Synthesis, ionisation potentials and electron affinities of hexaazatrinaphthylene derivatives

Article abstract of DOI:10.1002/chem.200601298

Several hexaazatrinaphthylene derivatives and a tris-(thieno) hexaazatriphenylene derivative have been synthesised by reaction of the appropriate diamines with hexaketocyclohexane. The crystal structure of 2,3,8,9,14,15-hexachloro-5,6,11,12,17,18-hexaazatrinaphthylene has been determined by X-ray diffraction; this reveals a molecular structure in good agreement with that predicted by density functional theory (DFT) calculations and π-stacking with an average spacing between adjacent molecular planes of 3.18 A. Solid-state ionisation potentials have been measured by using UV photoelectron spectroscopy and fall in the range of 5.99 to 7.76 eV, whereas solid-state electron affinities, measured using inverse photoelectron spectroscopy, vary in the range -2.65 to -4.59 eV. The most easily reduced example is a tris(thieno)hexaazatriphenylene substituted with bis(trifluoro-methyl)phenyl groups; DFT calculations suggest that the highly exothermic electron affinity is due both to the replacement of the outermost phenylene rings of hexaazatrinaphthylene with thieno groups and to the presence of electron-withdrawing bis(trifluoro-methyl)phenyl groups. The rather exothermic electron affinities, the potential for adopting π-stacked structures and the low intramolecular reorganisation energies obtained by DFT calculations suggest that some of these molecules may be useful electron-transport materials.

Full text of DOI:10.1002/chem.200601298

FULL PAPER
DOI: 10.1002/chem.200601298
Synthesis, Ionisation Potentials and Electron Affinities of
Hexaazatrinaphthylene Derivatives
Stephen Barlow,[a, b] Qing Zhang,[a] Bilal R. Kaafarani,[a, b, c] Chad Risko,[a] Fabrice Amy,[d]
Calvin K. Chan,[d] Benoit Domercq,[e] Zoya A. Starikova,[f] Mikhail Yu. Antipin,[f, g]
Tatiana V. Timofeeva,[g] Bernard Kippelen,[e] Jean-Luc BrØdas,[a] Antoine Kahn,[d] and
Seth R. Marder*[a, b]
Abstract: Several hexaazatrinaphthy-
lene derivatives and tris-
(thieno)hexaazatriphenylene derivative
have been synthesised by reaction of
tion potentials have been measured by
using UV photoelectron spectroscopy
and fall in the range of 5.99 to 7.76 eV,
whereas solid-state electron affinities,
measured using inverse photoelectron
spectroscopy, vary in the range ꢀ2.65
to ꢀ4.59 eV. The most easily reduced
example is a tris(thieno)hexaazatriphe-
nylene substituted with bis(trifluoro-
methyl)phenyl groups; DFT calcula-
tions suggest that the highly exother-
mic electron affinity is due both to the
replacement of the outermost phenyl-
ene rings of hexaazatrinaphthylene
with thieno groups and to the presence
of electron-withdrawing bis(trifluoro-
methyl)phenyl groups. The rather exo-
thermic electron affinities, the potential
for adopting p-stacked structures and
the low intramolecular reorganisation
energies obtained by DFT calculations
suggest that some of these molecules
may be useful electron-transport mate-
rials.
a
ACHTREUNG
the
hexaketo
appropriate
cyclohexane.
diamines
The
with
crystal
A
structure of 2,3,8,9,14,15-hexachloro-
5,6,11,12,17,18-hexaazatrinaphthylene
has been determined by X-ray diffrac-
tion; this reveals a molecular structure
in good agreement with that predicted
by density functional theory (DFT) cal-
culations and p-stacking with an aver-
age spacing between adjacent molecu-
lar planes of 3.18 Š. Solid-state ionisa-
Keywords: electron affinities · elec-
tron transport · heterocycles · ioni-
zation potentials · pi stacking
Introduction
attention as materials for organic electronic applications.
These compounds can, depending on the choice of substitu-
ent, form films with a wide range of morphologies, including
crystalline, columnar discotic liquid-crystalline and amor-
Derivatives of 5,6,11,12,17,18-hexaazatrinaphthylene (diqui-
noxalino[3,3-a:2’,3’-c]phenazine, 1) have recently attracted
[a] Dr. S. Barlow, Dr. Q. Zhang, Dr. B. R. Kaafarani, Dr. C. Risko,
Prof. J.-L. BrØdas, Prof. S. R. Marder
[e] Dr. B. Domercq, Prof. B. Kippelen
School of Electrical and Computer Engineering and
Center for Organic Photonics and Electronics
Georgia Institute of Technology, Atlanta, GA 30332 (USA)
School of Chemistry and Biochemistry and
Center for Organic Photonics and Electronics
Georgia Institute of Technology, Atlanta, GA 30332 (USA)
Fax :( +1)404-894-5909
[f] Dr. Z. A. Starikova, Prof. M. Y. Antipin
Institute of Organoelement Compounds
Russian Academy of Sciences, Moscow (Russia)
E-mail:seth.marder@chemistry.gatech.edu
[b] Dr. S. Barlow, Dr. B. R. Kaafarani, Prof. S. R. Marder
Department of Chemistry
[g] Prof. M. Y. Antipin, Prof. T. V. Timofeeva
Department of Natural Sciences
University of Arizona, Tucson, AZ 85721 (USA)
New Mexico Highlands University, Las Vegas, NM 87701 (USA)
[c] Dr. B. R. Kaafarani
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Current address:
Department of Chemistry
American University of Beirut, Beirut (Lebanon)
[d] Dr. F. Amy, C. K. Chan, Prof. A. Kahn
Department of Electrical Engineering
Princeton University, Princeton, NJ 08544 (USA)
Chem. Eur. J. 2007, 13, 3537 – 3547
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3537

phous. Most work has concerned esters derived from the
to be ꢀ4.59 eV, considerably more exothermic than that of
1.[10] In addition, this material can be n-doped with cobalto-
cene, which greatly enhances electron injection into the ma-
terial.[10]
2,8,14- and 2,8,15-tricarboxylic acid (2)[1,2] or 2,3,8,9,14,15-
hexa(alkylsulfanyl) (3)[3–6] and 2,3,8,9,14,15-hexa
(alkoxy) de-
A
rivatives (4).[7,8] Analogues with more extensive p-systems,
such as 5, have also been reported.[9]
One-dimensional charge-carrier mobilities have been ob-
tained by using the pulse-radiolysis time-resolved microwave
conductivity technique for the crystalline and liquid-crystal-
line phases formed by the hexa(alkylsulfanyl) derivatives
(3),[4,6] with the largest value so far, 0.9 cm2 Vꢀ1 sꢀ1, being
found at 858C for crystalline 3a (R=n-C10H21);[6] this tech-
nique measures the sum of hole and electron mobilities and
computational work suggests that both carriers contribute to
the observed mobility in these materials.[4] Recently we in-
vestigated the pentafluorobenzyl triester, 2b, and, using the
space-charge limited current technique, determined mobili-
ties of 0.07 and 0.02 cm2 Vꢀ1 sꢀ1 for the Cs isomer and as-syn-
thesised isomer mixture, respectively.[2] Energy-level consid-
erations based on electrochemical measurements suggest
that hexaazatrinaphthylene derivatives are likely to function
as electron-transport materials and that the mobilities we re-
ported for 2b in reference [2] are likely to represent elec-
tron mobilities. Further support for the ease of electron in-
jection relative to hole injection comes from direct measure-
ments of the solid-state ionisation potentials (IPs) of 1[2] and
In this paper we report
more fully on the solid-state
photoelectron (PES) and in-
verse-photoelectron
(IPES)
spectra of 1 and 6 and com-
pare these with the PES and
IPES spectra of the halogenat-
ed hexaazatrinaphthylenes
7
and 9 and of a representative
example of the hexa(alkylsul-
fanyl)hexaazatrinaphthylenes,
3a. The experimental results
are compared to quantum-
chemical estimates for these
and additional compounds.
Furthermore, we report the
syntheses of 6 and 9 for the
first time, along with the crys-
tal structure of 7; the p-stack-
ing seen in this structure sug-
gests that vapour-deposited 7
may be an interesting charge-
transport material.
3a[5] as 6.6 and 5.9 eV, respectively, and of the solid-state
electron affinity (EA) of 1 as ꢀ2.4 eV.[2] Recently, we re-
Results and Discussion
ported the EA of tris{2,5-bis(3,5-bis-trifluoromethyl-phenyl)-
A
thieno}[3,4-b,h,n]-1,4,5,8,9,12-hexaazatriphenylene (6; i.e.,
an analogue of a hexaazatrinaphthylene in which the outer
benzene rings are replaced with thiophene rings fused to the
Synthesis:Compounds 1 and 7–9 were synthesised from re-
action of the appropriate o-phenylene diamine with hexake-
tocyclohexane octahydrate in the presence of acetic acid
(Scheme 1).[11,12] The diamines were either commercially
hexaazatriphenyl
A
3538
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 3537 – 3547

Hexaazatrinaphthylene Derivatives
FULL PAPER
graphic symmetry (C1), mole-
cules of 7 appear highly sym-
metrical, with approximate D3h
symmetry.
Table 1 summarises the key
bond lengths (averaged over
chemically equivalent bonds
for each structure), defined ac-
cording to Figure 2, from the
structures of two chloroform
Scheme 1. General synthesis of hexaazatrinaphthylene derivatives used for the preparation of 1 (X=Y=H), 7
(X=Cl, Y=H), 8 (X=F, Y =H) 9 (X=Y=F).
available (for 1 and 7) or were synthesised using literature
procedures (for 8 and 9).[13–16] Compound 3a was obtained
from 7 according to the literature procedure.[6] These com-
pounds were very poorly soluble in organic solvents and
were, therefore, purified by gradient sublimation under high
vacuum.[17] The fluorinated derivatives 8 and 9 were espe-
1
cially insoluble and H and 19F NMR spectra could only be
obtained in the presence of strong acids (trifluoroacetic acid
or sulfuric acid).
Compound 6 was obtained in an analogous fashion to the
hexaazatrinaphthylene derivatives from the reaction of
hexaketocyclohexane octahydrate with the diamine S3
A
(Scheme 2), which was in turn obtained from the corre-
sponding dinitro compound S2, which was synthesised using
the Stille coupling of 2,5-dibromo-3,4-dinitrothiophene (S1)
and
3,5-bis(trifluoromethyl)phenyltributylstannane.
As
might be anticipated from the presence of the bis(trifluoro-
methyl)phenyl substituents, 6 is more soluble in common or-
ganic solvents than 7–9 and could, therefore, be purified by
column chromatography.
Figure 1. Molecular structure of 2,3,8,9,14,15-hexachloro-5,6,11,12,17,18-
hexaazatrinaphthylene (7) from X-ray crystallography.
Molecular geometry:Crystals of 2,3,8,9,14,15-hexachloro-
5,6,11,12,17,18-hexaazatrinaphthylene (7) suitable for X-ray
structure determination were obtained from gradient
vacuum sublimation. Crystallographic data and details of
the crystal structure solution and refinement are given in
the Experimental Section, with complete information pro-
vided in the Supporting Information; the molecular struc-
ture is shown in Figure 1. As in the previously reported
structures of the solvates of 1,[12,18] the molecules are essen-
tially planar (mean deviation of the C and N atoms from the
RMS molecular plane of 0.038 Š) and, despite low crystallo-
solvates of 1[12,18,19] and from that of 7. Also included in
Table 1 are the corresponding bond lengths from density
functional theory (DFT) calculations (B3LYP/6–31G**) for
the gas-phase molecules of 1, 6–9 and for the related model
compounds benzene (I), pyrazine (II), triphenylene (III)
and 1,4,5,8,9,12-hexaazatriphenylene (IV; Figure 2). The
table shows that the computational geometries are in good
agreement with the three experimental structures and that
the structural difference between 1 and 7 and the other
hexaazatrinaphthylene derivatives are relatively minor. Sev-
A
eral resonance forms can be envisaged for hexaazatrinaph-
thylenes; two of these are indicated in Figure 2 and differ in
Scheme 2. Synthetic route for compound 6.
Chem. Eur. J. 2007, 13, 3537 – 3547
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3539

S. R. Marder et al.
Table 1. Key bond lengths [defined in Figure 2 in Š] from B3LYP-31G** DFT calculations and X-ray crystallography (in italics) for the p-systems of
hexaazatrinaphthylenes and related structures.
[12]
[18]
I
II
III
IV
1·CHCl3
1·4CHCl3
1
Cs-2a[a]
6
6’
6’’
7
7
8
9
a
b
c
d
e
f
–
–
1.467
1.421
1.413
1.383
1.401
–
1.465
1.417
1.348
1.325
1.407
–
1.477(3)
1.424(4)
1.328(1)
1.360(1)
1.419(4)
1.414(1)
1.359(3)
1.405(3)
1.479(6)
1.425(2)
1.325(2)
1.359(6)
1.424(7)
1.413(2)
1.371(13)
1.414(6)
1.478
1.439
1.324
1.353
1.434
1.421
1.375
1.422
1.478
1.439
1.478
1.453
1.318
1.355
1.449
1.401
1.736
–
1.478
1.453
1.318
1.355
1.452
1.402
1.739
–
1.483
1.459
1.316
1.359
1.451
1.386
1.718
–
1.474(6)
1.427(6)
1.322(2)
1.351(1)
1.423(3)
1.412(1)
1.363(1)
1.426(2)
1.477
1.438
1.324
1.351
1.431
1.418
1.375
1.432
1.477
1.436
1.325
1.349
1.437
1.422
1.367
1.424
1.477
1.438
1.323
1.346
1.438
1.421
1.375
1.418
1.396
1.396
1.338
1.396
1.323–1.324
1.352–1.353
1.434
1.418–1.421
1.373–1.380
1.427
–
–
–
–
–
–
–
–
–
–
g
h
–
–
–
–
[a] For this compound, the reduced symmetry means there is more than one chemically inequivalent type of c, d, f and g bond length. Very similar bond
lengths are found for Cs-2b.
to structure A shown above for hexaazatrinaphthylenes) can
be drawn for the thieno species without separating charges.
Accordingly, the bond lengths in the thiophene rings of 6, 6’
and 6’’ show similar patterns to that of thiophene (at the
ꢀ
ꢀ
same level of theory S C, C=C and C C bond lengths are
1.735, 1.367 and 1.429 Š, respectively), with the differences
between the different compounds presumably arising from
steric effects associated with the aryl groups. These steric ef-
fects also result in a deformation of the core p-systems of 6
and 6’; a slight twist of the arms reduces the steric interac-
tions between the substituted aryl groups of neighbouring
arms.
Orbital structure, radical anion and cation geometries, and
reorganisation energies:We have also calculated the geome-
tries of the radical cations and anions of 1, Cs-2a, Cs-2b, 6’,
6’’ and 7–9;[20] full details are given in the Supporting Infor-
mation. Reduction of 1 leads to moderate bond-length
changes throughout the structure while maintaining D3h
symmetry; the largest changes are decreases of 0.010 Š in
the h bond lengths and increases of 0.008, 0.009 and 0.009 Š
in the c, d and g bond lengths, respectively, consistent with
the bonding or antibonding character of the LUMO across
the bonds in question. The LUMO of 1 (Figure 3) has a2’’
symmetry and is qualitatively similar to that previously re-
ported for 3 (R=H).[6] Qualitatively similar LUMOs are ob-
tained for Cs-2a, Cs-2b, 6’, 6’’ and 7–9, with the addition of
small coefficients on the p-accepting carbonyl groups of the
ester derivatives 2a and 2b (see Supporting Information for
reference [2]). The calculated changes in the bond lengths
between the neutral molecule and the radical anion for com-
pounds 2a, 2b, 6’, 6’’ and 7–9 are similar to those for 1.
The B3LYP/6-31G** HOMO of 1 (Figure 3) is a doubly
degenerate orbital of e’’ symmetry, as is the previously re-
ported qualitatively similar HOMO calculated for 3 (R=
H),[5] while the non-degenerate HOMOꢀ1 has a1’’ symme-
Figure 2. Top:general structures of arenes, triphenylenes, hexaazatri-
naphthylenes and tris(thieno)hexaazatriphenylenes showing labelling of
chemically inequivalent bonds as used in Table 1. Below:structures of
benzene (I), pyrazine (II), triphenylene (III) and hexaazatriphenylene
(IV) and two of the resonance structures possible for hexaazatrinaphthy-
lenes.
ACHTREUNG
which of the six-membered rings is formally “aromatic”.
While no single resonance structure adequately accounts for
the pattern of bond lengths, structures of type A (Figure 2)
are clearly important, as suggested by:the shortest bonds
being between the central six-membered ring and the nitro-
gen atoms (c, Figure 2); the bonds between the nitrogen
atoms and the outermost rings (d) being somewhat longer
than c; and the bonds within the central ring (a and b) being
very long. This picture is similar to that calculated for tri-
phenylene (III) or hexaazatriphenylene (IV) in that the cen-
tral ring has rather long bonds; however, these bonds are
longer in the hexaazatrinaphthylenes, while the c bonds are
shorter and the relative lengths of c and d bonds are re-
versed. The bond lengths calculated for the tris-
try. Accordingly, the structure of 1 +, in which an electron is
C
A
removed from one of these degenerate orbitals, is subject to
Jahn–Teller distortion. The optimised structure for the
cation does indeed lack the D3h symmetry of the neutral
molecule; the most dramatic bond-length changes are the
shortening of two of the c type bonds by 0.017 Š, one in
each of two arms (see Supporting Information for full de-
tails). The hexahalo derivatives 7 and 8 also have degener-
type resonance structures dominate even more strongly in
these compounds; in particular, the c bonds are slightly
shorter and the b and d bonds are slightly longer than the
corresponding bonds in the hexaazatrinaphthylenes. These
observations can be rationalised by using a simple valence-
bond picture; only a single resonance structure (analogous
3540
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 3537 – 3547

Hexaazatrinaphthylene Derivatives
FULL PAPER
sum of l1, the energy corre-
sponding to relaxation on the
neutral potential-energy sur-
face from the radical-ion state
geometry to the relaxed neu-
tral geometry and l2, the relax-
ation on the radical-ion poten-
tial energy surface from the
neutral geometry to the re-
laxed radical-ion geometry; the
calculated values are given in
Table 2. There is not a particu-
larly clear dependence of the
reorganisation
energy
on
chemical structure, although
for 1, 7 and 8, in which the rad-
ical cations are Jahn–Teller dis-
torted (vide supra), the values
for hole transfer exceed those
Figure 3. Frontier orbitals calculated at the B3LYP/6-31G** level for 1 (left) and 6’’ (right).
Table 2. Intramolecular reorganisation energies [eV] for hole and elec-
tron exchange calculated at the B3LYP/6–31G** level.[a]
ate HOMOs and undergo Jahn–Teller distortion on oxida-
tion. However, in the dodecafluoro species, 9, the a1’’ orbital
cation/neutral
neutral/anion
is the HOMO, albeit only 0.03 eV higher in energy than the
+
l1
l2
li
l1
l2
li
C
e’’ degenerate HOMO-1; accordingly, 9 retains D3h sym-
1
0.074
0.102
0.056
0.038
0.035
0.083
0.100
0.087
0.063
0.096
0.052
0.039
0.033
0.075
0.097
0.083
0.137
0.198
0.108
0.078
0.068
0.158
0.197
0.169
0.048
0.087
0.100
0.061
0.046
0.058
0.076
0.082
0.047
0.083
0.096
0.067
0.044
0.058
0.077
0.082
0.095
0.170
0.196
0.129
0.091
0.117
0.153
0.164
metry. The general appearance of the a1’’ and e’’ orbitals is
very similar for the halogenated derivatives and for 1, al-
though out-of-phase contributions from the p-donor halogen
atoms are seen in 7–9. In the case of Cs-2a and Cs-2b, the
HOMO is related to that of 1 (see Supporting Information
for reference [2]); inevitably the low symmetry of the substi-
tution factor means the degeneracy is lifted, albeit by only
2a
2b
6’
6’’
7
8
9
[a] The total intramolecular reorganisation energy, li, is the sum of the
relaxation energies of the neutral molecule, l1 and of the charged species,
l2, on electron transfer.
0.02–0.03 eV, and oxidation leads to Jahn–Teller-like dis-
+
C
tortion similar to that seen for
1
.
The tris-
A
broadly similar frontier orbitals to the hexaazatrinaphthy-
lene derivatives (shown for 6’’ in Figure 3; orbitals for 6 are
shown in the Supporting Information and are broadly simi-
lar, with only small coefficients on the aryl groups). The
frontier orbital ordering in 6, 6’ and 6’’ is the same as that in
9, but with a larger separation between the non-degenerate
HOMO and the degenerate HOMOꢀ1 and, accordingly,
for electron transfer. However, in general, the reorganisa-
tion energies for both hole and electron transfer are rather
small compared to those calculated at the same level of
theory for many other commonly used transport materials.
+
C
For example, a value of 0.29 eV was reported for TPD
/
TPD {TPD=N,N’-diphenyl-N,N’-di
U
+
+
C
C
6’ and 6’’ are predicted to retain threefold symmetry.
The geometric changes associated with oxidation and re-
duction determine the intramolecular components of the re-
organisation energies for the electron-exchange reactions
between the neutral molecules and their radical cations and
anions and, therefore, the barrier to hole or electron trans-
port between molecules in the solid state. According to
Marcus theory[21] and as shown pictorially elsewhere,[22] this
barrier is l/4, in which l is the sum of the internal reorgani-
sation energy (li) and external reorganisation energy (ls),
which is associated with changes in polarisation and posi-
tions of the surrounding molecules. We have used B3LYP/6-
31G** calculations to evaluate li for hole and electron
transfer for 1, 2a, 2b, 6’, 6’’, 7, 8 and 9 from the adiabatic
potential surfaces.[20,23] Specifically, we calculated li as the
quinolato)aluminium},[25] and 0.38–0.61 and 0.49–0.56 eV for
hole and electron exchange, respectively, in 1,1-diaryl-
2,3,4,5-tetraphenylsiloles.[26] Indeed, the reorganisation
energy of 0.091 eV calculated for 6’’ +/6’’ is even smaller
than that calculated for pentacene +/pentacene (0.098 eV);
C
the reorganisation energy of 0.068 eV for 6’’/6’’ is even
lower. These values are consistent with the high mobilities
that have been reported in some hexaazatrinaphthylene de-
rivatives.[2,4,6]
Packing in the structure ofcompound 7 :The electronic
properties of molecular transport materials depend on the
manner in which the molecules are organised in the solid
state.[27–29] In particular, p-stacking, although not a guarantee
Chem. Eur. J. 2007, 13, 3537 – 3547
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3541

S. R. Marder et al.
of good wavefunction overlap,[30] is a feature of many high
mobility materials, including many columnar discotic liquid-
crystalline mesophases,[31,32] including hexaazatrinaphthylene
derivatives,[6,8] phthalocyanines and perylene-based materi-
als,[33] polythiophenes,[34] extended porphyrins[35] and some
pentacene derivatives.[36] Both molecular sheets and molecu-
lar p-stacks are found in the crystal structure of 7. The struc-
ture can be described in terms of molecular sheets parallel
to the (102) crystallographic plane; Figure 4 shows the top
Figure 4. Two views of the molecular packing in crystals of 7 emphasizing
its description in terms of sheets of molecules. The lower view is along
the (102) direction.
Figure 5. Two views of the p-stacked columns in the structure of 7.
Ionisation potential and electron affinity:When considering
the use of an organic material for charge-transport applica-
tions it is important to have an understanding of its solid-
state ionisation potentials (IPs) and electron affinities
(EAs). This understanding can help in identifying suitable
partner organic transport materials and inorganic electrode
materials and in establishing whether holes or electrons are
dominant carriers in a particular device situation. Estimates
have often been made based on comparison of the electro-
chemistry, sometimes in conjunction with optical data, of
the species of interest with that of compounds of known IP
and/or EA. Aside from the uncertainties arising from the as-
sumptions involved in these estimates,[43] complications may
arise due to the presence of aggregates, which NMR spec-
troscopy suggests are present for 2b[2] and species of type
3[6] at the concentrations typically employed for electro-
chemical measurements. Moreover, some of the halogenated
hexaazatrinaphthylenes are insufficiently soluble for electro-
chemical investigations.
view and side views of these sheets, the latter revealing they
are slightly corrugated with the molecular planes deviating
from the mean plane of the sheet. Within the sheets there
are some short intermolecular Cl···Cl, N···Cl and C···H con-
tacts (as short as 3.323, 3.232 and 2.781 Š, respectively; see
Supporting Information for details). There is also a large
number of intersheet contacts (see Supporting Information);
molecules in successive sheets form slipped stacks propagat-
ing along the crystallographic z axis. Figure 5 shows two
views of how the molecules are arranged in the stacks, with
the average spacing between adjacent molecular planes
being 3.18 Š. This p-stacking distance is somewhat shorter
than seen in p-stacks of 1·CHCl3 (3.48 Š);[12] the structure
of 1·4CHCl3 is quite different, with p-interactions within
centrosymmetric dimers with an interplanar distance of
3.31 Š.[18] It is also rather short compared to the correspond-
ing distance in many other p-stacked organic electronic ma-
terials,[36–41] including discotic columnar hexaazatrinaphthy-
lene materials such as 3a[6] and 4.[8] However, an equally
short p-stacking distance of 3.18 Š has been reported in a
high-carrier-mobility hydrogen-bonded hexaazatripheny-
lene-based columnar material.[42]
In contrast, photoelectron spectroscopy (PES) and in-
verse-photoelectron spectroscopy (IPES) of solid films pro-
vide direct measures of IP and EA, respectively. We ac-
quired PES and IPES data for vacuum-deposited films of 1,
6, 7 and 9 and for a spin-coated film of one of the hexa(al-
kylsufanyl) class of materials, 3a, for comparison. The PES
3542
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 3537 – 3547

Hexaazatrinaphthylene Derivatives
FULL PAPER
and IPES spectra are shown in Figure 6; in general, qualita-
tively similar densities of states are observed for 1, 6, 7 and
9, consistent with the qualitatively similar orbital pictures
suggested by our calculations (the small HOMO/HOMOꢀ1
splittings are not resolved in the PES spectra). The PES
spectra for 3a look rather different, being dominated by an
intense feature with a peak corresponding to an IP greater
than 10 eV (binding energy vs. EF of about ꢀ6 eV) which
dwarfs and/or obscures some of the hexaazatrinaphthylene
ionisations. This is similar to previously reported PES spec-
tra for 3 (R=C6H13, here the ratio of intensities between
the > 10 eV feature and the lowest energy ionisation is
somewhat reduced)[5] and also to spectra of silicon-bound
alkyl monolayers (see Supporting Information for compari-
son of the PES of 3a and one of these monolayers);[44,45] in
all these systems the intense feature is presumably due to
ionisations from the side chains. In both cases, however, the
lowest energy IP attributed to ionisation from the HOMO
can be resolved; our value of the corresponding adiabatic IP
for the lowest energy ionisation of 3a is in good agreement
with the value of 5.9 eV previously reported for its hexyl an-
alogue.[5] The adiabatic IPs and EAs, extracted from the
PES and IPES spectra, respectively, are given in Table 3,
along with DFT estimates of the adiabatic and vertical gas-
phase IPs and EAs. The computed values follow the experi-
mental trends, although, as expected, they are consistently
more endothermic than the observed values. Solid-state IPs
are known to be as much as 1–1.5 eV lower than experimen-
tal gas-phase values, due to stabilisation of the cation by po-
larisation of the surrounding molecules in the solid,[46,47] and
analogous effects can be anticipated to make the solid-state
EA values more exothermic than gas-phase values. In addi-
tion, p-stacking effects may affect the solid-state spectra, as
previously noted for PES spectra of 3 (R=C6H13).[5]
The solid-state IPs are all higher than those of the widely
used hole-transport materials, TPD[47] and NPD {NPD=
N,N’-diphenyl-N,N’-di(1-naphthyl)(biphenyl-4,4’-diamine)}
(5.5 eV).[48,49] With the exception of 3a, the IPs are also in
excess of that of the commonly used electron-transport ma-
terial AlQ3 (5.93 eV),[47] suggesting they should be more ef-
fective as “hole-blocking” ma-
terials.
The EA of 1 is somewhat
more exothermic than that of
the widely used electron-trans-
port material, AlQ3 (ꢀ2.0 to
ꢀ2.5 eV)[50] and also more exo-
thermic than those measured
for 1,1-diaryl-2,3,4,5-tetraphe-
nylsiloles (ꢀ1.5 to ꢀ2.4 eV).[26]
As one would expect, the halo-
genated derivatives 7 and 9 are
both significantly more easily
reduced and less readily oxi-
dised than the parent heterocy-
cle, the IPs and EAs between
the halogenated and non-halo-
gentated materials differing by
around 1 eV. The similarity of
the experimental EAs of 7 and
9 might appear surprising at
first sight, given the greater
electronegativity of fluorine
relative to that of chlorine
(Pauling electronegativities are
3.98 and 3.16, respectively[51]).
However, this result closely
parallels IPES data showing
films of 1,3,5-trichlorobenzene
and hexafluorobenzene to
have rather similar EAs;[52]
thus, in both that study and
ours twice as many fluorine
substituents are required to
affect the EA to the same
extent as chlorine substituents,
presumably reflecting the inter-
Figure 6. PES and IPES spectra for some hexaazatrinaphthylene and tris
tives.
A
Chem. Eur. J. 2007, 13, 3537 – 3547
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3543

S. R. Marder et al.
Table 3. Adiabatic ionisation potentials (IPs) and electron affinities
(EAs) for some hexaazatrinaphthylene derivatives obtained using PES
and IPES, together with adiabatic and vertical values from B3LYP/6-
31G** calculations.
work in concert to make the EA of 6 1.3 eV more electro-
negative than that of 6’’.
IP [eV]
DFT
adiabatic vertical adiabatic adiabatic
EA[a] [eV]
DFT
vertical adiabatic
Summary
PES
IPES
In summary, we have synthesised several known and new
hexaazatrinaphthylenes and a related tris(thieno)hexaazatri-
phenylene. We have shown that both the solid-state IPs and
EAs of these materials can be tuned over ranges of nearly
2 eV through substitution. The results of DFT calculations
indicate rather small reorganisation energies in these materi-
als, suggesting low barriers to both electron and hole trans-
port. The low barriers and highly exothermic electron affini-
ties point towards potential applications as electron-trans-
port materials in organic electronics. Finally, we have shown
that one example adopts a p-stacked structure with a rather
short p-stacking distance of 3.18 Š.
1
2a
2b
6.58ꢁ0.1 7.63
7.56
7.72
7.78
ꢀ2.76ꢁ0.4 ꢀ1.40
ꢀ1.45
ꢀ1.97
ꢀ2.19
–
–
7.82
7.83
–
–
ꢀ1.89
ꢀ2.09
–
–
3a 5.99ꢁ0.1
–
ꢀ2.65ꢁ0.4
–
[b]
[b]
6
6.35ꢁ0.1 7.04
ꢀ4.59ꢁ0.4 ꢀ3.08
6’
6’’
7
8
9
–
–
6.08
7.24
6.04
7.20
7.91
7.87
8.04
–
–
ꢀ2.20
ꢀ1.78
ꢀ2.27
ꢀ1.82
ꢀ2.25
ꢀ1.88
ꢀ2.30
7.54ꢁ0.1 7.98
7.97
7.76ꢁ0.1 8.12
ꢀ4.07ꢁ0.4 ꢀ2.19
–
–
ꢀ1.80
ꢀ3.75ꢁ0.4 ꢀ2.22
[a] We have used the definition of EA as the energy change for the pro-
ꢀ
Cꢀ
cess M+e !M ; hence the negative values indicate exothermicity for
the reduction of a molecule. [b] Not calculated.[20]
play of s-acceptor and p-donor effects (fluorine is both a
better s-acceptor and p-donor than chlorine in organic sys-
tems).[53] Moreover, the calculated gas-phase data also sug-
gests similar EAs for 7 and 9, these values being more exo-
thermic than that calculated for the hexafluoro derivative, 8.
The most easily reduced material, according to the EA
Experimental Section
Synthesis ofliterature hexaazatrinaphthylene derivatives :Compound 3a
was obtained according to the literature procedure;[6] details of the spe-
cific synthetic and purification methods used for 1 and 7 are given in the
Supporting Information.
measurements, is the tris(thieno)hexaazatriphenylene com-
A
2,5-Bis-(3,5-bis-trifluoromethylphenyl)-3,4-dinitro-thiophene (S2):A mix-
ture of 2,5-dibromo-3,4-dinitro-thiophene[54] (S1; 6.00 g, 18.08 mmol), 3,5-
bis-trifluoromethyl-phenyl-tributyl-stannane (16.52 g, 36.15 mmol),[55]
tris(dibenzylideneacetone) dipalladium (0.66 g, 0.72 mmol) and tripheny-
larsine (0.88 g, 2.88 mmol) in toluene (100 mL) was heated to 808C for
12 h under nitrogen. The mixture was cooled to room temperature and
then a solution of potassium fluoride (30 mL, 2.0m) was added. The mix-
ture was stirred for 1 h. The organic layer was collected and the aqueous
layer was extracted with diethyl ether. The combined organic layer was
dried over magnesium sulfate and filtered. The solvent was removed
under reduced pressure. The residue was crystallised from methanol to
pound, 6, the EA of which is approximately 0.5 eV more
exothermic than that of 7 (around 1.8 eV more exothermic
than that of the parent 1). Indeed, it is this highly exother-
mic EA that prompted us to carry out n-type doping of this
material with cobaltocene.[10] Compound 6 is also slightly
more easily ionised than 1; this suggests a considerably re-
duced HOMO–LUMO gap relative to that of 1 and its halo-
genated derivatives, which is consistent with calculated orbi-
tal energies (see Table S7 in Supporting Information) and
with solution UV/Vis spectra, which show the onset of ab-
sorption for 6 to be approximately 1 eV lower than that of 1
or 7 (see Figure S5 in Supporting Information). To under-
stand the relative importance of the structural features of 6
leading to the differences in IPs and EAs relative to 1, we
used DFT to calculate gas-phase IPs and EAs for the model
compounds 6’ and 6’’ (Table 2), in which the trifluoromethyl
groups and the bis(trifluoromethyl)phenyl groups, respec-
tively, are removed. These calculations show that the core
give
[D]chloroform): d=8.08 (br, 2H), 8.01 ppm (br, 4H); 13C NMR
(125 MHz, [D6]acetone): d=140.0, 138.7, 132.9 (q, J(C,F)=34.0 Hz, Co),
131.7, 131.4 (q, (C,F)=3.13 Hz, Cm), 127.2, 125.5 (septet, (C,F)=
3.6 Hz, Cp), 124.0 ppm (q, J(C,F)=271.7 Hz, CF3); HRMS (EI): m/z
a
yellow solid (7.25 g, 67% yield). 1H NMR (300 MHz,
AHCTREUNG
J
G
J
ACHTREUNG
AHCTREUNG
calcd for C20H6F12N2O4S 597.9857; found 597.9843; elemental analysis
calcd (%) for C20H6F12N2O4S:C 40.15, H 1.01, N 4.68; found:C 39.95, H
0.88, N 4.72.
2,5-Bis-(3,5-bis-trifluoromethylphenyl)thiophene-3,4-diamine (S3):
A
mixture of S2 (2.22 g, 5.00 mmol) and tin(II) chloride dihydrate (11.28 g,
50.00 mmol) in ethanol (50 mL) was heated to reflux for 30 min under ni-
trogen atmosphere. After the mixture cooled to room temperature, it was
poured onto ice. The solution was made slightly basic by the addition of
5% aqueous sodium hydroxide solution. The solution was extracted with
ethyl acetate. The combined organic layer was dried over magnesium sul-
fate. The solvent was removed under reduced pressure to give a solid.
The solid was purified by flash chromatography on silica gel eluting with
hexane/ethyl acetate (10:1) to give 1.60 g (59%). 1H NMR (300 MHz,
[D6]benzene): d=7.72 (brs, 4H), 7.59 (brs, 2H), 2.62 ppm (brs, 4H);
tris(thieno)hexaazatriphenylene, 6’’, is both more easily oxi-
G
dised and more easily reduced than 1, in both cases by
about 0.4 eV. This is consistent with DFT calculations of the
IP and EA for thiophene and benzene, which show the
former species is both more easily oxidised and more easily
reduced (see Supporting Information). The more exothermic
EA of 6’’ versus 1 may also be related to the greater domi-
nance of A-type resonance structures noted in the optimised
structure of 6’’ (vide supra), this being enforced by the
bond-length distribution of thiophene. Comparison of the
data for 6, 6’ and 6’’ shows that the bis(trifluoromethyl)-
phenyl groups affect both IP and EA; phenylation and addi-
tion of the CF3 groups have opposing effects on the IP, but
13C NMR (75 MHz, [D6]benzene): d=136.6, 135.1, 132.5 (q, J
33.0 Hz, Cm), 127.0 (m, Co), 123.6 (q, J(C,F)=270.9 Hz, CF3), 120.1
(septet, (C,F)=4.5 Hz, Cp), 114.5 ppm; HRMS(EI): m/z calcd for
(C,F)=
AHCTREUNG
J
ACHTREUNG
C20H10F12N2S 538.0373; found 538.0379; elemental analysis calcd (%) for
C20H10F12N2S:C 44.62, H 1.87, N 5.20; found:C 44.72, H 1.80, N 5.21.
Tris{2,5-bis(3,5-bis-trifluoromethyl-phenyl)thieno}[3,4-b,h,n]-1,4,5,8,9,12-
hexaazatriphenylene (6):A mixture of hexaketocyclohexane octahydrate
(0.296 g, 0.95 mmol) and S3 (1.58 g, 2.85 mmol) was added to degassed
A
3544
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 3537 – 3547

Hexaazatrinaphthylene Derivatives
FULL PAPER
acetic acid (50 mL). The mixture was heated to 1008C for 12 h under ni-
trogen atmosphere. After the mixture was cooled to room temperature,
the solid was collected by filtration and was washed with acetic acid to
give a black solid. The product was purified by flash chromatography on
silica gel eluting with hexane/ethyl acetate (10:2) to give 1.00 g (63%) of
purified product. 1H NMR (300 MHz, [D2]dichloromethane) d=8.81 (br,
12H), 7.96 ppm (br, 6H); HRMS (MALDI-TOF, CHCA matrix): m/z
calcd for C66H18F36N6S3, 1674.0180; found:1674.0167; elemental analysis
calcd (%) for C66H18F36N6S3:C 47.32, H 1.08, N 5.02; found:C 47.35, H
0.99, N 5.14.
difference between Evac and the leading edge of the HOMO and LUMO
feature, respectively.
Crystal structure of 7:Yellow single crystals of 7 were obtained during
purification by vacuum sublimation. A 0.45”0.30”0.20 mm crystal was
selected and data collected at 96 K using a Bruker SMART diffractome-
ter with CCD detector with MoKa radiation (0.71073 Š). The crystal
was found to belong to the monoclinic space group P21/c with
a=10.9644(11), b=26.634(3), c=8.0161(8) Š, b=108.908(2)8, Z=4,
F
000 =1176, mcalcd =0.807 mmꢀ1 1calcd =1.773 gcmꢀ3
, . Least-squares re-
finement of F2 was carried out using all 6478 independent reflections
measured (4050 with [I>2s(I)]); refinement converged with R=0.828
and wR=0.1126 (all data). CCDC-620134 contains the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
2,3,8,9,14,15-Hexafluoro-5,6,11,12,17,18-hexaazatrinaphthylene (8):
A
mixture of hexaketocyclohexane octahydrate (2.78 g, 8.90 mmol) and 4,5-
difluoro-benzene-1,2-diamine[13, 14] (3.85 g, 26.71 mmol) was added to de-
oxygenated acetic acid (250 mL). The mixture was refluxed for 16 h.
After the mixture was cooled to room temperature, the solid was collect-
ed by filtration and was washed with acetic acid to give a yellow solid
(2.88 g, 58% crude yield). Gradient sublimation at high vacuum
(ca. 2508C) gave fine needle-like microcrystals, which, however, did not
give satisfactory elemental analysis. 1H NMR (300 MHz, [D]chloroform/
DFT calculations:Molecular geometries were optimised in the neutral,
radical anion and radical cation states, by means of density functional
theory (DFT) using the B3LYP functionals, in which Beckeꢁs three-pa-
rameter hybrid exchange functional[56,57] was combined with the Lee-
Yang-Parr correlation functional,[58] and a 6-31G** split valence plus
double polarisation basis set. Ionisation energies, electron affinities and
reorganisation energies were calculated from direct calculation of the en-
ergies of the relevant points on the potential-energy surfaces. Specifically,
vertical IPs were calculated as the energy difference between the energy
of the cation at neutral geometry and the neutral molecule at neutral ge-
ometry and adiabatic IPs as the difference between the cation at the re-
laxed cation geometry and the neutral molecule at neutral geometry. The
difference between vertical and adiabatic IPs corresponds to the relaxa-
tion energy, l2, associated with the cation surface. The parameter l1 is de-
fined as the difference between the neutral species at the cation geome-
try. The sum of l1 and l2, li is the Marcus intramolecular reorganisation
energy for the self-exchange reaction of a molecule and its cation. Simi-
larly, vertical and adiabatic EAs were calculated as the difference be-
tween the energy of the anion at neutral geometry and the neutral spe-
cies at neutral geometry and as the energy difference between the re-
laxed anion and neutral molecule, respectively. The relevant l2 contribu-
ting to li for self-exchange between a molecule and its cation is then the
difference between vertical and adiabatic EAs, while l1 is the energy dif-
ference between the anion at neutral geometry and at anion geometry.
All DFT calculations were carried out with the Gaussian suite of pro-
grams (both Gaussian 98, revision A.11[59] and Gaussian 03[60]).
[D]trifluoroacetic acid): d=8.42 ppm (t, J
(376.3 MHz, [D]chloroform/[D]trifluoroacetic acid): d=ꢀ121.26 ppm (t,
(F,H)=8.5 Hz); HRMS(EI): m/z calcd for C24H6F6N6 492.0558; found
A
J
ACHTREUNG
492.0585; elemental analysis calcd (%) for C24H6F6N6:C 58.55, H 1.23, N
17.07, F 22.34; found:C 56.98, H 1.39, N 16.81, F 22.36.
1,2,3,4,7,8,9,10,13,14,15,16-Dodecafluoro-5,6,11,12,17,18-hexaazatrinaph-
thylene (9):A mixture of hexaketocyclohexane octahydrate (2.50 g,
9.25 mmol) and 3,4,5,6-tetrafluoro-benzene-1,2-diamine[15,16] (5.00 g,
27.76 mmol) was added to deoxygenated acetic acid (150 mL). The mix-
ture was refluxed for 12 h under nitrogen. After the mixture was cooled
to room temperature, the resulting solid was collected by filtration and
was washed with acetic acid to give a yellow solid (4.05 g, 73% crude
yield). A portion was purified by gradient sublimation at high vacuum
(ca. 2508C) to give yellow crystals, which were not soluble in common or-
ganic solvents including CF3CO2H. However, the crystals were soluble in
concentrated sulfuric acid. 19F NMR (376.3 MHz, [D2]sulfuric acid): d=
ꢀ123.53, ꢀ138.35 ppm; HRMS(EI): m/z calcd for C24F12N6:599.9993;
found 599.9981; elemental analysis calcd (%) for C24F12N6:C 48.02, H
14.00; found:C 48.03, H 14.10.
PES and IPES measurements:The organic thin films were grown and
characterised with PES and IPES under UHV without any exposure to
air. The substrates were Si wafers covered with a 50 Š Ti layer (for adhe-
sion) and a 1200 Š Au layer that had been previously degreased by boil-
ing in acetone and methanol. Compounds 1, 6, 7 and 9 were thermally
evaporated from pyrolytic boron nitride crucibles after an extensive out-
gassing procedure. Films with thickness of 80 Š were deposited on the
Au surface at a rate of 1–2 Šsꢀ1. Film deposition was monitored using a
quartz crystal microbalance, assuming the same density of 1.5 gcmꢀ3 for
all four derivatives. Compound 3a could not withstand thermal evapora-
tion and was spin-coated (6000 rpm for 60 s) from a 0.08 wt% solution of
dichloromethane in a nitrogen-purged glove box (humidity <5%). The
film was then dried from solvent at 508C for one hour under nitrogen.
The thickness of the film prepared by spin coating was not measured but
was assumed to be less than 100 Š. Ambient exposure of the 3a film was
unavoidable when loading the sample into the chamber, but was limited
to less than a minute.
Acknowledgements
This work was supported in part by the National Science Foundation
through the Science and Technology Center Program (DMR-0120967),
through the Princeton MRSEC (DMR-0213706) and through grants
DRM-0408589, CHE-0211419 and DRM-0420863. We also thank Lintec
Corporation and the Office of Naval Research (N00014-04-1-0120) for
support.
[1] H. Bock, A. Babeau, I. Seguy, P. Jolinat, P. Destruel, ChemPhys-
Chem 2002, 3, 532.
The PES was performed with a He discharge lamp by using the He I
(21.22 eV) and He II (40.8 eV) photon lines with a total experimental
resolution of 150 meV. The IPES was carried out in the isochromat mode
by using a fixed-photon-energy detector centred at 9.2 eV and an electron
gun, resulting in a combined resolution of 500 meV. The IPES electron
beam current density was limited to around 1”10ꢀ6 Acmꢀ2 to minimise
degradation of the organic thin film. In all cases, we performed the PES
experiment first, then the IPES measurement, and repeated the PES to
assess any electron-beam-induced degradation.
[2] B. R. Kaafarani, T. Kondo, J. Yu, Q. Zhang, D. Dattilo, C. Risko,
S. C. Jones, S. Barlow, B. Domercq, F. Amy, A. Khan, J. L. BrØdas,
B. Kippelen, S. R. Marder, J. Am. Chem. Soc. 2005, 127, 16358.
[3] G. Kestemont, V. de Halleux, M. Lehmann, D. A. Ivanov, M.
Watson, Y. H. Geerts, Chem. Commun. 2001, 2074.
[4] V. Lemaur, D. A. de Silva Filho, V. Coropceanu, M. Lehmann, Y.
Geerts, J. Piris, M. G. Debije, A. M. van de Craats, K. Senthilkumar,
L. D. Siebbeles, J. M. Warman, J. L. BrØdas, J. Cornil, J. Am. Chem.
Soc. 2004, 126, 3271.
The PES and IPES energy scales were aligned by using the position of
the Fermi step measured on a gold substrate. The position of the vacuum
level (Evac) was measured on each sample using the onset of photoemis-
sion. The experimental values for IP and EA correspond to the energy
[5] X. Crispin, J. Cornil, R. Friedlein, K. K. Okudaira, V. Lemaur, A.
Crispin, G. Kestemont, M. Lehmann, M. Fahlman, R. Lazzaroni, Y.
Geerts, G. Wendin, N. Ueno, J. L. BrØdas, W. R. Salaneck, J. Am.
Chem. Soc. 2004, 126, 11889.
Chem. Eur. J. 2007, 13, 3537 – 3547
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3545

S. R. Marder et al.
[6] M. Lehmann, G. Kestemont, R. G. Aspe, C. Buess-Herman, M. H. J.
Koch, M. G. Debije, J. Piris, M. P. de Haas, J. M. Warman, M. D.
Watson, V. Lemaur, J. Cornil, Y. H. Geerts, R. Gearba, D. A.
Ivanov, Chem. Eur. J. 2005, 11, 3349.
[34] A. R. Murphy, J. S. Liu, C. Luscombe, D. Kavulak, J. M. J. FrØchet,
R. J. Kline, M. D. McGehee, Chem. Mater. 2005, 17, 4892.
[35] H. Xu, Y. Wang, G. Yu, W. Xu, Y. B. Song, D. Q. Zhang, Y. Q. Liu,
D. B. Zhu, Chem. Phys. Lett. 2005, 414, 369.
[36] J. E. Anthony, J. S. Brooks, D. L. Eaton, S. R. Parkin, J. Am. Chem.
Soc. 2001, 123, 9482.
[7] C. W. Ong, S.-C. Liao, T. H. Chang, H.-F. Hsu, Tetrahedron Lett.
2003, 44, 1477.
[37] X.-C. Li, H. Sirringhaus, F. Garnier, A. B. Holmes, S. C. Moratti, N.
Feeder, W. Clegg, S. J. Teat, R. H. Friend, J. Am. Chem. Soc. 1998,
120, 2206.
[38] J. E. Anthony, D. L. Eaton, S. R. Parkin, Org. Lett. 2002, 4, 15.
[39] K. Kobayashi, R. Shimaoka, M. Kawahata, M. Yamanaka, K. Yama-
guchi, Org. Lett. 2006, 8, 2385.
[8] C. W. Ong, S.-C. Liao, T. H. Chang, H.-F. Hsu, J. Org. Chem. 2004,
69, 3181.
[9] T. Ishi-i, K. Yaguma, R. Kuwahara, Y. Taguri, S. Mataka, Org. Lett.
2006, 8, 585.
[10] C. K. Chan, F. Amy, Q. Zhang, S. Barlow, S. R. Marder, A. Kahn,
Chem. Phys. Lett. 2006, 431, 67.
[40] A. M. van de Craats, J. M. Warman, H. Hasebe, R. Naito, K. Ohta,
J. Phys. Chem. B 1997, 101, 9224.
[41] S. Kumar, D. S. S. Rao, S. K. Prasad, J. Mater. Chem. 1999, 9, 2751.
[42] R. I. Gearba, M. Lehmann, J. Levin, D. A. Ivanov, M. H. J. Koch, J.
Barberµ, M. G. Debije, J. Piris, Y. H. Geerts, Adv. Mater. 2003, 15,
1614.
[11] S. Skujins, G. A. Webb, Tetrahedron 1969, 25, 3935.
[12] M. Du, X.-H. Bu, K. Biradha, Acta Crystallogr. Sect. C 2001, 57,
199.
[13] S. K. Kotovskaya, N. M. Perova, Z. M. Baskakova, S. A. Romanova,
V. N. Charushin, O. N. Chupakhin, Zh. Org. Khim. 2001, 37, 598.
[14] R. B. Baudy, L. P. Greenblatt, I. L. Jirkovsky, M. Conklin, R. J.
Russo, D. R. Bramlett, T. A. Emrey, J. T. Simmonds, D. M. Kowal,
R. P. Stein, R. P. Tasses, J. Med. Chem. 1993, 36, 331.
[15] A. Heaton, M. Hill, F. Drakesmith, J. Fluorine Chem. 1997, 81, 133.
[16] J. F. W. Keana, S. M. Kher, S. X. Cai, C. M. Dinsmore, A. G. Glenn,
J. Guastella, J. Huang, V. Ilyin, Y. Lu, P. L. Mouser, R. M. Wood-
ward, E. Weber, J. Med. Chem. 1995, 38, 4367.
[17] We were unable to obtain 8 analytically pure and so did not obtain
PES and IPES spectra.
[18] M. Alfonso, H. Stoeckli-Evans, Acta Crystallogr. Sect. E 2001, 57,
0242.
[43] This type of estimate assumes that the analyte and reference redox
system show comparable solvation energies and comparable solid-
state polarisation effects and ideally requires more or less reversible
electrochemistry to ensure that the observed potential represents
the thermodynamic potential. Moreover, differences in the magni-
tudes of solvation energies and solid-state polarisation effects mean
that IPs are best estimated from M+/M redox couples and a known
IP, while EAs are best estimated from M/Mꢀ and a known EA. For
example, we recently found that, in the case of 1,1-diaryl-2,3,4,5-tet-
raphenylsiloles, use of [silole]/
[silole]ꢀ potentials, the TPD+/TPD
A
potential and the IP of TPD gave EAs almost 1 eV more exothermic
than the values obtained by direct measurement. See reference [26].
[44] L. Segev, A. Salomon, A. Natan, D. Cahen, L. Kronik, F. Amy,
C. K. Chan, A. Kahn, Phys. Rev. B 2006, 74, 165323/1.
[45] F. Amy, C. K. Chan, W. Zhao, J. Hyung, A. Kahn, M. Ono, N. Ueno,
D. Cahen, L. Kronik, G. Nesher, A. Salomon, L. Segev, O. Seitz, H.
Shpaisman, A. Schçll, E. Umbach, J. Phys. Chem. B 2006, 110,
21826.
[46] N. Sato, H. Inokuchi, E. A. Silinish, Chem. Phys. 1987, 115, 269.
[47] J. D. Anderson, E. M. McDonald, P. A. Lee, M. L. Anderson, E. L.
Ritchie, H. K. Hall, T. Hopkins, E. A. Nash, J. Wang, A. Padias, S.
Thayumanavan, S. Barlow, S. R. Marder, G. Jabbour, S. Shaheen, B.
Kippelen, N. Peyghambarian, R. M. Wightman, N. R. Armstrong, J.
Am. Chem. Soc. 1998, 120, 9646.
[19] Bond
lengths
from
the
structure
of
[Ag3(1)]-
ACHTREUNG
M. Nishimura, T. Ito, K. Tanaka, M. Shionoya, Chem. Commun.
2000, 1953) are not included, since these were rather imprecisely de-
termined.
[20] The structures of the radical ions of 6, in which there are a large
number of degrees of freedom, could not be successfully minimised.
Accordingly, we were also unable to calculate reorganisation ener-
gies and adiabatic values of IP and EA for this compound.
[21] R. A. Marcus, J. Chem. Phys. 1956, 24, 966.
[22] For example, see:R. D. Hreha, C. P. George, A. Haldi, B. Domercq,
M. Malagoli, S. Barlow, J.-L. BrØdas, B. Kippelen, S. R. Marder,
Adv. Funct. Mater. 2003, 13, 967. While the barrier is strictly DH°,
DS° can be neglected due to similar vibrational degrees of freedom
associated with the two adiabatic surfaces, meaning l/4 is also a
good estimate of the free-energy barrier, DG°.
[23] The reorganisation energy can also be obtained from a full vibra-
tional analysis through summation of the contributions of the vibra-
tional modes involved (R. A. Marcus, N. Sutin, Biochim. Biophys.
Acta 1985, 811, 265).
[24] M. Malagoli, J.-L. BrØdas, Chem. Phys. Lett. 2000, 327, 13.
[25] B. C. Lin, C. P. Cheng, Z.-Q. You, C.-P. Hsu, J. Am. Chem. Soc.
2005, 127, 66.
[26] X. Zhan, C. Risko, F. Amy, C. Chan, W. Zhao, S. Barlow, A. Kahn,
J.-L. BrØdas, S. R. Marder, J. Am. Chem. Soc. 2005, 127, 9021.
[27] F. Garnier, G. Horowitz, D. Fichou, A. Yassar, Synth. Met. 1996, 81,
163.
[28] H. E. Katz, Z. Bao, J. Phys. Chem. B 2000, 104, 671.
[29] R. C. Haddon, X. Chi, M. E. Itkis, J. E. Anthony, D. L. Eaton, T.
Siegrist, C. C. Mattheus, T. T. M. Palstra, J. Phys. Chem. B 2002, 106,
8288.
[48] W. Y. Gao, A. Kahn, J. Appl. Phys. 2003, 94, 359.
[49] A. Kahn, N. Koch, W. Gao, J. Polym. Sci. Part B 2003, 41, 2529–
2548.
[50] D. Cahen, A. Kahn, Adv. Mater. 2003, 15, 271.
[51] These values are taken from J. E. Huheey, E. A. Keiter, R. L.
Keiter, Inorganic Chemistry: Principles of Structure and Reactivity,
4th ed., Harper Collins, New York, 1993, p. 187, in which values ac-
cording to a variety of other scales are also given.
[52] R. Dudde, B. Reihl, A. Otto, J. Chem. Phys. 1990, 92, 3930.
[53] The competing inductive and resonance effects of F and Cl can of
course be gauged by Hammett coefficients. See for example J.
March, Advanced Organic Chemistry, 3rd ed., Wiley, New York,
1985.
[54] D. D. Kenning, K. A. Mitchell, T. R. Calhoun, M. R. Funfar, D. J.
Sattler, S. C. Rasmussen, J. Org. Chem. 2002, 67, 9073.
[55] R. G. Kultyshev, G. K. S. Prakash, G. A. Olah, J. W. Faller, J. Parr,
Organometallics 2004, 23, 3184.
[56] A. D. Becke, Phys. Rev. A 1988, 38, 3098.
[30] Y. Olivier, V. Lemaur, J. L. BrØdas, J. Cornil, J. Phys. Chem. A 2006,
110, 6356.
[31] A. M. van de Craats, N. Stutzmann, O. Bunk, M. M. Nielsen, M.
Watson, K. Müllen, H. D. Chanzy, H. Sirringhaus, R. H. Friend,
Adv. Mater. 2003, 15, 495.
[32] Z. An, J. Yu, S. C. Jones, S. Barlow, S. Yoo, B. Domercq, P. Prins,
L. D. A. Siebbeles, B. Kippelen, S. R. Marder, Adv. Mater. 2005, 17,
2580.
[57] A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
[58] C. T. Lee, W. T. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.
[59] Gaussian98 (RevisionA.11), M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski,
J. A. Montgomery, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M.
Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J.
Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli,
C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q.
Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari,
[33] K. O. Sylvester-Hvid, J. Phys. Chem. B 2006, 110, 2618.
3546
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 3537 – 3547

Hexaazatrinaphthylene Derivatives
FULL PAPER
J. B. Foresman, J. Cioslowski, J. V. Ortiz, B. B. Stefanov, G. Liu, A.
Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin,
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
C. Gonzalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W.
Chen, M. W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle,
J. A. Pople, Gaussian, Inc., Pittsburgh, PA, 1998.
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, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
[60] Gaussian 03, Revision B.05, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomer-
y, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyen-
gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
Received:September 8, 2006
Published online:January 17, 2007
Chem. Eur. J. 2007, 13, 3537 – 3547
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3547