Kinetic study of synthesizing bisphenol A diallyl ether in a phase-transfer catalytic reaction

Article abstract of DOI:10.1246/bcsj.79.80

In the synthesis of bisphenol A diallyl ether from bisphenol A and allyl bromide, the liquid-liquid mode of operation catalyzed by quaternary ammonium salts was carried out in an alkaline solution/organic solvent two-phase medium. The mono-substituted product was not detected during or after the reaction. In this work, a rational reaction mechanism is proposed and a kinetic model then established. The apparent rate constant of the organic-phase reaction was obtained from experimental data. The effects of the reaction conditions, including agitation speed, organic solvents, quaternary ammonium salts, inorganic salts, temperature, alkali compounds, amount of potassium hydroxide, and water on the conversion of allyl bromide, were investigated in detail. The presence of a small amount of tetrabutylammonium bromide (TBAB) in the chlorobenzene/water system produced a rate over several folds larger than that of the reaction system in the absence of phase-transfer catalyst. The presence of alkali compounds promoted the formation of alkoxide and enhanced the extractive efficiency due to the salting out effect. The present etherification via phase-transfer catalyst could operate at lower temperatures to avoid Claisen rearrangement, which usually occurs at higher temperatures. Rational reasons to account for the absence of mono-substituted product are explained satisfactorily. Peculiar phenomena in investigating the effects of the volume of organic solvent, amount of KOH, and the volume of water and alkali compounds on the apparent rate constants are also explained.

Full text of DOI:10.1246/bcsj.79.80

8
0
Bull. Chem. Soc. Jpn. Vol. 79, No. 1, 80–87 (2006)
ꢀ 2006 The Chemical Society of Japan
Kinetic Study of Synthesizing Bisphenol A Diallyl Ether
in a Phase-Transfer Catalytic Reaction
Maw-Ling Wang 1 and Ze-Fa Lee2
ꢀ
1
Department of Environmental Engineering, Hung Kuang University,
Shalu, Taichung County, Taiwan 433, Taiwan
2
Department of Chemical Engineering, National Chung Cheng University,
Ming-Hsiung, Chiayi County, Taiwan 621, Taiwan
Received February 4, 2005; E-mail: chmmlw@sunrise.hk.edu.tw
In the synthesis of bisphenol A diallyl ether from bisphenol A and allyl bromide, the liquid–liquid mode of oper-
ation catalyzed by quaternary ammonium salts was carried out in an alkaline solution/organic solvent two-phase medi-
um. The mono-substituted product was not detected during or after the reaction. In this work, a rational reaction mech-
anism is proposed and a kinetic model then established. The apparent rate constant of the organic-phase reaction was
obtained from experimental data. The effects of the reaction conditions, including agitation speed, organic solvents, qua-
ternary ammonium salts, inorganic salts, temperature, alkali compounds, amount of potassium hydroxide, and water on
the conversion of allyl bromide, were investigated in detail. The presence of a small amount of tetrabutylammonium
bromide (TBAB) in the chlorobenzene/water system produced a rate over several folds larger than that of the reaction
system in the absence of phase-transfer catalyst. The presence of alkali compounds promoted the formation of alkoxide
and enhanced the extractive efficiency due to the salting out effect. The present etherification via phase-transfer catalyst
could operate at lower temperatures to avoid Claisen rearrangement, which usually occurs at higher temperatures. Ra-
tional reasons to account for the absence of mono-substituted product are explained satisfactorily. Peculiar phenomena in
investigating the effects of the volume of organic solvent, amount of KOH, and the volume of water and alkali com-
pounds on the apparent rate constants are also explained.
Phase-transfer catalysis (PTC)1–3 is an effective technique
for conducting heterogeneous reactions where the reagents are
chips, photoresists, tough impact resistant prepregs, mouldings
with high breaking toughness for fibre reinforced structures,
composites with high heat, water, and chemical resistance,
and coatings with high heat, water, and chemical resistance.8
It is also widely employed as a cross-linking agent in epoxy
4
located in different continuous phases. In order for phase-
transfer catalysis to take place, the quaternary ammonium cat-
ion must effectively transport the reactant anion from the aque-
ous phase through the interfacial region into the organic phase5
for an ion-pair extraction of a normal phase-transfer catalysis
9,10
resin.
The objective of this paper is to synthesize bisphenol A
dially ether by a new path, i.e. phase-transfer catalysis. In
the present work, bisphenol A diallyl ether was synthesized
from the reaction of bisphenol A and allyl bromide catalyzed
by quaternary ammonium salts used as PTC in an alkaline
solution/organic solvent two-phase medium. Mono-substituted
product was not observed during or after the reaction. The
greatest advantage of using PTC to synthesize bisphenol A
diallyl ether via PTC is that the reaction can be carried out
at a lower temperature. Therefore, the Claisen rearrangement,
which usually occurs at higher temperatures, can be avoided.
From the experimental data, a rational reaction mechanism is
proposed and pseudo-first-order rate kinetics are developed.
The effects of the reaction conditions, such as agitation speed,
organic solvents, quaternary ammonium salts, alkali com-
pounds, amount of KOH, TBAB catalyst, volume of organic
solvent, inorganic salts, and temperature, on the conversion
of allyl bromide are investigated in detail. Peculiar phenomena
in investigating the effects of the volume of organic solvents,
the volume of water, and amount of KOH, and alkali com-
pounds on the apparent rate constants are explained.
(NPTC). Undoubtedly, PTC offers many substantial advantag-
es for the practical execution of numerous reactions. Its wide
6
2
application in organic synthesis is well deserved. NPTC reac-
tions generally include the Starks extraction mechanism and
Makosza interfacial mechanism.1
–3,6
The model of Starks ex-
traction mechanism argues that: (1) The main reaction takes
place in the organic phase, (2) the desired anion is transferred
to the organic phase, and (3) the catalyst is important for both
the reaction step and the transfer step. The quaternary ammo-
nium salt forms lipophilic ion pairs with the nucleophilic re-
agent (anion) entering the organic phase for further reactions.
In principle, the reaction at high temperature and the use of
protic and aprotic solvents can be avoided in phase-transfer
catalytic reactions. Thus, the byproducts obtained from side re-
actions at higher temperatures and the impact to the environ-
mental in using protic and aprotic solvents are prevented. The
phase-transfer catalytic process consequently has lower safety
risks and is environmentally friendlier.7
Bisphenol A diallyl ether (BPADAE) is a product of high
economic-value, which is widely used for high-technology ap-
plications, including epoxidised adhesives for semiconductor
Published on the web January 13, 2006; DOI 10.1246/bcsj.79.80

M.-L. Wang et al.
Bull. Chem. Soc. Jpn. Vol. 79, No. 1 (2006)
81
kaq,1
Experimental
+K−
−
+
HO
OH
+
2KOH
O
O K
+ 2H2O
+ 2Q+X−
KQX
0
Materials. The reactants: 4,4 -isopropylidenediphenol (bis-
phenol A) and allyl bromide; the solvents: benzene, chloroben-
zene, cyclohexane, cyclohexanone, 1,4-dioxane, dibutyl ether, and
toluene; the alkali compounds: potassium hydroxide and sodium
hydroxide; the catalysts (QX): trioctylmethylammonium chloride
kaq,2
+
−
−
+
2K+
−
+
−
−
+
Q O
O Q
+
X
K O
O K
KQOROQ
(aqueous phase)
ð2Þ
(
Aliquat 336), tetrabutylammonium bromide (TBAB), tetrabutyl-
(
organic phase)
ammonium chloride (TBAC), tetrabutylammonium hydrogensul-
fate (TBAHS), tetraethylammonium bromide (TEAB), tetraoctyl-
ammonium bromide (TOAB), and phenyltrimethylammonium bro-
mide (PTMAB), and other reagents were all guaranteed grade
k1
QO
OQ
+
C3H5Br
C3H5Br
QO
OC H
+
QX
QX
3
5
k2
(
G. R.) chemicals.
Procedures. Synthesis of Bisphenol A Diallyl Ether and Its
QO
OC3H5
+
H5C3O
OC3H5
+
The mechanism was formulated on the basis of Starks’ extraction
model.1,3 Thus, for this instance, the reactive anion was produced
in base-initiated reactions by proton extraction from the substrate
Separation: Measured quantities of bisphenol A (15 mmol), allyl
bromide (300 mmol), cyclohexane (50 mL), TBAB catalyst (0.32
g, 1 mmol), water (20 mL), and potassium hydroxide (16.8 g) were
charged in the reactor of a 250-mL three-neck Pyrex flask. The
mixed solution was stirred continuously at an agitation speed of
þ
ꢃ
and then the formed active catalyst ion-pair Q O C6H4C(CH3)2-
ꢃ
þ
C6H4O Q , which is an organic soluble compound, was then
transferred into the organic phase. During the reaction, the mono-
6
00 rpm using a Yamato LR400C stirrer equipped with a PTFE
þ
ꢃ
substituted product Q O C6H4C(CH3)2C6H4OC3H5 was not
detected. This phenomenon indicates that the value of the second
intrinsic rate constant in the organic phase, k2 is larger than that of
the first intrinsic rate constant, k1. Then, the mono-substituted
stirring blade. At the same time, the reactor was submerged in a
constant-temperature water bath, HAAKE refrigerating circulator
F6-C40, in which the temperature was controlled to within ꢁ0:1
ꢂ
ꢂ
C at 30 C. After 2.5 h of reaction, the two-phase solution was
þ
ꢃ
product Q O C6H4C(CH3)2C6H4OC3H5 further reacted with
allyl bromide to produce bisphenol A diallyl ether. Afterwards,
separated and the portion of the organic solution was washed five
times with an alkaline solution to remove TBAB catalyst using a
separatory funnel. Vacuum evaporation was carried out by using a
Yamato RE-440 equipped with a BM-400 water bath. When the
organic solvent (cyclohexane) was evaporated, the residual syrupy
liquid was the product, bisphenol A diallyl ether (BPADAE). The
product was identified by mass spectroscope (VG Quattro GC/
þ
ꢃ
Q X transferred from the organic phase to the aqueous phase
for further regeneration. [QOC6H4C(CH3)2C6H4OQ]o and [QX]o
were kept at constant values using a large excess of bisphenol A.
From the experimental data, no other by-products were observed.
In addition, independent experiment was carried out to demon-
strate that bisphenol A would hardly dissolve in pure water or in
the organic solvent. According to the proposed mechanism and the
two-film theory, the material balances for the regenerated QX and
the active catalyst, QOROQ, (i.e. QOC6H4C(CH3)2C6H4OQ), in
the organic and aqueous phases are:
1
MS/MS/DS) for molecular weight and NMR ( H NMR and
C NMR) for functional groups, respectively.
Kinetics of the Etherification: The reactor was a 150 mL
13
0
three-neck Pyrex flask. Predetermined quantities of 4,4 -isopro-
pylidenediphenol, allyl bromide, potassium hydroxide, naphtha-
lene (as internal standard), organic solvent, water, and phase-
transfer catalyst were introduced into the reactor. The reactor was
kept in an isothermal water bath at a constant known temperature
and mechanically agitated by an electric motor powered agitator.
For a kinetic run, a sample (approximately 0.3 mL) was withdrawn
from the mixed solution at a predetermined time. The withdrawn
sample was then immediately added to 5 mL of cool methanol to
quench the reaction. High performance liquid chromatography
ꢀ
ꢁ
ð3Þ
d½QOROQꢄ
dt
½QOROQꢄo
O
þ
ꢃ
ꢃ
þ
¼
KQOROQA ½ Q ORO Q ꢄ ꢃ
a
MQOROQ
ꢃ k1½C3H5Brꢄ ½QOROQꢄ ;
o
o
þ
ꢃ
ꢃ
þ
a
d½ Q ORO Q ꢄ
dt
þ
ꢃ
ꢃ
þ
þ
ꢃ 2
¼
kaq,2½ K ORO K ꢄ ½Q X ꢄ
a
a
ꢀ
ꢁ
½
QOROQꢄ
o
þ
ꢃ
ꢃ
þ
ꢃ KQOROQAf ½ Q ORO Q ꢄ ꢃ
;
ð4Þ
ð5Þ
a
M
QOROQ
(
HPLC) was carried out with a Shimadzu SPD-10AVP using
d½QXꢄ
o
¼
k1½C3H5Brꢄ ½QOROQꢄ
o
o
Glass VP 5.0 analysis software with UV wavelength of 204 nm.
An RP-18e (5 mm) column (Applied Merck Co.) was used to sepa-
rate the components and to analyze the withdrawn sample exper-
imentally.
Reaction Kinetics and Mechanism. A total reaction for this
present system is
dt
þ k2½C3H5Brꢄ ½QOROC3H5ꢄ
ꢃ KQXAð½QXꢄ ꢃ MQX½Q X ꢄ Þ;
o
o
þ
ꢃ
a
o
þ
ꢃ
a
d½Q X ꢄ
dt
þ
ꢃ
¼
KQXAfð½QXꢄ ꢃ MQX½Q X ꢄ Þ
o
a
þ
ꢃ
ꢃ
þ
þ
ꢃ 2
ꢃ 2kaq,2½ K ORO K ꢄ ½Q X ꢄ ;
ð6Þ
ð7Þ
a
a
QX
2
KOH + 2C3H5Br + HO
OH
Q ¼ V ð2½QOROQꢄ þ ½QOROC H ꢄ þ ½QXꢄ Þ
0
o
o
3
5 o
o
þ
ꢃ
ꢃ
þ
þ
ꢃ
ð1Þ
þ Vað2½ Q ORO Q ꢄ þ ½Q X ꢄ Þ;
a
a
H5C3O
OC3H5 + 2H2O + 2KBr
where f is the volume ratio of the organic solution (Vo) to the
aqueous solution (Va). R denotes the -C6H4C(CH3)2C6H4- group.
The subscripts ‘‘o’’ and ‘‘a’’ denote the characteristics of the spe-
cies in the organic and aqueous phases, respectively. k1 and k2 are
the intrinsic rate constants of the two sequential reactions in the
organic phase. kaq,1 and kaq,2 are the intrinsic rate constants of the
two ionic reactions in the aqueous phase. A which is the interfacial
According to the experimental observation, mono-substituted
þ
ꢃ
product Q O C6H4C(CH3)2C6H4OC3H5 was not found. Only the
di-substituted product (bisphenol A diallyl ether) H5C3OC6H4C-
(CH3)2C6H4OC3H5 was produced from the reaction solution.
For this, a rational reaction mechanism is proposed as follows:

8
2
Bull. Chem. Soc. Jpn. Vol. 79, No. 1 (2006)
Synthesizing Bisphenol A Diallyl Ether in PTC
area between two phases is defined as the interfacial area between
Q0
fMQOROQ
2
3
the two phases/volume of organic solution, i.e. cm /cm . Q0 is
the total catalyst in the reaction solution. KQOROQ and KQX are
the mass-transfer coefficients of QOROQ and QX between two
phases, i.e. (cm/s). MQOROQ and MQX are the distribution coeffi-
cients of QOROQ and QX between two phases, respectively, i.e.
½QOROQꢄo ¼
:
ð16Þ
2Vo ð1 þ fMQOROQÞ
Material balances for the compounds in the reaction solution
are:
d½C3H5Brꢄ
dt
o
ꢃ
¼ k1½C3H5Brꢄ ½QOROQꢄ
o
o
½
QOROQꢄ
o,s
MQOROQ ¼
;
ð8Þ
ð9Þ
þ
ꢃ
ꢃ
þ
½
Q ORO Q ꢄ
þ k2½C3H5Brꢄo½QOROC3H5ꢄo;
ð17Þ
ð18Þ
a,s
½
QXꢄ
d½QOROC H ꢄ
o,s
3
5 o
MQX ¼
;
¼ k ½C H Brꢄ ½QOROQꢄ
þ
ꢃ
1
3
5
o
o
½
Q X ꢄ
dt
s,s
ꢃ k2½C3H5Brꢄ ½QOROC3H5ꢄ ;
d½H5C3OROC3H5ꢄo
where the subscript ‘‘s’’ represents the equilibrium condition of the
species at the interface.
o
o
¼
k2½C3H5Brꢄ ½QOROC3H5ꢄ : ð19Þ
o
o
On the basis of the experimental observation, the concentra-
tions of QOROQ and QX in the organic and aqueous phases
reached constant values at the beginning of the reaction.1
Therefore, a pseudo-steady-state hypothesis (PSSH) is applied, i.e.
dt
The mono-substituted product QOROC3H5 was not observed dur-
ing or after the reaction. Therefore, a pseudo-steady-state hypoth-
esis (PSSH) is applied, i.e.
1,12
þ
ꢃ
ꢃ
þ
a
d½QOROQꢄo
d½ Q ORO Q ꢄ
d½QOROC H ꢄ
3
5 o
¼
0;
¼ 0;
¼
0:
ð20Þ
dt
dt
dt
þ
ꢃ
d½QXꢄo
dt
d½Q X ꢄ
dt
a
¼
0;
¼ 0:
ð10Þ
Thus, Eq. 18 becomes
k1
k2
Combining Eqs. 3–7 and 10, we obtain
½QOROC H ꢄ ¼ ½QOROQꢄo:
ð21Þ
3
5 o
ꢀ
Q0
Vo
2
k1½C3H5Brꢄo
KQXA
¼
½QOROQꢄo 2 þ
Substituting Eq. 21 into Eq. 17, and applying Eq. 16, we obtain
fMQOROQ
ꢁ
d½C3H5Brꢄ
o
2
k1½C3H5Brꢄo
ꢃ
¼ 2kapp½C H Brꢄ ;
3
5
ð22Þ
ð23Þ
o
þ
dt
fKQOROQA
(
kapp ¼ k ½QOROQꢄ ¼ k ½QOROC H ꢄ
ꢀ
ꢁ
1
o
2
3
5 o
1
=2
fk1½C3H5Brꢄ
1
=2
o
k1 fMQOROQ Q0
þ ½QOROQꢄo
MQX
þ
ꢃ
ꢃ
þ
¼ 1
:
kaq,2½ K ORO K ꢄ
a
þ fMQOROQ 2Vo
Integrating Eq. 22 yields
)
ꢀ
ꢁ
1=2
k1½C3H5Brꢄ
fkaq,2½ K ORO K ꢄ
o
þ
:
ð11Þ
þ
ꢃ
ꢃ
þ
a
½C3H5Brꢄ
o
¼
expðꢃ2kapptÞ:
ð24Þ
The following Damkohler numbers, DaQOROQ and DaQX are
defined as,
½C3H5Brꢄo,i
The subscript ‘‘i’’ represents the initial concentration of the spe-
cies. Eq. 24 is rewritten as
k1½C3H5Brꢄo
KQOROQA
DaQOROQ ¼
;
ð12Þ
ð13Þ
ꢃ lnð1 ꢃ XÞ ¼ 2kappt;
where X is defined as the conversion of C H Br, i.e.
ð25Þ
ð26Þ
k1½C3H5Brꢄo
KQXA
DaQX ¼
:
3
5
½
½
C3H5Brꢄ
o
Ry is defined as the ratio of the organic-phase reaction rate to the
aqueous-phase reaction rate, i.e.
X ¼ 1 ꢃ
;
C3H5Brꢄ
o,i
in which [C3H5Br]o,i denotes the initial concentration of allyl bro-
mide in the organic phase. Thus, the value of kapp can be obtained
from experimental data in conjunction with Eq. 25.
k1½C3H5Brꢄ
o
Ry ¼
:
ð14Þ
þ
ꢃ
ꢃ
þ
a
kaq,2½ K ORO K ꢄ
Introducing Eqs. 12–14 into Eq. 11 yields a dimensionless form,
ꢀ
½QOROQꢄo 2 þ
(
ꢁ
Results and Discussion
Q0
Vo
2
2
¼
þ f DaQOROQ þ DaQX
In the present work, bisphenol A diallyl ether was synthe-
sized from the reaction of allyl bromide and bisphenol A cat-
alyzed by phase-transfer catalysis in an alkaline aqueous solu-
tion/organic solvent two-phase medium. Two sequential sub-
stitution reactions in the organic phase were carried out. Only
the final product bisphenol A diallyl ether was observed in
the organic phase. The mono-substituted product QOC6H4C-
fMQOROQ
)
ꢀ
ꢁ
1=2
Ry
1=2
MQX fR 1y =2
þ ½QOROQꢄo
þ
:
ð15Þ
f
Several independent experiments were carried out to measure the
concentrations of QOROQ in the aqueous and organic phases. It
was found that the concentration of QOROQ in the organic phase
remained at a constant value after one minute of reaction. This re-
sult indicates that the values of KQOROQA and KQXA are all larger
than that of k1[C3H5Br]o. Furthermore, the rate of aqueous ion-
exchange is more rapid than that of the organic phase. Thus, the
value of the aqueous-phase rate constant kaq,2 is larger than that of
the organic-phase rate constant k1. For this, Eq. 15 is simplified to
(
CH3)2C6H4OC3H5 (QOROC3H5) in the organic phase was
not obtained. This result indicates that the reactivity of QORO-
C3H5 is greater than that of QOROQ. Furthermore, this result
indicates that two hydroxy groups of each bisphenol A mole-
cule both react with quaternary ammonium salt and KOH to
form QOROQ. In order to find out the optimum reaction con-

M.-L. Wang et al.
Bull. Chem. Soc. Jpn. Vol. 79, No. 1 (2006)
83
1
0
8
6
4
2
0
7
6
5
4
3
2
1
kapp
kapp
0
0
200
400
600
800
1000
1200
0
20
40
60
80
100
120
Agitation speed / rpm
Volume of Organic Solvent / mL
Fig. 1. Effect of the agitation speed on the apparent rate
constant, kapp; 15 mmol of allyl bromide, 25 mmol of bis-
phenol A, 20 mL of chlorobenzene, 1 mmol of TBAB cat-
Fig. 2. Effect of the amount of organic solvent on the ap-
parent rate constant, kapp; 15 mmol of allyl bromide, 25
mmol of bisphenol A, 1 mmol of TBAB catalyst, 20 g of
ꢂ
alyst, 20 g of KOH, 40 mL of water, and 30 C.
ꢂ
KOH, 40 mL of water, 1000 rpm, and 30 C.
ditions, agitation speed, amount of organic solvents, phase-
transfer catalysts, potassium hydroxide, water, organic sol-
vents, alkali compounds, quaternary ammonium salts, inorgan-
ic salts, and reaction temperature were all considered. The ef-
fects of the operating conditions are discussed and summarized
below.
Effect of Agitation Speed. In the present study, experi-
ments were carried out to examine the effects of agitation
speed on the apparent rate constants. The values of the appa-
rent rate constants kapp under various agitation speeds are giv-
en in Fig. 1. The agitation speed had no significant influence
on the apparent rate constant in the range of 200–1200 rpm.
For agitation speed less than 200 rpm, the conversion is highly
affected by the agitation speeds in which the mass transfer and
chemical reaction both play important role. The conversion of
bisphenol A at 200–600 rpm was slightly lower than that in the
range of 800–1200 rpm. Thus, the agitation speed was set at
increase in the volume of organic solvent for 10–40 mL. For
a further increase in the volume of organic solvent, the dilution
effect becomes more important. The concentration of QOROQ
in the organic solvent is then decreased with an increase in
volume of organic solvent. Thus, the distribution coefficient
of QOROQ is then decreased when the volume of organic sol-
vent is larger than 40 mL. The main reason is that the volume
of the organic phase affects the environment of the reaction,
i.e., the concentration of the organic-phase reactant and the
concentration of the internal standard. In addition, the capacity
of the catalyst in the organic phase is also influenced by the
volume of the organic phase. Therefore, the distribution coef-
ficient of the active catalyst between two phases is affected by
the volume of the organic phase. The distribution coefficients
MQOROQ between organic and aqueous phases (40 mL of water)
are 0.60, 0.84, 7.57, 1.06, and 0.49, for 10, 20, 40, 80, and
120 mL of organic solvent, respectively. It is obvious that
the concentration of the active catalyst [QOROQ]o in the
organic phase is first increased when the volume of organic
solvents rose up to 40 mL. A further increase in the volume
of organic solvents reduced the concentration of QOROQ.
Therefore, there was a maximum in the apparent rate constant
at 40 mL of organic solvents. For this reason, Figure 2 exhibits
an optimal value of organic solvents that maximizes the con-
centration of active catalytic species.
1
000 rpm to eliminate the mass-transfer resistance and to as-
sess the effect of other variables on the reaction rate in the fol-
lowing experiments. The concentration of the active catalyst
[QOROQ]o in the organic phase was constant. In other words,
the rate of mass transfer is larger than that of reaction in the
organic phase.
Effect of Volume of Organic Solvents. In this study,
chlorobenzene was chosen as the organic solvent. The results
are shown in Fig. 2. The optimal volume of organic solvents
to obtain a maximum value of the apparent rate constant was
observed. The apparent rate constant is first increased and then
decreased with the increase in the volume of organic solvents.
The phenomenon can be explained with the aid of Eqs. 16 and
Effect of Quaternary Ammonium Salts. Quaternary am-
monium salts are generally used as phase-transfer catalysts to
promote reaction rate. In addition to TBAB, six other quater-
nary ammonium salts, such as TBAC, TBAHS, TEAB, TOAB,
Aliquat 336, and PTMAB, were investigated to test their reac-
tivities. The experimental results are listed in Table 1. The re-
2
3. In addition, the conditions for measuring the distribution
ꢃ
ꢃ
coefficients of QOROQ between organic phase and aqueous
phase are the same as those of the reaction conditions, i.e. bis-
phenol A, KOH, and QBr are all introduced to the measuring
vessel. Under this circumstance, KOROK in a large quantity
and QOROQ in a limited quantity are all produced in the aque-
ous phase. Then, the hydrophilic KOROK drives the hydro-
phobic QOROQ to the organic phase from the aqueous phase
in the region of 10 to 40 mL of organic solvent. Thus, more
QOROQ remains in the organic phase when the volume of or-
ganic phases is increased in the region of 10 to 40 mL. For this,
the distribution coefficient of QOROQ is increased with an
activity of the anion, ORO , depends on its degree of hydra-
tion and on the structure of its counter cation. Comparing
TEAB, TBAB, and TOAB reveals that the more lipophilic
the quaternary ammonium cation, the greater the effectiveness
in transferring nucleophilic anion into the organic media is. In
other words, the catalytic activities are mainly due to the solu-
þ
ꢃ
ꢃ
þ
bilities of their ion-pairs Q ORO Q in the organic phase,
which in turn can be attributed to the nature and bulkiness
þ
of Q and the properties of the medium. As shown in Table 1,
the reactivities of TBAB, TBAC, and TBAHS are not affected
significantly by the anions, X , with the symmetric tetrabutyl-
ꢃ

8
4
Bull. Chem. Soc. Jpn. Vol. 79, No. 1 (2006)
Synthesizing Bisphenol A Diallyl Ether in PTC
Table 1. Effects of the Phase-Transfer Catalysts on the
aÞ
1.0
0
5
g
g
Apparent Rate Constant (kapp)
0
0
0
0
0
.8
.6
.4
.2
.0
10 g
20 g
3
ꢃ1
Catalyst
kapp/10 min
30 g
35 g
40 g
TBAB
TBAC
TBAHS
TEAB
TOAB
6.59
6.59
5.67
3.22
9.90
7.84
2.30
Aliquat 336
PTMAB
0
20
40
60
80
100
120
140
a) Reaction conditions: 15 mmol of allyl bromide, 25 mmol of
bisphenol A, 3 mmol of internal standard (naphthalene), 20
mL of chlorobenzene, 20 g of KOH, 40 mL of water, 1000
Time / min
Fig. 4. Kinetics for studying the effect of the amount of po-
tassium hydroxide on the conversion; 15 mmol of allyl
bromide, 25 mmol of bisphenol A, 20 mL of chloroben-
zene, 1 mmol of TBAB catalyst, 40 mL of water, 1000
ꢂ
rpm, and 30 C.
3
3
2
2
1
1
5
0
5
0
5
0
5
0
ꢂ
kapp
rpm, and 30 C.
which the apparent rate constant is in proportion to the initial
amount of catalyst added. In the absence of TBAB catalyst,
6% of the conversion of allyl bromide was obtained after
150 min. Nevertheless, 36% of the conversion was obtained
only within 10 min of reaction when 6 mmol of TBAB catalyst
was added to the solution. It shows that the phase-transfer cat-
alyst is indeed capable of promoting the etherification effec-
tively.
3
0
1
2
3
4
5
6
Effect of the Amount of Potassium Hydroxide. The ef-
fect of the amount of potassium hydroxide on the reaction con-
version of etherification was investigated and the results are
shown in Fig. 4. No etherification was observed in the absence
of potassium hydroxide (only 1% conversion after 150 min).
However, the reaction was enhanced by adding KOH to the re-
action solution. Nevertheless, the conversion of the reactant
did not change significantly for the change in the amount of
KOH from 5 to 20 g. The addition of a large excess of potas-
sium hydroxide, 30 g, dramatically accelerated the reaction.
From the conversion of allyl bromide monitored by high per-
formance liquid chromatography (HPLC), it was confirmed
that etherification proceeded smoothly to reach almost 100%
conversion within 150 min. The reactivity of the anion,
TBAB / mmol
Fig. 3. Effect of the amount of TBAB catalyst on the appa-
rent rate constant, kapp; 15 mmol of allyl bromide, 25
mmol of bisphenol A, 20 mL of chlorobenzene, 20 g of
ꢂ
KOH, 40 mL of water, 1000 rpm, and 30 C.
þ
ammonium cation (Q ). However, the sulfate ion of TBAHS
ꢃ
catalyst is relatively weaker in basicity than those of Br
ꢃ
and Cl of TBAB and TBAC catalysts. Therefore, the dissoci-
ꢃ
ation of HSO4 from TBAHS is more difficult than those of
TBAB and TBAC catalysts. For this, the reactivity of TBAHS
is less than those of TBAB and TBAC catalysts. Comparing
PTMAB and TEAB shows that the symmetrical molecules
possess high reactivity although the number of carbon atoms
of PTMAB is greater than that of TEAB. It is because the
phenyl group is much less lipophilic than the corresponding
aliphatic straight chain containing the same number of carbon
atoms and the steric hindrance of phenyltrimethylammonium
ꢃ
ꢃ
ORO , depended both on its degree of hydration and on its
association with the quaternary ammonium cation. The degree
of hydration was affected by the presence of KOH in the aque-
ous phase. The activity of the transferred anion was thus en-
hanced compared with its reactivity in the aqueous media as
þ
13
cation. The activity of the lipophilic cation Q is determined
its degree of hydration was reduced. It is interesting to note
mainly by two factors: its extractability and the anion activa-
tion ability. Structural factors affect the formation of active
catalyst cation–anion pairs between the organic phase and
aqueous phase. Based on the above argument, the order of the
reactivities of these quaternary ammonium salts are: TOAB >
Aliquat 336 > TBAB > TBAC > TBAHS > TEAB > PTMAB.
Effect of the Amount of TBAB Catalyst. The effect of
the amount of TBAB catalyst on the reaction was studied by
conducting six experiments. As shown in Fig. 3, the apparent
rate constant (kapp) is plotted against the TBAB catalyst
amount ranging from 0 to 6 mmol. It is obvious that the appa-
rent rate constant (kapp) increased linearly with the amount of
TBAB catalyst. This result is in agreement with Eq. 23 in
that the conversion decreased conspicuously after adding 40 g
of potassium hydroxide. It is because the solution was saturat-
ed by the addition of 40 g of potassium hydroxide and 25 mmol
of bisphenol A to 40 mL of water, leading to the salting out of
bisphenol A as a white gel. Under this circumstance, the ether-
ification system would be changing from a liquid–liquid PTC
to a liquid–solid (bisphenol A) PTC. The change would de-
crease the reaction rate sharply. As shown in Fig. 5, the con-
version of allyl bromide is first increased with an increase in
the amount of KOH up to 30 g and then decreased with any
further increase in the amount of KOH.
Effect of the Amount of Water. The effect of the volume
of water on the apparent rate constant (kapp) was investigated.

M.-L. Wang et al.
Bull. Chem. Soc. Jpn. Vol. 79, No. 1 (2006)
85
Table 2. Effects of the Organic Solvents on the Apparent
aÞ
1
0
0
0
0
0
.0
.8
.6
.4
.2
.0
Rate Constants (kapp)
N
T
3
ꢃ1
Organic solvent
"
E
kapp/10 min
Dibutyl ether
Toluene
Benzene
2.8
2.4
2.3
5.6
2.2
2.0
18.3
0.07
0.10
0.11
0.19
0.16
0.01
0.28
3.57
4.49
4.13
6.59
5.21
3.58
4.93
Chlorobenzene
1,4-Dioxane
Cyclohexane
Cyclohexanone
0
10
20
30
40
a) Reaction conditions: 15 mmol of allyl bromide, 25 mmol of
bisphenol A, 3 mmol of internal standard (naphthalene), 20
mL of organic solvent, 1 mmol of TBAB catalyst, 20 g of
Amount of KOH / g
Fig. 5. Effect of the amount of potassium hydroxide on the
conversion of allyl bromide; 15 mmol of allyl bromide, 25
mmol of bisphenol A, 20 mL of chlorobenzene, 1 mmol of
ꢂ
KOH, 40 mL of water, and 30 C.
ꢂ
TBAB catalyst, 40 mL of water, 1000 rpm, and 30 C.
1
0
0
0
0
.0
.8
.6
.4
.2
NaHCO3
NaOH
KOH
7
6
5
4
3
2
1
0
kapp
0.0
0
20
40
60
80
100
120
140
Time / min
0
10
20
30
40
50
60
70
80
90
Volume of Water / mL
Fig. 7. Effect of the alkaline solution on the conversion of
allyl bromide; 15 mmol of allyl bromide, 25 mmol of bis-
phenol A, 20 mL of chlorobenzene, 1 mmol of TBAB, 357
equi-mmol of alkaline compound, 40 mL of water, 1000
Fig. 6. Effect of the volume of water on the apparent rate
constant, kapp; 15 mmol of allyl bromide, 25 mmol of bis-
phenol A, 20 mL of chlorobenzene, 1 mmol of TBAB cat-
ꢂ
rpm, and 30 C.
ꢂ
alyst, 20 g of KOH, 1000 rpm, and 30 C.
As shown in Fig. 6, there is an optimal volume of water in this
etherification. The apparent rate constant increased with an in-
crease in the volume of water (0–20 mL) and then decreased
with a further increase in the volume of water. A limited
amount of water (0–20 mL) formed a local saturated water
layer around the reacting species and consequently the reaction
rate increased because of the increase in interface area by add-
ing 0 to 20 mL of water. When an adequate amount of water
was added, all of the reactants in the aqueous phase, bisphenol
A and potassium hydroxide, dissolved. Therefore, all of the re-
agents facilitate the reaction in the optimum fraction between
the aqueous phase and the organic phase. Further increasing
the amount of water thus reduced both the alkaline concentra-
tion in the aqueous phase and the concentration of active cat-
alyst in the organic phase, because a greater amount of active
catalysts remained in the aqueous phase. Thus, the apparent
rate constant (kapp) is decreased with an increase in the amount
of water of more than 20 mL.
Effect of Organic Solvents. In the phase-transfer catalytic
reaction, the solvent dramatically influences the reactive activ-
ity. Dibutyl ether, toluene, benzene, chlorobenzene, 1,4-di-
oxane, cyclohexane, and cyclohexanone were used to investi-
gate the influence of organic solvents on reactivity. The results
are shown in Table 2. The order of the reactivities for these or-
ganic solvents was: dibutyl ether ; cyclohexane < benzene <
toluene < cyclohexanone < 1,4-dioxane < chlorobenzene. In
this reaction system, organic solvents are used to dissolve the
catalyst and reactive reagent. However, the active catalyst and
the reacting molecules in the solution often become inactive
because they are always surrounded by a number of solvent
molecules. In general, the more polar the solvent, the more
it can strip the bound water away from the catalyst. High sol-
vation to the active catalyst and reacting molecules often has a
negative effect on the acceleration of etherification. Further-
more, organic solvents also affect the distribution of the active
catalyst between the two phases and the chemical environment
of the reaction system. Therefore, the effect of organic solvents
on the reaction is complicated. It is not simple to predict
their effects simply by using only the dielectric constant or
Dimroth–Reichardt parameter of organic solvents.
Effect of the Alkali Compounds. The effect of different
alkali compounds, such as NaOH, KOH, and NaHCO3, on
the reaction rate was examined. As shown in Fig. 7, the ether-
ification rate is enhanced owing to the increase in extractive
efficiency of bisphenoxides by increasing the base concentra-
tion. In principle, the alkali compounds affect the distribution
of the active catalyst between the two phases, and the solu-
bility of bisphenol A in the aqueous phase. Both sodium hy-
droxide and potassium hydroxide exhibit strong basicity, and
sodium hydrogencarbonate exhibits weaker basicity. As shown

8
6
Bull. Chem. Soc. Jpn. Vol. 79, No. 1 (2006)
Synthesizing Bisphenol A Diallyl Ether in PTC
in Fig. 7, the reaction rate when using sodium hydroxide was
much lower than when using potassium hydroxide. The reason
is that the white gel (solid bisphenol A) was salted out when
using NaOH during the reaction, i.e. the solution was oversat-
urated when using 40 mL of water to dissolve 357 equi-mmol
of sodium hydroxide and 25 mmol of bisphenol A. Thus, for
convenience, potassium hydroxide was used in all experimen-
tal runs of the present study. The diether compound could be
obtained using a concentrated alkaline solution (NaOH and
KOH). However, the stabilities of these catalysts decreased
considerably in an aqueous–organic two-phase system with a
strong alkaline aqueous solution. Other highly efficient phase-
transfer catalysts, e.g. cryptands and crown ethers, are chemi-
cally stable in the presence of concentrated alkaline aqueous
solution, but their relatively high costs and/or toxicity limit
phase. Comparing the results for KCl, NaCl, KBr, and NaBr
with the same anion gives the order of the conversion of allyl
bromide as: KCl > NaCl and KBr > NaBr. The reason is that
the presence of sodium cations causes their salting out as the
white gel. From this viewpoint, the addition of sodium cations
is more unfavorable than the addition of potassium cations.
Moreover, comparing the results for KCl, KBr, NaCl, and
NaBr with the same cation gives the order of the conversion
of allyl bromide as: KBr > KCl and NaBr > NaCl. The reason
is that hydrophilic complexes associate more strongly with a
chloride ion among halide ions. In other words, the existence
of bromide ions would make the aqueous phase more lipophil-
ic than the existence of chloride ions which would make the
aqueous phase more compatible with the organic solvents.
Effect of Temperature. Etherification was studied at dif-
ferent temperatures. The results are shown in Fig. 9. The con-
version was found to increase substantially with increasing
temperature. The reaction follows the pseudo-first-order rate
law for all temperatures. At higher temperatures beyond 30
1
4–18
their usage.
On the other hand, quaternary ammonium
salts are widely available and inexpensive. For this reason,
quaternary ammonium salts are quite often the phase-transfer
catalysts of choice, even in strongly alkaline aqueous solutions
regardless of their low stability.1
9
ꢂ
C, allyl bromide was slightly evaporated and the kinetic data
is unreliable. In addition, Claisen rearrangement is an intra-
molecular thermal rearrangement at higher temperature, as
Effect of the Inorganic Salts. In phase-transfer catalytic
reactions, the inorganic salts added sometimes influence the
conversion. To investigate the effect of inorganic salts on
etherification, potassium bromide, potassium chloride, sodium
bromide, and sodium chloride were added to the reaction sys-
tem. In general, the hydration levels of all ions present in the
system tend to decrease when the salt concentration in the
aqueous phase is increased. The added salt ties up water mole-
cules for the dehydrating effect on the ions present in the sys-
ꢃ1
shown in Scheme 1. Plotting lnðkappÞ against T gives an ap-
ꢃ1
parent activation energy of 65.6 kJ mol from the Arrhenius
equation, as shown in Fig. 10. The apparent activation energy
confirmed that etherification is kinetically controlled.
Conclusion
Preparation of bisphenol A diallyl ether from the reaction of
biphenol A and allyl bromide using phase-transfer catalyst was
achieved successfully at room temperature and effectively
avoided the occurrence of Claisen rearrangement. The reaction
rate is greatly enhanced by adding a small quantity of phase-
transfer catalyst to the reaction system. The kinetic model
was constructed and the mass transfer and phase equilibrium
of the catalysts between the two phases were also constructed
by the two-film theory. Under the appropriate conditions, the
conversion reached as high as 100%. The reaction was found
to be kinetically controlled. Chlorobenzene was the best organ-
ic solvent for the present system of the etherification reaction
and tetraoctylammonium bromide (TOAB) was the best cata-
1
3
tem. Furthermore, the high-concentration inorganic salt in
the aqueous phase salts out the quaternary onium salts into
the organic phase and thereby alters the distribution coeffi-
cients of the phase-transfer catalysts. Thus, the inexpensive
inorganic salts have the potential of promoting reaction rate
when compared to the same reaction conditions without in-
organic salts. Unfortunately, as shown in Fig. 8, adding in-
organic salt did not increase the reaction rate. Instead a slight
decrease in reaction rate was observed in this etherification
system. It can be explained from the reaction mechanism that
the presence of more inorganic salts does not favor the forma-
þ
ꢃ
ꢃ
þ
tion of the active catalyst, Q ORO Q , in the aqueous
1
0
0
0
0
0
.0
.8
.6
.4
.2
.0
KBr
NaBr
NaCl
KCl
1
1
0
.2
.0
.8
1
1
2
2
3
0°C
5°C
0°C
5°C
0°C
No salt adding
0.6
0
0
0
.4
.2
.0
0
20
40
60
80
100
120
140
Time / min
0
20
40
60
80
100
120
140
Time / min
Fig. 8. Effect of salts added on the conversion of allyl bro-
mide; 15 mmol of allyl bromide, 25 mmol of bisphenol A,
Fig. 9. Effect of the temperature on the conversion of allyl
bromide; 15 mmol of allyl bromide, 25 mmol of bisphenol
A, 20 mL of chlorobenzene, 1 mmol of TBAB catalyst, 20
g of KOH, 40 mL of water, and 1000 rpm.
20 mL of chlorobenzene, 1 mmol of TBAB catalyst, 20 g
of KOH, 40 mL of water, 10 equi-mmol of salt, 1000 rpm,
ꢂ
and 30 C.

M.-L. Wang et al.
Bull. Chem. Soc. Jpn. Vol. 79, No. 1 (2006)
87
The authors would like to thank the CTCI Foundation,
Taiwan, Republic of China, and the National Science Council,
Taiwan, Republic of China for financial support under the
grant No. NSC-90-2214-E-029-006.
O
R
O
Heat
O
References
1
C. M. Starks, C. L. Liotta, M. E. Halpern, Phase-Transfer
Catalysis: Fundamentals, Applications, and Industrial Perspec-
R
tives, Chapman and Hall Publications, New York, U.S.A., 1994.
2
W. P. Weber, G. W. Gokel, Phase Transfer Catalysis in
Organic Syntheses, Springer Verlag, New York, U.S.A., 1977.
E. V. Dehmlow, S. S. Dehmlow, Phase Transfer Catalysis,
rd ed., VCH, New York, U.S.A., 1993.
Tautomerisation
OH
O
3
3
R
4
5
F. M. Menger, Chem. Soc. Rev. 1972, 1, 229.
J. J. Jarrouse, C. R. Hebd. Seances Acad. Sci., Ser. C 1951,
R
H
2
32, 1424.
Scheme 1. Claisen rearrangement for bisphenol A diallyl ether.
6 M. Makosza, M. Wawrzyniewicz, Tetrahedron Lett. 1969,
3, 4659.
5
7
New York, 2003.
A. G. Volkov, Interfacial Catalysis, Marcel Dekker, Inc.,
-
-
-
-
5.0
5.5
6.0
6.5
8
9
http://www.bimax.com/allyl.htm.
R. J. Morgan, E. E. Shin, B. Rosenberg, A. Jurek, Polymer
1
997, 38, 639.
J. M. Barton, I. Hamerton, J. R. Jones, J. C. Stedman, Poly-
mer 1996, 37, 4519.
T. Sakai, M. Tsubouchi, M. Nakagawa, M. Tanaka, Anal.
Chim. Acta 1977, 93, 357.
A. F. Lopez, M. T. Peralta de Ariza, O. A. Orio, J. High
1
0
1
1
1
2
Resolut. Chromatogr. 1989, 12, 503.
13 D. Landini, A. Maia, G. Podda, J. Org. Chem. 1982, 47,
3
.30
3.35
3.40
3.45
3.50
3.55
-
1
-3 -1
K
T
/ 10
2
264.
Fig. 10. Arrhenius plot for kapp vs 1=T; 15 mmol of allyl
bromide, 25 mmol of bisphenol A, 20 mL of chloroben-
zene, 1 mmol of TBAB catalyst, 20 g of KOH, 40 mL of
water, and 1000 rpm.
1
4
Handbook of Phase Transfer Catalysis, ed. by Y. Sasson,
R. Neumann, Blackie Academic and Professional, London, 1997.
1
4024.
5
D. J. Sam, H. E. Simmons, J. Am. Chem. Soc. 1972, 94,
C. L. Liotta, H. P. Harris, J. Am. Chem. Soc. 1974, 96,
M. C. V. Zwan, F. W. Hartner, J. Org. Chem. 1978, 43,
16
250.
2
2
þ
ꢃ
lyst. The anions in the ion pairs Q X do not affect catalyst
efficiency. Potassium hydroxide is better than sodium hydrox-
ide because sodium hydroxide salted out as a white gel in the
reaction system. It was also found that the existence of sodium
cations decreased the reaction rate when sodium hydroxide
1
7
655.
18 P. E. Scott, J. S. Bradshaw, W. W. Parish, J. Am. Chem.
Soc. 1980, 102, 4810.
19 R. A. Jones, Quaternary Ammonium Salts: Their Use in
Phase-Transfer Catalysed Reactions, Academic Press, U.K.,
2001.
or sodium salts were added. The apparent activation energy
ꢃ1
obtained is 65.6 kJ mol
.