The class II/III transition electron transfer on an infrared vibrational time scale for N,Na?2-diphenyl- 1,4-phenylenediamine structures

Article abstract of DOI:10.1021/jp0492265

Intramolecular electron transfer (ET) within the class II/III transition for the mixed-valence state of N,Na?2-diphenyl-1,4-phenylenediamine (PDA) derivatives which were substituted in the center or the outer phenyl ring and N,Na?2-diphenylbenzidine (BZ) was examined. Each compound showed two reversible redox couples. The splitting of the redox waves, ??E, is related to the interaction intensity between redox sites. The introduction of a substitutent into the central phenylene ring of the PDAs resulted in a decrease in ??E. A similar result was noted for the expansion of the distance between the redox centers such as in BZ. In opposition, the ??E of N,Na?2-bis(2,6-dimethylphenyl)-1,4-phenylenediamine (2,6-DMPDA) as a compound with substituents introduced into the outer phenyl rings was spread. The mixed-valence state of these compounds also exhibited an intervalence charge transfer (IV-CT) band in the near-IR region which provided the determination of the Marcus reorganization energy (??), the electron coupling (V), the thermal ET barrier (??G*), and the electron-transfer rate (k _th) using the Marcus-Hush theory. We first confirmed the electron-transfer rate of PDA derivatives in the class II/III transition state by two methods. The v(N-H) stretching vibrational spectra of the mixed-valence states were analyzed by a Bloch-type equation analysis using variable-temperature IR spectra measurements which were to be in good agreement with the those obtained from the Marcus-Hush theory. On the basis of this approach, the electron-transfer rate of PDA was determined to be 8.2 ?? 10¹¹ s^-1 at 298 K, yielding ??Ga?? = 420 cm ^-1 (the activation free energy from the Eyring plot) for the und erlying process.

Full text of DOI:10.1021/jp0492265

7992
J. Phys. Chem. B 2004, 108, 7992-8000
The Class II/III Transition Electron Transfer on an Infrared Vibrational Time Scale for
N,N′-Diphenyl-1,4-phenylenediamine Structures
Toyohiko Nishiumi, Yasuhiro Nomura, Yuya Chimoto, Masayoshi Higuchi, and
Kimihisa Yamamoto*
Department of Chemistry, Faculty of Science & Technology, Keio UniVersity, Yokohama 223-8522, Japan
ReceiVed: February 20, 2004; In Final Form: March 25, 2004
Intramolecular electron transfer (ET) within the class II/III transition for the mixed-valence state of N,N′-
diphenyl-1,4-phenylenediamine (PDA) derivatives which were substituted in the center or the outer phenyl
ring and N,N′-diphenylbenzidine (BZ) was examined. Each compound showed two reversible redox couples.
The splitting of the redox waves, ∆E, is related to the interaction intensity between redox sites. The introduction
of a substitutent into the central phenylene ring of the PDAs resulted in a decrease in ∆E. A similar result
was noted for the expansion of the distance between the redox centers such as in BZ. In opposition, the ∆E
of N,N′-bis(2,6-dimethylphenyl)-1,4-phenylenediamine (2,6-DMPDA) as a compound with substituents
introduced into the outer phenyl rings was spread. The mixed-valence state of these compounds also exhibited
an intervalence charge transfer (IV-CT) band in the near-IR region which provided the determination of the
Marcus reorganization energy (λ), the electron coupling (V), the thermal ET barrier (∆G*), and the electron-
transfer rate (kth) using the Marcus-Hush theory. We first confirmed the electron-transfer rate of PDA
derivatives in the class II/III transition state by two methods. The ν(N-H) stretching vibrational spectra of
the mixed-valence states were analyzed by a Bloch-type equation analysis using variable-temperature IR
spectra measurements which were to be in good agreement with the those obtained from the Marcus-Hush
theory. On the basis of this approach, the electron-transfer rate of PDA was determined to be 8.2 × 1011 s-1
at 298 K, yielding ∆Gq ) 420 cm-1 (the activation free energy from the Eyring plot) for the underlying
process.
Introduction
using the Marcus-Hush theory.12 To resolve these problems,
we employed IR spectra analysis, which is suitable for
confirmation of the results of IV-CT analysis, because the time
scale of the ET rate in the class II/III transition state reaches
picoseconds.15
The electron-transfer (ET) process is fundamentally one of
the most important processes in chemical reactions.1 Especially,
the construction of a fast ET system2 is considerably important
for the development of functional molecules possessing proper-
ties of electrical conductivity,3 ferromagnetism,4 and charge
separation.5 N,N′-Diphenyl-1,4-phenylenediamine6 (PDA), which
is a redox unit of polyaniline, as one of the most well-known
conducting polymers7 possesses a stable two-electron redox
couple in the presence of a Lewis acid. However, there have
been no reports on the physicochemical properties, such as
electron transfer and IV-CT in the mixed-valence (MV) state
of the polyaniline unit (e.g., the relationship between the
substitution effect and the conductivity of the polyaniline based
on IV-CT).
Organometallic and inorganic compounds, with MV states,
e.g., such as the Creutz-Taube ion8 and 1,4-pyrazine-bridged
ruthenium cluster complexes,9 have been analyzed using the
Marcus-Hush theory.10 The first direct investigation of ET rate
analysis was done by Ito and Kubiak et al. using a Bloch-type
equation which was independent of the major problems of the
MV state in class II/III transitions, such as the exact distance
between the redox centers due to charge delocalization or the
nuclear frequency factor (νn).9 Recently, many studies of organic
compounds in the MV state11-14 have been reported by Nelsen
et al. using both the Marcus-Hush theory and ESR measure-
ments.11 The discussion of ET has been expanded to the class
II/III transition state by Lambert et al. by simple IV-CT analysis
We herein report the analysis of the electron-transfer rate of
monocation radicals formed at the amine sites of PDA deriva-
tives using the Marcus-Hush theory in order to elucidate the
substituent effect at the outer N-phenyl or the inner phenylene
ring. The ET rate constant was determined by two methods.
One was IV-CT band analysis using the Marcus-Hush theory,
and the other was the ν(N-H) stretching absorption observed
in the temperature range from 193 to 298 K, which is analyzed
with regard to the ET rate by simulated dynamic effects (Bloch-
type equations).16 The important advantage of this IR method
lies in the independence from the concentration of the MV state
in the solution and the distance between two redox centers in
the MV compounds and an exact nuclear frequency factor.
IV-CT states are classified as classes I, II, and III on the basis
of the electronic coupling intensity between redox sites.17 The
fastest electron transfer takes place in the class II/III transition
state.2c,12a, No IV-CT band is observed for class I because the
ET coordinates are completely localized on only the diabatic
surfaces. For class II, the IV-CT absorption band is usually
observed in the near-IR region as being distorted from a
Gaussian shape in the low-energy region17 due to the cutoff
around the electric coupling. For class III, the electron coupling
is so strong that the redox centers are completely delocalized
* Corresponding author. E-mail yamamoto@chem.keio.ac.jp.
10.1021/jp0492265 CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/08/2004

N,N′-Diphenyl-1,4-phenylenediamine
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7993
CHART 1. Chemical Structures of PDA Derivatives
with Atom Labeling
Figure 1. Cyclic voltammograms of 1.0 mM PDA (solid curve) and
1.0 mM PDA-Br (dotted curve) in CH2Cl2 containing 0.2 M TBAP.
Scan rate: 0.1 V/s.
TABLE 1: Electrochemical Data for PDA Derivatives
compda
E°1b [V]
E°2c [V]
∆Ed [V]
Kco
e
PDA
0.14
0.16
0.27
0.48
0.33
0.27
0.35
0.67
0.57
0.63
0.71
0.55
0.48
0.55
0.525
0.412
0.360
0.231
0.219
0.213
0.204
7.4 × 108
9.1 × 106
1.2 × 106
8.0 × 103
5.0 × 103
4.0 × 103
2.8 × 103
PDA-Me
PDA-Ph
PDA-Br
PDA-Th
PDA-tBu
BZ
in the system, and collapsing of IV-CT absorption symmetry is
observed.2c,12a In this paper, the configuration of the IV-CT band
was compared by introducing a substituent into the inside or
the outside phenyl ring of PDA. We found that N,N′-bis(2,6-
dimethylphenyl)-1,4-phenylenediamine (2,6-DMPDA), which
was substituted in the outer phenyl rings, showed a redox center
completely delocalized class III property on the basis of the
solvent effect even though PDA was classified as a class II/III
transition compound.
a Cyclic voltammograms recorded in CH2Cl2 containing 0.2 M TBAP
at ca. 298 K. b 0/1+ couple vs Ag/Ag+. c 1+/2+ couple vs Ag/Ag+.
d E°2 - E°1. Kco ) 10(∆E/0.0592)
.
e
centers, the E°1, E°2, ∆E are 0.35, 0.55, 0.20. In the case of
PDA, the ∆E is 0.53 V, in which the comproportionation
constant19 (Kco) is calculated to be 9.6 × 108. A large Kco
indicates a high stability of the monocation radical state, which
is estimated to be a MV state (+1 and 0 state). The relationship
between the electrochemical value of ∆E and the electronic
coupling constant (V) has been discussed.20 In our case, a
monotonically increasing relationship was found except for
PDA-tBu•+. We simply treated the electrochemical processes
of PDA derivatives as an EE process (“electron transfer” is
abbreviated as E) because the solvent donor number of dichlo-
romethane is sufficiently lower, and in dichloromethane with
0.2 M TBAP, the second reversible redox couple was confirmed
without proton transfer. Generally, in the case of the two-step
redox reactions of PDA derivatives, the redox potentials and
the ∆E are dependent not on the Hammett rule such as a
substitution effect (electron-donating or -withdrawing nature),
but on the Nernst equation. The magnitude of ∆E is enlarged
by an electronic interaction such as Columbic repulsion through
the π-linked group but is made smaller by a conformational
change after electron removal based on steric hindrance and
the solvent dipole moment, resulting in the solvent reorganiza-
tion energy. When assuming that the series of the measurements
of the redox potentials of PDA derivatives is under the same
conditions, the twist of the orbital on the N atoms makes the
magnitude of ∆E small.
Spectroelectrochemistry. The MV states of PDA derivatives
were also confirmed by spectroelectrochemical (SEC) analysis
in dichloromethane containing 0.2 M TBAP. In the case of PDA,
the absorption band attributed to the monocation radical state
(νmax) appears at 14 100 cm-1 on applying a potential from 0.1
to 0.35 V (vs Ag/Ag+), and the absorption band disappeared
on applying a potential at ca. 0.8 V (vs Ag/Ag+), at which
potential the monocation radical species are further oxidized to
the dication (Figure 2). The νmax values of PDA-Ph, PDA-Th,
PDA-Me, PDA-tBu, PDA-Br, and BZ are at 13 900, 13 100,
14 800, 15 200, 13 300, and 8730 cm-1, respectively (Table 2).
Results and Discussion
The set of molecules that have a p-phenylenediamine moiety
twisted by the substituents is shown in Chart 1. The p-
phenylenediamine moiety was chosen so as to ensure a
reversible two-redox reaction with strong interaction between
the N atoms as redox centers and an accurate N-H stretching
vibration spectrum around 3400 cm-1. The syntheses of the
compounds are described in the Experimental Section and in
previous reports18 and were accomplished by palladium-
catalyzed C-C or C-N cross-coupling or the dehydration of
tBu-, methyl-, or bromo-substituted benzoquinone and aniline
in the presence of TiCl4 for the p-phenylenediamine moiety.
Because the electronic interaction intensities between the redox
centers are influenced dramatically by the twist of the highest
occupied molecular orbital (HOMO) on the N atoms, we
synthesized a series of compounds that have substitutions on
the inner or outer phenyl rings.
Electrochemistry. The cyclic voltammograms (CV) of inner-
phenyl-modified PDA derivatives show two reversible redox
processes in dichloromethane containing 0.2 M tetrabutylam-
monium perchlorate (TBAP). The CV of PDA and PDA-Br as
typical examples are shown in Figure 1. In the one-electron
oxidative state between the first and second redox potentials,
the cation radical is formed, and a stable IV-CT between two
amine sites takes place. All electrochemical results are shown
in Table 1. The first (E°1) and the second (E°2) half-wave
potential and the redox potential splitting (∆E) of nonsubstituted
PDA are 0.14, 0.67 (vs Ag/Ag+), and 0.53 V, respectively. The
E°1, E°2, and ∆E of PDA-Ph and PDA-Th, which are the
π-conjugated aryl-substituted PDA derivatives, are 0.27 V, 0.63
V, 0.36 V and 0.33 V, 0.55 V, 0.22 V, respectively. For PDA-
Me, PDA-tBu, and PDA-Br, which are nonconjugated-group-
substituted PDA derivatives, the E°1, E°2, ∆E are 0.16, 0.57,
0.41; 0.27, 0.48, 0.21; and 0.48, 0.71, 0.23, respectively. For
BZ, which has a different distance between the two redox

7994 J. Phys. Chem. B, Vol. 108, No. 23, 2004
Nishiumi et al.
1011, 7.6 × 1010, 1.3 × 1011, 1.1 × 1011, 3.9 × 1010, and 1.2 ×
1011 s-1, respectively (Table 2). Nonsubstituted PDA showed
the fastest ET. The ET rate decreases more efficiently with the
introduction of bulky substituent such as methyl and tert-butyl
due to the twisted orbitals on the N atoms rather than to the
increase in the distance between the two redox centers. The
electronic mixing coefficient R is an estimate of the electron
8d,12b
delocalization: R ) V/ν˜max
.
A plot of R vs kth, except for
BZ, gave a straight line (Figure 4); this result supports the idea
that the series of PDA derivatives have two local minima, and
these are not the MV state in the class III state but the class
II/III transition state.
Infrared Spectroelectrochemistry. In the case of PDA, the
ν(N-H) stretching absorption attributed to the monocation
radical was observed using IR-SEC21,22 in CH2Cl2/0.2 M TBAP
(Figure 5a). The absorption around 3420 cm-1 assigned to the
N-H stretching vibration in the neutral state decreases with
the applied potential from 0 up to 0.5 V vs Ag/Ag+. On the
other hand, a new absorption around 3247 cm-1 appeared. On
applying a potential over 0.8 V vs Ag/Ag+, this absorption
disappears. These results indicate that the absorption around
3247 cm-1 is attributed to the N-H vibration of PDA in the
monocation radical state. The intensity of the N-H vibration
signal in the monocation radical state is larger than that in neutral
state because the dipole moment of the N•+-H moiety is larger
than that of the N-H moiety. We ascertained the identification
of the absorption of this N-H vibration in the monocation
radical state around 3245 cm-1 by potential sweeping from both
directions on the oxidation side and on the reduction side. When
electrochemically reducing N,N′-diphenyl-1,4-benzoquinonedi-
imine (PDI), which is an oxidized form of PDA, in acetonitrile
containing 0.2 M TBAP and 25 mM trifluoroacetic acid (TFA),
a new absorption around 3245 cm-1, attributed to the N-H
vibration in the one-electron oxidized state, appeared at a
potential from 0.4 to 0.2 V vs Ag/Ag+. On applying a potential
under 0 V, the absorption disappears, and a new absorption
around 3376 cm-1 attributed to N-H stretching of the neutral
state appears. The IR-SEC results of PDA and PDI under the
same conditions are in accord; thus, the absorption around 3245
cm-1 is determined to be the N-H vibrational stretching in the
one-electron oxidized state. The adjacent weak signals (*) at
the low-energy side of the genuine signals are thought to be
the N-H vibration interacting with the supporting electrolyte,
such as TBAP, because these signals are shifted to the low-
energy side by a more basic supporting electrolyte such as
tetrabutylammonium tetrafluoroborate (TBABF4). Similar results
are obtained from the PDA-Br spectra (Figure 5b). The IR-
SEC spectra were fitted by the Voigt function,15 and the Voigt
profiles are collected in SI Tables 1 and 2. The shapes of these
signals are fitted by the Voigt function as a convolution of
Gaussian and Lorentzian function. These results indicate that
each spectrum consisted of a single peak by the deconvolution
of the spectrum the using the Voigt function because the
difference in the N-H vibration energy between the line
symmetry and the point symmetry of the molecular conforma-
tion is only about 2 cm-1 based on the MO frequency
calculation. In dichloromethane containing 0.2 M TBAP, the
N-H vibrational stretching absorption maxima in the MV state
of PDA•+, PDA-Me•+, PDA-Ph•+, PDA-Br•+, PDA-Th•+, and
Figure 2. Electronic absorption spectral data for PDA in neutral (dotted
curve), one-electron oxidized (solid curve), and two-electron oxidized
(dashed curve) state. Their potentials are 0.0, 0.3, and 0.8 V (vs Ag/
Ag+), respectively. Light path length: 1.0 mm; concentration: 1.0 mM
PDA, 0.2 M TBAP in CH2Cl2 at ca. 298 K.
Figure 3 shows the IV-CT bands of PDA•+ and PDA-Br•+ as
typical spectra and their fitted Gaussian curves. The resulting
bands are not symmetrical and possess a Gaussian shape in the
high-energy region, but the wave in the low-energy side did
not overlap with a Gaussian band shape. The reason why the
IV-CT band for the class II/III transition state is cut off at the
low-energy site was described by Lambert et al. in ref 12. These
results support the idea that the band is estimated to be an IV-
CT based on a MV state classified as a class II/III transition.
On the basis of the classical Marcus-Hush theory, electronic
coupling V (in cm-1) has been calculated by deconvolution of
the IV-CT band spectra as follows:10
2.06 × 10-2
V )
ν˜max ν˜1/2ꢀ
x
(1)
r
where ν˜max is the band energy in cm-1, ν˜1/2 is the bandwidth at
half-height in cm-1, ꢀ is the molar absorptivity, and r in Å is
taken as the distance between the interaction redox centers.
Equation 1 can cover only a true Gaussian-shaped curve within
a MV state in class II; two problems are conceivable: (1) The
exact distance between interacting redox centers is experimen-
tally obtained only for diabatic (interacting) redox centers. (2)
Because the vibrational levels are higher than those assumed
in class II, the IV-CT band shape collapses roughly from the
form of the Gaussian band. There is the possibility of under-
estimating V because of these problems; however, we obtained
V in this way in order to estimate the ET rate. When the
electronic coupling is sufficiently strong, the thermal ET barrier
∆G* and the thermal ET rate constant kth were estimated using
the following equations:10
λ
4
V2
λ
∆G* ) - V +
(2)
(3)
kth ) νn exp(-∆G*/RT)
where λ is the Marcus reorganization energy of the redox center
obtained as ν˜max and νn is the nuclear frequency factor, which
includes both the solvent and conformation reorganization
frequency required by charge transfer between the localized
valence state. The theoretical maximum ET rate constant for
an intramolecular ET can reach 1012-1013 s-1. We employ this
νn of PDA derivatives in dichloromethane as 9 × 1012 s-1
according to the previous literature by Lambert et al.12 The kth
values of PDA•+, PDA-Me•+, PDA-Ph•+, PDA-Br•+, PDA-Th•+,
PDA-tBu•+, and BZ•+ are calculated to be 2.4 × 1011, 1.1 ×
BZ•+ are 3247, 3257, 3333, 3320, 3322, and 3253 cm-1
,
respectively. All N-H vibrational spectra of the neutral and
MV state are summarized in Table 3. The intensity of the N-H
vibration signal in the monocation radical states is larger than
in the neutral state because the dipole moment of the N•+-H

N,N′-Diphenyl-1,4-phenylenediamine
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7995
TABLE 2: Band Shape Data for the IV-CT Band of PDA Derivatives in CH2Cl2 Containing 0.2 M TBAP at ca. 298 K
compd
PDA•+
ra/Å
ν
max/cm-1
ν1/2b/cm-1
ꢀc/M-1 cm-1
V d/cm-1
Re
∆G* f/cm-1
kthg/s-1
5.56
5.56
5.56
5.56
5.56
5.54
9.89
1.41 × 104
1.48 × 104
1.39 × 104
1.33 × 104
1.31 × 104
1.52 × 104
8.73 × 103
5.69 × 103
5.83 × 103
5.64 × 103
5.53 × 103
5.49 × 103
5.90 × 103
4.48 × 103
1.37 × 104
1.18 × 104
9.81 × 103
1.07 × 104
9.51 × 103
8.90 × 103
1.87 × 104
3.78 × 103
3.72 × 103
3.23 × 103
3.23 × 103
3.10 × 103
3.44 × 103
1.57 × 103
0.27
0.25
0.23
0.24
0.24
0.23
0.18
751
915
990
881
910
1130
896
2.4 × 1011
1.1 × 1011
7.6 × 1010
1.3 × 1011
1.1 × 1011
3.9 × 1010
1.2 × 1011
PDA-Me•+
PDA-Ph•+
PDA-Br•+
PDA-Th•+
PDA-tBu•+
BZ•+
a The N-N distance from an UB3LYP/6-31G* optimization (the 3-21G* basis set was substituted for 6-31G* in the case of the bromine atom).
b Observed bandwith at half-height. c For the highest absorption band given by [M+] ) [M+]0 K /(2 + K ), where [M+]0 ) [M]0 is the initial
x
x
co
co
and final concentration of M and of M+, respectively. d V ) (0.0206/r) ν˜maxν˜1/2ꢀ. e R ) V/ν˜max
.
f ∆G* ) λ/4 - V + V2/λ. kth ) νn exp(-∆G*/
g
x
RT), νn ) 9.0 × 1012 s-1, according to ref 12a.
Figure 3. IV-CT bands of PDA•+ (solid line) and PDA-Br•+ (dashed
line) from spectroelectrochemical analysis in CH2Cl2 containing 0.2
M TBAP at ca. 298 K. Light path length: 1.0 mm; Gaussian band fits
(dotted lines). Their potentials are 0.3 and 0.55 V (vs Ag/Ag+),
respectively.
Figure 4. Correlation of electron-transfer rate kth (s-1) vs electronic
mixing coefficient R for (b) PDA derivatives and ([) BZ.
Figure 5. ν(N-H) vibration spectra of several potentials of (a) PDA
and (b) PDA-Br in CH2Cl2 containing 0.2 M TBAP at ca. 298 K. Light
path length: 1.0 mm; Voigt function band fits (O). An Ag/Ag+ electrode
is used as the reference. The marked peaks (*) are assigned to the
ν(N-H) vibration interaction with the supporting electrolyte (TBAP)
as minor peaks.
moiety is larger than that of the N-H moiety. No N-H
stretching spectra of two-electron-oxidized PDA derivatives in
the dication state are detected in the area from 3500 to 3200
cm-1
.
Temperature Effect on ν(N-H) and the Electron-Transfer
Rate Constant. If the single peak which was obtained at room
temperature in the IR spectra measurement was separately
observed completely at low temperature due to coalescence, we
can obtain the ET rate estimation using a Bloch-type equation
on the IR spectra time scale as a second method. The variable-
temperature IR spectra of the amines in PDA in the monocation
radical state were determined from the temperature-controlled
IR spectra in a quartz 1 mm cell (Figure 6a). The absorption
band of PDA in the monocation radical state was observed at
298 K as a single peak due to the coalescence between the two
kinds of N-H vibrational absorptions attributed to the neutral
and monocation radical states. The single peak at 3247 cm-1 is
gradually separated into two peaks at 3275 and 3243 cm-1 on
decreasing the temperature to 193 K. The absorption band
analysis of PDA-Br in the monocation radical state gave the
TABLE 3: Spectroelectrochemical Analysis of ν(N-H)
Stretching Vibration
ν(N-H) in the
ν(N-H) in the monocation
∆ν
compd neutral state (cm-1
)
radical state (cm-1
)
(cm-1
)
PDA
3420
3414
3413
3401
3404
3421
3247
3257
3333
3320
3322
3253
173
157
80
81
82
PDA-Me
PDA-Ph
PDA-Br
PDA-Th
BZ
168
ET rate by the same procedure and was observed at 298 K as
a single peak at 3320 cm-1 and at 193 K as two peaks at 3342
and 3315 cm-1 (Figure 6b). No division of the peaks around
3420 cm-1 assigned to the N-H stretching vibration in PDA
in the neutral state was observed in each temperature from 298

7996 J. Phys. Chem. B, Vol. 108, No. 23, 2004
Nishiumi et al.
TABLE 4: Selected Geometrical Parameters in the
Monocation Radical State
dihedral angle (deg)a
compdb
C1C2N3C4
C2N3C4C5
C6C2N3H7
c
d
PDA•+
13.2
13.0
14.3
16.7
17.1
12.3
5.8
36.2
39.1
37.2
37.2
40.4
37.0
12.9
13.8
11.3
14.4
16.5
12.0
11.1
25.8
38.2
PDA-Me•+
PDA-Ph•+
PDA-Th•+
PDA-tBu•+
BZ•+
2,6-DMPDA•+
PDA
2,6-DMPDA
64.7
32.7 (29.9)e
2.3 (9.1)f
18.9 (23.9)e
70.8 (70.1)f
a Computed at the DFT UB3LYP/6-31G* level. b Atomic labels are
shown in Chart 1. c The inside C2-N3 bond. d The outside N3-C4 bond.
f
e From the X-ray structural analysis in ref 6c. From the X-ray structural
analysis.
energy (∆Gq) with which a Bloch-type equation plot and the
temperature-dependent IR spectra most agree. The ∆Gq of PDA
and PDA-Br were determined to be 420 and 522 cm-1
,
respectively (SI Figure 4). The ket values between the N-H
neutral and N•+-H at room temperature of PDA•+ and PDA-
Br•+ were estimated to be 8.2 × 1011 and 5.0 × 1011 s-1 (Figure
6), which are in good agreement with the order from the IV-
CT band analysis using the Marcus-Hush theory.
The results for ET from the Marcus-Hush theory (kth) are
smaller than that from the Bloch-type equation analysis (ket)
due to the overestimation of the distance between redox centers
(r) or the underestimation of the nuclear frequency factor (νn).
By making typical assumptions, we estimated the r′ from the
kEX using Marcus-Hush theory (νn ) 9.0 × 1012 s-1),12a and
the r′ value for PDA•+ and PDA-Br•+ are 4.79 (5.56) and 4.70
(5.56) Å (calculated), respectively. These results show a
reasonable underestimate of the electron coupling constant (V)
in the class II/III transition state by a factor of the 0.85-0.86.
Electronic Structure Calculations. Reasonable molecular
structure estimations for the PDA derivatives in the neutral and
MV states were provided by the UB3LYP/6-31G* level using
the Gaussian98 package.23-25 Because we are interested in the
location of the point symmetric geometry configuration (as a
reference configuration),14 the optimization of the monocation
radical states has been carried out with symmetry constraints.
The selected geometric parameters are summarized in Table 4.
The linear correlation between the experimental value of ∆E
in Table 1 and the dihedral angle of the C2-N3 bond inside in
the monocation radical state was found because the ∆E indicates
the magnitude of the interaction between the redox sites. To
increase the electronic coupling, the HOMO orbital, which
participates in the electron transfer, should be concentrated on
the phenyl ring at the center. The introduction of a nonconju-
gated bulky group, such as methyl or tert-butyl, to the center
phenyl ring causes the whole molecular structure to be out-of-
plane due to the twist of the inside C2-N3 bond. As a result,
the electric interaction between N atoms through the π-bond
becomes weak. The introduction of a π-conjugated group such
as phenyl or thienyl, which are expected to function as an
electron-donating group, does not result in an increase with the
HOMO orbital level on the phenyl ring at the center but also
twists the whole molecule. Because the electron densities of
the phenyl ring at the center decline mainly due to twisting of
the inside C2-N3 bond by the steric hindrance of a substituent
in the phenylene ring, the ET rate thus decreases.
Figure 6. (a) ν(N-H) vibrational in the variable-temperature IR spectra
of (a) PDA•+ and (c) PDA-Br•+. The recorded conditions are as
follows: (a) in CH2Cl2 solution containing 0.2 M TBAP at 193/203/
213/233/253/273/298 K; (c) in CD2Cl2 solution containing 0.2 M TBAP
at 193/213/233/253/273/298 K. Relative simulated deconvolution
spectra using the Bloch-type equation as a function of the intramolecular
ET rate constant (ket) of (b) PDA•+ and (d) PDA-Br•+
.
to 193 K (SI Figure 1). The simulated line widths are narrower
than the observed ones because the simulated ones consist of
the Lorentzian functions based on the well-established methods
of dynamic NMR spectroscopy,16 but observed ones are obtained
as the Voigt functions, which are broader than the simple
Lorentzian functions due to the IR absorptions in liquid-phase
systems.
Some reasons why the peak could separate clearly into two
in this way are considered. (1) The gap in the vibration spectra
(∆ν) between N-H and N•+-H is sufficiently widely separated
to be observed. (2) The electron coupling is weak so that the
compounds have two local minima in the valence trapped or
the charge localized state, and the electron-transfer rate is slower
than the IR time scale at low temperature due to electron
trapping at the amine site. The ET rate at room temperature
(298 K) is calculated as follows. Each IR spectrum was analyzed
by curve fitting to the Voigt Function shown in SI Figures 2
and 3. The information on the peak position and width was used
for the analysis. We estimated the ET rate at room temperature
by extrapolation of the Eyring plot using the activation free
In the case of PDA, which showed the fastest ET rate, the
inside C2-N3 bond is hardly twisted and the lobe of n electrons
on the N atoms are aligned in line in the same plane across the

N,N′-Diphenyl-1,4-phenylenediamine
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7997
Figure 8. IV-CT bands of PDA•+ with 0.5 M TFA (solid line) and
2,6-DMPDA•+ with 0.63 M TFA (dashed line) from spectroelectro-
chemical analysis in acetonitrile containing 0.2 M TBABF4 at ca. 298
K. Light path length: 1.0 mm; Gaussian band fits (dotted lines). Fitting
for 2,6-DMPDA•+ in high-energy side is impossible. Their potentials
are 0.30 V (vs Ag/Ag+).
to twisting of the outer C-N bond by steric hindrance.
Comparing the HOMO orbital of PDA with that of 2,6-DMPDA,
the HOMO electron density of 2,6-DMPDA is higher than that
of PDA (Figure 7a,b).
Electrochemistry and the UV-vis SEC of Outer Phenyl-
Modified PDA Derivatives. The introduction of a substitutent
into the central phenylene ring of PDA resulted in decreases in
the electron density on the amine atoms due to twisting of the
inner C-N bond of the homo orbital, in which the ET between
the two amines is suppressed. The cyclic voltammograms of
2,6-DMPDA showed two reversible redox processes in aceto-
nitrile containing 0.2 M tetrabutylammonium tetrafluoroborate
(TBABF4) and 0.63 M trifluoroacetic acid (TFA).26 The E°1,
E°2, and ∆E of PDA with 0.5 M TFA are 0.20 V, 0.48 V (vs
Ag/Ag+), and 0.28 V, respectively (SI Figure 5). The ∆E of
2,6-DMPDA, which is an ortho-dimethyl-substituted PDA, is
drastically increased. This result shows that the interaction
between redox sites is increased. The ∆E of 2,6-DMPDA in
the CV was larger than that of PDA. This result suggests a
higher interaction in 2,6-DMPDA between the two amines than
that in PDA, which shows the fastest ET rate among the PDA
derivatives.
Figure 7. Contour plots of the HOMO R molecular orbitals of (a)
PDA•+ and (b) 2,6-DMPDA•+ as obtained at the B3LYP/6-31G* level.
ORTEP diagrams of (c) 2,6-DMPDA in the neutral state.
phenyl at the center, so that the interaction of the orbital is
strong. However, the HOMO orbital is seen to intrude into the
outer phenyl rings (Figure 7a).
On the basis of these results, we designed 2,6-DMPDA which
is substituted in the outside phenyl ring. In the case of N,N′-
bis(2,6-dimethylphenyl)-1,4-phenylenediamine (2,6-DMPDA),
because the twist of the outside N3-C4 bond is large, due to
steric hindrance, the n electrons at the HOMO orbital intruding
into the outer phenyl are suppressed, and the interaction with
the outside phenyl is small. On the other hand, because the twist
of the inside C2-N3 bond is small, the n electrons on the N
atoms are aligned approximately in the same plane with the
phenyl ring at the center, and the concentration of the HOMO
orbital on the phenyl ring at the center occurs so that the
electronic interaction increases. There is little intrusion of the
HOMO orbitals outside of the phenyl compared with PDA, and
the electronic coupling between amines is expected to increase
more than in PDA.
We compared the structures between 2,6-DMPDA and PDA
in the monocation radical state from the viewpoint of “the twist
of the N orbitals” as two dihedral angles of inner and outer
C-N bonds. The twist of the inside C2-N3 bond of 2,6-
DMPDA•+ is smaller than that of PDA•+ (Table 4). Moreover,
the dihedral angles of the outer phenyl and N in PDA•+ and
2,6-DMPDA•+ are 36.2° and 64.7°, respectively (Figure 7a,b)
and so prevent the intrusion of the electrons of the N atom into
the outer phenyl. These results support the fact that it is easy to
localize the HOMO orbital on the phenyl ring at the center.
The X-ray structural analysis in the neutral state (Figure 7c)
and electronic structure calculations also support these results.
The n electrons on the N atom overlap the π electrons of the
inner phenyl ring on the same plane, whose structure prevented
intrusion of the HOMO orbital into the outer N-phenyl ring due
Under the same CV conditions, the IV-CT band of 2,6-
DMPDA•+ in the MV state increased around 555 nm (18 000
cm-1) from a potential of 0.05 to 0.3 V vs Ag/Ag+ from the
UV-vis SEC in acetonitrile (Figure 8). The νmax of PDA•+ is
at 698 nm (Figure 8). The high-energy side of the absorption
spectrum after one-electron oxidation did not fully agree with
the Gaussian curve for 2,6-DMPDA•+ because the large electron
coupling (when 2V > λ) pushes up a cutoff to the high-energy
side of the Gaussian curve.2c The electron coupling V, the
thermal ET barrier ∆G*, and the thermal ET rate constant kth
were estimated using the Marcus-Hush theory by analysis of
the absorption wave shapes. The kth of PDA•+ in acetonitrile
was 3.1 × 1011 s-1 27
.
On the other hand, the mechanically
calculated kth of 2,6-DMPDA•+ was 4.0 × 109 s-1 27
. This result
suggests that the 2,6-DMPDA•+ is in class III, and so the
Marcus-Hush theory does not apply.
Solvent Effect. The wavelength of the absorption maximum
νmax of the IV-CT band for 2,6-DMPDA•+ is independent of
the solvents. The correlation of νmax and (1/n2 - 1/D)8c,28 for
PDA•+ has a slope, but the νmax of 2,6-DMPDA does not obey
the equation λ ) (q2/8πD0)(1/a)(1/n2 - 1/D) from the Marcus
theory due to the electron delocalization in a class III state
(Figure 9). This means that the absorption maximum based on
the optical diabatic path as the ET path is greater than the
Marcus reorganization. The coalescence analysis of 2,6-DMPDA

7998 J. Phys. Chem. B, Vol. 108, No. 23, 2004
Nishiumi et al.
4H), 6.66 (t, J ) 7.3 Hz, 2H), 2.09 (s, 6H). 13C NMR (100
MHz, DMSO-d6, TMS standard, ppm) δ: 146.1, 135.9, 129.1,
128.9, 124.3, 117.6, 114.6, 17.7. IR (KBr, cm-1): 3396 (N-
H). MALDI-TOF-mass calcd: 288.4 [M+]. Found: 287.2.
Anal. Calcd for C20H20N2: C, 83.30; H, 6.99; N, 9.71. Found:
C, 83.33; H, 6.87; N, 9.74.
Synthesis of 2,5-Di-tert-butyl-N,N′-diphenyl-1,4-phenylene-
diamine (PDA-tBu). Aniline (9.48 g, 102 mmol) was dissolved
in chlorobenzene (40 mL) under nitrogen, and titanium(IV)
tetrachloride (3.40 g, 17.9 mmol) was added dropwise at 90
°C. The addition funnel was then rinsed with chlorobenzene
(20 mL). 2,5-Dimethyl-1,4-benzoquinone (1.5 g, 6.81 mmol)
dissolved in chlorobenzene (100 mL) was added dropwise over
15 min to the mixture. The mixture was stirred in an oil bath at
125 °C for 12 h. The precipitate was removed by filtration and
washed with chlorobenzene (20 mL). The filtrate was concen-
trated, and the compound PDA-tBu (0.406 g, 16%, white
powder) was isolated by silica gel chromatography (hexane:
Figure 9. Solvent effect of (O) 2,6-DMPDA•+ and (0) PDA•+ for
IV-CT band νmax recorded in PhCl, CHCl3, CH2Cl2, DMSO, DMF, and
MeCN.
in the temperature-dependent IR spectra measurement for a
Bloch-type equation analysis was unfortunately not observed
because complete delocalization of the monocation radical
occurred.
t
dichloromethane ) 2:1). PDA- Bu: 1H NMR (400 MHz,
CD3CN, TMS standard, ppm) δ: 7.24 (s, 2H), 7.16 (dd, J )
7.8, 7.8 Hz, 4 H), 6.72-6.69 (m, 6H), 5.82 (s, 2H), 1.30 (s,
18H). 13C NMR (100 MHz, CD3CN, TMS standard, ppm) δ:
148.5, 145.0, 138.1, 130.0, 128.2, 118.7, 115.3, 35.2, 30.8. IR
(KBr, cm-1): 3389 (N-H). MALDI-TOF-mass calcd: 372.6
[M+]. Found: 370.9. Anal. Calcd for C26H32N2: C, 83.82; H,
8.66; N, 7.52. Found: C, 83.70; H, 8.59; N, 7.51.
Conclusion
The electron-transfer rates of PDA derivatives in the MV state
were investigated using two methods: the Marcus-Hush theory
and a Bloch-type equation analysis. These results are in good
agreement with each other. The electronic coupling intensity
between the two amines is fixed by the twist of n electrons on
the N atoms based on the substituent effect. We first demon-
strated the PDA derivatives of IV-CT in classes II and III by
the introduction of a substituent into the inner or outer phenyl
rings.
Synthesis of N,N′-Bis(2,6-dimethylphenyl)-1,4-phenylenedi-
amine (2,6-DMPDA) Via N,N′-Bis(2,6-dimethylphenyl)-1,4-
benzoquinonediimine. 2,6-Dimethylaniline (0.73 g, 6.0 mmol)
and triethylamine (1.21 g, 12.0 mmol) were dissolved in
chlorobenzene (15 mL) under argon, and titanium(IV) tetra-
chloride (0.85 g, 4.5 mmol) was added dropwise at 60 °C. The
addition funnel was then rinsed with chlorobenzene (1 mL).
1,4-Benzoquinone (0.34 g, 3.0 mmol) dissolved in chloroben-
zene (7 mL) was added dropwise over 15 min to the mixture.
The mixture was stirred in an oil bath at 60 °C for 2 h. The
precipitate was removed by filtration and washed with chlo-
robenzene (5 mL). The filtrate was concentrated, and the
compound N,N′-bis(2,6-dimethylphenyl)-1,4-benzoquinonedi-
imine (0.69 g, 74%, orange powder) was isolated by silica gel
chromatography (CHCl3). N,N′-Bis(2,6-dimethylphenyl)-1,4-
benzoquinonediimine: 1H NMR (400 MHz, CDCl3, TMS
standard, ppm) as mixture of two syn-anti isomers, δ: 7.20
(m, 1H), 7.08-6.91 (m, 7H), 6.48 (dd, J ) 2.4, 10.3 Hz, 1H),
6.23 (m, 1H), 2.01 (s, 7H), 1.97 (s, 5H). 13C NMR (100 MHz,
CDCl3, TMS standard, ppm) as mixture of two syn-anti
isomers, δ: 158.9, 158.8, 148.0, 147.9, 136.7, 135.9, 127.8,
127.7, 125.7, 124.7, 124.5, 124.0, 123.9, 18.2. MALDI-TOF-
mass calcd: 315.4 [M + 1+]. Found: 315.1.
Experimental Section
General. NMR spectra were recorded on a JEOL JMN400
FT-NMR spectrometer (400 MHz) in DMSO-d6, CD3CN,
CDCl3 or THF-d8 + TMS internal standard) solution. MALDI-
TOF-mass spectra were obtained on a Shimadzu/Kratos KO-
MPACT MALDI mass spectrometer (positive mode; matrix:
dithranol). The infrared spectrum was obtained with a potassium
bromide pellet using a JASCO FT-IR-460Plus.
Materials. N,N′-Diphenyl-1,4-phenylenediamine (PDA) was
purchased from Aldrich Chemical Co., Inc. 2,5-Dimethyl-1,4-
benzoquinone and 2,5-di-tert-butyl-1,4-benzoquinone were from
Tokyo Kasei Co., Ltd. N,N′-Diphenylbenzidine (BZ) and all
other chemicals were purchased from Kantoh Kagaku Co. The
following compounds were prepared according to literature
procedures: 2,5-dibromo-N,N′-diphenyl-1,4-phenylenediamine
(PDA-Br),18 2,5-diphenyl-N,N′-diphenyl-1,4-phenylenediamine
(PDA-Ph),18 N,N′-diphenyl-2,5-di-thiophen-2-yl-1,4-phenylene-
diamine (PDA-Th).18
N,N′-Bis(2,6-dimethylphenyl)-1,4-benzoquinonediimine (0.314
g, 1.00 mmol) was dissolved in acetonitrile (30 mL) under
nitrogen, and tin(II) dichloride (1.21 g, 15 mmol) and trifluo-
roacetic acid (1.7 g, 15 mmol) were added. The mixture was
stirred for 13.5 h at room temperature, and triethylamine (1.5
g, 15 mmol) was added. The filtrate was concentrated, and the
compound 2,6-DMPDA (0.307 g, 97%, white powder) was
isolated by silica gel chromatography (hexane:THF ) 2:1). 2,6-
DMPDA: 1H NMR (400 MHz, THF-d8, TMS standard, ppm)
δ: 8.81 (d, J ) 7.3 Hz, 4H), 8.70 (t, J ) 7.3 Hz, 2H), 8.11 (s,
4H), 7.79 (s, 2H), 3.95 (s, 12H). 13C NMR (100 MHz, THF-d8,
TMS standard, ppm) δ: 140.3, 139.0, 134.5, 128.1, 124.0, 114.9,
17.8. IR (KBr, cm-1): 3396 (N-H). MALDI-TOF-mass
calcd: 316.4 [M+]. Found: 316.2. Anal. Calcd for C22H24N2:
C, 83.50; H, 7.64; N, 8.85. Found: C, 83.32; H, 7.67; N, 8.70.
Syntheses. Synthesis of 2,5-Dimethyl-N,N′-diphenyl-1,4-phe-
nylenediamine (PDA-Me). Aniline (9.48 g, 102 mmol) was
dissolved in chlorobenzene (40 mL) under nitrogen, and
titanium(IV) tetrachloride (3.66 g, 19.3 mmol) was added
dropwise at 90 °C. The addition funnel was then rinsed with
chlorobenzene (20 mL). 2,5-Dimethyl-1,4-benzoquinone (1.00
g, 7.3 mmol) dissolved in chlorobenzene (100 mL) was added
dropwise over 15 min to the mixture. The mixture was stirred
in an oil bath at 125 °C for 12 h. The precipitate was removed
by filtration and washed with chlorobenzene (20 mL). The
filtrate was concentrated, and the compound PDA-Me (0.99 g,
47%, white powder) was isolated by silica gel chromatography
(hexane:dichloromethane ) 1:1). PDA-Me: 1H NMR (400
MHz, DMSO-d6, TMS standard, ppm) δ: 7.27 (s, 2H), 7.12
(dd, J ) 7.3, 7.8 Hz, 4 H), 7.00 (s, 2H), 6.75 (d, J ) 7.8 Hz,

N,N′-Diphenyl-1,4-phenylenediamine
J. Phys. Chem. B, Vol. 108, No. 23, 2004 7999
(2) (a) Williams, R. D.; Hupp, J. T.; Ramm, M. T.; Nelsen, S. F. J.
Phys. Chem. A 1999, 103, 11172-11180. (b) Nelsen, S. F.; Trieber, D. A.,
II; Nagy, M. A.; Konradsson, A.; Halfen, D. T.; Splan, K. A.; Pladziewicz,
J. R. J. Am. Chem. Soc. 2000, 122, 5940. (c) Nelsen, S. F. Chem.sEur. J.
2000, 6, 581. (d) Nishikitani, Y.; Kobayashi, M.; Uchida, S.; Kubo, T.
Electrochim. Acta 2001, 46, 2035.
Electrochemical Measurements. Electrochemical analyses
were performed using an electrochemical workstation (BAS Co.,
Ltd., model ALS-660) under the following conditions. Cyclic
voltammetry (CV) was carried out under a N2 atmosphere. A
glassy carbon (GC) disk electrode (0.071 cm2) was used as the
working electrode and was polished with 0.3 µm alumina before
the experiments. The auxiliary electrode was a coiled platinum
wire. The reference electrode was a commercial Ag/Ag+ one,
which was placed in the main cell compartment. The formal
potential of the ferrocene/ferrocenium couple was 0.076 V vs
this reference electrode in acetonitrile. All potentials are quoted
with respect to this Ag/Ag+ reference electrode. The potential
was normalized to the ferrocene/ferrocenium couple in aceto-
nitrile. The scanning rate was 100 mV/s. In all cases, a 0.2 M
solution of tetrabutylammonium perchlorate (TBAP) in dichlo-
romethane was used. These PDA derivatives underwent a two-
step oxidation from the amine to the monocation radical to the
imine forms. In the case of using 0.2 M solution of TBABF4 in
acetonitrile without trifluoroacetric acid (TFA), the second redox
couple was not reversible because the oxidative half-reaction
of the second couple involves the loss of one electron and two
protons, while the cathodic half-reaction entails the gain of one
electron and two protons. However, in dichloromethane with
0.2 M TBAP, the two reversible redox couples were confirmed
without proton transfer because the solvent donor number of
dichloromethane is lower than that of acetonitrile.
(3) Venugopal, G.; Quan, X.; Johnson, G. E.; Houlihan, F. M.; Chin,
E.; Nalamasu, O. Chem. Mater. 1995, 7, 271.
(4) van Meurs, P. J.; Janssen, R. A. J. J. Org. Chem. 2000, 65, 5712.
(5) (a) Asahi, T.; Ohkohchi, M.; Matsusaka, R.; Mataga, N.; Zhang,
R. P.; Osuka, A.; Maruyama, K. J. Am. Chem. Soc. 1993, 115, 5665. (b)
Ohkohchi, M.; Takahashi, A.; Mataga, N.; Okada, T.; Osuka, A.; Yamada,
H.; Maruyama, K. J. Am. Chem. Soc. 1993, 115, 12137. (c) Imahori, H.;
Tamaki, K.; Guldi, D. M.; Luo, C.; Fujitsuka, M.; Ito, O.; Sakata, Y.;
Fukuzumi, S. J. Am. Chem. Soc. 2001, 123, 2607. (d) Imahori, H.; Guldi,
D. M.; Tamaki, K.; Yoshida, Y.; Luo, C.; Sakata, Y.; Fukuzumi, S. J. Am.
Chem. Soc. 2001, 123, 6617.
(6) (a) Wolf, J. F.; Forbes, C. E.; Gould, S.; Shacklette, L. W. J.
Electrochem. Soc. 1989, 136, 2887. (b) Boyer, M. I.; Quillard, S.; Louarn,
G.; Froyer, G.; Lefrant, S. J. Phys. Chem. B 2000, 104, 8952. (c) Boyer, I.;
Quillard, S.; Corraze, B.; Deniard, P.; Evain, M. Acta Crystallogr. 2000,
C56, e159. (d) Nishiumi, T.; Nomura, Y.; Higuchi, M.; Yamamoto, K.
Chem. Phys. Lett. 2003, 378, 18.
(7) (a) Li, X. G.; Huang, M. R.; Duan, W.; Yang, Y. L. Chem. ReV.
2002, 102, 2925. (b) Yamamoto, K.; Taneichi, D. Macromol. Chem. Phys.
2000, 201, 6.
(8) (a) Creutz, C.; Taube, H. J. Am. Chem. Soc. 1969, 91, 3988. (b)
Creutz, C.; Taube, H. J. Am. Chem. Soc. 1973, 95, 1086. (c) Creutz, C.
Prog. Inorg. Chem. 1983, 30, 1. (d) Brunschwig, B. S.; Creutz, C.; Sutin,
N. Chem. Soc. ReV. 2002, 31, 168.
(9) (a) Ito, T.; Hamaguchi, T.; Nagino, H.; Yamaguchi, T.; Washington,
J.; Kubiak, C. P. Science 1997, 277, 660. (b) Ito, T.; Hamaguchi, T.; Nagino,
H.; Yamaguchi, T.; Zavarine, I. S.; Richmond, T.; Washington, J.; Kubiak,
C. P. J. Am. Chem. Soc. 1999, 121, 4625-4632.
IV-CT Spectra Measurement. Controlled-potential absorp-
tion spectra were obtained with an optically transparent thin-
layer electrode quartz cell (light path length ) 1 mm). The work-
ing and the counter electrodes were platinum mesh and a platin-
um coil, respectively. The potential was applied with a Hokuto-
Denko potentio/galvanostat, model HA-501G, and referred to
Ag/Ag+. The spectra were measured with a Shimadzu UV-
3150PC spectrophotometer. All spectroelectrochemical measure-
ments were carried out at ca. 298 K under a nitrogen atmosphere.
IR-Spectroelectrochemistry. Controlled-potential absorption
spectra of ν(N-H) vibration spectra were obtaind using the same
electrochemical apparatus as in the IV-CT spectra measurement.
The spectra were measured with a JASCO FT-IR-460Plus.
Variable-Temperature IR Spectra of the Amines in the
Mixed-Valence State. Controlled-temperature absorption spec-
tra were obtained with 1 mM PDA•+/PDA-Br•+ in CH2Cl2/CD2-
Cl2 with 0.2 M TBAP prepared by electrochemical one-electron
oxidation in a quartz cuvette (light path length ) 1 mm). The
variable temperature was supplied by an OXFORD cryostat,
model Optistat DN. The spectra were measured with a JASCO
FT-IR-460Plus.
(10) (a) Hush, N. S. Prog. Inorg. Chem. 1967, 8, 391. (b) Sutin, N.
Prog. Inorg. Chem. 1983, 30, 441. (c) Brunschwig, B. S.; Sutin, N. In
Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-VCH: New York,
2001; Vol. 2, p 583. (d) Brunschwig, B. S.; Creutz, C.; Sutin, N. Chem.
Soc. ReV. 2002, 31, 168.
(11) (a) Nelsen, S. F.; Chang, H.; Wolff, J. J.; Adamus, J. J. Am. Chem.
Soc. 1993, 115, 12276. (b) Nelsen, S. F.; Adamus, J.; Wolff, J. J. J. Am.
Chem. Soc. 1994, 116, 1589. (c) Nelsen, S. F.; Ismagilov, R. F.; Powell,
D. R. J. Am. Chem. Soc. 1996, 118, 6313. (d) Nelsen, S. F.; Ramm, M. T.;
Wolff, J. J.; Powell, D. R. J. Am. Chem. Soc. 1997, 119, 6863. (e) Nelsen,
S. F.; Ismagilov, R. F.; Powell, D. R. J. Am. Chem. Soc. 1997, 119, 10213.
(f) Nelsen, S. F.; Tran, H. Q.; Nagy, M. A. J. Am. Chem. Soc. 1998, 120,
298. (g) Nelsen, S. F.; Ismagilov, R. F.; Powell, D. R. J. Am. Chem. Soc.
1998, 120, 1924. (h) Nelsen, S. F.; Ismagilov, R. F.; Teki, Y. J. Am. Chem.
Soc. 1998, 120, 2200. (i) Nelsen, S. F.; Ismagilov, R. F.; Gentile, K. E.;
Powell, D. R. J. Am. Chem. Soc. 1999, 121, 7108. (j) Nelsen, S. F.; Trieber,
D. A.; Ismagilov, R. F.; Teki, Y. J. Am. Chem. Soc. 2001, 123, 5684. (k)
Zwier, J. M.; Brouwer, A. M.; Keszthelyi, T.; Balakrishnan, G.; Offersgaard,
J. F.; Wilbrandt, R.; Barbosa, F.; Buser, U.; Amaudrut, J.; Gescheidt, G.;
Nelsen, S. F. J. Am. Chem. Soc. 2002, 124, 159.
(12) (a) Lambert, C.; No¨ll, G. J. Am. Chem. Soc. 1999, 121, 8434. (b)
Lambert, C.; No¨ll, G. Angew. Chem., Int. Ed. 1998, 37, 2107.
(13) Lindeman, S. V.; Rosokha, S. V.; Sun, D.; Kochi, J. K. J. Am.
Chem. Soc. 2002, 124, 843.
(14) Coropceanu, V.; Malagoli, M.; Andre, J. M.; Bredas, J. L. J. Am.
Chem. Soc. 2002, 124, 10519.
(15) Grevels, F. W.; Kerpen, K.; Klotzbucher, W. E.; McClung, R. E.
D.; Russell, G.; Viotte, M.; Schaffner, K. J. Am. Chem. Soc. 1998, 120,
10423.
(16) Sandstro¨m, J. Dynamic NMR Spectroscopy; Academic Press:
London, 1982.
(17) Robin, M. B.; Day, P. AdV. Inorg. Chem. Radiochem. 1967, 10, 247.
(18) Nishiumi, T.; Higuchi, M.; Yamamoto, K. Macromolecules 2003,
36, 6325.
(19) Richardson, D. E.; Taube, H. Inorg. Chem. 1981, 20, 1278.
(20) (a) Richardson, D. E.; Taube, H. Coord. Chem. ReV. 1984, 60, 107.
(b) Sutton, J. E.; Taube, H. Inorg. Chem. 1981, 20, 3125. (c) de la Rosa,
R.; Chang, P. J.; Salaymeh, F.; Curtis, J. C. Inorg. Chem. 1985, 24, 4229.
(d) Dong, Y.; Hupp, J. T. Inorg. Chem. 1992, 31, 3170.
(21) (a) Londergan, C. H.; Salsman, J. C.; Ronco, S.; Dolkas, L. M.;
Kubiak, C. P. J. Am. Chem. Soc. 2002. 124. 6236. (b) Londergan, C. H.;
Salsman, J. C.; Ronco, S.; Dolkas, L. M.; Kubiak, C. P. Inorg. Chem. 2003,
42, 926.
(22) DeRosa, M. C.; White, C. A.; Evans, C. E. B.; Crutchley, R. J. J.
Am. Chem. Soc. 2001, 123, 1396.
Acknowledgment. This work was partially supported by
Grants-in-Aid for Scientific Research (Nos. 15350073, 15655069)
and the 21st COE program (Keio-LCC) from the Ministry of
Education Science Foundation Culture, a Grant-in-Aid for
CREST from the Science and Technology Agency, and a
Kanagawa Academy of Science and Technology Research Grant
(Project No. 23).
Supporting Information Available: Fitting procedures and
parameters for spectroelectrochemical analysis, CIF file, elec-
trochemical analysis, and additional data. This material is
available free of charge via the Internet at http://pubs.acs.org.
References and Notes
(1) (a) Barbara, P. F.; Meyer, T. J.; Ratner, M. A. J. Phys. Chem. 1996,
100, 13148. (b) Yamamoto, K.; Taneichi, D. Chem. Lett. 2000, 1, 4. (c)
Higuchi, M.; Ikeda, I.; Hirao, T. J. Org. Chem. 1997, 62, 1072. (d)
Yamamoto, K.; Oyaizu, K.; Tsuchida, E. J. Am. Chem. Soc. 1996, 118,
12665.
(23) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(24) Hariharan, P. C.; Pople, J. A. Chem. Phys. Lett. 1972, 16, 217.

8000 J. Phys. Chem. B, Vol. 108, No. 23, 2004
Nishiumi et al.
(25) Gaussian 98 (Revision A.11): Frisch, M. J.; Trucks, G. W.;
Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski,
V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich,
S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.;
Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.;
Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui,
Q.; Morokuma, K.; Salvador, P.; Dannenberg, J. J.; Malick, D. K.; Rabuck,
A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.;
Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-
Gordon, M.; Replogle, E. S.; Pople, J. A.; Gaussian, Inc., Pittsburgh, PA,
2001.
(26) Because of the strong interaction between the redox centers, the
second oxidation of 2,6-DMPDA in dichloromethane without TFA is not
reversible accompanied by the proton removal due to higher potential around
0.7 V (vs Ag/Ag+). The addition of TFA restrains the proton removal, and
the reversible redox couples are observed.
(27) νn ) 1.35 × 1013 s-1 as an appropriate value in acetonitrile
according to ref 12b.
(28) Demadis, K. D.; Hartshorn, C. M.; Meyer, T. J. Chem. ReV. 2001,
101, 2655-2685.