Structural characterization of 2,3,5,6-tetramethyl-p-phenylenediamine radical cation and its dimer in molecular crystals

Article abstract of DOI:10.1016/j.molstruc.2014.12.005

Reaction between 2,3,5,6-tetramethyl-p-phenylenediamine (DDA) and bromanil (BA) gave two different crystalline products, needle-like and block-shaped crystals. The former was an alternate-stack 1:1 charge-transfer complex between DDA and BA with an electron-transfer ratio of about 0.7. The latter was a bromide salt of DDA, which was crystallized simultaneously with the decomposition of BA anions. The DDA molecules in the salt were dimerized even at room temperature. The similar products were obtained by a reaction between DDA and chloranil. The molecular and electronic structures are discussed with the calculation results of the density functional theory. This report has two additional meanings on the synthesis of charge-transfer complexes with using haloanils, which are commonly used as electron acceptors: (1) special attention must be paid for contaminating halide ions in characterizing reaction products and in discussing their properties, and (2) a new moderate synthetic root of halide complexes becomes available when using haloanils.

Full text of DOI:10.1016/j.molstruc.2014.12.005

Journal of Molecular Structure 1083 (2015) 260–267
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Structural characterization of 2,3,5,6-tetramethyl-p-phenylenediamine
radical cation and its dimer in molecular crystals
Tsukasa Takaiwa a, Atsushi Koyama a, Yoshihide Nagaishi a, Kiyohiko Nakajima b, Michinori Sumimoto c,
Kenji Hori c, Susumu Matsuzaki a, Hitoshi Fujimoto a,
⇑
a Department of Chemistry, Graduate School of Science and Technology, Kumamoto University, Kurokami 2-39-1, Kumamoto 860-8555, Japan
b Department of Chemistry, Aichi University of Education, Kariya, Aichi 448-8542, Japan
c Division of Materials Science and Engineering, Graduate School of Science and Engineering, Yamaguchi University, Tokiwadai, Ube 755-8611, Japan
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
ꢀ
ꢀ
ꢀ
We reported the crystal structures of
two reaction products between DDA
and bromanile.
X-ray crystal analyses showed that
these are 1:1 CT complexes and
bromide salts.
The electronic structures of DDA
complexes were investigated by UPS,
and the results were discussed with
the DFT calculations.
ꢀ
The degree of charge-transfer was
estimated to be about 0.7 for the CT
complex.
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 23 October 2014
Received in revised form 1 December 2014
Accepted 1 December 2014
Available online 6 December 2014
Reaction between 2,3,5,6-tetramethyl-p-phenylenediamine (DDA) and bromanil (BA) gave two different
crystalline products, needle-like and block-shaped crystals. The former was an alternate-stack 1:1
charge-transfer complex between DDA and BA with an electron-transfer ratio of about 0.7. The latter
was a bromide salt of DDA, which was crystallized simultaneously with the decomposition of BA anions.
The DDA molecules in the salt were dimerized even at room temperature. The similar products were
obtained by a reaction between DDA and chloranil. The molecular and electronic structures are discussed
with the calculation results of the density functional theory. This report has two additional meanings on
the synthesis of charge-transfer complexes with using haloanils, which are commonly used as electron
acceptors: (1) special attention must be paid for contaminating halide ions in characterizing reaction
products and in discussing their properties, and (2) a new moderate synthetic root of halide complexes
becomes available when using haloanils.
Keywords:
Crystal structure
Molecular structure
Charge-transfer complex
Ultraviolet photoelectron spectroscopy
Electronic structure
Ó 2014 Elsevier B.V. All rights reserved.
Introduction
Among the aromatic amines, p-phenylenediamine (PPD) and its
N-methyl derivatives, such as N,N-dimethyl-p-phenylenediamine
Aromatic amines are strong electron donors and used as elec-
tron-donating components of charge-transfer (CT) complexes.
(DMPD) and N,N,N0,N0-tetramethyl-p-phenylenediamine (TMPD)
are known as Würster’s dyes. Würster radicals show remarkably
interesting magnetic properties: they are usually paramagnetic at
high temperatures and diamagnetic at low temperatures [1]. These
phenomena are explained by the formation of Würster radical
⇑
Corresponding author. Tel.: +81 96 342 3383; fax: +81 96 342 3320.
E-mail address: fuji@sci.kumamoto-u.ac.jp (H. Fujimoto).
http://dx.doi.org/10.1016/j.molstruc.2014.12.005
0022-2860/Ó 2014 Elsevier B.V. All rights reserved.

T. Takaiwa et al. / Journal of Molecular Structure 1083 (2015) 260–267
261
Table 1
dimers in low temperature phases. The formation of dimers has
Elemental analysis for reaction products between DDA and BA.
also been reported in solutions of these radicals by Raman [2,3]
and absorption [4–9] spectroscopies. Thus, theoretical calculations
performed on cationic dimers have become available including
those for 2,3,5,6-tetramethyl-p-phenylenediamine (durendiamine;
DDA) [10–12].
Compounds
Analyses (%)
Yield (%)b
C
H
N
Needle-like crystal
32.85
49.33
34.30
32.69
49.19
2.64
6.61
3.09
2.74
6.61
4.69
11.50
5.41
4.77
11.47
48.6
13.3
–
–
–
Block-shaped crystal
Mixture as obtained
DDA–BA (C16H18N2O2Br4)a
DDA–Br (C10H16N2Br)a
On the other hand, benzoquinones and their halogen-
substituted derivatives are used widely as electron acceptors of
CT complexes, and the physical properties of resulting CT
complexes have also been studied rigorously. A series of tetra-
halo-p-benzoquinones (haloanils) are important for investigating
the physicochemical properties of CT complexes, because their
electron accepting abilities can be altered systematically by chang-
ing the substituted halogens. Complexes between haloanils and
PPD [13,14], TMPD [15] or DDA [16] show relatively high conduc-
tivities against the ionic nature of their ground state [17,18]. These
complexes sometimes show non-stoichiometric molar ratios of
components [19] and sometimes the adoption of solvent molecules
[20]. This may cause these materials to be investigated on their
physical properties with some difficulties.
Experimental information on molecular structures and their
stacking is crucial for discussing the physical properties of CT com-
plexes; however, structural information for halide complexes with
both PPD and DDA are still lacking due to their crystal qualities. To
our knowledge, among aromatic amine and haloanil complexes, a
crystal structure has only been reported on the TMPD-tetra-
chloro-p-benzoquinone (chloranil; CA) complex [21].
In this paper, we carried out X-ray analyses on complexes of
DDA bromide (DDA–Br) and DDA-tetrabromo-p-benzoquinon
(bromanil; BA), which are reaction products of DDA and BA. The
columnar structure and a third type of the molecular stacking in
DDA–Br will be discussed with the aid of the density functional
theory (DFT). The similar reaction products were obtained between
DDA and CA. The electronic structures of these DDA complexes
were investigated by ultraviolet photoelectron spectroscopy
(UPS), and the results were also discussed with the DFT calcula-
tions for model systems based on the crystal structures. In addition
to these results, part of our aims here is to emphasize a new syn-
thetic root for halide complexes using haloanils.
a
Calculated values for products expected under the reaction conditions.
Based upon starting DDA.
b
DDA was also reacted diffusively with CA in toluene or acetoni-
trile as a solvent. Contrary to the case of BA, it took a long time for
the reaction between DDA and CA. Needle-like and block-shaped
crystals were taken away from the reaction tube by filtration after
three months and one year, respectively, and dried over silica gel in
a desiccator. The former product was seemed to be uniform, but
the latter was a mixture of both types of crystals from an observa-
tion through a microscope. The results of chemical analyses for the
needle-like crystal and a typical mixture, which were obtained in
toluene and acetonitrile, respectively, are given in Table 2 along
with the theoretical values for some expected products under the
experimental conditions.
Infrared absorption spectroscopy was performed with a potas-
sium bromide disk method. Raman spectra were measured by a
quasi-back scattering alignment with excitation of 514.5 nm. The
electronic absorption spectra were also measured by a pressed disk
diluting with potassium bromide. UPS spectra were measured and
analyzed with the same system and the same method as the previ-
ous report [22]. Thin films were obtained by scrubbing the sample
on the polished copper substrate in air. All UPS measurements
were performed under the base pressure of less than 10ꢁ4 Pa.
The sample films were characterized by Raman spectra after UPS
measurements.
X-ray crystal analyses were performed on the reaction products
of DDA and BA. Block-shaped crystals were used as obtained. In the
case of needle-like crystals, a portion of the crystalline parts was
selected, excised and used in further analyses because of its poor
quality. Positions of hydrogen atoms were mapped by difference
Fourier syntheses, and refined by the least squares method using
an isotropic temperature factor. The structural data of these mate-
rials have been deposited to the Cambridge Crystallographic Data
Centre (CCDC Nos. 668457 and 668458).
The unrestricted DFT method with the B3PW91 [23–28] func-
tional was applied to a radical cation of DDA (DDA+). The geometry
was optimized under three types of molecular symmetries; C1
(without symmetry restriction), Cs and D2h symmetries. Any
remarkable differences were not observed in the total energy
among these three symmetries, where relative energies of both
Cs and D2h geometries were 0.004 kcal/mol higher than that of
the C1 structure. We used the higher symmetric D2h geometry for
Experimental
Acetonitrile was dehydrated with calcium hydride and purified
by distillation over phosphoric oxide. BA and CA were recrystal-
lized from an acetonitrile solution before succeeding reactions.
DDA and toluene with a purity better than 98% and 99%, respec-
tively, were used as received.
DDA was reacted diffusively with a molar equivalent of BA in
acetonitrile using an H-shaped glass apparatus in the dark. Nee-
dle-like crystals started to grow for several days, and after two
weeks block-shaped crystals appeared. Both crystals were
removed from the reaction tube by filtration after one week and
one month, respectively, and dried over silica gel in a desiccator.
The former product contained a mixture of both types of crystals,
but the latter seemed to be uniform under a microscope. If the sol-
vent was changed to toluene, the needle-like crystalline product
became to be uniform. Typical images of both crystals under a
microscope are shown in Fig. S1 of Electronic Supplementary Data.
Chemical analysis results for needle-like and block-shaped crys-
tals, which were obtained in toluene and acetonitrile, respectively,
are given in Table 1 along with typical yields and theoretical values
for some products that are expected under the experimental con-
ditions described herein. The values for the mixture, which was
obtained in acetonitrile, are also listed in the table by way of
example.
Table 2
Elemental analysis for reaction products between DDA and CA.
Compounds
Analyses (%)
Yield (%)b
C
H
N
Needle-like crystal
47.00
53.40
46.62
60.14
3.92
7.90
4.40
8.08
6.71
12.75
6.80
47.8
Mixture as obtained
DDA–CA (C16H18N2O2Cl4)a
DDA–Cl (C10H16N2Cl)a
–
–
–
14.03
a
Calculated values for products expected under the reaction conditions.
Based upon starting DDA.
b

262
T. Takaiwa et al. / Journal of Molecular Structure 1083 (2015) 260–267
Table 3
Table 4
Crystallographic data for DDA–Br.
Crystallographic data for DDA–BA.
DDA–Br
DDA–BA
Chemical formula
C
10H16N2ꢃBr
Chemical formula
Formula weight
Crystal system
Space group
T (K)
C16H16Br4N2O2
587.95
Formula weight
Crystal system
Space group
T (K)
244.15
Monoclinic
C2/m
Triclinic
ꢀ
P1
298
298
a (Å)
b (Å)
c (Å)
9.714 (5)
16.195 (9)
6.867 (4)
109.15 (4)
1020.5 (10)
4
4.00
1469
1218
999
0.040, 1.03
98
0.029
0.029
0.40, ꢁ0.38
a (Å)
b (Å)
c (Å)
6.704 (6)
7.733 (6)
9.027 (7)
77.26 (3)
72.40 (4)
87.99 (4)
434.8 (6)
1
9.27
1971
1971
1615
0.0000, 1.11
117
0.050
0.150
0.97, ꢁ0.76
b (°)
a
(°)
V (Å3)
b (°)
Z
l
c
(°)
(mmꢁ1
)
V (Å3)
Measured reflections
Z
l
(mmꢁ1
)
Independent reflections
Reflections with F2 > 2
Rint, S
r
(F2)
Measured reflections
Independent reflections
Reflections with I > 2
Rint, S
Number of parameters
r
(I)
R[F2 > 2
r
(F2)]
wR(F2)
Number of parameters
R[F2 > 2
wR(F2)
r
(F2)]
D
qmax
,
D
qmin (eÅꢁ3
)
CCDC No. 668458.
D
qmax
,
D
qmin (eÅꢁ3
)
CCDC No. 668457.
the discussion because of its small energy difference. Using the
obtained geometry of the DDA+ monomer, a dimer structure was
also optimized without counter anions for charge compensation.
The inter-plane distance between the DDA+ moieties and the
amount of sliding along the direction of two amino groups were
optimized, where the geometry of each DDA+ moiety was fixed
as that of the DDA+ monomer. Analytical vibrational frequency
computations at the optimized structure were then performed to
confirm that the optimized structure was at an energy minimum.
The Cartesian coordinates for the optimized geometries for the
DDA+ monomer and dimer are deposited as a list in Electronic Sup-
plementary Data. The vertical values of the ionization potential
consistent with results from an elemental analysis and the fact that
the block-shaped crystal showed no vibrations corresponding to
anionic and neutral molecules of BA (see Fig. S2 in Electronic Sup-
plementary Data). It is obvious that there is no source of bromide
anions (Brꢁ) under the reaction conditions except for BA. An earlier
study [31] of TMPD–CA complexes also showed that further reac-
tions with CA anions take place in ionizing solvents. Dissociative
electron attachment reactions of alkyl halides occur during photo-
ionization processes of TMPD in organic glasses at low tempera-
tures [32]. Coupling reactions between aromatic amines and CA
have also been reported [33]. This work clearly shows that decom-
position of haloquinon anions takes place and liberates halide ions
in polar solvents. It could be thought that this decomposition
increases slowly in concentrations of Brꢁ anions giving good reac-
tion fields for crystal growth. The same situation would be held at
the reaction between DDA and CA. This is important when synthe-
sizing CT complexes because haloquinons can be used as halide-
ion source reagents. In addition to this, it is noteworthy that the
quality of crystals is sufficient to determine the positions of the
hydrogen atoms.
were calculated by the
DSCF method, while the adiabatic ones
were estimated from the difference of the total energies between
the corresponding two charge states. The values of the electron
affinity were also calculated by the
DSCF method. All calculations
were performed with using the 6–311G(d) basis set. The Gaussian
09 program package [29] was employed for these calculations. All
molecular structures were drawn with the GaussView 5.0 [30].
Results and discussion
On the other hand, X-ray crystallography revealed that needle-
like crystals consist of 1:1 complexes between DDA and BA, and
there is no evidence of solution uptake. The complex of DDA–BA
Two types of crystals obtained from reactions between DDA and
CA resembled corresponding ones from DDA and BA in spectral
patterns of their infrared absorption (see Figs. S2 and S3 in Elec-
tronic Supplementary Data) and in crystalline appearances such
as a needle or a block; therefore, it would be expected that the
two types of reaction products are similar to each other in the
cases of CA and BA. The products between DDA and CA were usu-
ally obtained as powdery samples. Especially, the sample, which
were kept for a long period in the reaction system, had mainly
small block-shaped crystals, but contained a small portion of very
tiny needle-like crystals from observation through a microscope.
Contrary with this, the reaction between DDA and BA gave
relatively large crystals; then, crystallographic analyses were per-
formed to both types of crystals. Typical sizes were
0.5 ꢂ 0.25 ꢂ 0.06 mm3 and 0.2 ꢂ 0.2 ꢂ 0.2 mm3 for needle-like
and block-shaped crystals, respectively. Crystal X-ray analyses
clearly showed that block-shaped crystals obtained from the reac-
tion between DDA and BA are salt complexes of DDA bromide
(DDA–Br), which crystallizes in the monoclinic system, space
group C2/m. Crystallographic data are listed in Table 3. This is
ꢀ
crystallizes in the triclinic system, space group P1. Crystallographic
data are listed in Table 4. This is also consistent with results from
chemical analyses. The crystal quality of DDA–BA is usually worse
than that of DDA–Br. This is due to decomposing BAꢁ anions in sur-
face regions of crystals and due to effects from Brꢁ anions gener-
ated by the decomposition, and is the reason why reactions
between DDA and BA do not give rise to a single product in a polar
solvent. This situation was obvious for the case of the reaction
products between DDA and CA, because decomposition of CAꢁ
anions was slower than that of BAꢁ anions and the DDA–CA com-
plexes still remained after a year. Typical data for the mixture as
synthesized are shown in Tables 1 and 2. This shows that special
care should be taken when characterizing structures and when dis-
cussing results from this series of the charge-transfer complexes
with haloaniles.
Fig. 1 compares the molecular structure of DDA+ obtained by
the X-ray analysis of DDA–Br to the optimized geometry of the
DFT calculation. The position of each hydrogen atom in the X-ray

T. Takaiwa et al. / Journal of Molecular Structure 1083 (2015) 260–267
263
Fig. 1. Bond lengths given in angstroms (left) and bond angles in degrees (right) of DDA+: (a) experimental and (b) DFT optimized structures. A mirror plane lies perpendicular
to the molecular plane; therefore, values for an asymmetric unit are shown. Thermal motion ellipsoids are also depicted for carbon (green) and nitrogen (blue) atoms in the X-
ray structure. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
analysis was determined from difference Fourier syntheses, which
also attests to the high quality of the crystal used. It is obvious
from observed and calculated structures that the DDA+ molecule
adopts a quinoid structure as pointed out from AM1 calculations
[10]. From X-ray analyses, the central carbon ring is almost planar
within experimental error, and four methyl carbon atoms and one
of two amino groups also poses on this plane; however, the other
amino group is slightly removed from the plane. This is partly due
to interactions between DDA+ molecules, and in part to a hydro-
gen-like bonding between the amino group and the Brꢁ ion as dis-
cussed below. It should be mentioned that methyl groups do not
freely rotate about their C–CH3 axes even at room temperature.
DFT calculations show that hydrogen atoms from methyl groups
stay away from those of amino groups because of steric repulsion.
This provides a reason as to why methyl groups do not rotate
freely.
ecules resembles the one observed for the low temperature phase
of the TMPD perchlorate salt in spite of the difference in molecular
overlapping [34]. The inter-planar spacing of 3.1 Å is obviously
close to the van der Waals radius of a carbon atom, suggesting that
there would be strong interactions between unpaired electrons of
DDA+ molecules of DDA–Br form one-dimensional columns
along the c-axis. The stacking manner of DDA+ molecules is shown
in Fig. 2. All the molecules are equivalent in crystallography; there-
fore, there exists no evidence of doubly charged or neutral mole-
cules in the crystal. One remarkable feature of the column is an
alternation in the distance between molecular planes: shorter
and longer distances are 3.1 and 3.6 Å, respectively, which corre-
spond to intervals of 3.16 and 3.65 Å in a nearly straight line of
amino carbon atoms. In addition to this column of DDA+ molecules,
four Brꢁ anions closely locate at amino hydrogen atoms with dis-
tances of 2.65 and 2.77 Å in the asymmetric unit. This hydrogen-
like bonding combines DDA+ columns in order to form networks
in ab planes. It should be noted that this alternation in DDA+ mol-
Fig. 2. Stacking of DDA+ and Brꢁ (red ellipsoids) molecules in DDA–Br complexes
with characteristic inter-atomic contacts given in angstroms.

264
T. Takaiwa et al. / Journal of Molecular Structure 1083 (2015) 260–267
Fig. 3. Bond lengths given in angstroms (left) and bond angles in degrees (right) for DDA and BA molecules in DDA–BA crystals. Both molecules have inversion symmetry;
therefore, the values for an asymmetric unit are shown. Thermal motion ellipsoids are also depicted for carbon (green), nitrogen (blue), bromine (red), and oxygen (purple)
atoms. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
these two molecules. This spacing is similar to the value of 3.15 Å
for DMPD–Br [35] and is shorter than the value of 3.52 Å for
TMPD–Br [36]. Overlapping of DDA+ molecules is smaller than in
these Würster’s salts. The distance and the degree of overlapping
between molecules are determined by the repulsion of methyl
groups, which depends on the substituted position.
We would like to consider molecular overlapping with using
the results given from DFT calculations. Unpaired DDA+ electrons
occupy 2b3g orbitals (SOMOs) under D2h molecular symmetries,
The DFT method gives the optimized geometry for the singlet elec-
tron configuration of the dimeric (DDA+)2 system without counter
anions, while the structure for the triplet state never converged.
The optimized geometry of the dimer is similar to the closest pair
of DDA+ molecules in the DDA–Br crystal, and the closest carbon–
carbon distance is 3.027 Å, which is consistent to the observed
value of 3.16 Å. We confirmed that a dimer with an optimized
structure shows no imaginary frequencies in analytical vibrational
frequency computations. The total energy of the dimer at this opti-
mum point is larger than twice that of the monomer; however, it is
interesting to note that there exists a potential minimum at the
structure close to the experimental one. This is due to exchange
interactions of unpaired electrons on DDA+ molecules, which act
on attractive forces for adjacent molecules in the crystal. The
energy diagram and several molecular orbitals of the DDA+ dimer
are depicted as Fig. S6 in Electronic Supplementary Data. It is
hypothesized that this attractive force and the hydrogen-like
bonding between DDA+ and Brꢁ causes the alternate stacking of
DDA+ molecules in the one-dimensional column.
Fig. 3 shows the molecular structures of DDA and BA in the
DDA–BA crystal. DDA and BA molecules form one-dimensional col-
umns along the c-axis with constant distances of 3.25 Å. This short
contact shows strong interactions between adjacent DDA and BA
molecules in the column (see Fig. S7 in Electronic Supplementary
Data for molecular stacking and characteristic inter-atomic dis-
tances). Both molecules are nearly planar except for DDA hydrogen
atoms. In contrast with DDA–Br, positions for methyl group
hydrogen atoms could not be determined by difference Fourier
which are
ings of several molecular orbitals depicted as Fig. S4 in Electronic
Supplementary Data). The LUMO is also a -orbital, and transitions
p-orbitals (see the energy diagram and schematic draw-
p
between these two orbitals are optically allowed. Orbital charac-
ters are the same as those in Hartree–Fock calculations of PPD+
[37], except for contributions from hydrogen atoms to the LUMO.
It is obvious from the atomic contribution to the SOMO that the
most effective interaction between SOMOs of adjacent molecules
is achieved when molecules stack parallel upon each other. When
considering the repulsion of substituted methyl groups, the stack-
ing manner of DMPD–Br would be the next effective structure.
TMPD–Br shows the same molecular overlap to DMPD–Br; how-
ever, repulsion from dimethylamino groups distances these mole-
cules. These two types of configurations in PPD+ dimers have been
investigated by ab initio calculations [12]. A third type of configu-
ration is seen with DDA–Br, and the closest pair of DDA+ molecules
was deposited as Fig. S5 in Electronic Supplementary Data. Molec-
ular overlapping is less effective in comparison to DMPD–Br due to
repulsions in methyl groups substituted to central carbon rings.

T. Takaiwa et al. / Journal of Molecular Structure 1083 (2015) 260–267
265
hν = 9.18 eV
DDA-CA
DDA-BA
DDA+ calcd.
DDA-Cl
DDA-CA
DDA-Br
DDA-BA
hν = 7.75 eV
DDA-Cl
DDA-Cl
DDA-CA
DDA-Br
DDA-Br
DDA-BA
10
9
8
7
6
5
4
3
(DDA+)2 calcd.
Binding Energy / eV
4000 3500 3000 2500 2000 1500 1000
Wavenumber / cm−1
500
Fig. 5. UPS spectra of the CT complexes and halide salts measured at 7.75 eV (lower
panel) and 9.18 eV (upper panel) in photon energies (see Figs. S8–11 in Electronic
Supplementary Data for spectra obtained at other photon energies). The measure-
ments were performed with the thin film, which was obtained by scrubbing the
sample on the polished copper substrate in air.
Fig. 4. IR absorption spectra of the CT complexes and halide salts depicted by blue
and red curves, respectively. The results of the DFT calculations for DDA+ and
(DDA+)2 are also shown by black curves, where the scaling factor of 0.958 was used
to fit both experimental and simulated spectra. The peaks marked by arrows could
be assigned to carbonyl asymmetric stretching vibrations of BA and CA, which were
used in estimation for amounts of charge transfer (see text). (For interpretation of
the references to colour in this figure legend, the reader is referred to the web
version of this article.)
the same factor was used for simulating the theoretical spectrum
of DDA+ in the DDA–CA and DDA–BA complexes as shown in
Fig. 4. The N–H stretching vibrations were observed at the reduced
frequencies as compared to those in the simulated spectra for
DDA+ and (DDA+)2. This reduction would be due to hydrogen bond-
ing discussed above. The situations of the hydrogen-like bonding
are nearly similar to each other in two amino hydrogen atoms of
DDA–Br, but are quite different in those of DDA–BA, which showed
two features in each peak of the N–H stretching vibrations. It
would be concluded by comparison between observed and simu-
lated spectra that the carbonyl asymmetric stretching vibrations
syntheses; therefore, lengths and dihedral angles of C–H bonds
were fixed as 0.95 Å and 109.5°, respectively, and positions of
hydrogen atoms were refined using the riding model. As shown
in Fig. 3, two amino groups twist oppositely along an axis through
two nitrogen atoms. This is partly due to hydrogen bonds between
DDA and BA in the adjacent column. The central ring of DDA can be
thought to adopt a benzene-like structure, which is apparently dif-
ferent from the structure of DDA+ in the DDA–Br crystal. BA mole-
cules show strong quinoid structures, suggesting that DDA–BA is a
molecular CT complex. It is interesting to note that all of the short
inter-atomic contacts appear at atoms, which contribute largely to
the HOMO of DDA or the SOMO of DDA+. This may imply that the
large amount of the charge transfer is more likely to occur in adja-
cent molecules within the one-dimensional alternated column of
DDA and BA.
In order to characterize the nature of the DDA–CA and DDA–BA
complexes, the amounts of charge-transfer were estimated from
the carbonyl asymmetric stretching vibrations of chloranil and
bromanil, where we adopted an approximately linear relationship
between the position of this vibration and the degree of charge-
transfer [38,39]. Firstly, we compared the observed infrared spec-
trum of DDA–Br to that of the theoretical calculation for (DDA+)2,
where the scaling factor of 0.958 was used to fit both spectra to
each other. Consistency between them is fairly well except for
the region of N–H stretching vibration around 3300 cmꢁ1; then,
of CA and BA are assigned to peaks at 1691 cmꢁ1 and 1681 cmꢁ1
,
respectively, which are denoted by arrows in the figures. From
these vibrational peaks, the degrees of charge-transfer were esti-
mated to be 0.723 and 0.755 for DDA–CA and DDA–BA, respec-
tively. The value for DDA–CA is quite large as compared to 0.24
for the complex between tetrathiafulvalene (TTF) and CA at
300 K [38]. Although the observed values of the ionization poten-
tial have not been reported for DDA in a gaseous phase, the DFT
calculations estimated the ionization potential of DDA to be
6.29 eV and 5.77 eV for the vertical and adiabatic values, respec-
tively, which is about 0.6 eV smaller than the observed vertical
and adiabatic values of 6.92 eV and 6.4 eV, respectively, for TTF
in a gaseous phase [40–42]. This small ionization potential of
DDA and sufficient overlapping of electron donor and accepter
molecules in the one-dimensional column would be attributed to
the reason for the relatively large degree of charge-transfer.
The electronic structures of the obtained products were investi-
gated by UPS. The UPS spectra measured at 7.75 and 9.18 eV of
photon energies are compared in Fig. 5, and those measured at

266
T. Takaiwa et al. / Journal of Molecular Structure 1083 (2015) 260–267
Table 5
DDA and BA molecules alternated. X-ray analyses showed that
obtained products of DDA–BA were molecular crystals and DDA
molecules adopted benzenoid structures, while the amount of
the charge-transfer was estimated to be about 0.7 from infrared
spectra.
The valence electronic structures of DDA complexes were inves-
tigated by UPS, and the results were compared with the theoretical
calculations. Two spectral features were observed at 6 eV and 8 eV
in the low binding energy region for all compounds with and with-
out dimerization. The DFT calculations showed that the differences
between the first and the second ionization energy are nearly con-
stant during dimerization of DDA+, which is consistent with the
UPS observations.
Finally, results from this report have two additional implica-
tions on the synthesis of charge-transfer complexes with haloanils,
which are commonly used as electron acceptors: (1) special atten-
tion must be paid for contaminating halide ions when characteriz-
ing reaction products and in discussing their properties, and (2) a
new moderate synthetic root for halide complexes is now available
when using haloanils.
Calculated values of the vertical and adiabatic ionization potentials (Iv and Ia) and the
electron affinity (EA) in eV. The first two values are shown for Iv.
a
a
b
Compounds
EA
Iv
Ia
DDA
ꢁ2.0
5.6
8.6
6.5
11.2
13.1
7.8
12.6
14.6
6.0
11.0
–
DDA+
(DDA+)2
a
Vertical values were calculated by the
Adiabatic values were estimated from the difference in the total energy.
D
SCF method.
b
other photon energies are deposited as Figs. S8–11 in Electronic
Supplementary Data. All samples studied here showed the spectral
features around 6 eV and 8 eV, indicating that these features could
be assigned to the spectral structures along with ionization of the
common components. It should be noted that any spectral differ-
ences were not observed for two crystalline types of products, indi-
cating that the valence electronic structures of the DDA complexes
are not much affected along with dimerization. This would be also
confirmed from the fact that the UV–visible absorption spectra of
both crystal types show a similar pattern to each other (see
Fig. S12 in Electronic Supplementary Data). The calculated values
of the ionization potential are summarized in Table 5, where the
values for DDA+ and (DDA+)2 were calculated without any charge
compensation of counter anions. The first structure at 6 eV can
be assigned to ionization from the SOMO (3b3g) of DDA+ or the
HOMO of (DDA+)2, which is constructed by a bonding-type combi-
nation between the SOMOs of two DDA+ moieties in the dimer. It
might be concluded that the second feature around 8 eV originates
in ionization from the molecular orbital (2b2g) next to the SOMO of
DDA+ or the molecular orbital next to the HOMO of (DDA+)2. The
values of these two ionization energies were estimated by the
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molstruc.2014.
12.005.
References
[1] P.L. Nordio, Z.G. Soos, H.M. McConnell, Ann. Rev. Phys. Chem. 17 (1966) 237.
and references therein.
[2] K. Yokoyama, S. Maeda, Chem. Phys. Lett. 48 (1977) 59.
[3] E.E. Ernstbrunner, R.B. Girling, W.E.L. Grossmann, E. Mayer, K.P.J. Williams, R.E.
Hester, J. Raman Spectrosc. 10 (1981) 161.
[4] L. Michaelis, S. Granick, J. Am. Chem. Soc. 65 (1943) 1747.
[5] K.H. Hausser, Z. Naturforsch. A11 (1956) 20.
[6] K.H. Hausser, J.N. Murrell, J. Chem. Phys. 27 (1957) 500.
[7] K. Uemura, S. Nakayama, Y. Seo, K. Suzuki, Y. Ooshika, Bull. Chem. Soc. Jpn. 39
(1966) 1348.
[8] K. Kimura, H. Yamada, H. Tsubomura, J. Chem. Phys. 48 (1968) 440.
[9] S. Nakayama, K. Suzuki, Bull. Chem. Soc. Jpn. 46 (1973) 3694.
[10] G. Rauhut, T. Clark, J. Am. Chem. Soc. 115 (1993) 9127.
[11] A.M. Brouwer, J. Phys. Chem. A 101 (1997) 3626.
[12] T. Kamisuki, C. Hirose, Spectrochim. Acta A56 (2000) 2141.
[13] M.M. Labes, R. Sehr, M. Bose, J. Chem. Phys. 33 (1960) 868.
[14] M. Shwarz, H.W. Davis, B.J. Dohrlansky, J. Chem. Phys. 40 (1964) 3257.
[15] D.D. Eley, H. Inokuchi, M.R. Willis, Disc. Faraday Soc. 28 (1959) 54.
[16] P.L. Kronick, M.M. Labes, J. Chem. Phys. 35 (1960) 2916.
[17] Y. Matsunaga, J. Chem. Phys. 41 (1964) 1609.
D
SCF calculations to be 11.2 eV and 12.6 eV for DDA+ and to be
13.1 eV and 14.6 eV for (DDA+)2, respectively. It should be noted
from the calculations that the differences between the first and
the second ionization energies are nearly constant even if DDA+
molecules form dimers. This is consistent with the experimental
observations for the two types of reaction products between DDA
and haloanile. In passing, we failed to observe the UPS spectra of
DDA in the solid state because of its volatile nature and maybe of
the low ionization energy. Actually, as discussed above, it might
be expected from the calculated value of the ionization potential
that DDA shows the lower ionization threshold than TTF, of which
the value was reported to be 5.0 eV for the solid film at ꢁ38 °C
[43,44].
[18] Y. Matsunaga, Nature 205 (1965) 72.
[19] M.M. Labes, R. Sehr, M. Bose, J. Chem. Phys. 32 (1960) 1570.
[20] A. Ottenberg, C.J. Hoffman, J. Osiecki, J. Chem. Phys. 38 (1963) 1898.
[21] J.L. de Boer, A. Vos, Acta Cryst. B24 (1968) 720.
Conclusion
[22] T. Kimura, M. Sumimoto, S. Sakaki, H. Fujimoto, Y. Hashimoto, S. Matsuzaki,
Chem. Phys. 253 (2000) 125.
[23] A.D. Becke, Phys. Rev. A 38 (1988) 3098.
[24] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[25] J.P. Perdew, Y. Wang, Phys. Rev. B 45 (1992) 13244.
[26] J.P. Perdew, J.A. Chevary, S.H. Vosko, K.A. Jackson, M.R. Pederson, D.J. Singh, C.
Fiolhais, Phys. Rev. B 46 (1992) 6671.
[27] J.P. Perdew, J.A. Chevary, S.H. Vosko, K.A. Jackson, M.R. Pederson, D.J. Singh, C.
Fiolhais, Phys. Rev. B 48 (1993) 4978.
[28] J.P. Perdew, K. Burke, Y. Wang, Phys. Rev. B 54 (1996) 16533.
[29] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery, Jr., J.E. Peralta, F. Ogliaro,
M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, T. Keith, R.
Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J.
Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken,
C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R.
Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski,
G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B.
Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09 (Revision C.01),
Gaussian Inc, Wallingford CT, 2010.
Herein, we reported the crystal structures of two reaction prod-
ucts of DDA and BA for the first time. Both products were obtained
from the same reaction pot, but at different reaction times. One
structure was a complex salt of DDA–Br, which was crystallized
simultaneously with decomposing BAꢁ anions. Another structure
was a 1:1 charge-transfer complex between DDA and BA, DDA–
BA. The similar products were obtained with using CA instead of
BA.
DDA molecules from DDA–Br crystals showed quinoid struc-
tures and stacked in a one-dimensional column with alternating
molecular distances. The closest pair of DDA molecules was consis-
tent with the optimized geometry of DFT calculations. Steric repul-
sion in methyl groups caused DDA+ molecules to stack in a third
type of dimeric structure, which was different from that of
DMPD–Br and TMPD–Br. Brꢁ anions played a role in combining col-
umns of DDA through hydrogen bonds. Component molecules of
DDA–BA also stacked in a one-dimensional column, in which

T. Takaiwa et al. / Journal of Molecular Structure 1083 (2015) 260–267
267
[30] R. Dennington, T. Keith, J. Millam, Gauss View Version 5, Semichem Inc.,
Shawnee Mission KS, 2009.
[38] A. Girlando, F. Marzola, C. Pecile, J.B. Torrance, J. Chem. Phys. 79 (1983) 1075.
[39] A. Paineli, A. Girlando, J. Chem. Phys. 84 (1986) 5655.
[31] R. Foster, T.J. Thomson, Trans. Faraday Soc. 58 (1962) 860.
[32] R.F.C. Claridge, J.E. Willard, J. Am. Chem. Soc. 87 (1965) 4992.
[33] T. Nogami, T. Yamaoka, K. Yoshihara, S. Nagakura, Bull. Chem. Soc. Jpn. 44
(1971) 380.
[40] R. Gleiter, E. Schmidt, D.O. Cowan, J.P. Ferraris, J. Electron Spectrosc. Relat.
Phenom. 2 (1973) 207.
[41] A.J. Berlinsky, J.F. Carolan, L. Weiler, Can. J. Chem. 52 (1974) 3373.
[42] R. Gleiter, M. Kobayashi, J.S. -Larsen, J.P. Ferraris, A.N. Bloch, K. Bechgaad, D.O.
Cowan, Ber. Bunsenges. Physik. Chem. 79 (1975) 1218.
[34] J.L. de Boer, A. Vos, Acta Cryst. B28 (1972) 839.
[35] J. Tanaka, N. Sakabe, Acta Cryst. B24 (1968) 1345.
[36] J. Tanaka, M. Mizuno, Bull. Chem. Soc. Jpn. 42 (1969) 1841.
[37] D.M. Chipman, Q. Sun, G.N.R. Tripathi, J. Chem. Phys. 97 (1992) 8073.
[43] P. Nielsen, A.J. Epstein, D.J. Sandman, Solid State Commun. 15 (1974) 53.
[44] P. Nielsen, D.J. Sandman, A.J. Epstein, Solid State Commun. 17 (1974) 1067.