Electrofugalities of 1,3-diarylallyl cations

Article abstract of DOI:10.1021/jo302766k

Heterolysis rate constants k₁ of differently substituted 1,3-diarylallyl halides and carboxylates have been determined in various solvents. The linear free energy relationship log k₁ = s _f(N_f + E_f) was found to predict the heterolysis rates (log k₁) of 1,3-diarylallyl derivatives with a standard deviation of 0.26, corresponding to a factor of 1.82 in k₁, and maximum deviation in k₁ of a factor of 5. Some systematic deviations are evident, however. Thus, 1,3-diarylallyl carboxylates always react faster and 1,3-diarylallyl chlorides always react more slowly than calculated by the quoted correlation equation when both types of leaving groups were used to determine the electrofugality parameters E_f. As 1,3-diarylalyl cations are generated faster in solvolysis reactions and also react faster with nucleophiles than benzhydrylium ions of similar thermodynamic stabilities, i.e., Lewis acidities, one can conclude that the reactions involving 1,3-diarylallyl cations proceed with lower intrinsic barriers than those involving benzhydrylium ions. The electrofugality parameters E_f of 1,3-diarylallylium ions determined in this work were combined with the electrophilicity parameters E of the corresponding cations as well as with the results on ion pair dynamics reported in preceding papers for generating the full mechanistic spectrum of 1,3-diarylallyl solvolyses.

Full text of DOI:10.1021/jo302766k

Article
pubs.acs.org/joc
Electrofugalities of 1,3-Diarylallyl Cations
Konstantin Troshin and Herbert Mayr*
Department Chemie, Ludwig-Maximilians-Universitat Munchen, Butenandtstrasse 5-13 (Haus F), 81377 Munchen, Germany
̈
̈
̈
S
* Supporting Information
ABSTRACT: Heterolysis rate constants k1 of differently
substituted 1,3-diarylallyl halides and carboxylates have been
determined in various solvents. The linear free energy relation-
ship log k1 = sf(Nf + Ef) was found to predict the heterolysis rates
(log k1) of 1,3-diarylallyl derivatives with a standard deviation of 0.26, corresponding to a factor of 1.82 in k1, and maximum
deviation in k1 of a factor of 5. Some systematic deviations are evident, however. Thus, 1,3-diarylallyl carboxylates always react
faster and 1,3-diarylallyl chlorides always react more slowly than calculated by the quoted correlation equation when both types
of leaving groups were used to determine the electrofugality parameters Ef. As 1,3-diarylalyl cations are generated faster in
solvolysis reactions and also react faster with nucleophiles than benzhydrylium ions of similar thermodynamic stabilities, i.e.,
Lewis acidities, one can conclude that the reactions involving 1,3-diarylallyl cations proceed with lower intrinsic barriers than
those involving benzhydrylium ions. The electrofugality parameters Ef of 1,3-diarylallylium ions determined in this work were
combined with the electrophilicity parameters E of the corresponding cations as well as with the results on ion pair dynamics
reported in preceding papers for generating the full mechanistic spectrum of 1,3-diarylallyl solvolyses.
construction of comprehensive nucleophilicity and electro-
philicity2 as well as nucleofugality and electrofugality3 scales.
Because of the occurrence of allylic rearrangements,
solvolyses of allyl derivatives were of particular importance
for the development of the mechanistic understanding of
aliphatic nucleophilic substitutions.4 Pioneering studies by
Goering brought new insights into structure and reactivities
of ion pairs;5 however, the mechanistic investigations were
limited by the analytical methods available at that time. In a
recent study, we employed the capability of modern HPLC to
clarify the ion-pair dynamics during the solvolyses of unsym-
metrical 1,3-diarylallyl derivatives and showed how eq 1 can be
used to predict the extent of external and internal return during
solvolysis.6 We also demonstrated that the rates of the reactions
of 1,3-diarylallyl cations 1a−h with a manifold of π- and n-
nucleophiles in various solvents can be described by eq 1, and
we determined the E parameters of the 1,3-diarylallyl cations
1a−h which are depicted in Table 1.7 We now studied the
solvolysis rates of various derivatives of cations 1a−h in order
to determine their electrofugality parameters Ef according to eq
2. We will subsequently compare the electrofugalities Ef with
the corresponding electrophilicity parameters E in order to
elucidate general relationships between electrophilicities and
electrofugalities.
INTRODUCTION
■
In recent work, we have demonstrated that the rates of the
reactions of electrophiles with nucleophiles as well as the rates
of the reverse reactions, e.g., heterolyses of covalent esters,1
follow the three-parameter linear free energy relationships 1
and 2, respectively.
RESULTS AND DISCUSSION
■
Synthesis of the Substrates. The (E)-1,3-diarylallyl
chlorides 1(a−e)-Cl, the bromide 1a-Br, the 4-nitrobenzoates
1(c,e,f)-OPNB, and the 3,5-dinitrobenzoates 1(b−f)-ODNB
were synthesized from the corresponding alcohols (obtained as
A series of variously substituted benzhydrylium ions were
chosen as reference electrophiles and electrofuges for the
Received: January 2, 2013
Published: January 31, 2013
© 2013 American Chemical Society
2649
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660

The Journal of Organic Chemistry
Article
Table 1. 1,3-Diarylallyl Cations 1a−h and Their
Electrophilicity Parameters E (from Ref 7)
X
E
1a
1b
1c
1d
1e
1f
m,m-F2
m-F
6.11
4.15
p-Br
2.85
p-Cl
2.69
H
2.70
p-Me
p-OMe
p-NMe2
1.23
1g
1h
−1.45
−7.50
Figure 1. Linear dependence of the relative conductivity κrel on the
concentration of HCl (generated by solvolysis of 1d-Cl in 60% aq
acetone (60A40W)), 25 °C.
described in ref 7) by using standard procedures as depicted in
Scheme 1.
a
Scheme 1. Syntheses of the Substrates 1(a−f)-LG
acetone and acetonitrile solutions, solvolyses of such substrates
were performed in the presence of amines or pyridines to
ensure complete dissociation of the generated acids which
resulted in a linear dependence of κ on the acid concentration.
In many cases, kinetics, which did not follow monoexpo-
nential functions (Figure 2a), were observed due to
recombination of the carbocation with the anion of the leaving
group (k−1, Scheme 2, common ion rate depression).
a
Abbreviations used: DNB = 3,5-dinitrobenzoyl, PNB = 4-nitro-
benzoyl, DMAP = 4-dimethylaminopyridine.
The highly reactive (E)-1,3-diarylallyl acetates and benzoates
1(g,h)-OAc and 1(g,h)-OBz were generated in solution by
mixing the corresponding (E)-1,3-diarylallylium tetrafluorobo-
rates (1(g,h)-BF4) with an excess of the tetrabutylammonium
carboxylates until colorless solutions were obtained.
Kinetic Experiments. As the bis(dimethylamino)-substi-
tuted cation 1h formed by solvolysis of 1h-OAc or 1h-OBz in
aqueous acetone or acetonitrile is stable under the reaction
conditions, the solvolyses of these substrates were followed
spectrophotometrically at 704 nm (absorption maximum of
1h). All other reactions were followed conductimetrically using
conventional (k1 < 0.15 s−1) or stopped-flow techniques.
Because of the direct proportionality between conductivity and
concentration of the acid or salt (in the presence of amine
additives) produced during the solvolysis (Figure 1 and Figure
S2 of the Supporting Information), the rate constants kobs were
obtained by fitting the time-dependent conductivity (κt) to the
monoexponential function (3).
obst
κt = κ∞(1 − e−k ) + C
Figure 2. Increase of conductivity during solvolyses of 1,3-
diphenylallyl chloride (1e-Cl) in 90% aq acetonitrile (90AN10W) at
25 °C (a) in the absence of added nucleophiles and (b) in the
presence of 0.30 M 4-(N,N-dimethylamino)pyridine (DMAP).
(3)
As the carboxylic acids produced during solvolyses of
diarylallyl carboxylates are only partially dissociated in aqueous
2650
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660

The Journal of Organic Chemistry
Article
1,3-diarylallyl cations 1 (E from Table 1) can be calculated by
eq 1 to be in the range of (6.5−8.1) × 108 M−1 s−1 for 1c−e
and to be diffusion-controlled (≈ 5 × 109 M−1 s−1) for the
reactions with 1a,b. At [1-Cl]0 = 10−2 M, a typical initial
substrate concentration for stopped-flow experiments in this
work, one would expect the effective recombination rate
constants k−1[Cl−] at 50% conversion ([Cl−] = 5 × 10−3 M) to
be approximately (3.3−4.1) × 106 s−1 for 1c−e and ca. 2.5 ×
107 s−1 for 1a,b. These values are of similar magnitude as the
calculated first-order rate constants for the reactions of 1c−e
Scheme 2. Simplified Solvolysis Scheme
The addition of strong neutral nucleophiles (amines,
pyridines, phosphines) was shown to suppress common ion
rate depression in benzhydryl halide solvolyses, as these
nucleophiles rapidly trap the carbenium ions without attacking
the precursor substrates in an SN2 mode.8 In order to confirm
that the solvolysis rate constants determined in this work reflect
the rates of the heterolysis step, several measurements with
increasing concentrations of amines were performed. In cases
where the kinetics in the absence or the presence of small
concentrations of amines did not follow monoexponential
functions, estimates of kobs were derived from the best fit of the
experimental curve to the exponential function (eq 3) as
depicted in Figure 2a.
with water in 60% aq acetonitrile (N = 5.05, sN = 0.89)11 ksolv
=
(7.7−11) × 106 s−1; common ion rate depression is, therefore,
observed in these cases. Lowering the water concentration in
the solvent will increase the nucleophilicity of the anionic
leaving group (N value of Cl− is 13.30 in 80% aqueous
acetonitrile) and slightly decrease ksolv (for 90% aq acetonitrile,
N = 4.56, sN = 0.94). Both factors result in an increase of the
k−1/ksolv ratio and explain why common ion rate depression
increases with decreasing water concentration in the solvent.
The same reasoning also holds for aqueous acetone of variable
water concentration.
On the other hand, the diffusionally limited reactions of 1a,b
with Cl− (5 × 10−3 M) are significantly slower than the
reactions of these cations with the solvent, ksolv(1a) = 8.6 × 109
s−1 and ksolv(1b) = 1.5 × 108 s−1; the absence of common ion
rate depression in the solvolyses of 1(a,b)-Cl can thus be
explained.
The quality of the monoexponential fits improved with
increasing concentration of amines (Figure 2b), and the values
of kobs reached a plateau after an initial increase (Figure 3). The
As the typical substrate concentrations for the investigations
of solvolyses of 1(c−g)-OCOR with conventional conductim-
etry were at least 10 times lower than those used to follow the
solvolyses of 1(c−e)-Cl with stopped-flow techniques, and the
nucleophilicities of the carboxylate anions are generally lower
than those of the halide anions,2b the extent of common ion
return observed for solvolyses of 1(c−g)-OCOR was much
smaller than that for 1(c−e)-Cl.
By multiplication of the second order rate constant k2 = 103
M−1 s−1 calculated for the reaction of the bis(dimethylamino)-
substituted cation 1h with acetate anion in 80% aq acetone
using eq 1 (E = −7.50, N = 12.50, sN = 0.60) with 10−3 M (the
upper limit of [Bu4NOAc] used in solvolyses of 1h-OAc), one
obtains a first-order recombination rate constant of keff = 1 s−1
which is much smaller than the heterolysis rate constants k1 ≥
78 s−1 for 1h-OAc in aqueous acetone and acetonitrile (Table
2). Therefore, the concentration of Bu4NOAc, which was mixed
with 1h-BF4 in order to generate an anhydrous solution of 1h-
OAc, did not have an effect on the heterolysis rate constants
listed in Table 2, which were observed when the solutions of
1h-OAc in acetone or acetonitrile were mixed with water. The
same is true for heterolyses of 1h-OBz.
Winstein−Grunwald Analysis. The values of log k1 for
the solvolyses of 1-Cl in various solvents correlate well with the
corresponding solvent ionizing powers YBenzyl12 (Figure 4). The
m values given in Figure 4 are slightly smaller than those for
solvolyses of benzhydryl chlorides of similar reactivity in
aqueous acetone and acetonitrile (m = 0.96 for diphenylmethyl
chloride (Ef = −6.03), m = 0.93 for bis(p-tolyl)methyl chloride
(Ef = −3.44)).13
Figure 3. Observed rate constants kobs for the heterolysis of 1e-Cl at
variable concentrations of 4-(N,N-dimethylamino)pyridine (DMAP)
in 90% aq acetonitrile (90AN10W) at 25 °C.
nonlinearity of the kobs vs [Nu] plots excludes the operation of
SN2 or SN2′ mechanisms, and the plateaus correspond to the
concentrations where the common ion rate depression is
completely suppressed because the carbocations generated
during the heterolysis step are completely trapped by the
solvent or by amines before they can recombine with the
leaving group. The values of kobs at the plateaus, therefore,
correspond to the heterolysis rate constants k1. The slight
dependence of the heights of the plateaus on the nature of the
added nucleophiles,8,9 is negligible for the determination of the
Ef values considering the logarithmic character of eq 2.
For the solvolyses of the 1,3-diarylallyl chlorides 1(a−e)-Cl
in aqueous acetonitrile the amount of common ion return
depends on the electrophilicities of the carbocations and the
content of water in the solvent, as previously found for
solvolyses of benzhydryl8 and trityl9 derivatives. The
occurrence of common ion return can be rationalized on the
basis of eq 1. The second-order rate constants for the reactions
of Cl− in 60% aq acetonitrile (N = 12.00, sN = 0.60)10 with the
Increasing cation stability (ca. 3 orders of magnitude in
electrophilicity) does not change the m values significantly, and
their magnitudes indicate a high carbocation character in the
transition states of the heterolyses of 1(a−e)-Cl in aqueous
acetone and acetonitrile. This conclusion can also be reached
by consideration of the rate constants of the reactions of the
2651
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660

The Journal of Organic Chemistry
Article
Table 2. Heterolysis Rate Constants k1 of (E)-1,3-Diarylallyl Halides and Carboxylates (at 25 °C)
a
b
c
electrofuge
Ef
nucleofuge
solvent
Nf
sf
k1/s−1
kcalc
k1/kcalc
1a (X = m,m-F2)
−5.07
Cl−
Cl−
Cl−
Cl−
Cl−
Cl−
Br−
Br−
Br−
Br−
Br−
Br−
Br−
90A10W
80A20W
60A40W
90AN10W
80AN20W
60AN40W
90A10W
80A20W
60A40W
90AN10W
80AN20W
60AN40W
100E
1.14
2.03
3.30
2.23
2.96
3.84
2.29
3.01
d
1.11
1.05
0.97
1.08
1.00
0.96
1.01
0.90
d
8.95 × 10−5
7.44 × 10−4
1.95 × 10−2
8.37 × 10−4
6.86 × 10−3
6.32 × 10−2
1.35 × 10−3
1.03 × 10−2
1.95 × 10−1
1.26 × 10−2
8.83 × 10−2
5.54 × 10−1
1.81 × 10−2
1.05 × 10−5
7.31 × 10−5
1.03 × 10−4
6.38 × 10−3
9.96 × 10−2
2.91
4.34 × 10−5
6.43 × 10−4
1.92 × 10−2
8.57 × 10−4
7.76 × 10−3
6.59 × 10−2
1.56 × 10−3
1.40 × 10−2
2.06
1.16
1.02
0.98
0.88
0.96
0.87
0.74
d
d
d
d
5.23
2.93
−2.34
−2.20
−2.06
1.14
2.03
3.30
2.23
3.84
−2.79
−3.41
−3.30
−2.68
−2.20
1.14
2.03
3.30
2.23
2.96
−2.34
−2.20
−2.06
1.14
2.03
2.23
2.96
−3.40
−2.79
−3.30
−2.34
−2.20
d
0.99
0.93
1.10
0.90
0.97
1.11
1.05
0.97
1.08
0.96
1.11
0.98
0.91
1.13
0.90
1.11
1.05
0.97
1.08
1.00
1.10
0.90
0.97
1.11
1.05
1.08
1.00
1.16
1.11
0.91
1.10
0.90
d
1.44
0.38
1.77
3.67
1.88
4.27
0.34
0.50
0.76
0.41
0.72
3.22
1.34
2.10
1.01
1.76
0.54
0.62
0.66
0.53
0.62
1.95
2.19
2.72
0.52
0.63
0.50
0.62
1.92
2.67
2.07
1.27
1.70
1.02 × 10−2
2.86 × 10−6
3.89 × 10−5
2.41 × 10−5
1.86 × 10−2
1.98 × 10−1
3.82
e
1b (X = m-F)
−2.70
−ODNB
80A20W
60A40W
60AN40W
90A10W
80A20W
60A40W
90AN10W
60AN40W
60A40W
80AN20W
60AN40W
90A10W
60A40W
90A10W
80A20W
60A40W
90AN10W
80AN20W
80A20W
60A40W
60AN40W
90A10W
80A20W
90AN10W
80AN20W
80A20W
60A40W
60AN40W
80A20W
60A40W
90AN10W
80AN20W
60AN40W
90A10W
80A20W
90AN10W
80AN20W
80A20W
60A40W
90AN10W
80AN20W
60AN40W
90A10W
80A20W
60A40W
90AN10W
80AN20W
60AN40W
−ODNB
−ODNB
Cl−
Cl−
Cl−
Cl−
1.29 × 10−1
8.97
3.11 × 10−1
1.24 × 101
2.41 × 10−5
2.07 × 10−5
5.63 × 10−5
2.65 × 10−5
6.12 × 10−4
5.56 × 10−1
4.93
Cl−
f
1c (X = p-Br)
−1.37
−OPNB
7.76 × 10−5
2.77 × 10−5
1.18 × 10−4
2.68 × 10−5
1.08 × 10−3
3.01 × 10−1
3.04
−OPNB
−OPNB
−ODNB
−ODNB
Cl−
Cl−
Cl−
4.92 × 101
4.47
7.45 × 101
8.49
Cl−
Cl−
2.40 × 101
2.30 × 10−4
1.79 × 10−3
1.75 × 10−3
4.10 × 10−1
4.34
3.89 × 101
1.18 × 10−4
8.18 × 10−4
6.44 × 10−4
7.95 × 10−1
6.92
1.20 × 101
5.37 × 101
3.33 × 10−5
2.47 × 10−4
3.79 × 10−4
8.32 × 10−4
4.04 × 10−3
1d (X = p-Cl)
−1.23
−0.46
−ODNB
−ODNB
−ODNB
Cl−
Cl−
Cl−
6.01
Cl−
3.34 × 101
6.40 × 10−5
6.59 × 10−4
7.86 × 10−4
1.06 × 10−3
6.87 × 10−3
8.61 × 10−4
2.54 × 10−3
7.36 × 10−3
1.93
1e (X = H)
−OPNB
−OPNB
−OPNB
−ODNB
−ODNB
−ODNB
−ODNB
−ODNB
Cl−
d
d
−2.06
1.14
2.03
2.23
2.96
−3.40
−2.79
d
0.97
1.11
1.05
1.08
1.00
1.16
1.11
d
3.59 × 10−3
5.69
4.45 × 101
8.16 × 101
3.16 × 102
2.66 × 10−3
1.63 × 10−2
2.05
0.34
0.44
0.32
0.53
0.65
1.25
Cl−
1.96 × 101
2.62 × 101
1.66 × 102
1.74 × 10−3
2.04 × 10−2
2.16 × 10−3
6.75 × 10−3
2.19 × 10−2
1.26 × 10−2
4.30 × 10−2
1.42 × 10−1
3.51 × 10−2
8.21 × 10−2
1.92 × 10−1
Cl−
Cl−
1f (X = p-Me)
1.18
−OPNB
−OPNB
−OPNB
−OPNB
−OPNB
−ODNB
−ODNB
−ODNB
−ODNB
−ODNB
−ODNB
−3.41
−3.30
−2.68
−2.34
−2.20
d
0.98
0.91
1.13
1.10
0.90
d
6.53 × 10−3
1.18 × 10−2
2.02 × 10−2
5.30 × 10−2
1.21 × 10−1
1.03
1.86
0.62
0.81
1.17
d
d
−2.06
0.97
1.40 × 10−1
1.37
2652
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660

The Journal of Organic Chemistry
Article
Table 2. continued
a
b
c
electrofuge
Ef
nucleofuge
solvent
Nf
sf
k1/s−1
kcalc
k1/kcalc
1g (X = p-OMe)
2.87
−OAc
−OAc
−OAc
−OBz
−OBz
−OAc
−OAc
−OAc
−OAc
−OBz
−OBz
80A20W
−4.73
−4.52
−4.18
−4.46
−4.19
−4.73
−4.05
−4.52
−4.18
−4.46
−3.92
1.18
1.11
1.08
1.17
1.12
1.18
1.17
1.11
1.08
1.17
1.02
5.15 × 10−3
1.22 × 10−2
6.90 × 10−2
1.32 × 10−2
2.97 × 10−2
7.85 × 101
6.39 × 10−3
1.47 × 10−2
3.85 × 10−2
1.38 × 10−2
3.32 × 10−2
9.09 × 101
5.47 × 102
1.19 × 102
2.44 × 102
1.81 × 102
3.31 × 102
0.81
0.83
1.79
0.95
0.89
0.86
0.77
1.24
1.15
0.87
1.00
80AN20W
60AN40W
80A20W
80AN20W
80A20W
1h (X = p-NMe2)
6.39
g
60A40W
4.21 × 102
80AN20W
60AN40W
80A20W
1.48 × 102
2.81 × 102
1.57 × 102
3.31 × 102
60AN40W
a
The Ef parameters for 1a−h result from the least-squares minimization of Δ2 = Σ(log k1 − sf(Nf + Ef))2 which uses the heterolysis rate constants k1
(this table) and the Nf and sf parameters of the nucleofuges given in ref 3a and listed in this table. All solvent compositions are given in vol %, A =
b
c
acetone, AN = acetonitrile, E = ethanol, W = water. The values of kcalc were obtained using eq 2 and Ef, sf, and Nf parameters listed in this table.
d
e
f
g
The values of sf and Nf are not available. −ODNB = 3,5-Dinitrobenzoate. −OPNB = 4-Nitrobenzoate. This rate constant was not included in the
correlation since the heterolysis is too fast to derive a reliable value of k1.
indicating that there is no or only a very small barrier for the
ion combination, which is equivalent to a carbocation-like
transition states.
A different situation is found for the solvolyses of 1-OCOR.
For all carboxylates (plots for 1-ODNB and 1-OPNB are
shown in Figure 5), the m values are less than 0.6 and decrease
with increasing cation stability, which is most pronounced for
the 3,5-dinitrobenzoates.14 This situation is very similar to that
for benzhydryl15 and trityl9 carboxylates, where the m-values of
Ar2CH-OCOR and Ar3C-OCOR were reported to decrease
steadily with increasing electrofugalities of the carbocations. As
in the benzhydryl and trityl series, the low m-values for allyl
carboxylates and their decrease with increasing electrofugality
can be rationalized by incompletely developed carbocation
character in the transition states.9,15 It is noteworthy that the m-
values of the allyl carboxylates 1-OCOR are larger than those of
benzhydryl carboxylates of similar reactivities (Supporting
Information), while the solvolysis rates of 1-Cl and 1-Br
(Figure 4) are less sensitive to the changes in solvent ionizing
power (i.e., smaller m) than those of their benzhydryl analogs.
Hammett Analysis. The heterolysis rates of 1(a−f)-Cl in
90, 80, and 60% aqueous acetone and acetonitrile correlate with
the sum of Hammett−Brown’s σ+ parameters16 (Figure 6 and
the Supporting Information). The ρ-values (−3.4 to −3.1) are
lower than those for benzhydryl derivatives of comparable
Figure 4. Correlation between the solvolysis rate constants of 1-Hal in
acetone/water and acetonitrile/water mixtures (from Table 2) and the
corresponding YBenzyl values.12d The line for 1d-Cl (m = 0.85) is not
drawn to avoid overlap.
1,3-diarylallyl cations with Cl− (principle of microscopic
reversibility). As discussed above, 1e and the more reactive
carbocations 1a−d react with chloride ions in these solvents
with second-order rate constants k2 > 7 × 108 M−1 s−1,
Figure 5. Correlation between the solvolysis rate constants of (a) 1-ODNB and (b) 1-OPNB in acetone/water and acetonitrile/water mixtures
(from Table 2) and the corresponding YBenzyl values.12d
2653
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660

The Journal of Organic Chemistry
Article
squared deviations Δ2 = Σ(log k1 − sf(Nf + Ef))2, one can derive
the electrofugality parameters Ef for cations 1a−h listed in
Table 2, which correspond to the intercepts on the abscissa of
the correlation lines in Figure 7.
Since the use of eq 2 enforces slopes of 1.0 for these
correlations, carboxylates are generally above and halides are
below the correlation lines, when data for both classes of
nucleofuges are used for the correlations. As a consequence one
would arrive at better predictions of solvolysis rate constants,
when different electrofugality parameters would be employed
for allyl carboxylates and allyl halides. For the sake of simplicity
and unambiguity we refrain from proliferating the number of
parameters and refer to the last column of Table 2 which shows
that also with a single set of Ef parameters, experimental and
calculated rate constants never deviate by more than a factor of
5. As most rate constants agree much better, we consider these
deviations tolerable in view of the fact that the 1,3-diarylallyl
cations are covering a reactivity range of 12 orders of
magnitude.
Comparison of 1,3-Diarylallyl and Benzhydryl Cati-
ons. Depending on the substituents, the electrofugalities of 1,3-
diarylallyl cations exceed those of analogously substituted
benzhydryl cations by 1.5 to 7.5 orders of magnitude (Figure
8). As for the electrophilicities, the electrofugality range
covered by 1,3-diarylallylium ions is smaller than that for
analogously substituted benzhydryl systems, and the difference
in electrofugalities between the two series decreases with
increasing electron-donating abilities of the substituents (Figure
8).
Figure 6. Correlation of the heterolysis rate constants k1 of 1-LG in
80% aq acetone (80A20W) with the sum of σ+ parameters of the
corresponding aryl substituents (σ+ from ref 16).
reactivities (−4.2 to −4.1).17 As in the benzhydryl series,15
smaller values of ρ are also found for the solvolyses of the allyl
4-nitrobenzoates and 3,5-dinitrobenzoates 1-OCOR (−2.8 < ρ
< −2.3, Figure 6) again indicating that the transition states of
solvolyses of 1,3-diarylallyl chlorides are more carbocation-like
than those for 1,3-diarylallyl carboxylate solvolyses.
Determination of the Electrofugality Parameters.
Figure 7 shows that (log k1)/sf for the solvolyses of 1(a−h)-
LG correlates linearly with Nf. By minimizing18 the sum of the
The differences can be explained in the same way as in the
case of the electrophilicities:7 As the electron-demand of the
Figure 7. Plot of (log k1)/sf vs Nf for solvolyses of 1(a−h)-LG in various solvents. Abbreviations used for the solvents are A (acetone), AN
(acetonitrile), and W (water). All solvent ratios are given in % v/v (i.e., 80A20W means 80% v/v aq acetone). Abbreviations for leaving groups are:
−OPNB = 4-nitrobenzoate, −ODNB = 3,5-dinitrobenzoate.
2654
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660

The Journal of Organic Chemistry
Article
As previously reported,7 the electrophilicities of the allyl
cations 1a−h increase with increasing methyl anion affinities
(MAA, eq 4, red dots in Figure 10a).
Correspondingly, their electrofugalities decrease with in-
creasing methyl anion affinities (red dots in Figure 10b). It is
remarkable however, that in both correlations the red dots are
above the black squares (benzhydrylium ions) indicating that
1,3-diarylallyl cations are both better electrophiles (Figure 10a)
and better electrofuges (Figure 10b) than benzhydrylium ions
of equal methyl anion affinities.19 One, therefore, can conclude
that the reactions of 1,3-diarylallyl cations proceed with lower
intrinsic barriers than the corresponding reactions of
benzhydrylium ions, i.e., the heterolyses of the diarylallyl
derivatives 1-LG (and the reverse ion recombinations) require
less reorganization than the corresponding reactions of
benzhydryl derivatives. As a consequence of the different
intrinsic barriers, 1a−h are more electrophilic than benzhy-
drylium ions of similar electrofugality and better electrofuges
than diarylmethyl cations of similar electrophilicity. More
specific: The 1,3-bis(p-tolyl)allyl cation 1f is both a better
electrophile (E = 1.23) and a better electrofuge (Ef = 1.18) than
the 4,4′-dimethoxybenzhydryl cation (E = 0; Ef = 0).
This behavior can quantitatively be explained by Marcus
theory20 or the related principle of least nuclear motion:21 As
the formation of 1,3-diarylallyl cations from covalent precursors
as well as their reactions with nucleophiles require less changes
of charge densities and related movements of nuclei and solvent
molecules than the corresponding processes for benzhydrylium
Figure 8. Electrofugalities Ef of 1(a−h) compared with those of
analogously substituted benzhydryl cations.3a
carbenium centers is smaller in the 1,3-diarylallyl cations due to
the charge delocalization over two terminal allylic carbons,
substituents have a smaller effect on the rates of their reactions
as well as on the rates of their formation.
Figure 9 shows that the electrofugalities of analogously
substituted 1,3-diarylallyl and benzhydryl cations are linearly
ions of equal Lewis acidities (= methyl anion affinities), smaller
̂
reorganization energies are needed in the 1,3-diarylallylium
than in the benzhydrylium series.
Let us now analyze how the differences in intrinsic barriers
affect the correlations based on eq 2 (Figure 7). When the ion
recombination proceeds with diffusion-controlled rate, the
activation energy for the heterolytic cleavage (ΔG⧧) equals
ΔG0 of this reaction (principle of microscopic reversibility).
This is the case for many bromides and chlorides derived from
acceptor-substituted benzhydrylium and 1,3-diarylallylium ions.
As 3,5-dinitrobenzoate and 4-nitrobenzoate anions are less
nucleophilic than Cl− (≥100 times), several benzhydrylium
ions which have been used to derive the nucleofugality
parameters Nf and sf undergo diffusion-controlled recombina-
tions with Cl− and activation-controlled reactions with −ODNB
−
or OPNB. Consequently, the Nf and sf values of carboxylates
embed a larger portion of the intrinsic reactivities of the
benzhydrylium ions than the corresponding nucleofuge-specific
parameters of Cl− and Br−. As the intrinsic reactivities of the
1,3-diarylallyl cations are higher (i.e., the electrophile-specific
contributions to the barriers of their reactions are lower), the
experimental heterolysis rates of the corresponding carboxylates
are higher than those predicted based on eq 2 using the Nf and
sf parameters based on diarylmethyl cations (see Figure 7).
Acceleration of allyl carboxylate solvolysis by the interaction of
the carbonyl oxygen with the other allylic terminus cannot be
responsible for this deviation, since the contact ion pairs for
allyl systems have unsymmetrical structures.22
Figure 9. Correlation between the electrofugality parameters Ef of
analogously substituted 1,3-diarylallyl and benzhydryl cations.3a
correlated, as previously reported for the corresponding
electrophilicities of these carbocations. The slope of this
correlation (0.64) reflects the differences in electron demand
between the two series which has already been derived from
Figure 8.
Electrophilicity−Electrofugality Relationships. Figure
11, which plots the Ef values of the 1,3-diarylallyl cations 1a−h
versus their electrophilicity parameters E, demonstrates that,
2655
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660

The Journal of Organic Chemistry
Article
Figure 10. Correlations of (a) electrophilicities and (b) electrofugalities of benzhydryl and 1,3-diarylallyl cations with their calculated (B3LYP/6-
31G(d,p) level of theory) gas-phase methyl anion affinities (MAA) as defined in eq 4.
Figure 12. Pseudo-first-order rate constants observed for solvolyses of
1a-Br in 80% aq acetone in the presence of various amine
concentrations, 25 °C.
Figure 11. Electrofugality parameters of the cations 1a−h plotted vs
their electrophilicities E. The correlation line refers to the acceptor-
substituted cations 1a−d.
is given by the identical intercepts of the three correlation lines
(eq 5).
similar to the situation with benzhydryl3a,23 and trityl9 systems,
the Ef vs E correlation is not linear. The high intrinsic barriers
kobs = k1 + k2[amine]
(5)
for the reactions involving cations with strong electron donor
groups (high reorganization energies) reduce the rates of
heterolyses as well as the rates of the reactions with
nucleophiles, and this shifts 1h downward and to the left
from the extrapolated correlation line in Figure 11.
As shown in the Supporting Information, the operation of
the rate law (eq 5) has also been observed for the reaction of
1a-Br with these three amines in 90% aqueous acetone as well
as in 80% and 90% aq acetonitrile.
Table 3, which summarizes these results, shows that in all
solvents, DABCO undergoes the SN2 reaction approximately 1
order of magnitude faster than morpholine, and piperidine is
about two times more reactive than morpholine. A similar
reaction sequence for these three amines has previously been
reported for their reactions with benzhydryl bromides in
DMSO.24
In contrast to the strong solvent effects on the rates of the
SN1 reactions discussed above, the rates of the SN2 reactions are
not or only slightly affected by the solvent polarity. While the
nucleophilic substitutions proceed with equal rates in 90% and
80% aqueous acetone, they are even slightly slower in 80% than
in 90% aqueous acetonitrile in contrast to the expectations
based on the Hughes−Ingold rules, which predict a strong
Parallel SN1 and SN2 Mechanisms in Reactions of 1a-
Br with Amines. From the dependence of the hydrolysis rates
of 1,3-diphenylallyl chloride 1e-Cl in the presence of variable
concentrations of DMAP (Figure 3) we derived the operation
of an SN1 mechanism, as kobs became independent of the
concentration of DMAP when it exceeded a certain value. A
different behavior was observed when 1a-Br was treated with
variable amounts of DABCO, piperidine, or morpholine in
aqueous acetonitrile and aqueous acetone. As illustrated in
Figure 12, one now finds a linear increase of the pseudo-first-
order rate constants with increasing amine concentration,
indicating SN2 reactions of the amines with 1a-Br, which are
accompanied by an SN1 process, the rate constant of which (k1)
2656
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660

The Journal of Organic Chemistry
Article
Table 3. First-Order Rate Constants (s−1) for the Reactions of 1a-Br with Aqueous Acetone and Acetonitrile (SN1 solvolysis)
and Second-Order Rate Constants (M−1 s−1) for the Reactions of 1a-Br with Amines in these Solvents, 25 °C
k2 (M−1 s−1
)
solvent
k1/s−1
DABCO
piperidine
morpholine
90A10W
1.35 × 10−3
1.03 × 10−2
1.95 × 10−1
1.26 × 10−2
8.83 × 10−2
5.54 × 10−1
4.97 × 10−1
4.66 × 10−1
1.07 × 10−1
1.08 × 10−1
b
2.08 × 10−1
1.46 × 10−1
(3.24 × 10−1
4.49 × 10−2
80A20W
4.86 × 10−2
a
60A40W
(7.9 × 10−1
)
b
90AN10W
80AN20W
60AN40W
1.23
8.83 × 10−1
b
9.40 × 10−2
b
b
a
)
a
Due to the uncertainties caused by the fast reaction with the solvent, the k2 value cannot be precisely determined and should be considered as an
b
estimate. Not determined.
increase of the rates of SN2 reactions when ions are generated
from neutral reactants.25 Obviously, the reduction of the
nucleophilicities of the amines by water2b compensates
(aqueous acetone) or overcompensates (aqueous acetonitrile)
the expected acceleration due to the stabilization of the partially
ionic transition states by the solvent.
subsequent reaction. Crossing of the two correlation lines is
expected for an 1,3-diarylallyl chloride with a reactivity between
those of the dimethyl- and dimethoxy-substituted 1,3-diarylallyl
chlorides 1f-Cl and 1g-Cl. From log k at the point of
intersection one can derive that heterolysis of the correspond-
ing 1,3-diarylallyl chloride 1-Cl as well as solvent capture of the
generated carbenium ion occur on the submillisecond time
scale, where measurements of the rates of thermal heterolysis
are presently not possible.
As the 3,5-dinitrobenzoate anion is a weaker nucleofuge than
Cl−, the correlation line for the heterolysis rate constants of 1-
ODNB is lower, and crossing with the ksolv graph occurs for
allyl cations which are slightly better stabilized than the
dimethoxy-substituted system 1g. Because of the smaller log k
for such 3,5-dinitrobenzoates, the buildup of visually detectable
carbocation concentrations can be expected, as previously
observed in the benzhydrylium27 and tritylium9 series.
Acetate is such a weak leaving group that it was even possible
to measure the rate of the heterolytic formation of the highly
stabilized dimethylamino-substituted cation 1h from covalent
1h-OAc. Though Figure 13 suggests a slow subsequent reaction
of 1h with water, this reaction is highly reversible, and we have
observed the formation of the persistent allyl cation 1h in
aqueous acetone (and acetonitrile).
As 1a is by far the least stabilized diarylallyl cation of this
series, one can explain why amines, which are present as
trapping agents to suppress common ion return (Figures 2 and
3), do not undergo SN2 reactions with other 1,3-diarylallyl
derivatives 1(b−f)-LG.
Changes of Solvolysis Mechanisms of Allyl Deriva-
tives. Figure 13 combines the rate constants for the heterolyses
CONCLUSIONS
■
Though the linear free energy relationship log k1 = sf(Nf + Ef)
(eq 2) was found to predict heterolysis rates of 1,3-diarylallyl
bromides, chlorides, and carboxylates with an accuracy better
than factor of 5 from a single set of Ef parameters and a single
set of the benzhydrylium-based nucleofuge-specific parameters
Nf and sf, systematic deviations are evident. Thus, the 1,3-
diarylallyl carboxylates 1-OCOR react generally faster, while the
1,3-diarylallyl chlorides 1-Cl react generally more slowly than
predicted when solvolysis rates for both types of leaving groups
were used to determine the electrofugality parameters Ef.
As in the benzhydryl3a and trityl9 series, solvolyses of 1,3-
diarylallyl bromides and chlorides have significantly larger
Winstein−Grunwald m-values than carboxylates; this difference
has been explained by late and early transition states,
respectively. Variation of the substituents in the aromatic
rings affects solvolysis rates of 1,3-diarylallyl derivatives
significantly less than those of benzhydryl derivatives. As the
intrinsic barriers for heterolyses of 1,3-diarylallyl derivatives are
lower than those for the corresponding benzhydryl heterolyses,
one comes to the conclusion that 1,3-diarylallyl derivatives
ionize faster than benzhydryl derivatives of the same electro-
philicity E and, at the same time, 1,3-diarylallylium ions
Figure 13. Heterolysis rate constants k1 for 1-Cl, 1-ODNB, and 1-OAc
in 80% aq acetone and of the first-order rate constants ksolv for the
reactions of 1 with the solvent in correlation with calculated (B3LYP/
6-31G(d,p) level of theory) gas-phase methyl anion affinities (MAA)
of 1. Experimentally derived values of k1 (Table 2) are identified by
symbols, and the trend lines correspond to k1 and ksolv calculated for
1(a−h)-LG and 1a−h by eqs 2 and 1, respectively.
of allyl derivatives (k1 of Scheme 2) with the rate constants for
the reactions of the allyl cations with 80% aqueous acetone.
One can see that the acceptor-substituted allyl chlorides 1(a,b)-
Cl like the unsubstituted (E)-1,3-diphenylallyl chloride 1e-Cl
(as well as 1(c,d)-Cl) follow the typical SN1 mechanism: slow
formation of the carbocation, followed by rapid trapping by the
solvent. For k1 > ksolv (i.e., for the left part of Figure 13) the
SN2C+ mechanism26 is encountered, where a high concen-
tration of an intermediate carbocation is generated in a fast
initial step, which is slowly converted into an allyl alcohol in a
2657
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660

The Journal of Organic Chemistry
Article
and morpholine (>99%) were purchased and distilled prior to use.
Acetone (99.8%), acetonitrile (HPLC grade), DMAP (>99%), and
DABCO (98%) were purchased and used as received. Details of the
kinetic measurements are given in the Supporting Information.
combine faster with nucleophiles than benzhydryl cations of
equal electrofugality Ef.
Combination of the electrofugality parameters of 1,3-
diarylallylium ions derived in this work with their electro-
philicity parameters E7 and with the results on ion pair
dynamics reported in a preceding publication6 allows one to
construct the comprehensive solvolysis scheme.
1
Analytics. H and 13C NMR chemical shifts are expressed in δ
(ppm) and refer to CDCl3 (δH 7.26, δC 77.0), CD2Cl2 (δH 5.32, δC
54.0), or tetramethylsilane (δH 0.00, δC 0.0) as internal standards. The
coupling constants are given in Hz. Abbreviations used are: s (singlet),
d (doublet), t (triplet), q (quartet), m (multiplet). In case of 13C NMR
spectra, these abbreviations refer to the multiplicity in hydrogen
decoupled spectra and the hydrogen multiplicity (based on DEPT or
HSQC experiments) is shown as CH3, CH2, CH or C to avoid
ambiguity. For the compounds containing no fluorine, the letter “s” is
omitted. Assignments of NMR signals are based on the combined
As illustrated in Figure 14, the highly stabilized carbocation
1h in the left gray range is generated faster by heterolysis of the
1
analysis of H and 13C NMR, HSQC, HMBC, and COSY spectra.
HRMS in EI mode (70 eV) were determined with sector field
detectors. Melting points were determined using a standard melting
point unit with gradient of 0.5 °C.
All yields refer to nonoptimized procedures.
(E)-3-Bromo-1,3-bis(3,5-difluorophenyl)prop-1-ene (1a-Br).
Compound 1a-Br was obtained from 1a-OH (1.03 g, 3.69 mmol)
and phosphorus tribromide (420 μL, 1.21 g, 4.48 mmol) using a
procedure from ref 29: 708 mg (2.05 mmol, 56%), colorless solid (mp
1
3
57.0−59.5 °C). H NMR (CDCl3, 300 MHz): δ 5.68 (d, JHH = 7.1
Hz, 1 H, ArCHCHCH(Br)Ar), 6.49−6.63 (m, 2 H, ArCHCHCH-
(Br)Ar, ArCHCHCH(Br)Ar), 6.69−6.83 (m, 2 H, HAr), 6.86−6.98
(m, 2 H, HAr), 6.98−7.06 ppm (m, 2 H, HAr). 13C NMR (CDCl3, 75
MHz): δ 51.4 (t, JCF = 2.3 Hz, CH), 103.9 (t, JCF = 25.6 Hz, CH),
104.2 (t, JCF = 25.2 Hz, CH), 109.5−109.8 (m, CH), 110.6−111.0 (m,
CH), 130.6 (s, CH), 130.9 (t, JCF = 2.9 Hz, CH), 138.9 (t, JCF = 9.6
Hz, C), 143.4 (t, JCF = 9.0 Hz, C), 163.0 (dd, JCF = 248.6, 12.7 Hz, C),
163.2 ppm (dd, JCF = 247.2, 12.9 Hz, C). 19F NMR (CDCl3, 282
MHz): δ −109.7 to −109.4 (m), −108.4 to −108.1 ppm (m). HRMS
+
(EI+): calcd 265.0635 (C15H9F4 (M − Br−)), found 265.0623.
Figure 14. Changes of mechanism in solvolyses of 1,3-diarylallyl 4-
nitrobenzoates (1-OPNB) in 60% aq acetone ([−OPNB] = 5 mM).
(E)-3-Chloro-1,3-bis(3,5-difluorophenyl)prop-1-ene (1a-Cl).
Compound 1a-Cl was synthesized as reported in ref 30.
(E)-3-Chloro-1,3-bis(3-fluorophenyl)prop-1-ene (1b-Cl).
Compound 1b-Cl was synthesized from 1b−OH (530 mg, 2.15
mmol) and thionyl chloride (220 μL, 360 mg, 3.03 mmol) following
the protocol described in ref 29: 378 mg (1.26 mmol, 58%), colorless
corresponding covalent ester than it reacts with water or
carboxylate ions. As the latter two reactions are highly
reversible, 1h is thermodynamically stable under the reaction
conditions. When the stabilization of the carbocations is
decreased, one gets into the green SN2C+ range,26 where the
formation of the carbocation is faster than its reaction with the
solvent. Further destabilization of the carbocations (1g, 1f)
leads to SN1 reactions with external (common ion) return,
where the consumption of the carbocations is faster than their
generation from the covalent precursors, but the rate of
recombination with the leaving group LG− is comparable to the
rate of the reaction with the solvent (blue area). If the rate of
ion recombination is comparable to that of the diffusional
separation of the initially formed ion pairs, one observes
internal return (red area) which has extensively been studied
with unsymmetrically substituted 1,3-diarylallyl carboxylates.6
The extent of internal return is decreasing again when the
electrophilicity of the carbocation is high enough that the
solvent can attack at the stage of contact ion pairs (CIPs). An
enforced SN2 reaction can eventually be expected, when the
rate constant of the combination of the carbocation with water
(ksolv) exceeds the vibrational frequency (∼10−13 s−1, Richard−
Jencks criterion28). Equations 1 and 2 can analogously be
employed for predicting mechanistic changes in the solvolyses
of other series of substrates.
1
3
oil. H NMR (CDCl3, 400 MHz): δ 5.60 (d, JHH = 7.7 Hz, 1 H,
3
ArCHCHCH(Cl)Ar), 6.46 (dd, JHH = 15.6, 7.7 Hz, 1 H,
3
ArCHCHCH(Cl)Ar), 6.60 (d, JHH = 15.6 Hz, 1 H, ArCHCHCH-
(Cl)Ar), 6.93−6.99 (m, 1 H, HAr), 6.99−7.06 (m, 1 H, HAr), 7.06−
7.39 ppm (m, 6 H, HAr). 13C NMR (CDCl3, 101 MHz): δ 62.3 (d, JCF
= 2.0 Hz, CH), 113.3 (d, JCF = 22.3 Hz, CH), 114.5 (d, JCF = 22.7 Hz,
CH), 115.2 (d, JCF = 21.4 Hz, CH), 115.5 (d, JCF = 21.2 Hz, CH),
122.8 (d, JCF = 2.8 Hz, CH), 122.9 (d, JCF = 3.0 Hz, CH), 129.7 (d, JCF
= 0.5 Hz, CH), 130.1 (d, JCF = 8.4 Hz, CH), 130.3 (d, JCF = 8.3 Hz,
CH), 131.4 (d, JCF = 2.6 Hz, CH), 137.9 (d, JCF = 7.7 Hz, C), 142.4
(d, JCF = 7.2 Hz, C), 162.8 (d, JCF = 247.2 Hz, C), 163.0 ppm (d, JCF
=
245.9 Hz, C). HRMS (EI+): calcd 264.0512 (C15H11ClF2), found
264.0505.
Compounds 1c-Cl and 1e-Cl were synthesized as described in ref 7.
(E)-3-Chloro-1,3-bis(4-chlorophenyl)prop-1-ene (1d-Cl).
Compound 1d-Cl was synthesized from 1d-OH (487 mg, 1.74
mmol) and concentrated hydrochloric acid (1 mL) following the
protocol described in ref 31: 344 mg (1.16 mmol, 66%), colorless solid
1
(mp 77.5−80.0 °C). H HMR (CDCl3, 300 MHz): δ 5.59 (d, 1 H,
3
3JHH = 7.4 Hz, ArCHCHCH(Cl)Ar), 6.44 (dd, JHH = 15.7, 7.4 Hz, 1
H, ArCHCHCH(Cl)Ar), 6.56 (d, 3JHH = 15.7 Hz, 1 H, ArCHCHCH-
(Cl)Ar), 7.22−7.43 ppm (m, 8 H, HAr). 13C NMR (CDCl3, 75 MHz):
δ 62.6 (CH), 128.0 (CH), 128.7 (CH), 128.8 (CH), 128.9 (CH),
129.2 (CH), 131.2 (CH), 134.1 (2 × C), 134.3 (C), 138.5 ppm (C).
HRMS (EI+): calcd 296.9999 (C15H11Cl3 + H+), found 296.9987.
General Procedure for Synthesis of 1,3-Diarylallyl 3,5-
Dinitrobenzoates (1-ODNB). 3,5-Dinitrobenzoyl chloride (1−2
equiv) was added to a stirred solution of the corresponding alcohol
1-OH (1 equiv) and triethylamine (1.2 equiv with respect to DNBCl)
in toluene. After 12 h of stirring, the reaction mixture was quenched
EXPERIMENTAL SECTION
■
Materials for Kinetic Measurements. Dry ethanol was obtained
by distillation of commercially available absolute ethanol from
sodium/diethyl phthalate. Doubly distilled water (impedance 18.2
Ω) was obtained from a water purification system. Piperidine (>99%)
2658
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660

The Journal of Organic Chemistry
Article
with 0.1 M aq HCl, and the organic layer was separated, washed with
saturated NaHCO3 solution, and water, and freed from the solvent.
The crude product was purified by recrystallization from a CH2Cl2/
pentane mixture.
MHz): δ 21.48 (CH3), 21.51 (CH3), 80.0 (CH), 123.0 (CH), 125.8
(CH), 127.2 (CH), 127.7 (CH), 129.9 (CH), 130.0 (CH), 130.1
(CH), 133.6 (C), 134.4 (CH), 134.8 (C), 136.0 (C), 139.1 (C), 139.3
(C), 149.2 (C), 162.3 ppm (C). HRMS (EI+): calcd 432.1316
(C24H20N2O6), 432.1331.
(E)-1,3-Bis(3-fluorophenyl)allyl 3,5-Dinitrobenzoate (1b-ODNB).
From 1b-OH (300 mg, 1.22 mmol), DNBCl (562 mg, 2.44 mmol),
and Et3N (350 μL, 254 mg, 2.51 mmol): 199 mg (0.451 mmol, 37%),
yellowish powder (mp 130.0−131.5 °C). 1H NMR (CDCl3, 300
1,3-Diarylallyl 4-Nitrobenzoates (1-OPNB). 1-OPNB were ob-
tained following the protocol described in ref 6.
(E)-1,3-Bis(4-bromophenyl)allyl 4-Nitrobenzoate (1c-OPNB).
From 1c-OH (1.00 g, 2.72 mmol), PNBCl (760 mg, 4.10 mmol),
Et3N (500 μL, 436 mg, 4.30 mmol), and DMAP (33.0 mg, 270 μmol):
477 mg (0.923 mmol, 34%), colorless powder (mp 130.7−132.0 °C).
3
MHz): δ 6.48 (dd, JHH = 15.9, 7.1 Hz, 1 H, ArCHCHCH(ODNB)-
Ar), 6.67−6.79 (m, 2 H, ArCHCHCH(ODNB)Ar, ArCHCHCH-
(ODNB)Ar), 6.92−7.03 (m, 1 H, HAr), 7.03−7.36 (m, 6 H, HAr),
7.36−7.48 (m, 1 H, HAr), 9.18−9.22 (m, 2 H, HODNB), 9.22−9.25 ppm
(m, 1 H, HODNB). 13C NMR (CDCl3, 75 MHz): δ 78.1 (d, JCF = 2.0
Hz, CH), 113.3 (d, JCF = 22.0 Hz, CH), 114.1 (d, JCF = 22.5 Hz, CH),
115.6 (d, JCF = 21.4 Hz, CH), 116.0 (d, JCF = 21.1 Hz, CH), 122.6 (s,
CH), 122.7 (d, JCF = 3.1 Hz, CH), 122.9 (d, JCF = 2.9 Hz, CH), 126.6
3
1H HMR (CD2Cl2, 400 MHz): δ 6.47 (dd, JHH = 15.9, 6.7 Hz, 1 H,
3
ArCHCHCH(OPNB)Ar), 6.63 (d, JHH = 6.7 Hz, 1 H,
3
ArCHCHCH(OPNB)Ar), 6.69 (d, JHH = 15.9 Hz, 1 H,
ArCHCHCH(OPNB)Ar), 7.25−7.34 (m, 2 H, HAr), 7.37−7.44 (m,
2 H, HAr), 7.44−7.51(m, 2 H, HAr), 7.52−7.60 (m, 2 H, HAr), 8.23−
8.34 ppm (m, 4 H, HOPNB). 13C NMR (CD2Cl2, 101 MHz): δ 77.5
(CH), 122.7 (C), 123.0 (C), 124.2 (CH), 127.8 (CH), 128.9 (CH),
129.4 (CH), 131.4 (CH), 132.3 (CH), 132.5 (CH), 132.9 (CH),
135.5 (C), 136.0 (C), 138.3 (C), 151.3 (C), 164.2 ppm (C). HRMS
(EI+): calcd 514.9362 (C22H15Br2NO4), found 514.9379.
(E)-1,3-Diphenylallyl 4-Nitrobenzoate (1e-OPNB). From 1e-OH
(200 mg, 951 μmol), PNBCl (180 mg, 970 μmol), Et3N (160 μL, 116
mg, 1.15 mmol), and DMAP (20 mg, 164 μmol): 281 mg (782 μmol,
82%), colorless solid (mp 92.5−93.5 °C). The 1H and 13C NMR
spectra of 1e-OPNB agreed with those described in ref 32.
(s, CH), 129.5 (s, CH), 130.2 (d, JCF = 8.4 Hz, CH), 130.7 (d, JCF
8.2 Hz, CH), 133.7 (s, C), 133.8 (d, JCF = 2.6 Hz, CH), 137.7 (d, JCF
=
=
7.7 Hz, C), 140.1 (d, JCF = 7.1 Hz, C), 148.7 (s, C), 161.5 (s, C), 163.0
(d, JCF = 247.5 Hz, C), 163.1 ppm (d, JCF = 246.2 Hz, C). 19F NMR
(CDCl3, 282 MHz): δ −113.0 to −112.9 (m), −111.5 to −111.4 ppm
(m). HRMS (EI+): calcd 440.0814 (C22H14F2N2O6), found 440.0808.
(E)-1,3-Bis(4-bromophenyl)allyl 3,5-Dinitrobenzoate (1c-ODNB).
From 1c-OH (1.63 g, 4.43 mmol), DNBCl (1.03 g, 4.47 mmol), and
Et3N (750 μL, 545 mg, 5.38 mmol): 1.05 g (1.87 mmol, 42%),
yellowish powder (mp 161.5−163.1 °C). 1H NMR (CD2Cl2, 400
3
MHz): δ 6.51 (dd, JHH = 15.9, 7.0 Hz, 1 H, ArCHCHCH(ODNB)-
(E)-1,3-Di-p-tolylallyl 4-Nitrobenzoate (1f-OPNB). From 1f-OH
(1.00 g, 4.20 mmol), PNBCl (1.01 g, 5.44 mmol), Et3N (900 μL, 653
mg, 6.46 mmol), and DMAP (66.5 mg, 544 μmol): 727 mg (1.88
mmol, 45%), colorless powder (mp 92.6−94.0 °C). 1H HMR (CDCl3,
Ar), 6.68−6.74 (m,
2 H, ArCHCHCH(ODNB)Ar,
ArCHCHCH(ODNB)Ar), 7.29−7.36 (m, 2 H, HAr), 7.40−7.50 (m,
4 H, HAr), 7.56−7.62 (m, 2 H, HAr), 9.18−9.19 (m, 2 H, HODNB),
9.21−9.22 ppm (m, 1 H, HODNB). 13C NMR (CD2Cl2, 101 MHz): δ
78.8 (CH), 122.9 (C), 123.2 (CH), 123.3 (C), 127.1 (CH), 128.9
(CH), 129.5 (CH), 130.1 (CH), 132.4 (CH), 132.6 (CH), 133.8
(CH), 134.3 (C), 135.2 (C), 137.7 (C), 149.3 (C), 162.2 ppm (C).
HRMS (EI+): calcd 559.9213 (C22H14Br2N2O6), found 559.9220.
(E)-1,3-Bis(4-chlorophenyl)allyl 3,5-Dinitrobenzoate (1d-ODNB).
From 1d-OH (1.08 g, 3.87 mmol), DNBCl (900 mg, 3.90 mmol), and
Et3N (650 μL, 472 mg, 4.66 mmol): 0.851 g (1.80 mmol, 47%),
colorless powder (mp 150.0−151.5 °C). 1H NMR (CD2Cl2, 400
300 MHz): δ 2.35 (s, 3 H, CH3), 2.37 (s, 3 H, CH3), 6.43 (dd, 3JHH
=
15.9, 6.9 Hz, 1 H, ArCHCHCH(OPNB)Ar), 6.63−6.75 (m, 2 H,
ArCHCHCH(OPNB)Ar, ArCHCHCH(OPNB)Ar), 7.07−7.18 (m, 2
H, HAr), 7.19−7.26 (m, 2 H, HAr), 7.27−7.33 (m, 2 H, HAr), 7.34−
7.45 (m, 2 H, HAr), 8.24−8.31 ppm (m, 4 H, HOPNB). 13C NMR
(CDCl3, 75 MHz): δ 21.19 (CH3), 21.21 (CH3), 78.0 (CH), 123.5
(CH), 125.8 (CH), 126.7 (CH). 127.1 (CH), 129.3 (CH), 129.4
(CH), 130.8 (CH), 133.1 (C), 133.4 (CH), 135.8 (C), 135.9 (C),
138.2 (C), 138.4 (C), 150.5 (C), 163.7 ppm (C). HRMS (EI+): calcd
387.1465 (C24H21NO4), found 387.1482.
3
MHz): δ 6.50 (dd, JHH = 15.9, 7.0 Hz, 1 H, ArCHCHCH(ODNB)-
Ar), 6.66−6.77 (m,
2 H, ArCHCHCH(ODNB)Ar,
ArCHCHCH(ODNB)Ar), 7.29−7.35 (m, 2 H, HAr), 7.35−7.41 (m,
2 H, HAr), 7.41−7.46 (m, 2 H, HAr), 7.47−7.52 (m, 2 H, HAr), 9.18 (d,
3JHH = 2.1 Hz, 2 H, HODNB), 9.22 ppm (t, 3JHH = 2.1 Hz, 1 H, HODNB).
13C NMR (CD2Cl2, 101 MHz): δ 78.8 (CH), 123.2 (CH), 127.0
(CH), 128.7 (CH), 129.2 (CH), 129.4 (CH), 129.6 (CH), 130.1
(CH), 133.7 (CH), 134.4 (C), 134.7 (C), 134.8 (C), 135.2 (C), 137.2
(C), 149.3 (C), 162.2 ppm (C). HRMS (EI+): calcd 472.0223
(C22H14Cl2N2O6), found 472.0225.
ASSOCIATED CONTENT
* Supporting Information
Details of the kinetic experiments, additional Hammett and
Winstein−Grunwald correlations, and copies of the NMR
spectra of new compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
■
S
(E)-1,3-Diphenylallyl 3,5-Dinitrobenzoate (1e-ODNB). From 1e-
OH (1.00 g, 4.76 mmol), DNBCl (1.10 g, 4.77 mmol), and Et3N (800
μL, 581 mg, 5.74 mmol): 731 mg (1.81 mmol, 38%), colorless powder
(mp 154.1−155.5 °C). 1H HMR (CDCl3, 300 MHz): δ 6.52 (dd, 3JHH
= 16.0, 7.1 Hz, 1 H, PhCHCHCH(ODNB)Ph), 6.70−6.83 (m, 2 H,
PhCHCHCH(ODNB)Ph, PhCHCHCH(ODNB)Ph), 7.23−7.48 (m,
8 H, HAr), 7.48−7.56 (m, 2 H, HAr), 9.15−9.25 ppm (m, 3 H, HODNB).
13C NMR (CDCl3, 75 MHz): δ 79.3 (CH), 122.4 (CH), 125.9 (CH),
126.8 (CH), 127.1 (CH), 128.6 (CH), 128.7 (CH), 128.9 (CH),
129.0 (CH), 129.5 (CH), 134.1 (C), 134.5 (CH), 135.6 (C), 138.0
(C), 148.7 (C), 161.6 ppm (C). HRMS (EI+): calcd 404.1003
(C22H16N2O6), found 404.1005.
AUTHOR INFORMATION
Corresponding Author
*E-mail: herbert.mayr@cup.uni-muenchen.de.
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the Deutsche Forschungsgemeinschaft (Ma 673/22-
1) for financial support. We are grateful to Anna Antipova,
Johannes Ammer, and Dr. Armin R. Ofial for their help during
preparation of this manuscript.
(E)-1,3-Di-p-tolylallyl 3,5-Dinitrobenzoate (1f-ODNB). From 1f-
OH (1.00 g, 4.20 mmol), DNBCl (968 mg, 4.20 mmol), and Et3N
(700 μL, 508 mg, 5.02 mmol): 900 mg (2.08 mmol, 50%), colorless
powder (mp 84.5−86.0 °C). 1H HMR (CD2Cl2, 400 MHz): δ 2.33 (s,
REFERENCES
■
3
(1) More precisely, the generation of the free ions from a covalent
substrate should be termed as ionization plus dissociation: Reichardt, C.;
Welton, T. Solvents and Solvent Effects in Organic Chemistry, 4th ed.;
Wiley: Weinheim, 2011; p 52. To avoid the long terms, we used the
term heterolysis as a description of this process.
3 H, CH3), 2.37 (s, 3 H, CH3), 6.49 (dd, JHH = 15.9, 7.2 Hz, 1 H,
ArCHCHCH(ODNB)Ar), 6.66−6.76 (m, 2 H, ArCHCHCH-
(ODNB)Ar, ArCHCHCH(ODNB)Ar), 7.11−7.17 (m, 2 H, HAr),
7.22−7.28 (m, 2 H, HAr), 7.28−7.35 (m, 2 H, HAr), 7.40−7.46 (m, 2
H, HAr), 9.16−9.23 ppm (m, 3 H, HODNB). 13C NMR (CD2Cl2, 101
2659
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660

The Journal of Organic Chemistry
Article
(2) (a) Mayr, H.; Kempf, B.; Ofial, A. R. Acc. Chem. Res. 2003, 36,
66−77. (b) For a database of reactivity parameters E, N, and sN see:
www.cup.lmu.de/oc/mayr/DBintro.html, (last accessed on November
2, 2013).
(19) The values obtained at B3LYP/6-31G(d,p) level of theory were
used, as they are available both for 1a−h (ref 7) and benzhydrylium
ions: Singer, T. Dissertation, Ludwig-Maximilians-Universitat,
̈
Munchen, 2008.
̈
(20) (a) Marcus, R. A. Annu. Rev. Phys. Chem. 1964, 15, 155−196.
(b) Marcus, R. A. J. Phys. Chem. 1968, 72, 891−899. (c) Marcus, R. A.
J. Am. Chem. Soc. 1969, 91, 7224−7225. (d) Albery, W. J.; Kreevoy, M.
M. Adv. Phys. Org. Chem. 1978, 16, 87−157. (e) Albery, W. J. Annu.
Rev. Phys. Chem. 1980, 31, 227−263. (f) Marcus, R. A. Angew. Chem.,
Int. Ed. Engl. 1993, 32, 1111−1121. (g) Marcus, R. A. Pure Appl. Chem.
1997, 69, 13−29.
(3) (a) Streidl, N.; Denegri, B.; Kronja, O.; Mayr, H. Acc. Chem. Res.
2010, 43, 1537−1549. (b) Denegri, B.; Streiter, A.; Juric,
R.; Kronja, O.; Mayr, H. Chem.Eur. J. 2006, 12, 1648−1656;
M.; Juric, S.; Denegri, B.;
́
S.; Ofial, A.
Chem.Eur. J. 2006, 12, 5415. (c) Matic,
́
́
Kronja, O. Int. J. Mol. Sci. 2012, 13, 2012−2024. (d) Splitting up of
the general nucleofugality scale described in refs 3a and 3b into
separate scales for different types of electrofuges has been suggested:
Bentley, T. W. Chem.Eur. J. 2006, 12, 6514−6520.
(21) (a) Rice, F. O.; Teller, E. J. Chem. Phys. 1938, 6, 489−496.
(b) Hine, J. J. Org. Chem. 1966, 31, 1236−1244. (c) Hine, J. J. Am.
Chem. Soc. 1966, 88, 5525−5528. (d) Hine, J. Adv. Phys. Org. Chem.
1978, 15, 1−61.
(22) (a) Kantner, S. S.; Humski, K.; Goering, H. L. J. Am. Chem. Soc.
1982, 104, 1693−1697. (b) Thibblin, A. J. Chem. Soc. Perkin. Trans.
1987, 2, 1629−1632.
(23) A rationalization of this behavior by quantum chemical methods
turned out to be unsuccessful as the predicted electrofugality
parameters Ef for amino- and alkoxy-substituted benzhydrylium ions
deviated significantly from the corresponding experimental values
(4) See, for instance: (a) Hughes, E. D. Trans. Faraday Soc. 1941, 37,
603−631. (b) Roberts, J. D.; Young, W. G.; Winstein, S. J. Am. Chem.
Soc. 1942, 64, 2157−2164. (c) Young, W. G.; Winstein, S.; Goering,
H. L. J. Am. Chem. Soc. 1951, 73, 1958−1963. (d) De La Mare, P. B.
D.; Vernon, C. A. J. Chem. Soc. 1954, 2504−2510. (e) Sneen, R. A. J.
Am. Chem. Soc. 1960, 82, 4261−4269. (f) Jia, Z. S.; Ottosson, H.;
Zeng, X.; Thibblin, A. J. Org. Chem. 2002, 67, 182−187. For a review,
see: (g) Raber, D. J.; Harris, J. M.; Schleyer, P. v. R. in Ions and Ion
Pairs in Organic Reactions; Szwarc, M., Ed.; Wiley: New York, 1974;
Vol. 2.
published almost simultaneously (ref 3a): Ormazab
́
al-Toledo, R.;
(5) (a) Goering, H. L.; Blanchard, J. P.; Silversmith, E. F. J. Am.
Chem. Soc. 1954, 76, 5409−5418. (b) Goering, H. L.; Silversmith, E. F.
J. Am. Chem. Soc. 1955, 77, 1129−1133. (c) Goering, H. L.;
Silversmith, E. F. J. Am. Chem. Soc. 1955, 77, 6249−6253. (d) Goering,
H. L.; Takahashi Doi, J. J. Am. Chem. Soc. 1960, 82, 5850−5854.
(e) Goering, H. L.; Nevitt, T. D.; Silversmith, E. F. J. Am. Chem. Soc.
1955, 77, 5026−5032. (f) Goering, H. L.; Koermer, G. S.; Linsay, E. C.
J. Am. Chem. Soc. 1971, 93, 1230−1234.
(6) Troshin, K.; Mayr, H. J. Am. Chem. Soc. 2013, 135, 252−265.
(7) Troshin, K.; Schindele, C.; Mayr, H. J. Org. Chem. 2011, 76,
9391−9408.
(8) Streidl, N.; Antipova, A.; Mayr, H. J. Org. Chem. 2009, 74, 7328−
7334.
Camporodon
́
ico, P. R.; Contreras, R. Org. Lett. 2011, 13, 822−824.
(24) Phan, T. B.; Nolte, C.; Kobayashi, S.; Ofial, A. R.; Mayr, H. J.
Am. Chem. Soc. 2009, 131, 11392−11401.
(25) Reichardt, C.; Welton, T. Solvents and Solvent Effects in Organic
Chemistry, 4th ed.; Wiley: Weinheim, 2011; p 181.
(26) Gelles, E.; Hughes, E. D.; Ingold, C. K. J. Chem. Soc. 1954,
2918−2929.
(27) Schaller, H. F.; Mayr, H. Angew. Chem., Int. Ed. 2008, 47, 3958−
3961.
(28) Richard, J. P; Jencks, W. P. J. Am. Chem. Soc. 1984, 106, 1383−
1396.
(29) Nolte, C.; Mayr, H. Eur. J. Org. Chem. 2010, 1435−1439.
(30) Troshin, K.; Mayer, P.; Mayr, H. Organometallics 2012, 31,
2416−2424.
(31) Hayashi, T.; Yamamoto, A.; Yoshihiko, I.; Nishioka, E.; Miura,
H.; Yanagi, K. J. Am. Chem. Soc. 1989, 111, 6301−6311.
(32) Hollingworth, S.; Hazari, A.; Hopkinson, M. N.; Tredwell, M.;
Benedetto, E.; Huiban, M.; Gee, A. D.; Brown, J. M.; Gouverneur, V.
Angew. Chem., Int. Ed. 2011, 50, 2613−2617.
(9) Horn, M.; Mayr, H. J. Phys. Org. Chem. 2012, 25, 979−988 and
references therein.
(10) Minegishi, S.; Loos, R.; Kobayashi, S.; Mayr, H. J. Am. Chem.
Soc. 2005, 127, 2641−2649.
(11) These values were reported for 50% aq acetonitrile (ref 2b);
however, as the differences between the N values of 50% (N = 5.05, sN
= 0.89) and 80% (N = 5.02, sN = 0.89) aq acetonitrile are marginal and
the corresponding sN values are equal, one would expect similar N and
sN values for 60% aq acetonitrile.
(12) (a) Grunwald, E.; Winstein, S. J. Am. Chem. Soc. 1948, 70, 846−
854. (b) Bentley, T. W.; Carter, G. E. J. Am. Chem. Soc. 1982, 104,
5741−5747. (c) Kevill, D. N.; D’Souza, M. J. J. Chem. Res. 2008, 61−
66. The Ybenzyl parameters were taken from: (d) Bentley, T. W.; Dau-
Schmidt, J.-P.; Llewellyn, G.; Mayr, H. J. Org. Chem. 1992, 57, 2387−
2392.
(13) Winstein−Grunwald correlations for benzhydryl chlorides are
based on the kinetic data from ref 3a and can be found in the
Supporting Information.
(14) One reviewer raised the question of whether it is justified to use
the YBenzyl values (derived from 4-methoxybenzyl chloride solvolyses,
ref 12d) for the correlations for 1-OPNB and 1-ODNB considering
the availability of YBnOPNB scale (derived from solvolyses of 2-aryl-2-
adamantyl 4-nitrobenzoates: Liu, K.-T.; Chang, C.-W.; Chen, H.-I;
Chin, C.-P.; Duann, Y.-F. J. Phys. Org. Chem. 2000, 13, 203−207; J.
Phys. Org. Chem. 2000, 13, 306. In order to directly compare the
effects of electrofuge and nucleofuge variation on m, it was crucial to
use the same Y scale for all correlations. As the YBenzyl scale provided
the best correlations among the scales containing parameters for
aqueous acetonitrile, we have chosen it as the basis for our analysis.
(15) Denegri, B.; Kronja, O. J. Phys. Org. Chem. 2009, 22, 495−503.
(16) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165−195.
(17) Streidl, N.; Mayr, H. Eur. J. Org. Chem. 2011, 2498−2506.
(18) What’sBest!7.0 (May 25, 2004) software by Lindo Systems, Inc.,
2004, was used for this purpose.
2660
dx.doi.org/10.1021/jo302766k | J. Org. Chem. 2013, 78, 2649−2660