Charge delocalization in a homologous series of α,α- Bis(dianisylamino)-substituted thiophene monocations

Article abstract of DOI:10.1021/jp303989t

A homologous series of three molecules containing thiophene, bithiophene, and terthiophene bridges between two redox-active tertiary amino groups was synthesized and explored. Charge delocalization in the one-electron-oxidized forms of these molecules was investigated by a combination of cyclic voltammetry, near-infrared optical absorption spectroscopy, and EPR spectroscopy. All three cation radicals can be described as organic mixed-valence species, and for all of them the experimental data are consistent with strong delocalization of the unpaired electron. Depending on what model is used for analysis of the optical absorption data, estimates for the electronic coupling matrix element (H_AB) range from ～5000 to ～7000 cm^-1 for the shortest member of the homologous series. According to optical absorption and EPR spectroscopy, even the terthiophene radical appears to belong either to Robin-Day class III or to a category of radicals commonly denominated as borderline class II/class III systems. The finding of such a large extent of charge delocalization over up to three adjacent thiophene units is remarkable.

Full text of DOI:10.1021/jp303989t

Article
pubs.acs.org/JPCA
Charge Delocalization in a Homologous Series of α,α′-
Bis(dianisylamino)-Substituted Thiophene Monocations
Luisa G. Reuter,† Annabell G. Bonn,† A. Claudia Stuckl,† Bice He,† Palas Baran Pati,‡ Sanjio S. Zade,‡
̈
and Oliver S. Wenger*,†
†Institut fur Anorganische Chemie, Georg-August-Universitat Gottingen, Tammannstrasse 4, D-37077 Gottingen, Germany
̈
̈
̈
̈
‡Department of Chemical Sciences, Indian Institute of Science Education and Research, Kolkata, P.O. BCKV Campus Main Office,
Mohanpur 741252, Nadia, West Bengal, India
S
* Supporting Information
ABSTRACT: A homologous series of three molecules
containing thiophene, bithiophene, and terthiophene bridges
between two redox-active tertiary amino groups was
synthesized and explored. Charge delocalization in the one-
electron-oxidized forms of these molecules was investigated by
a combination of cyclic voltammetry, near-infrared optical
absorption spectroscopy, and EPR spectroscopy. All three
cation radicals can be described as organic mixed-valence
species, and for all of them the experimental data are
consistent with strong delocalization of the unpaired electron.
Depending on what model is used for analysis of the optical absorption data, estimates for the electronic coupling matrix element
(HAB) range from ∼5000 to ∼7000 cm−1 for the shortest member of the homologous series. According to optical absorption and
EPR spectroscopy, even the terthiophene radical appears to belong either to Robin−Day class III or to a category of radicals
commonly denominated as borderline class II/class III systems. The finding of such a large extent of charge delocalization over
up to three adjacent thiophene units is remarkable.
thiophene molecules (1−3) from Scheme 1. Further included
in the study was a reference compound (4) comprised of a
thiophene unit containing only one diphenylamino substituent.
From a combined electrochemical, optical spectroscopic, and
EPR study of the monocationic forms of these molecules, we
obtain detailed insight into charge-delocalization phenomena in
thiophene monocations. Of key interest in this study was the
question to which Robin−Day class the three monocations 1+,
2+, and 3+ belong.39 We anticipated that 1+ would be a fully
delocalized (class III) radical and wanted to explore whether
the longer bithiophene (2+) and terthiophene (3+) systems
would have a similar electronic structure, or whether there
would be a transition to partial charge localization (class II) or
even complete charge localization (class I).40,41 For phenylene-
bridged organic mixed-valence systems this question had been
answered previously,16,19 but to the best of our knowledge such
a length dependence of charge delocalization has not yet been
investigated in a homologous series of thiophene-bridged
organic mixed-valence compounds.34,35 In view of the wide
interest in thiophene materials mentioned above, it seemed
relevant to explore over how many adjacent thiophene units an
unpaired electron can be delocalized.
INTRODUCTION
■
The one-electron-oxidized forms of oligothiophenes exhibit a
remarkable electrical conductivity, and this property makes
them well suited as hole transport materials in a variety of
different (opto)electronic applications.1−5 Understanding dis-
tance-dependent charge delocalization in thiophene mono-
cations is therefore of significant interest. Numerous exper-
imental and theoretical studies already explored the electronic
structure of oxidized thiophenes at a fundamental level, but
mixed-valence approaches are comparatively rare.6−14 In a
purely computational study, Lacroix and co-workers arrived at
the conclusion that oligothiophene monocations may be
described as organic mixed-valence species.15 Indeed, the
one-electron-oxidized or one-electron-reduced forms of several
oligo-p-phenylene and oligo-p-phenylenevinylene systems
could be described well as organic mixed-valence compounds
in various experimental studies,16−25 but until now there has
been comparatively little conceptually analogous work on
thiophene systems.26−35 A notable exception is the recent study
of the oxidized forms of bis(4-(alkoxyphenyl)amino) deriva-
tives of dithienylethene and bithiophene and the finding that a
mixed-valence description of these cations is meaningful.36
Another exception is our own recent investigation of
photoswitchable organic mixed valence in dithienylethene
systems.37,38
Received: April 25, 2012
Revised: May 31, 2012
Published: June 26, 2012
Here, we report on a detailed experimental investigation of
the homologous series of α,α′-bis(diphenylamino)-substituted
© 2012 American Chemical Society
7345
dx.doi.org/10.1021/jp303989t | J. Phys. Chem. A 2012, 116, 7345−7352

The Journal of Physical Chemistry A
Article
Scheme 1. Chemical Structures of the Molecules Investigated in This Work
mV, respectively, and they are attributed to one-electron-
oxidation processes involving predominantly the two different
tertiary amino groups. The decrease in the voltage splitting
(ΔE) between the two waves from 450 mV in 1 to 250 mV in 2
is a manifestation of the increasing N−N distance and a
decrease in electronic communication between the two redox-
active moieties. For compound 3 only one reversible wave is
observable at first glance, but digital fitting of the relevant part
of the cyclic voltammogram from Figure 1 c yields ΔE = 70 mV
(see Supporting Information). For reference compound 4 there
is only one tertiary amino group; hence the observation of a
single oxidation wave in the potential range considered here is
no surprise (Figure 1d). Moreover, this particular wave is
irreversible, presumably due to electrochemically induced
polymerization chemistry of the 4+ species.
RESULTS AND DISCUSSION
■
Synthesis. Compound 1 was obtained using a palladi-
um(0)-catalyzed reaction between 2,5-dibromothiophene and
4,4′-dimethoxydiphenylamine. The longer congeners (2 and 3)
were synthesized in an analogous manner using 5,5′-dibromo-
2,2′-bithiophene and 5,5″-dibromo-2,2′:5′,2″-terthiophene as
reaction partners with 4,4′-dimethoxydiphenylamine. Reference
compound 4 was available from a prior investigation.37 Detailed
synthetic protocols and product characterization data for all
new compounds are found in the Supporting Information.
Electrochemistry. Figure 1 shows the results of cyclic
voltammetry experiments with compounds 1−4 in deoxy-
The electrochemical potentials for one- and two-electron
oxidation of molecules 1−4 are summarized in Table 1. The
Table 1. Electrochemical Potentials for One- and Two-
Electron Oxidation of the Four Compounds from Scheme 1
a
(in volts) vs Fc+/Fc
compd
E1/2(1)
E1/2(2)
ΔE [mV]
Kc
1
2
3
4
−0.23
−0.17
−0.08
0.16
0.22
0.08
450
250
70
4.2 × 107
1.7 × 104
1.5 × 101
−0.01
a
ΔE is the difference between the first (E1/2(1)) and second oxidation
potentials (E1/2(2)) in mV; Kc is the comproportionation constant.
Figure 1. Cyclic voltammograms of the four compounds from Scheme
1 in dry and deoxygenated acetonitrile measured in the presence of 0.1
M tetrabutylammonium hexafluorophosphate (TBAPF6) as a support-
ing electrolyte: (a) compound 1; (b) compound 2; (c) compound 3;
(d) compound 4. The waves at −0.51 V vs Fc+/Fc (dashed vertical
line) are due to decamethylferrocene, which was added for internal
voltage calibration.
fact that the first one-electron oxidation in the bis-
(diphenylamino) systems occurs at significantly more negative
potentials than the first one-electron oxidation of reference
compound 4 is attributed to the more electron-rich nature of
the 2-fold amino-substituted systems with respect to the singly
amino-substituted reference compound. We note that in the
α,α′-bis(diphenylamino)-substituted systems 1−3 the first
oxidation potential (E1/2(1)) increases to more positive values
with increasing thiophene bridge length, while the second
oxidation potential (E1/2(2)) is decreasing simultaneously to
more negative values.42 This leads to the above-mentioned
decrease in the overall oxidation potential splitting (ΔE), which
may be interpreted as a manifestation of decreasing electronic
communication with increasing length. In mixed-valence
genated dry acetonitrile solution in the presence of
tetrabutylammonium hexafluorophosphate (TBAPF6) as a
supporting electrolyte. The waves at −0.51 V vs Fc+/Fc
(vertical dashed line) are due to decamethylferrocene, which
was added in small quantities as an internal reference for
voltage calibration. The voltammograms of compounds 1
(Figure 1a) and 2 (Figure 1b) are comprised of two clearly
discernible reversible waves which are separated by 450 and 250
chemistry the comproportionation constant (Kc = 10ΔE/59 mV
)
7346
dx.doi.org/10.1021/jp303989t | J. Phys. Chem. A 2012, 116, 7345−7352

The Journal of Physical Chemistry A
Article
is often used as a measure for the stability of the mixed-valence
state,43,44 and we calculate values of 4.2 × 107 for 1+, 1.7 × 104
for 2+, and 1.5 × 101 for 3+ (Table 1). Classification of mixed-
valence systems based on Kc values is possible,39 but it requires
some caution.43−46 With our thiophene systems from Scheme 1
we are in the fortunate situation of having the possibility to gain
complementary insight into the electronic structures of the
monocations by optical absorption and EPR spectroscopies.
Therefore, we refrain from premature conclusions on the
exclusive basis of the cyclic voltammetry studies.
any low-energy absorptions in the oxidized forms of reference
molecule 4 (Figure 2d) is consistent with the interpretation of
the lowest energy absorptions of 1+, 2+, and 3+ as intervalence
absorption bands. However, we also note that (at least in cyclic
voltammetry, vide supra) compound 4 undergoes irreversible
oxidation.
Figure 3 shows the lowest energetic absorptions of the
monocations in more detail and on a properly calibrated
Optical Absorption Spectroscopy. The black traces in
Figure 2 are the optical absorption spectra of the charge-neutral
Figure 3. Black traces, low-energy portions of optical absorption
spectra of monocationic forms of the molecules from Scheme 1. (a)
1+; (b) 2+; (c) 3+; (d) 4+. Colored traces, Gaussian fits to the
experimental data. The parameters of the Gaussian functions
represented by the dotted green traces are reported in Table 2.
Figure 2. Optical absorption spectra of the four compounds from
Scheme 1 in acetonitrile solution in the presence of increasing
amounts of Cu(ClO4)2 as a chemical oxidant: (a) compound 1; (b)
compound 2; (c) compound 3; (d) compound 4. Black lines, neat
compounds before adding any oxidant; blue lines, after addition of 0.5
equiv of oxidant; green traces, after addition of 1 equiv of oxidant; red
traces, after addition of 2 equiv of oxidant.
extinction scale. For this purpose, the low-energy portions of
the optical absorption spectra measured after addition of 0.25
equiv of oxidant were employed. The use of data obtained in
the presence of a shortfall of chemical oxidant is a common
procedure to avoid disproportionation of the mixed-valence
state.36 The black traces in Figure 3 represent the actual
experimental data. While 1+ has only one absorption band in
the relevant spectral range (Figure 3a), the longer congeners
(2+ in Figure 3b and 3+ in Figure 3c) exhibit two bands, and the
reference compound (4+) is spectroscopically innocent (Figure
3d). By analogy to prior investigations on bithiophene and
dithienyl systems with diphenylamino substituents, it appears
plausible to interpret the lowest energetic bands in 1+−3+ as
intervalence absorptions.36,37,50 Indeed, these absorptions fall
into the spectral range in which intervalence absorptions of
bis(triarylamine) radical cations are commonly ob-
served.16,20,34,35,51−61 An important difference from the
intervalence absorptions of coordination compounds40,62−65 is
the comparatively large extinction coefficients of these bands,
but this is a (favorable) peculiarity of organic mixed-valence
compounds which has been noted many times before.34,35
In order to perform a quantitative analysis of the intervalence
absorptions, the experimental spectra in Figure 3 were fitted to
Gaussian functions. Due to the increasing complexity of the
absorption spectra along the series 1+ < 2+ < 3+, an increasing
number of Gaussians was required to obtain satisfactory fits:
Three Gaussians were sufficient for Figure 3a, four were needed
for the spectrum in Figure 3b, and five were necessary in the
case of the spectrum in Figure 3c. In all cases the lowest
energetic absorption band had to be fitted with two Gaussian
functions (dashed green traces). The characteristics of the
forms of molecules 1−4 in acetonitrile solution at room
temperature. The data show that none of the four compounds
from Scheme 1 has any significant absorption features below
17 000 cm−1 as long as the molecules are in their charge-neutral
forms. Absorption spectra of molecules 1−4 with extinction
coefficients are shown in Figure S1 of the Supporting
Information, and we find that the lowest energetic absorption
maxima are at 29 700 cm−1 for 1, 25 000 cm−1 for 2, 22 200
cm−1 for 3, and 33 300 cm−1 for 4. Thus, there is a red shift of
the lowest absorption band with increasing thiophene bridge
length, which is likely a manifestation of increasing π-
conjugation.47−49
The colored traces in Figure 2 were measured after the
addition of increasing amounts of copper(II) perchlorate as an
oxidant. The blue traces represent the spectra detected after
addition of 0.5 equiv, the green traces are those measured after
adding 1 equiv, and the red traces are the spectra obtained after
2 equiv of Cu(ClO4)2 had been added. We note that in each of
the three bis(dianisylamino) systems (1−3; Figure 2a−c)
absorption bands between 5000 and 15 000 cm−1 appear upon
oxidation. Importantly, in each case the lowest energetic
absorptions are only present in situations in which less than 2
equiv of oxidant has been added (blue and green traces),
indicating that the respective absorption bands are due to the
one-electron-oxidized (mixed-valence) species. The absence of
7347
dx.doi.org/10.1021/jp303989t | J. Phys. Chem. A 2012, 116, 7345−7352

The Journal of Physical Chemistry A
Article
Table 2. Parameters Obtained from Analysis of the Lowest Energetic Near-Infrared Absorptions in Figure 3: Twice the
Bandwidth on the High-Energy Side of the Lowest Absorption (ν1/2[high]), Twice the Bandwidth on the Low-Energy Side of the
a
Lowest Absorption (ν1/2[low]), and Ratio between the Two (ν1/2[high]/ν1/2[low]
)
compound
ν1/2[high] [cm−1
]
ν1/2[low] [cm−1
]
ν1/2[high]/ν1/2[low]
νmax,G1 [cm−1
]
ν1/2,G1 [cm−1
]
νmax,G2 [cm−1
]
ν1/2,G2 [cm−1
]
νclass II [cm−1
]
1+
2+
3+
5700
3550
2640
4300
3060
1940
1.33
1.16
1.36
12 945
10 112
7 652
3231
2426
1800
15 276
11 671
9 051
3723
2232
1850
5656
4953
4237
a
νmax,G1 and νmax,G2 are the energetic positions of the maxima of the two Gaussian functions required to fit the lowest energetic near-infrared
absorption. ν1/2,G1 and ν1/2,G2 are the full widths at half-maximum values of these two Gaussian fit functions. νclass II is the bandwidth which is
expected based on eq 4.
a
respective Gaussian functions are summarized in Table 2:
νmax,Gi is the energetic position of the maximum of the ith
Gaussian while ν1/2,Gi is its full width at half-maximum. Given
the fact that all of the lowest energetic absorptions have to be
fitted by two Gaussian functions, it is obvious that the shapes of
these bands are asymmetric, an observation that is commonly
made for class III systems or mixed-valence compounds that are
near the borderline of class II and class III.16,36,63,66 In order to
capture the asymmetry of an intervalence band in one single
number, it is customary to determine the bandwidth on the
high-energy side of the intervalence band (ν1/2[high] in Table 2)
as well as on its low-energy side (ν1/2[low] in Table 2), and to
build the ratio between the two measures. In our specific cases
this procedure yields ν1/2[high]/ν1/2[low] ratios between 1.16 and
1.36 (fourth column of Table 2), which are typical values for
systems near the borderline of class II and class III.16,36,63,66
There is no clear trend of ν1/2[high]/ν1/2[low] along the series of
compounds 1+, 2+, and 3+. One might expect the respective
ratio to decrease as the systems get longer and the class II
character of the relevant mixed-valence species increases. From
the finding that ν1/2[high]/ν1/2[low] ≠ 1, one can merely conclude
that all three systems deviate from pure class II behavior, but it
seems impossible to draw any quantitative conclusions
regarding the percentage of class III character.
Table 3. Nitrogen−Nitrogen Distance (dNN), Number of σ-
Bonds between Nitrogen Atoms (nσ), and Parameters
Extracted from the Near-Infrared Absorption Spectra of the
Monocations in Figure 3: Absorption Band Maximum
(νmax), Transition Dipole Moment (μge, in Units of debyes),
and Electronic Coupling Matrix Elements (HAB) Calculated
Using Different Models (See Text)
HAB [cm−1
]
class II;
(2/3)
a
dNN
compd
[Å]
nσ νmax [cm−1
]
νge [D] class II × dNN class III
1+
2+
3+
5.2
9.1
4
7
13 850
10 620
7 770
5.7
7.7
7.8
3160
1870
970
4790
2835
1470
6925
5310
3885
13.0
10
a
Estimated from molecular modeling of the charge-neutral parent
compounds.
importance is usually what to take for R. The effective electron
transfer distance can be substantially shorter than what might
be expected from purely geometric considerations. For
instance, the effective electron transfer distance in selected
bis(triarylamine) mixed-valence systems has been found to be
only roughly two-thirds of the geometric distance between
nitrogen atoms,34,61 and analogous investigations of mixed-
valent dinitroaromatic anions have come to similar conclu-
sions.70,71 In principle it is possible to estimate R with various
computational methods,72−76 but here we intend to rely
exclusively on experimental observations. Therefore, we give in
Table 3 the results of two calculations with eq 2: Once we
report HAB calculated on the basis of the geometric N−N
distance (dNN; second column of Table 3), and once we give
From the Gaussian fits to the intervalence bands in Figure 3
it is possible to determine the dipole moment (μge) associated
with the intervalence transition by employing eq 1.67,68
ε(ν) dν
νmax
∫
μ = 0.09584
ge
(1)
Equation 1 yields the transition dipole moment in units of
debyes and requires νmax and ε(ν) input values in units of cm−1
and M−1 cm−1, respectively. Here, νmax represents the
experimentally determined maximum of the intervalence
absorptions (fourth column in Table 3) while the integral
over ε(ν) was taken as the integral over the two Gaussians
needed to fit the lowest energetic near-infrared absorptions.
Thus we obtain μge values ranging from 5.7 D for 1+ to 7.8 D
for 3+ (Table 3).
The transition dipole moment μge is in direct relation with
the electronic coupling matrix element HAB describing the
electronic interaction between the two redox-active units of a
mixed-valence species.69 This direct relationship is captured by
eq 2, which was used here to estimate HAB on the basis of the
experimentally determined μge values.
H
AB calculated on the basis of R = (2/3)dNN. For compound 1+
we obtain HAB = 3160 cm−1 (dNN) and HAB = 4790 cm−1 ((2/
3)dNN). For reference, the N,N,N′,N′-tetraanisyl-p-phenylenedi-
amine monocation (TAPD+) has HAB = 3240 cm−1, and this is a
mixed-valence species which was classified as a borderline class
II/class III system.16 For our systems we consider the analysis
on the basis of R = (2/3)dNN more adequate than analysis
based on R = dNN, inter alia because our EPR data (see below)
suggest that there is significant unpaired spin density on the
thiophene bridges.
For class III systems, it is possible to estimate HAB directly
from the intervalence absorption band maximum (νmax):
HAB = νmax/2
(3)
The νmax values are given in the fourth column of Table 3, and
using eq 3 one obtains HAB values of 6925 (1+), 5310 (2+), and
3885 cm−1 (3+) (last column of Table 3). These are relatively
high HAB values even when compared to some of the most
strongly coupled organic mixed-valence systems known to
date.34,35 For reference, the N,N,N′,N′-tetramethyl-p-phenyl-
μ νmax
eR
ge
HAB
=
(2)
In eq 2, e is the elemental charge and R is the effective
distance between redox centers. A question of central
7348
dx.doi.org/10.1021/jp303989t | J. Phys. Chem. A 2012, 116, 7345−7352

The Journal of Physical Chemistry A
Article
enediamine cation (Wurster’s blue) has HAB ≈ 8100 cm−1,
while N,N,N′,N′-tetraanisyl-p-phenylenediamine cation has HAB
= 3240 cm−1.34,35 Here we find HAB values which are close to
4000 cm−1 even for the system comprised of three adjacent
thiophene units.
When attempting to classify mixed-valence species according
to the Robin−Day scheme,39 the solvent dependence of the
intervalence absorptions may be helpful. We have therefore
explored the solvent dependence of the near-infrared
absorptions of 1+, 2+, and 3+, and the results of our efforts
are shown in Figure 4. Due to the poor solubility of our
νmax to be independent of solvent. In our case there is a solvent
dependence, but it is a comparatively weak one, suggesting that
at least the shorter members of our cation series (1+ and 2+) are
indeed near the borderline of class II and class III. The
comproportionation constants (Kc) calculated above (4.2 × 107
for 1+ and 1.7 × 104 for 2+) do not contradict this classification.
For the longest congener (3+) the electrochemical data suggest
class I (Kc = 15), but the optical spectroscopic data are more
consistent with borderline class II/class III. We note that
comproportionation constants have often been employed as
measures for electronic couplings, but it has been demonstrated
that they can actually provide a poor guide and should be used
with caution.77 Therefore, we think that the optical absorption
(and EPR) data are more reliable for classifying our mixed-
valence compounds.
An additional argument which supports interpretation of 2+
and 3+ as borderline class II/class III systems is the relatively
narrow width of the intervalence absorption bands observed for
these compounds: For a pure class II system one would expect
the bandwidth to satisfy eq 4, but in the present cases the
calculated bandwidths (νclass II, last column of Table 2; 4953 and
4237 cm−1) are significantly larger than the experimentally
observed bandwidths (ν1/2[high], second column of Table 2;
3550 and 2640 cm−1).
ν
=
2310νmax
(4)
class II
EPR Spectroscopy. Given the challenges in classifying the
monocations 1+, 2+, and 3+ according to the Robin−Day
scheme based on electrochemical results and optical absorption
spectroscopy, we decided to perform EPR spectroscopy with
these systems in order to gain complementary insight into their
electronic structures. Furthermore, EPR spectroscopy can
potentially shed some light on the question of to what extent
these monocations may indeed be considered diamino mixed-
valence compounds, or to what extent they have to be regarded
as thiophene-bridge oxidized species.36
The solid lines in the upper half of Figure 5 are the
experimental X-band EPR spectra obtained at room temper-
ature from dilute dichloromethane solutions of 1+ (a), 2+ (b),
and 3+ (c). In all three cases, the EPR signals are centered at a
g-value of 2.005, which is typical for triarylamine radical
cations.78 The appearance of the spectra in Figure 5 resembles
that observed for related bithiophene and dithienylethene
Figure 4. Optical absorption spectra of (a) 1+, (b) 2+, (c) 3+, and (d)
4+ in acetonitrile (solid black traces) and in dichloromethane (red
dashed traces).
monocations in most common solvents, the solvent depend-
ence study remained restricted to acetonitrile (black traces) and
dichloromethane (red dashed traces). The key observation is
the same in all three cases: The lowest energy absorption band
is red shifted when going from acetonitrile to dichloromethane.
However, the spectral shifts are relatively modest and range
from 850 (1+) to 560 cm−1 (3+). For class II systems a red shift
upon solvent change from acetonitrile to dichloromethane is to
be expected because of a decrease of the solvent reorganization
energy, whereas for pure class III systems one would expect
Figure 5. Solid traces, experimental X-band EPR spectra obtained from radicals (a) 1+, (b) 2+, and (c) 3+ in dilute dichloromethane solution at room
temperature. Dashed traces, simulated EPR spectra using the hyperfine coupling parameters given in Table 4. The simulations occurred using the
WINEPR SimFonia Software.
7349
dx.doi.org/10.1021/jp303989t | J. Phys. Chem. A 2012, 116, 7345−7352

The Journal of Physical Chemistry A
Article
radicals which were end-capped with 4-alkoxyphenylamino
groups;36 the EPR spectrum of 2+ had even been reported
before.36 The hyperfine structure of our EPR spectra can be
understood on the basis of an interaction of the unpaired
electron spin with the nuclear spins of 14N (I = 1) and 1H (I =
1/2), whereby the magnitudes of the coupling constants (for
between the two ends of the molecule is rapid on the EPR time
scale, i.e., faster than ∼10−7 s. There have been several
comparative investigations of optical and thermal electron
transfer in organic mixed-valence systems (mostly on bis-
(hydrazine) radical cations),80−82 which have come to the
conclusion that the results obtained by the two fundamentally
different experimental techniques (i.e., optical absorption and
EPR spectroscopies) need not necessarily lead to the same
conclusion, mostly because these techniques probe electron
transfer on significantly different time scales. Be that as it may,
in our specific case the EPR data are consistent with the optical
absorption results in that both sets of data can be analyzed
adequately by assuming complete (in EPR) or nearly complete
(in absorption) delocalization of the unpaired electron.
31,36
1
interaction with 14N and H) are similar.
This statement is
corroborated by the simulated EPR spectra shown as dashed
traces in the lower half of Figure 5. In performing these
simulations, we attempted to use a minimum of adjustable
parameters. We found that when going from 1+ to 2+ and finally
to 3+, the experimental EPR data can only be simulated
satisfactorily when increasing the number of coupling constants
along this series of radicals (Table 4).
On a final note in this section we point out that the hyperfine
constant for coupling to the two equivalent nitrogen nuclei
decreases from aN = 4.3 G in 1+ to aN = 2.9 G in 2+, and finally
to aN = 2.2 G in 3+. It appears plausible that this decrease is a
manifestation of increasing spin delocalization away from the
nitrogen atoms toward the center of the thiophene bridge. The
expected increase in π-conjugation with increasing thiophene
Table 4. Hyperfine Coupling Constants (in Units of gauss)
Used To Simulate the Experimental X-Band EPR Spectra
Represented by the Solid Traces in the Upper Half of Figure
a
5
radical
aN [G]
aH [G]
aH′ [G]
aH″ [G]
bridge length seems compatible with this interpretation.83
,95
1+
2+
3+
4.3
2.9
2.2
2.6
2.4
2.1
2.0
2.0
1.8
SUMMARY AND CONCLUSIONS
■
a
The dashed lines in the lower half of Figure 5 illustrate the outcomes
One of the key findings from the research presented in this
paper is that a mixed-valence description of α,α′-bis-
(dianisylamino)-substituted thiophene monocations is mean-
ingful from an experimental point of view, as suggested by the
theoretical work performed by Lacroix and co-workers.15 Our
experimental work demonstrates that the mixed-valence
approach is in fact reasonable for analysis of electrochemical,
optical absorption, and EPR spectroscopic data obtained from
amine-decorated thiophene monocations.
of these simulations.
For 1+, it is sufficient to invoke coupling to two nitrogen
nuclei with a hyperfine coupling constant aN = 4.3 G and
simultaneous coupling to two hydrogen nuclei with aH = 2.6 G.
In other words, the experimental data for 1+ are in line with
uniform spin distribution over the entire molecule, consistent
with class III mixed-valence behavior on the EPR time scale.
This observation is similar to what has been reported for other
bis(triarylamine) radical cations.31,36,60,61 It appears plausible
that the two chemically equivalent hydrogen nuclei which
couple to the unpaired electron spin with aH = 2.6 G are in fact
those attached to the thiophene bridging unit of molecule 1.79
The experimental EPR spectrum of 2+ in Figure 5b can be
simulated using three different hyperfine coupling constants: aN
= 2.9 G, aH = 2.4 G, and aH′ = 2.0 G (second row of Table 4).
Importantly, all of these couplings involve two equivalent
nuclei, suggesting that the electron is again symmetrically
delocalized on the EPR time scale. The additional hydrogen
hyperfine constant in this case (aH′) is most likely due to the
fact that there are now two chemically distinct hydrogen atoms
attached to the thiophene bridge; indeed, this interpretation is
in line with a prior experimental and theoretical study of 2+ and
a series of related thiophene cations.36
The optical absorption and EPR data are consistent with
interpretation of the 1+, 2+, and 3+ radical species as strongly
delocalized systems: EPR spectroscopy suggests them to be
class III systems, while optical absorption spectroscopy is
compatible with interpretation as both class III and borderline
class II/class III species. The computational work by Lacroix
and co-workers clearly shows that a transition between different
Robin−Day classes is expected for long oligothiophenes.15 Our
current experimental study suggests that this is the case only for
molecules comprised of more than three adjacent thiophene
units.
EXPERIMENTAL SECTION
■
Synthetic protocols and product characterization data for 1, 2,
3, and all relevant isolable reaction intermediates are provided
in the Supporting Information. 1H and 13C NMR spectroscopy
was performed using Bruker Avance DRX300 and Bruker B-
ACS-120 instruments, electron ionization mass spectrometry
(EI-MS) occurred on a Finnigan MAT8200 instrument, and
elemental analyses were conducted on a Vario EL III CHNS
analyzer from Elementar. Chemical oxidation of the charge-
neutral molecules from Scheme 1 to their monocationic and
dicationic forms was effected with Cu(ClO4)2, which represents
a standard procedure for generating triarylamine radical
cations.37,84,85 Cyclic voltammetry was performed using a
potentiostat from Princeton Applied Research (Versastat3-200)
with a glassy carbon disk working electrode and two separate
silver wires as counter and quasi-reference electrodes. Optical
absorption spectra were measured on a Cary 5000 instrument
For radical 3+, four hyperfine coupling constants must be
invoked (third row of Table 4), namely one for two equivalent
nitrogen atoms (aN = 2.2) and three for pairs of equivalent
hydrogen nuclei (aH = 2.1 G, aH′ = 2.0 G, aH″ = 1.8 G). In
analogy to the shorter 1+ and 2+ radicals, the EPR simulation in
Figure 5c thus leads us to the conclusion that the unpaired
electron spin is symmetrically delocalized over the 3+ species on
the EPR time scale.
From the EPR spectra alone, one would thus arrive at the
conclusion that 1+, 2+, and 3+ are all class III mixed-valence
species. However, it is to be noted that even class II compounds
can in principle exhibit EPR spectra that suggest complete
delocalization of the unpaired electron spin.34 Specifically, this
is the case in systems in which thermal electron transfer
7350
dx.doi.org/10.1021/jp303989t | J. Phys. Chem. A 2012, 116, 7345−7352

The Journal of Physical Chemistry A
Article
(19) Rosokha, S. V.; Sun, D.-L.; Kochi, J. K. J. Phys. Chem. A 2002,
106, 2283−2292.
from Varian. The EPR spectra were recorded on a ELEXSYS
CW-EPR spectrometer E500 from Bruker, using a microwave
frequency of about 9.42 GHz. The modulation frequency was
100 kHz, the modulation amplitude was 1 × 10−4 G, and the
power of the microwave was 5−6 mW. The spectrometer was
equipped with a digital temperature control system ER 4131
VT, covering a temperature range from 110 to 350 K. UV−vis
and cyclic voltammetry data were analyzed using the Igor Pro
software (version 6.0.0.0) from WaveMetrics.
(20) Barlow, S.; Risko, C.; Chung, S. J.; Tucker, N. M.; Coropceanu,
V.; Jones, S. C.; Levi, Z.; Bred
2005, 127, 16900−16911.
(21) Funston, A.; Kirby, J. P.; Miller, J. R.; Pospísi
Hromadova, M.; Gal, M.; Pecka, J.; Valas k, M.; Zawada, Z.; Rempala,
́
as, J. L.; Marder, S. R. J. Am. Chem. Soc.
̌
l, L.; Fiedler, J.;
́
́
́ ̌
e
P.; Michl, J. J. Phys. Chem. A 2005, 109, 10862−10869.
(22) Nelsen, S. F.; Weaver, M. N.; Telo, J. P. J. Phys. Chem. A 2007,
111, 10993−10997.
(23) Zhou, G.; Baumgarten, M.; Mullen, K. J. Am. Chem. Soc. 2007,
̈
ASSOCIATED CONTENT
* Supporting Information
129, 12211−12221.
■
(24) Nelsen, S. F.; Weaver, M. N.; Telo, J. P. J. Am. Chem. Soc. 2007,
129, 7036−7043.
S
Synthetic protocols and characterization data for molecules 1−
3 and all intermediate reaction products. Additional optical
absorption data. Digital fit of the cyclic voltammogram from
Figure 1c. This material is available free of charge via the
Internet at http://pubs.acs.org.
(25) Lloveras, V.; Vidal-Gancedo, J.; Figueira-Duarte, T. M.;
Nierengarten, J. F.; Novoa, J. J.; Mota, F.; Ventosa, N.; Rovira, C.;
Veciana, J. J. Am. Chem. Soc. 2011, 133, 5818−5833.
(26) Bauerle, P.; Segelbacher, U.; Maier, A.; Mehring, M. J. Am.
̈
Chem. Soc. 1993, 115, 10217−10223.
(27) Zhu, Y. B.; Wolf, M. O. J. Am. Chem. Soc. 2000, 122, 10121−
10125.
AUTHOR INFORMATION
Corresponding Author
*E-mail: oliver.wenger@chemie.uni-goettingen.de.
■
(28) Le Stang, S.; Paul, F.; Lapinte, C. Organometallics 2000, 19,
1035−1043.
(29) Fraysse, S.; Coudret, C.; Launay, J.-P. J. Am. Chem. Soc. 2003,
125, 5880−5888.
Notes
The authors declare no competing financial interest.
(30) Casado, J.; Gonzalez, S. R.; Delgado, M. C. R.; Oliva, M. M.;
Navarrete, J. T. L.; Caballero, R.; de la Cruz, P.; Langa, F. Chem.Eur.
J. 2009, 15, 2548−2559.
ACKNOWLEDGMENTS
■
(31) Rohde, D.; Dunsch, L.; Tabet, A.; Hartmann, H.; Fabian, J. J.
Phys. Chem. B 2006, 110, 8223−8231.
Financial support from the Deutsche Forschungsgemeinschaft
(DFG) through Grant WE4815/1-1 is gratefully acknowledged.
We are grateful for financial support for the research visit of
(32) Gao, L. B.; Kan, J.; Fan, Y.; Zhang, L. Y.; Liu, S. H.; Chen, Z. N.
Inorg. Chem. 2007, 46, 5651−5664.
P.B.P. from the International Office, University of Gottingen.
̈
(33) Noll, G.; Avola, M.; Lynch, M.; Daub, J. J. Phys. Chem. C 2007,
̈
111, 3197−3204.
REFERENCES
(34) Hankache, J.; Wenger, O. S. Chem. Rev. 2011, 111, 5138−5178.
(35) Heckmann, A.; Lambert, C. Angew. Chem., Int. Ed. 2012, 51,
326−392.
■
(1) Roncali, J. Chem. Rev. 1992, 92, 711−738.
(2) Horowitz, G. Adv. Mater. 1998, 10, 365−377.
(3) Murphy, A. R.; Frechet, J. M. J. Chem. Rev. 2007, 107, 1066−
1096.
(36) Odom, S. A.; Lancaster, K.; Beverina, L.; Lefler, K. M.;
́
Thompson, N. J.; Coropceanu, V.; Bredas, J. L.; Marder, S. R.; Barlow,
S. Chem.Eur. J. 2007, 13, 9637−9646.
(4) Zhu, X. H.; Peng, J. B.; Caoa, Y.; Roncali, J. Chem. Soc. Rev. 2011,
(37) He, B.; Wenger, O. S. J. Am. Chem. Soc. 2011, 133, 17027−
17036.
40, 3509−3524.
(5) Ma, C. Q.; Pisula, W.; Weber, C.; Feng, X. L.; Mullen, K.;
̈
(38) Wenger, O. S. Chem. Soc. Rev. 2012, 41, 3772−3779.
(39) Robin, M. B.; Day, P. Adv. Inorg. Chem. Radiochem. 1967, 10,
247−422.
(40) Brunschwig, B. S.; Creutz, C.; Sutin, N. Chem. Soc. Rev. 2002,
31, 168−184.
Bauerle, P. Chem.Eur. J. 2011, 17, 1507−1518.
̈
(6) Miller, L. L.; Mann, K. R. Acc. Chem. Res. 1996, 29, 417−423.
(7) Katz, H. E.; Bao, Z. J. Phys. Chem. B 2000, 104, 671−678.
(8) Katz, H. E.; Bao, Z. N.; Gilat, S. L. Acc. Chem. Res. 2001, 34, 359−
369.
(41) Creutz, C. Prog. Inorg. Chem. 1983, 30, 1−73.
(42) The separation between the cathodic and anodic peaks in Figure
1 is significantly larger than 59 mV; the precise reasons for this are
unclear.
(43) D’Alessandro, D. M.; Keene, F. R. Dalton Trans. 2004, 3950−
3954.
(9) Hutchison, G. R.; Ratner, M. A.; Marks, T. J. J. Phys. Chem. B
2005, 109, 3126−3138.
(10) Hutchison, G. R.; Ratner, M. A.; Marks, T. J. J. Am. Chem. Soc.
2005, 127, 16866−16881.
(11) James, D. K.; Tour, J. M. Molecular wires. In Molecular Wires:
From Design to Properties; De Cola, L., Ed.; Topics in Current
Chemistry 257; Springer-Verlag: Berlin, 2005; pp 33−62.
(12) Gao, P.; Beckmann, D.; Tsao, H. N.; Feng, X. L.; Enkelmann,
(44) Kaim, W.; Klein, A.; Glockle, M. Acc. Chem. Res. 2000, 33, 755−
̈
763.
(45) One difficulty with using Kc values for characterizing mixed-
valence behavior arises from the fact that calculation of the
comproportionation constant relies on the assumption that the
solvation energies for monocations and dications is identical. However,
this does not necessarily have to be the case; see, for example, ref 46.
(46) Nelsen, S. F. Adv. Phys. Org. Chem. 2006, 41, 183−215.
(47) Jones, D.; Guerra, M.; Favaretto, L.; Modelli, A.; Fabrizio, M.;
Distefano, G. J. Phys. Chem. 1990, 94, 5761−5766.
V.; Baumgarten, M.; Pisula, W.; Mullen, K. Adv. Mater. 2009, 21, 213−
̈
216.
(13) Zade, S. S.; Zamoshchik, N.; Bendikov, M. Acc. Chem. Res. 2011,
44, 14−24.
(14) Speck, M. J.; Schaarschmidt, D.; Lang, H. Organometallics 2012,
31, 1975−1982.
̀
(15) Lacroix, J. C.; Chane-Ching, K. I.; Maquere, F.; Maurel, F. J. Am.
Chem. Soc. 2006, 128, 7264−7276.
(48) da Silva, D. A.; Coropceanu, V.; Fichou, D.; Gruhn, N. E.; Bill,
(16) Lambert, C.; Noll, G. J. Am. Chem. Soc. 1999, 121, 8434−8442.
̈
T. G.; Gierschner, J.; Cornil, J.; Bred
2007, 365, 1435−1452.
́
as, J. L. Philos. Trans. R. Soc. A
(17) Rovira, C.; Ruiz-Molina, D.; Elsner, O.; Vidal-Gancedo, J.;
Bonvoisin, J.; Launay, J.-P.; Veciana, J. Chem.Eur. J. 2001, 7, 240−
250.
(49) Salzner, U. J. Phys. Chem. A 2010, 114, 10997−11007.
(50) In using the term “intervalence” we do not intend to make a
distinction between partially and fully charge-delocalized systems. As is
(18) Lindeman, S. V.; Rosokha, S. V.; Sun, D.; Kochi, J. K. J. Am.
Chem. Soc. 2002, 124, 843−855.
7351
dx.doi.org/10.1021/jp303989t | J. Phys. Chem. A 2012, 116, 7345−7352

The Journal of Physical Chemistry A
Article
common practice, we use this term interchangeably for class II and
class III mixed-valence compounds.
(84) Sreenath, K.; Suneesh, C. V.; Gopidas, K. R.; Flowers, R. A. J.
Phys. Chem. A 2009, 113, 6477−6483.
(85) Sreenath, K.; Thomas, T. G.; Gopidas, K. R. Org. Lett. 2011, 13,
1134−1137.
(51) Lambert, C.; Noll, G. Angew. Chem., Int. Ed. 1998, 37, 2107−
̈
2110.
(52) Bonvoisin, J.; Launay, J. P.; Van der Auweraer, M.; De Schryver,
F. C. J. Phys. Chem. 1994, 98, 5052−5057.
(53) Bonvoisin, J.; Launay, J. P.; Verbouwe, W.; Van der Auweraer,
M.; De Schryver, F. C. J. Phys. Chem. 1996, 100, 17079−17082.
́
(54) Coropceanu, V.; Lambert, C.; Noll, G.; Bredas, J. L. Chem. Phys.
̈
Lett. 2003, 373, 153−160.
(55) Jones, S. C.; Coropceanu, V.; Barlow, S.; Kinnibrugh, T.;
Timofeeva, T.; Bredas, J. L.; Marder, S. R. J. Am. Chem. Soc. 2004, 126,
11782−11783.
́
(56) Sun, D.; Rosokha, S. V.; Kochi, J. K. Angew. Chem., Int. Ed. 2005,
44, 5133−5136.
(57) Barlow, S.; Risko, C.; Coropceanu, V.; Tucker, N. M.; Jones, S.
C.; Levi, Z.; Khrustalev, V. N.; Antipin, M. Y.; Kinnibrugh, T. L.;
Timofeeva, T.; Marder, S. R.; Bred
766.
́
as, J. L. Chem. Commun. 2005, 764−
(58) Noll, G.; Avola, M. J. Phys. Org. Chem. 2006, 19, 238−241.
̈
(59) Amthor, S.; Lambert, C. J. Phys. Chem. A 2006, 110, 1177−
1189.
(60) Kattnig, D. R.; Mladenova, B.; Grampp, G.; Kaiser, C.;
Heckmann, A.; Lambert, C. J. Phys. Chem. C 2009, 113, 2983−2995.
(61) Lancaster, K.; Odom, S. A.; Jones, S. C.; Thayumanavan, S.;
́
Marder, S. R.; Bredas, J. L.; Coropceanu, V.; Barlow, S. J. Am. Chem.
Soc. 2009, 131, 1717−1723.
(62) Brunschwig, B. S.; Sutin, N. Coord. Chem. Rev. 1999, 187, 233−
254.
(63) Demadis, K. D.; Hartshorn, C. M.; Meyer, T. J. Chem. Rev.
2001, 101, 2655−2685.
(64) D’Alessandro, D. M.; Keene, F. R. Chem. Soc. Rev. 2006, 35,
424−440.
(65) D’Alessandro, D. M.; Keene, F. R. Chem. Rev. 2006, 106, 2270−
2298.
(66) Nelsen, S. F. Chem.Eur. J. 2000, 6, 581−588.
(67) Creutz, C.; Newton, M. D.; Sutin, N. J. Photochem. Photobiol., A
1994, 82, 47−59.
(68) Cave, R. J.; Newton, M. D. Chem. Phys. Lett. 1996, 249, 15−19.
(69) Hush, N. S. Prog. Inorg. Chem. 1967, 8, 391.
(70) Nelsen, S. F.; Konradsson, A. E.; Weaver, M. N.; Telo, J. P. J.
Am. Chem. Soc. 2003, 125, 12493−12501.
(71) Nelsen, S. F.; Weaver, M. N.; Zink, J. I.; Telo, J. P. J. Am. Chem.
Soc. 2005, 127, 10611−10622.
(72) Nelsen, S. F.; Newton, M. D. J. Phys. Chem. A 2000, 104,
10023−10031.
(73) Johnson, R. C.; Hupp, J. T. J. Am. Chem. Soc. 2001, 123, 2053−
2057.
(74) Coropceanu, V.; Malagoli, M.; Andre,
Chem. Soc. 2002, 124, 10519−10530.
́ ́
J. M.; Bredas, J. L. J. Am.
(75) Renz, M.; Theilacker, K.; Lambert, C.; Kaupp, M. J. Am. Chem.
Soc. 2009, 131, 16292−16302.
(76) Kaupp, M.; Renz, M.; Parthey, M.; Stolte, M.; Wurthner, F.;
̈
Lambert, C. Phys. Chem. Chem. Phys. 2011, 13, 16973−16986.
(77) Barriere, F.; Camire, N.; Geiger, W. E.; Mueller-Westerhoff, U.
̀
T.; Sanders, R. J. Am. Chem. Soc. 2002, 124, 7262−7263.
(78) Bamberger, S.; Hellwinkel, D.; Neugebauer, F. A. Chem. Ber.
1975, 108, 2416−2421.
(79) DFT calculations could potentially shed more light on this issue,
but this is beyond the scope of our detailed experimental study.
(80) Nelsen, S. F.; Ismagilov, R. F.; Trieber, D. A. Science 1997, 278,
846−849.
(81) Nelsen, S. F.; Chang, H.; Wolff, J. J.; Adamus, J. J. Am. Chem.
Soc. 1993, 115, 12276−12289.
(82) Nelsen, S. F.; Ismagilov, R. F.; Powell, D. R. J. Am. Chem. Soc.
1996, 118, 6313−6314.
(83) Jenart, M.; Niebel, C.; Balandier, J. Y.; Leroy, J.; Mignolet, A.;
Stas, S.; Van Vooren, A.; Cornil, J.; Geerts, Y. H. Tetrahedron 2012, 68,
349−355.
7352
dx.doi.org/10.1021/jp303989t | J. Phys. Chem. A 2012, 116, 7345−7352