Low-temperature formation of self-assembled 1,5-diaminoanthraquinone
nanofibers: Substrate effects and field emission characteristics
Kuo-Jung Huang
a, Yu-Sheng Hsiao
b,⇑, Wha-Tzong Whang
a,⇑ aDepartment of Materials Science and Engineering, National Chiao Tung University, Hsinchu 30010, Taiwan, ROC
b
Research Center for Applied Sciences, Academia Sinica, Taipei 11529, Taiwan, ROC
a r t i c l e
i n f o
Article history:
Received 11 October 2010
Received in revised form 11 January 2011 Accepted 12 January 2011
Available online 23 February 2011
Keywords: DAAQ Self-assemble Nanofibers
Small organic molecules Field emission
a b s t r a c t
In this study, we used thermal evaporation with a low sublimation temperature (42 °C) to deposit 1,5-diaminoanthraquinone (DAAQ) in various morphologies—including nanorods, nanocornerstones, and nanofibers—onto various substrates. Three major factors influenced the growth of vertically aligned DAAQ nanofibers on the electrodes: a low water contact angle (WCA) for the substrate and intermolecular hydrogen bonding andp–pinteractions between the DAAQ molecules. On Au and Ti substrates (low-WCA), the DAAQ nanofibers DAAQ-Au and DAAQ-Ti, respectively, possessed great verticality and high aspect ratios; they also exhibited field emission characteristics, with maximum emission current densi-ties of 0.31 and 0.65 mA/cm2, respectively, at an applied electric field of 12 V/lm. The
turn-on electric fields for producing a current density of 10lA/cm2were 8.5 V/
lm for DAAQ-Au and 8.25 V/lm for DAAQ-Ti. From the slopes of Fowler–Nordheim plots, we calculated the field enhancement factors (b) of DAAQ-Au and DAAQ-Ti to be 447 and 831, respectively. Field emission stability studies revealed that the DAAQ nanofibers possessed outstanding anti-degrading capability. The emission current did not decrease, but rather increased slightly, after 3000 s. Given the advantages of this simple low-temperature process and the impressive anti-degrading field emission characteristics, such DAAQ nanofibers have great potential for use in various electronics applications (e.g., as organic field emitters).
Ó 2011 Elsevier B.V. All rights reserved.
1. Introduction
Nanostructures based small organic molecules have be-come the focus of intensive research efforts because of their unique applications in nanoscale devices [1–3]. One-dimensional (1D) organic nanostructures, including nanorods, nanofibers, nanowires, and nanotubes, can be fabricated using a wide range of methods; they often exhi-bit unique material properties that make them suitable for use in organic electronics applications (e.g., solar cells
[4–7], chemical vapor sensors[8], field emitters[9–15]) and in various optoelectronic devices [16–19]. Recently, several small organic molecules, including tris(8-hydrox-yquinolinato) aluminum (Alq3) [9,13,16,19], anthracene
(AN), perylene (PY)[20], coronene[10], copper phthalocy-anine (CuPc)[11], and 1,5-diaminoanthraquinone (DAAQ) [8], have been prepared in form of various nanostructures under conditions milder than those used to prepare related nanostructures from inorganic compounds. Be-cause of their suitability for use in inexpensive, low tem-perature-processed, highly flexible optoelectronic devices, increasing attention is being drawn to the development of nanostructures of small-molecule organic materials.
Several fabrication methods have been reported for the preparation of organic 1D nanostructures [1–3,21]. Nanowires and nanorods have been prepared through thermal evaporation (solid state chemical reactions)[20] 1566-1199/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.orgel.2011.01.010
⇑ Corresponding authors. Tel.: +886 2 27898000-21; fax: +886 2 27826680 (Y.S. Hsiao), tel.: +886 35 731873; fax: +886 35 724727 (W.T. Whang).
E-mail addresses: [email protected] (Y.-S. Hsiao),
[email protected](W.-T. Whang).
Contents lists available atScienceDirect
Organic Electronics
nanofibers for organic field emission devices[10–12]. We have previously applied high vacuum processing to fabri-cate nanostructures with highly uniform distributions on targeting substrates, without the need for catalysts or high temperatures. Such simple one-step approaches not only eliminate the effects of impurities but also potentially lower the fabrication temperature.
Recent investigations of DAAQ-based optoelectronic de-vices, such as chemical sensors [8], optical waveguides [27], and organic solar cells[28], have led to a rapid devel-opment of DAAQ fabrication technologies. Zhao et al.[27] recently reported that this organic dye prefers to undergo vertical deposition on substrates having high surface ener-gies (SEs) and hydrophilic domains. In this present study, we synthesized 1D DAAQ nanofibers through vacuum sub-limation on selected substrates. We characterized these 1D DAAQ nanofibers using scanning electron microscopy (SEM) and high-resolution transmission electron micros-copy (HRTEM). From investigations of the morphologies of DAAQ thin films formed on different substrates, we se-lected electrodes suitable for the vertical deposition of DAAQ nanofibers. To determine the practicality of employ-ing DAAQ nanofibers as cathode materials in field emission devices, we evaluated their field emission characteristics using a vacuum emission measurement (VEM) system.
Stabilized by intermolecular hydrogen bonding and
p
–p
interactions, DAAQ self-assembles onto substrates to form vertically out-of-plane nanostructures. Analogous to organic vapor-phase deposition (OVPD), the vacuum subli-mation of small organic molecules is also a vapor-solid (VS) process. Because sublimation proceeds at lower tem-peratures in higher vacuums, relative to those used for OVPD, its use opens up a new area of study and potentially increases the applications of organic small molecules in flexible electronic devices.2. Experimental
Commercially available DAAQ powder (purity: 92%; Tokyo Chemical Industry) and 1,1,1,3,3,3-hexamethyldisi-lazane (HMDS; Lancaster), for Si surface treatment, were used as received without further purification.
DAAQ films were deposited on various substrates (Ag, Al, Au, Ni, Si, Ti) using a thermal evaporator operated at a
Top and cross-sectional views of the morphologies of the DAAQ nanostructure films were surveyed using a JEOL JSM-6500F scanning electron microscope. Ultraviolet– visible (UV–Vis) spectra (250–900 nm) were recorded using a Shimadzu UV-3600 spectrophotometer. Contact angles (CAs) and SEs were determined on each substrate using a Krüss universal surface tester (model GH-100), the geometric mean approximation, and three standard liquids: water (H2O), diiodomethane (CH2I2), and ethylene
glycol [C2H4(OH)2]. The profiles and fine structure of
nanofibers were imaged and analyzed using a JEOL-2010 high-resolution transmission electron microscope and an internal charge-coupled device (CCD) camera. The crystal-line of DAAQ nanofibers was characterized by GIXRD, using Bruker D8 system and Cu K
a
radiation. The incident angle of the X-ray beam was fixed at 0.2°. The field emission characteristics of the DAAQ nanofibers were determined under a base pressure of 8 106torr. Indium tin oxide(ITO) plate glasses were used as anodes, positioned above the substrate surfaces at a distance of 80
l
m. The field emission instrument featured a plate-to-plate geometry; the cathode comprised DAAQ nanofibers deposited on Au and Ti substrates (area: 0.06 cm2). Current density–electric field (J–E) curves of the field emission devices were mea-sured using a Keithley 237 instrument (accuracy: 1013A). The emission currents of the DAAQ nanofibers were monitored as a function of the sweep bias.
3. Results and discussion
Fig. 1(a) displays the molecular structure of DAAQ. Assemblies of DAAQ molecules feature both intramolecular charge transfer[29]and intermolecular hydrogen bonding [23]. Intramolecular charge transfer occurs between the amino (NH2) and carbonyl (C@O) groups of the
anthraqui-none ring; intermolecular hydrogen bonding occurs between the oxygen atoms of the C@O group in one DAAQ molecule and the hydrogen atoms of the NH2 group in
another. For the nonpolar molecule coronene[10]to form 1D nanofibers, molecular stacking occurs mainly through
p
–p
and van der Waals interactions. In the stacking of DAAQ molecules, we suspected that both intra- and intermolecular hydrogen bonds would coexist. In addition top
–p
interactions and van der Waals contacts betweenstacking molecules, intermolecular hydrogen bond also plays an important role in 1D self-assembly.
The major advantage of vacuum sublimating DAAQ nanostructure films is the simplicity and accessibility of the process. No catalyst is used and no droplets of catalyst are found at the tips of the nanofibers; therefore, the growth mechanism of DAAQ nanostructure films is pre-sumed to be similar to that of the physical vapor transport method[27]. One strategy for achieving 1D structures, re-ported by Xia et al.[30], is the use of various templates with 1D morphologies to direct the assembly process. The growth mechanism normally involves three steps (Fig. 1(b)) (i) as the concentration of DAAQ increases, the molecular building blocks begin to aggregate into small nuclei; (ii) these nuclei become seeds for the anisotropic packing of DAAQ molecules; (iii) with a continuous supply of DAAQ, vapor molecules self-assemble to form 1D mor-phologies with vertically aligned nanostructures on the substrates. In this study, we used vacuum sublimation of DAAQ to construct 1D nanostructures on selected sub-strates. We investigated the growth mechanism of these DAAQ thin films to develop guidelines for the design of suitable structures for field emitters.Fig. 1(c–e) present cross-sectional SEM images of the early stage (6, 10, and 60 s of deposition at a rate of 0.5 Å/s, respectively) prod-ucts of the DAAQ nanofibers on a Si wafer. We observe a layer of vertically self-assembled and elongated nanoparti-cles, which acted as seeds for the anisotropic overgrowth of the nanofibers into columnar structures. Furthermore, varying the deposition time allowed us to adjust the length of the nanofibers on the substrates.
Fig. 2 presents SEM images (top and cross-sectional views) of the DAAQ nanostructure films formed on the Ag, Al, Au, Ni, Si, and Ti substrates. Clearly, the nature of the substrate influenced the morphology of its DAAQ thin film.Fig. 2(a and d) reveal entangled, in-plane, rod-like morphologies for the DAAQ films formed on the Ag and Ni substrates, respectively. In the top-view SEM images inFig. 2(b, c and e, f) we observe that most of the DAAQ nanofibers grew perpendicularly (slightly slanting) from
the substrates to form quasi-arrays. Repulsive forces among nuclei can result in a parallel-displaced conforma-tion, leading to a tilted columnar structure [10,31]. In Fig. 2(c, e, and f), the high-density packing morphologies of the DAAQ nanofibers on the Si and Ti substrates are sim-ilar to those on the Au surface. The densities of these nano-structures on all of the substrates ranged from 108to 109
cm2(Table 1). The DAAQ thin films deposited on the Ag
and Ni substrates both possessed tilted out-of-plane and in-plane nanorod morphologies, as revealed in their cross-sectional SEM images (insets to Fig. 2(a and d), respectively). The short, tilted nanorods on the Ag surface had a mean radius of 48 nm and a mean length of 226 nm; those on the Ni substrate, however, had larger dimensions (90 and 843 nm, respectively). The DAAQ nanostructures deposited on the Al substrate (inset to Fig. 2(b)) had the largest average radius (100 nm) and length (883 nm); we suspect these oversized nanostruc-tures comprised three or four nanofibers aggregated together in the form of nanocornerstones. In contrast, the DAAQ molecules packed on the Au, Si, and Ti substrates to present an out-of-plane morphology with uniform arrangement; these nanofibers had average lengths of 427, 794, and 587 nm, respectively (insets to Fig. 2(c, e, and f), respectively). The mean radii of these nanofibers, featuring smooth surfaces, ranged from 27 to 35 nm. The lengths of the nanofibers on these substrates could be ad-justed by varying the deposition time.Table 1summarizes the mean lengths (L), radii (rm), and nanostructural types of
the DAAQ assemblies deposited on the various substrates. The aspect ratios (ARs), defined as
AR ¼ L=rm ð1Þ
of the 1D DAAQ nanofibers deposited on the Au, Si, and Ti substrates were 12.6, 29.4, and 16.8, respectively. The high ARs of these DAAQ nanofibers suggest that they are poten-tially applicable nanostructures for use as organic field emitters.
The selective growth of vertical DAAQ nanowires gener-ally occurs through preferential deposition on geometrical Fig. 1. (a) Chemical structure of DAAQ. (b) Schematic representation of the growth of 1D DAAQ nanofibers through vacuum sublimation. SEM images (cross-sectional view) of DAAQ thin films grown on a Si substrate for (c) 6 s, (d) 10 s, and (e) 60 s at a deposition rate of 0.5 Å/s.
chemically modified surfaces; for example, substrates fea-turing sharp tips or edges, silica beads, or hydrophilic do-mains [27]. To realize the effects of CAs and to select appropriate metals for use as electrodes for potential appli-cability in organic field emission, we selected Ag, Au, Al, Ni, Si, and Ti as substrates and investigated the morphologies of the DAAQ nanostructure films formed on these surfaces. Table 2summarizes the CAs of the various substrates and their SEs, calculated using the Owens and Wendt method [32]. The water contact angles (WCAs) of the Ag, Al, and Ni substrates were 96.1°, 96.7°, and 100.9°, respectively, making them hydrophobic substrates. The majority of DAAQ molecules deposited on these substrates formed flat and entangled structures. In contrast, the DAAQ molecules deposited on the Au, Si, and Ti surfaces extended perpen-dicularly from the substrates, which had WCAs of 81.8°, 59.7°, and 78.8°, respectively (i.e., hydrophilic substrates). Therefore, it appears that the preferred growth of 1D verti-cal DAAQ nanofibers requires hydrophilic, rather than hydrophobic, surfaces. This result is similar to those
reported by Zhao et al.[27]who studied the effect using UV–Vis absorption spectroscopy. Next, we performed surface treatment of the Si substrate to confirm our Fig. 2. SEM images (top view) of DAAQ thin films deposited on (a) Ag, (b) Al, (c) Au, (d) Ni, (e) Si, and (f) Ti substrates. Insets are the cross-sectional SEM images.
Table 1
Mean lengths (L), mean radii (rm), ARs, nanostructure types, and distributions for the DAAQ thin films deposited on various substrates.
Substrate Mean length (nm) Mean radius (nm) AR Nanostructure type Distribution (1/cm2
) Ag 226 48 4.7 Nanorod 2.4 109 Al 883 100 8.8 Nanocornerstone 2.1 108 Au 427 34 12.6 Nanofiber 2.5 109 Ni 843 90 9.4 Nanorod 5.8 108 Si 794 27 29.4 Nanofiber 1.6 109 Ti 587 35 16.8 Nanofiber 1.6 109 Table 2
CAs and SEs of various substrates.
Substrate CA (°) SEb (mJ/m2 ) H2O CH2I2 C2H4(OH)2 Ag 96.1 52.7 66.5 34.8 Al 96.7 49.5 72.8 35.0 Au 81.8 19.9 56.2 45.1 Ni 100.9 39.5 69.9 46.4 Si 59.7 48.1 38.2 41.3 Ti 78.8 33.5 57.0 39.0 SiHa 88.5 58.5 69.9 27.6 a
Sample obtained after the bare Si substrate had been treated with HMDS vapor and annealed at 160 °C for 30 min.
b
Calculated using the Owens method from the geometric mean approximation (GMA) measurements of the static CAs of H2O, CH2I2, and
hypothesis.Fig. 3(a and d) present images of water drop-lets on the surfaces of bare Si and HMDS-treated Si (SiH), respectively; the WCAs were 59.7° and 88.5°, respectively. We suspect that the greater WCA for the SiH substrate re-sulted from a reduction in the density of interfacial trap-ping states and a lowering of the SE. Fig. 3(b and c) display SEM images of the DAAQ thin films deposited on the bare Si surface; as before, the structures extend away from the substrate. In contrast, some of the DAAQ nanofi-bers were deposited in-plane and entangled on the SiH substrate (Fig. 3(e and f)). This finding confirms that a low-WCA surface endows a preference for vertical deposi-tion of 1D DAAQ nanofibers.
Fig. 4(a and c) present TEM images of the DAAQ nanof-ibers formed on the Au and Ti substrates, respectively; these nanofibers were very smoothly faceted and uniform in length. High-resolution TEM images of single nanofibers revealed (Fig. 4(b and d)) indistinct lattice fringe spacing. We suspect that the DAAQ molecules were oriented in short-range order at the edges of the nanofibers, but entan-gled randomly within the nanofibers. The selected area electron diffraction (SAED) pattern in the inset to Fig. 4(a) features only amorphous rings, rather than dis-tinct diffractive spots, suggesting that an insignificant de-gree of crystallinity was responsible for the isotropic and randomly oriented molecules. Similarly, the electron dif-fraction pattern in the inset toFig. 4(c) also reveals the amorphous features of the DAAQ molecules deposited on the Ti substrates. In addition, these HRTEM images do not reveal the presence of any metal catalysts at the tips
of the fibers, confirming that the growth mechanism was different from that of carbon nanotubes (CNTs)[33]. In or-der to unor-derstand the lattice differences in our system, the XRD patterns of the DAAQ commercial powders and nano-fiber arrays were used to further study (Fig. 4(e)). Although the DAAQ nanofiber arrays feature lower peak intensity than the commercial powders, the relative percentages of (1 0 0), (1 0 2) and (1 1 2) lattice planes show some changes. Comparison of our results with the previous re-port of DAAQ nanowires [8], the DAAQ nanofiber arrays (formed by vacuum sublimation in low-temperature pro-cess) also prefer to grow in the orientation of (1 0 0) lattice plane.
To determine the differences in energy levels between the electrodes and the DAAQ materials for potential field emission applications, we used photoelectron spectros-copy in air (PESA; Fig. 5(a)) and UV–Vis spectroscopy (Fig. 5(b)) to measure the energy levels of DAAQ. When we bombarded our surface materials under a slowly increasing amount of UV light, photoelectrons were emit-ted from the surface (from a depth of several to hundred angstroms) at a certain energy level, due to the photoelec-tron effect. These emitted photoelecphotoelec-trons were then counted by a detector and open counter. Using this ap-proach, the valence band [highest occupied molecular orbi-tal (HOMO)] of DAAQ obtained through linear fitting from the PESA data was 5.60 eV relative to the vacuum level. The UV–Vis absorption spectra of the DAAQ films revealed an absorption band in the region 300–800 nm. From the on set wavelength at 614 nm, we derived an energy gap Fig. 3. (a, d) Photographs of water droplets on the (a) bare Si substrate (WCA: 59.7°) and (d) HMDS-treated SiH substrate (WCA: 88.5°). (b, c, e, f) SEM images of the DAAQ structures formed on the (b, c) bare Si and (e, f) SiH substrates; (b, e) cross-sectional views; (c, f) top views.
for the DAAQ of 2.02 eV. From this value, we estimated the energy of the conduction band [lowest unoccupied molec-ular orbital (LUMO)] to be 3.58 eV.Fig. 5(c and d) provide a schematic energy level diagram; the HOMO and LUMO states of DAAQ were 5.60 eV and 3.58 eV, respectively; the work functions of Au and Ti, for use as cathode sub-strates, were 5.1 and 4.33 eV, respectively. Therefore, elec-trons injected from the Au substrate into the LUMO state of the DAAQ layer must conquer a larger energy gap than those injected from the Ti substrate.
We examined the field emission properties of the DAAQ nanofibers using a parallel-plate configuration with the spacing of 80
l
m in a vacuum chamber at a pressure of 8 106torr. The anode ITO glass was connected to thesource monitor unit (SMU) of a Keithley 237 instrument; the cathode of Au and Ti substrates were grounded.Fig. 6 displays the field emission characteristics (J–E curve) of
of the emitter material. By plotting ln(J/E2) versus 1/E, the
slope of the line of best fit can be used to deduce the field enhancement factor (b):
b¼ B/
3=2
S ð3Þ
where S is the slope of the FN plot. According to FN theory, bis strongly dependent on the geometric structure of the field emitter. The linear FN plots in the insets toFig. 6(a and b) reveal that the J–E characteristics of the DAAQ nanofibers followed the FN field emission mechanism. The slopes of FN plots for DAAQ-Au and DAAQ-Ti were 103.5 and 55.67, respectively. Because the LUMO en-ergy level (/) of DAAQ was 3.58 eV, we estimated the val-ues of b of DAAQ-Au and DAAQ-Ti to be 447 and 831, respectively. To understand why these values of b differed so dramatically, we must consider the effect of the geom-etry of the nanofibers. Tarntair et al. [35] reported that the field enhancement factor b could be approximated by the length-to-radius ratio of the 1D nanostructure, ex-pressed as
b¼l
r ð4Þ
where r is the radius of curvature of the tip and l is the length of the nanofiber. This equation suggests that emit-ters having longer lengths and smaller radii, resulting in larger values of b, would have better field emission charac-teristics. Because our low-WCA surfaces (Au and Ti) favored DAAQ molecules self-assembling into standing nanofibers and because the value of b of DAAQ-Ti was greater than that of DAAQ-Au, we suspected that the high-er AR of the DAAQ nanofibhigh-ers deposited on Ti would be superior, to those formed on Au, for use in field emission devices. Cho et al.[36] reported the effective radius (re)
of an effective area at the tips of their Alq3nanostructural
thin film; the calculated radius (re= 2.07 nm) was smaller
than the measured radius (rm= 40 nm). The DAAQ
nanofi-bers of DAAQ-Au and DAAQ-Ti had effective radii of 0.96 and 0.7 nm, respectively (Table 3). From the differences be-tween the values of rm (34 nm for DAAQ-Au; 35 nm for
DAAQ-Ti) and re, we suspect that the electrons were
emit-ted from a localized area at the tip of a nanofiber, rather than from the entire measured area of the nanofiber. From the viewpoint of the energy levels,Fig. 5(c and d) reveal Fig. 4. (a) HRTEM image of a single DAAQ nanofiber formed on the Au
substrate; inset: corresponding SAED pattern. (b) Magnified image of the nanofiber in (a). (c) HRTEM image of a single DAAQ nanofiber formed on the Ti substrate; inset: corresponding SAED pattern. (d) HRTEM image of the smooth surface of a single nanofiber. (e) XRD patterns of DAAQ commercial powder (top) and nanofiber arrays (bottom). Inset: magnified XRD pattern of DAAQ nanofiber arrays.
that the work function (5.1 eV) of Au was much higher than that (4.33 eV) of Ti; therefore, the energy barrier (1.52 eV) at the Au-DAAQ interface was larger than that (0.75 eV) at the Ti-DAAQ contact. These energy barriers
limited the injection of electrons from the conducting sub-strates to organic nanofibers; consequently, a higher ap-plied field was required to reach the same level of emission current.
We performed stability tests of our DDAQ nanofibers under an applied field of 11 V/
l
m for 3000 s. The calcu-lated mean current densities were ca. 0.07 mA/cm2 forDAAQ-Au (Fig. 7(a)) and 0.19 mA/cm2 for DAAQ-Ti
(Fig. 7(b)), with perturbations of less than one order. In addition, we observed a slowly increasing DAAQ-Ti emis-sion current over time and two apparent increasing DAAQ-Au emission currents from 150 to 250 s and from Fig. 5. (a) PESA analysis of the DAAQ thin films. (b) UV–Vis absorption spectra of DAAQ thin films. (c, d) Schematic representations of the energy levels and electrons field emitting through the DAAQ nanofibers of the (c) Au-DAAQ and (d) Ti-DAAQ contacts.
Fig. 6. Field emission J–E curves of (a) DAAQ-Au and (b) DAAQ-Ti. Insets: Corresponding FN plots.
Table 3
Values of Eturn-on, slopes of FN plots, field emission enhancement factors (b),
and effective radii (re) for the DAAQ nanofibers.
Substrate Eturn-on(V/lm) Slope b re(nm)
Au 8.50 103.50 447 0.96
2440 to 2540 s. The latter presumably arose from a training effect [37,38]; that is, contaminants were expelled from the nanofibers after an adapted biasing on the nanofibers for a period of time. Thus, the organic semiconductor sus-tained a stable field emission current, without any decay, during the measurement period, demonstrating that DAAQ organic nanofibers have great potential for use in cold field electron-emitting devices.
4. Conclusions
By using DAAQ as a starting material, we prepared ver-tical organic nanofibers through low-temperature (42 °C) vacuum sublimation. The morphologies of the structures formed from the DAAQ molecules were controlled by the surface properties of the substrates, with vertically aligned 1D DAAQ structures growing preferentially on low-WCA surfaces, such as Au, Si, and Ti. The DAAQ nanofibers exhib-ited unique field emission characteristics and followed FN behavior. The maximum emission current densities of DAAQ-Au and DAAQ-Ti were 0.31 and 0.65 mA/cm2when
biased at 960 V (E = 12 V/
l
m), respectively. The field enhancement factors b of the DAAQ nanofibers on the Au and Ti substrates were 447 and 831, respectively. In field emission stability tests, the field emission current was sta-ble, without any decay, during the duration of themea-[3] Y.S. Zhao, H. Fu, A. Peng, Y. Ma, Q. Liao, J. Yao, Acc. Chem. Res. 43 (2010) 409.
[4] F. Yang, M. Shtein, S.R. Forrest, J. Appl. Phys. 98 (2005) 014906. [5] F. Yang, S.R. Forrest, Adv. Mater. 18 (2006) 2018.
[6] R.R. Lunt, J.B. Benziger, S.R. Forrest, Appl. Phys. Lett. 90 (2007) 181932.
[7] Y.S. Hsiao, W.T. Whang, S.C. Suen, J.Y. Shiu, C.P. Chen, Nano-technology 19 (2008) 415603.
[8] Y.S. Zhao, J. Wu, J. Huang, J. Am. Chem. Soc. 131 (2009) 3158. [9] J.J. Chiu, C.C. Kei, T.P. Perng, W.S. Wang, Adv. Mater. 15 (2003) 1361. [10] S.C. Suen, W.T. Whang, B.W. Wu, Y.F. Lai, Appl. Phys. Lett. 84 (2004)
3157.
[11] S.C. Suen, W.T. Whang, F.J. Hou, B.T. Dai, Org. Electron. 7 (2006) 428. [12] S.C. Suen, W.T. Whang, F.J. Hou, B.T. Dai, Org. Electron. 8 (2007) 505. [13] C.P. Cho, T.P. Perng, Org. Electron. 11 (2010) 115.
[14] H. Liu, Q. Zhao, Y. Li, Y. Liu, F. Lu, J. Zhuang, S. Wang, L. Jiang, D. Zhu, D. Yu, L. Chi, J. Am. Chem. Soc. 127 (2005) 1120.
[15] S. Cui, Y. Li, Y. Guo, H. Liu, Y. Song, J. Xu, J. Lv, M. Zhu, D. Zhu, Adv. Mater. 20 (2008) 309.
[16] J.J. Chiu, W.S. Wang, C.C. Kei, C.P. Cho, T.P. Perng, P.K. Wei, S.Y. Chiu, Appl. Phys. Lett. 83 (2003) 4607.
[17] Y.S. Zhao, J. Xu, A. Peng, H. Fu, Y. Ma, J. Yao, Adv. Mater. 20 (2008) 1661.
[18] Y.S. Zhao, H. Fu, F. Hu, A.D. Peng, J. Yao, Adv. Mater. 19 (2007) 3554. [19] W. Chen, Q. Peng, Y. Li, Adv. Mater. 20 (2008) 2747.
[20] H. Liu, Y. Li, S. Xiao, H. Gan, T. Jiu, H. Li, L. Jiang, D. Zhu, D. Yu, B. Xiang, Y. Chen, J. Am. Chem. Soc. 125 (2003) 10794.
[21] S. Cui, H. Liu, L. Gan, Y. Li, D. Zhu, Adv. Mater. 20 (2008) 2918. [22] H. Gan, H. Liu, Y. Li, Q. Zhao, Y. Li, S. Wang, T. Jiu, N. Wang, X. He, D.
Yu, D. Zhu, J. Am. Chem. Soc. 127 (2005) 12452.
[23] H. Liu, Y. Li, S. Xiao, H. Li, L. Jiang, D. Zhu, B. Xiang, Y. Chen, D. Yu, J. Phys. Chem. B 108 (2004) 7744.
[24] J.J. Chiu, W.S. Wang, C.C. Kei, T.P. Perng, Appl. Phys. Lett. 83 (2003) 347.
[25] Y.S. Zhao, C.A. Di, W. Yang, G. Yu, Y. Liu, J. Yao, Adv. Funct. Mater. 16 (2006) 1985.
[26] Y.S. Zhao, D. Xiao, W. Yang, A. Perng, J. Yao, Chem. Mater. 18 (2006) 2302.
[27] Y.S. Zhao, P. Zhan, J. Kim, K. Sun, J. Huang, ACS Nano 4 (2010) 1630. [28] H. Derouiche, V. Djara, Sol. Energy Mater. Sol. Cells 83 (2004) 125. [29] G. Suganthi, C. Meenakshi, V. Ramakrishnan, J. Fluoresc. 20 (2010)
95.
[30] Y.N. Xia, P.D. Yang, Y.G. Sun, Y.Y. Wu, B. Mayers, B. Gates, Y.D. Yin, F. Kim, H.Q. Yan, Adv. Mater. 15 (2003) 353.
[31] J.D. Wright, Molecular Crystals, second ed., Cambridge University Press, 1995. p. 22.
[32] D.K. Owens, R.C. Wendt, J. Appl. Polym. Sci. 13 (1969) 1741. [33] V. Vinciguerra, F. Buonocore, G. Panzera, L. Occhipinti,
Nano-technology 14 (2003) 655.
[34] R. Fowler, L.W. Nordheim, Proc. R. Soc. London 119 (1928) 173. [35] F.G. Tarntair, C.Y. Wen, L.C. Chen, J.J. Wu, K.H. Chen, P.F. Kuo, S.W.
Chang, Y.F. Chen, W.K. Hong, H.C. Cheng, Appl. Phys. Lett. 76 (2000) 2630.
[36] C.P. Cho, T.P. Perng, Nanotechnology 18 (2007) 125202.
[37] J.M. Bonard, N. Weiss, H. kind, T. Stöckli, L. Forró, K. Kern, A. Châtelain, Adv. Mater. 13 (2001) 184.
[38] S.K. Patra, G.M. Rao, J. Appl. Phys. 100 (2006) 024319. Fig. 7. Emission current stability plots recorded at a constant voltage for
