Metal-interrupted perylene diimide: Toward a new class of tunable n-type inorganic-organic hybrid semiconductors

Article abstract of DOI:10.1021/ic100785r

In organic thin-film transistors (OTFTs), organic electron-transport materials (n-type semiconductors) are well behind the advances in development of hole-transport materials (p-type semiconductors). Currently, one class of organic n-type semiconductor materials that is widely utilized is N,N′-dialkyl-3,4,9,10-perylenetetracarboxylic diimide (PTCDI-R) derivatives with high electron affinities (EAs), such as N,N′-dioctyl-3,4, 9,10-perylenetetracarboxylic diimide with a reported EA as high as 4.4 eV. The PTCDI-R derivatives have been manipulated by adding substituents on the perylene moiety or at the amine position to afford more stable compounds and higher EAs. On the basis of these materials, we have developed metal-containing perylenediimide analogues, placing a salpen ligand for metal ion chelation between two n-isobutylnaphthalimides. We demonstrate here that the electronic properties of this class of materials can be systematically tuned in a divergent manner by simply changing the metal center. The synthesis, characterization, electrochemistry, and band-gap analysis are discussed herein.

Full text of DOI:10.1021/ic100785r

Inorg. Chem. 2010, 49, 6787–6789 6787
DOI: 10.1021/ic100785r
Metal-Interrupted Perylene Diimide: Toward a New Class of Tunable n-Type
Inorganic-Organic Hybrid Semiconductors
Kate R. Edelman and Bradley J. Holliday*
Department of Chemistry and Biochemistry and Center for Electrochemistry, The University of Texas at Austin,
1 University Station, A5300, Austin, Texas 78712-0165
Received April 27, 2010
In organic thin-film transistors (OTFTs), organic electron-transport
materials (n-type semiconductors) are well behind the advances in
development of hole-transport materials (p-type semiconductors).
Currently, one class of organic n-type semiconductor materials that
is widely utilized is N,N0-dialkyl-3,4,9,10-perylenetetracarboxylic
diimide (PTCDI-R) derivatives with high electron affinities (EAs),
such as N,N0-dioctyl-3,4,9,10-perylenetetracarboxylic diimide with
a reported EA as high as 4.4 eV. The PTCDI-R derivatives have
been manipulated by adding substituents on the perylene moiety or
at the amine position to afford more stable compounds and higher
EAs. On the basis of these materials, we have developed metal-
containing perylenediimide analogues, placing a salpen ligand for
metal ion chelation between two n-isobutylnaphthalimides. We
demonstrate here that the electronic properties of this class of
materials can be systematically tuned in a divergent manner by
simply changing the metal center. The synthesis, characterization,
electrochemistry, and band-gap analysis are discussed herein.
difficult to create robust OTFTs, which perform well when
doped. In addition, improving electron injection and trans-
port in organic devices requires materials with high electron
affinities (EAs).5,6 Specifically, organic transistors require
EA values of about 4.0 eV or higher to create ohmic contacts
to aluminum electrodes and even higher for gold elec-
trodes.7,8 Finally, tuning of the EA of organic semiconduc-
tors usually involves the addition of electron-withdrawing
substituents (i.e., -CN or F), and this often requires lengthy
tedious synthetic procedures.
To date, examples of organic n-type materials with high
EAs include hexadecafluorocopper phthalocyanine (F16CuPc)
and N,N0-dialkyl-3,4,9,10-perylenetetracarboxylic diimide de-
rivatives. The EA value for F16CuPc is 3.4 eV, and that of
N,N0-dioctylperylene-3,4,9,10-tetracarboxylic diimide (PTCDI-
C8) is 4.4 eV.9,10 The perylene diimides are of particular interest
not only because of high EAs but also because of the fact
that the electronic properties can be tuned in a rational way
through synthetic manipulations of the substituents around
the aromatic core.11,12 Herein, we describe a new class of
materials with the potential for use as n-type organic electro-
nics whereby a perylene diimide aromatic structure is inter-
rupted by a Schiff-base metal complex (Scheme 1). This
strategy has been developed, in part, to take advantage of
the ability to tune the electronic properties of these complexes
by simply changing the metal center.
Organic semiconductors were identified as early as 1940,
but low charge mobilities hindered the development of
practical applications of these materials. It was not until
the late 1980s when work on organic thin-film transistors
(OTFTs) reignited research on both polymer- and small-
molecule-based organic electronics.1-3 Organic materials are
advantageous over the inorganic counterparts because of
their ability to create flexible devices that can be produced on
a large scale and at relatively low cost.4
While a wide variety of organic hole-transport materials
(p-type semiconductors) for applications such as OTFTs
have been successfully developed, electron-transport materi-
als (n-type semiconductors) still have to overcome many
challenges. When highly charged, organic n-type materials
are reactive toward oxygen and moisture, thus making it
Our approach to developing metal-containing perylene
diimide analogues (Scheme 1) involves placing a salpen
ligand (salicyaldehydes connected through a 1,3-propylene-
diamine backbone) betweentwonaphthalimide moieties with
(5) Bao, Z.; Lovinger, A. J.; Dodabalapur, A. Appl. Phys. Lett. 1996, 69,
3066.
(6) Babel, A.; Jenekhe, S. A. Adv. Mater. 2002, 14, 371.
(7) Sirringhaus, H.; Tessler, N.; Friend, R. H. Science 1998, 280, 1741.
(8) Zhu, Y.; Yen, C.-T.; Jenekhe, S. A.; Chen, W.-C. Macromol. Rapid
Commun. 2004, 25, 1829.
(9) Hosoi, Y.; Tsunami, D.; Ishii, H.; Furukawa, Y. Chem. Phys. Lett.
2007, 436, 139.
*To whom correspondence should be addressed. E-mail: bholliday@
cm.utexas.edu.
(10) Kim, K.; Kwak, T. H.; Cho, M. Y.; Lee, J. W.; Joo, J. Synth. Met.
2008, 158, 553.
(11) Chesterfield, R. J.; McKeen, J. C.; Newman, C. R.; Ewbank, P. C.; da
Silva Filho, D. A.; Bredas, J.-L.; Miller, L. L.; Mann, K. R.; Frisbie, D. C. J.
Phys. Chem. 2004, 108, 19281.
(12) Liang, B.; Zhang, Y.; Wang, Y.; Xu, W.; Li, X. J. Mol. Struct. 2008,
917, 133.
(1) Horowitz, G. J. Mater. Res. 2004, 19, 1946.
(2) Organic Electronics: Materials, Manufacturing and Applications;
Klauk, H., Ed.; Wiley-VCH: New York, 2006.
(3) Newman, R. C.; Frisbie, C. D.; da Silva Filho, D. A.; Bredas, J.-L.;
Ewbank, P. C.; Mann, K. R. Chem. Mater. 2004, 16, 4436.
(4) Organic Field-Effect Transistors; Bao, Z., Locklin, J., Eds.; CRC Press:
Boca Raton, FL, 2007.
r
2010 American Chemical Society
Published on Web 06/29/2010
pubs.acs.org/IC

6788 Inorganic Chemistry, Vol. 49, No. 15, 2010
Edelman and Holliday
Table 1. λmax Values, Solution Reduction Potentials, Derived LUMO and
HOMO Energies, and Band-Gap Analysis
band gapc
a
b
c
c
λmax
E1/2 (red.)
ELUMO
EHOMO
4
5
6
7
229, 369
236, 427
236, 410
235, 407
-1.90
-1.93
-1.86
-1.85
-3.12
-3.19
-3.38
-3.71
-5.50
-5.03
-5.37
-5.21
2.38
1.84
1.99
1.50
Figure 1. ORTEP diagram of 5 showing the labeling scheme of selected
atoms with thermal ellipsoids drawn at the 30% probability level.
Hydrogen atoms and solvent molecules are omitted for clarity.
a In nm. b Volts vs Fc/Fcþ. c Electronvolts vs vacuum.
Scheme 1. Synthetic Scheme for Bis(isobutylnaphthalimide)salpen
Metal Complexesa
a (a) 3-Formyl-4-methoxybenzeneboronic acid, Na2CO3, PdCl2-
(PPh3)2, dry DMF. (b) 1.0 M BBr3 in CH2Cl2, dry CH2Cl2. (c) 2,2-
Dimethylpropane-1,3-diamine, EtOH, CH2Cl2. (d) M(OAc)2 xH2O,
3
EtOH, CH2Cl2.
isobutyl groups used to improve the solubility. The salpen
moiety was chosen because of its ease of metalation as well as
evidence that there is electronic communication through the
salen bonds to the metal in conducting metallopolymer
systems.13-15 By variation of the metal center, the relative
redox properties of the naphthalimide, in conjunction with
these metal centers, will be affected and studied. We have
chosen to specifically study nickel(II) (5), copper(II) (6), and
cobalt(II) (7) because of the II f I redox couples previously
reported for the respective salen complexes (vide infra).
The metalated salpen isobutylnaphthalimide complexes
(5-7, Scheme 1), as well as the precursors (1-4), have been
Figure 2. UV-vis absorption spectra of 1-7 in DCM at room tem-
perature.
crystal structure of a nickel(II) salen complex (Ni-Nav
=
16
˚
˚
1.871 A; Ni-Oav =1.846 A). The phenyl and naphthali-
mide ring planes are twisted with respect to each other, with
the torsion angles between the two moieties being 55.79° for
the left side of the molecule and 66.22° for the right side, as
displayed in Figure 1. A torsion angle of 21.94° is observed
between the individual naphthalimide rings.
1
synthesized and fully characterized by H and 13C NMR
spectroscopy (when practical), mass spectrometry, UV-vis
spectroscopy, cyclic voltammetry, and combustion analysis,
and all data are fully consistent with the proposed structures.
The solid-state structure of the nickel(II) metal complex (5)
was also determined by single-crystal X-ray diffraction anal-
ysis (Figure 1).15 This analysis reveals several interesting
features about complex 5. The four-coordinate NiII center
is in a twisted square-planar geometry (28.14°) that is defined
by the two nitrogen atoms and two oxygen atoms from the
The electronic absorption spectra of 1-7 have been studied
in dichloromethane (DCM) at room temperature. The λmax
values for 4-7 are displayed in Table 1. The absorption spectra
of 1-4 display two broad bands from 225 to 285 nm and from
300 to 400 nm (Figure 2, top). The absorption maxima (λmax
=
∼238 nm) of 1-4 are attributed to a π f π* transition of the
salicylaldehyde moiety, and the second displayed absorption
maxima (λmax=∼358 nm) are attributed to a π f π* transition
of the naphthalimide moiety.17,18 The absorption spectra of
5-7 display three broad bands: from 225 to 305 nm, from 300
to 375 nm, and from 380 to 470 nm (Figure 2, bottom). The two
higher-energy absorption maxima (λmax = ∼236 and ∼350
nm) of 5-7 are again attributed to π f π* transitions from the
salpen and naphthalimide moieties, respectively, while the
new lower-energy absorption maxima (λmax =∼414 nm) are
˚
salpen ligand. The Ni-Nav bond distance (1.882 A) is longer
˚
than the Ni-Oav bond distance (1.842 A). This trend and
these bond distances are consistent with a previouslyreported
(13) Kingsborough, R. P.; Swager, T. M. Adv. Mater. 1998, 10, 110.
(14) Kingsborough, R. P.; Swager, T. M. J. Am. Chem. Soc. 1999, 9, 8825.
(15) Crystal data for 5: C51H46N4NiO6 C3H7NO H2O, M = 960.74,
3
3
˚
˚
triclinic, space group P1, a = 9.6700(4) A, b = 14.0580(7) A, c =
˚
18.1070(10) A, R = 107.225(2)°, β = 94.799(3)°, γ = 104.624(2)°, V =
2241.24(19) A , Z = 2, Dcalcd = 1.424 g cm-3, μ = 0.498 mm-1, T = 153(2)
3
˚
K, R1 = 0.1727, R2 = 0.2209.
(16) Cengiz, A.; Filiz, E.; Kurtaran, R.; Atakol, O. Acta Crystallogr.,
Sect. C: Cryst. Struct. Commun. 2001, C57, 812.
(17) Wintgens, V.; Valat, P.; Kossanyi, J. New J. Chem. 1996, 20, 1149.
(18) Nandhikonda, P.; Begaye, M. P.; Cao, Z.; Heagy, M. D. Chem.
Commun. 2009, 4941.

Communication
Inorganic Chemistry, Vol. 49, No. 15, 2010 6789
E1/2 = -0.58 V, which should have absolutely no overlap with
the ligand redox couple.20-23
Cyclic voltammetry of 5 illustrates a slight shift of the rever-
sible reduction to E1/2 =-1.93 V and displays good overlap
between the metal and ligand reduction potentials. Complex 6
overlaps more than predicted, with a peak at -1.76 V that is
associated with the reduction of copper from copper(II) to
copper(I), while the ligand redox couple displays a shift to
E1/2=-1.86 V. The cobalt complex 3 shows two well-separated
redox couples: one that is ligand-based at E1/2 = -1.85 V and
one that is metal-based at -0.80 V, corresponding to the
cobalt(II/I) redox couple.
The EAs [lowest unoccupied molecular orbital (LUMO)
level] of 4-7, estimated from the onset of the reduction
potential by conversion to SCE and taking the SCE energy
level to be -4.8 eV relative to the vacuum level (EA = Eonset red
þ 4.8), are 3.12, 3.19, 3.38, and 3.71 eV, respectively.24 These
EA values are comparable with F16CuPc values but are lower
than the value reported for PTCDI-C8 (vide supra).
Band-gap analysis, i.e., determination of the highest occu-
pied molecular orbital (HOMO) and LUMO positions, is
necessary for determining the majority of charge carriers and
inferring the stability of the charge carrier in organic semi-
conductors. HOMO and LUMO levels were calculated from
the onset of oxidation and reduction in the electrochemical
data for complexes 4-7 with respect to the vacuum level
(Table 1).24 The band gaps appear to be on the same order as
those of several previously reported peryelene derivatives
studied for photovoltaic devices by Shin et al.25 The calcu-
lated band gaps for 4-7 are 2.38, 1.84, 1.99, and 1.50 eV,
respectively.
In summary, we have synthesized and characterized a
series of metal-containing perylene diimide analogues and
fully investigated their electrochemical behavior. These com-
plexes exhibit EAs comparable to those of n-type materials
currently used in OTFT fabrication. With further inves-
tigation and manipulation of the ligand structure in the
future, we plan to try to tune the electronics of this potential
new class of hybrid semiconductors to improve upon the
obtained EAs.
Figure 3. Cyclic voltammograms of compounds 4-7.
attributed to M f L charge-transfer transitions, which was
confirmed by solvatochromism (see the Supporting Informa-
tion).19
In order to determine the EA of the salpen isobutylnaphtha-
limide materials, the electrochemical properties were investi-
gated. Cyclic voltammetry was performed over a window of
þ0.50 to -2.05 V (vs Fc/Fcþ) at a scan rate of 100 mV/s in
anhydrous dimethylformamide (DMF) to determine the re-
duction potentials of the various compounds. Additionally,
separate cyclic voltrammetry experiments were performed
over a window of -0.5 to þ1.5 V (vs Fc/Fcþ) at a scan rate
of 100 mV/s in anhydrous DCM to locate the onset of
oxidation for band-gap analysis. The ligand (4) shows a
reversible reduction wave with E1/2 = -1.90 V (Figure 3).
With this information, metals were chosen according to how
well the redox couples of these metal ions in salen ligands
overlapped with the reduction potential of 4. By varying the
electrochemical overlap, we can examine what role the metal
might play in the realized electronic properties of the materials
and how its redox chemistry in relation to the ligand affects the
realized EA. From previous literature reports, the nickel(II/I)
salen redox couple is E1/2 = -1.69 V, which should create
overlap with the redox couple of the flanking organic groups,
the copper(II/I) salen redox couple is E1/2 = -1.27 V, which
should not overlap well, and the cobalt salen redox couple is
Acknowledgment. We gratefully acknowledge the Welch
Foundation (F-1631), the PRF/ACS (47022-G3), the NSF
(CHE-0639239, CHE-0741973, and CHE-0847763), the
THECB (ARP 003658-0010-2006), the UT-CNM, The Na-
tional Institute for Nano-Engineering (NINE) Program at
Sandia National Laboratories, and UT-Austin for support-
ing this research.
Supporting Information Available: Experimental details for
the synthesis and characterization of 1-7, electrochemical data
for 1-4, a solvatochromism UV-vis study of 5, X-ray diffrac-
tion tables, and crystallographic data for 5 in CIF format. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(21) Gach, P. C.; Karty, J. A.; G., P. D. J. Electroanal. Chem. 2008,
612, 22.
(22) Hoferkamp, L. A.; Goldsby, K. A. Chem. Mater. 1989, 1, 348.
(23) Kingsborough, R. P.; Swager, T. M. Chem. Mater. 2000, 12, 872.
(24) Tonzola, C. J.; Alam, M. M.; Jenekhe, S. A. Adv. Mater. 2002, 14,
1086.
(19) Bella, S. D.; Fragala, I.; Ledoux, I.; Diaz-Garcia, M. A.; Marks, T. J.
J. Am. Chem. Soc. 1997, 119, 9550.
(20) Fritzche, S.; Lonnecke, P.; Hocher, T.; Hey-Hawkins, E. Z. Anorg.
Allg. Chem. 2006, 632, 2256.
(25) Shin, W. S.; Jeong, H.-H.; Kim, M.-K.; Jin, S.-H.; Kim, M.-R.; Lee,
J.-K.; Lee, J. W.; Gal, Y.-S. J. Mater. Chem. 2006, 16, 384.