
almost the same with 150 rpm of stirring speed at silent condition.
Table 2 also shows the distribution of catalyst Q+ in various
phases after 4-h on stream for different frequencies and powers
of ultrasound. With ultrasound (28 kHz, 300 W), a fraction 76.6%
of the initial added catalyst would stay within the reactor, and
the total catalyst flowing out of the reactor was 5.9% in the aqueous
stream and 17.5% in the organic stream. The overall fractions Q+/
QBr,0 in the outlet for 0, 24, 40, 50 and 80 kHz were 16.8%, 23.4%,
29.2%, 27.6%, and 24.1%, respectively. The fraction QþðthirdÞ=QBrð;0Þ re-
vealed that much Q+ was regenerated from the faster intrinsic reac-
tion in the third-liquid phase and then distributed into the aqueous
and organic phases. Similar trends for the distribution of Q+ were
4.2. Comparison of effects of ultrasound irradiation and stirring
Ultrasound irradiation can generate a high temperature at hot
spot and offers high energy dissipation at local medium to enhance
the reaction rate. In a multiphase system under sufficient stirring,
the resistance of mass-transfer between phases is commonly ne-
glected, and the overall reaction rate is expected to be reaction
control. The effects of ultrasound irradiation, stirring speeds and
their combination on the reaction rate were explored for flow rate
at 0.3 cm3/min using toluene as the solvent. The reaction simply
promoted by countercurrent mixing of two-phase flow was also
compared. The results are shown in Fig. 3 and Table 1. Without

H.-M. Yang, G.-Y. Peng / Ultrasonics Sonochemistry 17 (2010) 239–245
243
Table 1
Comparison of effects of ultrasound and stirring on the third-liquid catalyzed reaction.
Stirring speed
(rpm)
Ultrasonic frequency
(kHz)
Volume of third-liquid before
reaction (cm3)
Volume of third-liquid after
reaction (cm3)
Product yielda in organic
exit (%)
kapp
(10ꢀ2 minꢀ1
)
0
0
0
28
0
28
28
28
28
7.9
7.6
7.9
7.8
8.0
7.8
7.8
3.8
1.5
4.0
1.6
2.6
2.5
2.7
49.7
58.2
71.5
73.4
78.2
79.2
78.7
0.59
0.84
1.50
1.66
2.15
2.28
2.22
150
100
150
200
300
Conditions: 0.25 M of ArCOONa, 3.0 M of RBr, 1 mol of NaBr, 0.0125 mol of TBPB, temperature at 70 °C, flow rate (aqueous or organic phase) at 0.3 cm3/min, 300 W of
ultrasound.
a
Average value based on ArCOONa after 4 h of duration (at steady state).
Table 2
Effects of ultrasonic frequency and power on the distribution of catalyst after reaction.
Power
(W)
Frequency
(kHz)
QþðthirdÞ=QBr
reactor (%)
in
QþðaqÞ=QBr
(QþðorgÞ=QBr ) in
QþðaqÞ=QBr
outlet (%)
(QþðorgÞ=QBr ) in
Product yielda in organic
exit (%)
kapp
ð;0Þ
ð;0Þ
ð;0Þ
ð;0Þ
ð;0Þ
reactor (%)
(10ꢀ2 minꢀ1
)
0
300
300
300
300
100
200
0
28
40
50
80
28
28
56.8
39.5
39.0
41.5
45.3
49.3
45.2
5.7 (20.7)
16.3 (20.8)
10.6 (21.2)
7.7 (23.2)
9.0 (21.6)
6.7 (21.4)
7.3 (20)
4.8 (12.0)
5.9 (17.5)
9.6 (19.6)
8.4 (19.2)
7.6 (16.5)
5.7 (17.0
8.0 (19.5)
65.1
78.2
71.5
71.2
69.3
64.3
69.8
1.12
2.15
1.50
1.48
1.35
1.08
1.39
Conditions: 0.25 M of ArCOONa, 3.0 M of RBr, 1 mol of NaBr, 0.0125 mol of TBPB, stirring speed at 150 rpm, temperature at 70 °C, flow rate (aqueous or organic phase) at
0.3 cm3/min.
a
Average value based on ArCOONa after 4 h of duration (at steady state).
observed for different powers applied. At 100 W of power of ultra-
sound, the third-liquid phase in the reaction zone kept about 50%
of the catalyst Q+, showing the slower reaction rate in the reaction
zone.
300 W)-assisted third-liquid catalyzed reaction. These four cata-
lysts all can be used to form the third-liquid phase. The volume
of third-liquid phase were 8.0 cm3 for TBPB, 6.2 cm3 for TBAB,
6.7 cm3 for TBAI, and 6.6 cm3 for BTBAB, in the condition of
0.0125 mol of catalyst, 0.25 mol of NaBr, 0.25 M of ArCOONa, and
50 cm3 of toluene at 70 °C. The volume of third-liquid phase pro-
duced by TBPB was the largest. With the conditions of 0.25 M of
ArCOONa, 3.0 M of RBr, flow rate at 0.3 cm3/min, and temperature
at 70 °C the product yields at steady state and the apparent rate
constants estimated for TBPB, TBAB, TBAI, and BTBAB were 78.2%
(0.0215 minꢀ1), 61.7% (0.0097 minꢀ1), 62% (0.0098 minꢀ1), and
55.3% (0.0074 minꢀ1), respectively, as shown in Fig. 4. TBPB shows
the highest activity among the tested catalysts, because of the eas-
ieþst formation of ArCOOQ. After 4-h on stream, the fractions
QðthirdÞ=QBrð;0Þ were 39.5% for TBPB, 32% for TBAB, 37.2% for TBAI,
and 51.2% for BTBAB. Apparently, the steric hindrance by bulky
BTBAB reduced the transfer rate and the catalytic activity, leaving
the highest content of Q+ in the third-liquid phase. The tempera-
ture effect was tested using TBPB in the same conditions as de-
scribed above. The product yields and the rate constants for
50 °C, 60 °C, and 80 °C were 42.4% (0.00442 minꢀ1), 58.43%
(0.00843 minꢀ1), and 81.79% (0.027 minꢀ1), respectively. The
apparent activation energy by Arrhenius’ equation was 14.5 kcal/
mol and kapp = 2.96 Â 107 exp (ꢀ14,500/RT) for TBPB, implying
the controlling step being the intrinsic reaction.
4.4. Effect of phase-transfer catalyst and temperature
In the present flow system without using catalysts, the product
yield in the organic outlet at steady state was only 2.3%, exhibiting
that the reaction was conducted by interfacial contact of the two
immiscible reactants. Four catalysts, TBPB, TBAB, TBAI, and BTBAB
were compared for their activity in this ultrasound (28 kHz,
100
80
60
40
20
0
The catalytic behaviors for different amounts of TBPB were ex-
plored. The third-liquid phase was prepared with 0.25 mol of NaBr,
0.0125 mol of ArCOONa, 50 cm3 of water, and 50 cm3 of toluene at
70 °C. The volumes of third-liquid phase as-prepared and after 4-h
on stream were measured. The results are shown in Fig. 5. The vol-
ume of third-liquid phase increased with increasing the amount of
TBPB. Approximately a 65–69% reduction of volume in the third-li-
quid phase was observed; but for the case of 0.0075 mol of TBPB,
the third-liquid phase disappeared after a period of 4-h on stream.
The reduction in the volume of third-liquid phase resulted from the
0
30
60
90
120 150 180 210 240
270
Time on stream (min)
Fig. 4. Effects of phase-transfer catalysts on reaction rates. Conditions: 0.25 M of
ArCOONa, 3.0 M of RBr, 1 mol of NaBr, 0.0125 mol of PTC, toluene solvent, 70 °C,
150 rpm, flow rate (aq. or org. phase) at 0.3 cm3/min, 28 kHz/300 W of ultrasound;
PTC: (h) TBPB, (}) TBAI, (4)TBAB, and (s) BTBAB.

244
H.-M. Yang, G.-Y. Peng / Ultrasonics Sonochemistry 17 (2010) 239–245
were 48.0% (kapp = 0.00553 minꢀ1), 68.1% (0.0128 minꢀ1), and
10
8
71.1% (0.0147 minꢀ1), respectively. The small increments on the
reaction rate for higher amounts of TBPB were considerably related
to the formation of ArCOOQ within the reactor. The fractions of Q+
distributed in various phases before and after reaction, and Ar-
COOQ in the third-liquid phase for different amounts of TBPB are
shown in Fig. 5. Before the start of reaction, the distribution of
Q+ between phases wþ as not so distinctive for each amount of TBPB
with the fraction QðthirdÞ=QBrð;0Þ above 90% and ArCOOQ(third)/QBr,0
around 23–26% (except 8.87% for 0.0075 mol of TBPB). After 4-h
on stream, the third-liquid phase disappeared for 0.0075 mol of
TBPB, and a high fraction 62.9% of QþðorgÞ=QBrð;0Þ was distributed in
the organic outlet. Obviously, the consumption of ArCOOQ from
reacting with RBr was not compensated sufficiently, allowing the
disappearance of the third-liquid phase and to give a low product
yield (4þ8.0%). For the cases of TBPB at 0.00875 mol above, the frac-
tions QðthirdÞ=QBrð;0Þ kept at around 40% (39.5–42.5%), and the cata-
lyst retaining in the reactor was around 75–80%. Moreover, the
fraction QþðaqÞ=QBrð;0Þ in the reactor would increase with the in-
crease of TBPB, representing a faster intrinsic reaction in the
third-liquid phase to allow much QBr transferring into the aqueous
phase.
6
4
2
0
0.006
0.009
0.012
0.015
Amount of TBPB used (mol)
Fig. 5. Dependence of volume of third-liquid phase on the amount of TBPB.
Conditions: 0.25 M of ArCOONa, 3.0 M of RBr, 1 mol of NaBr, flow rate (aq. or org.
phase) at 0.3 cm3/min, toluene solvent, 150 rpm, 70 °C, 28 kHz/300 W of ultra-
sound, 4-h on stream; (4) before reaction and (s) after reaction.
4.5. Effect of molar ratio of RBr to ArCOONa
100
80
The feeding ratio of RBr to ArCOONa influenced the existence of
ArCOOQ in the third-liquid phase that in turn affected the intrinsic
reaction rate. The third-liquid phase was first formed and used to
conduct the reaction under ultrasound (28 kHz, 300 W) for various
concentrations (0.25, 1.0, 2.0, 3.0, and 4.0 M) of RBr in toluene. The
results are shown in Table 3. When RBr to ArCOONa was 1:1, the
product yield in the organic outlet was only 15.6% and
QþðthirdÞ=QBrð;0Þ still kept at 61.1%. Increasing the ratio of RBr to
ArCOONa up to 12:1 increased the product yield up to 78.2% but
with QþðthirdÞ=QBrð;0Þ reduced. For RBr to ArCOONa at 16:1, the prod-
uct yield diminished to 73.5% and QþðthirdÞ=QBrð;0Þ still reduced. It is
suspected that the faster reaction from a higher concentration of
RBr would make a lot of QBr transferring into the aqueous phase
and increase the out-flowing of catalyst (10% of QþðaqÞ=QBrð;0Þ). Other
ratios below 12, the catalysts flowing out were all less than 5.9%. In
addition, the inverse proportiþon between the volume of third-li-
quid phase and the fraction QðthirdÞ=QBrð;0Þ was observed. At a ratio
ofþRBr to ArCOONa below 1.0, the third-liquid phase kept a fraction
QðthirdÞ=QBrð;0Þ greater than 80% in the reactor, only 0.7% of QBr
flowing out of the reactor. Such phenomena can be applied to re-
tain the third-liquid phase within the reactor by properly adjusting
the reaction conditions.
Product Yield
60
40
ArCOOQ
20
0
0.006
0.009
0.012
0.015
Amount of TBPB (mol)
Fig. 6. Distributions of QBr and ArCOOQ in phases, and the product yield for various
amounts of TBPB. Conditions: 0.25 M of ArCOONa, 3.0 M of RBr, 1 mol of NaBr, flow
rate (aq. or org. phase) at 0.3 cm3/min, toluene solvent, 150 rpm, 70 °C, 28 kHz/
300 W of ultrasound; Symbols (solid: before reaction, blank: after reaction) (Â)
product yield,
(
) , , , (N4)
Qðþorg;outÞ=QBr;0 (+) Qðþaq;outÞ=QBr;0 (ds) QðþthirdÞ=QBr;0
ꢀ
QþðaqÞ=QBr;0, (jh) Qþ =QBr;0, (ꢁ) ArCOOQ(th)/QBr,0
.
ðorgÞ
4.6. Effect of organic solvent
consumption of ArCOOQ and the out-flowing of QBr. The degree of
increase in product yield was less significant as the employed TBPB
exceeding 0.01 mol. As shown in Fig. 6, for 0.0075, 0.00875, and
0.01 mol of TBPB, the product yields and apparent rate constants
The polarity of organic solvent is important in the formation of
third-liquid phase. The third-liquid phase is difficult to be formed
Table 3
Effect of concentrations of RBr and ArCOONa on the reaction rate after reaction.
RBr (M)
ArCOONa (M)
Product yielda (%)
QþðthirdÞ=QBr in reactor (%)
QþðaqÞ=QBr
in aq. outlet (%)
ð;0Þ
Volume of third-liquid (cm3)
kapp (10ꢀ2 minꢀ1
)
ð;0Þ
0.25
1.0
2.0
3.0
4.0
0.25
0.25
0.25
0.25
0.25
0.50
1.00
15.6
50.4
60.5
78.2
73.5
12.2
7.2
61.1
55.8
44.6
39.5
30.8
79.6
81.6
4.8
5.8
5.6
5.9
10
4.7
4.5
3.6
2.6
2.2
6.3
7.7
0.11
0.61
0.92
2.15
1.67
0.083
0.046
0.25
0.25
0.7
0.6
Conditions: 1 mol of NaBr, 0.0125 mol of TBPB, stirring speed at 150 rpm, temperature at 70 °C, flow rate (aqueous or organic phase) at 0.3 cm3/min, 28 kHz/300 W of
ultrasound.
a
Average value based on ArCOONa after 4 h of duration (at steady state).

H.-M. Yang, G.-Y. Peng / Ultrasonics Sonochemistry 17 (2010) 239–245
245
Table 4
Effect of organic solvent on the distribution of Q+ between phases.
Organic
solvent
QþðthirdÞ=QBr
before reaction (after QþðaqÞ=QBr
before reaction (after
QþðaqÞ=QBr
before reaction (after
ð;0Þ
reaction) (%)
Q+/QBr,0 in aq. outlet (org.
outlet) (%)
Product
ð;0Þ
ð;0Þ
reaction) (%)
reaction) (%)
yielda (%)
Toluene
n-Heptane
Isooctane
91.4 (39.5)
95.0 (67.6)
98.3 (79.2)
1.5 (16.3)
4.6 (11.1)
0.9 (4.1)
7.1 (20.8)
0.4 (3.7)
0.8 (3.3)
5.9 (17.5)
12.1 (5.5)
6.6 (6.8)
78.2
59.4
62.5
Conditions: 0.25 M of ArCOONa, 3.0 M of RBr, 1 mol of NaBr, stirring speed at 150 rpm, flow rate (aqueous or organic phase) at 0.3 cm3/min, 28 kHz/300 W of ultrasound,
temperature at 70 °C.
a
Average value based on ArCOONa after steady state reached.
was more significant. It was also observed that the time on stream
to reach the steady state almost corresponded to the space time, as
100
shown in Fig. 7, demonstrating the countercurrent two-phase-flow
in good stability. The volumes of the third-liquid phase after reac-
tion were 2.6, 1.5, 0.4, and 0.3 cm3 for flow rates at 0.3, 0.5, 0.7, and
1.0 cm3/min, respectively. This showed that the higher flow rate
made a much severe countercurrent mixing of phases, leading to
the entrained catalyst flowing out of reactor. Hence, the too high
flow rate not only lowered the contact time between reactants in
the third-liquid phase but also reduced the effectiveness of ultra-
sound irradiation.
80
60
40
20
0
5. Conclusions
A green technology in organic synthesis, the ultrasound-as-
sisted third-liquid phase-transfer catalysis, was developed in a
continuous two-phase-flow reactor for synthesizing butyl salicy-
late. With ultrasound irradiation (28 kHz, 300 W), space time at
168 min and stirring at 150 rpm, the product yield was greatly en-
hanced to 78.2%, a rise by 57.3% times higher than that without any
assistance of stirring and ultrasound. The molar ratio of organic to
aqueous reactant and the type of solvent play important roles in
forming and keeping the third-liquid phase in the continuous flow
reactor. A kinetic model was proposed to estimate the apparent
rate constants, and the apparent activation energy correlated by
Arrhenius’ equation for TBPB was 14.5 kcal/mol. The stability test
shows the feasibility of third-liquid catalyzed esterification in
phase-transfer catalysis assisted by ultrasound irradiation.
0
30
60
90
120 150 180 210 240
270
Time on stream (min)
Fig. 7. Effect of flow rate on the reaction rate. Conditions: 0.25 M of ArCOONa, 3.0 M
of RBr, 1 mol of NaBr, 0.0125 mol of TBPB, toluene solvent, 150 rpm, 70 °C, 28 kHz/
300 W of ultrasound; flow rate (aq. or org. phase) (cm3/min): (h) 0.3, (}) 0.5, (4)
0.7, (s) 1.
in a high-polarity solvent. Three types of solvents, toluene, n-hep-
tane, and isooctane were tested. The volumes of third-liquid phase
were 8.0 cm3 for toluene, 5.5 cm3 for n-heptane, and 5.4 cm3 for
isooctane using TBPB. As shown in Table 4, the fractions
QþðthirdÞ=QBrð;0Þ before reaction were all above 90%, especially
98.3% using isooctane; while the product yields were 59.4% for
n-heptane, 62.5% for isooctane and 78.2% for toluene. Although tol-
uene was used to dissolve much Q+ than using n-heptane and iso-
octane, the faster reaction still released much QBr into the organic
solvent and the aqueous phase. The order of reaction rate for sol-
vents was toluene > isooctane > n-heptane. It was quite consistent
with the reduction of QðþthirdÞ=QBrð;0Þ after reaction. A fraction 79.2%
of QþðthirdÞ=QBrð;0Þ after reaction was observed with isooctane, and
the order of fractions Q+/QBr,0 flowing out of the reactor was isooc-
tane (13.4%) < n-heptane (17.6%) < toluene (23.4%). Apparently,
isooctane has the potential to be used in ultrasound-assisted
third-liquid catalyzed esterification.
Acknowledgments
The authors acknowledge the financial support of the National
Science Council, Taiwan, ROC (Grant No. NSC 95-2221-E-005-
141-MY2). This work is also supported in part by the Ministry of
Education, Taiwan, ROC, under the ATU plan.
References
[1] C.M. Starks, C.L. Liotta, M. Halpern, Phase-transfer Catalysis: Fundamental,
Applications, and Industrial Perspectives, Chapman & Hall, New York, 1994.
[2] H.M. Yang, H.S. Wu, Catal. Rev.-Sci. Eng. 45 (2003) 463–540.
[3] D.H. Wang, H.S. Weng, Chem. Eng. Sci. 50 (1995) 3477–3486.
[4] G.D. Yadav, C.A. Reddy, Ind. Eng. Chem. Res. 38 (1999) 2245–2253.
[5] P.J. Lin, H.M. Yang, J. Mol, Catal. A: Chem. 235 (2005) 293–301.
[6] C.C. Huang, H.M. Yang, Appl. Catal. A: Gen. 290 (2005) 65–72.
[7] H.M. Yang, C.C. Huang, Ind. Eng. Chem. Res. 46 (2007) 7915–7920.
[8] J.L. Luche, Synthetic Organic Sonochemistry, Plenum Press, New York, 1998.
[9] A. Loupy, J.L. Luche, in: Y. Sasson, R. Neumann (Eds.), Handbook of Phase-
transfer Catalysis, Blackie Academic & Professional, London, 1997, pp. 369–
404.
[10] M.H. Entezari, A.A. Shameli, Ultrason. Sonochem. 7 (2000) 169–172.
[11] M.L. Wang, V. Rajendran, Ultrason. Sonochem. 14 (2007) 46–54.
[12] M.L. Wang, V. Rajendran, J. Mol, Catal. A: Chem. 273 (2007) 5–13.
[13] J.T. Li, X.L. Li, Ultrason. Sonochem. 14 (2007) 677–679.
[14] N.S. Nandurkar, M.J. Bhanushali, S.R. Jagtap, B.M. Bhanage, Ultrason.
Sonochem. 14 (2007) 41–45.
4.7. Effect of flow rate
The flow rates of aqueous and organic solutions affect the resi-
dence of reactants in the reaction zone. Fig. 7 shows the effect of
flow rate on the product yield in the organic outlet. The space
times for 0.3, 0.5, 0.7 and 1 cm3/min of flow rates were 168, 100,
72, and 50 min, respectively. The product yields and apparent rate
constants were 42% (0.0145 minꢀ1
)
for 1 cm3/min, 53.1%
(0.0159 minꢀ1) for 0.7 cm3/min, 61.6% (0.016 minꢀ1) for 0.5 cm3/
min, and 78.2% (0.0215 minꢀ1) for 0.3 cm3/min. Increasing the
space time up to 100 min only slightly increased the apparent reac-
tion rate. At a low flow rate, the effect of ultrasound on the reaction