Synthesis and structural characterization of crystalline nonacenes

Article abstract of DOI:10.1002/anie.201102671

Nine in a line: A multifaceted approach, including a silylethylene functionalization strategy, to the synthesis of stable nonacene derivatives (see picture, F green, Si blue) allows an unprecedented level of characterization, including electrochemical, photophysical, and crystallographic studies. The nonacenes show a strong S₀-S₁ transition and no fluorescence in the visible region. Copyright

Full text of DOI:10.1002/anie.201102671

DOI: 10.1002/anie.201102671
Oligoacenes
Synthesis and Structural Characterization of Crystalline Nonacenes**
Balaji Purushothaman, Matthew Bruzek, Sean R. Parkin, Anne-Frances Miller, and
John E. Anthony*
Acenes and functionalized acenes have played a pivotal role
in the development of organic electronic materials.[1] From
the first report of high-mobility organic transistors based on
vapor-deposited pentacene,[2] to high-efficiency organic light-
emitting diodes based on anthracene,[3] and the remarkable
transport properties of rubrene,[4] these linearly fused hydro-
carbons have made significant contributions to the under-
standing of electronic processes in organic semiconductors.
Despite the utility of these compounds, the selection of
materials suitable for exploration essentially stops at penta-
cene. Although numerous studies have predicted enticing
electronic properties for larger acenes,[5] the combination of
poor solubility and low stability have hindered detailed
studies of these materials.
The potential technological importance of higher acenes
has created a resurgence of interest in these systems.[6]
Elegant low-temperature matrix-isolation techniques have
recently allowed the spectroscopic characterization of hexa-
cene,[7] heptacene,[8] octacene, and nonacene,[9] as well as a few
more detailed studies of the electronic structure of hepta-
cene.[10] Moreover, new functionalization strategies are being
developed to allow the isolation and full structural and
electronic characterization of larger acenes. The selection of
substituents for such strategies is critical to the success of the
approach, for example, substitution of hexacene with various
aryl groups resulted in no stable materials,[7] while silylethyne
functionalization has provided numerous soluble hexacenes,
which have now been extensively characterized.[11] The
silylethyne functionalization approach to engineer solid-
state order[12] has also produced a number of crystalline
heptacene derivatives such as 1a.[13] Materials prepared using
this strategy are often sufficiently crystalline for X-ray
diffraction studies, which are critical for the unambiguous
structural confirmation of these reactive species. The addition
of phenyl substituents to a silylethyne-functionalized hepta-
cene (1b) disrupted the p-stacking motif beneficial to charge
transport, but in exchange produced a highly stable deriva-
tive, which was also fully structurally characterized.[14] The
enhanced stability likely arises from the perpendicular
orientation of the phenyl groups relative to the acene
backbone; this orientation hinders the rate of face-to-face
dimerization. Later reports demonstrated that the use of
electron-deficient arene substituents (1c) produced a further
incremental improvement to acene stability.[15]
A more recent functionalization strategy involves the use
of extensive thioether substitution around the acene chromo-
phore.[16] This innovative approach led to the report of a
reasonably stable heptacene derivative,[17] and more recently,
a so-called “persistent” nonacene derivative (2).[18] This
purported functionalized nonacene is highly unusual because
of an almost non-existent absorption corresponding to the S0–
S1 transition, and fluorescence at remarkably short wave-
length. These findings raise interesting questions about the
electronic structure and optoelectronic properties of non-
acene, and heighten the need for a derivative that is stable
enough for full structural and spectroscopic characterization.
[*] Dr. B. Purushothaman,[+] M. Bruzek, Dr. S. R. Parkin, Prof. A.-F. Miller,
Prof. J. E. Anthony
Department of Chemistry, University of Kentucky
Lexington, KY 40506-0055 (USA)
E-mail: anthony@uky.edu
[+] Current address:, School of Chemistry, Bio21 Institute, The
University of Melbourne, Parkville, VIC 3010 (Australia)
[**] We thank the National Science Foundation (CHE-0749473 and CHE-
0319176) for support of this research.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201102671.
Angew. Chem. Int. Ed. 2011, 50, 7013 –7017
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7013

Communications
With these recent reports in mind, we decided to explore
the olive-colored solutions of 3a–c (Figure 1a) are markedly
different from the spectrum reported for 2 (Figure 1c).[18]
Nonacenes 3 show a prominent S0–S1 transition at 1014 nm,
and only weak absorptions throughout the visible region
the preparation of a structurally characterizable nonacene
derivative, in order to better understand the nature of the
optoelectronic properties of this chromophore. From our
work on hexacene derivatives,[11] we already knew that the
substituents on the alkynes needed to be sufficiently bulky to
prevent Diels–Alder reactions between the alkynes and the
reactive acene chromophore, and that the solid-state order
must be carefully tuned to prevent close contacts between the
highly reactive central aromatic rings of the acene backbone.
It was also clear from the outset that the incorporation of
phenyl substituents, as used by Wudl and co-workers for their
highly stable heptacene,[14] would be critical to the prepara-
tion of a stable nonacene. However, our strategy places these
substituents on the central ring of the acene in order to leave a
reasonable expanse of the acene periphery available for
p stacking. Such interactions are critical for the rapid growth
of acene crystals, which we reasoned would be necessary for
the isolation of a stable, crystalline compound. Finally, our
work on partially fluorinated pentacenes[19] and anthradithio-
phenes[20] taught us that fluorine substitution both increases
acene stability and enhances p-stacking interactions—per-
haps mitigating the impact of the orthogonal phenyl sub-
stituents. With these criteria in mind, we envisioned non-
acenes 3 as reasonable target molecules.
The synthesis of derivatives
3 was straightforward
(Scheme 1), and involved the alkynylation of the correspond-
Figure 1. a) Absorption spectrum and image of a sample of 3c in
toluene. b) Absorption (blue trace) and emission (red trace) spectra of
the strongly fluorescent decomposition product of 3c, along with an
image of the photodecomposed solution. Absorption spectrum of pure
3c (black trace) shown for reference. c) Reported UV/Vis/NIR (blue
trace) and fluorescence (purple trace) spectra for nonacene 2. Data for
2[23] reprinted from Ref. [18] with permission. Copyright 2010 American
Chemical Society.
Scheme 1. Synthesis of nonacenes 3.
down to 500 nm, contrasting with the intense absorptions
from 500–700 nm and barely observable S0–S1 transition
reported for 2. The absorption onset of 3a (1033 nm) gives
a HOMO–LUMO gap of 1.2 eV (HOMO = highest occupied
molecular orbital, LUMO = lowest unoccupied molecular
orbital), a value corroborated by electrochemical analysis of
3a by differential pulse voltammetry (Figure S6 in the
Supporting Information), showing oxidation and reduction
onsets versus ferrocene/ferrocenium of + 0.78 V and À0.41 V
(Egap = 1.19 eV). The relatively high oxidation potential is a
result of the extensive fluorine substitution, and likely
enhances the oxidative stability of nonacenes 3. While there
is little match between the absorption spectrum of 3 and the
ing quinone 4, which in turn was prepared in four steps from
durene (details in the Supporting Information). The addition
of a variety of silyl acetylides to 4 produced an isomeric
mixture of the corresponding tetraols 5 in good yield,
although extensive purification of the crude mixtures was
not carried out. Because the functionalized larger acenes are
significantly more stable in the crystalline state,[13] we
optimized the reaction conditions to crystallize the nonacenes
directly from the deoxygenation reaction mixture (see the
Experimental Section). Upon re-dissolving these crystals in
toluene, we found that the absorption spectra obtained from
7014
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 7013 –7017

spectrum reported for 2, there is reasonable but red-shifted
structural correspondence between the spectra for nonacenes
3 and the spectrum reported for matrix-isolated nonacene,[9]
which also exhibits a strong S0–S1 transition and similar
spacing of weak transitions through visible wavelengths.
In the report on 2,[18] and in subsequent media releases,[21]
the intense red fluorescence of the isolated compound was
noted. It was claimed that this red fluorescence was a
diagnostic tool for the presence of the nonacene chromo-
phore. We did not observe red emission from nonacenes 3,
nor did we observe (or expect to observe) any fluorescence in
the visible region. Kashaꢀs rule[22] states that alternant hydro-
carbons emit from the first singlet excited state (S1), thus, with
a S0–S1 transition beyond 1000 nm, emission of visible light is
most unexpected. Intrigued by this prior report of red
fluorescence, we investigated the absorption and emission
properties of the decomposition products of 3c, which were
formed by exposing solutions of 3c to air under bright
laboratory lighting. Under these conditions, the nonacene
decomposed completely within 6 h to give a dark blue-black
solution that did indeed exhibit bright red fluorescence
(Figure 1b). The decomposition product was isolated and
purified by column chroma-
from the reaction mixture, and showed essentially no
decomposition when stored in the dark at 108C over more
than 2 days. All three nonacene derivatives reported here
formed crystals suitable for X-ray analysis (Figure S2), and all
structures were amenable to at least some degree of refine-
ment (Figure 2).[24] Triisopropylsilyl derivative 3a gave a
reasonable data set that showed that the linear acene
chromophore is slightly twisted by the backbone substituents
and warped by crystal packing effects. This derivative adopted
a 2D p-stacked motif similar to that observed for some
pentacene derivatives,[12] although in this case there is
minimal p overlap—the spaces above and below the reactive
center of the nonacene chromophore are occupied by
disordered solvent molecules. Crystals of 3b were much
thicker than those of 3a or 3c, and gave significantly better
data. This nonacene was almost exactly planar, and exhibited
only a small amount of twist due to interactions between the
aryl and alkynyl substituents. The nonacene backbones pack
as perpendicular 1D “slipped” stacks, with only the terminal
rings of the acenes overlapping. As with 3a, the center of the
acene is surrounded by incorporated solvent molecules.
Crystals of 3c were extremely thin, and the data were
tography, and crystallo-
graphic analysis of this
decomposition
product
(Figure S4) indicates that
the nonacene forms an
endoperoxide on ring 4 of
the structure to produce a
pentacene
chromophore.
The absorption spectrum
for this decomposition
product (Figure 1b) exhib-
its strong transitions in the
visible region (500–700 nm)
and has vibronic structure
typical of a pentacene. We
also note a striking similar-
ity between both the
absorption and emission
spectra of this decomposi-
tion product and those
reported for nonacene
(Figures 1b and c).
2
A key benefit of the
silylethyne functionaliza-
tion approach used here is
the ability to engineer easily
crystallized materials, and
this factor was crucial to
the preparation of the first
stable hexacene and hepta-
cene derivatives.[13] We
exploited this same ration-
ale to obtain nonacene
derivatives as surprisingly
stable solids that were
formed as crystals directly
Figure 2. Thermal ellipsoid plots for 3a (top), 3b (center), and 3c (bottom), and packing diagrams (right).
F green, Si blue. Alkyl groups and fluorine substituents omitted from packing diagrams for clarity. Thermal
ellipsoids set at 50% probability.
Angew. Chem. Int. Ed. 2011, 50, 7013 –7017
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7015

Communications
correspondingly weak. The nonacene chromophores in this
spectroscopic characterization. Crystallographic analysis
revealed conclusively the presence of the conjugated non-
acene chromophores. The spectroscopic properties of these
compounds differ significantly from those previously reported
for purported nonacene (2), and show a strong S0–S1
transition near 1000 nm and no fluorescence in the visible
region. Electrochemical analysis supports the observed opti-
cal gap, and the absorption maxima and peak structures
correlate well with those of the previously reported series of
silylethyne-substituted acenes (Figure 4).[26] While the photo-
physical differences between nonacenes 3 and the reported
crystal are highly distorted, adopting a 1D “slipped-stack”
motif, in this case with significantly more overlap between the
nonacene chromophores than observed for 3a or 3b. Crystals
of 3c incorporated chlorobenzene, and, as with the other two
derivatives, the central part of the chromophore was sur-
rounded by these solvent molecules. The solid-state stability
of all of these nonacene crystals is likely enhanced by the
presence of low-reactivity solvent molecules that surround
the most reactive acene rings in the crystal and thus hinder
acene dimerization.
Finally, we note that it proved impossible to collect
1H NMR spectra for any of these nonacene derivatives.
Although the compounds are sufficiently soluble for the
experiment, initial scans in the NMR spectrum showed only
noise, and over time, broad peaks corresponding to the
independently characterized decomposition product began to
appear. After leaving a sample exposed to laboratory lighting
for a few days, thus leading to complete decomposition of the
sample, sharp peaks corresponding to the nonacene decom-
position product were the only visible features of the NMR
spectrum. Recent theoretical work on larger acenes predicts
open-shell structures for species greater than octacene,[25] thus
1
explaining the lack of signal in the H NMR spectrum. We
thus subjected 3b to EPR analysis, and at both room
temperature and at 115 K, a signal was observed in the EPR
spectrum with g = 2.0060 (Figure 3) that disappeared in a time
frame corresponding to the typical lifetime of our nonacene
Figure 4. The long-wavelength portion of the absorption spectra for
silylethyne-substituted acenes from anthracene to heptacene[1b] and
nonacene 3c.
derivative 2 may arise from the extensive thiol substitution
present in 2, the strong correspondence between the absorp-
tion and emission spectra of decomposed samples of non-
acenes 3 and the spectra reported for 2 suggest that the weak
S0–S1 transition and unprecedented short-wavelength fluo-
rescence observed for 2 may arise from a decomposition
product that makes up the bulk of the sample. The ability to
synthesize and characterize larger acenes will provide sig-
nificant insight into the nature of aromaticity in linearly fused
aromatic hydrocarbons, and presents an interesting oligomer
study[27] for the narrowest of zigzag graphene nanoribbons.[28]
However, it must be emphasized that structures must be
characterized unambiguously for these studies to be relevant.
Figure 3. EPR spectrum of 3b in 2-methyltetrahydrofuran (9.45 GHz,
115 K). Solid line: pristine sample. Dashed line: sample after 48 h
exposure to air and light.
derivatives in solution. The signal indicates the presence of a
free radical at low concentration, thus explaining the diffi-
culty in obtaining 1H NMR spectra for 3, as well as the limited
lifetime of the compounds. The radical signal displays
evidence of unresolved hyperfine coupling or anisotropy,
and is not easily saturated. The oxidation potential of these
derivatives, combined with the use of rigorously deoxygen-
ated solvents, make it unlikely that this signal arises from a
radical cation oxidation product. The origin of this radical
signal, including whether it stems from an intermediate or
product of decomposition, or is an intrinsic characteristic of
these large acenes, is currently under investigation.
Experimental Section
Tetraols 5 were prepared as described in the Supporting Information.
For conversion to the nonacenes, solvents were optimized to
encourage the nonacenes 3 to crystallize as they were formed. For
derivative 3a, 5a was deoxygenated by the addition of solid
SnCl2·2H2O to a solution of 5a in 1,2,4-trichlorobenzene/ethyl-
benzene/acetonitrile (2:1:1 by volume), followed by brief, vigorous
stirring of the solution. When this mixture was left to stand in the dark
overnight, crystals of 3a grew from the solution. Under similar
conditions, 3b and 3c both crystallized from chlorobenzene/acetoni-
trile (1:1).
We have demonstrated that the silylethyne functionaliza-
tion strategy developed to produce stable hexacene and
heptacene derivatives can also be used to prepare nonacene
derivatives that are stable enough for detailed structural and
Received: April 18, 2011
Published online: June 29, 2011
7016
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 7013 –7017

[15] H. Qu, C. Chi, Org. Lett. 2010, 12, 3360.
Keywords: acenes · alkynes · aromatic compounds ·
organic materials
.
[16] I. Kaur, W. Jia, R. P. Kopreski, S. Selvarasah, M. R. Dokmeci, C.
Pramanik, N. E. McGruer, G. P. Miller, J. Am. Chem. Soc. 2008,
130, 16274.
[17] I. Kaur, N. N. Stein, R. P. Kopreski, G. P. Miller, J. Am. Chem.
Soc. 2009, 131, 3424.
[18] I. Kaur, M. Jazdzyk, N. N. Stein, P. Prusevich, G. P. Miller, J. Am.
Chem. Soc. 2010, 132, 1261.
[19] C. R. Swartz, S. R. Parkin, J. E. Bullock, J. E. Anthony, A. C.
Mayer, G. G. Malliaras, Org. Lett. 2005, 7, 3163.
[20] S. Subramanian, S. K. Park, S. R. Parkin, V. Podzorov, T. N.
Jackson, J. E. Anthony, J. Am. Chem. Soc. 2008, 130, 2706.
[21] a) New Materials Produced for Future Organic Electronics. Plus
Plastic Electronics, Feb 16 2010; b) Nonacene Shows Promise for
Organic Electronics. OSA Direct Newsletter, Feb 18 2010.
[22] M. Kasha, Discuss. Faraday Soc. 1950, 9, 14.
[23] The data for 2 are reproduced exactly as reported in Ref. [18].
We note that the axis labeling is not linear at long wavelengths
(1000–1200 nm), presumably because of a typographic error.
[24] CCDC 826567 (3a), 826568 (3b), and 826569 (3c) contains the
supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[25] a) M. Bendikov, H. M. Duong, K. Starkey, K. N. Houk, E. A.
Carter, F. Wudl, J. Am. Chem. Soc. 2004, 126, 7416; b) K. N.
Houk, P. S. Lee, M. Nendel, J. Org. Chem. 2001, 66, 5517; c) S.
Kivelson, O. L. Chapman, Phys. Rev. B 1983, 28, 7236; d) C.
Raghu, Y. A. Pati, S. Ramasesha, Phys. Rev. B 2002, 65, 155204;
e) J. Hachmann, J. J. Dorando, M. Avilꢁs, G. K.-L. Chan, J.
Chem. Phys. 2007, 127, 134309; f) B. Hajgatꢃ, D. Szieberth, P.
Geerlings, F. De Proft, M. S. Deleuze, J. Chem. Phys. 2009, 131,
224321.
[1] a) M. Bendikov, F. Wudl, D. Perepichka, Chem. Rev. 2004, 104,
4891; b) J. E. Anthony, Angew. Chem. 2008, 120, 460; Angew.
Chem. Int. Ed. 2008, 47, 452; c) R. W. I. de Boer, M. E.
Gershenson, A. F. Morpurgo, V. Podzorov, Phys. Status Solidi
A 2004, 201, 1302.
[2] S. F. Nelson, Y.-Y. Lin, D. J. Gundlach, T. N. Jackson, Appl. Phys.
Lett. 1998, 72, 1854.
[3] J. Shi, C. W. Tang, Appl. Phys. Lett. 2002, 80, 3201.
[4] J. Takeya, M. Yamagishi, Y. Tominari, R. Hirahara, Y. Naka-
zawa, T. Nishikawa, T. Kawase, T. Shimoda, Appl. Phys. Lett.
2007, 90, 102120.
[5] a) W.-Q. Deng, W. A. Goddard III, J. Phys. Chem. B 2004, 108,
8614; b) J.-L. Brꢁdas, D. Beljonne, V. Coropceanu, J. Cornil,
Chem. Rev. 2004, 104, 4971; c) F. Valiyev, W.-S. Hu, H.-Y. Chen,
M.-Y. Kuo, I. Chao, Y.-T. Chao, Chem. Mater. 2007, 19, 3018.
[6] S. S. Zade, M. Bendikov, Angew. Chem. 2010, 122, 4104; Angew.
Chem. Int. Ed. 2010, 49, 4012.
[7] R. Mondal, R. M. Adhikari, B. K. Shah, D. C. Neckers, Org. Lett.
2007, 9, 2505.
[8] R. Mondal, B. K. Shah, D. C. Neckers, J. Am. Chem. Soc. 2006,
128, 9612.
[9] C. Tꢂnshoff, H. F. Bettinger, Angew. Chem. 2010, 122, 4219;
Angew. Chem. Int. Ed. 2010, 49, 4125.
[10] a) H. F. Bettinger, R. Mondal, D. C. Neckers, Chem. Commun.
2007, 5209; b) R. Mondal, C. Tꢂnshoff, D. Khon, D. C. Neckers,
H. F. Bettinger, J. Am. Chem. Soc. 2009, 131, 14281.
[11] B. Purushothaman, S. R. Parkin, J. E. Anthony, Org. Lett. 2010,
12, 2060.
[12] J. E. Anthony, D. L. Eaton, S. R. Parkin, Org. Lett. 2002, 4, 15.
[13] M. M. Payne, S. R. Parkin, J. E. Anthony, J. Am. Chem. Soc.
2005, 127, 8028.
[26] F. Wꢄrthner, R. Schmidt, ChemPhysChem 2006, 7, 793.
[27] J. Wu, W. Pisula, K. Mꢄllen, Chem. Rev. 2007, 107, 718.
[28] M. Fujita, K. Wakabayashi, K. Nakada, K. Kisakabe, J. Phys.
Soc. Jpn. 1996, 65, 1920.
[14] D. Chun, Y. Cheng, F. Wudl, Angew. Chem. 2008, 120, 8508;
Angew. Chem. Int. Ed. 2008, 47, 8380.
Angew. Chem. Int. Ed. 2011, 50, 7013 –7017
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7017