How iodide anions inhibit the phase-transfer catalyzed reactions of carbanions

Article abstract of DOI:10.1016/j.tet.2008.04.042

The inhibitory effect of iodide anions on the phase-transfer catalyzed reactions of carbanions generated in liquid-liquid two-phase systems by aqueous NaOH is due to preferential location of these anions in the interfacial region of the two-phase system-organic phase/concd aqueous NaOH. In such situation, the basic activity of NaOH in this region is decreased and the equilibrium of deprotonation of the carbanion precursors is disfavoured.

Full text of DOI:10.1016/j.tet.2008.04.042

Tetrahedron 64 (2008) 5925–5932
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
How iodide anions inhibit the phase-transfer catalyzed reactions
of carbanions
*
Mieczys1aw Ma˛kosza , Alexey Chesnokov
Institute of Organic Chemistry PAN, Kasprzaka 44/52, 01-224, Warsaw, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
The inhibitory effect of iodide anions on the phase-transfer catalyzed reactions of carbanions generated
in liquid–liquid two-phase systems by aqueous NaOH is due to preferential location of these anions in the
interfacial region of the two-phase systemdorganic phase/concd aqueous NaOH. In such situation, the
basic activity of NaOH in this region is decreased and the equilibrium of deprotonation of the carbanion
precursors is disfavoured.
Received 15 January 2008
Received in revised form 27 March 2008
Accepted 10 April 2008
Available online 12 April 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Phase-transfer catalysis
Carbanions
Iodide anions
Reaction mechanism
1. Introduction
by bromide and particularly iodide anions produced in the reaction,
which overrides the intrinsic reactivity of R–X in the two-phase
In a short communication by Jarrousse,1 it was reported that
alkylation of phenylacetonitrile proceeds efficiently in the presence
of 50% aqueous NaOH and 20 mol % of benzyltriethyl ammonium
chloride (TEBA) with ethyl chloride, much less with ethyl bromide
and ‘does not proceed’ with ethyl iodide. On the basis of this com-
munication, a general catalytic methodology for the alkylation of
arylacetonitriles via carbanions generated in an immiscible two-
phase systemdin the presence of 50% aqueous NaOH and tetraal-
kylammonium (TAA) salt acting as catalyst, was elaborated2,3 and
applied in industry.4,5 This two-phase catalytic methodology was
subsequently expanded to alkylation and other reactions of carb-
anions6 as well as inorganic anions and named Phase-Transfer
Catalysis (PTC).7 The catalytic action of TAA salts in these two-phase
systems consists of the continuous formation of lipophilic ion pairs
of the lipophilic TAA cations with the reacting anions that then
enter the organic phase where further reactions take place. Pres-
ently PTC is a well established general methodology applicable to a
variety of reactions of inorganic and organic anions, metalloorganic
compounds, etc.8 Contrary to what might be expected, the order of
reactivity of ethyl halides namely Et–Cl>Et–Br>Et–I in the alkyl-
ation of phenylacetonitrile in this system observed by Jarrousse1
was confirmed and explained in one of our first papers in this field.2
It was shown that this is an artefact and that the reaction is inhibited
systems. On the other hand, in the reaction of TAA salts of carban-
ions normal order of activity R–I>R–Br>R–Cl is observed. Similar
observation was made for the PTC reaction of alkyl halides with
inorganic anions. For example, PTC reactions of alkyl halides with
cyanide anions proceed efficiently with R–Cl, more slowly with
R–Br, and do not proceed (or are extremely slow) with R–I. Also in
these cases, it was shown that the iodide anions produced inhibit
the process. They remain as the TAA iodide in the organic phase,
hence cyanide anions are not transferred into the organic phase and
the catalytic process is arrested.7
Consider a reaction of an alkyl halide with an inorganic salt
carried out in a two-phase system: an aqueous solution of inorganic
salt NaþYꢀdthe nonpolar alkyl halide R–X or its solution in a
nonpolar solvent catalyzed by a lipophilic TAA salt QþXꢀ (Eq. 1a).
Assuming that all this salt stays in the organic phasedone can
show that concentration of inorganic anions Xꢀ and Yꢀ in the or-
ganic phase is governed by the ion-exchange equilibrium presented
in Eq. 1b.
þXꢀ
R—Xorg þ NaþYꢀ Q/ R—Yorg þ NaþXꢀ
(1a)
aq
aq
Q
þXꢀorg þ NaþYꢀaq!QþYꢀorg þ NaþXaꢀq
(1b)
The position of this equilibrium, and hence the concentration
of Xꢀ and Yꢀ anions in the organic phase is determined by the
differences in energy of hydration and energy of solvation by the
* Corresponding author. Fax: þ48 22 6326681.
E-mail address: icho-s@icho.edu.pl (M. Ma˛kosza).
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.04.042

5926
M. Ma˛kosza, A. Chesnokov / Tetrahedron 64 (2008) 5925–5932
organic solvent of these anions, the former is, as a rule, a much
more important factor. On this basis, the explanation of the in-
hibitory effect of iodide anions on the PTC reactions of inorganic
anions is straightforward. The energy of hydration of iodide anions,
due to their low charge density, is much smaller than that of ma-
jority of other inorganic anions, thus when Xꢀ¼Iꢀ the equilibrium
(Eq. 1b) is shifted to the left and concentration of Yꢀ in the organic
phase becomes negligible, hence the catalytic reaction of Yꢀ anions
with R–X is inhibited.
the difference of the free energy change for Brꢀ and Iꢀ anions in the
system (Eq. 3),
DDG0ðBrL=ILÞ [
L
D
G0ðBrLÞL
D
G0ðILÞ [ ½
D
D
G0hydrðBrL
Þ
D
G0solvðBrLÞꢁL½
G0hydrðILÞL
D
G0solvðILÞꢁ
[
DDGh0ydrðBrL=ILÞLDDGs0olvðBrL=IL
Þ
ð3Þ
where
D
G0
and
D
G0
stand for the standard free energies of
solv
hydr
The numerous observations of the inhibitory effect of Iꢀ anions
on the PTC reactions of carbanions generated in situ from C–H acids
in the presence of concentrated aqueous alkali (PTC/OHꢀ two-
phase systems) were rationalized similarly as for the reactions of
inorganic anions, namely, because their high lipophilicity iodide
anions compete successfully with carbanions for lipophilic TAA
cations of the catalyst, hence, concentration of the ion pairs:
carbanion–TAA cation (nCꢀQþ) in the organic phase is low and the
catalytic process retarded.8 Although indeed the concentration
of these ion pairs in the organic phase in the presence of Iꢀ ions is
low, such a rationalization appears insufficient and oversimplified.
Obviously, one cannot explain the preference for Iꢀ anions in
competition with carbanions for TAA cations in the organic phase
on the basis of differences in hydration and solvation energies of Iꢀ
and carbanions, similarly as in the case of PTC reactions of inorganic
anions governed by the ion-exchange equilibrium (Eq. 1b). Such
explanation would be based on the assumption that the lip-
ophilicity of Iꢀ is higher than the lipophilicity of carbanions, which
can hardly be accepted.
Moreover there are numerous reports that alkyl iodides can
be successfully applied for the PTC alkylation of carbanions.9 Ob-
viously, in these cases, the iodide anions produced during the
reaction do not inhibit the catalytic process.
In this paper, we clarify these controversies and present
a mechanistic picture of the inhibitory action of Iꢀ anions on PTC
reactions of carbanions.
The key difference between the PTC reactions of inorganic
anions and carbanions is due to the fact that in the former case, the
catalyst acts simply as an agent transferring inorganic anions
located in the aqueous phase into the organic phase, whereas the
PTC reactions of carbanions that are generated in situ are more
complicated. In the typical PTC systems, the carbanion precursor
located in the organic phase is deprotonated by base located in the
aqueous phase and the generated carbanions, thanks to the cata-
lyst, are present in the organic phase in the form of lipophilic ion
pairs with the catalyst cations. Thus, without mechanistic discus-
sion of how this process proceeds and where the deprotonation
takes place, we can write for a given CH acid a general equilibrium,
which is responsible for concentration of the carbanions in the
organic phase (Eq. 2).
hydration and solvation of Xꢀ in the aqueous and organic phases,
respectively. In a similar way as presented in Eq. 3, the effect of
differences in the C–H acidity of the various carbanion precursors
on the equilibrium can be estimated keeping Xꢀ constant:
DDG0(nC–H0/nC–H00)¼
D
G0(nCꢀ0/nC–H0)ꢀ
D
G0(nCꢀ00/nC–H00). These
reasonings were semiquantitatively confirmed by experiments in
which the equilibrium contents of carbanions in the organic phase
according to Eq. 2. was measured as a function of acidity of the
carbanion precursors and kind of Xꢀ in QþXꢀ. Ring substituted phe-
nylacetonitriles ArCH2CN (pKa’s in the range 16–24 in DMSO10) and
also phenylsulfonyl acetonitrile (pKa 1210) were used as the model
carbanion precursors, they were reacted with equimolar amounts
of tetrabutylammonium (TBA) halides in two-phase system, chlo-
robenzene/50% aq NaOH (Table 1).
The results obtained are presented in a graphical form in Fig-
ure 1. The contents of carbanions and halides in the samples of the
organic phase taken from the corresponding two-phase systems
equilibrated at þ35 ꢂC were determined by titration and given in
Table 1 as fractions (in %) of the total amount of Qþ, which was
considered to be 100%. It is assumed and verified experimentally
that when the aqueous phase consists of a 50% solution of NaOH, all
TBA cations are in the organic phase in the form of nCꢀQþ or QþXꢀ.
The assumption that all nCꢀQþ and QþXꢀ are in the organic phase,
and that no other anionic species are present in this phase, requires
that the sum of the found contents of the carbanions and halides in
the organic phase should be 100% in regard to the total starting
amount of TBA halides, which is perfectly the case. The basicity
of the organic phase determined by acidometric titration was
assigned to nCꢀQþ ion pairs because, if no ArCH2CN was added to
the system, extraction of OHꢀ anions into the organic phase by
QþXꢀ salts was negligible in the case of Xꢀ¼Brꢀ, Iꢀ (Table 1, entries
2 and 3) or took place to a very small extent in the case of TBA
chloride (entry 1). In a similar way, when no QþXꢀ was added to the
system there was no NaþOHꢀ or nCꢀNaþ detected in the organic
phase by the titration (entries 4, 8, 12, 16, 20 and 23). We have
confirmed that the results are related to the truly equilibrated
system by an experiment (entry 17 in the brackets) in which
a mixture of 50% aq NaOH and chlorobenzene solution of the
ArCH2CN and QþClꢀ was treated in a standard way to show 82%
content of the carbanions in the organic phase. Subsequent addi-
tion to the system of NaI in the amount equimolar to QþClꢀ fol-
lowed by equilibration resulted in decrease of the carbanion
contents in the organic phase from 82% to 5%, which was similar to
the value obtained when a solution of the ArCH2CN and QþIꢀ in
chlorobenzene was treated with an excess of 50% aq NaOH (Table 1,
entry 19).
KðXꢀÞ
nC—Horg þ NaþOHꢀaq þ QþXoꢀrg ! nCꢀOoþrg þ NaþXaꢀq þ H2Oaq
(2)
The position of this equilibrium, or in more strict terms, the free
energy difference (D
G0) between the right and left sides, is a func-
tion of many parameters:
D
G0 of the deprotonation reaction:
G0 of the H2O transfer: H2Oorg$H2Oaq
;
Thus, it was shown that the equilibrium ratio nCꢀ/Xꢀ in the
organic phase is a function of acidity of ArCH2CN and a nature of Xꢀ,
decreasing for a given C–H acid in the order Xꢀ¼Clꢀ, Brꢀ, Iꢀ (Fig. 1).
The same picture was observed when ‘catalytic’ amounts of QþXꢀ
(10% molar) in regard to ArCH2CN was employed, though the
fractions of Qþ accompanied with carbanions were somewhat
higher (Table 1, entries 13–15 in the brackets).
nC–HþOHꢀ$nCꢀþH2O;
D
D
G0 of the Xꢀ transfer: Xꢀorg$Xꢀaq; etc. Keeping all other com-
ponents and conditions constant, the effect of Xꢀ on the position of
the equilibrium, which governs the concentration of the carbanions
in the organic phase and is obviously responsible for the inhibitory
effect of Iꢀ on the PTC reactions can be easily determined by
comparing the equilibrium concentrations of nCꢀ according to Eq. 2
for different Xꢀ. Thus, DDG0(Brꢀ/Iꢀ), which for a given C–H acid
decides on the difference between equilibrium concentrations of
nCꢀQþ in the organic phase for QþBrꢀ and QþIꢀ is in fact equal to
On the basis of equation
D
G0¼ꢀRTln K and Eq. 3 one could
express change of the equilibrium constant
D
ln K(Xꢀ1 /Xꢀ2 ) for
a given C–H acid and two different TAA halides QþX1ꢀ and QþX2ꢀ as
(Eq. 4).

M. Ma˛kosza, A. Chesnokov / Tetrahedron 64 (2008) 5925–5932
5927
Table 1
Equilibrium contents of tetrabutylammonium salts of carbanions (nCꢀQþ) and halides (QþXꢀ) in the organic phase and the equilibrium constant K(Xꢀ) in two-phase system:
a solution of ArCH2CN and TBA halides 1:1 molar ratio in chlorobenzene/50% aq NaOH, at 35 ꢂCa
KðXꢀÞ
ArCH2CNorg þ NaþOHaꢀq þ QþXoꢀrg
ArCHCNꢀQoþrg þ NaþXaꢀq þ H2Oaq
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ!
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
PhCl=50% aq NaOH 35 ꢂC; argon
Entry
Ar
pKa of ArCH2CN
Xꢀ
nCꢀorg (%)
Xꢀ (%)
nCꢀorgþXꢀ (%)
K(Xꢀ
)
org
org
(DMSO)b
1
2
3
No ArCH2CN
d
Clꢀ
Brꢀ
Iꢀ
4(0.6)c(OHꢀ
)
94(98)c
100
98(99)c
100
d
d
d
<0.5(OHꢀ
<0.5(OHꢀ
)
)
99
99
4
5
6
7
4-MeOC6H4
4-MeC6H4
Ph
w23.4
No QþXꢀ
Clꢀ
w0
22
9
d
d
d
79
92
97
101
101
99
w2.88ꢅ10ꢀ1
w3.55ꢅ10ꢀ2
w1.57ꢅ10ꢀ3
Brꢀ
Iꢀ
2
8
9
10
11
w22.9
21.9
No QþXꢀ
Clꢀ
w0
36
11
4
d
d
d
65
90
95
101
101
99
w1.14ꢅ100
w5.54ꢅ10ꢀ2
w6.57ꢅ10ꢀ3
Brꢀ
Iꢀ
12
13
14
15
No QþXꢀ
Clꢀ
w0
d
d
d
56(83)d
16(30)d
5(8)d
44(18)d
85(71)d
98(93)d
100(101)d
101(101)d
103(101)d
w6.05ꢅ100
w1.31ꢅ10ꢀ1
w9.64ꢅ10ꢀ3
Brꢀ
Iꢀ
16
17
18
19
4-ClC6H4
w20.8
No QþXꢀ
Clꢀ
w0
82(5)e
33
d
d
100(100)e
100
d
18(Iꢀꢀ95)e
w7.78ꢅ101
w9.02ꢅ10ꢀ1
w4.68ꢅ10ꢀ2
Brꢀ
67
89
Iꢀ
10
99
20
21
22
3-ClC6H4
w19.7
w18.5
No QþXꢀ
Brꢀ
w0
53
14
d
47
87
d
d
w4.74ꢅ100
100
101
Iꢀ
w9.60ꢅ10ꢀ2
23
24
25
2,4-Cl2C6H3
No QþXꢀ
Brꢀ
w0
76
34
d
d
d
w3.20ꢅ101
26
68
102
102
Iꢀ
w9.30ꢅ10ꢀ1
26
27
C6F5
15.8
12.0
No QþXꢀ
Iꢀ
w0
58
d
d
d
w7.47ꢅ100
41
99
28
29
PhSO2
No QþXꢀ
Iꢀ
w0f
100
d
d
d
d
1
101
a
The 0.12 M solutions each of ArCH2CN and TBA halides in chlorobenzene were used unless indicated otherwise. Initial ratio nC–H/OHꢀz1/100.
b
According to Bordwell.10 Sign w denotes an estimated pKa value that was calculated from the Hammett equation on the basis of pKa of phenylacetonitrile and
r values
reported by Bordwell10 and the literature data on
s
and sꢀ values of substituents.16
c
NaCl (5 equiv) in regard to Qþ was added to the system.
d
e
f
Initial ratio nC–H/QþXꢀ¼10/1, i.e., 1.2 M in regard to the C–H acid and 0.12 M in regard to QþXꢀ solution in chlorobenzene was used.
NaI (1 equiv) in regard to QþClꢀ was added to the system, which was equilibrated previously with QþClꢀ.
Three-phase system: organic phase, aqueous NaOH and dispersion of white precipitatedPhSO2CHCNꢀNaþ.
R¼8.314 J/mol K) and the literature data on DDG0hydr(X1ꢀ/X2ꢀ) are 3.4
(Brꢀ/Iꢀ) and 2.1 (Clꢀ/Brꢀ) that is in a reasonably good agreement
with the experimental results obtained in our studies.
D
lnKðXL1 =XL2 Þ [ lnKðX1LÞLlnKðXL2 Þ [ LDDG0ðX1L=XL2 Þ=RT
[ ½DDGs0olvðX1L=XL2 ÞLDDG0hydrðXL1 =XL2 Þꢁ=RT
Assuming that jDDG0solvjꢃjDDG0hydrj, because of relatively low
ð4Þ
ForthecarbanionprecursorsofrelativelyhighC–Hacidity, pKa<10
in DMSO, the position of the equilibrium presented in Eq. 2 is strongly
values of the solvation energy of inorganic anions Xꢀ in an aprotic
organic medium of low polarity such as chlorobenzene, Eq. 4 can be
substantially reduced to Eq. 5.
shifted to the right, sothe differences D
ln K(Xꢀ1 /Xꢀ2 ) cannot be reliably
measured in direct experiments. This was confirmed in experiments
similar to those presented in Table 1dthe determination of the
contents of carbanions and halide anions in the organic phase when
a solution of much stronger CH acid such as phenylmalononitrile
(pKa¼4.2 in DMSO10) and equimolar amount of TBA halides (Xꢀ¼Brꢀ
or Iꢀ) in chlorobenzene was equilibrated with an excess of 50% aq
NaOH. Titrimetric analysis of the organic phase indicated that prac-
tically all the Qþ cations are accompanied with the carbanions re-
gardless of the kind of Xꢀ (Iꢀ or Brꢀ) anions (Table 2). Similar results
gave experiments in which sodium hydroxide solution contained
large quantities of dissolved NaI (Table 2 entries 4–6). Phenyl-
malononitrile being a very strong C–H acid is completely deproto-
nated with an excess of aqueous NaOH, but the formed sodium salt is
not soluble in the organic phase, in our case chlorobenzene, thus are
no carbanions or other basic anions in the organic phase unless
quaternary ammonium salt QþXꢀ is added (Table 2, entry 1).
D
lnKðXL1 =X2LÞyL½DDGh0ydrðX1L=X2LÞꢁ=RT
(5)
According to Eq. 5,
a kind of a C–H acid that is confirmed by the experimental data
shown in Table 1. The average experimental values of
ln K(Xꢀ1 /Xꢀ2 )
for the tested C–H acids calculated on the basis of Eq. 6 were as
follows:
D
ln K(Xꢀ1 /Xꢀ2 ) values should not depend on
D
D
ln K(Brꢀ/Iꢀ)¼3.3ꢄ0.7;
D
ln K(Clꢀ/Brꢀ)¼3.4ꢄ1.6.
KðXLÞyð½nCLQDꢁorg½XLꢁaq½H2Oꢁaq=½nC—Hꢁorg½OHLꢁaq
equilibrium
½QDXLꢁorg
Þ
ð6Þ
Coe et al.11 reported the following values of DDG0hydr(X1ꢀ/Xꢀ2 ) for
anions hydrated by one molecule of water as it is in 50 wt % aqueous
NaOH: ꢀ8.6 kJ/mol for (Brꢀ/Iꢀ) and ꢀ5.3 kJ/mol for (Clꢀ/Brꢀ). The
calculated values of
D
ln K(Xꢀ1 /Xꢀ2 ) according to Eq. 5 (T¼308.15 K,

5928
M. Ma˛kosza, A. Chesnokov / Tetrahedron 64 (2008) 5925–5932
100
90
80
70
60
50
40
30
20
10
0
(Cl−)
(Br−)
C−Q+
(%)
(I−)
(org)
12
14
16
18
20
22
24
pK of ArCH CN (DMSO)
a
2
Figure 1. Equilibrium contents of tetrabutyl salts of carbanions (nCꢀQþ) in the organic phase as a function of acidity of ArCH2CN and a kind of Xꢀ anions in two-phase system:
a solution of ArCH2CN and TBA halides 1:1 molar ratio in chlorobenzene/50% aq NaOH (see also Table 1).
The data presented in Tables 1 and 2 show that the equilibrium
concentration of nCꢀQþ in the organic phase generated from CH
acids and QþXꢀ in the two-phase system: nonpolar solvent/50% aq
NaOH according to Eq. 2 is determined by two factors: CH acidity of
the carbanion precursor and the hydration energy of Xꢀ. However,
for the equilibrium concentration of carbanions in the organic
phase according to Eq. 2. The alkylation rate constant kRx increases
13
in the order kRCl<kRBr<kRI whereas due to inhibitory effects of
produced Xꢀ, the observed rate of PTC alkylation of carbanions
generally decreases from R–Cl to R–I,1,2 because the equilibrium
and effective concentration of carbanions in the organic phase in
the course of PTC alkylation of a given C–H acid with R–I in the
presence of aq NaOH should be significantly lower than in the
course of alkylation with R–Br. A similar relation should be ob-
served for the alkylation with R–Br and R–Cl. A rigorous description
of the kinetics of PT catalyzed reactions is therefore a complicated
task requiring consideration of the mass transfer and chemical re-
actionstepsinvolvedintheprocess.However, onecouldsupposethat
the effective concentration of ionpairs nCꢀQþ in the organic phase in
the course of PTC alkylation should correlate with the equilibrium
constant K(Xꢀ) of these ion pairs formation process (Eq. 2).
in the case of CH acids of high acidity, the gain of energy
D
G0 of
the deprotonation nC–HþOHꢀ#n CꢀþH2O is so large that it offsets
the differences of the hydration energy of Xꢀ, thus the differences
between equilibrium concentrations of carbanions in the organic
phase as a function of a kind of Xꢀ become negligible or rather
beyond the range of determination by the used method. In this
case, the equilibrium content of carbanions in the organic phase is
almost 100% in regard to TAA cations regardless of a kind and
concentration of Xꢀ anions present in the system (see Table 2).
Since in two-phase systems, nonpolar solvents/50% aq NaOH,
the phases might be regarded as completely mutually immiscible12
assuming that rate of deprotonation and transfer of ion pair is a fast
process (equilibrium establishment is a fast process), the observed
rate of alkylation for a given of C–H acid can be written down as
presented in Eq. 7.
On the basis of this reasoning, one could assume that the final
outcome of PTC alkylation of CH acids with various alkyl halides,
R–X, X¼Cl, Br, I, should be rather a complicated function of a few
parameters: acidity of CH acid, nucleophilicity of the corresponding
carbanions, intrinsic activity of R–X, inhibitory effect of Xꢀ, etc.
In order to demonstrate the complication of the system and
interplay of these parameters, we have carried out PT catalyzed
alkylation of three arylacetonitriles differing in CH acidity with
n-Pr–X, X¼Cl, Br, I, under identical conditions. The results are
shown in Table 3.
eff
Rate of alkylation [ kRx½nCLQD
ꢁ
½RXꢁ
(7)
org
org
It is obvious that during the course of the alkylation, the effec-
tive concentration of the carbanions is lower than equilibrium
concentration [nCꢀQþ]efforgꢆ[nCꢀQþ]eqorg, where the latter stands
Phenylacetonitrile is the weakest of these three CH acids, so the
equilibrium concentration of the corresponding ion pairs ArCH-
CNꢀQþ is the most affected by the nature of Xꢀ anions generated in
Table 2
Equilibrium contents of tetrabutylammonium salts of carbanions (nCꢀQþ) and ha-
lides (QþXꢀ) in the organic phase in system: solid sodium salt of phenyl-
malononitrile–a solution of TBA halides in chlorobenzene–aqueous NaOH, at 35 ꢂ
C
Table 3
ꢀ
ꢀ
þ
PhCðCNÞ2 Naþsolid þ QþXorg
PhCðCNÞ Q
ꢀ
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ!
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
2
org
Influence of C–H acidity of ArCH2CN on PTC alkylation of ArCH2CN with n-propyl
halides in chlorobþenꢀzene/50% aq NaOH two-phase system in the presence of 2 mol %
of TBA halides (Q X ) at 35 ꢂCa
PhCl=aq NaOH; 35 ꢂC; argon
þ NaþXꢀaq
CN
Entry Aqueous
phase
Xꢀ
Ratio nCꢀ/Xꢀ nCꢀ
Xꢀ
nCꢀorgþXꢀ
2% mol.Q+X−
PhCl / 50% aq. NaOH
35 °C
org
org
org
X
+
Ar
CN
Ar
(%)
(%)
(%)
1
2
3
50% aq NaOH
No QþXꢀ
Brꢀ
d
<0.5
100
100
d
3
d
1:1
1:1
103
102
Iꢀ
2
Entry Ar
pKa of ArCH2CN Time
Conversion of ArCH2CNa (%)
(DMSO)
4
5
6
NaX (2.0 M) in Iꢀ
1:11
1:101
1:101
99
85
96
4
12
4
103
97
100
X¼Cl
51
2
0
X¼Br
13
66
4
X¼I
2
20
24
50% aq NaOH
Iꢀ
1
4
5
Ph
2,4-Cl2C6H3 18.5
C6F5 15.8
21.9
1 h
10 min
24 h
Brꢀ
7
NaX (2.0 M) in Iꢀ
20% aq NaOH
1:101
98
6
104
a
By GLC.

M. Ma˛kosza, A. Chesnokov / Tetrahedron 64 (2008) 5925–5932
5929
the course of the reaction. On the other hand, its carbanions are
strong nucleophiles thus they react satisfactorily with n-Pr–Cl.
The highest conversion with n-Pr–Cl indicates that the process is
controlled by the equilibrium concentration of ArCHCNꢀQþ in the
organic phase.
According to the interfacial mechanism the crucial stepd
deprotonation of the CH acidsdproceeds in the interfacial region. It
is followed by the ion exchange of interfacially located nCꢀNaþ with
QþXꢀ to produce lipophilic ion pair that enters subsequently the
organic phase Eq. 10, Figure 1.
Pentafluoroacetonitrile is the strongest CH acid, so the equilib-
rium concentration of the ion pairs ArCHCNꢀQþ in the organic
phase is high, close to the starting concentration of the QþXꢀ re-
gardless of the kind of Xꢀ and is not affected by the produced Xꢀ.
On the other hand, carbanions of this nitrile are weak nucleophiles
thus the observed conversion, highest for n-Pr–I is determined by
chemical activity of R–X (value of KRx in Eq. 7). Due to high acidity of
this carbanion precursor, the process is not inhibited by Iꢀ anions.
2,4-Dichlorophenylacetonitrile occupies the intermediate posi-
tion. It is a stronger CH acid than PhCH2CN, so the equilibrium con-
centration of ArCHCNꢀQþ in the organic phase is higher but still
significantly affected by Iꢀ. On the other hand, n-Pr–Br is a more
active alkylating agent than n-Pr–Cl. The observed conversion is
governed partially by the rate constant of alkylation (for n-Pr–Cl) and
partiallybythe equilibriumconcentrationoftheionpairs(forn-Pr–I).
First experiments in which degree of PTC alkylation of ring
substituted phenylacetonitriles with various alkyl halides was
connected with CH acidity of the nitriles and kind of leaving groups
were carried out by W. Lasek, Ph.D. thesis, Institute of Organic
Chemistry, 1994.
These results indicate unambiguously that the inhibitory effect
of the bromide and iodide anions on the PTC alkylation of carb-
anions with alkyl halides is not connected with the competition
between these anions and carbanions for the lipophilic cations of
the catalyst. This conclusion is unambiguously confirmed by direct
competition between Iꢀ and carbanions of phenylmalononitrile
that are less lipophilic than that of phenylacetonitrile. In such
a competition, only nCꢀQþ is transferred to the organic phase
(Table 2). Thus, one can conclude that Iꢀ ions exert an inhibitory
effect not on the extraction of the carbanions but on deprotonation of
the carbanion precursorsdhence the inhibitory effect should be
correlated with the CH acidity of the latter. For analysis of this
question, the crucial equilibrium (Eq. 2) can be divided intotwo parts:
deprotonation of the carbanion precursor and the ion exchange.
nCHint þ NaOHint!nCꢀNaþ þ H2Oint
(10a)
(10b)
(10c)
(10d)
int
nCꢀNaþint þ QþXiꢀnt!nCꢀQiþnt þ NaXint
nC—Horg/nC—Hint NaOHaq/NaOHint
nCꢀQþint/nCꢀQþorg etc:
The chemical steps 10a and 10b are connected with transfer
processes 10c and 10d.
For reactions proceeding according to the extraction mecha-
nism, the reason for the inhibitory effect of iodide and bromide
anions is obviousdthe unfavourable equilibrium 10a for the
extraction of OHꢀ anions into the organic phase. However, this
explanation is inconsistent with significant differences between
effect of Brꢀ and Iꢀ anions and also differences of these effects for
CH acids of different pKa. On the other hand, the explanation of the
inhibitory effect of iodide (and bromide) anions on PT catalyzed
reactions of carbanions and also results presented in Tables 1 and 2
are reasonably rationalized by the interfacial mechanism.
In 50% aqueous NaOH there is about one molecule of water per
each ion of Naþ or OHꢀ that is much less than normal hydration
number of these ions.
The low hydration energy of Xꢀ compared with that of OHꢀ
(Table 4) suggests that Xꢀ anions (especially Iꢀ) would preferen-
tially occupy the surface (interfacial region) of the aqueous phase,
hence they would significantly reduce the interfacial concentration
(activity) of OHꢀ anions in the two-phase systems. This phenom-
enon that can be expressed as in Eq. 11, should strongly affect the
deprotonation equilibrium (8a).
Xꢀaq þ OHꢀint>Xꢀint þ OHaꢀq
(11)
Adsorption of lipophilic Iꢀ anions at the interface has been al-
ready suggested in order to explain the observed inhibitory effect of
Iꢀ anions on the rate of hydroxide ion promoted interfacial deute-
rium exchange and alkylation reactions of phenylacetonitrile car-
ried out in the unagitated two-phase systems (‘flat’ interface) in the
absence of a phase-transfer catalyst.15 However, in such a system,
the measured (observed) rate of the deuterium exchange does not
reflects rate of deprotonation because deuteration of the in-
termediate carbanions can be the rate limiting step. Therefore, we
have studied effect of iodide anions on the rate of the isotope ex-
change of arylacetonitriles in the vigorously stirred two-phase
systems under conditions that assure that observed rate is indeed
determined by the rate of deprotonation (Eq. 12).
nC—Horg þ NaOHaq!nCꢀNaþ þ H2Oaq
(8a)
(8b)
int
nCꢀNaiþnt þ QþXꢀorg!nCꢀQþorg þ NaXaq
As was already shown, Eq. 8b is shifted to the right regardless of
ꢀ and thus cannot be responsible for the inhibitory effect of iodide
X
anions. On the other hand, Xꢀ is not present in Eq. 8a, thus their
effect on the position or the equilibrium 8a should be connected
with deprotonation of the CH acids and hence a function of pKa and
concentration of NaOH. Thus, simple thermodynamic consideration
does not offer sufficient explanation of the inhibitory effect, because
it does not takes into account how the process occurs.
K
ꢀD
KH
nC—Dint þ OHꢀint % nCꢀint þ HODintðH2OintÞ % nC—Hint
In order to clarify the situation, mechanistic concepts should be
included in the consideration. There are two accepted distinct
mechanisms of operation of PTC: extraction7 and interfacial14 mech-
anisms. In application to the reactions of carbanions, the extraction
mechanism comprises transfer of TAA hydroxides to the organic
phase where they act as base abstracting protons from CH acids Eq. 9.
KD
K
ꢀH
À
Á
þ ODꢀint OHꢀint
(12)
It should be stressed that experimental verification of the
above-mentioned effect in the D/H exchange model reaction (Eq.
12) is straightforward only under the condition kH[kD[kꢀD. If so,
QþXoꢀrg þ NaþOHaꢀq!QþOHoꢀrg þ NaþXꢀaq
(9a)
Table 4
Bulk average differences of enthalpies and free energies of hydration of OHꢀ and Xꢀ
anions (kJ/mol)11
nCHorg þ QþOHoꢀrg!nCꢀQþorg þ H2Oorg/aq
(9b)
Xꢀ
D
H0aq(OHꢀ)ꢀ
D
H0aq(Xꢀ
)
D
D
G0aq(OHꢀ)ꢀ G0aq(Xꢀ)
Thepositionoftheion-exchangeequilibriumwhenXꢀ¼Brꢀ andIꢀ
is very unfavourable for formation of QþOHꢀ and the concentration of
OHꢀ anions in the organic phase is negligible. In fact, the difference of
this concentration between Brꢀ and Iꢀ is meaningless.
Clꢀ
Brꢀ
Iꢀ
ꢀ152.7
ꢀ183.7
ꢀ224.9
ꢀ126.6
ꢀ153.4
ꢀ190.8

5930
M. Ma˛kosza, A. Chesnokov / Tetrahedron 64 (2008) 5925–5932
then for a C–H acid such as, for instance, phenylacetonitrile
(pKa¼21.9 in DMSO10 when kD[kꢀD) the observed rate of deute-
rium exchange is determined by the step of proton (deuteron)
abstraction from the substrate (rate¼kꢀD[nC–D]int[OHꢀ]int) and
therefore directly depends on the interfacial concentration of
hydroxide. If kHꢃkD, then the effective concentration of carbanions
in the course of the reaction is almost as in equilibrium (rate¼
kH[nCꢀ]eqint[H2O]int) and generally should be reduced with a
decrease of the interfacial concentration of OHꢀ anions, if Xꢀ anions
would occupy the interface. However, such adsorption should also
increase the activity of water in the interfacial area because
bonding of H2O with OHꢀ anions is much stronger than with
Xꢀ anions (Table 4). That should in turn increase the rate of
protonation of the formed carbanions and therefore the observed
rate of the isotope exchange reaction in such a case might not be
practically affected by eventual adsorption of Xꢀ at the interface.
For experimental verification of these suppositions, we have
chosen as a model the CH acid 4-methylphenylacetonitrile (pKa
22.9 in DMSO) fully deuterated in the methylene group. The
progress of deuterium exchange was determined using an 1H NMR
technique following the change of intensity of signal of the
methylene group protons in relation to the protons of the methyl
though much less profound, was observed in the case of bromide
anions.
These results support the hypothesis that the effective concen-
tration of carbanions of such C–H acids as phenylacetonitrile in the
course of PTC alkylation in the presence of aqueous NaOH depends
on the interfacial concentration of OHꢀ anions (see Eq. 11), which in
turn depends on a kind of Xꢀ anions produced in the reaction, being
the lowest for iodide anions.
It must be noted that the adsorption of iodide anions at
the interface generally should result in a high effective concentra-
tion of Iꢀ anions involved in the processes of the ion-exchange
equilibria (e.g., nCꢀ/Iꢀ competition for TAA cations), which take
place between the phases. Entries 5 and 7 from Table 2 clearly
demonstrate this effect. The only parameter varied in these two
entries was concentration of aqueuos NaOH (50 and 20 wt %,
respectively). Whereas in the case of diluted NaOH, for which effect
of interfacial adsorption of Iꢀ anions might be neglected, the
equilibrium content of carbanions in the organic phase was almost
100%, in the case of highly concentrated NaOH this value was sig-
nificantly lower. However, this effect apparently is not important
for explanation of the inhibition in PTC/OHꢀ alkylation with R–I,
since it becomes considerable only with high conversion of the
substrates, so that the ratio Iꢀ/nCꢀ becomes very high in the
system.
The results and discussion presented in this paper un-
ambiguously shows that the inhibitory effect of iodide anions and
much weaker of bromide anions on PTC alkylation of carbanions
in two-phase liquid–liquid system containing concd NaOH as the
aqueous phase is not due to competition of these anions with
carbanions for the lipophilic TAA cations. As it could be supposed
a priori, and we have shown experimentally, direct competition
between carbanions and iodide anions is always favourable for the
more lipophilic anions, that is, carbanions. Since the degree of
inhibition is a function of CH acidity of CH acids, iodide anions
undoubtedly hinder deprotonation of the CH acids. Thus in-
hibition of PT catalyzed reactions of carbanions in liquid–liquid
two-phase systems containing 50% aq NaOH is due to accumula-
tion of iodide anions at the interface. In such a situation, the
activity of OHꢀ anions in the interfacial regions is decreased
group of 4-methylphenylacetonitrile (
internal standard. When 2.8 M solution of p-methylphen-
ylacetonitrile-
-d2 (pKay22.9 in DMSO10) in chlorobenzene was
d 2.4 ppm) used as the
a
a
vigorously stirred with excess of 50% aq NaOH (saturated or not
saturated with NaI), we did not observe any significant effect of
the iodide anions on the rate of isotope exchange (t1/2¼51.4 min
in the absence of NaI and t1/2¼67.8 min in the case of 2.0 M NaI in
50% aq NaOH).
The absence of the effect of iodide anions in this model system
indicates that the rate of the exchange is determined by the rate of
protonation of the generated carbanion (kHꢃkD). In such a system,
Iꢀ at the interface can affect both deprotonation and protonation
steps, thus observed results of these controversial effects could be
meaningless. In order to avoid these difficulties, we measured the
effect of Iꢀ and Brꢀ anions on the interfacial D–H exchange between
two CH acids of similar acidity: phenylacetonitrile (pKa 21.9) and 4-
methylphenylacetonitrile, Eq. 13.
-
50% NaOH, I
p-CH3C6H4CD2CN + C6H5CH2CN
p-CH3C6H4CHDCN + C6H5-CHDCN
C6H5Cl
time of the half exchange (t 1/2 min)
-
-
-
-
-
-
-
a
-
a
+
+
ð13Þ
No I (5 % Q I )
no I
2.0 M I
0.2 M I (5 % Q I ) 2.0 M Br
0.2 M Br
0.8
fast
0.8
8.1
7.4 1.6 2.9
amolar % in respect to the nitriles, Q+ = Bu4N+
The rate of protonation (migration of proton between the
carbanion and the nitrile) is not affected by the iodide anions
adsorbed at the interface. The results of the isotope exchange
and acid–base equilibrium for generation of the carbanions
disfavoured.
between 4-methylphenyl-acetonitrile-a-2D and phenylacetonitrile
in the presence of 50% aq NaOH and some anions are shown under
Eq. 13. In such a system, we observed strong inhibition of D/H
exchange between the nitriles even when the aqueous phase
contains a small amount of NaI. Introduction of catalytic amounts
of PT catalyst QþIꢀ strongly accelerates the rate of the exchange,
apparently due to an increase of the effective concentration of
carbanions in the system via formation of the nCꢀQþ ion pairs,
though addition to such system of iodide anions inhibits the rate
of the isotope exchange. This indicates that the rate of the for-
mation of nCꢀQþ ion pairs in two-phase systems containing 50%
aq NaOH is significantly reduced when the aqueous phase contains
even small amounts of iodide anions. A similar inhibition effect,
2. Experimental
2.1. General
GLC analyses were performed on Shimadzu GC-14A gas chro-
matograph. 1H, 13C and 19F NMR spectra were recorded in chloro-
form-d1 on Varian Gemini spectrometer (200 MHz for 1H, 50 MHz
for 13C and 375 MHz for 19F). Chemical shifts were assigned using
signals from CHCl3 (7.26 and 77.0 ppm for 1H and 13C, respectively)
or CFCl3 (0 ppm for 19F) as internal standards. MS spectra were
recorded on AMD 604 mass-spectrometer. Elemental analyses
were performed on Perkin–Elmer 240 instrument. Melting points

M. Ma˛kosza, A. Chesnokov / Tetrahedron 64 (2008) 5925–5932
5931
were not corrected. All experiments were carried out using mag-
netic stirrer equipped with a temperature-controlled oil bath.
C9H6N2 Calculated: for C 76.04, H 4.25, N 19.71. Found: C 76.17, H
4.04, N 19.82.
p-Methylphenylacetonitrile-a-d2 was obtained by stirring a mix-
2.2. Reagents and substrates
ture of neat p-methylphenylacetonitrile with w2 wt % solution of
NaOD in D2O (prepared by careful dissolving of Na in D2O under
argon) for 12 h at room temperature, which assures complete
Chlorobenzene, methanol, sodium chloride, bromide, iodide and
hydroxide were analytical reagent grade, 50 wt % aqueous NaOH
was prepared using distilled water and kept as a stock solution.
Substituted phenylacetonitriles (assay 97–99%) were purchased
from AldrichÒ. Tetrabutylammonium (TBA) bromide, iodide and
hydrogen sulfate (assayꢇ98%) were purchased from FlukaÒ. Com-
mercial n-propyl halides and n-dodecane were distilled before use.
Tetrabutylammonium chloride (assay Clꢀꢇ98% by potentiometric
titration with aq AgNO3) was synthesized from TBA hydrogen sul-
fate according to published procedure.17
equilibration of the hydrogen atoms on the
a-carbon of the nitrile
with deuterium. After two such exchanges with a 20:1 ratio of D2O
per nitrile in each exchange, more than 99% of the
a
-carbon
hydrogens of the nitrile (1H NMR
deuterium.
d 3.7 ppm) were replaced by
2.3. Determining of equilibrium contents of carbanions and
halides in the organic phase generated from equimolar
mixtures of ArCH2CN, PhSO2CH2CN or PhCH(CN)2 and TBA
halides (QDXL) in chlorobenzene/50% aq NaOH two-phase
system at 35 8C (Table 1)
Phenylacetonitrile (assay ꢇ99.9% by GLC) was obtained by
distillation of the commercially available product (assay 98%,
AldrichÒ).
2,4-Dichlorophenylacetonitrile (assay ꢇ99.9% by GLC) was
obtained by recrystallization of the commercially available product
(assay 98%, AldrichÒ) from EtOH. Mp 61.5–62.5 ꢂC (lit.18 62–62.5 ꢂC).
Calculated for C8H5NCl2 (%): C 51.65, H 2.71, N 7.53, Cl 38.11. Found
(%): C 51.74, H 2.84, N 7.59, Cl 38.25.
A 50 mL three-neck round-bottom flask, equipped with a gas
inlet and a magnetic stirring bar, connected to a source of argon and
an oil pump (w1 Torr), was charged with a solution of equimolar
mixture of corresponding nitrile and TBA halide in chlorobenzene
prepared in a volumetric flask (w25 mL, 0.012–0.12 M, 0.3–3 mmol
each of the components). In the case of TBA chloride and iodide,
which generally were not completely soluble in chlorobenzene at
room temperature to give a 0.12 M solution, the solution in volu-
metric flask was warmed up to 35–40 ꢂC prior to addition to the
flask. Under cooling with a dry ice–acetone bath the flask was
degassed in vacuum and filled with argon (three times). In the
stream of argon, 50% aq NaOH (15 mL, w300 mmol NaOH) or a so-
lution of NaX (X¼I, Br) in 50% aq NaOH (15 mL, 0.2–2 M, 3–30 mmol
NaX) was added and the flask was additionally degassed in the
same manner and finally filled with argon. Then the cooling bath
was replaced with an oil bath and the flask was allowed to warm up
to þ35 ꢂC for 1 h period time. The mixture was then vigorously
stirred for 15min at 35ꢄ1 ꢂC. After short separation of the phases
(2–3 min), three samples (2ꢅ3–5 mL, 1ꢅ5–10 mL) of the clear up-
per organic layer were taken by a volumetric pipette. The use of
centrifuge for accurate separation of the phases was not necessary
since the phases were well separated by gravitation. Two samples
were mixed with methanol (3–5 mL) and titrated with 0.05 M HCl
using bromophenol blue as the indicator. Chlorobenzene was
evaporated under reduced pressure from the third sample. The
residue was dissolved in methanol (10.0 mL total volume) and
potentiometrically titrated with 0.01 M AgNO3 in order to de-
termine the content of Xꢀ in the organic phase.
Pentafluorophenylacetonitrile was obtained from hexa-
fluorobenzene and ethyl cyanoacetate according to published
procedure.19 The sample obtained after distillation of the crude
product contained w2 mol % (by 1H NMR) of C6F5CH2CO2Et. The
ester impurity was removed by hydrolysis with 50% aq NaOH
according to following procedure. A solution of the contaminated
sample (5 g) in benzene (20 mL) was mixed with excess of 50% aq
NaOH (20 mL) and vigorously stirred at 40 ꢂC for 7 h. The organic
phase was separated. The aqueous phase was diluted with water
(5–10 mL) and extracted with benzene (3ꢅ10 mL). The combined
organic extracts were washed with saturated Na2CO3 (4ꢅ10 mL),
water (10 mL) and dried over MgSO4. After removal of the solvent,
the residue was distilled under reduced pressure to give 4.0 g of the
pure pentafluorophenylacetonitrile as colourless oil. Bp 89–
90 ꢂC/w12 Torr (lit.19 105 ꢂC/8 Torr). 1H NMR:
d
3.8 (m, CH2). 13C
2
n>2
NMR:
d
10.9 (
a
-C), 104.5 (td, Ar(C1), JC1F2¼17.8 Hz,
J
¼
C1Fn
1
4.2 Hz), 114.4 (CN), 137.6 (dm, Ar, JCF¼254.8 Hz), 141.6 (dm, Ar,
1JCF¼256.1 Hz),144.9 (dm, Ar, 1JCF¼250.4 Hz). 19F NMR:
d
ꢀ160.9 (m,
3
2F, F3), ꢀ152.7 (t, 1F, F4, Jortho¼20.8 Hz), ꢀ141.8 (m, 2F, F2). Cal-
culated for C8H2F5N (%): C 46.40, H 0.97, N 6.76. Found (%): C 46.31,
H 0.79, N 6.65.
Phenylsulfonylacetonitrile was synthesized on the basis of pub-
lished procedure for alkylation of sulfinates under PTC conditions.20
A mixture of PhSO2Na (8.21 g, 0.05 mol), chloroacetonitrile (3.78 g,
0.05 mol), TBA chloride (0.7 g, w2.5 mmol, w5 mol %) in DME
(10 mL) was stirred at 90–95 ꢂC for 2 h. After cooling the intensively
coloured reaction mixture was passed with Et2O through a thin (1–
2 cm) layer of SiO2. After removal of the solvents the crude product
was recrystallized twice from EtOH (20 mL and 10 mL, respectively)
to give 4.91 g (54%) of the product as bright grey crystals. Mp 113–
2.4. Comparative experiments on alkylation of ArCH2CN
(Ar[Ph, 2,4-dichlorophenyl or pentafluorophenyl) with
n-propyl halides carried out in chlorobenzene/50% aq NaOH
two-phase system in the presence of 2 mol % of TBA halides
(QDXL) at 35 8C (Table 3)
114 ꢂC (lit.20 113–114 ꢂC). 1H NMR:
d
4.1 (s, 2H, CH2), 7.7 (m, 2H, Ph),
45.7 (CH2), 110.4 (CN),
A 10 mL round-bottom flask, equipped with a magnetic stirring
bar, was charged with 50% aq NaOH (1.0 g, 12.5 mmol). The flask
was thermostated at 35 ꢂC and additionally charged with a homo-
geneous mixture of ArCH2CN (4 mmol) and 5.0 mL of chloroben-
zene solution containing n-propyl halide (1.6 M, 8 mmol), QþXꢀ
(0.016 M, 0.08 mmol, 2 mol %) and n-dodecane (0.16 M, 0.8 mmol,
20 mol %)dinternal standard. The three chlorobenzene solutions,
each containing corresponding n-PrX, QþXꢀ and the internal
standard, were prepared once in 25 mL volumetric flasks and used
for alkylation of all nitriles. Since some TBA salts (Xꢀ¼Clꢀ and Iꢀ)
partially crystallized from the chlorobenzene solutions upon stor-
age at ambient temperature, the solutions were warmed up to 35–
40 ꢂC to ensure complete dissolving of QþXꢀ salt crystals prior
7.8 (m, 1H, Ph), 8.1 (m, 2H, Ph). 13C NMR:
d
128.8 (Ph), 129.8 (Ph), 135.4 (Ph), 136.6 (Ph). Calculated for
C8H7NO2S (%): C 53.03, H 3.89, N 7.73, S 17.69. Found (%): C 53.01, H
4.03, N 7.70, S 17.90.
Phenylmalononitrile was obtained in 78% yield on the basis
of published procedure21 via the reaction of iodobenzene with
malononitrile in DMSO/K2CO3 at 120 ꢂC catalyzed by CuCl
(20 mol %). The crude product contained w5 mol % of CH2(CN)2 and
some amounts of tars was sufficiently purified by passing with
toluene through a thin (1–2 cm) layer of SiO2. Mp 66–67 ꢂC (lit.21
67–68 ꢂC). 1H NMR:
d
5.1 (s, 1H, H2), 7.5 (mzs, 5H, Ph). 13C NMR:
d
28.1 (C2), 111.7 (CN), 126.1 (Ph), 127.1 (Ph), 129.9 (Ph), 130.3 (Ph).

5932
M. Ma˛kosza, A. Chesnokov / Tetrahedron 64 (2008) 5925–5932
to addition to the reaction flask. After mixing the reagents, the
mixtures were vigorously stirred at 35ꢄ1 ꢂC. Small samples of the
organic phases were periodically taken after short stops of stirring
(see Table 4). The samples were diluted with CH2Cl2 and analyzed
by GLC. Assigned conversions of ArCH2CN were confirmed by iso-
lation of the alkylation products and unreacted nitriles by column
chromatography (SiO2, hexane/hexane/EtOAc (20/1)).
5 mol %) had been dissolved in 50% aq NaOH (1.0 mL, 1.52 g,
w19 mmol) before addition of p-methylphenylacetonitrile- -d2.
a
The mixture was vigorously stirred at 25ꢄ1 ꢂC. Small samples of the
organic phases were periodically taken after short stops of stirring
(e.g., 0.5 min, 1 min, 2 min, 5 min, 10 min, 20 min), rapidly mixed
with excess of 5% aq HCl and vigorously shaken. After separation
from the aqueous phase the samples were mixed with CDCl3 and
placed in an NMR tube by passing through a thin layer of MgSO4 in
a Pasteur pipette. The degree of deuterium exchange was de-
termined by 1H NMR technique on the basis of intensity of H signal
of the methylene group using signal of p-methylphenylacetonitrile
The following products of alkylation were isolated:
2-Phenylvaleronitrile: 1H NMR
d
1.0 (t, 3H, H5, 3JH5H4¼7.2 Hz), 1.5
3
(m, 2H, H4), 1.9 (m, 2H, H3), 3.8 (dd, 1H, H2, JH2H3¼8.2 Hz,
3
0
JH2(H3) ¼6.6 Hz), 7.3–7.4 (m, 5H, Ph) is consistent with the litera-
ture data.22 13C NMR:
d
13.4 (C5), 20.2 (C4), 37.1 (C3 or C2), 37.8 (C2
(d 2.4 ppm) used as the internal standard. Chemical shifts (ppm)
or C3), 120.8 (CN), 127.1 (Ph), 127.9 (Ph), 128.9 (Ph), 136.0 (Ph). MS
(EI, 70 eV) m/z (%): 159(30, [M]þ), 117(100, [MꢀC3H6]þ). Calculated
for C11H13N (%): C 82.97, H 8.23, N 8.80. Found (%): C 82.81, H 8.42, N
8.64.
of the methylene protons in ArCH2CN:
p-methylphenyl).
d
3.8 (Ar¼Ph) and 3.7 (Ar¼
2-(2,4-Dichlorophenyl)valeronitrile: 1H NMR
d 1.0 (t, 3H, H5,
References and notes
3JH5H4¼7.2 Hz), 1.6 (m, 2H, H4), 1.8 (m, 2H, H3), 4.3 (t (dd), 1H, H2,
3
3
3JH2H3z JH2(H3) z7.2 Hz), 7.3 (dd, 1H, Ar(H5), Jortho¼8.4 Hz, 4Jmeta
¼
¼
1. Jarrousse, J. Compt. Rend. Hebd. Seances Acad. Sci. Ser. C 1951, 232, 1424–1426.
2. Ma˛kosza, M.; Serafinowa, B. Rocz. Chem. 1965, 39, 1223–1231; Chem. Abstr. 1966,
64, 12595h.
3. Ma˛kosza, M.; Serafinowa, B. Rocz. Chem. 1965, 39, 1401–1408; Chem. Abstr. 1966,
64, 17474g.
0
1.9 Hz), 7.4 (d, 1H, Ar(H3), 4Jmeta¼1.9 Hz), 7.5 (d, 1H, Ar(H6), 3Jortho
8.4 Hz). 13C NMR:
d
13.2 (C5), 20.3 (C4), 33.9 (C3 or C2), 36.0 (C2 or
C3), 119.8 (CN), 127.8 (Ar), 129.6 (Ar), 129.7 (Ar), 132.4 (Ar), 133.2
(Ar), 134.5 (Ar). MS (EI, 70 eV) m/z (%): 229(24, [M]þ), 227(37, [M]þ),
187(67, [MꢀC3H6]þ), 185(100, [MꢀC3H6]þ), 43(28, [C3H7]þ). Calcu-
lated for C11H11NCl2 (%): C 57.92, H 4.86, N 6.14, Cl 31.08. Found (%):
C 57.78, H 4.70, N 6.00, Cl 31.41.
_
´
4. Urbanski, T.; Be1zecki, C.; Lange, J.; Ma˛kosza, M.; Piotrowski, A.; Serafinowa, B.;
Wojnowska, H, Polish Patent 46030, 1962.
5. Ma˛kosza, M.; Serafin, B.; Urban´ ski, T. Chim. Ind. 1965, 93, 537–539.
6. Ma˛kosza, M. Tetrahedron Lett. 1966, 4621–4624; Ma˛kosza, M. Tetrahedron
Lett. 1969, 673–676, 677–678; Ma˛kosza, M.; Wawrzyniewicz, M. Tetrahedron
Lett. 1969, 4659.
2-Pentafluorophenylvaleronitrile: 1H NMR
d 1.0 (t, 3H, H5,
7. Starks, C. M. J. Am. Chem. Soc. 1971, 93, 195–199.
3JH5H4¼7.3 Hz), 1.5 (m, 2H, H4), 1.8 (m, 1H, H3), 2.1 (m, 1H, (H3)0), 4.1
8. Ma˛kosza, M.; Fedorynski, M. Advances in Catalysis; Academic: New York, NY,
1987; Vol. 35, 375–422; Ma˛kosza, M.; Fedoryn´ ski, M. Catal. Rev. 2003, 45, 321;
Dehmlow, E. V.; Dehmlow, S. S. Phase-Transfer Catalysis, 3rd ed.; Chemie:
Weinheim, 1993; Starks, C. M.; Liotta, C.; Halpern, M. Phase Transfer Catalysis.
Fundamentals, Applications and Industrial Perspectives; Chapman & Hall: New
York, NY, 1994.
´
3
(t (dd), 1H, H2, 3JH2H3z JH2(H3) z7.9 Hz). 13C NMR:
d
13.1 (C5), 20.5
0
(C4), 25.7 (C3 or C2), 34.6 (C2 or C3),109.8 (td, Ar(C1), 2JC1F2¼15.9 Hz,
n>2
J
¼4.3 Hz),117.4 (CN),138.0 (dm, Ar,1JCF¼253.5 Hz),141.3 (dm,
C1Fn
Ar, 1JCF¼256.3 Hz), 144.7 (dm, Ar, 1JCF¼250.3 Hz). 19F NMR:
ꢀ155.8
d
(m, 2F, F3), ꢀ148.2 (tt, 1F, F4, 3Jortho¼21.0 Hz, 4Jmeta¼2.1 Hz), ꢀ136.9
(m, 2F, F2). MS (EI, 70 eV) m/z (%): 249(19, [M]þ), 207(100,
[MꢀC3H6]þ), 43(72, [C3H7]þ), 41(31, [C3H5]þ). HRMS (EI) 249.05870;
calculated for C11H8NF5: 249.05769. Calculated for C11H8NF5 (%):
C 53.02, H 3.24, N 5.62. Found (%): C 53.14, H 2.87, N 6.10.
9. Masuyama, Y.; Keno, Y.; Okawara, M. Tetrahedron Lett. 1976, 2967; Ma˛kosza, M.;
Danikiewicz, W.; Wojciechowski, K. Liebigs Ann. 1988, 203; Diez-Barra, E.; De La
Hoz, A.; Sanchez-Migallon, A.; Sanchez-Verdu, P.; Bram, G.; Loupy, A.;
Pedoussant, M.; Pigeon, P. Synth. Commun. 1989, 19, 293.
10. Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456–463.
11. Tissandier, M. D.; Cowen, K. A.; Feng, W. Y.; Gundlach, E.; Cohen, M. H.; Earhart,
A. D.; Coe, J. V.; Tuttle, T. R., Jr. J. Phys. Chem. A 1998, 102, 7787–7794.
12. Ma˛kosza, M.; Bia1ecka, E. Tetrahedron Lett. 1977, 183–186.
13. Smith, M. B.; March, J. March’s Advanced Organic Chemistry, 6th ed.; Wiley: New
York, NY, 2001; pp 445–449.
2.5. Measurements of rate of D/H exchange in
p-methylphenylacetonitrile-a-d2 in chlorobenzene/50% aq
NaOH two-phase system (Eq. 13)
14. Ma˛kosza, M. Pure Appl. Chem. 1975, 43, 439–462; Ma˛kosza, M.; Lasek, W. J. Phys.
Org. Chem. 1993, 6, 412.
15. Sawarkar, C. S.; Juvekar, V. A. Ind. Eng. Chem. Res. 1996, 35, 2581–2589.
16. Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
17. Brandstrom, A. Preparative Ion Pair Extraction: An Introduction to Theory and
Practice; Apotekarsocieteten Hassle, Lakemedel: Stockholm, 1974; p 144.
18. Reeve, W.; Pickert, P. E. J. Am. Chem. Soc. 1957, 79, 1932–1934.
19. Filler, R.; Woods, S. M.; Ferring, A. E.; Sheppard, W. A. Org. Synth. 1977, 57,
80–83.
20. Wildeman, J.; van Leusen, A. M. Synthesis 1979, 733–734; Bram, G.; Loupy, A.;
Sansoulet, M. C.; Strza1ko, T.; Seyden-Penne, J. Synthesis 1987, 56–59.
21. Okuro, K.; Furuune, M.; Miura, M.; Nomura, M. J. Org. Chem. 1993, 58, 7606–
7607.
A solution of p-methylphenylacetonitrile-a-d2 in chlorobenzene
(0.67 mL, 4.24 M, 2.8 mmol) was rapidly added with vigorous stir-
ring to a 10 mL flask charged with 50% aq NaOH (1.0 mL, 1.52 g,
w19 mmol) and phenylacetonitrile (0.33 mL, 2.8 mmol, 100 mol %)
or chlorobenzene (0.33 mL) to keep constant the total volume of
the organic phase (1.0 mL), if phenylacetonitrile was not added.
When effect of Xꢀ anions (Xꢀ¼Iꢀ, Brꢀ) or Qþ cations on the rate of
deuterium exchange was studied, corresponding salt NaX (0.2–
2.0 mmol, 7–70 mol %) and/or TBA iodide (0.052 g, 0.14 mmol,
22. Freerksen, R. W.; Selikson, S. J.; Wroble, R. R.; Kyler, K. S.; Watt, D. S. J. Org.
Chem. 1983, 48, 4087–4096.