Kinetics of the reaction between 2-phenylpropionitrile and 2-chloro-5-nitrotrifluoromethylbenzene under phase-transfer catalysis

Article abstract of DOI:10.1021/jo0483320

Phase-transfer catalysis (PTC) is a widely accepted methodology in organic synthesis. Although a great number of organic syntheses were reported in PTC conditions, systematic kinetic studies are scarce. In the present report, a detailed study of the kinet

Full text of DOI:10.1021/jo0483320

Kinetics of the Reaction between 2-Phenylpropionitrile and
-Chloro-5-nitrotrifluoromethylbenzene under Phase-Transfer
Catalysis
2
Sof ´ı a Varela Calafat, Edgardo N. Durantini, Stella M. Chiacchiera, and Juana J. Silber*
Departamento de Qu ı´ mica, Universidad Nacional de R ´ı o Cuarto, Agencia Postal # 3,
R ı´ o Cuarto X5804BYA, Argentina
jsilber@exa.unrc.edu.ar
Received September 20, 2004
Phase-transfer catalysis (PTC) is a widely accepted methodology in organic synthesis. Although a
great number of organic syntheses were reported in PTC conditions, systematic kinetic studies are
scarce. In the present report, a detailed study of the kinetics of the reaction between 2-chloro-5-
-
nitrotrifluoromethylbenzene (CNTFB) and 2-phenylpropionitrile anion (HPP ), under PTC, was
performed under several conditions. The reaction was carried out either in toluene or chlorobenzene
as the organic phase, in the presence of a concentrated aqueous solution of NaOH using
+
-
tetraalkylammonium (Q X ) salts as phase-transfer catalysts. The major product was 2-(4-nitro-
-trifluoromethylphenyl)-2-phenylpropionitrile, and its yield depends on the experimental conditions.
2
Different aspects of the mechanism are discussed and quantified. Kinetic data were explained by
means of an interfacial mechanism that involves the deprotonation of the adsorbed nucleophile
precursor followed by its catalyst-mediated extraction to the organic phase. A multicomponent
-
Langmuir-type interface was assumed. Although the extraction of OH by catalyst to the organic
phase is usually disregarded, the formation of the substrate hydrolysis product that leads to catalyst
poisoning was also investigated. The influence of this side reaction on the yield of the main product
was established. A discussion about the influence of this side process on the main reaction and the
operating mechanism is presented.
Introduction
ology allows the synthesis of many compounds not only
from poorly reactive starting materials but also with an
increase in yield or selectivity. Irrespective of the mecha-
nistic details, these processes offer a number of advan-
tages over homogeneous alternatives, and most of these
have been recognized as “green” alternatives in synthetic
Phase-transfer catalysis (PTC) is a widely accepted
methodology in organic synthesis, where reagents in
different phases come in contact by means of a phase-
transfer catalyst. This catalyst is capable of dissolving
or extracting the reagent into the organic phase, in the
form of an ion pair, where the reaction with the substrate
takes place. In the present report, the practical and
phenomenological Dehmlow’s definition for the term
phase-transfer catalysis is used irrespective of the oper-
ating mechanism.1
3-5
organic chemistry. Moreover, the use of chiral PTC to
satisfy the increasing demand of enanteomerically en-
4,6
riched compounds has been thoroughly demonstrated.
The reactivity of in situ generated anionic nucleophiles
under PTC conditions is usually enhanced since they
result unsolvated and therefore strongly activated for the
A great number of different organic reactions are
carried out under PTC conditions because this method-
1,7,8
substitution reactions.
The reaction of carbanions with
2
haloaromatic compounds offers important synthetic pos-
*
To whom correspondence should be addressed.
1) Dehmlow, E. V.; Dehmlow, S. S. Phase Transfer Catalysis, 3rd
revised and enlarged ed.; VCH Publishers: New York, 1993.
2) Sasson, Y., Newman, R., Eds. Handbook of Phase Transfer
Catalysis; Blackie Academic & Professional: London, 1997.
(3) Makosza, M. Pure Appl. Chem. 2000, 72, 1399-1403.
(4) Lygo, B.; Andrews, B. I. Acc. Chem. Res. 2004, 37, 518-525.
(5) Bielski, R.; Joyce, P. J. Org. Process Res. Devel. 2003, 7, 551-
552.
(
(
(6) O’Donnell, M. J. Acc. Chem. Res. 2004, 37, 506-517.
1
0.1021/jo0483320 CCC: $30.25 © 2005 American Chemical Society
Published on Web 05/11/2005
J. Org. Chem. 2005, 70, 4659-4666
4659

Calafat et al.
sibilities as it provides a way to introduce a carbon
tetraethylammonium chloride (TEAC), tetrabutylammonium
chloride (TBAC), tetrabutylammonium hydrosulfate (TBAHS),
and benzyldodecyldimethylammonium bromide were used
without further purification. Chlorobenzene, toluene, aceto-
nitrile, n-hexane, 2-propanol, and dichloromethane (HPLC
quality) were used as received. Pure water, conductivity 0.3-
.5 µΩ at 298 K, was obtained in a WaterPro Mobile for HPLC
system.
Procedures. The kinetic experiments were carried out in
a 20 mL three-necked flask equipped with a condenser and a
mechanical stirrer. The stirrer contains a mixing shaft of 6
mm diameter equipped with three six-bladed turbine-type
impellers. The reactor was immersed in a water-bath at 40.0
skeleton in an aromatic ring, which is often accompanied
by the introduction or change in functionality.9,10 The
reactions of R-phenyl alkylacetonitriles with chloroni-
trobenzenes were reported to proceed with yields ranging
from 59 to 92%.1
,11,12
An interfacial mechanism, which
0
involves the carbanion formation by proton abstraction
from the CH acid dissolved in a nonpolar solvent by a
concentrated aqueous NaOH phase boundary, is usually
accepted for this type of reaction.1
2,13
Thus, in situ
generated carbanions are ion-exchanged and extracted
to the organic phase as fully lipophilic ion pairs with the
PT cation and further reacted with the substrate, even
though a mechanistic picture of the way these reactions
take place was proposed in the early 1970s.14 In this
work, the kinetics of the reaction between 2-chloro-5-
nitrotrifluoromethylbenzene (CNTFB) and 2-phenylpro-
(
0.1 °C unless otherwise specified. All of the concentrations
given apply to a single phase.
To start a kinetic run, known quantities of HPPN, CNTFB,
+
-
and tetraalkylammonium salt (Q X ), dissolved in 5 mL of
toluene or chlorobenzene, were introduced into the reactor at
the desired temperature. The stirring rate was adjusted by a
rotor revolution counter to 1400 rpm unless otherwise speci-
fied. At zero time, a measured quantity of a thermostated
aqueous NaOH solution was added to the reactor. At given
times, the stirring was stopped, the two phases were allowed
to separate, and 50 µL of the organic phase was withdrawn
from the reaction mixture, diluted to 5 mL, and quenched with
0.2 mL of hydrochloric acid (30% v/v). The content of the
organic phase was quantitatively analyzed either by GC, using
-
pionitrile anion (HPP ) under PTC to yield 2-(4-nitro-2-
trifluoromethylphenyl)-2-phenylpropionitrile has been
studied. To the best of our knowledge, no kinetic data
have been reported for this reaction. This type of data
can provide the basis for future process design simula-
tions to assess possible industrial applications.15-17
The reaction was performed either in toluene or
chlorobenzene as the organic phase in the presence of a
concentrated aqueous solution of NaOH using tetraalky-
1
-methyl naphthalene (MN) as an internal standard, or by
HPLC.
Reaction kinetics was studied by following the disappear-
+
-
lammonium (Q X ) salts as PT catalyst. The effect of
variables such as base concentration, type, and concen-
tration of catalyst were explored. Different aspects of the
mechanism are discussed and quantified. Since catalyst-
ance of the substrate by GC using MN as internal standard.
The chromatographic area of the product, 2-(4-nitro-2-trifluo-
romethylphenyl)-2-phenylpropionitrile, was not quite repro-
ducible by GC. Thermal discrimination or partial thermal
decomposition in GC split/splitless injector was thought to be
the cause of the lack of reproducibility in the product areas.
Thus, the rate of product formation was followed by HPLC.
The amount of the product was obtained through direct
comparison of the peak areas against a calibration curve. The
pseudo-first-order rate constants (kobs) were obtained by a
nonlinear least-squares fit of the experimental concentration
vs time data.
-
mediated OH extraction to the organic phase can occur
under certain conditions, the substrate hydrolysis in the
reaction media was also investigated. To find the influ-
ence of this side reaction on the yield of the main reaction,
the rate of formation of the phenol derivative was
investigated under certain experimental conditions. A
discussion about the influence of this side process on the
main reaction and the operating mechanism is presented.
Calculation of Φ. To evaluate Φ (eq 17), experiments were
carried out the same way as described before but with no
18
CNTFB added. Under such experimental conditions, only eqs
-5 are applicable. The value of Φ was determined taking into
account the mass balance for the catalyst and the value of the
Experimental Section
3
General Methods. GC columns 30 m × 0.32 mm, phase
thickness 0.25 mm and 25 m × 0.25 mm, phase thickness 0.25
mm, were used. NMR spectra were acquired on a 200 MHz
spectrometer. Normal-phase chromatography was performed
with 99% hexane-2-propanol as eluent.
+
-
Q PPN concentration in the organic phase. The later was
determined by titration of the bromide anion in the organic
phase, according to Mohr’s methodology, keeping all of the
o
experimental conditions constant except Q . A constant value
-
3
Materials. 2-Chloro-5-nitro-trifluoromethylbenzene (CNTFB),
-phenylpropionitrile (HPPN), 1-methylnaphthalene (MN),
of 0.70 ( 0.05 was obtained within the range 5 × 10 to 25 ×
-3
2
10 M for TBAB concentrations.
tetrabutylammonium bromide (TBAB), benzyltriethylammo-
nium bromide (BTAB), tetrahexylammonium chloride (THAC),
Preparative Synthesis of 2-(4′-Nitro-2′-trifluorometh-
ylphenyl)-2-phenylpropionitrile. A mixture of CNTFB
(
3.38 mmol), HPPN (4.05 mmol), and TBAB (1.01 mmol)
dissolved in 3 mL of chlorobenzene was placed in a three-
necked flask equipped with a vertical condenser and a
thermometer. After a short period of efficient magnetic stir-
ring, 4 mL of a 50% w/w aqueous NaOH solution was added.
The mixture was kept at 40.0 ( 0.1 °C for 24 h. The crude
mixture was extracted with dichloromethane and filtered on
Florisil. Removal of the solvent and flash chromatography
(silica gel, 60 mesh, eluent: petroleum ether-dichloromethane
gradient) provided the pure compound. The purity was con-
firmed by HPLC and GC-MS.
(
7) Starks, C. M.; Liotta, C. L.; Halpern, M. Phase Transfer catalysis.
Fundamentals, applications and industrial perspectives; Chapman and
Hall: New York, 1994.
(
8) Rabinovitz, M.; Cohen, Y.; Halpern, M. Angew. Chem., Int. Ed.
Engl. 1986, 25, 960-970.
9) Calafat, S. V.; Durantini, E. N.; Chiacchiera, S. M.; Silber, J. J.
(
Organometallics 1999, 18, 2727-2730.
(
(
(
10) Jo n´ czyk, A. ARKIVOC 2004, 3, 176-178.
11) Makosza, M. Tetrahedron Lett. 1969, 673-676.
12) Makosza, M.; Jagusztyn-Grochowska, J. M.; Ludwikow, M.;
Jawdosiuk, M. Tetrahedron 1974, 30, 3723-3735.
(
(
(
13) Makosza, M.; Krylowa, I. Tetrahedron 1999, 55, 6395-6402.
1
The product was characterized by FT-IR, GC-MS, HNMR,
14) Makosza, M. Pure Appl. Chem. 1975, 43, 439-462.
1
3
-1
15) Satrio, J. A. B.; Doraiswamy, L. K. Chem. Eng. J. 2001, 82,
and C NMR. FT-IR: ν (cm , KBr) ) 3095.6, 3056.5, 2411.2,
4
1
5
3-56.
2
1618.7, 1534.3, and 1360.8 (NO ), 1483.1, 1418.8, 1400.1,
(
16) Bahattacharya, A.; Mungikar, A. J. Mol. Catal. A: Chem. 2002,
81, 243-256.
(
17) Glatzer, H. J.; Doraiswamy, L. K. Chem. Eng. Sci. 2000, 55,
(18) Durantini, E. N.; Chiacchiera, S. M.; Silber, J. J. J. Org. Chem.
1993, 58, 7115-7119.
149-5160.
4660 J. Org. Chem., Vol. 70, No. 12, 2005

Phase-Transfer Catalysis Reactions
1
317.1 (CF
MS (EI, 70 eV): m/z (relative intensity) 320 (84, M ), 305 (100,
M - CH ), 259 (24, 305 - NO ) 239 (14, 259 - FH), 190 (68,
3
), 1295.2, 1177.8, 1141.6, 1123.1, 650-800. GC-
+
3
2
+
+
C
6
H
3
NO
2
CF
3
), 130 (84, C
6
H
5
CCNCH
). H NMR (200.13 MHz, CDCl
2.23 (s, 3H), 7.20 (d, J ) 7.8, 2H), 7.34 (m, 3H), 8.63 (d, J )
3
), 103 (46, 130 - CNH),
+
1
7
7 (44, C
6
H
5
3
, TMS): δ (ppm)
)
Different salts were evaluated as catalyst such as
THAC, TBAC, TEAC, TBAB, TBAHS, and BTAB. The
salt counteranion seemed to be very important to achieve
the desire solubility in the organic phase. Chlorides and
bisulfate salts are scarcely soluble in the organic phase;
in fact, the chloride salt of the most organophilic assayed
cation, which is THAC, is almost at the solubility edge
of the catalyst. Therefore, although the use of highly
hydrophilic anions, such as chloride or bisulfate, is
preferable to avoid catalyst poisoning, they were rejected
due to their low solubility. Moreover, it has been reported
that PT-catalyst bearing alkyl groups with 16 or more
carbon atoms are very effective due to their high lipo-
philicity.23 Consequently, taking into account the greater
lipophilicity and therefore the higher solubility of TBAB
in the organic phase, the kinetic studies were conducted
using this catalyst.
Preliminary kinetic experiments were conducted by
mechanically stirring different concentrations of TBAB
in 5 mL of a toluene solution (0.1 M of CNTFB and 0.526
M HPPN) with 5 mL of NaOH aqueous solution (50%
w/w). The plot of the pseudo-first-order rate constants
(kobs) vs TBAB concentration is shown in Figure 1. At
TBAB concentrations higher than 0.025 M, the reaction
becomes excessively fast to be measured. Below 0.005 M,
the reaction goes to a halt with a rather low percentage
of conversion.
2
1
4
1
.56, 1H), 8.49 (dd, J ) 2.56, J ) 8.95, 1H), 8.05 (d, J ) 8.95,
13
H). C NMR (50.32 MHz, CDCl
3
, TMS): δ (ppm) ) 31.1,
5.7, 120.3 (CN), 124.5, 124.7 (CF
30.8, 131.4, 140.2, 144.3, 148,2.
3
), 125.7, 126.5, 128.4, 129.0,
Preparative Synthesis of 4-Nitro-2-trifluoromethylphe-
nol. A standard procedure for the synthesis of phenol deriva-
19
tives was used. Thus, a mixture of 8 mmol (1.8 g) of CNTFB,
.5 mmol (0.2 g) of benzyldodecyldimethylammonium bromide
0
in 4 mL of methanol, and 4 g of a 50% w/w NaOH solution
were thoroughly stirred by means of a mechanic stirrer. The
reaction was followed by TLC (silica F254, dichloromethane-
petroleum ether 1:1) until complete CNTFB disappearance.
The reaction mixture was extracted with dichloromethane with
a previous adjustment of the pH to 5 with a 50% v/v aqueous
hydrochloric acid solution. Staining of the phase was observed
upon acidification. Thin-layer chromatography (silica F254,
diethyl ether-dichloromethane, 1:1) of the organic phase
shows two spots; the largest one was the phenol derivative,
which remained at the origin of the TLC run. The organic
solvent was evaporated under vacuum. The residue dissolved
in water-acetonitrile shows that the characteristic pH depen-
dence of the absorption bands of the phenol derivatives was
2
0,21
also observed.
The product was reacted with bis(trimeth-
ylsilyl)trifluoracetamide to obtain trimethyl(4-nitro-2-trifluo-
romethylphenoxy)silane which was analyzed by GC-MS. GC-
+
MS (EI, 70 eV): m/z (relative intensity) 279 (5, M ), 264 (100,M
-
2 3 2 2
15), 218 (14, 264 - NO ), 77 (36, M - 15 - OSi(CH ) NO -
+
+
3 3 3 3
CF ), 71 (18, Si(CH ) ), 69 (2, CF ).
Kinetics. The reaction rates were measured by the initial
rate method in order to avoid longer time complications. The
sampling time could not be smaller than 1 min. Therefore, a
To investigate the reason for the observed low conver-
sion, different experimental conditions were assayed. A
33% conversion was obtained after 24 h when 5 mL of a
-
3
-1
pseudo-first-order rate constant of ca. 5 × 10
s
is ap-
proximately the highest value that could be measured by this
methodology with a percentage of error below 10%.
-
3
-3
toluene solution (0.103 × 10 , 0.0425 × 10 , and 3.93
-3
Catalyst Stability. The catalyst in the organic phase, in
either the presence or absence of the nucleophile, was stirred
at a controlled temperature with the basic aqueous phase.
According to the methodology suggested by Dehmlow22 for
chloride, aliquots were analyzed at different times by Mohr’s
titration of bromide in the organic phase. An attempt to
perform the experiment under the same experimental condi-
tions as the kinetics runs failed because the concentrations
were below the limit of sensitivity of the method. Therefore, a
concentration of ammonium salt at least 10 times greater than
that of the kinetic runs was used. Chlorobenzene was used as
solvent to avoid solubility problems.
× 10 M of HPPN, CNTFB, and TBAB, respectively) was
stirred with 5 mL of a 50% w/w aqueous NaOH. Then,
experiments with a higher TBAB/CNTFB ratio were
performed keeping the other experimental conditions
constant. Enhancing the ratio by a factor of 3 increases
the conversion almost 2-fold. In every case, the reaction
reaches 100% conversion if an excess of the fresh catalyst
is added to the reaction medium. Similar results were
obtained if BTAB was used as catalyst.
1
Since the catalyst decomposition is one of the most
frequent causes of observed inhibition, studies were
conducted in order to test catalyst stability under the
experimental conditions.
Results and Discussion
Catalyst Stability. Quaternary onium salts decom-
pose by two major mechanisms, Hoffman elimination and
nucleophilic displacement. Both reactions could be pro-
The reaction of CNTFB with HPPN under PTC condi-
tions afforded 2-(4-nitro-2-trifluoromethyl-phenyl)-2-phe-
nylpropionitrile (eq 1) as the main product, as shown by
TLC, GC, and HPLC analysis of the reaction mixtures
at infinite time.
The organic solvents used were either toluene or
chlorobenzene as described. This solvent proved to be the
most versatile to solubilize the reactants and catalyst
used in this study.
1
moted by either high-temperature exposure or the pres-
ence of strong alkaline medium or powerful nucleophilic
compounds. The results obtained for TBAB in the pres-
ence of NaOH (50% w/w) are collected in Table 1. As can
be observed, negligible decomposition seems to be operat-
ing within the first hour of reaction. Moreover, it can be
assumed that the organophilic catalyst remains in the
organic phase, since the amount of bromide in this phase
before phase contact equals the amount found after
contact.
(
19) Marhold, A.; Kcanke, E. (Bayen AG) Ger. Offen. 2,733,682, 1979
(
CI: C07 C79/32).
(
(
(
20) Cowles, E. J. J. Org. Chem. 1988, 53, 657-660.
21) Cowles, E. J. J. Org. Chem. 1986, 51, 1336-1340.
22) Dehmlow, E. V.; Knufinke, J. Chem. Res., Synop. 1989, 224-
(23) Desikan, S.; Doraiswamy, L. K. Chem. Eng. Sc. 2000, 55, 6119-
6127.
2
25.
J. Org. Chem, Vol. 70, No. 12, 2005 4661

Calafat et al.
tetrafluorobenzene with base.25 A precipitate of 2,4,6-
tribromophenolate was also observed under similar ex-
perimental conditions in a related system.26
Kinetic of the Reaction of CNTFB with the Base.
To establish the importance of the side reaction repre-
sented in eq 2, a kinetic study was performed.
Equimolecular amounts of CNFTB/TBAB were used in
order to avoid catalyst poisoning. The kinetic studies
were performed in the absence of HPPN using chloroben-
zene as solvent in order to improve catalyst solubility.
Thus, 5 mL of chlorobenzene solution (0.1 M in CNTFB
and TBAB) was mechanically stirred with 5 mL of 50%
w/w aqueous NaOH. The kinetics was followed by moni-
toring the disappearance of CNTFB by GC-MS.
FIGURE 1. Effect of the TBAB concentration on the kinetics
of the reaction: 5 mL of a toluene solution (CNTFB 0.1 M and
HPPN 0.526 M) with 5 mL of NaOH aqueous solution (50%
w/w).
The effect of the nucleophile on the catalyst stability
is shown in Table 2. As can be observed, the amount of
catalyst remains constant during the experiment, al-
though part of it should be forming ion pairs with PPN
anions. Previous studies showed that TBAB was also
stable in the presence of the related phenylacetonitrile
The reaction is slow and the initial rate method had
to be applied. Assuming a pseudo-first-order dependence
on the substrate, an observed rate constant of (6.0 ( 0.3)
-
5
-1
×
10
s
was calculated. It should be pointed out that,
1
8
despite the more favorable conditions of this experiment
compared to the ones of the results displayed in Figure
anion. Since these results demonstrated that the cata-
lyst is stable under the present experimental conditions,
other factors should therefore be responsible for the
observed inhibition. When phenylacetonitrile was reacted
1
, that is equimolecular amounts of catalyst and the use
of a better solvent, the estimated kobs value is 2 orders of
magnitude lower than the values reported in the former
experiment. Therefore, the formation of the phenol
derivative only can be a problem at long reaction times
or under experimental conditions that makes the main
reaction slow, such as low concentration of either catalyst
and/or base.
24
with CNTFB under PTC conditions, catalyst poisoning
was observed. Makosza et al.24 pointed out that the ipso-
substitution product, 2-(4-nitro-2-trifluoromethylphenyl)-
2
-phenylacetonitrile, is more acidic and lipophilic than
the nucleophile itself and causes displacement from the
lipophilic ion pair with tetrabutylammonium cation. In
this case, the reaction comes to a halt with low conver-
sion. 5-Chloro-3-phenyl-6-trifluoromethylbenzo[c]isox-
azole is formed instead of the ipso-dechlorination sub-
stitution product in a faster phase-transfer-catalyzed
reaction. However, in the present case, the nonionizable
nature of the ipso-substitution product prevents the
product mediate inhibition. The presence of side products
which could be both strong acids and lipophilic enough
to poison the catalyst was therefore investigated.
Catalyst Poisoning. The presence of a noticeable
amount of precipitate was observed and isolated from the
reaction medium when 5 mL of a toluene solution (3 M
in HPPN, 0.1 M in CNTFB, and 0.005 M in TBAB) was
stirred with 5 mL of 41% w/w of aqueous NaOH. Under
these experimental conditions, which amount to high
HPPN concentration but low concentration of catalyst
and NaOH, in the presence of less polar organic solvent,
a rapid but not complete disappearance of the substrate
took place. The isolated precipitate was identified as a
sodium salt of 4-nitro-2-trifluoromethylphenol. The struc-
Effect of Different Experimental Variables on
Substrate Consumption. The effects of different vari-
ables in the percentages of conversion, defined as the
percentage of CNTFB reacted to give products, were
determined by varying the solvent, the base, and sub-
strate concentration.The results are shown in Table 3.
As can be observed, runs 1-6, the percentage of conver-
sion decreases with the concentration of base in the
aqueous phase. It is known that hydroxide ion extraction
to the organic phase is favored at low base concentration
because of its greater degree of hydration of the extracted
ion pair.25 In addition, due to lower acidity of HPPN
related to water, a more hydrated organic phase brings
-
about a decrease in the PPN concentration in the phase.
Therefore, the lower percentage of conversion observed
at low base concentration could be the contribution of
both catalyst poisoning and a slower phenylpropionitrile
nitroarylation reaction.
The solvent also seems to have a great influence on
the percentage of conversion. Thus, a lower percentage
of conversion was found when toluene was used (run 7)
ture of the phenol was confirmed by comparison with the
3
compound synthesized by an independent procedure.19
instead of chlorobenzene (run 1). Wang et al. reported
the influence of solvent polarity on the PTC reaction of
The highly acidic phenolic derivative can be deprotonated
at the phase boundary and efficiently compete with the
nucleophile poisoning the catalyst. The phenol formation
was also found in the PTC reaction involving 1,2,3,4-
27
dibromo-o-xylene with n-butanol under basic conditions.
(
25) Landini, D.; Maia, A.; Rampoldi, A. J. Org. Chem. 1986, 51,
187-3191.
26) Wang, M.-L.; Yung, C.-Y. Tetrahedron. 1999, 55, 6275-6288.
(27) Wang, M.-L.; Tseng, Y.-H. J. Mol. Catal. A: Chem. 2002, 179,
17-26.
3
(
(
24) Makosza, M.; Tomasehewskij, A. A. J. Org. Chem. 1995, 60,
5
425-5429.
4662 J. Org. Chem., Vol. 70, No. 12, 2005

Phase-Transfer Catalysis Reactions
TABLE 1. Effect of Strong Basic Medium on TBAB Stability
time (min)
0
15
24
49
60
TBAB (mmol)a
0.525 ( 0.005
b
0.521 ( 0.005
0.526 ( 0.005
0.528 ( 0.005
0.522 ( 0.005
a
mL of 0.105 M TBAB in chlorobenzene was stirred at 40 °C and 1000 rpm with 5 mL of 50 wt % aqueous NaOH. b Before contact
5
among the organic and aqueous phases.
TABLE 2. Influence of the HPPN on the TBAB Stability
time (min)
TBAB (mmol)
0
5
15
25
78
b
0.495 ( 0.005
0.465 ( 0.005
0.470 ( 0.005
0.468 ( 0.005
0.470 ( 0.005
a
5
mL of chlorobenzene solution, 0.105 M TBAB, and 0.155 M PPN were stirred at 40 °C and 1000 rpm with 5 mL of 50 wt % aqueous
b
NaOH. Before contact among phases.
TABLE 3. Effect of Different Experimental Conditions
on the Percentage of Conversiona
Over 30% of NaOH the rate is too fast to be measured
by the initial rate method, and a roughly lower limit of
its value could be estimated, as discussed in the Experi-
mental Section. A rather similar effect was previously
observed for a related system, and it was attributed to
-
b
run
[OH ] (w/w, %)
conversion (%)
1
2
3
4
5
6
7
8
40
35
30
28
27
25
100
100
90
1
8
the amount of water present in the organic phase. At
high base concentration, the organic phase dehydration
by the hydroxide ions and the increase in the rate of the
80
40
35
2
9
c
deprotonation-extraction steps might occur. Both ef-
fects in turn allow an explanation of the increase in the
observed rate coefficient with the base concentration.
40
35
50
d
46
a
5
mL of organic phase, 0.1 M CNTFB, 3 M HPPN unless
-
3
specified, and 5 × 10 M in TBAB were mechanically stirred at
A competition between both substitution reactions, the
main and the side reactions, occurs at each base concen-
tration; the faster the main reaction, the higher the
percentage of conversion that is achieved. At small base
1
000 rpm with 5 mL of aqueous NaOH. Chlorobenzene was used
b
c
d
as solvent unless specified. At 24 h of reaction. Toluene. 0.97
M HPPN.
concentration poor deprotonation and a small selectivity
These authors found that the reaction in chlorobenzene
is faster than in toluene. In light of our results, the lower
conversion found in toluene could hence be due to the
slower rate of the ipso-substitution reaction compared
with the reaction which leads to catalyst poisoning.
Keeping constant the other experimental conditions,
a decrease in the HPPN concentration also decreases the
percentage of conversion (runs 2 and 8 in Table 3). The
observed results are probably due to the differential rate
of the competitive reactions under the different experi-
mental conditions, i.e., the slow phenol formation leading
to catalyst poisoning and the HPPN substitution. Thus,
when the main reaction becomes slow enough, the side
reaction progressively poisons the catalyst until the
reaction comes to a halt. The present results seem to be
in agreement with the peculiar dependence of the con-
centration of tetrabutylammonium 2,4,6-tribromopheno-
, (see eq 5) could be expected.2
5,28
Actually,
constant, K
s
at 15% w/w of NaOH the catalyst-promoted extraction
of solvated hydroxide to the organic phase might occur
but, taking into account that the more hydrated hydrox-
ide anion is a poorer nucleophile than the necked one,
the phenol derivative formation is almost negligible. This
side reaction seems to be slightly more significant when
the percentage of base is increased to 30% w/w. However,
as previously discussed, the rate of the hydrolysis reac-
tion is slower than the corresponding main reaction and
did not proceed to complete conversion even when sto-
ichiometric amount of TBAB is used.On the basis of the
previous results, the kinetic studies were performed in
3
0% w/w base solution to obtain a cleaner and most
complete reaction.
Effect of Substrate Concentration in the Reac-
tion Rate. Kinetic runs were performed by stirring 5 mL
of a toluene solutions of HPPN 0.526 M, TBAB 0.02 M,
and varying concentrations CNTFB (between 0.04 and
late with the concentration of the potassium hydroxide
observed by Wang et al.26 These authors suggested that
the true concentration of tetrabutylphenolate in the
organic phase is highly dependent on the phenol dis-
sociation and salting-out effect.
The best experimental conditions to carry out the
kinetic study should be a compromise between a high
conversion and a measurable rate. Therefore, to choose
the best experimental conditions for performing a thor-
ough kinetic study, the effect of the base concentration
upon reactivity were further investigated.
0
.16 M) with 5 mL of 30% w/w aqueous sodium hydrox-
ide. A first-order kinetics on substrate, characteristic of
a chemical-controlled rate, was obtained with a pseudo-
-
3
-1
first-order rate constant of (3.75 ( 0.05) × 10
s
for
all the CNTFB concentrations studied. These results
imply that the protonation-extraction equilibriums are
faster than the rate of the S
N
Ar reaction in the organic
phase.2
9,30
Effect of the NaOH Concentration in the Aqueous
Phase upon Reactivity. This studies were performed
using 5 mL of a toluene solution, 0.1 M in CNTFB, 0.02
M TBAB, 0.06 M of methylnaphthalene, and 0.526 M of
HPPN were stirred with 5 mL of different concentrations
of aqueous NaOH.
Effect of the TBAB Concentration on the Reac-
tion Kinetics. The kinetics of the reaction was studied
(
28) Sasson, Y.; Arrad, O.; Dermeik, S.; Zahalka, H.; Weiss, M.;
Wiener, H. Mol. Cryst. Liq. Cryst. Inc. Nonlin. Opt. 1988, 161, 495-
4
99.
(
(
29) Wu, H.-S.; Meng, S.-S. Chem. Eng. Sci. 1998, 53, 4073-4084.
30) Satrio, J. A.; Doraiswamy, L. K. Chem. Eng. Sci. 2002, 57,
At 15% w/w of NaOH, the reaction rate was negligible.
-
3
-1
At 30% w/w a kobs of (3.7 ( 0.3) × 10
s
was obtained.
1355-1377.
J. Org. Chem, Vol. 70, No. 12, 2005 4663

Calafat et al.
FIGURE 3. Effect of the 2-phenylpropionitrile concentrations
upon the reactivity: 5 mL of toluene solution, 0.1 M CNTFB
and 0.015 M TBAB, and 5 mL of 30 wt % NaOH as aqueous
phase.
FIGURE 2. Effect of the catalyst concentration on the kinetics
of the reaction between CNTFB and HPPN under basic PTC
conditions: 5 mL of toluene solution, 0.526 M HPPN, 0.1 M
CNTFB; 5 mL of 30% wt % of aqueous NaOH.
deprotonated by the strong base and is further extracted,
as a fully lipophilic ion pair with the PT-cation, to the
organic phase. The following mechanistic steps can be
written:
varying the TBAB concentration and keeping the other
experimental conditions constant. The results are shown
in Figure 2. Two different behaviors become apparent.
At low initial concentrations of catalyst, a linear depen-
dence on kobs is observed. Such linear tendency is
characteristic when the nucleophile deprotonation occurs
â
HPPN + Θ h Θ
(3)
(4)
s
HPPN
at the interface and the rate-limiting step is the nucleo-
Ka
philic substitution.31 The intercept of the plot is negli-
-
-
ΘHPPN + HO y
\
z Θ
+ H O
PPN
2
gible, and it would correspond to an undetectable rate of
the noncatalyzed reaction. Over certain concentration of
catalyst, ca. 0.015 M, the reaction rate becomes constant.
This could be attributed to interface saturation, which
means that the mass transfer of the active species into
the organic phase reaches a maximum value. At this
point, the addition of more catalyst will not help the
reaction to proceed faster. However, since the reaction
in the organic phase becomes too fast to be measured by
the initial rate method (see the Experimental Section),
the possibility of an underestimated observed rate con-
stant should also be considered. Nevertheless, as ex-
pected, the slope of the linear part of the plot in Figure
Ks
ΘPPN- + Q X
+
-
y\ z Q PPN
+
-
+ ΘX-
(5)
(6)
org
org
k
+
-
+
-
Q PPN
+ ArX h ArPPN + Q X
org
org
where Θ , ΘHPPN, ΘPPN
s
-
, and ΘX are the fractions of the
-
-
-
interface cover by solvent, HPPN, PPN , and X , respec-
tively. The subscript “org” is used to denote ion pairs in
the organic phase.
Assuming the interfacial excesses of HPPN and PPN-
and the corresponding fractions coverage can be ex-
pressed by the Langmuir monolayer isotherm of a mul-
2
5
(0.26 ( 0.01 M- s ) is lower than the one obtained at
1
-1
ticomponent system, the constant for eq 1 can be
32
-
1
-1
0% w/w NaOH in Figure 1 (0.40( 0.02 s M ).
expressed as follows:
Effect of the HPPN Concentration upon Reactiv-
ΘHPPN
ity. The kinetics of the reaction was studied at different
HPPN concentrations, keeping the other experimental
conditions fixed. The experimental results are shown in
Figure 3. Under these conditions, the observed rate
constant increases with the nucleophile concentration.
Mechanism. When “in situ” generated nucleophiles,
by deprotonation a lipophilic weak acid precursor (HPPN),
are involved in PTC reactions, both the unchanged
nucleophile and its anion are extracted to the organic
medium due to the strong salting out effect of the
aqueous phase. The extraction should be accomplished
with the aid of the PT-catalyst; otherwise, a bilayer-like
arrangement of alkali metal cations on the aqueous side
and deprotonated nucleophiles on the organic side of the
â )
(7)
[
HPPN] Θs
Taking into account all surface-active compounds
Θ ) 1 -
Θi
(8)
s
∑i
The acidity constant for HPPN deprotonation at the
phase boundary, K (in eq 4), is defined as
a
ΘPPN
-
[H O]
2
K )
(9)
a
-
ΘHPPN[HO ]
1
phase boundary is formed instead. In such cases, and
As already discussed, the generated carbanion, as a
Na salt, can neither migrate into the organic phase,
because the highly hydrophilic counterion, nor move with
carbanions into the aqueous phase, due to the strong
+
according to the hitherto presented data, an interfacial
mechanism has been postulated.1
3,14
At first, the nucleo-
phile precursor is adsorbed at the interface where it is
(
31) Solaro, R.; D’Antone, S.; Chiellini, E. J. Org. Chem. 1980, 45,
(32) Juang, R.-S.; Liu, S.-C. Ind. Eng. Chem. Res. 1997, 36, 5296-
5301.
4
179-4183.
4664 J. Org. Chem., Vol. 70, No. 12, 2005

Phase-Transfer Catalysis Reactions
salting out effect.33 The selectivity equilibrium constant,
could be neglected in eq 14. Besides, at short reaction
times, when both nucleophile precursor and the base are
in great excess over the catalyst, [Q PPN ]org remains
constant and eq 14 can be rewritten as follows
-
K
s
(in eq 5) for the extraction of PPN with respect to
-
+
-
that of X , entails the nucleophile extraction to the
organic phase as a fully lipophilic ion pair with the onium
ion. Its mathematic expression is shown in eq 10.
+
-
r ) k[Q PPN ]org[ArX] ) kobs[ArX]
(15)
+
-
[
Q PPN ]orgΘX
-
K )
(10)
s
where kobs is the observed pseudo-first-order rate coef-
ficient.
+
-
ΘPPN-[Q X ]
org
Assuming a fast equilibrium previous to the determin-
ing step, the substitution reaction in the organic phase,
and taking into account the mass balance for the catalyst,
The intrinsic second-order rate coefficient for the main
substitution reaction in the organic phase is represented
by k (eq 6).
Processes depicted in eqs 11-13 could also be taking
place.
+
-
+
-
Q
o
) [Q X ]org + [Q PPN ]org, eq 16 can be obtained.
-
kâK K Q Θ [OH ][HPPN]
a
s
o
s
kobs
)
(16)
-
+
-
+
-
-
-
OH + Q X
h Q OH
+ X
(11)
(12)
org
org
[H O]Θx + âK K Θ [OH ][HPPN]
a s s
-
2
k′
+
-
org
+
-
For the sake of simplicity, eq 16 can be written in the
form of eq 17
Q OH
+ ArX
9
8 Q X
+ ArOH
org
+
-
org
+
-
Q OH
+ HPPN h Q PPN
+ H O (13)
org 2
kQ [HPPN]
o
kobs
)
(17)
-
-1
Even though X are more lipophilic than the very
φ
+ [HPPN]
-
hydrophilic OH , and eq 11 should be shifted to the
1
,7,33
where Φ ) (âK K [HO ]Θ )/([H O]Θ -)
-
left,
the importance these equation is stressed since
a s s 2 x
the phenol derivative is found as side product in the
reaction medium. Although it has been pointed out that
the hydroxide ion transfer to the organic phase only
occurs when both very weak acids and highly hydrophilic
Estimation of the Intrinsic Second-Order Rate
Coefficient. At short reaction times, it can be assumed
that the coverage of the interface remains almost un-
s x
changed and, hence, Θ and Θ as well. Data adjustment
-
-
34-36
catalyst counterion HSO
4
are involved,
it seems
requires evaluating Φ at the working conditions. To
accomplish the task, the experiments described in the
Experimental Section were carried out. Under these
circumstances, only steps 3-5 of the mechanism can
occur and Φ can be evaluated from eq 18.
that, at least under experimental conditions that make
reaction 6 slow enough, eq 12 yields the surface-active
phenol derivative, whose deprotonated form finally leads
to catalyst poisoning. Taking into account the lower acid
strength of HPPN compared to water, the equilibrium
+
-
(
13) would be almost completely displaced to the left side,
[Q PPN ]
org
Φ )
(18)
and therefore, it would not be a source of nucleophile in
the reaction medium.
+
-
(
Q - [Q PPN ]org)[HPPN]
o
The kinetic rate (r) for ArX disappearance, eq 14,
should therefore be the contribution of two reactions, one
that leads to the arylation of the HPPN (eq 6) and other
that produces the phenol derivative (eq 12). It is impor-
tant to note that the reactions at the phase boundary
were ignored in view of the fact that no reaction at all
was observed in the absence of catalyst.
Thus, by the introduction of the proper value of Φ (0.7
M- under the present conditions) in eq 17, the value of
k can be estimated from the slope of the linear portion of
1
the kobs vs Q
o
plot (Figure 3). A value of k ) 0.95 ( 0.07
for [NaOH] ) 30% w/w was obtained.
Other estimations could be performed provided the Φ
-
1
-1
s
M
value is almost independent of the HPPN concentration
operating range. Nonlinear least-squares fitting of data
in Figure 3 by means of eq 16 gave a k value of 0.85 (
r ) -d[ArX]/dt )
+
-
+
-
(
k[Q PPN ] + k′[Q OH ]org)[ArX] (14)
org
-1
-1
0
.08 s M . As can be appreciated, good estimates for
k within experimental error were obtained by two dif-
ferent data sets.
However, as described previosly, the rate of phenol
formation measured in the absence of HPPN and using
equimolecular amounts of catalyst and substrate has a
Effect of Stirring Rate upon the Reactivity. The
results are shown in Figure 4. As can be noticed, a rise
in the stirring rate produce a typical increase in the
pseudo-first-order rate constant. This behavior is char-
acteristic of reaction believed to proceed through an
interfacial mechanism.37 This effect can be attributed to
-
5
-1
value of (6.0 ( 0.3) × 10 s . This is at least 2 orders of
magnitude slower than the one obtained (i.e., (7.5 ( 0.4)
-
3
-1
×
10 s ) under the same conditions but using four
timesless catalyst (Figure 1). Thus, this side reaction
17,38-40
the related enhance in the interfacial surface area.
(
33) Makosza, M.; Fedorynski, M. In Advances in Catalysis; Eley,
D. D., Pines, H., Weisz, P. B., Eds.; Academic Press: London, 1987;
Vol. 35, p 375.
(37) Jayachandran, J. P.; Wang, M.-L. Appl. Catal. A: Gen. 2001,
296, 19-28.
(
34) Halpern, M.; Sasson, Y.; Rabinovitz, M. J. Org. Chem. 1983,
4
5
3
8, 1022-1025.
35) Feldman, D.; Halpern, M.; Rabinovitz, M. J. Org. Chem. 1985,
0, 1746-1749.
36) Feldman, D.; Rabinovitz, M. J. Org. Chem. 1988, 53, 3779-
784.
(38) Wang, M.-L.; Hsieh, Y.-M.; Chang, R.-Y. J. Mol. Catal. A:
Chem. 2003, 198, 111-124.
(
(39) Sirovski, F.; Gorokhova, M.; Ruban, S. J. Mol. Catal. A: Chem.
2003, 197, 213-222.
(
(40) Starks, C. M. Tetrahedron 1999, 55, 6261-6274.
J. Org. Chem, Vol. 70, No. 12, 2005 4665

Calafat et al.
the nucleophiles.4
1-44
The GC analysis of the reaction
mixture at 40 °C shows a clean reaction, and the ipso-
substitution product is the only one observed. The rates
of product formation and CNTFB disappearance agree
within the experimental error with values of (4.3 ( 0.3)
-
3
-3 -1
×
10 and (3.9 ( 0.3) × 10 s , respectively.
Conclusions
HPPN and CNTFB react under basic phase-transfer-
catalysis conditions to give mainly the chloro-ipso-
substitution product. Even though the phase-transfer
catalyst is stable to experimental conditions, the depro-
tonated phenol side product slowly poisons it. This side
reaction becomes important at rather low catalyst con-
centration, i.e., under experimental conditions where the
main reaction is excessively slow. The system is very
sensitive to the base concentration that mainly changes
the degree of nucleophile hydration in the organic phase
and consequently strongly affects the rate of reaction. The
reaction is sped up by an increase in the concentration
of catalyst or nuclephile precursor and by the stirring
rate. An interfacial mechanism was assumed, and the
data were adjusted taking into account a Langmuir
monolayer isotherm of a multicomponent system at the
phase boundary. The intrinsic second-order rate coef-
ficients estimated from two different sets of data agree
within the experimental error. As a final remark, one can
state that the reported data allowed us to draw a
complete kinetic formulation of the reaction mechanism
and the main hints that should be considered while using
it as a synthetic tool.
FIGURE 4. Effect of stirring rate on the pseudo-first-order
reaction rate: 5 mL of toluene solution, 0.526 M MBC, 0.1 M
2CNTFB, 0.02 M TBAB, were mechanically stirred with 5 mL
of 30 wt % NaOH.
Acknowledgment. Financial support from the Con-
FIGURE 5. Effect of the temperature upon the reactivity of
CNTFB and HPPN under PTC conditions: 5 mL of toluene
solution, 0.526 M HPPN, 0.1 M CTFNB, 0.015 M TBAB, 5 mL
of 30% wt % aqueous NaOH.
sejo Nacional de Investigaciones Cient ´ı ficas y T e´ cnicas
(
CONICET) and the Consejo de Investigaciones Cien-
t ´ı ficas y Tecnol o´ gicas de la Provincia de C o´ rdoba (CONI-
COR) is gratefully acknowledged. S.V.C. thanks CONI-
COR for a fellowship. S.M.C., J.J.S., and E.N.D. are
scientific members of CONICET.
At stirring rates higher than 1450 rpm the reaction is
too fast to be followed by the initial rate method.
Temperature Effect on the Observed Rate Con-
stant. An Arrhenius plot of the data at different tem-
peratures is shown in Figure 5. From this plot, the
JO0483320
(
41) de la Zerda Lerner, J.; Cohen, S.; Sasson, A. J. Chem. Soc.,
Perkin Trans. 2 1990, 1-6.
q
-1
q
activation parameters ∆H ) 80 ( 5 kJ mol and ∆S )
(42) Wang, M.-L.; Wu, H.-S. J. Chem. Soc., Perkin Trans. 2 1991,
841-845.
-
1
-1
-
212 ( 20 J mol
K
were calculated in the range of
(
(
43) Kuo, C.-S.; Jwo, J.-J. J. Org. Chem. 1992, 57, 1991-1995.
44) Wang, M.-L.; Wu, H.-S. J. Polym. Sci., Part A: Polym. Chem.
temperature used. These values are consistent with the
complexity of proposed mechanism, where carbanions are
1992, 30, 1393-1399.
4666 J. Org. Chem., Vol. 70, No. 12, 2005