Substituent effects on the formation and nucleophile selectivity of ring-substituted phenonium ions in aqueous solution

Article abstract of DOI:10.1002/poc.3145

The reaction of 2-(4-methyphenyl)ethyl tosylate (Me-1-OTs) in 50/50 (v/v) trifluoroethanol/water at 25 °C is first order in the concentration of azide anion nucleophile. A carbon-13 NMR analysis of the products of the reactions of Me-1-[α-¹³C]OTs in 50/50 (v/v) trifluoroethanol/water at 25 °C shows the formation of Me-1-[β-¹³C]OH, Me-1-[β- ¹³C]OCH₂CF₃ and Me-1-[β-13C]N3 from the trapping of a symmetrical 4-methylphenonium ion reaction intermediate Me-2 ⁺. The formation of Me-1-[α-13C]N3 by concerted bimolecular displacement of azide ion at Me-1-[α-¹³C]OTs (k_N = 3.8 × 10^-6 M^-1 s^-1) and of Me-1-[α-¹³C]OH and Me-1-[α-¹³C]OCH ₂CF₃ by concerted bimolecular displacement of solvent (k_solv = 1.8 × 10^-8 s^-1) is also observed. An analysis of the rate and product data provides a value of k _az/k_s = 32 M^-1 for partitioning of Me-2 ⁺ between addition of azide ion and solvent that is nearly three-fold smaller than k_az/k_s = 83 M^-1 reported in an earlier study on the partitioning of MeO-2⁺ [J. Org. Chem. 2011, 76, 9568]. This change is attributed to a decrease in nucleophile selectivity with increasing electrophile reactivity for the activation-limited addition of solvent and azide anion to X-2⁺. These data set a limit of 1/k _s ≥ 10^-7 s for the lifetime of Me-2+ in aqueous solution. Copyright 2013 John Wiley & Sons, Ltd. A limit of 1/k _s ≥ 10^-7 s is established for the lifetime of the bridging 2-methylphenonioum ion in aqueous solution. Copyright

Full text of DOI:10.1002/poc.3145

Special Issue Article
Received: 18 March 2013,
Revised: 20 April 2013,
Accepted: 9 May 2013,
Published online in Wiley Online Library: 18 June 2013
(wileyonlinelibrary.com) DOI: 10.1002/poc.3145
Substituent effects on the formation and
nucleophile selectivity of ring-substituted
phenonium ions in aqueous solution†
Yutaka Tsujia, Shin Ogawaa and John P. Richardb*
The reaction of 2-(4-methyphenyl)ethyl tosylate (Me-1-OTs) in 50/50 (v/v) trifluoroethanol/water at 25 ꢀC is first
order in the concentration of azide anion nucleophile. A carbon-13 NMR analysis of the products of the reactions of
Me-1-[a-13C]OTs in 50/50 (v/v) trifluoroethanol/water at 25 ꢀC shows the formation of Me-1-[b-13C]OH, Me-1-[b-13C]
OCH2CF3 and Me-1-[b-13C]N3 from the trapping of a symmetrical 4-methylphenonium ion reaction intermediate
Me-2+. The formation of Me-1-[a-13C]N3 by concerted bimolecular displacement of azide ion at Me-1-[a-13C]OTs
(kN = 3.8 ꢁ 10ꢂ6 Mꢂ1 sꢂ1) and of Me-1-[a-13C]OH and Me-1-[a-13C]OCH2CF3 by concerted bimolecular displacement
of solvent (ksolv = 1.8 ꢁ 10ꢂ8 sꢂ1) is also observed. An analysis of the rate and product data provides a value of kaz/ks
= 32 Mꢂ1 for partitioning of Me-2+ between addition of azide ion and solvent that is nearly three-fold smaller than
kaz/ks = 83 Mꢂ1 reported in an earlier study on the partitioning of MeO-2+ [J. Org. Chem. 2011, 76, 9568]. This change
is attributed to a decrease in nucleophile selectivity with increasing electrophile reactivity for the activation-limited
addition of solvent and azide anion to X-2+. These data set a limit of 1/ks ≥ 10ꢂ7 s for the lifetime of Me-2+ in aqueous
solution. Copyright © 2013 John Wiley & Sons, Ltd.
Keywords: nucleophilic substitution; electrophile; phenonium ion; lifetime; selectivity
The barriers for addition of the solvent water to many simple
carbocations[1] and carbocation–anion pairs have been deter-
mined.[2–5] There is good evidence that solvolysis at secondary
carbon is by an enforced concerted reaction mechanism,
because secondary carbocations are too unstable to exist in water
for the time of a bond vibration.[6–9] A value ks ꢃ 3 ꢁ 1012 sꢂ1
(Scheme 1) has been estimated for addition of a largely aqueous
solvent to a simple tertiary carbocation, and it was proposed that
solvent reorganization with a rate constant kr ꢃ 1011 sꢂ1 is the
rate-limiting step for the overall carbocation–nucleophile combi-
nation reaction.[10,11] Electron-donating a-substituents, such as
methoxy[12] and 4-methoxyphenyl,[13–15] (Scheme 1), are sufficient
to give a carbocation with a lifetime long enough to permit
diffusion-controlled bimolecular trapping by added nucleophiles.
Surprisingly, the lifetimes of a-substituted 4-methoxybenzyl
carbocations are not strongly affected by a strongly electron-
withdrawing substituent at the a-carbon, such as a-CF3 and
a-COOEt.[16–20] This is because the increase in ks due to the
strong inductive substituent effect is offset by a decrease in
ks due to an increasing intrinsic reaction barrier that arises
from the tendency of these strongly electron-withdrawing
groups to enhance resonance delocalization of charge onto
the 4-methoxyphenyl ring substituent.[21,22]
Bridging phenonium ions were at one time the subject of
intense interest in studies to characterize internal [anchimeric]
nucleophilic assistance to solvolysis,[23] and the chemical com-
munity continues to study the generation and reaction of these
electrophiles.[24,25] However, relatively little is known about the
absolute barrier for addition of solvent to ring-substituted
phenonium ions. We are interested in designing experiments
to examine these reaction barriers, so that they may be com-
pared with barriers to nucleophile addition to related ring-
substituted 1-phenylethyl carbocations.[26–29]
Isomerization of the parent phenonium cation to the 1-phenylethyl
carbocation in a super acidic medium is estimated to be favored
by 12 kcal/mol.[30] This predicts a larger thermodynamic driving
force for addition of solvent to the phenonium cation, which will
reduce the kinetic barrier for ks for solvent addition, compared
with ks for addition to the 1-phenylethyl carbocation (Scheme 2).
On the other hand, nucleophile addition to phenonium ions
presumably proceeds by a concerted bimolecular displacement
reaction through a pentavalent transition state that avoids
formation of the more unstable primary carbocation intermediate
of a stepwise reaction. This requirement for the development of
unfavorable bonding for pentacoordinate carbon should result in
an increase of the intrinsic barrier for bimolecular substitution at
the phenonium cation, compared to the intrinsic barrier for
thermodynamically favorable collapse of the 1-phenylethyl
carbocation to the solvent adduct (Scheme 2).[21,31]
We recently reported the generation and trapping of the
4-methoxyphenonium ion MeO-2+ as an intermediate of the
solvolysis reaction of 2-(4-methoxyphenyl)ethyl tosylate in
* Correspondence to: J. P. Richard, Department of Chemistry, University at
Buffalo, SUNY, Buffalo, NY 14260, USA.
E-mail: jrichard@chem.buffalo.edu
†
This article is published in Journal of Physical Organic Chemistry as a Special
issue: In Memoriam edited by Michael Page (University of Huddersfield,
Chemistry, Queensgate, Huddersfield, HD1 3DH, United Kingdom).
a Y. Tsuji, S. Ogawa
Department of Biochemistry and Applied Chemistry, Kurume National College
of Technology, Komorinomachi, Kurume 830-8555, Japan
b J. P. Richard
Department of Chemistry, University at Buffalo, SUNY, Buffalo, NY, 14260, USA
J. Phys. Org. Chem. 2013, 26 970–976
Copyright © 2013 John Wiley & Sons, Ltd.

RING-SUBSTITUTED PHENONIUM IONS IN AQUEOUS SOLUTION
Scheme 1
EXPERIMENTAL
[
13C]-Carbon dioxide (99% enriched) was purchased from Aldrich. All
other organic and inorganic chemicals were reagent grade from
commercial sources and were used without further purification.
Syntheses
2-(4-Methylphenyl)-[1-13C]ethanol [Me-1-[a-13C]OH] was synthesized by
the procedure of Tsuno and coworkers.[33] The product was purified by
silica-gel column chromatography eluting by n-hexane/ether; yield
54%, 1H NMR (400 MHz, CDCl3): d 7.13 (m, 4H, ArH), 3.84 (dq, 2H, JH13C
= 143 Hz, JHCCH = JHCOH = 6 Hz, 13CH2OH), 2.84 (dt, 2H, J13CCH = 5 Hz, JHCCH
= 6 Hz, CH2), 2.33 (s, 3H, PhCH3), 1.39 (1H, dt, J13CCOH = 3 Hz, JHCOH = 6 Hz,
OH), 13C NMR (100.4 MHz, CDCl3): d 63.8.
Scheme 2
2-(4-Methylphenyl)ethyl 1-13
C
tosylate [Me-1-[a-13C]OTs] was
the largely aqueous solvent of 50/50 (v/v) trifluoroethanol
water.[32] The relative yields of the products of addition of
azide anion and solvent to MeO-2+ were determined and
used to calculate a value of kaz/ks = 83 Mꢂ1 for partitioning
of the electrophile between capture by azide anion and solvent.
Combining this rate constant ratio and an upper limit of
synthesized from the corresponding alcohol and 4-toluenesulfonyl
chloride by the procedure of Tipson[34]; yield 68%, mp 67–68 oC, 1H
NMR (400 MHz, CDCl3): d 7.69 (A2B2, 2H, J = 8 Hz, ArH), 7.28 (A2B2,
2H, J = 8 Hz, ArH), 7.06 (A2B2, 2H, J = 8 Hz, ArH), 6.99 (A2B2, 2H, J
= 8 Hz, ArH), 4.17 (dt, 2H, JH13C = 151 Hz, JHCCH = 7 Hz, 13CH2OTos),
2.91 (q, 2H, J13CCH = JHCCH = 7 Hz, CH2), 2.44 (s, 3H, PhCH3), 2.31
(s, 3H, PhCH3), 13C NMR (100.4 MHz, CDCl3): d 70.8.
kaz ≤ 5 ꢁ 109
M
ꢂ1sꢂ1 gave ks ≤ 6 ꢁ 107 sꢂ1 for addition
of an aqueous solvent to MeO-2+. [29] Arguments were
presented that the actual value of ks lies below this limit.[32]
Carbon-13 NMR analyses
Proton-decoupled 13C NMR spectra were recorded on a JEOL AL-400
FT-NMR spectrometer operating at 100.4 MHz. Chemical shifts were
reported in ppm, relative to a value of d = 77.0 ppm for CDCl3. The
signals for the [a-13C] and [b-13C]-positions of the products of solvolysis
of Me-1-[a-13C]OTs were monitored using a pulse angle of 45o and
relaxation delay of 70 s to obtain spectra centered at 55.0 ppm, which
spanned 7042.25 Hz and contained 65,000 data points (0.107 Hz/pt).
Generally, >1000 FT transients were averaged before determining the
final spectra.
Kinetic studies
The solvolysis reactions of Me-1-[a-13C]OTs in the absence or in the
presence of azide anion were initiated by making a 200-fold dilution of
a solution of substrate in acetonitrile into 50/50 (v/v) TFE/H2O (I = 0.5,
NaClO4) to give a final substrate concentration of 0.2 mM. The reactions
were carried out under pseudo-first-order conditions, where the concen-
tration of azide anion is >10-fold larger than the concentration of
substrate. The reaction progress was monitored for 60 days by HPLC
analyses as described in earlier work.[13,17,29] Values of the observed
first-order reaction rate constants for these reactions were determined
as the slopes of semilogarithmic plots of reaction progress against time.
We report here the generation of Me-2+ as an intermediate of
the solvolysis reaction of 2-(4-methylphenyl)ethyl tosylate in the
largely aqueous solvent of 50/50 (v/v) trifluoroethanol/water.
The 4-Me- for 4-MeO- substitution results in a < three-fold
decrease in kaz/ks that is consistent with a modest decrease in
nucleophile selectivity, with increasing electrophile reactivity,
due to a shift to a transition state with weakened bonding of
the nucleophile/leaving group to the central carbon. The results
provide strong evidence that there is a significant chemical
activation barrier to concerted bimolecular displacement reac-
tions of azide anion and solvent at the strained cyclopropyl rings
of X-2+ so that the absolute lifetimes of these intermediates
generated by laser flash photolysis might be obtained by direct
measurement.
HPLC product analyses
The solvolysis reactions of Me-1-OTs were initiated by making a 200-fold
dilution of a solution of substrate in acetonitrile into 50/50 (v/v) TFE/H2O
(I = 0.5, NaClO4) to give a final substrate concentration was 0.20 mM. The
J. Phys. Org. Chem. 2013, 26 970–976
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

Y. TSUJI, S. OGAWA AND J. P. RICHARD
reactions were carried out at room temperature, and after a reaction time
of ca two months the relative yields of the azide anion (Me-1-N3), water
(Me-1-OH) and trifluoroethanol (Me-1-OCH2CF3) adducts were deter-
mined directly from the relative peak areas from HPLC analysis, with
peak detection at 265 nm. This is lmax for the 2-(4-methylphenyl)ethyl
alcohol, and it was shown in an earlier study of the reactions of
1-(4-methylphenyl)ethyl derivatives that the molar extinction coeffi-
cients for products of the nucleophilic substitution reactions of
azide anion (4-MeC6H4CH(CH3)N3 and solvent [(4-MeC6H4CH(CH3)
OH), (4-MeC6H4CH(CH3)OCH2CF3) adducts are the same.[29]
Me-1-N3, Me-1-OH and Me-1-OTFE labeled with carbon-13 in
the a-position [a-13C] and the b-position [b-13C] were deter-
mined by carbon-13 NMR analyses, and the product ratios
[a-13C]/[b-13C] are reported in Table 2.
DISCUSSION
The observed first-order rate constant (ksolv obs
for solvolysis of Me-1-OTs in 50/50 (v/v) TFE/H2O (I = 0.5, NaClO4)
)
= 2.5 ꢁ 10ꢂ7 sꢂ1
and at 25 ꢀC is eight-fold smaller than (ksolv obs
)
= 1.95 ꢁ 10ꢂ6 sꢂ1
Carbon-13 NMR product analyses
determined for solvolysis of MeO-1-OTs under the same reac-
tion conditions.[32] This is consistent with internal nucleophilic
push by the methoxyphenyl ring to ionization of MeO-1-OTs
to form MeO-2+. The solvolysis of MeO-1-[a-13C]OTs in 50/50
(v/v) TFE/H2O gives equal yields of products MeO-1-OH and
MeO-1-OTFE labeled with [a-13C] and with [b-13C], so that
there is no detectable solvolysis by a concerted bimolecular
The solvolysis reactions of Me-1-[a-13C]OTs in the absence or in the
presence of Nꢂ3 were initiated by preparing 20 mg of substrate in aceto-
nitrile and adding this to 400 mL of 50/50 (v/v) TFE/H2O (I = 0.5, NaClO4)
to give a final substrate concentration of 0.18 mM. After a ca two-month
reaction time, the reaction products were extracted into toluene. The
toluene layer was washed with water, dried over MgSO4 and removed
by evaporation. The resulting residue was dissolved in 0.6 mL of CDCl3
and analyzed by 13C NMR to determine relative yields of products labeled
with 13C at the a- and b-positions. The relative product yields were deter-
mined by integration of the following peaks for the [13C]-enriched
carbons: Me-1-[a-13C]OH, 63.77 ppm; Me-1-[b-13C]OH, 38.72 ppm;
Me-1-[a-13C]OCH2CF3, 73.77 ppm; Me-1-[b-13C]OCH2CF3, 35.67 ppm;
Me-1-[a-13C]N3, 52.56 ppm; Me-1-[b-13C]N3, 34.19 ppm.
displacement mechanism. This shows that (ksolv obs is equal to
)
kΔ for intramolecular displacement of tosylate anion by the
aromatic ring to give the phenonium cation MeO-2+.[32]
By contrast, carbon-13 NMR analyses of the products of solvol-
ysis of Me-1-[a-13C]OTs in 50/50 (v/v) TFE/H2O show a small
excess of carbon-13 label at the a-carbon for both Me-1-OH
and Me-1-OTFE, for which [[a-13C]/[b-13C] = 1.16 and 1.08,
respectively (Table 2). This solvolysis reaction therefore proceeds
mainly by a unimolecular stepwise DN + AN mechanism,[35]
with rate constant kΔ and scrambling of the label at the a- and
b-carbons; and, by a slower concurrent concerted bimolecular
ANDN mechanism,[35] with rate constant ksolv, for direct addition
of solvent to the a-carbon (Scheme 2) to form products labeled
with carbon-13 at this carbon. The relationship between the
product ratio [a-13C]/[b-13C] and the rate constants kΔ and ksolv
is given by eqn 1. Substitution of [a-13C]/[b-13C] = 1.16 for the
reaction of HOH into eqn 1 gives a value of ksolv/kΔ = 0.080.
RESULTS
The first-order rate constant (ksolv obs
)
= 2.5 ꢁ 10ꢂ7 sꢂ1 was deter-
mined for solvolysis of Me-1-OTs in the absence of azide anion
by monitoring the disappearance of substrate by HPLC. Figure 1
shows a plot of values of the pseudo-first-order rate constants
kobs against the concentration of added azide anion for reactions
of Me-1-OTs in 50/50 (v/v) TFE/H2O (I = 0.5, NaClO4) at 25 ꢀC. The
second-order rate constant (kNu obs
)
= 3.8 (ꢄ 0.1) ꢁ 10ꢂ6 Mꢂ1 sꢂ1
was determined as the slope of a linear plot of kobs against [Nꢂ3 ]
(Figure 1). The yields of the products of the nucleophilic substitu-
tion reactions of azide anion (Me-1-N3), water (Me-1-OH) and
trifluoroethanol (Me-1-OTFE) with Me-1-OTs in 50/50 (v/v)
TFE/H2O (I = 0.5, NaClO4) and at 25 ꢀC, determined by HPLC
analyses, are reported in Table 1. The reactions of Me-1-
[a-13C]OTs with azide anion were examined in 50/50 (v/v)
TFE/H2O (I = 0.5, NaClO4) at 25 ꢀC. The relative yields of
Combining the observed rate constant (ksolv obs
)
= 2.50 ꢁ 10ꢂ7 sꢂ1
= ksolv + kΔ with ksolv/kΔ = 0.080 gives values of kΔ = 2.32 ꢁ 10ꢂ1
and ksolv = 1.8 ꢁ 10ꢂ8 sꢂ1 (Scheme 3).
Equations 2 and 3 were used to partition the total yield of
solvolysis reaction products, fMe-1-OSol (Table 1), into the product
yields from the DN + AN [ðfROSol
Þ
] and ANDN [ðfROSol
Þ
]
DNþAN
ANDN
solvolysis reactions (Table 3).[35] The yield of Me-1-OTFE is low
(7% of total solvolysis products, and we therefore make the
simplifying assumption that ksolv/kΔ = 0.080 for the reactions of
both water and trifluoroethanol). The use of the smaller
observed ratio of ksolv/kΔ
= 0.04 for the reaction of
trifluoroethanol does not cause a significant change (≤0.2%) in
the product yields for Me-1-OSol = Me-1-OH + Me-1-OTFE
reported in Table 3.
ꢀ
ꢁ
ꢀ
ꢁ
13
a ꢂ C
ksolv þ 0:5kΔ
0:5kΔ
¼
(1)
13
b ꢂ C
ꢀ
ꢁ
kΔ
ðfROSol
Þ
¼ ðfROSolÞobs
(2)
(3)
DNþAN
ksolv þ kΔ
ðfROSol
Þ
¼ ðfROSolÞobs- ðfROSolÞ
DNþAN
ANDN
Figure 1. The increase in the observed pseudo-first-order rate constant
for the reactions of Me-1-OTs as the initial concentration of the azide
anion nucleophile is increased
The second-order rate constant kNu = 3.8 ꢁ 10ꢂ6
M
ꢂ1sꢂ1 for
bimolecular substitution of azide anion at Me-1-OTs in 50/50
wileyonlinelibrary.com/journal/poc
Copyright © 2013 John Wiley & Sons, Ltd. J. Phys. Org. Chem. 2013, 26 970–976

RING-SUBSTITUTED PHENONIUM IONS IN AQUEOUS SOLUTION
Table 1. The yields of the products and the observed product rate constant ratios (kaz/ks)obs for the reaction of Me-1-OTs in 50/50
ꢀ
(v/v) TFE/H2O (I = 0.5, NaClO4) at 25 C and different fixed concentrations of azide anion
[Nꢂ3 ]/M
fRN3
fROH
fROTFE
fROSol
(kaz/ks)obs (Mꢂ1
)
a
a
a
b
c
0.0050
0.0060
0.0080
0.010
0.192
0.223
0.279
0.328
0.751
0.725
0.673
0.626
0.056
0.052
0.048
0.045
0.807
0.777
0.721
0.671
48
48
48
49
aFractional product yields determined by HPLC analyses.
bThe sum of the fractional yields of the water and trifluoroethanol adducts.
cObserved product rate constant ratio determined from the ratio of product yields using eqn 4.
constant ratios (kaz/ks)obs (Mꢂ1) for the reaction of azide anion
and solvent with Me-1-OTs, calculated from the product yields
using eq 4. The observed product rate constant ratio (kaz/ks)obs
= 48 Mꢂ1 is substantially larger than the rate constant ratio
Table 2. The ratio of the yields of product labeled with
carbon-13 in the a- and b-positions for the reactions of
MeO-1-[a-13C]OTs in 50/50 (v/v) TFE/H2O (I = 0.5, NaClO4)
ꢀ
(kNu obs/(ksolv)obs
= 15.2 Mꢂ1 determined by kinetic analyses. This
)
at 25 C and different fixed concentrations of azide anion
shows that a large fraction of the total yield of Me-1-N3 forms by
[a-13C]/[b-13C] a
trapping of the symmetrical intermediate Me-2+.
[Nꢂ3 ]/M
MeO-1-N3
MeO-1-OH
Me-1-OTFE
!
fRN3
ꢂ
ꢃ
0.00
1.16
1.15
1.14
1.16
1.18
1.08
1.06
1.08
1.08
1.06
ðkaz=ksÞobs
¼
(4)
-
fROSol N3
0.0050
0.0060
0.080
0.010
2.25
2.29
2.45
2.54
!
ꢂ
ꢃ
-
kNu N3
ꢂ
ꢃ
ðfRN3
Þ
¼
(5)
(6)
ANDN
-
ðksolv
Þ
obs þ kNu N3
aDetermined as the ratio of the carbon-13 peak area for the
products labeled with carbon-13 at the a- and b-carbon.
ðfRN3
Þ
¼
ðfRN3
Þ
obs- ðfRN3
Þ
DNþAN
ANDN
!
ꢀ
ꢁ
a-13C
b-13C
ðfRN3
Þ
þ 0:5ðfRN3Þ
DNþAN
ANDN
(v/v) TFE/H2O (I = 0.5, NaClO4) is similar to kNu = 3.6 ꢁ 10ꢂ6
¼
(7)
(8)
0:5ðfRN3
Þ
DNþAN
M
ꢂ1sꢂ1 for the substitution reaction at MeO-1-OTs under the
same conditions.[32] This shows that there is no significant push
by the 4-OMe group to nucleophilic displacement of azide anion
at MeO-1-OTs. Table 1 reports the observed product rate
!
ðfRN3
Þ
DNþAN
ꢂ
ꢃ
ðkaz=ksÞCþ
¼
-
ðfROSol
Þ
N3
DNþAN
Scheme 3
J. Phys. Org. Chem. 2013, 26 970–976
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

Y. TSUJI, S. OGAWA AND J. P. RICHARD
Table 3. Calculated yields of the azide anion and solvent adducts to MeO-1-[a-13C]OTs by competing DN + AN and ANDN reaction
mechanisms in 50/50 (v/v) TFE/H2O (I = 0.5, NaClO4) at 25 ꢀC determined by an analysis of the dependence of the observed rate
constants and reaction products on the concentration of azide anion
"
#
13
a- C
½
ꢅ
[Nꢂ3 ]
ðfRN3
Þ
ðfRN3
Þ
ðfROSol
Þ
ðfROSolÞ
DNþAN
d
(kaz/ks)C+
e
f
13
b- C
ANDN
DNþAN
ANDN
½
ꢅ
a
b
c
RN3
Observedg
Calculatedh
0
0
0
0.074
0.060
0.058
0.053
0.050
0.926
0.747
0.719
0.668
0.621
–
–
–
0.005
0.006
0.008
0.010
0.071
0.084
0.108
0.132
0.121
0.139
0.171
0.197
32
32
32
32
2.25
2.29
2.45
2.54
2.24
2.21
2.26
2.34
aCalculated using eqn 5.
bCalculated using eqn 6.
cCalculated using eqn 3.
dCalculated using eqn 2.
eCalculated using eqn 8.
fThe ratio of the yields of MeO-1-[a-13C]-N3 and MeO-1-[b-13C]-N3.
gDetermined by carbon-13 NMR analyses.
hCalculated using eqn 7.
Equations 5 and 6, derived for Scheme 3, were used to
partition the total yield of azide anion adduct Me-1-N3 (Table 1),
determined for the reaction of azide anion and solvent with methyl
bromide in water.[36] This difference in nucleophile selectivity is
consistent with a stronger stabilization of the transition state for
bimolecular substitution at aliphatic carbon by interactions of reac-
tive nucleophiles with the bromide anion leaving group, compared
to tosylate anion leaving group. Such transition state stabilization
by interactions between nucleophile and leaving group across a
pentavalent carbon transition state has been noted in other work
and given the name “symbiosis”.[6,37–40]
Chart 1 shows the very different effects of replacing a 4-Me by
a 4-OMe ring substituent on nucleophile selectivity for addition
of azide anion and a solvent of aqueous/trifluoroethanol to
phenonium cations and to the ring substituted 1-phenylethyl
carbocations MeO-3+ and Me-3+. This difference is not related
to a difference in the effect of these substitutions on the reaction
thermodynamic driving force, because the 4-OMe substituent
provides a similar stabilization of the Me-2+ cation in bromide
ion transfer reaction in the gas phase (Scheme 4A),[41] and the
Me-3+ cation in a hydroxide ion transfer reaction in aqueous/
trifluoroethanol (Scheme 4B).[29]
into the product yields from the stepwise DN + AN [ðfRN3
Þ
]
DNþAN
and concerted ANDN [ðfRN3
Þ
] substitution reactions (Table 3).
ANDN
Substitution of these product yields into eqn 7, derived for Scheme 3,
gives the theoretical ratios [a-13C]/[b-13C] for Me-1-N3 labeled
with carbon-13 at the a- and b-carbons. The good agreement
between the calculated and observed ratios shows that Scheme 3
provides a complete and self-consistent rationalization of the exper-
imental data. Table 3 also reports values for the rate constant ratio
(kaz/ks)C+ for trapping of the phenonium cation intermediate X-2+,
determined from the product yields for the stepwise DN + AN
reaction using eqn 8.
Reactivity–selectivity relationships
Chart 1 reports the nucleophile selectivity kaz/ks (Mꢂ1) deter-
mined for addition of azide anion and solvent to several electro-
philes, and the absolute rate constant ks in the cases where these
are known. The nucleophile selectivity (kNu obs/ksolv = kaz/ks =
)
[3.8 ꢁ 10ꢂ6 Mꢂ1sꢂ1]/[1.8 ꢁ 10ꢂ8 sꢂ1] = 210 Mꢂ1 for the reaction
of Me-1-OTs serves as a benchmark for a concerted bimolecular
displacement reaction at a primary alkyl tosylate. The value of log
(kaz/ks) = 2.3 for this reaction is significantly smaller than the
Swain–Scott nucleophilicity parameter n = log(kaz/ks)MeBr = 3.9
The large decrease in nucleophile selectivity from (kaz/ks) =
100 to 1.1 Mꢂ1 for nucleophile addition to MeO-3+ and Me-3+
(Chart 1) reflects the increase in ks for the activation-limited
reaction of solvent, relative to the invariant value of kaz = 5 ꢁ 109
for diffusion-limited addition of azide ion to MeO-3+ and
Chart 1
wileyonlinelibrary.com/journal/poc
Copyright © 2013 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2013, 26 970–976

RING-SUBSTITUTED PHENONIUM IONS IN AQUEOUS SOLUTION
Scheme 4 suggests that there is a smaller effect of changing
phenyl ring substituents on the reactivity of X-2+ compared
with X-3+ that would be reflected as an increase in the observed
103-fold difference in the reactivity of 4 and 5, with increasing
electrophile reactivity. If this were the case, then ks ≤ 6 ꢁ 106 sꢂ1
and kaz ≤ 2 ꢁ 108 Mꢂ1sꢂ1 for nucleophile addition to CH3-2+. In
any case, these data set an approximate lower limit of 1/ks ≥ 10ꢂ7
s for the lifetime of Me-2+ in aqueous solution and a longer lifetime
for CH3O-2+. These results suggest that the absolute lifetimes of
these phenonium may be readily determined by experiment,
provided efficient procedures for their generation by laser flash
photolysis can be developed.[48]
Scheme 4
Me-3+.[26,29,42,43] By comparison, the smaller 2.6-fold change
in nucleophile selectivity from kaz/ks = 83 Mꢂ1 for MeO-2+
to 32 Mꢂ1 for Me-2+ is consistent with a shift to a “loose”
transition state, with weakened bonding of the nucleophile
and leaving group to the central carbon, as the pKa of the
carbon leaving group is decreased (Scheme 5).[44] The alternative
explanation is that the change in selectivity reflects a 2.6-fold
increase in ks relative to kaz = 5 ꢁ 109 for the diffusion-limited
reactions of azide ion. However, such a small 2.6-fold increase
in ks is inconsistent with the larger eight-fold difference in
the values of kΔ for the ionization of Me-1-OTs and MeO-1-OTs
(ΔΔG†Δ ꢃ 1.2 kcal/mol) and still larger ca 3000-fold effect of
Me- for MeO- substitution on Keq = kf/kr for a bromide anion
transfer reaction (ΔΔGo ꢃ 4.7 kcal/mol, Scheme 4). This ca.
3000-fold effect on Keq must be expressed as changes in both
the rate constant kΔ for formation of the phenonium ion and
in the rate constant for the reactions in the reverse direction,
so that the Me- for MeO- substitution is predicted to cause a
>100-fold increase in ks for solvent addition to MeO-2+.
Acknowledgements
We acknowledge the National Institutes of Health (GM 39754) for
generous support of this work. This paper is dedicated to Rory A.
More O’Ferrall. We acknowledge the purity and significance of
his work to define the mechanisms organic reactions, his efforts
to promote collegial interactions within this field of study and
his friendship.
REFERENCES
[1] R. More O’Ferrall, Adv. Phy. Org. Chem. 2010, 44, 19.
[2] M. Teshima, Y. Tsuji, J. P. Richard, J. Phys. Org. Chem. 2010, 23, 730.
[3] Y. Tsuji, J. P. Richard, Chem. Rec. 2005, 5, 94.
[4] J. P. Richard, T. L. Amyes, M. M. Toteva, Y. Tsuji, Adv. Phys. Org. Chem.
2004, 39, 1.
[5] J. P. Richard, Y. Tsuji, J. Am. Chem. Soc. 2000, 122, 3963.
[6] P. E. Dietze, W. P. Jencks, J. Am. Chem. Soc. 1986, 108, 4549.
[7] P. E. Dietze, W. P. Jencks, J. Am. Chem. Soc. 1987, 109, 2057.
[8] W. P. Jencks, Acc. Chem. Res. 1980, 13, 161.
Lifetime of phenonium ions
We conclude that the decrease in kaz/ks from 83 Mꢂ1 for MeO-2+
to 32 Mꢂ1 for Me-2+ reflects a 2.6-fold decrease in the nucleo-
phile selectivity for the activation-limited addition of solvent
and azide anion to MeO-2+. This sets upper limits of kaz ≤ 5 ꢁ
[9] W. P. Jencks, Chem. Soc. Rev. 1981, 10, 345.
[10] M. M. Toteva, J. P. Richard, J. Am. Chem. Soc. 1996, 118, 11434.
[11] J. P. Richard, M. M. Toteva, T. L. Amyes, Org. Lett. 2001, 3, 2225.
[12] T. L. Amyes, W. P. Jencks, J. Am. Chem. Soc. 1989, 111, 7888.
[13] T. L. Amyes, J. P. Richard, J. Am. Chem. Soc. 1990, 112, 9507.
[14] M. M. Toteva, J. P. Richard, J. Am. Chem. Soc. 2002, 124, 9798.
[15] T. L. Amyes, J. P. Richard, J. Chem. Soc., Chem. Commun. 1991, 200.
[16] J. P. Richard, T. L. Amyes, L. Bei, V. Stubblefield, J. Am. Chem. Soc.
1990, 112, 9513.
[17] J. P. Richard, J. Am. Chem. Soc. 1989, 111, 1455.
[18] J. P. Richard, J. Am. Chem. Soc. 1986, 108, 6819.
[19] J. P. Richard, T. L. Amyes, I. W. Stevens, Tetrahedron Lett. 1991,
34, 4255.
[20] T. L. Amyes, I. W. Stevens, J. P. Richard, J. Org. Chem. 1993, 58, 6057.
[21] J. P. Richard, Tetrahedron 1995, 51, 1535.
[22] J. P. Richard, T. L. Amyes, M. M. Toteva, Acc. Chem. Res. 2001, 34, 981.
[23] C. J. Lancelot, D. J. Cram, P. v. R. Schleyer, Vol. 3(Ed.: G. A. Olah),
Wiley, 1972, 1347.
109
M
ꢂ1sꢂ1 and ks ≤ 1.6 ꢁ 108 sꢂ1 for addition of azide
anion and solvent to Me-2+. The lifetime of the neutral
spiro[2,5]-octa-1,4-diene-3-one 4 in water (ks = 3.4 ꢁ 10ꢂ3 sꢂ1 [45]
)
is 1000-fold longer than for the simple p-quinone methide
5.[46,47] If there were a constant 1000-fold difference in reactivity
for changing phenyl ring substituents, then the value of ks = 6 ꢁ
109 sꢂ1 estimated for addition of aqueous/trifluoroethanol to
CH3-3+[27,29] and (kaz/ks) = 32 Mꢂ1 for nucleophile addition to
CH3-2+ gives values of ks = 6 ꢁ 106 sꢂ1 and kaz = 2 ꢁ 108 Mꢂ1sꢂ1
for addition of solvent and azide anion to CH3-2+.
[24] S. Protti, D. Dondi, M. Mella, M. Fagnoni, A. Albini, Eur. J. Org. Chem.
2011, 17, 3229.
[25] I. Manet, S. Monti, G. Grabner, S. Protti, D. Dondi, V. Dichiarante,
M. Fagnoni, A. Albini, Chem. Eur. J. 2008, 14, 1029.
[26] J. P. Richard, W. P. Jencks, J. Am. Chem. Soc. 1982, 104, 4689.
[27] J. P. Richard, W. P. Jencks, J. Am. Chem. Soc. 1984, 106, 1373.
[28] J. P. Richard, W. P. Jencks, J. Am. Chem. Soc. 1984, 106, 1396.
[29] J. P. Richard, M. E. Rothenberg, W. P. Jencks, J. Am. Chem. Soc. 1984,
106, 1361.
[30] E. del Rio, M. I. Menendez, R. Lopez, T. L. Sordo, J. Phys. Chem. A
Scheme 5
2000, 104, 5568.
J. Phys. Org. Chem. 2013, 26 970–976
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

Y. TSUJI, S. OGAWA AND J. P. RICHARD
[31] J. P. Richard, T. L. Amyes, K. B. Williams, Pure Appl. Chem. 1998, 70, 2007.
[32] Y. Tsuji, D. Hara, R. Hagimoto, J. P. Richard, J. Org. Chem. 2011, 76, 9568.
[33] M. Fujio, Y. Seki, Y. Tsuno, Mem. Fac. Sci., Kyushu Univ., Ser. C
1978, 11, 119.
[41] M. Mishima, Y. Tsuno, M. Fujio, Chem. Lett. 1990, 19, 2277.
[42] R. A. McClelland, F. L. Cozens, S. Steenken, T. L. Amyes, J. P. Richard,
J. Chem. Soc., Perkin Trans. 1993, 2, 1717.
[43] R. A. McClelland, V. M. Kanagasabapathy, N. S. Banait, S. Steenken,
J. Am. Chem. Soc. 1991, 113, 1009.
[34] R. S. Tipson, J. Org. Chem. 1944, 9, 235.
[35] R. D. Guthrie, W. P. Jencks, Acc. Chem. Res. 1989, 22, 343.
[36] J. Koskikallio, Acta Chem. Scand. 1969, 23, 1477.
[37] J. P. Richard, W. P. Jencks, J. Am. Chem. Soc. 1984, 106, 1383.
[38] R. G. Pearson, J. Songstad, J. Am. Chem. Soc. 1967, 89, 1827.
[39] R. G. Pearson, J. Songstad, J. Org. Chem. 1967, 32, 2899.
[40] E. S. Lewis, S. H. Vanderpool, J. Am. Chem. Soc. 1978, 100, 6421.
[44] W. P. Jencks, Chem. Rev. 1985, 85, 511.
[45] R. Baird, S. Winstein, J. Am. Chem. Soc. 1957, 79, 4238.
[46] M. M. Toteva, M. Moran, T. L. Amyes, J. P. Richard, J. Am. Chem. Soc.
2003, 125, 8814.
[47] M. M. Toteva, J. P. Richard, Adv. Phy. Org. Chem. 2011, 45, 39.
[48] R. A. McClelland, Tetrahedron 1996, 52, 6823.
wileyonlinelibrary.com/journal/poc
Copyright © 2013 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2013, 26 970–976