Kinetics of 'initial burst' in the solid-liquid phase-transfer catalysis. Nucleophilic substitution of 2-octyl mesylate with potassium bromide

Article abstract of DOI:10.1002/poc.375

The S_N2 substitution of 2-octyl mesylate with solid KBr under the conditions of phase-transfer catalysis was studied kinetically using the model approach of 'initial burst.' It is suggested that such kinetics reflect the contribution of mass transfer and surface poisoning. The proposed model is used to explain the influences of catalyst, solvent, stirring speed, activation and agitation effects. The mechanistic scheme suggests that the reaction is described by two separate kinetic stages, one of which reflects the intrinsic rate-limited step and the other the mass transfer-controlled step. Copyright

Full text of DOI:10.1002/poc.375

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2001; 14: 343–348
DOI:10.1002/poc.375
Kinetics of `initial burst' in the solid±liquid phase-transfer
catalysis. Nucleophilic substitution of 2-octyl mesylate with
potassium bromide
S. S. Yu®t*and S. S. Zinovyev
1N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Leninsky pr. 47, Moscow 119992, Russia
Received 8 August 2000; revised 14 February 2001; accepted 17 February 2001
ABSTRACT: The SN2 substitution of 2-octyl mesylate with solid KBr under the conditions of phase-transfer catalysis
epoc
was studied kinetically using the model approach of ‘initial burst.’ It is suggested that such kinetics reflect the
contribution of mass transfer and surface poisoning. The proposed model is used to explain the influences of catalyst,
solvent, stirring speed, activation and agitation effects. The mechanistic scheme suggests that the reaction is described
by two separate kinetic stages, one of which reflects the intrinsic rate-limited step and the other the mass transfer-
controlled step. Copyright  2001 John Wiley & Sons, Ltd.
Additional material for this paper is available from the epoc website at http://www.wiley.com/epoc
KEYWORDS: solid–liquid phase transfer; kinetics; catalysis; nucleophilic substitution; 2-octyl mesylate; potassium
bromide
INTRODUCTION
occurs in the organic phase, at the solid–liquid interface
or in the immediate vicinity to the surface. It has been
shown6 that small additions (or residual traces) of water
may have a substantial effect on the kinetics by either
accelerating or slowing the reaction. Generally, water
results in softening of the salt lattice and, as a result, in
homogenization of the reaction step. We focus on the
‘dry’ systems where the lattice ‘strength’ does not allow
separation of the anion, so the reaction proceeds
preferably on the surface.
The method of phase-transfer catalysis (PTC) in the
presence of a solid phase as an inorganic reagent, the so-
called solid–liquid (s/l) PTC, is a convenient technique in
organic synthesis that has been widely studied over the
last 20 years.1 Occasionally, it has advantages over the
common liquid–liquid (l/l) PTC and is sometimes
successfully employed on an industrial scale.2 There
are a number of papers3 which have reported on the
kinetics and varied mechanistic questions, although there
is still no unanimity regarding either the mechanism or
the topology of s/l PTC.
The complexity of s/l systems arises from the
contributions of various side effects, such as adsorption
and diffusion on the surface,4 modification of the surface
by the regent/substrate adsorption that sometimes causes
surface poisoning,4,5 etc. These effects make s/l PTC
resemble heterogeneous catalysis. In fact, the contribu-
tion of surface factors is sometimes very pronounced, so
the reaction may not be limited by a chemical step but
rather by the mass transfer through the solid–liquid
interface barrier.
Kinetics of initial burst in s/l PTC
We have previously reported on the kinetics of
nucleophilic substitution of hexyl bromide with solid
MCl under PTC conditions.7 The reaction was shown to
proceed on the solid surface via the consecutive
formation of binary and ternary adsorption complexes
formed by the substrate, onium salt and MCl.
We recently reported on the kinetics of the analogous
substitution of 2-octyl mesylate with solid potassium
halides under s/l PTC conditions.8 The kinetics of this
reaction in an l/l system were previously studied by
Landini and co-workers.9 However, in the s/l system it
appears more complicated and informative and allows
one to draw more detailed conclusions about the reaction
mechanism. To describe the reaction kinetics we
proposed the approach of ‘initial burst’ (IB),8,10 which
gives a very accurate approximation of the data obtained.
Also, one of the most challenging issues in s/l PTC is
the topology of the chemical reaction step: whether it
*Correspondence to: S. S. Yufit, N. D. Zelinsky Institute of Organic
Chemistry, Russian Academy of Sciences, Leninsky pr. 47, Moscow
119992, Russia.
E-mail: yufit@ioc.ac.ru
Copyright  2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 343–348

344
S. S. YUFIT AND S. S. ZINOVYEV
Scheme 1
Avrami–Erofeev equation.13 This approach is based on
the continuous deceleration of the reaction due to the
growing contribution of diffusion through the solid
product ‘crust,’ which constantly accumulates over the
reacting particles. Recently, this approach has also been
applied to describe s/l PTC kinetics,11 although the
reported case is not identical with ours. In the
topochemical model, the amount of ‘crust’ grows
constantly, so the reaction does not become zero-order.
Another attempt to describe similar kinetics in
gaseous–solid phase substitution was made by Mitchenko
and Dadali,12 who suggested that the reaction is
described by two kinetic regimes because of two different
types of catalytic sites on the surface. However, neither
this model nor the Avrami–Erofeev model11,13 can be
used in the present case, because they do not describe the
stationary kinetic regime, which corresponds to the linear
part of the kinetic curve.
We propose a model which is based on the principles
of the three-stage Michaelis–Menten kinetics10a and is
applicable in some enzyme-catalyzed reactions. As a
starting point, we assume that the active site (AS) on the
KBr surface plays a role similar to that of an enzyme.8 As
in enzyme kinetics, these AS are consumed in the initial
stage and regenerated thereafter. The proposed mechan-
ism is given in Scheme 2.
Figure 1. Kinetics of IB in the substitution of 2-octyl mesylate
+1 M in 10 ml toluene solution) with solid KBr +12 g, <0.1 mm
fraction) catalyzed by Aliquat 336 +for other conditions, see
Experimental section). The curve is the non-linear least-
squares ®t +NLSF) of the experimental data +dots) using Eqn.
+2)
The IB kinetics have not been studied with respect to
PTC before, although there are papers reporting on
similar kinetic features in analogous systems.11,12 This
investigation was directed towards a more detailed study
of such kinetics in the reaction of 2-octyl mesylate with
solid KBr [Eqn. (1)]. This simple model reaction was
chosen to gain an insight into the basic mechanistic
picture, although the results obtained cannot be general-
ized for other systems.
n-C6H13CHꢀCH3†OMs
toluene/ 95ꢁC
À!
‡ KBr
n-C6H13CHꢀCH3†Br
Aliquat 336
‡ KOMs
ꢀ1†
The first kinetic regime is conditioned by the
adsorption of PT agent (QX) and substrate (ROMs) on
the KBr surface resulting in the adsorption (subscript
‘ads’ in Scheme 2) complex (KBr)n À 1[KBr QX
ROMs]ads, which then decomposes and gives RBr. These
steps overall are reflected in the pseudo-first-order rate
constant k‡2, which defines the reaction rate in the pre-
steady state. The second regime includes regeneration of
the PT agent and renewal of the KBr surface. Decom-
position of the imaginary complex (KBr)n À 1[KOMs]ads
and regeneration of a new KBr AS lead to the removal of
A typical reaction profile representing the IB kinetics
is shown in Fig. 1. The reaction initially proceeds at a
high rate of first order. After reaching a certain degree of
conversion, it turns into a regime which is described by a
lower rate and zero-order kinetics, and proceeds to
complete conversion in such a fashion.
As we suggested earlier,8 the solid product KOMs
forms a so-called ‘crust’ over the KBr particle and
gradually poisons the reacting surface. The amount of
‘crust’ grows to a certain extent until the rate of ‘crust’
accumulation becomes equal to the rate of its removal.
Once this equilibrium amount of ‘crust’ has been
reached, the reaction is no longer limited by the chemical
step, but proceeds in the mass transfer regime. A pictorial
representation of the exchange processes in our system is
shown in Scheme 1.
solid ‘crust’ and are defined by the k constant. If k
ꢂ k the reaction would turn into the steady-state as
‡3
‡2
‡3
soon as all AS are consumed, i.e. when the surface
becomes poisoned by the ‘crust’. In this regime, the
reaction rate is mostly defined by k‡3. We assign [E]0 to
the number of KBr sites, which react in the pre-steady-
state regime and most of which then become bound in
(KBr)n À 1 [KOMs]ads in the steady state. A zero-order
Similar kinetics are observed in some heterogeneous
topochemical reactions which are described by the
Copyright  2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 343–348

KINETICS OF ‘INITIAL BURST’ IN PHASE-TRANSFER CATALYSIS
345
ROMs
ꢀKBr† ‡ QX „ ꢀKBr† ‰KBr QXŠ
„
ꢀKBr† ‰KBr QX ROMsŠ
n
nÀ1
ads
nÀ1
ads
k
‡2
ꢀKBr† ‰KBr QX ROMsŠ À! ꢀKBr† ‰KOMs QXŠ ‡ RBr
nÀ1
ads
nÀ1
ads
ꢀKBr† ‰KOMs QXŠ À! ꢀ KBr† ‰KOMsŠ ‡ QX
nÀ1
ads
nÀ1
ads
k
‡3
ꢀKBr† ‰KOMsŠ À! ꢀKBr† ‡ KOMs
nÀ1
ads
nÀ1
Scheme 2
dependence is observed because the rates of the AS
consumption and regeneration become equal in the
steady state, i.e. the concentration of (KBr)n À 1[KOM-
s]ads is stationary (and is close to the [E]0 value). The
following integral rate expression can be derived for this
model:8,10
the stirring rate was kept near 2000 rpm. After the reaction
had been started, the stirring was stopped periodically and,
aftertherequiredseparationoforganicandsolidphaseshad
been achieved, samples of the organic phase (ꢃ0.1 ml)
were taken with a pipette through the reactor neck. The
reaction was terminated after acceptable kinetic profiles
had been obtained (40–70% conversion for most experi-
ments) or was allowed to proceed to completion. All the
reagentsandsolventswereof‘chemical bypure’gradeand
were used without further purification.
k
‡2k ‰EŠ t
‡3
‡ k
0
‰PŠ ˆ
k
‡2
‡3
k2 ‰EŠ f1 À exp‰Àꢀk ‡ k †tŠg
‡2
†
‡3
‡2
0
‡
ꢀ2†
2
Gas chromatographic +GC) analyses and kinetic
measurements. GC analyses were conducted at a column
temperature of 80–150°C depending on the type of a
compound under study. A glass column (2.4 Â 0.005 m
i.d.) packed with Chromaton N-AW DMCS (0.16–
0.2 mm) saturated with a 5% solution of SE-30 was
used. The concentrations of 2-octyl mesylate and 2-
bromooctane were calculated from the ratios of the GC
peak areas for the analyzed compound and the internal
standard (undecane) using the known concentration of
undecane in the organic phase and the calibration
coefficients 1.67 and 1.8 for 2-octyl mesylate and 2-
bromooctane, respectively. If the reaction proceeded to
completion the GC yields were near 90%.
ꢀk ‡ k
‡2
‡3
where [P] is the monitored RBr concentration in organic
phase, t is time and k‡2, k and [E]0 are kinetic
‡3
parameters. The use of this expression allows one to
calculate three independent kinetic parameters (k‡2, k
and [E]0) from a single run.
‡3
This work was aimed at further investigation of the IB
kinetics in application to s/l PTC. A number of
experiments were performed to study the effects of
different reaction conditions on the calculated k‡2, k
and [E]0 values.
‡3
The kinetic data, i.e. the product concentrations (mol
l
À1) plotted against time (min), were then processed using
EXPERIMENTAL
MicroCal Origin v. 3.5 software. The non-linear least-
squares fitting (NLSF) method was performed to compute
the kinetic parameters. Equation (2) was used for the
fitting of most of the experimental data, except the
experiments which were described by the first-order
kinetics or where an initial rate analysis was performed.
The values of k‡2, k‡3 and [E]0 were calculated with an
accuracy corresponding to 20% and lower standard errors
of NLSF.
General procedure and sampling. SolidKBrwasground
and sieved to obtain fractions of salt with particle sizes
<0.1, 0.1–0.125, 0.125–0.2 and >0.2 mm. Each fraction
was then thoroughly dried under vacuum prior to the
reaction.AportionofdriedKBr(acertainfractionofsieved
or non-sieved salt) was loaded into a water-jacketed three-
necked reactor thermostated at 95°Csupplied with a reflux
condenser and a mechanical stirrer. After the salt had been
pre-activatedbystirringinthereactorfor30 min,asolution
of racemic 2-octyl mesylate prepared by a standard
procedure14 (2.04 g, 0.01 mol) and undecane (0.3 ml) in
toluene (7 ml) was placed in the reactor. The reaction
mixture was stirred for a few seconds, whereupon the
stirring was stopped and the first reference sample (zero-
time point) was taken. Immediately after the first sample
had been taken, tricaprylmethylammonium chloride (Ali-
quat 336) (0.15 ml, 0.25 mmol) was placed in the reactor
and the stirring was continued. If not indicated otherwise,
RESULTS AND DISCUSSION
Stirring and preliminary activation
It has been shown by different authors7,15 that mechanical
agitation during or prior to the reaction may have a
pronounced influence on the kinetics of PTC. The typical
S-like dependences of the rate constants on stirring speed
Copyright  2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 343–348

346
S. S. YUFIT AND S. S. ZINOVYEV
mechanistic activation of salt had an accelerating effect
and led to the elimination of the induction period on
kinetic profiles.7a However, in the present case the
preliminary activation has little or no effect on kinetics
(for results, see Supplementary Material). Consequently,
the KBr surface is developed enough without any
preliminary agitation. On the other hand, if the reaction
is started with a coarse salt (>0.2 mm sieved particles)
and proceeds for a long time some acceleration is
observed during the steady state (see Supplementary
Material). Seemingly, the larger salt particles are easily
ground, so the rate increases because of the increase in
the surface area and in the amount of AS. A larger
amount of AS means a higher [E]0 value, which defines
the steady-state rate.
Figure 2. Effect of stirring speed on rate constants k‡2 and
k
‡3
in the substitution of 2-octyl mesylate +0.76 M in 10 ml
toluene solution) with solid KBr +12 g, salt fraction 0.1±
0.125 mm) catalyzed by Aliquat 336 +0.025 M). Error bars
indicate the standard errors of k and k calculation by
Effect of initial substrate concentration
‡2
‡3
The influence of the initial substrate concentration ([S]0)
on the reaction kinetics was also studied (see Supple-
NLSF
mentary Material). The observed rate constants k and
‡2
k
‡3 depend linearly on [S]0. The [E]0 value is not affected
are observed in most cases. Basically, the rate vs stirring
curves reach saturation at comparatively low agitation
speeds (100–300 rpm) if the reaction is not much affected
by mass transfer. In contrast, if the observed constant
reflects the mass transfer-limited step, as is typical for s/l
PTC, saturation is reached at about 1000 rpm or higher.
Once the saturation of the rate constant has been reached,
the reaction is considered to be in a kinetic regime,1c
which means the absence of any diffusion contribution
for the true chemically controlled reaction. As for the
mass transfer-controlled reaction, the rate constant at
which the saturation is reached can be conditioned by the
in posse degree of dispersion/emulsification, as has been
demonstrated by Starks,15a or by the form of the stirrer
and/or reactor.
by the changes in substrate concentration, which is
consistent with the proposed model (Scheme 2). Indeed,
the degree of surface poisoning by the ‘crust’ would be
defined by the surface properties only and not by the
component concentrations, except that of QX, which can
modify the reacting surface.8 Obviously, an increase in
[S]0 would lead to acceleration of the steps responsible
for the consumption of AS and accumulation of the
‘crust,’ which explains the observed increase in k
Therefore, k‡2, as an intrinsic constant, is first order in
substrate.
.
‡2
On the other hand, an increase in substrate concentra-
tion causes the same substantial increase in k‡3, which is
at variance with the model, because k‡3 represents for the
surface renewal and is not connected with any chemical
step involving substrate. The observed increase in k‡3 vs
[S]0 is the result of deviation from the IB model in the
experiments with low concentrations of substrate
([S]0 = 0.25, 0.4, and 0.6 M; see Supplementary Ma-
terial). In these cases, approximation of the steady state
becomes misleading, because [S]0 must not be compar-
able to the ‘burst’ magnitude ([E]0), which is approxi-
mately 0.2 M, so the model cannot be used for the
The calculated constants for the reaction under study
vs agitation speed are presented in Fig. 2. The k‡2 value
reaches a plateau starting with a very slow stirring speed,
whereas it takes nearly 2000 rpm for k
to reach
‡3
saturation. The observed dependences indicate the
different natures of the observed constants. According
to the proposed mechanism, k‡2 represents the consump-
tion of AS, which is a purely intrinsic chemical process
and therefore would not depend much on stirring. In
contrast, k‡3 describes the step of AS regeneration, which
is virtually the mass transfer-controlled process of
surface renewal. It is more affected by stirring because
its nature is more physical. The [E]0 value does not
depend on stirring, which means that the amount of
working AS does not increase in the studied range of
stirring.
accurate estimation of k
.
‡3
In¯uence of PT agent on reaction kinetics
The PT agent usually plays an essential role in the
kinetics of PTC. We studied a number of onium salts (see
Supplementary Material) having different quaternary
cation structures and coupled with ClÀ, BrÀ, IÀ and
HSO4À anions. The use of crown ethers was also studied.
The kinetic profiles of reaction (1) catalyzed by these
We have previously reported that in the bromine–
chlorine substitution of hexyl bromide with solid metal
chlorides under s/l PTC conditions, the preliminary
Copyright  2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 343–348

KINETICS OF ‘INITIAL BURST’ IN PHASE-TRANSFER CATALYSIS
347
Figure 3. Kinetic pro®les of Eqn. +1) in the presence of
different PT agents: ^, TBAB, +C4H9)4NBr; !, TBAH,
+C4H9)4NHSO4; &, TBAI, +C4H9)4NI; *, TCMAC,
CH3+C8H17)3NCl; *, TEAC, +C2H5)4NCl; ^, TEAB,
+C2H5)4NBr;
&,
TEAI,
+C2H5)4NI;
~,
TEAAB,
C15H31+C2H5)3NBr; ~, TEBAC, PhCH2+C2H5)3NCl; !,
TPMPB, CH3+Ph)3PBr; ‡, TOAB, +C8H17)4NBr; Â, DB18C6,
dibenzo-18-crown-6; ‡Â, CTMAB, C16H33+CH3)3NBr. Condi-
tions: 1 M ROMs and 0.03 M PT agent in 10 ml toluene
solution, 12 g KBr +<0.125 mm salt fraction), stirring speed
1500 rpm
Figure 4. `Surface af®nity' as a function of the quaternary
cation structure and its in¯uence on the IB kinetics. *, k
&, [E]0
;
‡2
surface. This would lead to a higher activation barrier of
the transition state and to stronger surface poisoning by
the onium salt.
catalysts are presented in Fig. 3 (for calculated kinetic
parameters, see Supplementary Material).
The role of anion is not so pronounced. Judging from
the first-order constants for TEAC, TEAB, TEAI and
Visually, all the studied catalysts can be subdivided
into two major groups: catalysts that cause the IB kinetics
and catalysts that result in a slow first-order reaction. The
first group is constituted by the onium salts, which are
lipophilic enough and have no aromatic substituents. The
second group is the onium salts, which either are not
lipophilic enough and poorly soluble in the reaction
medium (toluene) or have aromatic substituents. First-
order kinetics are observed in this case, because the
intrinsic reaction (consumption of AS, k‡2) is slower than
surface renewal (regeneration of AS, k‡3) from the
beginning and, therefore, the steady state is not reached.
The use of dibenzo-18-crown-6 results in an even
slower reaction. Apparently, crown ethers represent a
single class of catalysts, in the presence of which the
reaction preferably proceeds via the extraction mechan-
ism,16 as has also been suggested elsewhere.3a,9b Indeed,
crown ethers belong to a different structural and
functional group of catalysts since they bind the cation
of an inorganic salt.
from k for TBAB, TBAI and TBAH (see Supplemen-
‡2
tary Material), only a slight increase in the reactÀion rate is
observed in the order BrÀ > ClÀ > IÀ > HSO4
.
Considering the results obtained for the onium salts,
which result in the IB kinetics, we have arrived at some
interesting correlations. In Fig. 4, the calculated rate
constant k and [E]0 value are plotted against the quat
‡2
structures placed in the order TOctA, TBA, TEAA,
TCMA, CTMA. On moving from TOctA to CTMA, there
is an increase in [E]0, which takes place simultaneously
with a decrease in the reaction rate. We assume that these
correlations arise from a so-called ‘surface affinity’ of
these quats, which increases in this order (Fig. 4). We use
this abstract term to describe the comparative strength of
the quat adsorption on the surface. On the one hand, this
property would be affected by the hydrophilicity of the
cations; on the other hand, it would also depend on
structural geometric factors or how close nitrogen can
approach the surface,15b as visually demonstrated in Fig. 4.
A possible explanation for the observed correlations is
that the bulky lipophilic and symmetric quats cannot
closely approach the surface, and therefore the resulting
transition complex is weaker and the rate (k‡2) is higher.
On the other hand, bulky quats cover more surface area
per each reacting AS, so the amount of working AS is
lower, which results in a lower [E]0 value.
Surprisingly, onium salts containing one or more
aromatic substituents, such as TPMPB and TEBAC, also
seemingly form a single group of PT agents (see caption
of Fig. 3 for compound abbreviations). These salts, unlike
their aliphatic analogues, also result in a comparatively
slow first-order reaction. Low reaction rates in the
presence of these catalysts might be connected with a
stronger adsorption of the aromatic quats on the KBr
Remarkably, the catalysts, which are less efficient
from the standpoint of reaction rate (k‡2) but have a
Copyright  2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 343–348

348
S. S. YUFIT AND S. S. ZINOVYEV
higher IB magnitude ([E]0) basically result in a higher
overall rate (in terms of total conversion per time). It is
clearly seen in Fig. 3 that the higher ‘burst’ magnitude
means a faster reaction, because the k‡3 constant, which
mainly affects the steady-state rate, is virtually equal in
most cases (see Supplementary Material).
activation, PT catalyst and solvent, two figures depicting
the deviation from the IB model due to salt agitation and
the effect of initial substrate concentration, and the
accompanying text are available as supplementary data at
the EPOC website at http://www.wiley.com.epoc.
Solvent In¯uence
REFERENCES
We also investigated the influences of a number of
solvents on the kinetics of Eqn. (1). Since the intrinsic
rate constant (k‡2) exerts more influence in the initial part
(‘burst’) of the kinetic profiles, only the initial rate
analysis was performed (see Supplementary Material).
Organic solvents of various types were studied, namely
non-polar (carbon tetrachloride, toluene, anisole and
hexane), polar protonic (methanol) and aprotic (DMSO,
DMF, acetonitrile, pyridine). The majority of the solvents
studied do not influence the reaction rate much. We found
no acceptable correlations of the reaction rate with
various properties of the solvents (see Supplementary
Material). However, some increase in the rate is observed
for DMSO, while the reaction is slower in methanol and
acetonitrile. These data overall suggest that the reaction
proceeds on the surface, in contrast to the homogeneous
processes where the solvent influence is more pro-
nounced. In general, the absence of a pronounced solvent
effect seems to be typical for PTC, as has also been
demonstrated by a few other papers on this topic.11,17
1. (a) Starks CM, Liotta CL. Phase-Transfer Catalysis: Principles
and Techniques. Academic Press: New York, 1978; (b) Dehmlow
EV, Dehmlow SS. Phase Transfer Catalysis (3rd rev., Engl. edn.).
VCH: Weinheim, 1993; (c) Yufit SS. Mekhanizm Mezhfaznogo
Kataliza [Phase Transfer Catalysis Mechanism]. Nauka: Moscow,
1984 (in Russian).
2. Sirovski F. Org. Proced. Res. Dev. 1999; 3: 437–441.
3. (a) Wong K, Wai APW. J. Chem. Soc., Perkin Trans. 2 1983; 317;
(b) Zwan MC, Hartner FW. J. Org. Chem. 1978; 43: 2655; (c)
Zahalka HA, Sasson Y. Chem. Soc., Chem. Commun. 1984; 1652;
(d) Perosa A, Selva M, Tundo P. J. Org. Chem. 1999; 64: 3934.
4. Melville JB, Goddard JD. Chem. Eng. Sci. 1985; 40: 2207–2215.
5. (a) Yee HA, Palmer HJ, Chen SH. Chem. Eng. Prog. 1987; 83: 33–
39; (b) Arrad O, Sasson Y. J. Chem. Soc., Chem. Commun. 1988;
148–149; (c) Esikova I. Izv. Akad. Nauk SSSR, Ser. Khim. 1989;
38: 2690, 2697; [Bull. Acad. Sci. USSR, Div. Chem. Sci. 1989;
38:2463, 2469].
6. (a) Arrad O, Sasson Y. J. Am. Chem. Soc. 1988; 110: 185–189; (b)
Sasson Y, Weiss M, Loupy A, Bram J, Pardo C. J. Chem. Soc.,
Chem. Commun. 1986; 1251; (c) Liotta CL, Berkner J, Wright J,
Fair B. ACS Symp. Ser. 1997; 659: 29–40; (d) Dehmlow EV, Raths
HCh. J. Chem. Res. (S) 1988; 384–385; (M) 1988; 2901–2917.
7. (a) Yufit SS, Kryshtal GV, Esikova IA. ACS Symp. Ser. 1997; 659:
52–67; (b) Esikova IA, Yufit SS. J. Phys. Org. Chem. 1991; 4:
149–157; (c) Esikova IA, Yufit SS. J. Phys. Org. Chem. 1991; 4:
341–345; (d) Yufit SS, Esikova IA. J. Phys. Org. Chem. 1991; 4:
336–340.
CONCLUSION
8. (a) Yufit SS, Zinovyev SS. Tetrahedron 1999; 55: 6319–6328; (b)
Yufit SS, Zinovyev SS. Rus. J. Org. Chem. 1998; 34: 1282–1285
(in Russian).
We applied the kinetic model of ‘initial burst’ to study
model substitution reactions under s/l PTC conditions.
This novel approach gives a more selective and accurate
analysis of kinetic data and allows one to estimate not
only the intrinsic rate constant, but also the potential
reactivity of the surface and the rate of surface renewal.
The results obtained suggest that the reaction takes place
on the solid salt surface and is described by two different
kinetic regimes reflecting (1) the intrinsic reaction step on
the surface, which is first order in substrate, and (2) the
physical process of surface renewal, which is more
influenced by mechanical agitation. It has also been
shown that the catalysts of the choice are lipophilic
onium salts with no aromatic substituents, while the
reaction is very slow in the presence of crown ethers.
The proposed model may find a use in other PTC
processes which are described by similar kinetic profiles,
by giving a more detailed mechanistic picture and,
occasionally, clues on how to improve the reaction rate.
9. (a) Landini D, Quici S, Rolla F. Synthesis 1975; 430; (b) Landini
D, Montanari F. J. Chem. Soc., Chem. Commun. 1974; 880; (c)
Anelli P, Quici S. J. Chem. Soc., Perkin Trans. 1988; 2: 1469–
1471.
10. (a) Cornish-Bowden A. Principles of Enzyme Kinetics. Butter-
worths: London, 1976; (b) Oullet L, Stewart JA. Can. J. Chem.
1959; 737.
11. Sirovski F, Reichardt C, Gorokhova M, Ruban S, Stoikova E.
Tetrahedron 1999; 55: 6363–6374.
12. Mitchenko SA, Dadali YuV. Russ. J. Org. Chem. 1998; 34: 190–
194 (in Russian).
13. (a) Rozovsky YaA. Geterogennye
i Khimicheskie Reaktsii.
Kinetika i. Makrokinetika [Heterogeneous and Chemical Reac-
tions. Kinetics and Macrokinetics]. Nauka: Moscow, 1980; 180–
300 (in Russian);; (b) Kutepov AM, Bondareva TI, Berengarten
MG. Obschaya Khimicheskaya Tekhnologiya [General Chemical
Technology]. Vysshaya Shkola; Moscow, 1990; 169–204 (in
Russian); (c) Frank-Kamenetsky DA, Diffuziya i Teploperedacha
v Khimicheskoy Kinetike [Diffusion and Heat-Transfer in Chemi-
cal Kinetics]. Nauka: Moscow, 1987 (in Russian).
14. Weiner H, Sneen RA. J. Am. Chem. Soc. 1965; 87: 287.
15. (a) Starks CM. Tetrahedron 1999; 55: 6261–6274; (b) Starks CM.
ACS Symp. Ser. 1997; 659: 10–28.
16. (a) Starks CM, Owens RM. J. Am. Chem. Soc. 1973; 95: 3613; (b)
Starks CM. J. Am. Chem. Soc. 1971; 93: 195.
17. (a) Danilova OI, Yufit SS. Mendeleev Commun. 1993; 165–166;
(b) Savelova VA, Vakhitova LN, Rybak VV, Kosmynin VV.
Russ. J. Org. Chem. 1996; 32: 729. (in Russian); (b) Antoine JP,
De Aquirre I, Janssens F, Thyrion F. Bull. Soc. Chim. Fr. 1980; 5–
6: II–207.
Supplementary material
Three tables describing the influences of preliminary salt
Copyright  2001 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2001; 14: 343–348