Reversible formation of aryloxenium ions from the corresponding quinols under acidic conditions

Article abstract of DOI:10.1002/poc.2999

Quinols, 1, are products of the hydration of O-aryloxenium ions, 2, and N-arylnitrenium ions, 3, and they are being investigated for medical uses. Under acidic conditions (pH 1-3) kinetics and products of Br^- trapping demonstrate that 1a, 4-phenyl-4-hydroxy-2,5-cyclohexadienone, and 1b, 4-p-tolyl-4-hydroxy-2,5-cyclohexadienone, generate the corresponding oxenium ions 2a and 2b, respectively, as steady-state intermediates. Formation and trapping of the oxenium ions occurs in competition with the acid catalyzed dienone-phenol rearrangement. Because oxenium ion formation is reversible, the ion can only be detected by trapping with a nucleophile. Br^- is an efficient trap under acidic conditions because, unlike N₃ ^-, it is not protonated under those conditions. Attempts to detect the oxenium ions 2a and 2b at pH 4.6 and 7.1 with N₃^- were unsuccessful indicating that oxenium ion formation only occurs under acidic conditions. The oxenium ion 2c could not be detected under acidic conditions from the quinol 1c, 4-(benzothiazol-2-yl)-4-hydroxy-2,5-cyclohexadienone, by Br^- trapping methods, even though this ion can be detected during hydrolysis of the corresponding ester, 4c. Although the benzothiazol-2-yl group is a resonance electron donor that is capable of stabilizing an O-aryloxenium ion, it is also a strong inductive electron withdrawing group that hinders the formation of 2c from 1c by decreasing the extent of protonation of 1c to generate 1cH⁺ and by destabilizing the transition state for ionization of 1cH⁺. Generation of an oxenium ion from the corresponding quinol is feasible under acidic conditions as long as the 4-substituent of the quinol is both a resonance and inductive electron donor. Copyright

Full text of DOI:10.1002/poc.2999

Research Article
Received: 02 May 2012,
Revised: 01 June 2012,
Accepted: 19 June 2012,
Published online in Wiley Online Library: 2012
(wileyonlinelibrary.com) DOI: 10.1002/poc.2999
Reversible formation of aryloxenium ions from
the corresponding quinols under
acidic conditions
a
a
a
Mrinal Chakraborty , Christopher F. Brzozowski and Michael Novak *
Quinols, 1, are products of the hydration of O-aryloxenium ions, 2, and N-arylnitrenium ions, 3, and they are being
–
investigated for medical uses. Under acidic conditions (pH 1–3) kinetics and products of Br trapping demonstrate
that 1a, 4-phenyl-4-hydroxy-2,5-cyclohexadienone, and 1b, 4-p-tolyl-4-hydroxy-2,5-cyclohexadienone, generate
the corresponding oxenium ions 2a and 2b, respectively, as steady-state intermediates. Formation and trapping
of the oxenium ions occurs in competition with the acid catalyzed dienone–phenol rearrangement. Because oxenium
–
ion formation is reversible, the ion can only be detected by trapping with a nucleophile. Br is an efficient trap under
–
acidic conditions because, unlike N
3
, it is not protonated under those conditions. Attempts to detect the oxenium
–
ions 2a and 2b at pH 4.6 and 7.1 with N were unsuccessful indicating that oxenium ion formation only occurs
3
under acidic conditions. The oxenium ion 2c could not be detected under acidic conditions from the quinol 1c,
–
4
-(benzothiazol-2-yl)-4-hydroxy-2,5-cyclohexadienone, by Br trapping methods, even though this ion can be
detected during hydrolysis of the corresponding ester, 4c. Although the benzothiazol-2-yl group is a resonance
electron donor that is capable of stabilizing an O-aryloxenium ion, it is also a strong inductive electron withdrawing
+
group that hinders the formation of 2c from 1c by decreasing the extent of protonation of 1c to generate 1cH and
+
by destabilizing the transition state for ionization of 1cH . Generation of an oxenium ion from the corresponding
quinol is feasible under acidic conditions as long as the 4-substituent of the quinol is both a resonance and inductive
electron donor. Copyright © 2012 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this paper.
Keywords: aryloxenium ions; bromide trapping; dienone–phenol rearrangement; quinols (4-substitued-4-hydroxy-2,5-
cyclohexadienones); steady state intermediates
their known decomposition in acidic aqueous solution via
INTRODUCTION
the dienone–phenol rearrangement into hydroquinone deriva-
[
29–32]
Quinols (4-substituted-4-hydroxy-2,5-cyclohexadienones), 1, are
the ubiquitous hydration products of 4-substituted-O-
aryloxenium ions, 2, and N-arylnitrenium ions, 3, generated
tives.
would necessarily be a reversible process, the rapid formation of
quinols from hydration of oxenium ions (via k of Scheme 2)
having been demonstrated on numerous occasions,
Because the generation of the oxenium ion in water
s
[
1–17]
[1–6]
under aqueous conditions (Scheme 1).
They are increasingly
the
being investigated as antitumor drugs and as trypanocides
detection of the oxenium ion intermediate presents a potentially
challenging problem. Fortunately, low steady state concentrations
of aryloxenium ions can be detected by the well characterized
reactions of oxenium ions (via knu of Scheme 2) with the
[18–25]
(Chart 1).
It is thought that the quinols act as double
Michael acceptors in biological systems, targeting proteins
[
20]
containing vicinal cysteine units. A number of target proteins
–
– [1–6]
have been identified, including the redox-regulatory protein
anionic nucleophiles N
3
and Br ,
and by the kinetics of quinol
[
20,22]
thioredoxin.
Quinols are also useful synthetic intermediates
decomposition in the presence of these nucleophiles.
[
26–28]
that have been utilized in a variety of applications.
In this study, we examined the ability of three quinols, 1a–c,
(Scheme 1 and Chart 1) to generate oxenium ions in aqueous
Our interest in the chemistry of quinols and related compounds
stems from a long-term involvement in the chemistry of
ꢀ
solution under mild reaction conditions: 30 C, pH 1–7. The
[
1–8,12–17]
arylnitrenium and aryloxenium ions.
We were intrigued
quinols 1a and 1b were chosen because the kinetics of their
formation from the hydration of the corresponding oxenium
ions, 2a and 2b, respectively, is well characterized under just
by the possibility that quinols could be precursors to aryloxenium
ions under mild conditions in aqueous solution (Scheme 2). Quinol
esters, 4, have been shown to be efficient precursors to
ꢀ
aryloxenium ions at 30 C in aqueous solution without need of
[
1–3,6]
catalysts (Scheme 1).
The hydroxide of the quinol is a
significantly poorer leaving group than the carboxylate of the
corresponding ester, so the quinol would likely require acid
catalysis for efficient generation of the oxenium ion. We are not
aware of any reports of such reactions. This unexplored area of
quinol chemistry could serve as an alternative mode of action
of quinols as biological electrophiles, and as an alternative to
* Correspondence to: M. Novak, Department of Chemistry and Biochemistry,
Miami University, Oxford, OH 45056, USA.
E-mail novakm@muohio.edu
a M. Chakraborty, C. F. Brzozowski, M. Novak
Department of Chemistry and Biochemistry, Miami University, Oxford, OH,
45056, USA
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.

M. CHAKRABORTY ET AL.
[1,4]
such conditions.
The benzothiazole quinol, 1c, was chosen
accelerate their slow decomposition at pH 7.1. The azide-
because it was the first quinol demonstrated to have antitumor
adducts, 5a, 5b, and 6c, for all three oxenium ions, 2a–c, are
[18]
[1,3,6]
activity,
and it is also the known hydration product of the
available from previous studies (Chart 2).
None of these
[6]
oxenium ion 2c.
compounds could be detected by HPLC in pH 7.1 reaction
mixtures of the respective quinols. The rate constant for
–
3
trapping 2b by N
, kaz, has been directly measured under these
RESULTS
reaction conditions by fast UV methods following laser flash
9
–1 –1
photolysis of 4b as (6.6 ꢂ 0.2) ꢁ 10 M s , while the measured
6
–1
3
–1 [4]
Kinetics of the decomposition of 1a–c were monitored in pH 7.1
.02 M Na HPO /NaH PO buffer, pH 4.6 0.02 M NaOAc/HOAc
buffer, pH 4.6 0.10 M N
solutions in 5 vol% CH CN–H
k
s
is (5.8 ꢂ 0.4) ꢁ 10 s , so kaz/k
The cation 2b would have been detected in 50 mM N
observed slow decomposition of 1b led to 2b. Trapping
efficiencies (kaz/k ) for 2a and 2c are less favorable, and the rates
s
for this ion is 1.1 ꢁ 10 M .
–
3
0
2
4
2
4
if the
–
3
/HN
3
buffer, and 0.003–0.10 M HClO
4
ꢀ
3
2
O, m = 0.5 (NaClO
4
–
)) at 30 C, with
s
5
initial concentrations of quinols of ~5 ꢁ 10 M, by high-
of disappearance of 1a and 1c are slow under these conditions,
so conclusions are less certain in these cases.
At pH 4.6 the quinols are stable for periods up to 1 month.
performance liquid chromatography (HPLC) and UV methods
[
1–3,6]
previously described.
At pH 7.1 the quinols underwent very slow decomposition.
The p-tolyl quinol, 1b, was investigated by HPLC in detail
HPLC data for 1b are shown in Figure S2. Incubation of 1b in
–
3
pH 4.6 100 mM N
/HN
3
buffer did not lead to accelerated
(
Figure S1). Over three weeks about 10% of 1b disappeared from
the reaction mixture. The rate constant for the decomposition of
b, determined from the initial rate by dividing the initial
slope (in units of peak area/time) by the initial HPLC peak area
decomposition of 1b. The known azide-adduct of 1b, 5b, is
stable under the reaction conditions, and it would have been
detected by HPLC at concentrations ~5% of the initial
concentration of 1b. No evidence for trapping of the oxenium
1
–
8 –1
for the quinol, is (10.0 ꢂ 0.3) ꢁ 10 s . In the presence of
ion 2b was found at this pH.
At pH < 3.5 it is not possible to use N as a trap because it is
completely protonated under these conditions, but Br is an
efficient cation trap under acidic conditions. At pH 1.0, Br traps
–
–
3
5
0 mM N
is (6.0 ꢂ 0.3) ꢁ 10 s . At pH 7.1, N
rate of decomposition of 1b. The other two quinols were
3
, the measured rate constant for decomposition of 1b
–8
–1
–
3
–
does not accelerate the
–
–
3
examined in less detail, but it was also clear that N does not
the cation 2a produced from acid catalyzed decomposition of 4a
[
2]
–
to generate 7a at the expense of 1a. The Br /solvent selectivity,
–
1 [2]
k /k , for 2a is (46 ꢂ 5) M . This is somewhat smaller than
Br
s
–
1
k /k of (77 ꢂ 5) M for 2a, but still large enough that
az
s
–
[1,2]
Br trapping is feasible.
Under mildly acidic conditions the
dienone–phenol rearrangement of 1a and 1b occurs at a modest
Scheme 1. Generation of quinols from hydration of aryloxenium and
arylnitrenium ions
Scheme 2. Possible generation of 2 from 1 in competition with die-
none–phenol rearrangement
Chart 1. Selected quinols investigated as antitumor drugs and/or
trypanocides
Chart 2. Azide and bromide adducts of 2a–c, and dienone–phenol
rearrangement products of 1a and 1b
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)

OXENIUM ION CHEMISTRY
–
–
rate. Addition of Br to these reaction mixtures causes an
constant for rearrangement of the quinol, k /k is the Br /solvent
Br s
increase in the rate of disappearance of both quinols, and leads
selectivity ratio defined in Scheme 2, and the composite rate
constant k depends on the ionization equilibria of the quinol.
Both composite rate constants are fully defined in the Discussion
[
2]
to the formation of the bromide adducts 7a and 7b, along
with the rearrangement products 8a and 8b (Chart 2). Table 1
provides a summary of rate constants, kobs, obtained for these
two compounds by monitoring the disappearance of the HPLC
peak for the quinol as a function of time under acidic conditions
t
2
–1
Section. The data provide a value of kBr/k
s
of (2.9 ꢂ 0.2) ꢁ 10 M .
Because k is known by direct measurement for 2b under
s
6
–1
these conditions ((5.8 ꢂ 0.4) ꢁ 10 s ) this provides a value of
–
9
–1 –1
in the presence and absence of Br . At pH 1, reactions were
(1.7 ꢂ 0.2) ꢁ 10 M
s
for kBr, about four-fold smaller than the
[4]
followed to completion and peak area versus time data were fit
to the standard first-order rate equation. Under less acidic
conditions initial reaction rates were monitored for the first
known value of kaz for 2b.
ꢃ
ꢃ
kobs ¼ krear þ k ðk =k Þ½Br ꢄ=ð1 þ ðk =k Þ½Br ꢄÞ
t
Br
s
Br
s
(1)
~
5% of the reaction, and rate constants were obtained by the
initial rates method described above. When the reaction was
followed to completion, rate constants were also obtained from
fitting the peak area data for 7a or 7b and 8a or 8b to the first-
order rate equation. These rate constants, provided in Table S1
are in good agreement with those obtained from the quinols
If this interpretation is correct k /ks obtained from this
Br
–
experiment should be in agreement with k /k obtained by Br
Br
s
trapping of 2b formed during the uncatalyzed decomposition
of 4b (Scheme 1). Figure 2 shows that the yields of the quinol
–
1a or 1b. The rate constants obtained by the initial rates method
1b and the trapping product 7b vary with [Br ] as expected in
and the fit to the first-order rate equation were compared for 1b
at pH 1.0. Results obtained from the two methods are in good
agreement (Table 1). Because kobs for 1b is larger under all
conditions, more detailed kinetics results were obtained for
that compound.
an experiment performed at pH 4.6 in NaOAc/AcOH buffer.
Kinetic data confirm that the rate constant for decomposition
–
4 –1
of 4b under these conditions (8.0 ꢁ 10 s ) is independent of
–
–
[Br ], showing that Br traps an intermediate formed after the
rate limiting step of the reaction. Fitting the product yield data
[
33]
The rate constant, kobs, is pH dependent in the presence or
to the standard ‘azide clock’ equations
provides a value of
–
2
–1
absence of Br for both 1a and 1b. Figure S3 shows that log k
k /k of (2.6 ꢂ 0.2) ꢁ 10 M , in excellent agreement with the
obs
Br
s
–
is linearly dependent on pH with a slope of ꢃ1.0 at [Br ] = 0.0 M
selectivity ratio obtained from the kinetics experiment with 1b.
(
1b) and 0.1 M (1b) or 0.4 M (1a), confirming that both the die-
none–phenol rearrangement and the bromide trapping are sub-
ject to specific acid catalysis. Figure 1 shows that at constant pH,
–
kobs exhibits saturation behavior at high [Br ]. This result is con-
–
sistent with Br trapping of a steady state intermediate
generated in competition with the pathway leading to the
rearrangement product (Scheme 2). The data in Fig. 1 were fit
to Eqn (1) to obtain k /k where k is the composite rate
rear
Br
s
a
Table 1. Rate constants for decomposition of 1a and 1b
–
–1
Substrate
pH
[Br ], M
kobs, s
–
–
6
6
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
a
a
a
1.0
1.0
2.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.5
2.0
2.5
1.0
1.5
2.0
2.5
0
(1.47 ꢂ 0.02) ꢁ 10
(2.22 ꢂ 0.02) ꢁ 10
(2.30 ꢂ 0.38) ꢁ 10
(5.93 ꢂ 0.14) ꢁ 10
(1.37 ꢂ 0.02) ꢁ 10
(1.63 ꢂ 0.02) ꢁ 10
(1.90 ꢂ 0.02) ꢁ 10
(2.15 ꢂ 0.03) ꢁ 10
(2.24 ꢂ 0.06) ꢁ 10
(2.24 ꢂ 0.05) ꢁ 10
(2.34 ꢂ 0.02) ꢁ 10
(5.33 ꢂ 0.13) ꢁ 10
(1.32 ꢂ 0.14) ꢁ 10
(5.07 ꢂ 0.22) ꢁ 10
(1.02 ꢂ 0.23) ꢁ 10
(2.38 ꢂ 0.08) ꢁ 10
(1.47 ꢂ 0.21) ꢁ 10
(2.58 ꢂ 0.52) ꢁ 10
(6.65 ꢂ 1.27) ꢁ 10
0.40
0.40
0
b
–7
–
Figure 1. Plot of kobs versus [Br ] for 1b at pH 1.0. Data were fit to Eqn (1)
–
–
–
–
–
–
–
–
6
5
5
5
5
5
5
5
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
0.0025
0.0050
0.010
0.020
0.040
0.060
0.10
0
0
0
0
0.10
0.10
0.10
0.10
b
b
b
b
b
b
b
b
–6
–6
–7
–7
–5
–5
–6
–7
a
ꢀ
Conditions: 5 vol% CH
3
CN–H
2
O (m = 0.5 (NaClO
4
), 30 C.
–
b
Figure 2. Br trapping data for the decomposition of 4b at pH 4.6. Product
Determined by the initial rates method.
yield data for 1b and 7b were fit to standard ‘azide clock’ equations
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

M. CHAKRABORTY ET AL.
This confirms that the same intermediate is trapped in both
experiments, and that the oxenium ion 2b is formed reversibly
from 1b under acidic pH conditions.
These fitting parameters can be used to define the rate
constants k and k if the ionization constants of Scheme 2 are
r
o
known. Less extensive product studies were performed for 1a,
but HPLC data taken during the decomposition of 1a at pH 1.0
Product yield data for 7b and 8b taken at the completion of
the kinetics experiments for 1b are summarized in Fig. 3. These
are the only products detected from the decomposition of 1b,
–
in the presence of 0.4 M Br confirmed that ~50% of the product
mixture is the bromide-adduct, 7a. The only other detectable
product is the rearranged product, 8a.
–
and together they account for 90%–100% of 1b at all Br concen-
trations. Qualitatively the data are in agreement with the mech-
We have previously shown that 1c undergoes acid catalyzed
–
ꢀ
–3 –1 –1 [6]
anism of Scheme 2. In the absence of Br only the rearranged
decomposition at 80 C with a second-order rate constant of
–
ꢀ
product 8b is observed, but as [Br ] increases, the yield of 7b
(2.5 ꢂ 0.4) ꢁ 10
rate constant for decomposition of 1c, determined by UV
M s . At 30 C and pH 1.0 the first-order
increases at the expense of 8b until the yield of both products
–
–6 –1
2
reaches a constant value as the dependence of kobs on [Br ] for
methods, is (2.5 ꢂ 0.2) ꢁ 10 s , ~10 -fold smaller than at
ꢀ
–4 –1 –
decomposition of 1b reaches saturation as shown in Fig. 1.
The product yield data for 7b and 8b were fit to Eqns (2) and
80 C at pH 1.0 (2.5 ꢁ 10 s ). In the presence of 0.4 M Br , the
observed first-order rate constant for decomposition of 1c is
–6 –1
(
3), respectively, where [7b]max is the maximum product concen-
(2.1 ꢂ 0.2) ꢁ 10 s . There may be a small specific salt
–
tration of 7b at high [Br ], [8b]max is the maximum concentration
of 8b at [Br ] = 0 M, K7b and K8b are constants that can be related
to the equilibrium and rate constants of Scheme 2, and 290 M
is the measured value of kBr/k taken from the experiment shown
effect, but there is no evidence for rate acceleration in the
–
–
presence of Br . HPLC chromatograms of reaction mixtures of
–
1
–
1c at pH 1.0 in the presence and absence of Br show no
s
evidence for the formation of a bromide-adduct. The only
detected product of the acid catalyzed decomposition of 1c is
2(3 H)-benzothiazolone, 10c (Scheme 3).
in Fig. 1. All constants obtained from fitting the data of Figs 1
and 3 to Eqns (1)–(3) are gathered in Table 2.
ꢀ
ꢁ
ꢃ
ꢃ
½
7bꢄ ¼ ½7bꢄ =ð1 þ K7bÞ ½290½Br ꢄ=ð1 þ 290½Br ꢄÞꢄ
(2)
max
ꢃ
ꢃ
DISCUSSION
=
½K7b þ 290½Br ꢄ=ð1 þ 290½Br ꢄÞꢄ
If it is assumed that the oxenium ion 2 in Scheme 2 is a steady-
ꢀ
ꢁ
ꢃ
ꢃ
+
+
0
½
8bꢄ ¼ ½8bꢄ =K8b =½K8b þ 290½Br ꢄ=ð1 þ 290½Br ꢄÞꢄ (3)
max
state intermediate, and that 1, 1 H , and 1H are in equilibrium
with each other, the steady-state rate expression for the
mechanism of Scheme 2 is:
ꢂꢀ
ꢁ ꢀ
ꢁꢃ
0
0
kobsd ¼ kr aHþ=K a = 1 þ aHþ=Ka þ aHþ=K a
ꢃ
ꢃ
þ½koðkBr=ksÞ½Br ꢄ=ð1 þ ðkBr=ksÞ½Br ꢄÞꢄ
(4)
ꢂ
ꢀ
ꢁꢃ
0
ðaHþ=KaÞ= 1 þ aHþ=Ka þ aHþ=K a
At constant aH+, Eqn (4) has the same form as Eqn (1) that was
used to fit the data in Fig. 1. The composite rate constant krear of
Eqn (1) is defined by the first term of Eqn (4), and k of Eqn (1) is
t
defined by comparison of the second terms of Eqns (1) and (4).
Because no spectroscopic evidence for protonation of 1a was
observed at acidities up to H
served in a plot of log kobs versus H
arrangement of 1a up to H
= ꢃ2, both pK
o
= ꢃ2, and no curvature was ob-
for the dienone–phenol re-
s are well below
o
o
a
Figure 3. Dependence of the yields of 7b and 8b obtained during the
decomposition of 1b as a function of [Br ]. Data were fit to Eqns (2)
–
and (3) described in the text
Table 2. Derived constants for fits to Eqns (1)–(3) at pH 1.0
a
Derived constant
Equation
Value
–
6
–1
krear
kt
(1)
(1)
(1)
(2)
(2)
(3)
(3)
(6.0 ꢂ 0.3) ꢁ 10
s
s
–
5
–1
–1
(1.78 ꢂ 0.04) ꢁ 10
2
k /k
(2.9 ꢂ 0.2) ꢁ 10 M
Br
s
–
5
[7b]max
(3.4 ꢂ 0.4) ꢁ 10
0.48 ꢂ 0.15
M
M
K
7b
–
5
[8b]max
(4.31 ꢂ 0.04) ꢁ 10
0.32 ꢂ 0.01
K
8b
a
Derived constants obtained from non-linear least-squares
fits to the appropriate equation.
Scheme 3. Protonation of N in 1c leads to a unique hydrolysis
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)

OXENIUM ION CHEMISTRY
[
29]
ꢃ2. Because the quinols are quite reactive under strong acid
composite rate constants in Eqns (1)–(6). The magnitudes of kr
0
0
conditions, it is impractical to measure pK or pK by titrimetric
and k depend critically on the estimates of pK and pK for
a
a
o
a
a
[
29]
or kinetic methods, but it is possible to estimate the pK
a
s from
at 50%
ionization) for a series of four protonated 4-substituted-4-
1b, so they could vary by ꢂ two orders of magnitude in addition
to the error limits estimated in Table 3 from the quality of the fit
to Eqn (4). The estimates show that within these limitations
[30,34–38]
0
a
The pK (more correctly H
known correlations.
o
[
30]
methyl-2,5-cyclohexadienones
the Swain–Lupton F parameters for both substituents. A short
extrapolation to the 4-p-tolyl and 4-hydroxyl substituents of
bH provides a pK
a rapid hydroxide-dependent decomposition, but spectrophoto-
correlate well with the sum of
the magnitudes of k
r o
and k are chemically reasonable. The ratio
[
34]
k
r
/k was obtained from three independent fits of kinetic or
o
product data to Eqns (4)–(6). The individual ratios are in good
0
+
0
ꢀ
2
1
a
of ꢃ6.6. At pH > 11 at 30 C, 1b undergoes
agreement with each other (ranging from 1.6 ꢁ 10 to
2
2.4 ꢁ 10 ), indicating the internal self-consistency of the rate
metric titration of 1b in NaOH solution can be performed by
and product data. The results show that, within the limitations
extrapolation of UV absorbance back to t = 0. A pK
a
of 13.6,
was obtained for 1b
under these conditions. If the same substituent effect applies
of the pK
a
estimates, the rate constant for dienone–phenol
[
35]
0
+
1
.9 pK units lower than that of MeOH,
rearrangement of 1bH , k , is larger than that for the loss of
H O from 1bH , k . The estimated k for 1bH is ~3 orders of
2 o o
magnitude larger than the uncatalyzed rate constant for
ionization of the acetic acid ester 4b to generate the same
oxenium ion. The magnitudes of k
to justify the assumption made in the derivation of Eqns (4)–(6)
that 1b, 1bH , and 1bH , are in equilibrium with each other at
pH 1. The further assumption that 2b is a steady state intermedi-
r
+
+
[
36]
+
to the pK
ꢃ3.9, compared with ꢃ1.98 for MeOH
estimated that the error limits for similarly extrapolated pK
are ꢂ 2,
a
of the protonated alcohols,
the pK
We have previously
a
of 1bH is
+
[37]
2
.
[
3]
a
s
r o
and k are small enough
[
38]
so the error limits on rate constants derived from
+
0
+
these numbers will be relatively large. Nonetheless, the errors
are small enough that the validity of the assumptions used to
analyze the data according to Scheme 2 can be tested.
ate is also justified by the magnitude of k
o
. The detection of 2b
–
0
Expected product yields for 7 and 8 can be calculated from
the mechanism of Scheme 2. Equations (5) and (6), which are
derived from Scheme 2, are of the same form as Eqns (2) and
by Br trapping is possible because k /K and k /K are of
r
a
o
a
–
–
comparable magnitude and k [Br ] > > k at [Br ] ≥ 30 mM.
Br
s
Although both 1a and 1b exhibit evidence for reversible
generation of an oxenium ion under acidic conditions, the
benzothiazole quinol, 1c, does not. The cation 2c exhibits an
(
8
3), which were used to fit the product yield data for 7b and
b in Fig. 3. These equations provide additional ways to estimate
the rate constant ratio k /k of Scheme 2 from the product yield
–
1
azide/solvent selectivity of 310 M , intermediate between that
r
o
0
–1
3
–1
data and the estimates of pK and pK .
of 2a (77 M ) and 2b (1.0 ꢁ 10 M ), so if it was generated
a
a
under the reaction conditions it should have been detected by
h
ꢄ
ꢅ
0
–
[6]
ꢃ
ꢃ
Br trapping. Although the benzothiazol-2-yl group has a
strong resonance electron donating effect that stabilizes the
cation 2c, in other situations in which its inductive effect
dominates it behaves as an electron withdrawing group. The
½
7ꢄ ¼ ½1ꢄ ½ðkBr=ksÞ½Br ꢄ=ð1 þ ðkBr=ksÞ½Br ꢄÞꢄ= Kakr= koK a
(5)
(6)
o
i
ꢃ
ꢃ
þðkBr=ksÞ½Br ꢄ=ð1 þ ðkBr=ksÞ½Br ꢄÞ
h
ꢄ
ꢅi h
ꢄ
ꢅ
sm and s substituent constants for benzothiazol-2-yl are 0.27
0
0
p
[
39,40]
½
8ꢄ ¼ ½1ꢄo Kakr= koK a = Kakr= koK a
and 0.29, respectively,
and the Swain–Lupton F parameter
i
ꢃ
ꢃ
for this substituent is 0.27 compared with 0.12 for phenyl and
þðkBr=ksÞ½Br ꢄ=ð1 þ ðkBr=ksÞ½Br ꢄÞ
[
34]
p-tolyl.
Ionization of 4c to generate 2c occurs 24-fold more
slowly than ionization of 4a to generate 2a and 520-fold more
slowly than ionization of 4b to generate 2b at 80 C. Similar
retardation of the rate of ionization of 1cH via k of Scheme 2
o
is expected. Equilibrium protonation of 1c to form 1cH will also
be disfavored by the inductive electron withdrawing effect of
the benzothiazol-2-yl substituent. The same will be true for
ꢀ
[6]
The magnitudes of the estimated pK s and rate constant for
a
+
1b for Scheme 2 are shown in Table 3. The rate constants are
+
determined from the composite constants shown in Table 2 that
are obtained from the fits to the kinetic and product data in
Figs 1 and 3, the estimated pK s, and the definitions of the
a
+
0
protonation of C=O to form 1cH , but that inductive effect will
be somewhat attenuated in comparison with the effect on
protonation of the OH group. The significant electron withdraw-
ing inductive effect of benzothiazol-2-yl should significantly
reduce its migratory aptitude compared with phenyl or p-tolyl.
The inductive effect of benzothiazol-2-yl is expected to reduce
the rate of both the dienone–phenol rearrangement and the
reversible formation of 2c.
Table 3. Estimated constants of Scheme 2
Constant
Value
Source
–
1
6
9
2
[[16]]
k (s )
(5.8 ꢂ 0.4) ꢁ 10
(1.7 ꢂ 0.2) ꢁ 10
(1.9 ꢂ 0.4) ꢁ 10
direct measurement
Eqn (1)
Eqn (4), (5), (6)
Eqn (4)
Eqn (4)
Estimate, see text
Estimate, see text
s
–
1 –1
kBr (M s )
The acid–base chemistry and reactions of 1c are affected by
an additional protonation site for this compound. A spectropho-
a
kr/ko-
=1
2b
kr (s
)
(2.4 ꢂ 0.2) ꢁ 10
tometric pK
detected at ꢃ0.49 ꢂ 0.05 (Figure S4) is assigned to
a
–
1
b
+
0
0
k (s )
1.4 ꢂ 0.1
deprotonation of 1cH (Scheme 3). The benzothiazole-N is
significantly more basic than either O of 1c. The inductive effect
of the benzothiaol-2-yl substituent and protonation at N reduces
the equilibrium concentrations of 1cH and 1cH in the mildly
acidic pH range, and also leads to a unique acid catalyzed
decomposition of 1c (Scheme 3) generating 10c, apparently
through its less stable tautomer 9c. This reaction is not
o
pKa
ꢃ3.9 ꢂ 2.0
ꢃ6.6 ꢂ 2.0
0
pK
a
+
0
+
a
Average value of the ratio taken from three independent fits
of experimental data to Eqns (4)–(6).
Error limits are taken from the quality of the fit to Eqn (4)
b
a
and do not include estimated error limits for the two pK s.
00
inherently rapid; based on the observed pKa of ꢃ0.49, and kobs
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

M. CHAKRABORTY ET AL.
–
6 –1
at pH 1.0 of 2.5 ꢁ 10 s , the magnitude of kw of Scheme 3 is
(7.11), 126.5 (7.45), 127.6 (7.47), 129.5 (7.27), 130.2 (7.73), 135.3, 136.5,
–
5 –1
1
37.0, 151.4; high-resolution MS (ESI, negative) C13H10BrO (M–H) calcd.
8
.0 ꢁ 10 s . This reaction dominates for 1c because both the
2
60.9915, 262.9895, found 260.9915 (0.0 ppm), 262.9892 (ꢃ1.1 ppm).
dienone–phenol rearrangement and the formation of 2c are
suppressed by the inductive effect of the benzothiazol-2-yl group.
0
4
-methylbiphenyl-2,5-diol (8b)
CONCLUSIONS
2
mmol of 2-bromohydroquinone (378 mg), 2.1 mmol of p-tolylboronic
acid (285 mg) and Pd(PPh
4
2
times. The tube was kept in an oil bath at 90 C for 24 h. The reaction
mixture was diluted with 20 mL of EtOAc, followed by 10 mL of 1 M
HCl. The solution was stirred for 10 min. The organic layer was separated
3
)
4
(46.2 mg, 2 mol%) was mixed with 5 mL of
: 1 toluene : EtOH in a Schlenk tube followed by addition of 3 mL of
M aqueous Na CO . A freeze–thaw cycle under N was repeated four
The reversible generation of the oxenium ions 2a and 2b is
detected at pH 1 to 3 by the effect of Br on the kinetics of the
–
2
3
2
ꢀ
decomposition of 1a and 1b, respectively, and by the generation
–
of the Br adducts derived from the trapping experiments.
Kinetic and product data for 1b were quantitatively fit to a
reaction scheme (Scheme 2) in which formation of the oxenium
ion 2b as a steady state intermediate occurs in competition with
the dienone–phenol rearrangement of 1b. The magnitudes of
microscopic rate constants derived from the fit, and from
and washed with distilled water (2 ꢁ 15 mL), dried over Na
2 4
SO , and
subjected to rotary evaporation to obtain the crude solid. The product
was purified by column chromatography on silica gel, eluent 60/40
EtOAc/hexanes, to obtain 220 mg (55%) of 8b: m.p. 119–121 C; lit m.p.
1
ꢀ
ꢀ
[42]
–1 1
20–121 C ; IR 3519, 3264, 1502, 1437, 1365, 1192, 821, 797 cm ; H
a
estimates of the pK s of the hydroxyl and carbonyl protonated
3
NMR (500 MHz, CDCl ) d 2.43(3 H, s), 4.88(2 H, s(br)), 6.75(2 H, m), 6.87
conjugate acids of 1b, are reasonable and are consistent with
other rate constants already known for 2b and for its formation
from the acetic acid ester 4b. The reversible formation of
the oxenium ion is critically dependent on the nature of the
13
(
1 H, m), 7.30(2 H, d, J = 8 Hz), 7.35(2 H, d, J = 8 Hz); C NMR (125.8 MHz,
3
CDCl ) d 21.1, 115.5, 116.5, 116.6, 128.7, 128.8, 129.9, 133.8, 137.8,
146.4, 149.2.
4
-substituent. The 4-(benzothiazol-2-yl) substituent ,which is a
resonance electron donor that stabilizes the oxenium ion 2c, is
also a strong inductive electron withdrawing group that slows
the pathway for formation of 2c from 1c to the extent that 2c
cannot be detected under conditions in which both 2a and 2b
can be detected by the trapping experiments. A new decomposi-
tion pathway for 1c made possible by protonation of N of the
benzothiazol-2-yl substituent also complicates the detection of
the oxenium ion pathway. The pH range in which the formation
of the oxenium ions derived from 1a or 1b can be detected is
Kinetics and bromide trapping experiments with 1a and 1b
Purification of solvents and general procedures for kinetics experiments
[1–7]
have been described previously.
Kinetics experiments were carried
(5 vol% CH CN–H O,
ꢀ
out at 30 C in aqueous 0.1 M to 0.003 M HClO
4
3
2
m = 0.5 (NaClO )). Solutions were incubated in teflon capped round
4
bottom vials immersed in a water bath for 20 min prior to the addition
of 15 mL of a 0.01 M solution of 1a or 1b in CH
the reaction mixture were withdrawn periodically and analyzed by HPLC
C-8 reverse phase column, 70/30 MeOH/H O eluent, 1mL/min, monitored
3
CN. 20 mL aliquots from
(
2
by UV absorbance at 212 nm and 240 nm). HPLC peak areas for 1a or 1b
were plotted against time and fit to the first-order rate equation, or rate
constants were taken from initial slopes of HPLC peak area for 1a or 1b
restricted. At pH 4.6 and 7.1 neither of the oxenium ions 2a or
–
3
2
b can be detected even with the use of the superior trap, N that
is known to trap 2b at the diffusion controlled limit. It is clear
from these studies that formation of oxenium ions from the
corresponding quinols does not occur from chemical processes
under physiological conditions. Enzyme catalyzed processes
cannot be ruled out.
versus time divided by the initial peak area. Solutions containing NaBr
–
(
0.0025M to 0.40 M) were used for the Br trapping experiments under
–
the same conditions. The rearrangement products 8a and 8b, and the Br
adducts 7a and 7b were identified by co-injection with authentic samples.
First-order rate constants were obtained as above, and were also obtained
from a fit of the HPLC peak areas for 7a and 8a or 7b and 8b versus time to
the first-order rate equation.
EXPERIMENTAL
Similar procedures were followed to monitor the stability of 1a and 1b in
–
pH 4.6 0.02 M NaOAc/HOAc buffer, in pH 4.6 0.10 M N
pH 7.1 0.02 M Na HPO /NaH PO
3
/HN
3
buffer, and in
The quinols 1a, 1b, and 1c and the quinol ester 4b were available from
2
4
2
4
buffer containing 0 mM or 50 mM NaN
3
.
–
[1,3,6]
previous studies, as were the N
3
adducts 5a, 5b, and 6c.
A sample
–
[41]
of the Br adduct 7a, synthesized via a published procedure, was also
[
2]
available from a previous study. The dienone–phenol rearrangement
product 8a is commercially available, as is 10c. Other products were
synthesized or isolated as described below.
Kinetics and trapping experiments with 4b
ꢀ
Kinetic experiments were carried out at 30 C in pH 4.6 0.02 M NaOAc/
HOAc buffer (5 vol% CH
.1 M NaBr. The solution was incubated in a cuvette in the thermostated
cell holder of a UV–Vis spectrophotometer for ~20 min before the kinetic
run was initiated by injecting 15 mL of a 0.01 M solution of 4b in CH CN
3 2 4
CN–H O, m = 0.5 (NaClO )) containing 0 M or
0
0
3
-bromo-4 -methylbiphenyl-4-ol (7b)
0
2
CHCl
.5 mmol of 4 -methylbiphenyl-4-ol (460 mg) was dissolved in 4 mL of
3
ꢀ
into 3 mL of the buffer. Data were collected at 254 nm, 245 nm, 225 nm,
3
in a round bottom flask, which was kept in a 60 C water bath. A
(130 mL) in 144 mL of CHCl was added in a
dropwise fashion to the solution of 4 -methylbiphenyl-4-ol. The mixture
–
solution of 2.5 mmol of Br
2
3
and 211 nm in the absence of Br and at 273 nm, 261 nm and 234 nm
–
0
in the presence of Br . Absorbance versus time data were fit to a first-
ꢀ
was kept at 60 C for 72 h. After cooling, the brownish yellow solution
SO , and
order rate equation at all wavelengths.
was washed with distilled water (3 ꢁ 5 mL), dried over Na
2
4
For product studies, pH 4.6 NaOAc/HOAc buffers containing 1.00 mM
to 0.10 M NaBr were utilized under conditions identical to those used
for the kinetics experiments. A control experiment in the absence of NaBr
subjected to rotary evaporation to obtain a crude product. The product
was subjected to column chromatography on silica gel, eluent 20/80
ꢀ
ꢀ
EtOAc/hexanes, to obtain 178 mg (27%) of 7b: m.p. 115–117 C; IR
was also performed. Reaction mixtures were incubated for 3 h at 30 C
–
1 1
3
(
7
(
304, 3028, 1494, 1424, 1394, 1277, 1226, 1038, 805, 699 cm
500 MHz, CDCl ) d 2.43(3 H, s), 5.61(1 H, s), 7.11(1 H, d, J = 8.5 Hz),
.27(2 H, d, J = 8 Hz), 7.45(2 H, d, J = 8 Hz), 7.47(1 H, dd, J = 8.5, 2 Hz), 7.73
;
H NMR
prior to HPLC analysis (C-8 reverse phase column, conditions identical
to those described above) of each solution in duplicate. HPLC peak areas
3
–
for 1b and 7b were plotted against Br and fit to the ‘azide clock’
1
3
[33]
1 H, d, J = 2 Hz); C NMR (125.8 MHz, CDCl
3
) d 21.0 ( 2.43), 110.5, 116.2
s
equations to obtain kBr/k .
wileyonlinelibrary.com/journal/poc
Copyright © 2012 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. (2012)

OXENIUM ION CHEMISTRY
00
+
Kinetics of decomposition of 1c, pK determination for 1cH , and
a
REFERENCES
isolation and characterization of 10c
[
[
1] M. Novak, S. A. Glover, J. Am. Chem. Soc. 2004, 126, 7748.
2] M. Novak, S. A. Glover, J. Am. Chem. Soc. 2005, 127, 8090.
[3] M. Novak, M. J. Poturalski, W. L. Johnson, M. P. Jones, Y. Wang, S. A.
Glover, J. Org. Chem. 2006, 71, 3778.
ꢀ
Kinetic experiments were carried out at 30 C in 3 mL of 0.1 M HClO
4
(
5 vol% CH
in a thermostated cell holder of a UV–Vis spectrophotometer prior to a
5 mL injection of a 0.01 M solution of 1c in CH CN. Absorbance versus
time data were monitored at 230, 212, and 207 nm. Data were plotted
and fit to the first-order rate equation. Similar experiments in 0.1 M HClO
3
CN–H
2
O, m = 0.5 (NaClO
4
)) incubated for 15 min in a cuvette
1
3
[4] Y.-T. Wang, J. Wang, M. S. Platz, M. Novak, J. Am. Chem. Soc. 2007,
1
29, 14566.
[
5] Y.-T. Wang, K. J. Jin, S. H. Leopold, J. Wang, H.-L. Peng, M. S. Platz, J.
4
Xue, D. L. Phillips, S. A. Glover, M. Novak, J. Am. Chem. Soc. 2008,
containing 0.40 M NaBr were carried out with absorbance versus time
data monitored at 300, 260, and 230 nm. Data were plotted and fit to
the first-order rate equation. The stability of 1c was monitored by HPLC
methods at pH 4.6 and 7.1 as described above for 1a and 1b.
1
30, 16021.
6] Y.-T. Wang, K. J. Jin, L. R. Myers, S. A. Glover, M. Novak, J. Org. Chem.
009, 74, 4463.
7] M. Novak, M. J. Kahley, E. Eiger, J. S. Helmick, H. E. Peters, J. Am.
Chem. Soc. 1993, 115, 9453.
[8] P. A. Davidse, M. J. Kahley, R. A. McClelland, M. Novak, J. Am. Chem.
Soc. 1994, 116, 4513.
9] R. A. McClelland, P. A. Davidse, G. Hadzialic, J. Am. Chem. Soc. 1995,
[
[
2
For the pK
a
measurement, solutions of appropriate pH or H
o
(ranging
ꢀ
from ꢃ2.3 to 6.14) were incubated at 30 C in a cuvette in the
thermostated cell holder of a UV–Vis spectrophotometer for ~15 min.
[
prior to injection of a solution of 1c in CH
80 nm was plotted against pH/H and fit to a standard titration equation
to obtain the pK . For HClO concentrations > 0.10 M. the H acidity
function values were used instead of pH.
A volumetric flask containing 250 mL of 0.1 M HClO
3
CN. Initial absorbance at
1
17, 4173.
2
o
[
[
[
10] D. Ren, R. A. McClelland, Can. J. Chem. 1998, 76, 78.
11] R. A. McClelland, T. A. Gadosy, D. Ren, Can. J. Chem. 1998, 76, 1327
12] M. Novak, S. Kazerani, J. Am. Chem. Soc. 2000, 122, 3606.
[13] M. Novak, S. Rajagopal, Adv. Phys. Org. Chem. 2001, 36, 167.
[14] M. Novak, K. Toth, S. Rajagopal, M. Brooks, L. L. Hott, M. Moslener,
J. Am. Chem. Soc. 2002, 124, 7972.
[15] M. Novak, S. Rajagopal, L. Xu, S. Kazerani, K. Toth, M. Brooks, T.-M.
Nguyen, J. Phys. Org. Chem. 2004, 17, 615.
a
4
o
[
43]
4
(5 vol% CH
3
CN–H
2
O,
ꢀ
4
m = 0.5 (NaClO )) was incubated at 30 C in a water bath, while a
9
2
mM solution of 4c in CH
4 h intervals for 5 days. The water bath was raised to 40 C before
3
CN was added to the flask in 1 mL portions at
ꢀ
adding 1.25 mL portions of the solution of 4c at 12 h intervals for
[
[
[
16] M. Chakraborty, K. J. Jin, S. C. Brewer, H.-L. Peng, M. S. Platz, M.
Novak, Org. Lett. 2009, 11, 4862.
17] M. Chakraborty, K. J. Jin, S. A. Glover, M. Novak, J. Org. Chem. 2010,
2
days (10mL of the solution of 4c added over the entire experiment).
ꢀ
After the last addition the reaction mixture was kept at 40 C for
three days. The reaction mixture was cooled to room temperature
7
5, 5296.
before neutralization with Na
2
HPO
4
(3.5g). The solution was extracted
SO and rotary
18] G. Wells, J. M. Berry, T. D. Bradshaw, A. M. Burger, B. Wang, A. D.
Westwell, M. F. G. Stevens, J. Med. Chem. 2003, 46, 532.
[19] J. M. Berry, T. D. Bradshaw, I. Fichtner, R. Ren, C. H. Schwalbe, G.
2
ꢁ 100mL) with CH
2
Cl . The extract was dried over Na
2
2
4
evaporated to obtain a solid yellow crude product that was purified by
radial chromatography on silica gel (20/80 EtOAc/hexanes) to obtain
Wells, E.-H. Chew, M. F. G. Stevens, A. D. Westwell, J. Med. Chem.
2
005, 48, 639.
10c. The compound was characterized by NMR and HPLC comparison to
[
20] T. D. Bradshaw, C. S. Matthews, J. Cookson, E.-H. Chew, M. Shah, K.
Bailey, A. Monks, E. Harris, A. D. Westwell, G. Wells, C. A. Laughton,
M. F. G. Stevens, Cancer Res. 2005, 65, 3911.
an authentic commercial sample.
[
21] C. J. Lion, C. S. Matthews, G. Wells, T. D. Bradshaw, M. F. G. Stevens,
A. D. Westwell, Bioorg. Med. Chem. Lett. 2006, 16, 5005.
22] A. Mukherjee, K. Huber, H. Evans, N. Lakhani, S. Martin,
Br. J. Pharmacol. 2007, 151, 1167.
a
pK determination for 1b
[
The pK
a
measurement of 1b was performed using solutions of NaOH in
for
NaOH] ≤ 0.5 M). Solutions were incubated at 30 C in a cuvette in the
thermostated cell holder of a UV–Vis spectrophotometer for ~15 min
prior to injection of a solution (0.01 M) of 1b in CH CN. Absorbance at
20 nm was monitored for ~5 min and extrapolated back to zero time
to obtain the initial absorbance. Initial absorbance at 220 nm was plotted
against pH to obtain the pK of 1b.
the pH range 11–14 (at constant ionic strength of 0.5 using NaClO
[
4
[23] D. P. W. Wong, G. Wells, T. Hagen, Eur. J. Pharmacol. 2010, 643, 188.
[24] J. König, S. Wyllie, G. Wells, M. F. Stevens, P. G. Wyatt, A. H. Fairlamb,
J. Biol. Chem. 2011, 286, 8523.
ꢀ
[
25] A. Capes, S. Patterson, S. Wyllie, I. Hallyburton, I. T. Collie, A. J.
McCarroll, M. F. G. Stevens, J. A. Frearson, P. G. Wyatt, A. H. Fairlamb,
I. H. Gilbert, Bioorg. Med. Chem. 2012, 20, 1607.
3
2
[
26] S. Barradas, M. C. Carreño, M. González-López, A. Latorre, A. Urbano,
Org. Lett. 2007, 9, 5019.
a
[
[
[
[
[
[
[
[
27] I. Kim, K. Kim, J. Choi, J. Org. Chem. 2009, 74, 8492.
28] M. C. Redondo, M. Ribagorda, M. C. Carreño, Org. Lett. 2010, 12, 568.
29] J. H. Seong, Y.-S. Paik, J. Korean Chem. Soc. 1994, 38, 924.
30] V. P. Vitullo, E. A. Logue, J. Org. Chem. 1972, 37, 3339.
31] V. P. Vitullo, N. Grossman, J. Am. Chem. Soc. 1972, 94, 3844.
32] K. L. Cook, A. J. Waring, J. Chem. Soc. Perkin Trans. 2 1973, 88.
33] J. P. Richard, W. P. Jencks, J. Am. Chem. Soc. 1984, 106, 1383.
34] C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1991, 91, 165.
SUPPORTING INFORMATION
Figures S1 and S2 showing the stability of 1b at pH 7.1 and 4.6,
respectively, Figure S3 showing the pH dependence of the
decomposition of the quinols 1a and 1b, Figure S4 showing
the spectrophotometric titration of 1c, and Table S1 showing
the rate constants determined by high-performance liquid
chromatography methods for formation of 7b and 8b generated
from 1b at pH 1.0.
[35] P. Ballinger, F. A. Long, J. Am. Chem. Soc. 1960, 82, 795.
[36] L. H. Funderburk, L. Aldwin, W. P. Jencks, J. Am. Chem. Soc.1978,
1
00, 5444.
[
[
[
[
37] P. Bonvicini, A. Levi, V. Lucchini, G. Modena, G. Scorrano, J. Am.
Chem. Soc.1973, 95, 5960.
38] M. Novak, G. A. Bonham, J. J. Mulero, M. Pelecanou, J. N. Zemis, J. M.
Buccigross, T. C. Wilson, J. Am. Chem. Soc. 1989, 111, 4447.
39] V. F. Bystrov, Zh. N. Belaya, B. E. Gruz, G. P. Syrova, A. I. Tolmachev,
L. M. Shulezhko, L. M. Yagupol’skii, Z. Obshchei Khimii 1968, 38, 1001.
40] J. Durmis, M. Karvas, Z. Manasek, Z. Collect. Czech. Chem. Commun.
Acknowledgements
1
973, 38, 215.
[41] C. D. Gutsche, K. H. No, J. Org. Chem. 1982, 47, 2708.
[42] S. Fujita, K. Sano, J. Org. Chem. 1979, 44, 2647.
[43] R. A. Cox, S.-O. Lam, R. A. McClelland, T. T. Tidwell, J. Chem. Soc.,
Perkin Trans. 2 1979, 272.
We thank the NIH/NIGMS (Grant # R15 GM088751-01) for support
of this work. High resolution mass spectra were obtained at The
Ohio State University’s Campus Chemical Instrument Center.
J. Phys. Org. Chem. (2012)
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc