Solution-shear-processed quaterrylene diimide thin-film transistors prepared by pressure-assisted thermal cleavage of swallow tails

Article abstract of DOI:10.1021/ja110486s

A scalable synthesis of swallow-tailed quaterrylene diimides (STQDIs) and a method for the solution processing of sparingly soluble quaterrylene diimide (QDI) thin films are described. The pressure-assisted thermal cleavage of swallow tails yields crystalline QDI layers with electron mobility up to 0.088 cm² V^-1 s^-1. The developed method opens up a new route toward the solution processing of higher rylene diimides with poor solubility.

Full text of DOI:10.1021/ja110486s

COMMUNICATION
pubs.acs.org/JACS
Solution-Shear-Processed Quaterrylene Diimide Thin-Film Transistors
Prepared by Pressure-Assisted Thermal Cleavage of Swallow Tails
||
Joon Hak Oh,†, Wen-Ya Lee,†,^ Torsten Noe,‡ Wen-Chang Chen,^ Martin K€onemann,*,‡ and Zhenan Bao*,†
†Department of Chemical Engineering, Stanford University, Stanford, California 94305, United States
‡BASF SE, 67056 Ludwigshafen, Germany
School of Nano-Bioscience and Chemical Engineering, Ulsan National Institute of Science & Technology, Ulsan 689-798, Korea
^Department of Chemical Engineering, National Taiwan University, Taipei, Taiwan
S Supporting Information
b
a postprocessing step. One promising approach to this end is the
ABSTRACT: A scalable synthesis of swallow-tailed quater-
rylene diimides (STQDIs) and a method for the solution
processing of sparingly soluble quaterrylene diimide (QDI)
thin films are described. The pressure-assisted thermal cleavage
of swallow tails yields crystalline QDI layers with electron
mobility up to 0.088 cm2 V-1 s-1. The developed method
opens up a new route toward the solution processing of
higher rylene diimides with poor solubility.
thermal cleavage of solubilizing groups, which was used in photo-
resists,14 soluble small-molecule derivatives of pentacene,15
oligothiophene,16 and porphyrin17 as well as in polymers such as
polyacetylene,18 polythiophene,19 polyphenylenevinylene,20 and
benzothiadiazole/pyrrole-based polymers.21 Recently, thermally
cleavable groups have been used for PDIs.22 However, the
removal of bulky pendant groups often causes void spaces and
high surface roughness in the annealed film, and careful control
of process conditions is required to optimize the performance.23
Furthermore, although the electrical performance of several
swallow-tailed rylene diimides was reported,24,25 to our knowl-
edge, there is no report on the successful fabrication of devices
from sparingly soluble QDI derivatives with no substituent or a
low degree of alkyl or aryl substitution, which are more desirable
for high-performance applications.
Herein, we describe a scalable synthesis of soluble QDI deriva-
tives and a solution-based approach to the fabrication of high-
performance n-channel organic thin-film transistors (OTFTs) based
on crystalline QDI thin films. Two swallow-tailed QDIs (namely,
STQDIs: see Scheme 1) were synthesized and solution pro-
cessed to form thin films by solution shearing26 or spin coating.
In the solution shearing, a small volume of an organic semicon-
ductor solution is sandwiched between two preheated silicon
wafers that move relative to each other at a controlled speed
(Figure 1a). This method yields crystalline and aligned thin films
from various soluble organic semiconductors.26 It has been
reported that 4 exhibits phase transition from plastic crystalline
phase (Colp) to columnar hexagonal ordered liquid crystalline
phase (Colho) at 188 ꢀC (Table S1 and Figure S1 [Supporting
Information {SI}]).4 However, thermal cleavage of the swallow
tails was not reported. We found that the thermal cleavage of the
swallow tails of the STQDIs takes place at a relatively lower
temperature ranging from 350 to 400 ꢀC. A very clean conversion
to an ordered crystalline QDI layer was found. n-Channel
OTFTs were produced by the approach of cleaving off solubliz-
ing R groups. The crystallinity of the resulting thin films could be
further enhanced by pressure-assisted thermal cleavage, that is,
a modified annealing technique with a flat substrate pressed
on the film during the thermal cleavage (Figure 1b). This
method also prevents film dewetting commonly observed for a
ylene diimides have received much attention in dye che-
R
mistry,1 supramolecular assemblies,2 and optoelectronic
applications3 due to their chemical and photochemical stability
as well as property tunability.4 Compared to naphthalene dii-
mides (NDIs)5 and perylene diimides (PDIs),6,7 whose absorp-
tion wavelengths are limited in the visible region, quaterrylene
diimides (QDIs) have near-infrared (NIR) absorbing character-
istics with a high extinction coefficient (∼105 M-1 cm-1).8,9
Expansion of conjugation along the long molecular axis from
NDI to QDI raises the highest occupied molecular orbital
(HOMO) level significantly, while the lowest unoccupied mo-
lecular orbital (LUMO) level is not affected substantially.10
Using Marcus theory, Feng et al. found that, although the maxima
of the overlap integral may decrease with the increase of the
polyaromatic hydrocarbon core size, the overall hopping rate
increases owing to the simultaneous decrease of the reorganiza-
tion energy.11
The above features make QDIs suitable for a wide range of
applications such as solar cells, NIR photodetectors, heat block-
ers, laser welding or marking of polymers, noninvasive imaging
probes, IR photography, electrogenerated chemiluminescence
labels, and single-molecule field-effect transistors.12 Neverthe-
less, they have been far less employed for device applications
because of the very high sublimation temperature and extremely
low solubility. The substitution of branched alkyl chains at the
imide position has been demonstrated not only to increase
solubility but also to lower the temperature of the liquid crystal-
line phase for enabling facile processing.4,13 However, these
nonconjugated substituents reduce the density of chromophores
and disturb the π-stacking, resulting in a low charge carrier
mobility. Therefore, it is desirable to remove these substituents in
Received: November 22, 2010
Published: March 04, 2011
r
2011 American Chemical Society
4204
dx.doi.org/10.1021/ja110486s J. Am. Chem. Soc. 2011, 133, 4204–4207
|

Journal of the American Chemical Society
COMMUNICATION
Scheme 1. Synthesis of STQDIs
Figure 2. (a) Thermogravimetric analysis of STQDIs. The thermal
cleavage of 5 (red dashed line) takes place at a temperature lower than
that of 4 (black solid line) by about 40 ꢀC. The second weight loss above
650 ꢀC is due to the decomposition of QDI 6. (b) MALDI-TOF MS
spectra of 4 (black solid line) and 5 (red dashed line) after annealing at
400 and 370 ꢀC for 1 h, respectively. (c) The thermal cleavage
mechanism of STQDIs.
Figure 1. Schematic diagrams for (a) solution shearing and (b)
pressure-assisted thermal cleavage.
branched amines, respectively. Sterical reasons could account for
the difference in yield. The base-induced dimerization of PMIs
using KOtBu, DBU, and ethanolamine yielded 30% and 23% of 4
and 5, respectively. Our synthetic approach describes the first
synthesis applicable to the industrial-scale production of these
materials.
high-temperature annealing24 and improves the uniformity of the
device performance. It was reported that drop-cast 4 thin films
exhibit ambipolarity with the balanced hole and electron mobi-
lities of about 1 Â 10-3 cm2 V-1 s-1 25
.
After the thermal
annealing at 100 ꢀC, the hole conduction disappeared, and the
electron mobility decreased by an order of magnitude due to the
increased trapping by the morphological change.25,27,28 In this
study, we significantly improved the electron mobility of QDI-
based TFTs up to 0.088 cm2 V-1 s-1, through the pressure-
assisted thermal cleavage of the solution-processed STQDI thin
films. Our approach demonstrates a new route toward the
solution processing of QDIs and can be expanded to the
fabrication of other sparingly soluble rylene diimide thin films.
The synthesis of 5 has not been reported previously. The
synthesis of 4 analogue (R = CH(C6H13)2) was reported by
Langhals et al. in one step with a yield of 4% using a KOH melt at
300 ꢀC.9 M€ullen et al. synthesized 4 in four steps with an overall
yield of 46% from perylene-3,4-dicarboxylic monoimide 2 (PMI).4
The base-induced dimerization of PMIs to QDIs using potas-
sium-tert-butoxide (KOtBu) was described earlier,29 but only for
aryl-substituted compounds which are more stable toward base-
induced saponification compared to swallow-tailed R groups.
A one-step base-induced dimerization of PMIs to QDIs using
sodium-tert-butoxide (NaOtBu) was proposed;27 however, no
demonstration has been reported. Here we synthesized 4 with a
yield of 30% via a modified “green route”29,30 in mild basic
conditions using 8-diazabicyclo[5.4.0]undec-7-ene (DBU) and
ethanolamine. We replaced 1,5-diazabicyclo[4.3.0]non-5-ene
(DBN) used in ref 30 with the more readily available DBU.
Starting from perylene-3,4-dicarboxylic monoanhydride (1),
the imidization was carried out using a standard procedure in
quinoline with zinc acetate as the catalyst (Scheme 1). Yields of
66% and 34% were obtained for the less branched and more
The thermal cleavage of the swallow tails of 4 and 5 took place
at 394 and 353 ꢀC, respectively, as shown in Figure 2a. The mass
loss was approximately 40%, which corresponds well to the loss
of the two alkyl chains. Mass spectrometry (MS) analyses on the
residues after the thermogravimetry analysis (TGA) of 4 and 5
showed strong peaks respectively at 638.1 and 639.0 g mo1-1
3
(Figure 2b), which agrees well with the molecular weight (638.6
g mo1-1) of 6. Whereas W€urthner et al. described the cleavage
3
of branched groups from rylene diimides by using strong Lewis
acid BBr3 in a nonquantitative yield,31 we found that the thermal
cleavage of STQDIs surprisingly leads to a very clean conversion
to materials with semiconductor-grade purity without the use of
chemical reagents in a short reaction time (<1 h). In order to
lower the cleaving temperature of 4, 4,6-dipropylnon-5-yl-amine
that provides highly substituted and more stable olefin upon
thermal cleavage was used for 5 (Figure 2c).
UV-vis spectra of pristine 6 and annealed 4 at 420 ꢀC
exhibited similar absorptions in sulfuric acid with the identical
λmax of 872 nm and the very high extinctions in the NIR region
(see Figure S2 [SI]), indicating that there is no visible degrada-
tion due to the thermal annealing. Spin-coated thin films of 4 and
5 showed a long absorption wavelength up to ∼1000 nm with
λmax of 688 and 707 nm, respectively (Figure S3 [SI]). After
thermal cleavage, the absorption edge exhibited a bathochromic
shift of 150 nm, indicating a very small band gap of ∼1.1 eV. The
broader absorption range of the annealed STQDI films indicates
that the thermal cleavage not only removes the alkyl chains but
also enhances π-π stacking of the molecules, i.e. the aggregation
of 6 chromophores. Ultraviolet photoelectron spectroscopy
4205
dx.doi.org/10.1021/ja110486s |J. Am. Chem. Soc. 2011, 133, 4204–4207

Journal of the American Chemical Society
COMMUNICATION
Table 1. Electronic and OTFT Characteristics of Solution-
Sheared STQDI Thin Films
suppressed the 3D growth of the grains and forced the grains to
coalesce, leading to less empty space and enhanced mobility.
Intriguingly, we found that the surface treatment of the cover
wafer for pressing plays an important role in the performance of
the resulting devices. A PTS-treated cover wafer exhibited higher
μ, compared to the bare or n-octadecyltrimethoxysilane (OTS)-
treated one (Figure S11 [SI]). This is presumably related to the
difference in the wetting properties of the molecules on the
surface, which affects molecular reorganization during the an-
nealing (Figure S12 [SI]).
In summary, we developed a scalable synthesis of soluble
QDIs and demonstrated a solution fabrication of crystalline,
NIR-absorbing QDI thin films. The swallow tails were used as the
solubilizing agent for the insoluble QDI moiety and could be
subsequently removed by thermal treatment at temperatures
ranging from 350 to 400 ꢀC, sufficiently low to provide QDI 6
without degradation. The pressure-assisted thermal cleavage
significantly reduced the surface roughness and void space of
the resulting films, leading to μ up to 0.088 cm2 V-1 s-1. The
developed method opens up a new route toward the solution-
based fabrication of the crystalline thin films of sparingly soluble
higher rylene diimides.
λmax
Eg
μ
Ion/Ioff
Vt
cmpd annealinga [nm] [eV] [cm2 V-1 s-1
]
[V]
4
5
4
5
before
before
after
684
705
597
609
1.3
1.3
1.1
1.1
2.9 Â 10-5
2.3 Â 103 -19
1.1 Â 10-4
0.088
8.8 Â 103
5
2.5 Â 103 -1.8
4.9 Â 104 -0.5
after
0.061
a The annealings of 4 and 5 were performed at 400 and 370 ꢀC for 1 h
under the pressure of 10 and 20 kPa, respectively.
(UPS) revealed the HOMO levels of -5.4 to -5.6 eV for the
annealed films. This means the LUMO levels of the annealed
films correspond to -4.3 to -4.5 eV, which are located in the
empirical LUMO range for air-stable n-channel operation.6
To investigate the effects of thermal cleavage on the charge
transport, top-contact OTFTs were fabricated with a SiO2
(300 nm)/n-doped Si wafer with the surface modification layer
of phenyltrichlorosilane (PTS). The as-prepared OTFTs before
annealing showed electron mobility (μ) ranging from 10-4 to
10-5 cm2 V-1 s-1, whereas after the thermal cleavage they
’ ASSOCIATED CONTENT
yielded 2-3 orders of magnitude higher μ up to 0.088 cm2 V-1 -1
s .
This indicates that the thermal cleavage of the branched chains
indeed facilitates the molecular packing and thus improves
charge transport. The optimized annealing temperatures of 4
and 5 were 400 and 370 ꢀC, respectively. The solution-sheared
thin films typically yielded 1 order of magnitude higher μ com-
pared to that by spin-coated thin films under pressure, due to
their aligned and crystalline nature. The optimized performance
of solution-sheared OTFTs and the annealing conditions are
listed in Table 1 (for the performance details, see Table S2 and
Figure S4 [SI]). The STQDI TFTs were not as air-stable as other
rylene diimides with similarly low-lying LUMOs.6 In addition,
the solution-sheared OTFTs showed relatively higher air stability
with the same order of magnitude mobility after exposure to air
for a week, while the spin-coated devices degraded more rapidly
with about 1 order of magnitude decrease (Figure S5 [SI]). The
solution-sheared thin films benefit from the kinetic factors due to
the larger crystalline grains. This is consistent with our recent
finding that the film thickness and morphology also impact the
air stability in addition to the LUMO level of the molecule.32
The out-of-plane X-ray diffraction (XRD) measurements of 5
thin films annealed at 370 ꢀC exhibited a sharp primary peak at
2θ = 3.54ꢀ (Figure S6 [SI]). This corresponds to a d(001)-
spacing of 24.94 Å, close to the geometry-optimized molecular
length of 6 (23.7 Å).33 This result indicates that the molecules
adopt an edge-on orientation in the thin films and have favorable
molecular orientations for charge transport between source-to-
drain electrodes. The morphology of as-annealed thin films
showed some portions of a void area and a high roughness
(Figures S7 and S8 [SI]), thereby giving rise to some device-to-
device variations. However, the pressure-assisted thermal clea-
vage substantially improved the performance and uniformity
(Figure S9 [SI]). Upon annealing under a pressure of 20 kPa, the
average mobilities of solution-sheared and spin-coated films of 5
were enhanced by 4-8 times compared to those prepared
without pressure. The roughness of 5 thin films decreased
significantly from 12.8 nm, down to 6.7 nm for 10 kPa and
5.7 nm for 20 kPa (Figure S10 [SI]). The applied pressure
S
Supporting Information. Experimental details, OTFT
b
data, UV-vis spectra, XRD data, optical microscope images,
and AFM images. This material is available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
martin.koenemann@basf.com; zbao@stanford.edu
’ ACKNOWLEDGMENT
This work was supported by BASF SE and the Sloan Research
Fellowship. J.H.O. acknowledges partial financial support from
Basic Science Research Program through the National Research
Foundation of Korea (NRF) funded by the Ministry of Educa-
tion, Science and Technology (Grant No. 2010-0025292).
’ REFERENCES
(1) Hunger, K. Industrial Dyes: Chemistry, Properties, Applications;
Wiley-VCH: Weinheim, 2003; Vol. 1.
(2) W€urthner, F. Chem. Commun. 2004, 1564–1579.
(3) Schmidt-Mende, L.; Fechtenkotter, A.; M€ullen, K.; Moons, E.;
Friend, R. H.; MacKenzie, J. D. Science 2001, 293, 1119–1122.
(4) Nolde, F.; Pisula, W.; M€uller, S.; Kohl, C.; M€ullen, K. Chem.
Mater. 2006, 18, 3715–3725.
(5) Oh, J. H.; Suraru, S.-L.; Lee, W.-Y.; K€onemann, M.; H€offken,
H. W.; R€oger, C.; Schmidt, R.; Chung, Y.; Chen, W.-C.; W€urthner, F.;
Bao, Z. Adv. Funct. Mater. 2010, 20, 2148–2156.
(6) Jones, B. A.; Facchetti, A.; Wasielewski, M. R.; Marks, T. J. J. Am.
Chem. Soc. 2007, 129, 15259–15278.
(7) Oh, J. H.; Lee, H. W.; Mannsfeld, S.; Stoltenberg, R. M.; Jung, E.;
Jin, Y. W.; Kim, J. M.; Yoo, J.-B.; Bao, Z. Proc. Natl. Acad. Sci. U.S.A. 2009,
106, 6065–6070.
(8) (a) Quante, H.; M€ullen, K. Angew. Chem., Int. Ed. 1995, 34,
1323–1325. (b) Adachi, M.; Nagao, Y. Chem. Mater. 2001, 13, 662–669.
(c) Lee, S. K.; Zu, Y.; Herrmann, A.; Geerts, Y.; M€ullen, K.; Bard, A. J.
J. Am. Chem. Soc. 1999, 121, 3513–3520.
4206
dx.doi.org/10.1021/ja110486s |J. Am. Chem. Soc. 2011, 133, 4204–4207

Journal of the American Chemical Society
COMMUNICATION
(9) Langhals, H.; Sch€onmann, G.; Feiler, L. Tetrahedron Lett. 1995,
36, 6423–6424.
(10) Lee, S. K.; Zu, Y.; Herrmann, A.; Geerts, Y.; M€ullen, K.; Bard,
A. J. J. Am. Chem. Soc. 1999, 121, 3513–3520.
(11) Feng, X.; Marcon, V.; Pisula, W.; Hansen, M. R.; Kirkpatrick, J.;
Grozema, F.; Andrienko, D.; Kremer, K.; M€ullen, K. Nat. Mater. 2009,
8, 421–426.
(12) (a) Fabian, J.; Nakazumi, H.; Matsuoka, M. Chem. Rev. 1992,
92, 1197–1226. (b) Zhao, W.; Carreira, E. M. Angew. Chem., Int. Ed.
2005, 44, 1677–1679.
(13) (a) Wicklein, A.; Lang, A.; Muth, M.; Thelakkat, M. J. Am.
Chem. Soc. 2009, 131, 14442–14453. (b) Prehm, M.; Liu, F.; Zeng, X.;
Ungar, G.; Tschierske, C. J. Am. Chem. Soc. 2008, 130, 14922–14923.
(14) Frꢀechet, J. M. J.; Eichler, E.; Ito, H.; Willson, C. G. Polymer
1983, 24, 995–1000.
(15) (a) Herwig, P. T.; M€ullen, K. Adv. Mater. 1999, 11, 480–483.
(b) Afzali, A.; Dimitrakopoulos, C. D.; Breen, T. L. J. Am. Chem. Soc.
2002, 124, 8812–8813.
(16) Murphy, A. R.; Frꢀechet, J. M. J.; Chang, P.; Lee, J.; Subramanian,
V. J. Am. Chem. Soc. 2004, 126, 1596–1597.
(17) Aramaki, S.; Sakai, Y.; Ono, N. Appl. Phys. Lett. 2004,
84, 2085–2087.
(18) Edwards, J. H.; Feast, W. J.; Bott, D. C. Polymer 1984, 25, 395–398.
(19) Liu, J.; Kadnikova, E. N.; Liu, Y.; McGehee, M. D.; Frꢀechet,
J. M. J. J. Am. Chem. Soc. 2004, 126, 9486–9487.
(20) Wagaman, M. W.; Grubbs, R. H. Macromolecules 1997,
30, 3978–3985.
(21) Edder, C.; Armstrong, P. B.; Prado, K. B.; Frꢀechet, J. M. J. Chem.
Commun. 2006, 1965–1967.
(22) Ando, S.;Facchetti, A.;Marks, T. J. Org. Lett. 2010, 12, 4852–4855.
(23) Chang, P. C.; Lee, J.; Huang, D.; Subramanian, V.; Murphy,
A. R.; Frꢀechet, J. M. J. Chem. Mater. 2004, 16, 4783–4789.
(24) Liu, C.; Liu, Z.; Lemke, H. T.; Tsao, H. N.; Naber, R. C. G.; Li,
Y.; Banger, K.; M€ullen, K.; Nielsen, M. M.; Sirringhaus, H. Chem. Mater.
2010, 22, 2120–2124.
(25) Tsao, H. N.; Pisula, W.; Liu, Z.; Osikowicz, W.; Salaneck, W. R.;
M€ullen, K. Adv. Mater. 2008, 20, 2715–2719.
(26) Becerril, H. A.; Roberts, M. E.; Liu, Z.; Locklin, J.; Bao, Z. Adv.
Mater. 2008, 20, 2588–2594.
(27) Avlasevich, Y.; Li, C.; M€ullen, K. J. Mater. Chem. 2010, 20
3814–3826.
(28) Schmaltz, B.; Weil, T.; M€ullen, K. Adv. Mater. 2009, 21
1067–1078.
(29) K€onemann, M.; B€ohm, A.; Bidlingmaier, H.; Rieger, R.; Blaschka,
P.; Reichelt, H.; Krieger, M. U.S. 7,741,487, 2010.
(30) Sakamoto, T.; Pac, C. J. Org. Chem. 2000, 66, 94–98.
(31) Wang, H.; Kaiser, T. E.; Uemura, S.; W€urthner, F. Chem.
Commun. 2008, 1181–1183.
(32) Oh, J. H.; Sun, Y.-S.; Schmidt, R.; Toney, M. F.; Nordlund, D.;
K€onemann, M.; W€urthner, F.; Bao, Z. Chem. Mater. 2009, 21, 5508–
5518.
(33) Wagner, C.; Franke, R.; Dienel, T.; Forker, R.; Jacob, R.; Fritz,
T. Appl. Phys. Lett. 2007, 91, 113111.
4207
dx.doi.org/10.1021/ja110486s |J. Am. Chem. Soc. 2011, 133, 4204–4207