The role of ion-molecule pairs in solvolysis reactions. Nucleophilic addition of water to a tertiary allylic carbocation

Article abstract of DOI:10.1021/jo016162a

The acid-catalyzed solvolysis of 2-methoxy-2-phenyl-3-butene (1-OMe) in 9.09 vol % acetonitrile in water provides 2-hydroxy-2-phenyl-3-butene (1-OH) as the predominant product under kinetic control along with the rearranged alcohol 1-hydroxy-3-phenyl-2-butene (2-OH) and a small amount of the rearranged ether 2-OMe. The more stable isomer 2-OH is the predominant product after long reaction time, K_eq = [2-OH]_eq/[1-OH]_eq = 16. The ether 2-OMe reacts to give 2-OH and a trace of 1-OH. Solvolysis of 1-OMe in ¹⁸O-labeled water/acetonitrile shows complete incorporation of ¹⁸O in the product 1-OH, confirming that the reaction involves cleavage of the carbon-oxygen bond to the allylic carbon. A completely solvent-equilibrated allylic carbocation is not formed since the solvolysis of the corresponding chloride 1-chloro-3-phenyl-2-butene (2-C1) yields a larger fraction of 1-OH. This may be attributed to a shielding effect from the chloride leaving group. Quantum chemical calculations of the geometry and charge distribution show that the cation should rather be described as a vinyl-substituted benzyl cation than as an allyl cation, which is in accord with its higher reactivity at the tertiary carbon.

Full text of DOI:10.1021/jo016162a

182
J . Org. Chem. 2002, 67, 182-187
Th e Role of Ion -Molecu le P a ir s in Solvolysis Rea ction s.
Nu cleop h ilic Ad d ition of Wa ter to a Ter tia r y Allylic Ca r boca tion
Zhi Sheng J ia, Henrik Ottosson, Xiaofeng Zeng, and Alf Thibblin*
Institute of Chemistry, University of Uppsala, P.O. Box 531, SE-751 21 Uppsala, Sweden
Alf.Thibblin@kemi.uu.se
Received October 1, 2001
The acid-catalyzed solvolysis of 2-methoxy-2-phenyl-3-butene (1-OMe) in 9.09 vol % acetonitrile
in water provides 2-hydroxy-2-phenyl-3-butene (1-OH) as the predominant product under kinetic
control along with the rearranged alcohol 1-hydroxy-3-phenyl-2-butene (2-OH) and a small amount
of the rearranged ether 2-OMe. The more stable isomer 2-OH is the predominant product after
long reaction time, Keq ) [2-OH]eq/[1-OH]eq ) 16. The ether 2-OMe reacts to give 2-OH and a
trace of 1-OH. Solvolysis of 1-OMe in 18O-labeled water/acetonitrile shows complete incorporation
of 18O in the product 1-OH, confirming that the reaction involves cleavage of the carbon-oxygen
bond to the allylic carbon. A completely solvent-equilibrated allylic carbocation is not formed since
the solvolysis of the corresponding chloride 1-chloro-3-phenyl-2-butene (2-Cl) yields a larger fraction
of 1-OH. This may be attributed to a shielding effect from the chloride leaving group. Quantum
chemical calculations of the geometry and charge distribution show that the cation should rather
be described as a vinyl-substituted benzyl cation than as an allyl cation, which is in accord with its
higher reactivity at the tertiary carbon.
Sch em e 1
In tr od u ction
Stepwise solvolytic reactions of substrates with neutral
leaving groups produce carbocation molecule pairs (ion-
dipole pairs) by cleavage of the bond between the carbon
and the leaving group (Scheme 1).1-11 The diffusional
separation (k-d) to give a solvent-equilibrated carbocation
is expected to be fast (k-d ∼ 2 × 1010 s-1) in aqueous
solution which means that a bimolecular reaction with
a dilute reactant should not be able to compete with the
separation. Accordingly, a substantial amount of reaction
of the ion-molecule pair R+X should require either an
intramolecular reaction, e.g., a rearrangement reaction
to give R′X+ (Scheme 1) or an elimination reaction in
which the leaving group acts as the hydron-abstracting
base, or a bimolecular reaction with a species which is
close to R+X when it is born.
to an allylic carbocation and to its complexes with
methanol and other leaving groups.
The study is closely related to the question about the
lifetime of simple tertiary carbocations in water and
highly aqueous solvents. Recently, Toteva and Richard
discussed the lifetime of the tert-butyl carbocation.12
Reactivity data obtained from studies of less reactive
carbocations were extrapolated by employing the ex-
tended Hammett relationship to give an estimated rate
constant for reaction of the tert-butyl carbocation with
solvent (50 vol % 1,1,1-trifluoroethanol in water) of ks ∼
1.6 × 1012 s-1. This value suggests that the carbocation
is not diffusionally equilibrated but reacts at the ion-pair
stage within the pool of solvent molecules that are
present when the bond to the leaving group is broken.
The results of the present study suggest that also a
more stable benzallylic cation reacts to some extent with
a solvent water molecule of the solvation shell that is
present at the time of ionization, before separation takes
place.
We have previously reported studies on reactions which
give products directly through the ion-molecule pair.2,5-9
These studies suggested that the nature of the carbo-
cation and the basicity of the leaving group are of crucial
importance for the product composition. The present
report concerns the mechanisms of nucleophilic addition
(1) Sneen, R. A.; Felt, G. R.; Dickason, W. C. J . Am. Chem. Soc.
1973, 95, 638.
(2) Thibblin, A. J . Chem. Soc., Perkin Trans. 2 1987, 1629.
(3) Katritzky, A. R.; Brycki, B. E. J . Phys. Org. Chem. 1988, 1, 1
and references therein.
(4) (a) Szele, I. and Zollinger, H. J . Am. Chem. Soc. 1978, 100, 2811.
(b) Hashida, Y.; Landells, R. G. M.; Lewis, G. E.; Szele, I.; Zollinger,
H. J . Am. Chem. Soc. 1978, 100, 2816.
(5) Thibblin, A. J . Org. Chem. 1993, 58, 7427.
Resu lts
(6) Thibblin, A. J . Phys. Org. Chem. 1993, 6, 287.
(7) Sidhu, H.; Thibblin, A. J . Phys. Org. Chem. 1994, 578, and
references therein.
(8) Thibblin, A.; Sidhu, H. J . Phys. Org. Chem. 1993, 6, 374.
(9) Thibblin, A.; Saeki, Y. J . Org. Chem. 1997, 62, 1079.
(10) Bleasdale, C.; Golding, B. T.; Lee, W. H. L.; Maskill, H.;
Riseborough, J .; Smits, E. J . Chem. Soc., Chem. Commun. 1994, 93.
(11) Filippi, A.; Gasparrini, F.; Speranza, M. J . Am. Chem. Soc.
2001, 123, 2251.
R ea ct ion s of E t h er s a n d Alcoh ols. The acid-
catalyzed solvolysis of 2-methoxy-2-phenyl-3-butene (1-
OMe) in 50 vol % acetonitrile in water at 25 °C yields
(12) Toteva, M. M.; Richard, J . P. J . Am. Chem. Soc. 1996, 118,
11434.
10.1021/jo016162a CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/13/2001

Role of Ion-Molecule Pairs in Solvolysis
J . Org. Chem., Vol. 67, No. 1, 2002 183
Ta ble 1. Ra te Con sta n ts for th e Acid -Ca ta lyzed
Rea ction s of 1-OMe, 2-OMe, a n d 1-OH in Aqu eou s
Aceton itr ile a t 25 °C
a
vol % 106kobs
substrate MeCN
,
kH,
106kAB, 106kAC, 106kAD
,
s-1
M-1s-1
s-1
s-1
s-1
1-OMe
9.09
50
91.09
50
103.0b
54.8c
72.1d
1.4e
88 × 10-3
82.0
39.0
52.2
21.0
15.5
18.8
0.16
0.3
1.1
4.3 × 10-3
2-OMe
1-OH
4.6 × 10-6
357 × 10-6
50
4.5c,f
(568)g (931 × 10-6
)
a
b
Estimated maximum errors: ( 5%. pH 2.93 and [NaClO4]
) 0.10 M before mixing with acetonitrile. c pH 1.90 and [NaClO4]
d
) 0.10 M before mixing with MeCN. [HClO4] ) 51 mM before
mixing with MeCN. e [HClO4] ) 0.304 M and [NaClO4] ) 0.40 M
before mixing with MeCN. f Equilibrium constant [2-OH ]eq
/
g
[1-OH]eq ) 16 ( 2. [HClO4] ) 0.61 M and [NaClO4] ) 0.40 M
before mixing with MeCN; rate constant measured by UV spec-
trophotometry.
Ta ble 2. Ra te Con sta n ts for th e Solvolysis Rea ction s of
1-P NB, 2-P NB, a n d 2-Cl in 50 Vol % Aceton itr ile in Wa ter
a t 25 °C
F igu r e 1. The time dependence of the product composition
of the acid-catalyzed solvolysis of 1-OMe (Scheme 2) in 50 vol
% acetonitrile in water at 25 °C: pH 1.90 and [NaClO4] ) 0.10
M. The points are the experimental data, and the curves are
computer simulated using eqs 5-7. The formation of the small
amount of 2-OMe is not included.
a
106kobs
s-1
,
106kr, [1-OH]/ [1-SCN]/[2-SCN] or
substrate addition
1-PNB none
s-1
[2-OH]
[1-OMe]/[2-OMe]
179
33
12
- - -
2.8 ( 0.1
2.8 ( 0.2
2.5 ( 0.1
- - -
0.11 ( 0.05
6.0 ( 0.4
- - -
- - -
1.8 ( 0.1
1-PNB SCN- b 370
1-PNB MeOHc - - -
2-PNB none
Sch em e 2
∼0.03d
- - - 0.9d ( 0.1
2-Cl
2-Cl
none
fast
- - -
- - -
4.1 ( 0.2
2.1 ( 0.1
MeOHc fast
a
b
Estimated maximum errors: (7%. [NaSCN] ) 0.25 M; pH
) 6.8. Reaction with 0.25 M NaClO4 gave kobs ) 357 × 10-6 s-1
,
pH ) 6.8. c [MeOH] ) 0.75 M. 5% reaction after 20 days. e [1-
d
OH]/[2-OH] ) 1.0 at 60 °C.
It was not possible to carry out trapping experiments
with azide ion owing to the low pH required for the
reactions. Addition of chloride or bromide ion (0.20 M)
did not give rise to any detectable substitution product.
Solvolysis experiments in oxygen-labeled water were
performed in order to establish which of the two carbon-
oxygen bonds in the substrate is broken during the
reaction. Thus, the acid-catalyzed solvolysis of 1-OMe in
a mixture of 18O-labeled water with 50 vol % acetonitrile
was followed by a GC/MS sampling procedure. The
analysis showed complete incorporation of the labeled
oxygen into the product 1-OH, as well as in the product
2-OH, confirming that it is the tertiary carbon-oxygen
bond that is broken (see Experimental Section for de-
tails).
the alcohols 2-hydroxy-2-phenyl-3-butene (1-OH) and
1-hydroxy-3-phenyl-2-butene (2-OH) and a trace of 1-meth-
oxy-3-phenyl-2-butene (2-OMe) (Scheme 2). The change
in product composition as a function of time is shown in
Figure 1. The alcohol product 1-OH is less stable than
its allylic isomer 2-OH and after a long reaction time
most of 1-OH has been converted to 2-OH. The equilib-
rium constant was measured in a separate experiment
by studying the acid-catalyzed formation of 2-OH from
1-OH, Keq ) [2-OH]eq/[1-OH]eq ) 16 ( 2. The solvolysis
of 1-OMe in a less aqueous solvent mixture (91.09 vol %
acetonitrile in water) gives somewhat more of the rear-
rangement product 2-OMe, but has only a small effect
on the alcohol product ratio. A more aqueous solvent (9.09
vol % acetonitrile in water) increases the alcohol product
ratio [1-OH]/[2-OH].
The kinetics of the reaction of 1-OMe as well as of its
allylic isomer 2-OMe and the corresponding alcohols
1-OH and 2-OH were studied by a high-performance
liquid chromatography sampling procedure. The reaction
of 1-OH was also followed by UV spectrophotometry. The
rate constants and reaction conditions are recorded in
Table 1. The observed rate constant for the reaction of
1-OMe was divided into the separate phenomenological
rate constants by employing a computer simulation tech-
nique (Figure 1; see Experimental Section for details).
Rea ction s of Ester s a n d Ch lor id e. The solvolysis
of 2-(4-nitrobenzoyl)-2-phenyl-3-butene (1-P NB) in 50 vol
% acetonitrile in water at 25 °C provides the alcohols
1-OH and 2-OH and some of the rearranged substrate
1-(4-nitrobenzoyl)-3-phenyl-2-butene (2-P NB). The rear-
rangement of the ester 1-P NB is about 4 times slower
than the formation of the alcohols (Table 2). This rear-
ranged ester reacts much more slowly, and no trace of
1-P NB was observed along with the alcohols after 20
days of reaction (∼5% reaction). The very fast solvolysis
of 1-chloro-3-phenyl-2-butene (2-Cl) exclusively provides
the alcohols. The results of these reactions are shown in
Table 2.
Azide-trapping experiments with the ester 1-P NB did
not work owing to a complex reaction product mixture.
The solvolysis of 1-P NB in the presence of thiocyanate
ion (0.25 M) does not increase reaction rate compared
with perchlorate ion but gives two major products, 1-SCN

184 J . Org. Chem., Vol. 67, No. 1, 2002
J ia et al.
Sch em e 4
F igu r e 2. CC Bond distances (Å) and atomic charges (elec-
tron) for 1a calculated at B3LYP/6-31G(d) (normal print) and
PCM B3LYP/6-31G(d) (italics) levels.
conformation to the s-trans isomer 1a is 33.8 kcal/mol at
B3LYP/6-31G(d) level. However, when going from cumyl
cation to 1a (reaction 2, Scheme 3), only 11.3 kcal/mol
are released.
Sch em e 3
Discu ssion
The ether 1-OMe is relatively unreactive in acidic
solution, it reacts about 103 times slower than the
structurally related 1-methoxy-1-methyl-1,4-dihydronaph-
thalene (3 in Scheme 4).13 This is presumably the result
of extensive charge delocalization accompanying the
ionization of the latter compound (vide infra). The car-
bocation is stabilized by enhanced resonance in the cyclic
carbocation.14 A comparison with the structurally related
cumyl methyl ether (4 in Scheme 4) shows that 1-OMe
is 102 times more reactive.15
The reactivity of the carbocation has been measured
by trapping the intermediate(s) formed from the ester
1-P NB with thiocyanate ion (see Results). A rate con-
stant for reaction of the carbocation of kw ) 7 × 108 s-1
was obtained. This could be compared with the reactivity
of other structurally related carbocations. There are a
few relevant reports in the literature. For example, the
allylic carbocation 5+ (Scheme 4)2,16 reacts with water in
75% aqueous acetonitrile with a rate constant of kw ) 4
× 109 s-1 which is similar to that of the cumyl cation 4+.15
The cation 6+ in 20% aqueous acetone shows a lower
and 2-SCN, along with the alcohols. A trace of 2-NCS is
also formed. These substitution products are stable under
the reaction conditions. The nucleophilic selectivity be-
tween thiocyanate ion and water was derived from the
measured product composition as kSCN/kHOH ) 190 (ratio
of second-order rate constants, eq 1). Assuming a diffu-
sion-controlled reaction of thiocyanate ion (kd ) 5 × 109
M-1 s-1) with the carbocation yields kw ) 7 × 108 s-1 (eq
2).
kSCN /kHOH ) ([thiocyanate products]/[alcohol
products])/([SCN-]/[HOH]) (1)
kw ) ([alcohol products]/[thiocyanate products]) ×
kd × [SCN-] (2)
reactivity, kw ) 4 × 108 s-1 17
.
This should correspond to
a value of about 1 × 109 s-1 in 50 vol % acetonitrile. The
tertiary benzallylic carbocation is expected to be more
stable than these ions, but it should be less stable than
the cation 7+ which reacts with water with a rate
Ca lcu la tion s. The s-trans conformer (1a in Figure 2)
of the cation is 1.0 kcal/mol more stable than the s-cis
isomer at the B3LYP/6-31G(d) level of calculation. The
optimal CC bond lengths and atomic charges at the C
atoms, as calculated with natural population analysis,
are given for 1a in Figure 2. The B3LYP geometry reveals
a bond length difference between the C1-C2 and C2-
C3 bonds by 0.084 Å, contrary to the allyl cation where
these bond lengths are identical (1.385 Å). In addition,
C3 is more positively charged and C1 is more negatively
charged than the corresponding C atoms in the allyl
cation (0.003 e). Geometry optimization of the cation,
using the polarizable continuum model (PCM) of Tomasi
and co-workers with the dielectric constant set as for
water, did neither change geometry to any major extent
nor charge distribution (Figure 2). The relative stabilities
of the two conformers were changed slightly, but 1a is
more stable by 1.4 kcal/mol. Support for the rather
extensive stabilizing effect of the allyl cation by the
phenyl group in 1 is obtained through the isodesmic
reaction 1 shown in Scheme 3. The additional stabiliza-
tion when going from the 1-methylallyl cation in the s-cis
constant of kw ∼ 1 × 107 s-1 18
.
All these estimated
reactivities are based upon the assumption that the
cations are diffusionally equilibrated. However, this is
presumably not quite correct (vide infra).5
The pKR for the pseudo acid-base equilibrium for
cation hydration (eqs 1 and 2) is a measure of the
thermodynamic stability of a carbocation R+.
kw
\
R+ + H2O y
z ROH + H+
(3)
(4)
kH
pKR ) log(kH/kw)
(13) J ia, Z. S.; Thibblin, A. J . Chem. Soc., Perkin Trans. 2 2001 247.
(14) Boyd, D. R.; McMordie, R. A. S.; Sharma, N. D.; More O’Ferrall,
R. A.; Kelly, S. C. J . Am. Chem. Soc. 1990, 112, 7822.
(15) Thibblin, A. J . Phys. Org. Chem. 1989, 2, 15.
(16) Thibblin, A. J . Chem. Soc., Perkin Trans. 2 1986, 313 and 321.
(17) Ta-Shma, R.; Rappoport, Z, J . Am. Chem. Soc. 1983, 105, 6082.
(18) Pirinccioglu, N.; Thibblin, A. J . Am. Chem. Soc. 1998, 120, 6512.

Role of Ion-Molecule Pairs in Solvolysis
J . Org. Chem., Vol. 67, No. 1, 2002 185
Sch em e 5
to that for the dielectric relaxation of water which occurs
with a rate constant of ∼ 1011 s-1 19,20
.
The rate-limiting
step for addition of a solvent water molecule to a very
unstable carbocation is expected to be this rotation of the
water molecule into a reactive orientation. However, the
absence of a significant bonding barrier for the addition
of a water molecule to the cation of the benzallylic ion-
molecule pair 1+O(H)Me may not seem reasonable. Let
us look at the other alternative explanation to the
predominant formation of 1-OH.
The second alternative that the alcohols are formed
from a solvent-equilibrated allylic carbocation which has
a smaller barrier for addition of the water to the tertiary
than to the primary carbon is in accord with the large
fraction of 1-OH formed from 2-Cl. This large fraction
(80%, Table 2) is not expected to be caused by a SN2′
reaction owing to the steric hindrance at the benzylic
carbon, but is reasonably the result of a stepwise car-
bocation mechanism. The larger fraction of 1-OH formed
from 2-Cl than from 1-OMe is consistent with a shielding
effect of the leaving group in the allylic ion pair.5 Thus,
the carbocation is not a completely free solvent-equili-
brated species but reacts to a significant extent with a
solvent water molecule of the solvation shell that is
present at the time of ionization. The very small amount
of 1-OH formed from 2-OMe may seem to be inconsistent
with a common carbocation intermediate. However,
2-OMe is much less reactive, and any 1-OH formed may
quickly undergo further reaction to give the other,
thermodynamically more stable alcohol product 2-OH.
The product ratio of [1-OH]/[2-OH] ) 2.5 corresponds to
a difference in kinetic barrier of 0.5 kcal/mol. A similar
product ratio was measured with 1-P NB (Table 2).
However, the ratio measured with thiocyanate ion was
much smaller, [1-SCN]/[2-SCN] ) 0.1 (Table 2). This
value probably reflects the thermodynamic stabilities of
these products.
The estimated reactivity of the solvent-equilibrated
benzallylic carbocation with water of kw ) 7 × 108 s-1
(vide supra) and kH ) 357 × 10-6 M-1s-1 (Table 1) yields
pKR ) -12.3. This could be compared with the value for
the cyclic benzallylic carbocation 7+ (Scheme 4), which
has been estimated as pKR ∼ -7.8.18
The ionization rate constant k1 should be relatively
insensitive to the solvent polarity in sharp contrast to
the cleavage of a bond between carbon and a negatively
charged leaving group. Also, the rearrangement of the
two localized ion-molecule pairs (kr, Scheme 5) is not
expected to show a high sensitivity to the solvent polarity.
This allylic rearrangement of the ion-molecule pairs
involves breaking of the dipole interaction between the
carbocation and the leaving group and formation of a new
one between carbon 3 and the leaving group combined
with adjustments of electron density and geometry of the
allylic system. These processes are expected to be rela-
tively fast with a maximum rate constant of about 1 ×
1011 s-1 19,20
A rate constant of this magnitude has been
.
proposed for the reorganization of thionobenzoate pairs
in 50% aqueous trifluoroethanol.21
The primary allylic alcohol 2-OH is more stable than
its allylic tertiary isomer 1-OH, Keq ) [2-OH]eq/[1-OH]eq
) 16. However, 1-OH is the predominant product under
kinetic control. It is formed 2.5 times faster than 2-OH
(Table 1). The ester 1-P NB shows a similar rate-ratio
but exhibits more allylic rearrangement than 1-OMe
(Table 2). Two plausible alternative explanations for the
predominant formation of 1-OH are the following. (i)
Direct reaction of the localized ion-molecule pair
1+O(H)Me with solvent water (kw′, Scheme 5). (ii) Reac-
tion through the solvent-equilibrated allylic carbocation;
a smaller kinetic barrier for addition of water to the
tertiary carbon than to the primary carbon.
The first alternative is supported by the fact that there
is a significant increase in the product alcohol ratio
[1-OH]/[2-OH] with increasing fraction of water in the
solvent (from 2.5 to 3.9 for an increase in water content
from 50 to 91%); an approximately constant ratio is
expected for reaction through the solvent-equilibrated
carbocation. A substantial amount of direct reaction of
1+O(H)Me with water before dissociation (k-d) requires
that kw′ is comparable to k-d, which should have a value
of about 2 × 1010 s-1, and also comparable to kr, which is
expected to have a maximum value of about 1 × 1011 s-1
(Scheme 5, vide supra).
The rate constant for the collapse of the ion-molecule
pair to give 2-OMe should have a rate constant of about
1 × 108 s-1 based upon k-d ) 2 × 1010 s-1 and the
measured rate constants in 50% aqueous acetonitrile
(Table 1).
Another alternative explanation for the large amount
of 1-OH formed might have been an SN2 reaction in
which water attacks the methyl group of the hydronated
substrate. However, this possibility is excluded since
solvolysis experiment with 18O-labeled water showed
complete incorporation of the labeled oxygen in 1-OH.
Do quantum chemical calculations support the conclu-
sion that the addition of water to the tertiary carbon has
the lowest barrier? Our results indicate a much higher
charge density at the tertiary carbon. The results reveal
that the cation should rather be described as a vinyl-
substituted R-methylbenzyl carbocation than as an allyl
cation (Figure 2), consistent with the higher reactivity
of the tertiary carbon. The charge distribution and
geometry are very similar in gas phase and in water. The
substitution of a methyl with a vinyl group in the cumyl
carbocation 4+ (Scheme 4) corresponds to a rate increase
of the methyl ether of 2 orders of magnitude and to a
5-fold decrease in kw.15
The rate constant for rotation of a water molecule into
a reactive position in a solvation shell should be similar
(19) See for example: Chiang, Y.; Kresge, A. J . J . Am. Chem. Soc.
1985, 107, 6363. Paradisi, C.; Bunnett, J . F. J . Am. Chem. Soc. 1985,
107, 8223.
(20) Giese, K.; Kaatze, U.; Pottel, R. J . Phys. Chem. 1970, 74, 3718.
Kaatze, U. J . Chem. Eng. Data 1989, 34, 371. Kaatze, U.; Pottel, R.;
Schumacher, A. J . Phys. Chem. 1992, 96, 6017.
The result of the present work is closely related to an
earlier study on acid-catalyzed solvolysis of allylic ethers
in highly aqueous solution (Scheme 6) in which we
reported that allylic rearrangement of the methyl ethers
accompanies formation of the alcohols.2 The isomerization
(21) Richard, J . P.; Tsuji, Y. J . Am. Chem. Soc. 2000, 122, 3963.

186 J . Org. Chem., Vol. 67, No. 1, 2002
J ia et al.
UV gradient quality, respectively. All other chemicals used for
the kinetic experiments were of reagent grade and used
without further purification. The oxygen labeled water was
delivered from Isotec Inc. (>95 at. % 18O).
Sch em e 6
2-Hyd r oxy-2-p h en yl-3-bu ten e (1-OH) was prepared by
addition of vinylmagnesium bromide to acetophenone in tet-
rahydrofuran.22 Distillation of the product gave pure material.
1-Hyd r oxy-3-p h en yl-2-bu ten e (2-OH ) was prepared by
Reformatsky reaction of acetophenone with ethyl bromoacetate
followed by reduction with lithium aluminum hydride.23,24
2-Meth oxy-2-p h en yl-3-bu ten e (1-OMe) was prepared
from 1-OH by methylation with methyl iodide in tetrahydro-
furan and distilled. Purity was confirmed by NMR25 and
HPLC.
1-Meth oxy-3-p h en yl-2-bu ten e (2-OMe).25 A methanol
solution (30 mL) of the alcohol 1-OH (0.37 g) and perchloric
acid (1.5 g, 75%) was kept at room temperature for 3 h.
Extraction with diethyl ether, drying, and flash chromatog-
raphy with pentane-diethyl ether on a silica column gave a
product free from alcohols and 1-OMe.
2-(4-Nitr oben zoyl)-2-p h en yl-3-bu ten e (1-P NB). Methyl-
lithium (0.6 mL, 1.6 M in diethyl ether) was added to a solution
of 1-OH (0.122 g, 0.82 mmol) in dry tetrahydrofuran under
nitrogen at 0 °C; the solution was stirred for further 30 min.
A solution of 4-nitrobenzoyl chloride (0.178 g, 0.96 mmol) in
tetrahydrofuran (5 mL) was then added. After 5 h of stirring,
the reaction mixture was poured into water, extracted with
ether, and dried over magnesium sulfate. Removal of solvent
and recrystallization twice from pentane gave yellow crystals
(yield 70 mg, 25%). The purity was confirmed by NMR and
HPLC; 1H NMR (CDCl3): 2.056 (s, 3 H), 5.36 (d, 1 H, J ) 10.8
Hz), 5.403 (d, 1 H, J ) 17.6), 6.40 (q, 1 H, J ) 17.6 Hz, 10.8
Hz).
1-(4-Nitr oben zoyl)-3-p h en yl-2-bu ten e (2-P NB). A solu-
tion of butyllithium (2.2 mL, 2.5 M in hexane) was added
dropwise to a solution of 2-OH (0.81 g, 5.5 mmol) in 10 mL of
dry diethyl ether at 0 °C. After 10 min of further stirring, a
solution of 4-nitrobenzoyl chloride (1.02 g, 5.5 mmol) in 10 mL
of dry diethyl ether was added. The reaction mixture was
worked up after 16 h. Purification by flash chromatography
on a silica column gave pure material, yield 0.58 g, (36%);
1H NMR (CDCl3) 2.20 (s, 3 H), 5.09 (d, 2 H, 7.2 Hz), 6.03
(ddd, 1 H, J ) 1.6 Hz, 7.2 Hz), 7.27-7.45 (m, 5 H), 8.22-8.31
(m, 4 H).
1-Ch lor o-3-p h en yl-2-bu ten e (2-Cl) was prepared from
2-OH and thionyl chloride followed by distillation.23
Kin etics a n d P r od u ct Stu d ies. The reaction solutions
were prepared by mixing acetonitrile with aqueous perchloric
acid solutions at room temperature, ca. 22 °C. The reactions
were initiated by addition of a few microliters of the substrate
dissolved in acetonitrile to a 2-mL HPLC vial containing 1.2
mL of the prethermostated solvent mixture to give a final
substrate concentration of about 1 mM. The reaction flask was
sealed with a gastight PTFE septum and placed in a thermo-
stated aluminum block in the HPLC apparatus. At appropriate
intervals, samples were automatically injected onto the column
and analyzed. The rate constants for the disappearance of the
substrates were calculated from plots of substrate peak area
versus time by means of a nonlinear regression computer
program. The kinetics of 1-OH was also studied by UV
spectrophotometry by following the increase in absorbance at
246 nm. Very good pseudo-first-order behavior was observed
for all the reactions studied.
of 5-OMe is much more competitive with alcohol forma-
tion than the isomerization of 1-OMe. It was suggested
that two discrete ion-molecule pairs are involved, but,
owing to fast equilibration of the allylic alcohols, it was
not possible to detect any possible fast direct reaction of
the localized ion-molecule pair with solvent water. A
single common ion-molecule pair mechanism was con-
cluded to be inconsistent with the data since the collapse
ratio for formation of the two ethers derived from the
kinetic data was found to be smaller than the one
measured for the corresponding tertiary chloride 8-Cl in
aqueous acetonitrile with added methanol (Scheme 6).
However, alternatively, the results may be rationalized
by a shielding effect of the chloride ion as was discussed
above.
Recently, we reported studies on acid-catalyzed sol-
volysis of 1-methoxy-1,4-dihydronaphthalene and 1-meth-
oxy-1-methyl-1,4-dihydronaphthalene (3, Scheme 4) in
highly aqueous solution.13,18 No formation of the corre-
sponding 1-hydroxy derivatives was found in these stud-
ies. These observations were rationalized by postulating
that the carbon-oxygen bond cleavage is accompanied
by extensive delocalization of charge. This is in accord
with the approximately 103 times greater reactivity of 3
compared with 1-OMe and with the large driving force
of the reactions giving naphthalene and methylnaphtha-
lene; the reaction heats for the corresponding alcohols
were measured as -23.7 and -21.7 kcal/mol, respec-
tively.13,18 The extensive charge delocalization which
accompanies the ionization results in a large barrier for
the back reaction (the formation of the reactant ether)
as well as for the collapse to give the corresponding
alcohol.
Exp er im en ta l Section
Gen er a l P r oced u r es. NMR spectra were recorded at 25
°C with
a Varian Unity 300 or 400 MHz spectrometer.
Chemical shifts are indirectly referenced to TMS via the
solvent signal (chloroform-d 7.26 and 77.0 ppm). The high-
performance liquid chromatography analyses were carried out
with a Hewlett-Packard 1090 liquid chromatograph equipped
with a diode-array detector on an Inertsil 3 ODS-3 (3 × 100
mm) reversed-phase column. The mobile phase was a solution
of acetonitrile in water. The reactions were run at constant
temperature controlled by a HETO 01 PT 623 thermostat bath.
The kinetics using UV spectrophotometry were carried out
with a Kontron Uvicon 930 spectrophotometer equipped with
an automatic cell changer kept at constant temperature with
water from the thermostat bath. The pH was measured using
a Radiometer PHM82 pH meter with an Ingold micro glass
electrode. The pH values given are those measured before mix-
ing with the organic solvent. The mass spectrometric analyses
were carried out on a GC/MS instrument of quadrupole ion-
trap type (Finnigan MAT GCQ equipped with autosampler AS-
2000 and an OV-5, 30 m × 0.25 mm GC column).
The separate rate constants for the reaction of 1-OMe
(Scheme 2 and Figure 1) were derived by computer simulation
(least-squares fit) based upon the product composition data,
(22) Marvel, C. S.; Woodford, R. G. J . Org. Chem. 1958, 23, 1658.
Hou, Z.; Fujiwara, Y.; J intoku, T.; Mine, N,: Yokoo, K.; Taniguchi, H.
J . Org. Chem. 1987, 52, 3524.
(23) J enkins, S. J . Am. Pharm. Assoc. 1948, 37, 118.
(24) Murphy, J . A.; Pattersson, C. W. J . Chem. Soc., Perkin Trans.
1. 1993, 405.
Ma ter ia ls. Merck silica gel 60 (240-400 mesh) was used
for flash chromatography. Diethyl ether and tetrahydrofuran
were distilled under nitrogen from sodium and benzophenone.
Methanol and acetonitrile were of HPLC quality and HPLC
(25) Uccello-Barretta, G.; Bernardini, R.; Lazzaroni, R.; Salvadori,
P. J . Organomet. Chem. 2000, 598, 174.

Role of Ion-Molecule Pairs in Solvolysis
J . Org. Chem., Vol. 67, No. 1, 2002 187
which were obtained from the HPLC peak areas and the
relative response factors, and the kinetic and equilibrium data
of the alcohols. The small amount of the rearranged ether
2-OMe was not included in the simulation.
When starting from pure 1-OMe, the concentrations of
1-OMe, 1-OH, and 2-OH are described by the following
equations:26
parameter hybrid functional (B3)27 for exchange and the Lee-
Yang-Parr (LYP) gradient corrected correlation functional.28
The 6-31G(d) valence double-ú basis set of Pople and Hariha-
ran was used.29 Atomic charges were calculated using natural
population analysis,30 and nonspecific solvent interaction was
considered by the self-consistent reaction field model of Tomasi
and co-workers.31 All calculations were done with the Gauss-
ian98 program package.32
mol % 1-OMe ) 100e-m t
(5)
1
Ack n ow led gm en t. We thank the Swedish Natural
Science Research Council for supporting this work. H.O.
is grateful to the Wenner-Gren Foundations for financial
support and to the National Supercomputer Center in
Linko¨ping, Sweden, for generous allotment of computer
time.
mol % 1-OH ) 100(kCB/m2)(1 - e-m t) + 100{(kAB - kCB)/
2
(m2 - m1)} (e-m t - e-m t) (6)
1
2
mol % 2-OH ) 100 - mol % 1-OMe - mol % 1-OH (7)
where m1 ) kAB + kACm2 ) kBC + kCB
Solvolysis in 18O-La beled Wa ter . The reaction medium
was 50% aqueous acetonitrile prepared by mixing 150 µL of
labeled water with the same volume of an acetonitrile solution
of the substrate 1-OMe and 0.1 µL of 98% sulfuric acid. The
reactions were run at room temperature (ca. 22 °C) in 2 mL
GC autosampler vials sealed with PTFE septa. Aliquots (2 µL)
were injected at appropriate intervals directly onto the GC/
MS column. All components were well separated. The products
1-OH and 2-OH were detected and analyzed (30 eV) under
“positive ion/full mass” mode as well as with electron impact
and chemical ionization mode, respectively. For comparison,
the same procedure was applied to the reaction of 1-OMe in
unlabeled water/acetonitrile. Full incorporation of the labeled
oxygen in the products 1-OH and 2-OH was found.
The relative abundance of ions (in % of the reference base
peak m/z 105) in analysis of 1-OH were as follows: m/z 148
(M+) 0.7%, m/z 150 (M+) 17.3%, m/z 147 ([M - H]+) 1.1%, m/z
149 ([M - H]+) 13.0%, m/z 133 ([M - CH3]+) 3.6%, m/z 135
([M - CH3]+) 60.6%. For reaction in unlabeled water, the
following values (recorded in the same order) were measured:
15.4%, 0.2%, 12.2%, 1.3%, 56.9%, 0.7%.
Su p p or tin g In for m a tion Ava ila ble: Tables of absolute
energies and atomic coordinates from B3LYP/6-31G(d) geom-
etry optimization. This material is available free of charge via
the Internet at http://pubs.acs.org.
J O016162A
(27) Becke, A. D. J . Chem. Phys. 1993, 98, 5648.
(28) Lee, C.; Yang, W.; Parr, R. G.Phys. Rev. B 1988, 37, 785.
(29) Hariharan, P. S.; Pople, J . A. Theor. Chim. Acta 1973, 28, 213.
(30) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,
899.
(31) Barone, V.; Cossi, M.; Tomasi, J . J . Comput. Chem. 1998, 19,
404.
(32) Gaussian 98, Revision A.3: Frisch, M. J .; Trucks, G. W.;
Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J . R.;
Zakrzewski, V. G.; Montgomery, J . A. J r.; Stratmann, R. E.; Burant,
J . C.; Dapprich, S.; Millam, J . M.; Daniels, A. D.; Kudin, K. N.; Strain,
M. C.; Farkas, O.; Tomasi, J .; Barone, V.; Cossi, R.; Cammi, R.;
Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J .;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J . B.; Cioslowski, J .; Ortiz,
J . V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi,
I.; Gomberts, R.; Martin, R. L.; Fox, D. J .; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P.
M. W.; J ohnson, B.; Chen, W.; Wong, M. W.; Andres, J . L.; Head-
Gordon, M.; Replogle, E. S.; Pople, J . A. Gaussian, Inc., Pittsburgh,
PA, 1998.
Com p u ta tion a l Meth od s. Geometry optimization was
carried out using the B3LYP combination of the Becke’s three-
(26) Alberty, R. A.; Miller, W. G. J . Chem. Phys. 1957, 26, 1231.