Rapid formation and slow collapse of a carbocation - Anion pair to a neutral molecule

Article abstract of DOI:10.1021/jo702492r

(Chemical Equation Presented) The 4,4′,4″-trimethoxytrityl cation (TMT⁺) was observed to react with acetate ion in acetic acid reversibly to give the corresponding ester (TMT-OAc). The rate of the reaction was found to be independent of [NaOAc]

Full text of DOI:10.1021/jo702492r

Rapid Formation and Slow Collapse of a Carbocation-Anion Pair
to a Neutral Molecule
Weifang Hao and Vernon D. Parker*
Utah State UniVersity, Department of Chemistry and Biochemistry, Logan, Utah 84322
Vparker@cc.usu.edu
ReceiVed NoVember 20, 2007
The 4,4′,4′′-trimethoxytrityl cation (TMT+) was observed to react with acetate ion in acetic acid reversibly
to give the corresponding ester (TMT-OAc). The rate of the reaction was found to be independent of
[NaOAc] over a 25000-fold range. Similar results were observed in the presence of Bu4N+ in acetic acid
as well as in HOAc/AN (1/1). It was concluded that {TMT+ (HOAc/AcO-)} is an ion pair that forms
essentially completely from free TMT+ and HOAc/AcO- during the time of mixing under stopped-flow
conditions. The process which was studied kinetically is the intramolecular collapse of the ion pair to
TMT-OAc which takes place in two steps involving a kinetically significant intermediate. The remarkably
close resemblance of this reaction to the Winstein scheme for solvolysis reactions is noted. In analogy
to the Winstein scheme, it was proposed that the intermediate could be an intimate ion pair formed upon
extrusion of solvent from the solvent separated ion pair. The product-forming step could then correspond
to the intimate ion pair reacting further to form a covalent bond between the two moieties within the
complex. The values of the thermodynamic and the activation parameters as well as the apparent rate
constants for the reaction in the presence of either sodium or tetrabutylammonium ions suggest that
these counterions play insignificant roles in the reactions. However, the equilibrium constant for the
intramolecular step (K4) was observed to be two times greater in the presence of Bu4N+ than in the
presence of Na+. The rate of the reaction in HOAc was observed to be about four times as great as that
in HOAc/AN (1/1).
Introduction
reactions take place in a single rate-determining step ac-
companied by the formation of a covalent bond between the
two moieties. The main recent interest in these reactions has
been on developing reactivity scales for both the cations and
the nucleophiles.19,20
The study of the reactions of carbocations with nucleophiles
has been a topic of intense interest for more than half a
century.1-18 There appears to be general agreement that the
(1) Turgeon, V. C.; LaMer, V. K. J. Am. Chem. Soc. 1952, 74, 5988.
(2) Duynstee, E. F. J.; Grunwald, E. J. Am. Chem. Soc. 1959, 81, 4342.
(3) Diffenback, R. A.; Sano, K.; Taft, R. W. J. Am. Chem. Soc. 1966,
88, 4747.
(4) Ritchie, C. D.; Skinner, G. A.; Badding, V. G. J. Am. Chem. Soc.
1967, 89, 2063.
We regard the cation-anion combination reaction as one of
the simplest covalent bond forming reactions possible. The
reactions result in the formation of a single bond without
disrupting any other covalent bonds. This suggests these
(5) Hill, E. A.; Mueller, W. J. Tetrahedron Lett. 1968, 2565.
(6) Noyce, D. S.; Brauman, S. K. J. Am. Chem. Soc. 1968, 90, 5218.
(7) Brauman, J. I.; Archie, W. C., Jr. J. Am. Chem. Soc. 1970, 92, 5981.
(8) Ritchie, C. D.; Virtanen, P. O. I. J. Am. Chem. Soc. 1972, 94, 4963,
4966.
(9) Bunton, C. A.; Huang, K. J. Am. Chem. Soc. 1972, 94, 3536.
(10) Ritchie, C. D. Acc. Chem. Res. 1972, 5, 348.
(11) Ritchie, C. D.; Virtanen, P. O. I. J. Am. Chem. Soc. 1973, 95, 1882.
(12) Bunton, C. A.; Haung, S. K. J. Am. Chem. Soc. 1974, 96, 515.
10.1021/jo702492r CCC: $40.75 © 2008 American Chemical Society
Published on Web 12/08/2007
48
J. Org. Chem. 2008, 73, 48-55

Collapse of a Carbocation-Anion Pair to a Neutral Molecule
SCHEME 1
and is the simplest complex mechanism involving consecutive
reactions.
The IRC-time profile for the pre-association mechanism (a)
in Figure 1 intercepts the 0 time axis at 2.00, which corresponds
to kf and approaches the steady-state value (0.94 s-1). That for
the single-step mechanism (b) (k1 ) 1.00 s-1) is a straight line
with zero slope. The ease of distinguishing between the two
mechanisms under the conditions of the calculated data is
apparent from Figure 1.
The objective of this study was to determine the mechanism
of a cation-anion combination reaction. We selected the 4,4′,4′′-
trimethoxytrityl cation (TMT+), a carbocation of moderate
reactivity toward nucleophiles,12 as the cationic reactant and
acetate ion as the anionic reactant. In order to moderate the
rate of the cation-anion combination reaction, we chose acetic
acid as the solvent for this study. It is well-known that acetate
ion exists in acetic acid as the hydrogen-bonded complex,
hydrogen biacetate ion (HOAc/AcO-),35 and this is expected
to markedly affect the nucleophilicity of the ion. Choosing
conditions under which the rates of the anion-cation combina-
tions are moderate is of special importance in attempting to
determine whether the reactions follow a single step or more
complex reaction mechanisms. This is because of the fact that
the initial portion of the reaction, i.e., the pre-steady-state time
period, must be accessed in order to differentiate between the
mechanisms. Since cation-anion combination reactions are
inherently rapid reactions, most of the previous kinetic studies
have out of necessity been carried out under conditions where
the initial time periods could not be accessed.
FIGURE 1. IRC plots for pre-association (a) and single-step (b)
mechanisms.
reactions as an obvious target for our ongoing research,21-32
the results of which show that many fundamental organic
reactions previously believed to take place in a single step
actually take place by a mechanism with more than one
transition state.
The ability to distinguish between single-step and more
complex reaction mechanisms has been enhanced by the
development of instantaneous rate constant (IRC) analysis33
which allows the course of the reaction to be analyzed in terms
of the apparent instantaneous rate constant (kinst) as a function
of time. For the single step mechanism in the absence of any
complications, kinst is a true constant independent of time. On
the other hand, kinst varies with time in the pre-steady-state time
period for a more complex mechanism and approaches the
steady-state value (ks.s.) with increasing time. These relationships
are illustrated in Figure 1 for a reaction following the pre-
association mechanism (Scheme 1) in which the intermediate
reactant complex is kinetically significant. This mechanism was
chosen for the illustration since it has been widely discussed34
Results
The reaction between TMT+ and NaOAc in acetic acid to
form the corresponding ester is reversible. The approach to
equilibrium is illustrated by the spectra in Figure 2 for the
decrease in [TMT+]. The reversible reaction is illustrated in
Scheme 2. No reaction between TMT+ and acetic acid could
be detected in the absence of acetate ion.
Definitions of Rate Constants. A number of different
methods were used to obtain rate constants in this work. In order
to avoid confusion as to the meaning of the abbreviations used
to define the rate constants, the definitions are given below:
kinst ) An apparent pseudo first-order rate constant obtained
by an IRC procedure over short time intervals. These can be
time dependent in the manner illustrated in Figure 1.
(13) Ritchie, C. D. J. Am. Chem. Soc. 1975, 97, 1170.
(14) Ritchie, C. D.; Sawada, M. J. Am. Chem. Soc. 1977, 99, 3754.
(15) Ritchie, C. D.; Van Verth, J. E.; Viranen, P. O. I. J. Am. Chem.
Soc. 1982, 104, 3491.
(16) Ritchie, C. D. J. Am. Chem. Soc. 1984, 106, 7187.
(17) Ritchie, C. D. Can. J. Chem. 1986, 64, 2239.
(18) Arnett, E. M.; Flowers, R. A.; Ludwig, R. T.; Meekhof, A.; Walek,
S. Pure Appl. Chem. 1995, 67, 729.
(19) Mayr, H.; Kempf, B.; Ofial, A. R. Acc. Chem. Res. 2003, 36, 66
and references cited therein.
(20) Minegishi, S.; Mayr, H. J. Am. Chem. Soc. 2003, 125, 286.
(21) Parker, V. D.; Zhao, Y.; Lu, Y.; Zheng, G. J. Am. Chem. Soc. 1998,
120, 12720.
(22) Lu, Y.; Zhao, Y.; Parker, V. D. J. Am. Chem Soc. 2001, 123, 5900.
(23) Zhao, Y.; Lu, Y.; Parker V. D. J. Chem. Soc., Perkin Trans. 2001,
2, 1481.
(24) Parker, V. D.; Zhao, Y. J. Phys. Org. Chem. 2001, 14, 604.
(25) Zhao, Y.; Lu, Y.; Parker, V. D. J. Am. Chem. Soc. 2001, 123, 1579.
(26) Lu, Y.; Zhao, Y.; Handoo, K. L.; Parker, V. D. Org. Biomol. Chem.
2003, 1, 173.
(27) Handoo, K. L.; Lu, Y.; Zhao, Y.; Parker, V. D. Org. Biomol. Chem.
2003, 1, 24.
(28) Lu, Y.; Handoo, K. L.; Parker, V. D. Org. Biomol. Chem. 2003, 1,
36.
ks.s. ) An apparent pseudo first-order rate constant evaluated
in the plateau region of the kinst-time profile.
kN ) An apparent pseudo-first-order rate constant evaluated
in the conventional way over a segment of the (1 - E.R.)-
time profile where E.R. is the extent of reaction. The subscript
N signifies the segment number, and kN can be time dependent.
The Effect of Acetate Ion Concentration on the Reaction
Rate. Kinetic experiments were carried out by stopped-flow
spectrophotometry under a nitrogen atmosphere in a glove box.
Our initial experiments in glacial acetic acid gave rise to
(29) Parker, V. D.; Lu, Y. Org. Biomol. Chem. 2003, 1, 2621.
(30) Handoo, K. L.; Lu, Y.; Zhao, Y.; Parker, V. D. J. Am. Chem. Soc.
2003, 125, 9381.
(31) Parker, V. D.; Lu, Y.; Zhao, Y. J. Org. Chem. 2005, 70, 1350.
(32) Parker, V. D. Pure Appl. Chem. 2005, 77, 1823.
(33) Parker, V. D. J. Phys. Org. Chem. 2006, 19, 714.
(34) Ridd, J. H. AdV. Phys. Org. Chem. 1978, 16, 1.
(35) Clark, J. H.; Emsley, J. J. Chem. Soc., Dalton Trans. 1973, 2760.
J. Org. Chem, Vol. 73, No. 1, 2008 49

Hao and Parker
TABLE 2. Apparent Steady-State Rate Constants for the Reaction
between TMT+ and Acetate Ion as a Function of [Bu4N+ HOAc/
AcO-] in Acetic Acid at 298 Ka
[TMT-OAc]equil/[TMT+]equil
[Bu4N+ HOAc/OAc-]/mM
ks.s./s-1 b
() K4)
20.0
10.0
5.0
0.0668
0.0701
0.0710
0.0679
1.81
1.89
2.12
1.97
2.0
b
a From absorbance-time data at 480 nm. ks.s. was evaluated from 200
data points near the half-life of the reaction.
SCHEME 3. Equilibria Involved in the Reaction of TMT+
and HOAc/AcO-
FIGURE 2. Diode-array visible absorption spectra of TMT+ (0.02
mM) in acetic acid containing sodium acetate (0.01 M): start time, 0
s; cycle time, 2.5 s; total run time, 100 s.
TABLE 3. Ratios of Conventional Pseudo First-Order Rate
Constants Evaluated over Initial Data Segments Divided by the
Apparent Steady-State Rate Constant (ks.s. ) 0.065 s-1) for the
Reaction of TMT+ with HOAc/AcO- (0.5 M) in Acetic Acid at
298 K
SCHEME 2. Reversible Reaction between TMT+ and
Sodium HOAc/AcO- in Acetic Acid
segment/N
kN/ks.s.(21point)
kN/ks.s.(31point)
kN/ks.s.(51point)
1
2
3
4
5
6
7
2.64
1.12
1.13
1.05
1.09
1.07
1.09
2.02
1.15
1.09
1.07
1.09
1.07
1.08
1.51
1.10
1.09
1.07
1.06
1.05
1.05
TABLE 1. Effect of Sodium Acetate Concentration on the
Apparent Steady-State Rate Constant for the Reaction of TMT+
with Sodium Acetate in Acetic Acid at 298 K
[TMT+]/mM
[NaOAc]/mM
ks.s./s-1 a
0.20
0.20
0.20
0.20
0.20
0.010
0.010
500
250
125
10.0
2.50
0.10
0.020
0.0645
0.0586
0.0597
0.0529
0.0537
0.0593
0.0767
direct relationship between the latter and the concentration of
the product (TMT-OAc) at equilibrium. The most likely expla-
nation of zero-order kinetics in acetate ion for this reaction is
that HOAc/AcO- very rapidly interacts with TMT+ to form an
ion-pair, i.e., equilibrium (3) is saturated at all [HOAc/AcO-]
in Table 1 and 2. The ion-pair then reversibly undergoes an
intramolecular combination reaction (eq 4) resulting in the
formation of the corresponding acetate ester. Since equilib-
rium (3) is established during the time of mixing in the
stopped-flow spectrophotometer, the process that takes place
during our kinetic experiments is illustrated by equilibrium (4)
(Scheme 3).
The Results of Conventional Kinetic Studies of the
Combination Reaction in Acetic Acid. The method of data
treatment involved evaluating pseudo first-order rate constants
over consecutive intervals consisting of three different series
of segments. The 21 point procedure included the following
segments: data points 1-21(1), 21-41(2), 41-61(3), 61-81(4),
81-101(5), 101-121(6), and 121-141(7) for absorbance-time
data obtained by stopped flow spectrophotometry. The 31- and
51-point segments were constructed in the same manner as those
in the 21-point procedure. The pseudo first-order rate constants
(kN) evaluated for each time segment are finally divided by the
apparent steady-state rate constant (ks.s.) calculated from data
at times approaching the half-life of the reaction resulting in
dimensionless ratios.
a ks.s. was evaluated from 200 data points near the half-life of the reaction.
remarkable results. The rate of the reaction was observed to be
independent of sodium acetate concentration over a wide range
for the reaction of TMT+ with acetate ion in acetic acid at 298
K (Table 1). [NaOAc] was varied over the entire applicable
range, from 0.50 to 0.00002 M with only random variations in
ks.s. The data in the last entry in Table 1 where [NaOAc] and
[TMT+] were equal to 0.02 and 0.01 mM, respectively, were
subject to significant error due to the small absorbance change
for the reaction. The latter data are included to show that the
zero-order in [NaOAc] appears to continue as long as there is
enough acetate ion present for TMT+ to completely react.
The data in Table 2 show that the rate of the reaction of
TMT+ with Bu4N+ HOAc/AcO- in acetic acid is also inde-
pendent of the salt concentration. The value of ks.s. obtained at
480 nm was observed to be constant within experimental error
as [Bu4N+ HOAc/AcO-] was varied from 2 to 20 mM.
The data in Tables 1 and 2 suggest that TMT+ and HOAc/
AcO- in acetic acid react during the time of mixing. This
reaction is reversible (see Figure 1), but because of saturation
of equilibrium (3) with respect to [HOAc/AcO-] there is not a
The initial rate data in Table 3 were obtained by the
consecutive segment treatment of kinetic data for the pseudo-
50 J. Org. Chem., Vol. 73, No. 1, 2008

Collapse of a Carbocation-Anion Pair to a Neutral Molecule
TABLE 4. Apparent Pseudo-First-Order Rate Constants Ratios
Observed for the Reactions of TMT+ with NaOAc (0.50 M) in
Acetic Acid at 298 K
TABLE 5. Ratios of Conventional Pseudo-First-Order Rate
Constants Evaluated over Initial Data Segments Divided by ks.s.
(0.0155 s-1) for the Reaction of TMT+ with Acetate Ion in HOAc/
AN (1/1 vol) at 298 K
a
λ/nm
[TMT]/mM
(k1(31)/ks.s.
)
ks.s./s-1 b
K4
segment/N
kN/ks.s.(21point)
kN/ks.s.(31point)
kN/ks.s.(51point)
420
430
440
450
460
470
480
490
500
510
520
0.10
0.04
0.04
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.10
1.35
1.38
1.67
1.41
1.45
1.44
1.44
1.55
1.72
2.29
2.60
0.0646
0.0679
0.0681
0.0729
0.0725
0.0735
0.0711
0.0687
0.0730
0.0650
0.0698
0.79
0.94
0.83
0.89
0.91
0.93
0.92
0.94
0.92
0.98
0.78
1
2
3
4
5
6
7
1.31
1.24
1.21
1.17
1.04
1.03
1.01
1.26
1.18
1.11
1.04
1.01
0.98
0.99
1.21
1.12
1.01
0.98
0.99
1.00
0.99
TABLE 6. Kinetic and Thermodynamic Parameters for the
Collapse of {TMT+ HOAc/AcO-}
a Pseudo-first-order rate constant over the first 31 data points in the 2000
∆Hq
kcal mol-1 cal mol-1K-1 kcal mol-1 cal mol-1 K-1
∆Sq
∆H°
∆S°
point array divided by ks.s..
b Evaluated from 200 points near the apparent
solvent
HOAc
HOAc/AN
(1/1)
half-life of the reaction.
12.7
13.7
-21.5
-20.7
-4.3
-1.4
-14.3
-5.2
first-order reaction of TMT+ (0.0002 M) with acetate ion (0.50
M) in acetic acid at 298 K. The expected result for a single-
step mechanism is that all of the rate constant ratios (the value
difference in the results in the two solvents is that ks.s. in HOAc
is about four times greater than in the mixed solvent.
of the observed pseudo-first-order rate constant divided by ks.s.
)
in Table 3 are expected to be the same and equal to 1.00 within
experimental error; i.e., pseudo-first-order rate constants for the
single-step mechanism are expected to be time independent. This
is equivalent to the straight line with zero slope in Figure 1a.
The observed result is that the rate constants in all three sets of
data exceed ks.s. by 2.64, 2.02, and 1.51 for the initial 21-, 31-,
and 51-point segments, respectively, and the ratios in all three
columns approach unity in subsequent segments. The data rule
out the simple single-step mechanism for the intramolecular
combination reaction between TMT+ and HOAc/AcO- in acetic
acid.
When an intermediate absorbs in the same spectral region as
that where the kinetics of a reaction is monitored, the kinetic
results will usually be wavelength dependent. This is because
the kinetic analysis assumes that only reactant or product absorb
at the wavelength where measurements are made. Conducting
kinetic measurements over a range of wavelength then becomes
a very effective tool in demonstrating whether a reaction takes
place in a single step or by a more complex mechanism. The
data in Table 4 summarize the results over the wavelength region
from 420 to 520 nm for the reaction of TMT+ with NaOAc
(0.50 M) in acetic acid at 298 K. The variation in the k1(31)/
The Effect of Temperature on the Rate and Equilibrium
for the Collapse of {TMT+ HOAc/AcO-}. The kinetics of
the collapse of {TMT+ HOAc/AcO-} were studied in HOAc
and in mixed solvent HOAc/AN (1/1 vol) in the temperature
range from 293 to 323 K. The activation and thermodynamic
parameters derived from these studies based upon ks.s. data are
summarized in Table 6. The corresponding Eyring and van’t
Hoff plots for the reactions carried out in HOAc are shown in
Figure 3 and those obtained from data in mixed solvent are
illustrated in Figures S3 and S4 (Supporting Information).
Application of IRC Analysis to the Collapse of {TMT+
HOAc/AcO-}. The IRC plots shown in Figure 4 are for reaction
of TMT+ with sodium acetate in acetic acid at 298 K for decay
of absorbance due to reactant measured at 500 nm. The IRC-
time profile for the first 3% of reaction, which was obtained
from the fifth-order polynomial smoothed data is illustrated in
Figure 4a. The curve appears to intersect the zero time axis at
about 0.24 s-1 which provides an estimate of kf in Scheme 4.
The IRC-time profile begins to approach a plateau value at
about 170 ms. The latter is more clearly illustrated by the IRC-
time profile shown in Figure 4b. The latter was obtained using
the 51-point sliding procedure described in the experimental
section.
The IRC-time profiles shown in Figure 5 were constructed
from data obtained at 430 nm during the reaction of TMT+ (0.04
mM) with sodium acetate (0.5 M) in acetic acid at 298 K. The
zero-time intercept in Figure 5a is similar to that observed in
Figure 4a. The sine-wave-like behavior in the plateau portion
of the IRC-time profile in Figure 5a is often observed in these
plots and is due to spectrometer output noise which is not
completely smoothed out by the fifth-order polynomial function.
The effect of spectrometer output noise is illustrated even more
clearly in the plateau regions of Figures 4b and 5b.
k
s.s. ratio with wavelength suggests that there may be interference
from absorbance due to an intermediate since the ratio is
otherwise expected to be wavelength independent.
Estimation of the Equilibrium Constant for eq 4. The last
columns in Tables 2 and 4 give estimates of K4 (eq 4), the
equilibrium constant for the reaction of the ion pair. These were
obtained from the stopped flow experiments and were evaluated
as the difference in initial (Ainit) and infinity absorbances (Ainf)
divided by (Ainf); (Ainit - Ainf)/Ainf). It is of interest to note that
K4 in HOAc at 298 K is about two times as great in the presence
of Bu4N+ HOAc/AcO- (2.0 ( 0.1) as compared to that in the
presence of Na+ HOAc/AcO- (0.89 ( 0.09)
Effect of Cosolvent on the Kinetics of the Reactions of
TMT+ with HOAc/AcO-. Consecutive segment kinetic data
for the reaction of TMT+ (0.02 mM) with HOAc/AcO- (5.0
mM) in HOAc/AN (1/1 vol) where AN is acetonitrile are
summarized in Table 5. The results in the mixed solvent are
similar to those in acetic acid in terms of the decrease in the
rate constant ratios with the extent of reaction. The greatest
Discussion
The reaction coordinate (RC) diagram for the pre-association
mechanism is illustrated in Scheme 5. Beginning with separated
reactants on the far left, reactant and substrate diffuse and begin
to collide resulting in what is called the encounter complex,
which amounts to the two reactants in a solvent cage where it
J. Org. Chem, Vol. 73, No. 1, 2008 51

Hao and Parker
FIGURE 3. Eyring plot (a) for the collapse of the ion pair in acetic acid. Van’t Hoff-like plot (b) for equilibrium (4) in HOAc.
FIGURE 4. IRC time profiles for the reaction of TMT+ (0.02 mM) with sodium acetate (0.50 M) in acetic acid. Data for the first 3% of the
reaction (a) and that over the first-half-life (b).
SCHEME 4. Mechanism of the Combination Reaction
the time of mixing in the stopped-flow spectrophotometer and
that the process that can be studied kinetically is the intramo-
lecular collapse of {TMT+ HOAc/AcO-} to the product, TMT-
OAc (eq 4). The degree of reversibility of this process depends
strongly on the solvent composition. The data in Tables 3-5
and in Figures 4 and 5 show that the intramolecular collapse of
{TMT+ HOAc/AcO-} takes place in at least two steps and that
the data are consistent with the reversible consecutive two-step
mechanism illustrated in Scheme 4.
is thought that up to about 10 collisions occur before the
reactants separate.34 The diffusion process is illustrated by a
double arrow within a circle on the RC. The encounter complex,
which has a lifetime of a nanosecond or less depending upon
the solvent, is transformed through the first transition state (TS
A) to the reactant complex. We believe that this transformation
consists of solvation and geometry changes leading to a
configuration ready for covalent bond formation. The latter then
is transformed through the second transition state (TS B) to
products. This presentation of the RC diagram for pre-
association uses the bar graph approach employed by Guthrie36
in his discussion of reactions passing through tetrahedral
intermediates. We use the diffusion element to emphasize how
the encounter complex34 is defined.
All of the kinetic data reported in Tables 1-5 and Figures
3-5 are inconsistent with a single-step mechanism for the
reaction between TMT+ with hydrogen biacetate ion either in
the presence of Na+ or Bu4N+ in acetic acid or in mixed
solvents. Furthermore, the data are consistent with the pre-
association mechanism under all circumstances. The data in
Tables 1 and 2 show that the initial reaction takes place during
Since the mechanism depicted in Scheme 4 is the simplest
mechanism consistent with the data, we adopt it for the sake of
simplicity since there is no evidence currently available for a
more complex mechanism. With the latter in mind, the reaction
coordinate diagram illustrated in Scheme 6 was constructed to
describe the intramolecular reaction. The reactant in the lower
left-hand corner of the diagram is {TMT+ HOAc/AcO-}. The
first transition state (TS#1) lies between the reactant and an
intermediate which we refer to as the “Reactant Complex” for
which we have no structural data but it is required by the kinetics
observed. In the following paragraphs we relate our work to
Winstein’s scheme for the dissociation of R-X, where R is an
alkyl or an arylalkyl group and X is a leaving group, and provide
a plausible suggestion for the structure of the “Reactant
Complex”. The second transition state (TS#2) lies between the
“Reactant Complex” and the product TMT-OAc.
The concept of ion pairs in organic reactions owes much to
the pioneering work of Winstein’s group more than 50 years
ago.37 Using kinetic salt effects, they were able to provide
(36) Guthrie, J. P. J. Am. Chem. Soc. 1978, 100, 5892; 1974, 96, 3608;
1973, 95, 6999.
52 J. Org. Chem., Vol. 73, No. 1, 2008

Collapse of a Carbocation-Anion Pair to a Neutral Molecule
FIGURE 5. IRC time profiles for the reaction of TMT+ (0.02 mM) with sodium acetate (0.50 M) in acetic acid. Data for the first 3% of the
reaction (a) and that over the first half-life (b).
SCHEME 5. Reaction Coordinate Diagram for the
Pre-association Mechanism
SCHEME 6. Reaction Coordinate Diagram for the
Intramolecular Collapse of {TMT+ HOAc)/AcO-} to
TMT-OAc and Acetic Acid
convincing evidence for Scheme 7 for the solvolysis reactions
of RX (X ) halide or other leaving groups). Ionization to the
intimate ion pair which does not undergo solvolysis is the first
step in this process. The binding between the ions is weakened
and solvent separates the two in the solvent separated ion pair.
At this stage R+ can react irreversibly with solvent to give the
substitution product. Alternatively, the solvent separated ion pair
can dissociate to the free ions and R+ can react to form the
substitution product. The solvent separated ion pair was observed
to undergo exchange with other anions intentionally incorporated
in the solution for purposes of studying the process and the
substitution reaction could be nearly completely suppressed in
the presence of the common ion, X-.
SCHEME 7. Winstein Solvolysis Scheme
The reaction coordinate diagram (Scheme 8) illustrates the
reactions shown in the horizontal line in Scheme 7 compared
to the dissociation of TMT-OAc. The free energies (G) of wide
bars are drawn to approximately correspond to the relative
energies of the various species which we are able to estimate
from our kinetic data for the reaction of TMT+ with acetate
ion in acetic acid. The designation of the three transition states,
TS(mix), TS#1, and TS#2; refer to the TMT+ system and
TS(mix) is labeled to show that this step takes place during the
time of mixing in the stopped-flow experiments. The labeling
of the other transitions states (TS#1 and TS#2) is consistent
with that in Scheme 6.
The correspondence of the reaction coordinate deduced from
kinetic data for the dissociation of TMT-OAc in acetic acid to
that for the Winstein scheme is truly remarkable. This suggests
that {TMT+ HOAc/AcO-} is really a solvent (HOAc) separated
ion pair and that the “Reactant Complex” is an intimate ion
pair. All of our data can be reconciled with this formulation.
When considered in this context this work can be considered
to be a direct kinetic verification of the Winstein37 scheme which
(37) Winstein, S.; Clippenger, E.; Fainsberg, A. H.; Heck, R.; Robinson,
G. C. J. Am. Chem. Soc. 1956, 78, 328 and references cited therein.
J. Org. Chem, Vol. 73, No. 1, 2008 53

Hao and Parker
formation of {TMT+ HOAc/AcO-}, which we regard as a
solvent separated ion-pair, during the time of mixing. The
intramolecular collapse of the solvent separated ion-pair then
takes place by a two-step mechanism (Scheme 3) involving the
formation of a kinetically significant “Reactant Complex” as
intermediate which we propose to be the intimate ion pair,
{TMT+ -OAc}. This mechanism conforms remarkably well to
the Winstein37 scheme for the ionization of RX (TMT-OAc in
this case) as shown in the comparison illustrated in Scheme 7.
In other work, we have proposed that the “Reactant Complex”
is a tight complex formed by extrusion of solvent and appropri-
ate geometry changes from the encounter complex or in other
cases from the corresponding charge-transfer complex.26,30,32 It
would appear that the latter description can apply equally well
to conversion of a solvent separated ion pair to an intimate ion
pair. Obviously, our findings are unexpected from studies of
carbocation-anion combinations that have appeared in the
interim between the 1950s and the present time. Our observation
that the HOAc/AcO- solutions are ideal systems to obtain
detailed kinetic and mechanism information on these important
reactions will eventually provide the means to greatly increase
our knowledge of how these reactions take place. Work in
progress is expected to provide more detailed information on
this and other cation-nucleophile combination reactions.
SCHEME 8. Comparison of the Winstein Scheme to the
TMT+ System
was based on the more indirect evidence provided by the
addition of three different types of salts; (a) a common ion salt
(LiX for solvolysis of RX), (b) salts with nonreactive anions
(LiY for solvolysis or RX) the effect of which can be attributed
to increasing the ionic strength and (c) salts of reactive anions
(for example LiN3 for the solvolysis of RX) which result in the
formation of a stable compound.
Analysis of ks.s. data in the temperature range from 293 to
323 K resulted in ∆Hq and ∆Sq equal to 12.7 kcal/mol and
-21.5 cal/mol K, respectively. Over the same temperature range,
∆H° and ∆S° for equilibrium (4) were observed to be equal to
-4.28 kcal/mol and -14.3 cal/mol K, respectively. It is of
interest to note that the cation - anion combination reaction
between TMT+ and NaOAc in acetic acid is exothermic (∆H°
) -4.28 kcal/mol) with a significantly large barrier (∆Hq )
12.7 kcal/mol, calculated from ks.s. data). Since our evidence
strongly suggest that this cation-anion combination is a
unimolecular reaction of the solvent separated ion-pair and takes
place in two steps, the activation parameters refer to two
consecutive transition states. The data are illustrated in Figure
3. The corresponding plots for the reactions in HOAc/AN
(1/1) are shown in Figure S3 and S4 (Supporting Information),
from which ∆Hq and ∆Sq were found to be equal to 13.7
kcal/mol and -20.7 cal/mol K, respectively. Values of ∆H°
and ∆S° for equilibrium (4) were observed to be equal to -1.40
kcal/mol and -5.2 cal/mol K, respectively. The activation
parameters in the mixed solvent in the presence of Bu4N+
ion are very nearly the same as those observed in HOAc in
the presence of Na+ ion. On the other hand, both thermody-
namic parameters for equilibrium (4) are considerably more
negative in the mixed solvent than were observed in HOAc as
solvent.
It was mentioned earlier that the changes in the kN/ks.s. ratios
with wavelength in Tables 2 and 3 can be explained by
interference from the absorbance by the intermediate. This
appears to be consistent with the intermediate being the intimate
ion pair {TMT+ -OAc}. We do not find any significant
differences in the visible absorption spectrum of {TMT+ HOAc/
AcO-} in HOAc/AN (1/1) and of {TMT+ ClO4-} in CHCl3 or
of the former in HOAc (Figure 2). However, it is conceivable
that that the spectrum of {TMT+ -OAc} is shifted slightly to
lower wavelengths and that the observed absorbance at the
higher wavelengths, during the reaction, is more reflective of
that of the reactant, {TMT+ HOAc/AcO-}.
Experimental Section
Kinetic Analysis. The appropriate integrated rate law for the
pseudo first-order reversible reaction (6) is given by eq 7 where
R
Ao , AeqR, and AtR are the reactant absorbances at zero time, at
equilibrium and at time (t), respectively. Equation 7 is directly
applicable at wavelengths where the reactant is the only absorbing
species. Under these circumstances Aeq is directly measured as
the plateau absorbance at long times.
R
R yk
\
z P
(6)
(7)
k′
AoR - Aeq
R
ln
) (k + k′)t
AtR - Aeq
R
[
]
The first step in our analysis is to convert the (Ao - AeqR)-
R
time profile to an extent of reaction-time profile by dividing (AtR
- AeqR) by (Ao - AeqR) at each time point in the absorbance-
R
time array. Any kinetic operations on the extent of reaction time
profile defined in this way give results equivalent to those when
using the pseudo first-order relationship but in this case the apparent
rate constant is equal to the sum of forward and reverse rate
constants, kapp ) (k + k′).
Experimental Procedure for Kinetic Studies. Solvents and
Reactants. Acetic acid and acetonitrile were of the highest purity
level commercially available and distilled before use. Sodium
acetate was ACS reagent grade and used without further purifica-
tion.
Bu4N+(HOAc/AcO-). A 40% aqueous solution of Bu4NOH
(0.008 mol) was allowed to react with 2 equiv of HOAc (0.016mol)
in CH2Cl2 at room temperature. After stirring for about 30 min,
anhydrous sodium sulfate (about 10 g) was added to remove most
of the water. The CH2Cl2 solution was filtered through anhydrous
sodium sulfate before removing the solvent under vacuum. The
resulting white solid (Bu4N+ HOAc/OAc-, 0.0048 mol) was washed
thoroughly with anhydrous ether and the residual ether was removed
under vacuum.
Preparation of 4,4′,4′′-Trimethoxytrityl Perchlorate. 4,4′,4′′-
Trimethoxy-trityl chloride (1mmol) was allowed to react with silver
perchlorate (1mmol) in acetonitrile (50 mL) at room temperature.
Our overall conclusion is that the cation-anion combination
reactions of TMT+ with acetate salts in acetic acid involve the
54 J. Org. Chem., Vol. 73, No. 1, 2008

Collapse of a Carbocation-Anion Pair to a Neutral Molecule
After the precipitation of silver chloride, the solution was filtered
to remove AgCl, and the solvent was removed under vacuum and
dried to gave a red solid (0.89 mmol TMT+ ClO4-) identified by
(1 - E.R.)/time profiles in the manner described in the previous
section. The reactant extinction coefficient was calculated from Abso
and the known reactant concentration.
1
it is HNMR spectrum (solvent: CD3CO2D/CDCl3 ) 1/2). The
Data Handling Procedures. The procedures used have been
described in detail.24,33 Additional information specific to the IRC
procedure are presented here. The application of the IRC method
on calculated data (Figure 1) was used to illustrate the response of
the single step and the pre-association mechanisms. The application
of the method is somewhat more complex on experimental data
since calculating IRC between successive points amplifies the effect
of instrumental noise.33 We have found that smoothing (1 - E.R.)
time profiles by fitting to fifth order polynomial equations does
not give noticeable distortion in the data as long as the fit is limited
to the first 3% of the profile (or if the initial 3% of the data are
excluded and the remainder of the profile is fit precisely by that
treatment).33 An alternative method to approximate IRC is to carry
out a sliding 51 point analysis in which the IRC for the midpoint
of the initial interval (point 26) is estimated from the pseudo first-
order rate constant derived from the 51 data points. The IRC for
subsequent points (27 to 1975 for a 2000 point array) are obtained
by sequentially sliding the data segment analyzed by one point.
The IRC for point 27 is obtained from the points ranging from 2
to 52 that for point 28 from points ranging from 3 to 53, and so
forth until IRC for point 1976 are obtained from the data segment
from point 1950 to 2000. The disadvantage of the latter method is
that IRC for short times are not available from the analysis.
signals at 1.881 to 1.856, 11.533 and 7.254 are due acetic acid and
CHCl3, respectively. The 1HNMR assignments are summarized
below; and the spectrum is illustrated in Figure S5 (Supporting
Information).
Kinetic Measurements. A Hi-Tech Scientific SF-61 Stopped
Flow Spectrophotometer with a Techne Flow Cooler FC-200
thermostat to control the temperature of the cell block within (
0.2 °C in a glove box under a nitrogen atmosphere was used for
kinetic measurements. Data, 2000 points over a time range about
10% greater than the first half-life, were collected under pseudo
first-order conditions at 298 K.
Kinetic Procedure. The stopped-flow instrument used a single-
beam light path. This necessitated subtraction of background
absorbance which was obtained by having pure solvent in the
reactant syringe and the excess reagent in the other syringe in the
same concentration used in the kinetic experiments. Three back-
ground shots were recorded before each set of kinetic shots and all
data points were averaged. A minimum of twenty shots under
kinetic conditions were recorded in each set of measurements. The
kinetic absorbance-time curves were first averaged, background
absorbance was subtracted and the zero time absorbance (Abso)
was calculated by a linear least-squares procedure using the first
seven data points. The absorbance-time arrays were converted to
Acknowledgment. We are grateful to the National Science
Foundation (CHE-0313615) for support of this work. We thank
a manuscript reviewer for suggesting the analogy with the
Winstein scheme for solvolysis of RX.
Supporting Information Available: Tables of kinetic data and
figures of instantaneous rate constant analysis. 1H NMR spectrum
-
1
of TMT+ClO4 in CD3CO2D/CDCl3 (1/2) and HNMR spectrum
of the equilibrium reaction mixture in HOAc. This material is
available free of charge via the Internet at http://pubs.acs.org.
JO702492R
J. Org. Chem, Vol. 73, No. 1, 2008 55