Efficient unimolecular deprotonation of aniline radical cations

Article abstract of DOI:10.1021/jo047813g

Deprotonation of the radical cations of aromatic amines, such as anilines, generally occurs much more slowly than other fragmentation reactions. Here we report a stereoelectronic effect involving twisting of the anilino group out of the plane of the benze

Full text of DOI:10.1021/jo047813g

Efficient Unimolecular Deprotonation of Aniline Radical Cations
Gary W. Dombrowski and Joseph P. Dinnocenzo*
Department of Chemistry, University of Rochester, Rochester, New York 14627-0216
Paul A. Zielinski and Samir Farid*
Eastman Kodak Company, Research Laboratories, Rochester, New York 14650-2109
Zofia M. Wosinska and Ian R. Gould*
Department of Chemistry and Biochemistry, Arizona State University, Tempe, Arizona 85287-1604
igould@asu.edu
Received December 14, 2004
Deprotonation of the radical cations of aromatic amines, such as anilines, generally occurs much
more slowly than other fragmentation reactions. Here we report a stereoelectronic effect involving
twisting of the anilino group out of the plane of the benzene ring that results in a significantly
increased rate of reactivity toward deprotonation. Quantitative studies of the rate constants for
deprotonation as a function of aniline radical cation pKa (Brønsted plots) demonstrate that the
effect is not simply due to a change in the reaction thermodynamics. By combining this
stereoelectronic effect with covalent attachment of carboxylate as a base, aniline radical cations
that undergo unimolecular deprotonation with rate constants as high as 108 s-1, even in unfavorable
protic media, are described.
I. Introduction
Deprotonation reactions of amine radical cations to
form R-amino radicals have attracted considerable at-
tention because of interest in their mechanisms,1 and
because of their potential synthetic utility.2 This simplest
fragmentation reaction of a radical cation is illustrated
in eq 1 for a generic amine radical cation that undergoes
deprotonation by a base, B-. The R-amino radical product
of deprotonation is a very strong reducing agent, which
has added to the interest in amine radical cation frag-
mentation reactions.3
Deprotonation of amine radical cations is also impor-
tant in several practical applications of photoinduced
electron-transfer reactions.4 One recent example in silver
halide photography that capitalizes on the strong reduc-
(2) (a) Mariano, P. S.; Stavinola, J. L. In Synthetic Organic Photo-
chemistry; Horspool, W. M., Ed.; Plenum Press: New York, 1984;
Chapter 3. (b) Jeon, Y. T.; Lee, C. P.; Mariano, P. S. J. Am. Chem.
Soc. 1991, 113, 8863. (c) Xu, W.; Mariano, P. S. J. Am. Chem. Soc.
1991, 113, 1431. (d) Yoon, U. C.; Mariano, P. S. Acc. Chem. Res. 1992,
25, 233. (d) Yoon, U. C.; Mariano, P. S. Acc. Chem. Res. 2001, 34, 523.
(e) Yoon, U. C.; Mariano, P. S. J. Photosci. 2003, 10, 89.
(3) (a) Wayner, D. D. M.; Dannenberg, J. J.; Griller, D. Chem. Phys.
Lett. 1986, 131, 189. (b) Wayner, D. D. M.; McPhee, D. J.; Griller, D.
J. Am. Chem. Soc. 1988, 110, 132. (c) Jockusch, S.; Timpe, H.-J.;
Schnabel, W.; Turro, N. J. J. Photochem. Photobiol. A: Chem. 1996,
96, 129.
(1) See, for example: (a) Cohen, S. G.; Parola, A.; Parsons, G. H.
Chem. Rev. 1973, 73, 141. (b) Lewis, F. D.; Ho, T.-I.; Simpson, J. T. J.
Org. Chem. 1981, 46, 1077. (c) Lewis, F. D.; Ho, T.-I.; Simpson, J. T.
J. Am. Chem. Soc. 1982, 104, 1924. (d) Davidson, R. S. Adv. Phys. Org.
Chem. 1983, 19, 1. (e) Lewis, F. D. Acc. Chem. Res. 1986, 19, 401. (f)
Nelsen, S. F.; Ippoliti, J. T. J. Am. Chem. Soc. 1986, 108, 4879. (g)
Kavarnos, G. J.; Turro, N. J. Chem. Rev. 1986, 86, 401. (h) Ci, X.;
Silveira da Silva, R.; Nicodem, D.; Whitten, D. G. J. Am. Chem. Soc.
1989, 111, 1337. (i) Pienta, N. J. Photoinduced Electron Transfer; Fox,
M. A., Chanon, M., Eds.: Elsevier: New York, 1988; Part C. (j) Kellett,
M. A.; Whitten, D. G.; Gould, I. R.; Bergmark, W. R. J. Am. Chem.
Soc. 1991, 113, 358. (k) Dinnocenzo, J. P.; Karki, S. B.; Jones, J. P. J.
Am. Chem. Soc. 1993, 115, 7111. (l) Yoon, U. C.; Mariano, P. S.; Givens,
R. S.; Atwater, B. W. Adv. Electron Transfer 1994, 4, 117. (m) Yoon,
U. C.; Mariano. P. S. J. Photosci. 2003, 10, 89.
(4) (a) Reiser, A. Photoreactive Polymers, the Science and Technology
of Resists; Wiley: New York, 1989. (b) Paczkowski, J.; Neckers, D. C.
In Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-VCH: New
York, 2001; Vol. 5, p 516.
10.1021/jo047813g CCC: $30.25 © 2005 American Chemical Society
Published on Web 04/12/2005
J. Org. Chem. 2005, 70, 3791-3800
3791

Dombrowski et al.
SCHEME 2. The Angle θ Is the Dihedral Angle
between the Plane Containing the Benzene Ring
and That Containing the Nitrogen and the
Carbons of the r-Methyl Groups
SCHEME 1. Schematic Illustration of the
Two-Electron-Sensitization Scheme, Showing
Injection of Two Electrons into the Conduction
Band of Silver Halide, Ag+X- a
a The trapped electrons are indicated as Ag0.
ing power of the R-amino radical is two-electron sensi-
tization.5 Here, light absorption to form the excited state
of a photographic sensitizing dye, D*, which is adsorbed
on the surface of microcrystalline silver halide, Ag+X-,
results in transfer of an electron into the conduction band
and formation of the dye radical cation, D•+, Scheme 1.
In the presence of an appropriate amine additive, A-H,
donation of an electron to D•+ can occur to form the amine
radical cation, A-H•+. Fragmentation (in this case depro-
tonation) of A-H•+ forms an R-amino radical, A•, that
injects a second electron into the conduction band,
Scheme 1. Thus, two electrons can be transferred per
absorbed photon, thereby doubling the photographic
response.5
Decarboxylation and desilylation fragmentation reac-
tions of amine radical cations have been successfully
implemented in two-electron sensitization schemes;6
however, the corresponding deprotonation reaction ini-
tially posed a challenge. Alkylamines, whose radical
cations have been shown to undergo relatively facile
deprotonation,7 cannot be used in aqueous photographic
emulsions because they would be protonated and thus
not function as electron donors. In addition, Mariano et
al. have clearly shown that deprotonation of aromatic
amine radical cations tends to be much slower than other
possible fragmentation reactions, and is thus often less
useful.8 The kinetics of deprotonation, as well as other
fragmentation reactions of the radical cations, are ex-
pected to increase with increasing oxidation potential of
the amine.9 This is not a generally useful solution to the
problem of low kinetic reactivity, especially in the two-
electron sensitization application, since increasing the
oxidation potential reduces the ability of the amine to
act as an electron donor. Thus, an alternate method was
required for controlling the reactivity of amine radical
cations. We have found a stereoelectronic effect that can
significantly increase the rate constants for deprotonation
of aniline radical cations.10 Based upon this effect, aniline
derivatives have been designed with a covalently at-
tached base that undergo efficient unimolecular (in-
tramolecular) deprotonation upon one-electron oxidation.
Here we describe anilines that have relatively low
oxidation potentials, and whose radical cations deproto-
nate with controllable rate constants over 4 orders of
magnitude. Reactions with rate constants approaching
108 s-1 have been obtained, even in unfavorable protic
media.
The first part of this paper describes kinetic and
thermodynamic studies of the rate constants for bimo-
lecular deprotonation of aromatic amine radical cations
that delineate the stereoelectronic effect. The second part
describes structures that undergo unimolecular fragmen-
tation. The paper concludes with a discussion of other
factors that control the deprotonation reactions of amine
radical cations.
II. Results and Discussion
II. A. Bimolecular Deprotonation. Anilines in which
ortho substituents (R, Scheme 2) twist the nitrogen out
of conjugation with the benzene ring are known to be
more basic than simple anilines.11 This is understood as
a consequence of localization of the nonbonding electrons
on nitrogen, which increases their basicity. It seemed
reasonable that similar twisting might also help to
localize the spin and positive charge close to the nitrogen
in the corresponding radical cations, thus enhancing the
electrophilicity of hydrogen atoms on carbons R to the
nitrogen. The kinetics of amine radical cation deproto-
nation have been studied previously with a variety of
techniques.8,12 Rate constants are expected to depend on
the thermodynamics of the reaction, i.e., on the strength
of the base and the pKa of the radical cation, and this
has been shown to be the case for several amines, and
other radical cations.8a,13,14 Indeed, two previous studies
have specifically addressed the relationship between
kinetics and thermodynamics for deprotonation of sub-
(10) Deprotonation would also be expected to be faster with increas-
ing stability of the radical product, and this has also been confirmed
experimentally.8a
(5) (a) Gould, I. R.; Lenhard, J. R.; Muenter, A. A.; Godleski, S. A.;
Farid, S. J. Am. Chem. Soc. 2000, 122, 11934. (b) Gould, I. R.; Lenhard,
J. R.; Muenter, A. A.; Godleski, S. A.; Farid. S. Pure Appl. Chem. 2001,
73, 455.
(6) (a) Gould, I. R.; Godleski, S. A.; Zielinski, P. A.; Farid, S. Can.
J. Chem. 2003, 81, 777. (b) Gould, I. R.; Farid, S. J. Phys. Chem. A
2004, 108, 10949.
(7) See, for example: Davidson, R. S. Adv. Phys. Org. Chem. 1983,
19, 1.
(8) (a) Zhang, X.; Yeh, S.-R.; Hong, S.; Freccero, M.; Albini, A.;
Falvey, D. E.; Mariano, P. S. J. Am. Chem. Soc. 1994, 116, 4211. (b)
Su, Z.; Falvey, D. E.; Yoon, U. C.; Mariano, P. S. J. Am. Chem. Soc.
1997, 119, 5261. (c) Su, Z.; Mariano, P. S.; Falvey, D. E.; Yoon, U. C.;
Oh, S. W. J. Am. Chem. Soc. 1998, 120, 10676. (d) Yoon, U. C.; Kwon,
H. C.; Hyung, T. G.; Choi, K. H.; Oh, S. W.; Yang, S.; Zhao, Z.; Mariano,
P. S. J. Am. Chem. Soc. 2004, 126, 1110.
(9) (a) Nicholas, A. M. P.; Arnold, D. R. Can. J. Chem. 1982, 60,
2165. (b) Arnold, D. R.; Snow, M. S. Can. J. Chem. 1985, 66, 3012. (c)
Gaillard, E. R.; Whitten, D. G. Acc. Chem. Res. 1996, 29, 292.
(11) Ingold, C. K. Structure and Mechanism in Organic Chemistry;
Cornell University Press: Ithica, NY, 1953; Chapter XIII.
(12) (a) Powell, M. F.; Wu, J. C.; Briuce, T. C. J. Am. Chem. Soc.
1984, 106, 3850. (b) Sinha, A.; Bruice, T. C. J. Am. Chem. Soc. 1984,
106, 7291. (c) Fukuzumi, S.; Kondo, Y.; Tanaka, T. J. Chem. Soc.,
Perkin Trans. 2 1984, 673. (c) Das, S.; Von Sonntag, C. Z. Naturforsch.
1986, 416, 505. (d) Hapiot, P.; Moiroux, J.; Saveant, J.-M. J. Am. Chem.
Soc. 1990, 112, 1337. (e) Anne, A.; Hapiot, P.; Moiroux, J.; Neta, P.;
Saveant, J.-M. J. Phys. Chem. 1991, 95, 2370. (f) Fukuzumi, S.;
Tokuda, Y.; Kitano, T.; Okamoto, T.; Otera, J. J. Am. Chem. Soc. 1993,
115, 8960. (g) Anne, A.; Fraoua, S.; Hapiot, P.; Moiroux, J.; Saveant,
J.-M. J. Am. Chem. Soc. 1995, 117, 7412. (h) Goez, M.; Sartorius, I. J.
Phys. Chem. A 2003, 107, 8539.
(13) (a) Dinnocenzo, J. P.; Banach, T. E. J. Am. Chem. Soc. 1989,
111, 8646. (b) Parker, V. D.; Tilset, M. J. Am. Chem. Soc. 1991, 113,
8778. (c) Agnes, A.; Hapiot, P.; Moiroux, J.; Neta, P.; Saveant, J.-M. J.
Am. Chem. Soc. 1992, 114, 4694.
3792 J. Org. Chem., Vol. 70, No. 10, 2005

Deprotonation of Aniline Radical Cations
TABLE 1. Spectral, Kinetic, and Thermodynamic
Properties Related to Deprotonation of Aromatic Amine
Radical Cations with Tetrabutylammonium Acetate as
the Base in Acetonitrile Containing 0.5 M Water and 0.5
M Tetrabutylammonium Perchlorate
FIGURE 1. Plots of observed rate of pseudo-first-order decay,
kobs, versus concentration of tetrabutylammonium acetate for
deprotonation of the radical cations of (open circles) N,N-
dimethylaniline, 1c, and N,N,-dimethylaniline with (filled
circles) a 4-methyl substituent, 1b, (filled squares) a 2-methyl
substituent, 2, and (open squares) a 2-tert-butyl substituent,
3, in acetonitrile with 0.5 M water and 0.5 M tetramethylam-
monium perchlorate, at room temperature.
cations.6,17 The deprotonation rate constants were deter-
mined from the slopes of plots of the observed pseudo-
first-order decay rate constants of the radical cations
monitored at their absorption maxima, kobs, as a function
of the concentration of added acetate ion. The basicity of
the acetate anion is expected to be strongly dependent
upon adventitious water, and therefore to ensure repro-
ducibility in the measured rate constants the reactions
were performed in the presence of 0.5 M water. To
eliminate the effect of changing ionic strength with
increasing acetate concentration, the reactions were also
performed in the presence of 0.5 M tetrabutylammonium
perchlorate. Further details are given in the Experimen-
tal Section. Typical data are given in Figure 1 and the
rate constants are summarized in Table 1.
The observed bimolecular reactions between the radical
cations and the acetate were assigned to deprotonation
based on the similarity of the results to those previously
described,8a on the response of the reaction rate constants
to changes in structure of the anilines, and also on the
basis of observed primary deuterium isotope effects
discussed below.
Relating the kinetics and thermodynamics of the
deprotonation reactions requires determination of the pKa
of the radical cations.9 The usual approach uses a
thermodynamic cycle, as shown in Scheme 3, and eq 3a
to determine the C-H bond dissociation energies of the
radical cations, BDE•+.13a The corresponding pKa•+ values
for the radical cations are obtained from the BDE•+
according to eq 3b. The relevant data are collected in
Table 1.
a Absorption maximum of the radical cation in the visible region.
b Rate constant for deprotonation at room temperature. c Obtained
by comparing ArN(CH3)2 to ArN(CD3)2. d Oxidation potentials
measured as described in the Experimental Section. e Data taken
from ref 15, unless otherwise indicated. f Calculated with eq 3.
g Not measured, value taken as the average of the measured values
for related compounds. h Not measured, kinetics determined at 470
nm. i The absorption maximum for this radical cation varied
somewhat with time.20b
stituted dimethylaniline radical cations.8a,13b However,
the deprotonation rate constants are found to depend
strongly upon reaction conditions.8a To clearly delineate
the stereoelectronic effect, we revisited the kinetics of
these deprotonation reactions using a quantitative Brøn-
sted analysis that allowed comparison of the twisted and
nontwisted anilines under identical conditions. The pKa
values for the radical cations have also been recalculated
by using our recently revised R-C-H bond dissociation
energies for the N,N-dimethylanilines.15
+
Bimolecular rate constants, k-H , for deprotonation of
the radical cations of the 4-substituted N,N-dimethyl-
anilines 1a-e, Table 1, were determined in acetonitrile
solution at room temperature, using tetrabutylammo-
nium acetate as the external base, eq 2. The radical
cations were generated in time-resolved experiments by
using cyanoanthracene/cosensitization,16 as described in
detail elsewhere for studies of related aniline radical
BDE•+ ) ∆G°BDE + ENHE + Eox
pKa•+ ) BDE•+/2.303 RT
(3a)
(3b)
(14) (a) Schlesener, C. J.; Amatore, C.; Kochi, J. K. J. Am. Chem.
Soc. 1984, 106, 3567. (b) Schlesener, C. J.; Amatore, C.; Kochi, J. K.
J. Am. Chem. Soc. 1984, 106, 7472. (c) Schlesener, C. J.; Amatore, C.;
Kochi, J. K. J. Phys. Chem. 1986, 90, 3747. (d) Anne, A.; Fraoua, S.;
Grass, V.; Moiroux, J.; Saveant, J.-M. J. Am. Chem. Soc. 1988, 110,
2951.
(16) Gould, I. R.; Ege, D.; Moser, J. E.; Farid, S. J. Am. Chem. Soc.
1990, 112, 4290.
(17) Experimental details are given in ref 5a.
(15) Dombrowski, G. W.; Dinnocenzo, J. P.; Farid, S.; Goodman, J.
L.; Gould, I. R. J. Org. Chem. 1999, 64, 427.
J. Org. Chem, Vol. 70, No. 10, 2005 3793

Dombrowski et al.
SCHEME 3
From the data in Table 1 it is clear that, as previously
observed, increasing the pKa•+ of the aniline radical cation
results in a corresponding decrease in the rate constant
for deprotonation.8a,13b This effect is illustrated graphi-
FIGURE 2. Brønsted plot of the logarithm of the bimolecular
rate constant for deprotonation of the radical cations of N,N-
dimethylaniline derivatives: (closed circles) 4-substituted-N,N-
dimethylanilines, 1a-1e; (open circle) N,N,2-trimethylaniline,
2; (open squares) 2-tert-butyl-N,N-dimethylaniline, 3; and
N,N,2,6-tetramethylaniline, 4. The straight line represents the
best fit to the 4-substituted anilines and is characterized by a
slope of -0.23 and intercept of 12.3.
+
cally in Figure 2, which shows a Brønsted plot of log(k-H
versus pKa for these reactions.
)
•+
These data are in general agreement with several other
quantitative studies of deprotonation of amine radical
cations.8a,13b The difference in the rate constants for the
slowest (1a) and the fastest (1e) of the 4-substituted
anilines measured in this work is ca. 2 orders of magni-
tude. It is interesting that Mariano et al. obtained a very
similar relative rate difference for these same two
reactions,8a although their absolute rate constants are
roughly 2 orders of magnitude smaller than those re-
ported here. Presumably, this is because the solvent
mixture used by Mariano et al. was more “protic” (i.e.
60:40 methanol:acetonitrile) than that used in the present
experiments (0.5 M water in acetonitrile corresponds to
ca. 1% water in acetonitrile), which should reduce the
basicity of the carboxylate.18
SCHEME 4. Structures of Sterically Twisted
N,N-Dimethylaniline Radical Cations, and the
Dihedral Angles between the Planes Containing
the Aromatic Ring and the Amino Nitrogen, θ, As
Defined in Scheme 2
Included in Table 1 are rate constants for reaction of
acetate anion with several of the aniline radical cations,
where the hydrogens on the N-methyl groups have been
deuterated. Primary isotope effects were found, consis-
tent with C-H bond breaking in the transition state, i.e.,
deprotonation. Mariano et al. observed a somewhat larger
isotope effect for deprotonation of the unsubstituted N,N-
dimethylaniline radical cation 1c.8a The relatively small
kinetic isotope effects are also consistent with a fairly
early transition state,19 and of course this is quite
reasonable given the large absolute values of the rate
constants, Table 1. Consistent with this conclusion is the
slope of the Brønsted plot for the 4-substituted anilines
of -0.23. Such a small slope would normally be associated
with a small extent of proton transfer in the transition
state.19
The Brønsted plot is quite linear over 2 orders of
magnitude in rate constant. Previously, Saveant et al.
analyzed the rate constants for deprotonation of a series
of NADH derivatives using a Marcus-type analysis, i.e.,
a nonlinear free energy relationship.13c In the present
case, however, since the data fit reasonably well to a
linear relationship, there is obviously no additional
information that can be obtained from a nonlinear fit.
The radical cations of several dimethylanilines with
substituents in the ortho positions were examined by
using B3LPY/6-31G* computations. The structures of
these compounds, 2-6, are summarized in Scheme 4,
together with the dihedral angles between the plane
containing the aromatic ring and that containing the
nitrogen, θ (hereafter referred to as the twist angle), as
defined in Scheme 2. Also given is the corresponding
angle for the parent dimethylaniline radical cation, 1c•+.
Not surprisingly, the twist angle increases with increas-
ing bulk of the ortho substituent, with the N,N,2,6-
tetramethylaniline radical cation, 4•+, being the most
twisted.
The radical cations of three of these, 2•+-4•+, were also
studied experimentally under the same conditions as for
the para-substituted anilines. As illustrated in Figure 3,
the transient absorption spectrum of the radical cation
of 2 is virtually identical with that of the parent N,N-
dimethylaniline 1c. The more highly twisted 3•+ and 4•+,
however, give radical cations with increasingly broadened
spectra and red-shifted absorption maxima.20
(20) (a) In addition to red-shifting, the extinction coefficient at the
absorption maxima also decrease with increased twisting, although
accurate measurements of this effect were not made. (b) The absorption
spectrum of 3•+ exhibited a slight red-shift over a 1-2 µs time scale
under the experimental conditions; see Supporting Information, Figure
S3.
(18) Isaacs, N. Physical Organic Chemistry; Longman: Harlow,
England, 1987.
(19) Ritchie, C. D. Physical Organic Chemistry. The Fundamental
Concepts; Marcel Dekker: New York, 1975.
3794 J. Org. Chem., Vol. 70, No. 10, 2005

Deprotonation of Aniline Radical Cations
TABLE 2. Kinetic Parameters for Intramolecular
Deprotonation of the Radical Cations of Aniline
Derivatives, 7a-f, with Varying Substituents in the
Ortho Position, Measured in Acetonitrile with 20% Water
a
k-H
Ea
log(A)a
+
R
R′
compd
(s-1
)
(kcal/mol)
(s-1
)
b
H
H
7a
<5 × 104
b
b
(5 × 105)c
2.0 × 105
5.9 × 105
1.8 × 106
(11.9)c
12.9
12.6
11.6
(14.3)c
14.9
Me
Et
H
7b
7c
7d
7e
7f
H
15.1
14.8
i-Pr
H
d
t-Bu
H
∼4 × 107
∼9 × 107
d
Me
Me
FIGURE 3. Absorption spectra of the radical cations of (closed
circles) N,N-dimethylaniline, 1c, and N,N,-dimethylaniline
with (open circles) a 2-methyl substituent, 2, (closed squares)
a 2-tert-butyl substituent, 3, and (open squares) 2,6-dimethyl
substituents, 4, in acetonitrile containing 0.5 M water and 0.5
M tetrabutylammonium perchlorate, at room temperature.20
a Arrhenius activation energy, Ea, and intercept of Arrhenius
plot, log(A). b This reaction under these conditions was too slow
to be measured.21 c Data for acetonitrile with 0.5 M water.
d Estimated value, see text.
Rate constants for deprotonation of the radical cations
of 2-4 were measured under the same conditions used
for compounds 1a-1e, Table 1. Compared to the unsub-
stituted aniline 1c•+, the reactivity increases roughly with
increasing twist of the dimethylamino group, with depro-
tonation of the more highly twisted 3•+ and 4•+ being ca.
30 times faster than that of the parent aniline 1c•+.
Twisting the aryl group results in an increase in the
oxidation potential of the amines compared to that of the
parent N,N-dimethylaniline 1c, Table 1. Thus, deproto-
nations of 2•+-4•+ are more exothermic than 1c•+. How-
ever, the Brønsted plot of Figure 2 shows that, whereas
the influence on the rate constant is hardly noticeable
for the less twisted derivative, 2•+, the rate constants for
the strongly twisted 3•+ and 4•+ are approximately an
order of magnitude larger than would have been pre-
dicted from the 4-substituted aniline data. This clearly
indicates that the rate increase is not simply due to an
increase in the driving force for these reactions, but is a
true stereoelectronic effect. Indeed, the rate constants for
deprotonation of 3•+ and 4•+ are sufficiently high that they
are close to the diffusion-controlled limit (ca. 4 × 109 M-1
s-1), which suggests that an order of magnitude probably
represents a lower limit to the increase in the rate
constant as a result of charge localization.
carbon chain, since this both prevents decarboxylation
in the radical cation6b,8a and also allows a six-membered
ring in the transition state for deprotonation, as discussed
further below.
II. B. i. Reaction Kinetics. These reactions were
studied in acetonitrile with a higher water content than
used in the bimolecular studies for two reasons. First,
the observed rates of decay of the radical cations in the
unimolecular reactions were found to be significantly
larger than those of the pseudo-first-order rate constants
in the bimolecular reactions at the external base concen-
trations used. The use of 20% water, compared to the 0.5
M (ca. 1%) water used in the bimolecular reactions,
decreases the rate of reaction into a range that can be
readily determined by using our time-resolved apparatus.
Second, the kinetics of the unimolecular deprotonation
reactions could be directly compared to corresponding
decarboxylation and desilylation reactions for two-
electron sensitization application, since these were also
studied in 20% aqueous acetonitrile.6,17
In the solvent mixture 20% water in acetonitrile, the
radical cations of the anilines 7a-f exhibited transient
absorption spectra that were similar to those of the
corresponding N,N-dimethylanilines given in Figure 3
(data not shown). The time-resolved decays of the radical
cation absorptions were observed to be first order, and
were assigned to intramolecular deprotonation on the
basis of the fact that the decay rates responded to
molecular structure in a manner consistent with this
reaction (see below). In addition, the R-amino radical, the
product of deprotonation, eq 4, was shown to transfer an
electron to 9,10-dicyanoanthracene, as described else-
where for related systems.17 Finally, at the concentrations
of the anilines used in the experiments (ca. 2-4 mM),
the rates of intermolecular deprotonation were much
slower than the corresponding rates of intramolecular
deprotonation (see also below).
II. B. Intramolecular Deprotonation of Twisted
Anilines. The utility of a deprotonation reaction that
does not require an external base is obvious. We have
investigated the reactivity of aniline radical cations in
which a carboxylate functional group is included as a
substituent, structures 7, eq 4. The clear result from the
bimolecular reactions is that deprotonation is faster when
delocalization of the charge is minimized, thus we
expected that the rate of intramolecular deprotonation
should increase with increasing twist angle in structures
7. The carboxylate is attached to the nitrogen via a three-
Table 2 summarizes the rate constants for unimolecu-
lar deprotonation in the radical cations of 7a-f. With
increasing twist angle, the rate constant increases by
more than 3 orders of magnitude: from <5 × 104 for the
J. Org. Chem, Vol. 70, No. 10, 2005 3795

Dombrowski et al.
TABLE 3. Rate Constants and Activation Parameters
for Intramolecular Deprotonation of Aniline Radical
Cations as a Function of the Length of the Methylene
Chain Connecting the Nitrogen Atom to the Carboxylate
Group, Measured in Acetonitrile with 20% Water
a
log(A)a
(s-1
+
k-H
Ea
compd
(s-1
)
(kcal/mol)
)
n ) 2
n ) 3
n ) 4
n ) 5
n ) 6
7g
7f
7h
7i
1.8 × 106
∼9 × 107 b
1.3 × 107
1.4 × 106
2.3 × 105
12.4
15.4
10.2
11.0
11.2
14.6
14.3
13.7
7j
a Arrhenius activation energy, Ea, and intercept of Arrhenius
plot, log(A). b Estimated value, see text.
SCHEME 5
FIGURE 4. Plot of the logarithm of the rate constant for
+
unimolecular deprotonation (log(k-H )) versus the cosine squared
of the dihedral angle θ, defined in Scheme 1, for the simple
dimethylanilines, assumed to be appropriate for the intramo-
lecular deprotonating anilines here. The point for the radical
cation of 7a (indicated) is a maximum value.21
unsubstituted compound 7a to ∼9 × 107 s-1 for the o,o′-
dimethyl compound 7f.21
The combined stereoelectronic effect and increase in
oxidation potential with increasing twisting in the uni-
molecular systems is quite dramatic. The extent of charge
delocalization is likely to be related to the extent of orbital
overlap of the aromatic π system and the nonbonding
orbital on nitrogen. In turn, this can be related to cos2θ,
Not surprisingly, the maximum rate constant is at-
tained for 7f, with n ) 3, which has a six-membered ring
transition state for radical cation deprotonation, Scheme
5a. With each additional methylene group in the chain,
the rate constant decreases by an average factor of ∼7-
8, Figure 5. It is noteworthy that with n ) 2 (compound
7g), which has a five-membered ring transition state for
deprotonation from the chain containing the carboxylate,
proton transfer from the R-carbon of the ethyl group
would have a seven-membered-ring transition state,
Scheme 5b. The rate constant for 7g, however, is lower,
by a factor of ∼7, compared to that for 7h, which reacts
through a seven-membered transition state for deproto-
nation on the same chain as the carboxylate, Scheme 5c.22
An unfavorable conformation for the intrachain reaction
in 7g, relative to that for interchain reaction in 7h, is
likely to be the reason for the different reactivities of
these two seven-membered-ring transition state reac-
tions. The most favorable transition state will presum-
ably be favored by two structural features: (1) a linear
or near-linear O-H-C bond angle and (2) a collinear or
near-collinear alignment of the C-H bond and the
p-orbital at nitrogen.
Pictorial representations of the transition states for the
two reactions are given in Figure 6. These were obtained
from a B3LYP/6-31G* optimized structure for N,N,2,6-
tetramethylaniline, with addition of the appropriate side
chains to make 7g and 7h. The appropriate torsional
angles were adjusted manually to create transition state
structures that optimize the structural features cited
above (the structures are not optimized in any other way).
The structures show that for the “same chain” proton
transfer in 7h, a structure with a O-H-C angle close to
2
+
where θ is the twist angle. A plot of log(k-H ) versus cos θ,
where the values of θ are assumed to be the same as
those for the corresponding simple dimethylanilines from
Scheme 4, is shown in Figure 4. This plot is remarkably
linear, even though there is no reason to expect a strictly
linear relationship, and clearly indicates how twisting
can be used to control the rate of deprotonation over
nearly 4 orders of magnitude.
II. B. ii. Effect of Tether Length. Another way in
which the rate constant for unimolecular deprotonation
could in principle be controlled is by varying the length
of the methylene chain, (CH2)n, between the carboxylate
group and the nitrogen. The shortest tether is -(CH2)2-,
because the radical cations of derivatives with one
methylene unit are likely to undergo decarboxylation
rather than deprotonation.6b,8c In addition to 7f, we
investigated the deprotonation of the radical cations of
other derivatives, compounds 7g-j, Table 3, in which the
two o-methyl groups enhanced charge localization around
the nitrogen, and where the number of methylene groups,
n, was varied from 2 to 6. The results are summarized
in Table 3.
(21) For compound 7a, intramolecular deprotonation in acetonitrile
with 20% water was too slow to measure, because of the combined low
kinetic acidity of this untwisted amine radical cation moiety, and the
low kinetic basicity of the carboxylate under these conditions. The
observed rate of decay at room temperature was 5 × 104 s-1; however,
the reaction had Arrhenius activation parameters that were very
different from the other reactions, i.e., Ea ) 3.8 kcal/mol and log(A) )
7.7. These parameters would be more consistent with a bimolecular
reaction, for example, quenching by impurities. The activation param-
eters with 0.5 M water, where the carboxylate basicity is much larger,
Table 2, are consistent with deprotonation under these conditions.
Thus, we assign the observed rate of 5 × 104 s-1 in 20% water to be
the maximum rate constant for unimolecular deprotonation.
(22) The factor of ∼7 is, of course, a lower limit for the difference in
reactivity between the two seven-membered transition states in
compounds 7g and 7h, because the intrachain reaction in 7g, the five-
membered-ring transition state, is unlikely to be negligible.
3796 J. Org. Chem., Vol. 70, No. 10, 2005

Deprotonation of Aniline Radical Cations
TABLE 4. Rate Constants for Intramolecular
Deprotonation of Nontwisted Aniline Radical Cations as
a Function of the Length of the Methylene Chain
Connecting the Nitrogen Atom to the Carboxylate
Group, Measured in Acetonitrile with 1%, 2%, and 4%
Water
-1
+
k-H (s
)
compd 1% H2O
2% H2O
4% H2O
n ) 3
n ) 4
n ) 5
n ) 6
8a
8b
8c
8d
1.3 × 106 4.8 × 105 1.6 × 105
6.8 × 105 2.2 × 105 9.3 × 104
6.5 × 105 2.4 × 105 9.0 × 104
3.4 × 105
FIGURE 5. Plot of the logarithm of the rate constant for
+
unimolecular deprotonation (log(k-H )) versus the number of
methylenes separating the nitrogen and carboxylate.
FIGURE 7. Plot of the logarithm of the rate constant for
FIGURE 6. Tube models of schematic transition states for
the different seven-membered cyclic ring intramolecular depro-
tonations in the radical cations from 7g and 7h (see text for
explanation), showing how a collinear arrangement of the
O-H-C atoms is impossible in 7g. For clarity, the hydrogen
and methyl substituents have been removed from the benzene
rings.
+
unimolecular deprotonation (log(k-H )) of the radical cations
of ring-unsubstituted aniline derivatives (8) with varying
length of the methylene chain.
Whereas the rate constant for deprotonation of the
twisted, o,o′-dimethyl derivative with 3 methylene groups,
7f, at room temperature and in the presence of 20% water
is ca. 108 s-1, that of the corresponding unsubstituted
compound 8a at 41 °C, and in the presence of only 4%
water, is ca. 105 s-1. Since the rate constant for depro-
tonation decreases with increasing water content and
with decreasing temperature, the reactivity difference
between these two compounds appears to be more than
4 orders of magnitude!
In addition to the much lower rate constants for
deprotonation, the radical cations of the nontwisted
compounds 8a-d show very little dependence on the
length of the methylene chain. This is in contrast to the
corresponding twisted derivatives, 7f-j, for which the
rate constants vary by two and half orders of magnitude
between the (CH2)3 and (CH2)6 analogues.
180° is possible (the O, H, and C atoms are close to
collinear). In contrast, this is not possible in the transi-
tion state for proton transfer from the N-ethyl substituent
in 7g.
These structural differences can be understood in a
simple way. Shown in Scheme 5, parts b and c, are both
seven-membered deprotonation transition states. The key
structural difference between the two transition states
is the bond angle R, which is ca. 120° and ca. 109° for 7g
and 7h, respectively. The larger angle for 7g is a
consequence of rehybridization at nitrogen (flattening)
upon oxidation. The smaller angle for 7h allows the
carboxylate group to access a more optimal geometry in
the transition state.
II. C. Intramolecular Reactions of Nontwisted
Anilines. Intramolecular reactivity in a series of sub-
stituted anilines without ortho substituents, i.e., non-
twisted anilines 8, Table 4, was also investigated. In-
tramolecular deprotonation in the radical cations of 8a-d
was significantly slower than that in the radical cations
of the corresponding twisted anilines, 7f-j. Indeed, these
reactions could only be studied in solutions with much
lower water content and above room temperature. Some
kinetic data for these compounds are shown in Figure 7,
in acetonitrile solutions with varying water content and
at 41 °C, as a function of varying chain length of the
methylene tether. The data are summarized in Table 4.
For 8a-d, intramolecular deprotonation of the radical
cations was so slow that it could be clearly distinguished
from the corresponding intermolecular reaction. Experi-
ments were performed in which the observed decay rate
of the radical cations was measured as a function of the
concentration of 8 (data not shown), with 1% water at
41 °C. The slopes of these plots gave intermolecular rate
constants for the radical cations of 8a-d of ∼9 ( 2 ×
107 M-1 s-1. These can be compared to a bimolecular rate
constant of 2.5 × 108 M-1 s-1 determined for deprotona-
tion of the radical cation of N,N-dimethylaniline (1c•+)
by acetate anion, also measured at 41 °C and under
similar reaction conditions (data not shown). The simi-
J. Org. Chem, Vol. 70, No. 10, 2005 3797

Dombrowski et al.
TABLE 5. Kinetic Parameters for Deprotonation of
larity of the rate constants confirms that the carboxylate
base and the aniline radical cation parts of 8a•+-d•+
exhibit normal reactivity in a bimolecular reaction.
The intercepts of the plots of rate versus concentration
of 8 correspond to the intramolecular rate constants. The
ratio of the intercepts to the slopes thus yields the
“effective molarities” for the reactions.23 The values
obtained, ca. 0.01 M, are extremely low, since values
considerably greater than unity are often found when
comparing equivalent inter- and intramolecular reac-
tions.23 Quite different behavior was observed for the
corresponding reactions of the twisted radical cations.
Indeed, it was not possible to determine the correspond-
ing effective molarity in these cases since the intramo-
lecular reactions were too fast. For example, the rate
constant for intramolecular reaction of 7f•+ in 20%
aqueous acetonitrile at room temperature is ca. 108 s-1
(Table 3). The rate constant for the closest corresponding
bimolecular reaction, i.e., 4•+ with acetate, is approxi-
mately 30 times larger, 3.6 × 109 M-1 s-1 (Table 1).
However, this is in the presence of only 0.9% water. With
20% water, the bimolecular rate constant would probably
be much more than 30 times smaller than this (see, for
example, Figure S4 in the Supporting Information). Thus,
the effective molarities for the radical cations of the
twisted anilines are almost certainly much greater than
1 M.
Aniline Radical Cations with Tetrabutylammonium
Acetate as the Base, in Acetonitrile Containing 0.5 M
Water (0.9%) and 0.5 M Tetrabutylammonium
Perchlorate
+
compd
k-H a (M-1 s-1
)
Eab (kcal/mol)
log(A)b (s-1
)
1b
4.1 × 107
(2.3 × 107)c
1.2 × 108
1.7 × 108
1.4 × 109
3.6 × 109
(1.9 × 109)c
5.4 (6.6)c
11.8
(12.5)c
13.3
1c
1d
1e
4
7.0
7.6
13.9
6.6
14.0
4.4
12.7
(4.3)c
(12.7)c
a Bimolecular rate constant for deprotonation at room temper-
ature. b Arrhenius activation energy, Ea, and intercept of Arrhe-
nius plot, log(A). c Measured in acetonitrile/propionitrile solvent
mixtures to simulate constant solvent polarity as a function of
temperature; see Experimental Section.
Together, these observations suggest that the rate-
determining steps for the inter- and intramolecular
reactions may be quite different compared with those in
the reactions of the nontwisted anilines. The compara-
tively small rate constants for the 8•+ suggest that the
rate-determining steps are associated with reasonably
large activation barriers that do not depend on the ring
size of the transition state for deprotonation. The origin
of the difference between the chain length dependencies
for the twisted and untwisted aniline radical cation
deprotonations is unclear. It may involve rotamers as-
sociated with the methylene chain that joins the anilino
and carboxylate moieties, or the difference may result
from different stereoelectronic requirements for the
deprotonations.
FIGURE 8. Plots of the logarithm of the rate constants for
+
unimolecular deprotonation (log(k-H )) versus reciprocal tem-
perature, in acetonitrile with 20% water, for the ortho-
substituted radical cations of the aniline derivatives 7b-7e.
II. D. Energetic Considerations. Several of the rate
constants for both unimolecular and bimolecular depro-
tonation reactions were measured as a function of tem-
perature. The Arrhenius parameters are summarized in
Tables 2 and 3 for the intramolecular reactions of the
radical cations of compounds 7a-f, and in Table 5 for
the bimolecular reactions of the 4-substituted dimethyl-
anilines 1-4. Typical plots for the bimolecular reactions
are included in the Supporting Information, and for the
unimolecular reactions in Figures 8 and 9.
FIGURE 9. Plots the logarithm of the rate constants for
+
unimolecular deprotonation (log(k-H )) versus reciprocal tem-
perature, in acetonitrile with 20% water, for the radical cations
of the aniline derivatives 7f-j.
An unexpected observation from the temperature-
dependence experiments on the bimolecular reactions is
that the slowest reacting of the p-substituted aniline
radical cations for which measurements were made, 1b•+,
has the smallest Arrhenius activation energy (Table 5).
The other anilines have similar activation energies,
within experimental error, and the difference in room
temperature rates seems to be reflected more in the
apparent Arrhenius preexponential factors (log(A), Table
5), which also seem to be larger than usually observed
for bimolecular reactions in solution.24
In related work on decarboxylation reactions of aniline
radical cations, we observed unusual Arrhenius param-
eters due to temperature-dependent changes in solvation
of the carboxylate.6b The basicity of the acetate anion is
expected to depend strongly upon solvent polarity. For
example, the pKa of acetic acid increases dramatically
upon going from water to acetonitrile.25 We thus at-
(23) (a) Kirby, A. J. Adv. Phys. Org. Chem. 1980, 17, 183. (b)
Mandolini, L. Adv. Phys. Org. Chem. 1986, 22, 1.
(24) Moore, J. W.; Pearson, R. G. Kinetics and Mechanism, 3rd ed.;
Wiley: New York, 1981.
3798 J. Org. Chem., Vol. 70, No. 10, 2005

Deprotonation of Aniline Radical Cations
tempted to measure the kinetics of the reactions at
different temperatures, but at constant polarity. The
dielectric constant of most solvents decrease with in-
creasing temperature.26 Thus, mixtures of acetonitrile
and propionitrile with 0.5 M water were prepared to
obtain solutions that compensated for changes in solvent
polarity with temperature, according to measurements
with Reichardts dye (see the Experimental Section for
details). However, the data given in Table 5 indicate very
little change in the Arrhenius parameters compared to
experiments using only acetonitrile with 0.5 M water.
This may be because specific hydration of the carboxylate
with water dominates the solvation effects in all of the
solvent mixtures used. Because of possible temperature-
dependent carboxylate hydration, we have not attempted
a detailed analysis of the Arrhenius data using, for
example, transition state theory.
One general point can be made, however. In previous
work on the radical cations of para-substituted N,N-
dimethylaniline radical cations, Parker and Tilset re-
ported extremely large activation entropies for deproto-
nation using acetate as the base, with values for ∆Sq of
35-53 eu.13b The unusual activation entropies were
interpreted in terms of a preequilibrium between the
radical cation and the base.13b Eyring analysis of the data
in Table 5 (not shown) yields activation entropies that
are all within 3 entropy units of zero, which is much
closer to expected values for bimolecular reactions,24 and
suggests that there is no need to invoke a complex
between the acid and the base, at least under the present
conditions.
Even though we have not made a detailed analysis of
the temperature-dependence data for the intramolecular
reactions, we can see some clear empirical trends. For
the ortho-substituted anilines, the intercepts of the
Arrhenius plots (indicated as log(A) in the tables) are
observed to be essentially the same within experimental
error, i.e., ca. 15.27 The differences in the rate constants
at room temperature for these anilines are thus empiri-
cally due to differences in the slopes of the Arrhenius
plots, Table 2. Deprotonation in the o-tert-butyl com-
pound 7e was too fast to measure at room temperature
under the experimental conditions, and in fact could only
be measured at one temperature, corresponding to the
single point for this compound in Figure 8. If we assume
that the temperature dependence for this reaction is the
same as for 7b-d, as indicated in Figure 8, then we can
estimate a rate constant at room temperature of ca. 4 ×
107 s-1, Table 2.
Corresponding temperature-dependent data for several
of the compounds with varying methylene chain length
are shown in Figure 9. Compared to the compounds in
Figure 8, which all have a six-membered transition state,
these reactions have different sized cyclic transition
states, which may explain the larger differences in the
intercepts of the Arrhenius plots in these cases. The
deprotonation rate constant for the most reactive com-
pound, 7f, was too high to measure at room temperature;
however, a value was obtained at 0 °C (the single point
in Figure 9). Assuming that the temperature dependence
of this reaction is the same as for 7g-j, as indicated in
Figure 9, then we can estimate a rate constant of ca. 9 ×
107 s-1 at room temperature, Table 3.
Determination of the absolute energetics of the proton-
transfer reactions would require knowledge of the pKa
of the acetate anion and the aniline radical cations under
the experimental conditions. As mentioned above, the pKa
of acetic acid increases dramatically from 4.75 in water
to 22.3 in acetonitrile as a consequence of hydrogen
bonding by water.25 The corresponding increase in pKa
of the radical cation of dimethylaniline is much smaller,
i.e. from 11.4 to 19.4.28 Thus, proton transfer from the
N,N-dimethylaniline radical cation to acetate is endo-
thermic in water by ca. 7 pKa units, and exothermic in
acetonitrile by ca. 3 pKa units. However, the exother-
micities of the reactions under the present conditions are
difficult to determine, since the extent to which specific
hydration reduces the basicity of the anionic acetate with
0.5 M water is not known.
As an illustration of the influence of water on the
basicity of the acetate anion, the rate constant for
deprotonation of the p-trifluoromethyl-N,N-dimethyl-
aniline radical cation, 1e, by acetate was measured in
acetonitrile as a function of added water.29 The bi-
molecular rate constant was ca. 6 × 109 M-1 s-1 with no
added water and decreased to 1.4 × 109 M-1 s-1 at 0.9%
water (0.5 M, see Table 1). The rate constant continues
to decrease with increasing water concentration, becom-
ing weakly dependent above ca. 4% water (2.2 M water),
where the rate constant is ca. 108 M-1 s-1 29
Similar
.
observations were made by Mariano et al.8a The rate
constant with 0.5 M water is ca. 1 order of magnitude
smaller than the diffusion-controlled limit, suggesting
that this reaction may be close to thermoneutral, al-
though it is not possible to say this with any certainty.
III. Summary
Deprotonation of aniline radical cations with an ex-
ternal base is typically slower than other fragmentation
reactions of aniline radical cations under normal condi-
tions, and is thus often less useful.8a Reactivity toward
deprotonation in N,N-dimethylanilines can be increased
by the use of ortho substituents, which twist the non-
bonding electrons on nitrogen out of conjugation with the
ring. Based on this stereoelectronic effect, a series of
aniline radical cations that undergo rapid unimolecular
(intramolecular) deprotonation have been designed with
rate constants for fragmentation controllable over 4
orders of magnitude, up to a value of ca. 108 s-1 in
aqueous acetonitrile at room temperature. Control over
the rate constant for fragmentation is important in the
two-electron sensitization application of amine radical
cations. The use of the intramolecular deprotonating
(25) Izutzu, K. Acid-Base Dissociation Constants in Dipolar Aprotic
Solvents; Blackwell Scientific: Oxford, UK, 1990.
(26) Smyth, C. P. Dielectric behavior and structure; dielectric
constant and loss, dipole moment and molecular structure; McGraw-
Hill: New York, 1955.
(27) The meanings of the empirical Ea and log(A) values are clearly
questionable if the basicity of the carboxylate is significantly temper-
ature dependent.
(28) (a) Based on the thermodynamic calculations, the pKa differ-
ences in water and acetontrile are principally due to differences in the
free energies of solvation and the standard oxidation potentials of the
hydrogen atom in the two solvents.28b (b) Wayner, D. D. M.; Parker,
V. D. Acc. Chem. Res. 1993, 26, 287.
(29) See Figure S4 in the Supporting Information.
J. Org. Chem, Vol. 70, No. 10, 2005 3799

Dombrowski et al.
ments, no change in the measured rate due to variation in ionic
strength is expected.
anilines described here as two-electron sensitizers in
silver halide photographic systems is described else-
where.30
The temperature dependencies of the rate constant for
deprotonation of the radical cations of p-methyl-N,N-dimethyl-
aniline, 1b, and 2,6-dimethyl-N,N-dimethylaniline, 4, were
determined both with 0.5 M water in acetonitrile and in
solvent mixtures that were designed to keep the polarity as
constant with temperature as possible. These latter experi-
ments were performed at 5, 25, 45, and 65 °C. The dielectric
constants of propionitrile at 5 °C and acetonitrile at 65 °C are
essentially identical, ca. 31-32.36 The absorption maximum
of Reichardt’s dye37 in acetonitrile with 0.5 M water at 65 °C
is 592 nm, and in propionitrile with 0.5 M water at 5 °C it is
605 nm. Mixtures of acetonitrile and propionitrile with 0.5 M
water at intermediate temperatures were determined empiri-
cally to give a solution of Reichert’s dye with absorption
maxima between these two values. At 25 °C, the solvent
mixture 60:40 propionitrile:acetonitrile with 0.5 M water gave
an absorption maximum at 596 nm. At 45 °C, the solvent
mixture 30:70 propionitrile:acetonitrile with 0.5 M water gave
an absorption maximum at 602 nm. Determination of the rate
constant for deprotonation at these four temperatures in these
solvents gave the Arrhenius data shown in Table 5.
The errors in the rate constants and Arrhenius activation
parameters are estimated to be less than 15%, based upon
repeated measurements.
The oxidation potentials of the anilines were measured by
using square-wave voltammetry, and details are as described
in ref 16. Measurements were performed in dry acetonitrile
with use of a 10 µm Pt working electrode. The reference
potential was established with use of the ferrocene/ferrocenium
couple. The oxidation potentials were then corrected to the
SCE electrode by the addition of 0.4 V.
Electronic structure calculations were performed with the
Spartan ‘02 software package (version 1.0.6),38 and the opti-
mizations were done by using the B3LYP DFT method with a
6-31G* basis set.
IV. Experimental Section
Acetonitrile was spectrograde and used as received. Most
of the para-substituted N,N-dimethylanilines were available
from a previous study.15 N,N-Dimethyl-o-toludine was obtained
commercially and was distilled before use. N,N-Dimethyl-p-
anisidine was prepared according to a literature procedure.31
The N,N-bis(trideuteriomethyl)anilines were prepared by ei-
ther methylation of the corresponding anilines with CD3I or
reductive alkylation.32 Tetra-n-butylammonium acetate and
tetra-n-butylammonium perchlorate were also obtained com-
mercially. The acetate was recrystallized from dry ethyl
acetate, dried in vacuo, and dispensed in a glovebox. The
perchlorate was used as received. 9,10-Dicyananthracene,
9-cyanoanthracene, biphenyl, and 1,4-dimethoxybenzene were
available from previous experimental work.17 The anilines with
covalently attached carboxylates were prepared as described
in the Supporting Information. Purity was checked with proton
NMR spectroscopy.
The radical cations were generated with a time-resolved
absorption apparatus and method that has been described in
detail elsewhere,16 using either 9-cyanoanthracene (9-CA) or
9,10-dicyanoanthracene (DCA) as the sensitizers, with either
1,2-dimethoxybenzene or biphenyl as cosensitizers, respec-
tively.16 In one case, o-tert-butyl-N,N-dimethylaniline, 3, the
N-methylquinolinium (NMQ)/toluene cosensitization system
was used, to ensure rapid and irreversible oxidation of this
sterically hindered molecule.33 All of the anilines used in this
work have lower oxidation potentials than that of the cosen-
sitizers (Eox ) 1.45 V vs SCE for 1,2-dimethoxybenzene, Eox
)
1.96 V vs SCE for biphenyl, and Eox e 2.3 V vs SCE for
toluene),16,34 thus electron transfer from the aniline to the
cosensitizer radical cation is energetically favorable and occurs
with a rate constant that is essentially diffusion controlled.
In all cases, the solutions were purged with oxygen. Oxygen
oxidizes the 9-CA and DCA radical anions and NMQ radical
by one-electron transfer to form the superoxide anion within
ca. 100 ns under the conditions of the experiments, which
would otherwise interfere with the absorptions of the radical
cations.6,34a
Acknowledgment. We thank the National Science
Foundation for research support (CHE-9812719 and
CHE-0213445) and Deniz Schildkraut (Eastman Kodak
Company) for performing the electrochemical measure-
ments.
Supporting Information Available: Synthetic details for
compounds 7 and 8, plots showing kobs vs [Bu4N+ -OAc] at
Experiments in which the rate constant for deprotonation
of the radical cation 1c•+ was measured at varying concentra-
tions of tetrabutylammonium perchlorate demonstrated that
at first the rate of the reaction decreases rapidly with increas-
ing salt concentration, but then becomes essentially indepen-
dent of salt concentration above 0.4 and 0.6 M (data provided
as Supporting Information, Figures S1 and S2).35 Since the
concentration of the acetate was e0.1 M in all of the experi-
various [Bu4N+ -ClO4] for 1c•+, the dependence of k-H as a
+
function of [Bu4N+ -ClO4] for 1c•+, the transient absorption
spectrum of 3•+ as a function of time, the dependence of k-H
+
for 1e•+ in acetonitrile on added water, Arrhenius plots for
deprotonation of radical cations 1•+ by acetate in acetonitrile
with 0.5 M water and 0.5 M tetrabutylammonium perchlorate,
and Cartesian coordinates for the minimum geometries of
radical cations 1•+-5•+. This material is available free of charge
via the Internet at http://pubs.acs.org.
(30) Gould, I. R.; Farid, S.; Godleski, S. A.; Lenhard, J. R.; Muenter,
A. A.; Zielinski, P. A. U.S. Patent 5 994 051, Nov. 30, 1999.
(31) Davies, W. C. Bull. Soc. Chim. Fr. 1935, 2, 295.
(32) Dinnocenzo, J. P.; Karki, S. B.; Jones, J. P. J. Am. Chem. Soc.
1993, 115, 7111.
(33) Dockery, K. P.; Dinnocenzo, J. P.; Farid, S.; Goodman, J. L.;
Gould, I. R.; Todd, W. P. J. Am. Chem. Soc. 1997, 119, 1876.
(34) (a) Guirado, G.; Fleming, C. N.; Lingenfelter, T. G.; Williams,
M. L.; Zuilhof, H.; Dinnocenzo, J. P. J. Am. Chem. Soc. 2004, 126,
14086. (b) Zweig, A.; Hodgson, W. G.; Jura, W. H. J. Am. Chem. Soc.
1964, 86, 4124.
JO047813G
(35) See also ref 8a.
(36) Wohlfarth, C. In CRC Handbook of Chemistry and Physics, 76th
ed.; Lide, D. R., Ed.; CRC Press: Baco Raton, FL, 1995; p 6-193.
(37) Reichardt, C. Solvents and Solvent Effects in Organic Chem-
istry, 2nd ed.; VCH: Weinheim, Germany, 1988; pp 285-311.
(38) Wavefunction, Inc., Irvine, CA.
3800 J. Org. Chem., Vol. 70, No. 10, 2005