Synthesis of Dinaphtho[2,3-d:2',3'-d']anthra[1,2-b:5,6-b']dithiophene (DNADT) Derivatives: Effect of Alkyl Chains on Transistor Properties.

To investigate organic field-effect transistor (OFET) properties, a new thienoacene-type molecule, 4,14-dihexyldinaphtho[2,3-d:2’,3’-d’]anthra[1,2-b:5,6-b’]dithiophene (C6-DNADT), consisting of π-conjugated nine aromatic rings and two hexyl chains along the longitudinal molecular axis has been successfully synthesized by sequential reactions, including Negishi coupling, epoxidation, and cycloaromatization. The fabricated OFET using thin films of C6-DNADT exhibited p-channel FET properties with field-effect mobilities (µ) of up to 2.6 × 10−2 cm2 V−1 s−1, which is ca. three times lower than that of the parent DNADT molecule (8.5 × 10−2 cm2 V−1 s−1). Although this result implies that the installation of relatively short alkyl chains into the DNADT core is not suitable for transistor application, the origins for the FET performance obtained in this work is fully discussed, based on theoretical calculations and solid-state structure of C6-DNADT by grazing incidence wide-angle X-ray scattering (GIWAXS) and atomic force microscopy (AFM) analyses. The results obtained in this study disclose the effect of alkyl chains introduced onto the molecule on transistor characteristics.


Introduction
Organic field-effect transistors (OFETs) using thin films and single crystals of π-extended thienoacene and thienophenacene molecules have attracted much attention as the key elements for realizing future ubiquitous electronics because they are known to display excellent hole transport properties [1][2][3][4][5]. In terms of single crystal OFETs, rubrene has provided excellent FET characteristics with carrier mobility (µ) as high as 40 cm 2 V -1 s -1 [6], but the rubrene molecule has scarcely been available for thin-film FETs, i.e., very few rubrene thin-film FETs have been operated [7]. On the other hand, the highest µ of thin-film OFETs reported among the state-of-art materials is currently 43 cm 2 V −1 s −1 for 2,7-dioctyl [1]benzothieno [3,2-b] [1]benzothiophene (C8-BTBT) [8]. It has been reported that the µ value increases as the number of aromatic rings increases, i.e., more extension of the π-system is suitable for better transistor properties owing to their greater intermolecular overlaps in the molecular network [9]. The extension of the π-system derives relatively small reorganization energies (λ) that can lead to the high-performance OFET devices [10,11]. Furthermore, highest occupied molecular orbital

Theoretical Calculations for Molecular Design
The positions to introduce alkyl chains in the molecular framework is an important issue for improvement of OFET properties of the materials. In some cases, the installation of alkyl substituents along the longitudinal molecular axis dramatically increased carrier mobility [33,34]. Thereby, we designed the molecule having two hexyl groups in 4,14-positions, expecting a good balance between crystallinity and solubility.

Synthesis of C6-DNADT
First, we considered how to introduce alkyl chains into the DNADT framework. Starting from commercially available 2-bromo-6-methoxynaphthalene (1), one of coupling partners, 6-hexylnaphtho[2,3-b]thiophene (7) was synthesized according to the synthetic method for anthra [2,3-b]thiophene (Scheme 1) [41]. The palladium-catalyzed Kumada-Tamao-Corriu coupling of 1 with hexyl Grignard reagent afforded 2. Successively, regioselective bromination at the 3-position of naphthalene via lithiation gave 3-bromo-6-hexyl-2-methoxynaphthalene (3), which was demethylated with boron tribromide to afford 3-bromo-6-hexylnaphthalen-2-ol (4). Then, a hydroxy group of 4 was converted into the corresponding triflate 5, which was then utilized in Sonogashira-Hagihara coupling with trimethylsilylethyne to afford the precursor 6. The excellent chemoselective alkynylation for the Sonogashira-Hagihara coupling at a triflate over a bromine moiety of 6 was achieved using N,N-dimethylformamide (DMF) as the solvent. Finally, thienoannulation reaction was accomplished with sodium sulfide nonahydrate (Na 2 S·9H 2 O) to give the target product 7 in 85% isolated yield. It is noteworthy that since the common starting compound 1 can be commercially available, the synthesis of other types of alkylated derivatives of compound 7 could be possible. The synthetic route of C6-DNADT from compound 7 is illustrated in Scheme 2. This 3-step synthetic method has been established by us [18,19]. First, palladium-catalyzed Negishi coupling of organozinc reagent, prepared in situ by lithiation of 7 by treatment with n-BuLi followed by zincation, with 8 afforded dialdehyde 9 in 74% yield. Subsequently, epoxidation of 9 and a sequential indiumcatalyzed intramolecular cycloaromatization of 10 gave C6-DNADT as an orange solid, albeit in 19% yield. Unexpectedly, even though two hexyl chains were introduced onto the DNADT core, solubility of C6-DNADT was found to be very poor, which is unable to measure NMR in solution. To prepare a pure sample suitable for further evaluation of the physicochemical and FET properties, the synthesized C6-DNADT was further purified twice by a gradient vacuum sublimation. The synthetic route of C6-DNADT from compound 7 is illustrated in Scheme 2. This 3-step synthetic method has been established by us [18,19]. First, palladium-catalyzed Negishi coupling 4 of 16 of organozinc reagent, prepared in situ by lithiation of 7 by treatment with n-BuLi followed by zincation, with 8 afforded dialdehyde 9 in 74% yield. Subsequently, epoxidation of 9 and a sequential indium-catalyzed intramolecular cycloaromatization of 10 gave C6-DNADT as an orange solid, albeit in 19% yield. Unexpectedly, even though two hexyl chains were introduced onto the DNADT core, solubility of C6-DNADT was found to be very poor, which is unable to measure NMR in solution.
To prepare a pure sample suitable for further evaluation of the physicochemical and FET properties, the synthesized C6-DNADT was further purified twice by a gradient vacuum sublimation.
The synthetic route of C6-DNADT from compound 7 is illustrated in Scheme 2. This 3-step synthetic method has been established by us [18,19]. First, palladium-catalyzed Negishi coupling of organozinc reagent, prepared in situ by lithiation of 7 by treatment with n-BuLi followed by zincation, with 8 afforded dialdehyde 9 in 74% yield. Subsequently, epoxidation of 9 and a sequential indiumcatalyzed intramolecular cycloaromatization of 10 gave C6-DNADT as an orange solid, albeit in 19% yield. Unexpectedly, even though two hexyl chains were introduced onto the DNADT core, solubility of C6-DNADT was found to be very poor, which is unable to measure NMR in solution. To prepare a pure sample suitable for further evaluation of the physicochemical and FET properties, the synthesized C6-DNADT was further purified twice by a gradient vacuum sublimation. Scheme 2. Synthetic route of C6-DNADT.

UV-Vis Absorption Spectrum and Cyclic Voltammogram
To evaluate physicochemical properties of C6-DNADT, UV-vis absorption spectrum was measured for its vapor-deposited thin film ( Figure 2a). The maximum absorption was observed at Scheme 2. Synthetic route of C6-DNADT.

UV-Vis Absorption Spectrum and Cyclic Voltammogram
To evaluate physicochemical properties of C6-DNADT, UV-vis absorption spectrum was measured for its vapor-deposited thin film ( Figure 2a). The maximum absorption was observed at 467 nm and the optical energy gap estimated from an absorption edge was 2.48 eV, which is similar to that of DNADT (2.51 eV) [32], indicating that the introduction of alkyl groups did not affect its optical energy gap in thin film. However, the two obvious peaks appeared at 367 and 388 nm in thin film of C6-DNADT and the shape of UV-vis absorption spectrum in C6-DNADT is quite different from that of the parent DNADT, implying the formation of different structure in the solid state.
Cyclic voltammogram of C6-DNADT in dichloromethane solution was measured to estimate its frontier energy level (Figure 2b). C6-DNADT showed a very weak oxidation wave with oxidation onset (E ox onset ) of +1.02 V (vs. Ag/Ag + ) due to its poor solubility. The estimated HOMO energy level of C6-DNADT was −5.29 eV, which is similar to that of the result of DFT calculation ( Figure 1). As expected, this HOMO energy level is close to the work function of gold (5.1 eV) [42], which could be expected to lead to the smooth hole injection in OFETs [43]. In addition, this HOMO value is sufficiently deep to achieve the high air-stability. Thus, C6-DNADT-based OFET may show the good transistor property under ambient conditions. frontier energy level (Figure 2b). C6-DNADT showed a very weak oxidation wave with oxidation onset (E ox onset) of +1.02 V (vs. Ag/Ag + ) due to its poor solubility. The estimated HOMO energy level of C6-DNADT was −5.29 eV, which is similar to that of the result of DFT calculation ( Figure 1). As expected, this HOMO energy level is close to the work function of gold (5.1 eV) [42], which could be expected to lead to the smooth hole injection in OFETs [43]. In addition, this HOMO value is sufficiently deep to achieve the high air-stability. Thus, C6-DNADT-based OFET may show the good transistor property under ambient conditions.

Thermal Stability
In order to evaluate a thermal stability of C6-DNADT, thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) were measured ( Figure 3). The temperature of 5% weight loss (Td 5 ) was 462 °C. Furthermore, no transition peaks were observed of up to 270 °C in the DSC curve, despite having the two flexible alkyl chains. These results indicate that C6-DNADT has high thermal stability due to a high rigidity and a large π-extended electron system of the DNADT core, which is beneficial for practical application in OFETs.

Thermal Stability
In order to evaluate a thermal stability of C6-DNADT, thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) were measured ( Figure 3). The temperature of 5% weight loss (T d 5 ) was 462 • C. Furthermore, no transition peaks were observed of up to 270 • C in the DSC curve, despite having the two flexible alkyl chains. These results indicate that C6-DNADT has high thermal stability due to a high rigidity and a large π-extended electron system of the DNADT core, which is beneficial for practical application in OFETs.
onset (E ox onset) of +1.02 V (vs. Ag/Ag + ) due to its poor solubility. The estimated HOMO energy level of C6-DNADT was −5.29 eV, which is similar to that of the result of DFT calculation ( Figure 1). As expected, this HOMO energy level is close to the work function of gold (5.1 eV) [42], which could be expected to lead to the smooth hole injection in OFETs [43]. In addition, this HOMO value is sufficiently deep to achieve the high air-stability. Thus, C6-DNADT-based OFET may show the good transistor property under ambient conditions.

Thermal Stability
In order to evaluate a thermal stability of C6-DNADT, thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) were measured ( Figure 3). The temperature of 5% weight loss (Td 5 ) was 462 °C. Furthermore, no transition peaks were observed of up to 270 °C in the DSC curve, despite having the two flexible alkyl chains. These results indicate that C6-DNADT has high thermal stability due to a high rigidity and a large π-extended electron system of the DNADT core, which is beneficial for practical application in OFETs.

OFET Properties
To investigate FET properties of C6-DNADT, typical bottom-gate top-contact devices based on its thin films have been fabricated using SiO 2 gate dielectrics with the channel length (L) of 87 µm and width (W) of ca. 1980 µm. The surface of the n + -Si/SiO 2 substrate was treated with n-octyltrichlorosilane (OTS) or n-octadecyltrichlorosilane (ODTS) as the self-assembled monolayer (SAM). The active layers were deposited on the SAM treated substrate by vapor deposition at a rate of 0.5 Å s −1 under reduced pressure (5 × 10 −5 Pa). The substrate temperature was room temperature. Thermal annealing of the active layer was performed at 50, 100, and 150 • C for 30 min under an inert atmosphere. The measurements were conducted under ambient conditions in the dark. The transfer and output curves are shown in Figure 4 and the obtained FET properties, including anneal temperatures, hole mobility (µ), threshold voltage (V th ), and on-off ratio (I on /I off ), are summarized in Table 1. Typical p-channel FET properties were observed in the transfer and output curves of all devices. As we expected, C6-DNADT-based OFETs exhibited smaller threshold voltage (V th ) than that of DNADT-based devices due to its high-lying HOMO energy level [32]. In the case of ODTS-treated OFETs, the OFET based on as-deposited thin film showed hole mobility of up to 1.4 x 10 −2 cm 2 V −1 s −1 . Further increasing the temperature of a thermal annealing process did not enhance the hole mobility. With OTS as the SAM, all devices exhibited higher hole mobilities than that of the corresponding ODTS-treated OFETs. The FET device without a thermal annealing exhibited hole mobility of 1.9 × 10 −2 cm 2 V −1 s −1 , whereas a thermal annealing at 100 • C improved the maximum mobility of up to 2.6 × 10 −2 cm 2 V −1 s −1 , which is highest hole mobility in this system. However, the obtained hole mobility of C6-DNADT was still lower than that of DNADT (µ = 8.5 × 10 −2 cm 2 V −1 s −1 ) [32]. The reason why the installation of alkyl chains providing such worse FET characteristics is discussed based on the topological and electronic features of its thin film. thin films have been fabricated using SiO2 gate dielectrics with the channel length (L) of 87 μm and width (W) of ca. 1980 μm. The surface of the n + -Si/SiO2 substrate was treated with noctyltrichlorosilane (OTS) or n-octadecyltrichlorosilane (ODTS) as the self-assembled monolayer (SAM). The active layers were deposited on the SAM treated substrate by vapor deposition at a rate of 0.5 Å s −1 under reduced pressure (5 × 10 −5 Pa). The substrate temperature was room temperature. Thermal annealing of the active layer was performed at 50, 100, and 150 °C for 30 min under an inert atmosphere. The measurements were conducted under ambient conditions in the dark. The transfer and output curves are shown in Figure 4 and the obtained FET properties, including anneal temperatures, hole mobility (µ), threshold voltage (Vth), and on-off ratio (Ion/Ioff), are summarized in Table 1. Typical p-channel FET properties were observed in the transfer and output curves of all devices. As we expected, C6-DNADT-based OFETs exhibited smaller threshold voltage (Vth) than that of DNADT-based devices due to its high-lying HOMO energy level [32]. In the case of ODTStreated OFETs, the OFET based on as-deposited thin film showed hole mobility of up to 1.4 x 10 −2 cm 2 V −1 s −1 . Further increasing the temperature of a thermal annealing process did not enhance the hole mobility. With OTS as the SAM, all devices exhibited higher hole mobilities than that of the corresponding ODTS-treated OFETs. The FET device without a thermal annealing exhibited hole mobility of 1.9 × 10 −2 cm 2 V −1 s −1 , whereas a thermal annealing at 100 °C improved the maximum mobility of up to 2.6 × 10 −2 cm 2 V −1 s −1 , which is highest hole mobility in this system. However, the obtained hole mobility of C6-DNADT was still lower than that of DNADT (µ = 8.5 × 10 −2 cm 2 V −1 s −1 ) [32]. The reason why the installation of alkyl chains providing such worse FET characteristics is discussed based on the topological and electronic features of its thin film.

AFM Images
To investigate the thin-film structure of C6-DNADT, we performed an atomic force microscope (AFM) analysis of the vapor-deposited thin film. AFM images of thin films as-deposited and annealed at 100 • C are shown in Figure 5. Obviously, the two surface morphologies were quite different. In as-deposited thin film, no distinct domain, many dark spots, and the smooth surface with root-mean-square (RMS) of 0.68 nm was observed. In contrast, thin film treated by a thermal annealing at 100 • C formed well-defined domain and has slightly higher roughness of RMS = 0.77 nm, resulting in the highest hole mobility due to its appropriate morphology. However, even thin film of C6-DNADT treated by thermal annealing at 100 • C has drastically smaller domain size (ca. 80 nm) than that of parent DNADT (ca. 300-500 nm) [32]. Moreover, many grain boundaries were also found, leading to its poor interlayer connectivity, which may inhibit an effective carrier transport [44]. Thus, C6-DNADT-based OFET exhibited poor hole mobility than that of the DNADT-based devices.

AFM Images
To investigate the thin-film structure of C6-DNADT, we performed an atomic force microscope (AFM) analysis of the vapor-deposited thin film. AFM images of thin films as-deposited and annealed at 100 °C are shown in Figure 5. Obviously, the two surface morphologies were quite different. In asdeposited thin film, no distinct domain, many dark spots, and the smooth surface with root-meansquare (RMS) of 0.68 nm was observed. In contrast, thin film treated by a thermal annealing at 100 °C formed well-defined domain and has slightly higher roughness of RMS = 0.77 nm, resulting in the highest hole mobility due to its appropriate morphology. However, even thin film of C6-DNADT treated by thermal annealing at 100 °C has drastically smaller domain size (ca. 80 nm) than that of parent DNADT (ca. 300-500 nm) [32]. Moreover, many grain boundaries were also found, leading to its poor interlayer connectivity, which may inhibit an effective carrier transport [44]. Thus, C6-DNADT-based OFET exhibited poor hole mobility than that of the DNADT-based devices.

GIWAXS Images
To further understand the difference of OFET performances between the parent DNADT and C6-DNADT, we investigated the grazing incidence wide-angle X-ray scattering (GIWAXS) analysis in thin film. Two-dimensional (2D) GIWAXS image and one-dimensional (1D) profiles extracted from GIWAXS image are shown in Figure 6. In the qz direction, two series of (00l) diffractions were observed. The calculated interlayer distance (d001) was 30.7 Å, which is smaller than a molecular length estimated from a theoretical calculation (35.9 Å, Figure 1). Therefore, C6-DNADT is tilted at an angle of 31° with respect to the substrate. Furthermore, in the qxy axis direction, three characteristic reflections were observed at 1.25, 1.57, and 1.84 Å −1 , implying that C6-DNADT forms a herringbone structure as similar to that of the parent DNADT [32]. Although a weak (001) diffraction was also observed at the qxy direction, indicating a contamination of an unsuitable face-on crystallite, the intensity of this face-on crystallite was much weaker than that of the parent DNADT. This result suggests that the introduction of two alkyl side chains along the longitudinal molecular axis can suppress the construction of the unfavorable face-on crystallite. However, the diffraction intensity of C6-DNADT was obviously weaker than that of DNADT. Such low crystalline nature is consistent with the result of AFM images, leading to the lower hole transporting ability than that of DNADT. One possible reason for such low crystalline nature of C6-DNADT might be attributed to the length of alkyl side chains. In general, the introduction of alkyl side chains along the longitudinal molecular axis can enhance the construction of densely packing structure owing to hydrophobic interaction (i.e.,

GIWAXS Images
To further understand the difference of OFET performances between the parent DNADT and C6-DNADT, we investigated the grazing incidence wide-angle X-ray scattering (GIWAXS) analysis in thin film. Two-dimensional (2D) GIWAXS image and one-dimensional (1D) profiles extracted from GIWAXS image are shown in Figure 6. In the q z direction, two series of (00l) diffractions were observed. The calculated interlayer distance (d 001 ) was 30.7 Å, which is smaller than a molecular length estimated from a theoretical calculation (35.9 Å, Figure 1). Therefore, C6-DNADT is tilted at an angle of 31 • with respect to the substrate. Furthermore, in the q xy axis direction, three characteristic reflections were observed at 1.25, 1.57, and 1.84 Å −1 , implying that C6-DNADT forms a herringbone structure as similar to that of the parent DNADT [32]. Although a weak (001) diffraction was also observed at the q xy direction, indicating a contamination of an unsuitable face-on crystallite, the intensity of this face-on crystallite was much weaker than that of the parent DNADT. This result suggests that the introduction of two alkyl side chains along the longitudinal molecular axis can suppress the construction of the unfavorable face-on crystallite. However, the diffraction intensity of C6-DNADT was obviously weaker than that of DNADT. Such low crystalline nature is consistent with the result of AFM images, leading to the lower hole transporting ability than that of DNADT. One possible reason for such low crystalline nature of C6-DNADT might be attributed to the length of alkyl side chains. In general, the introduction of alkyl side chains along the longitudinal molecular axis can enhance the construction of densely packing structure owing to hydrophobic interaction (i.e., a fastener effect) [14]. However, the length of two hexyl side chains in C6-DNADT is much shorter than that of the central DNADT core. In this case, a fastener effect is not sufficient, because a hydrophobic interaction would be small [45]. Thus, the introduction of alkyl side chains with an appropriate length is highly important to develop the high-performance organic semiconductors for FET applications. a fastener effect) [14]. However, the length of two hexyl side chains in C6-DNADT is much shorter than that of the central DNADT core. In this case, a fastener effect is not sufficient, because a hydrophobic interaction would be small [45]. Thus, the introduction of alkyl side chains with an appropriate length is highly important to develop the high-performance organic semiconductors for FET applications.

Instrumentation
All the reactions were carried out under an Ar atmosphere using standard Schlenk techniques. Glassware was dried in an oven (130 °C) and heated under reduced pressure prior to use. For thin layer chromatography (TLC) analyses throughout this work, Merck pre-coated TLC plates (silica gel 60 GF254, 0.25 mm) were used. Silica gel column chromatography was carried out using silica gel 60 N (spherical, neutral, 40−100 μm) from Kanto Chemicals Co., Ltd. The 1 H and 13 C{ 1 H} NMR spectra were recorded on a Varian Mercury-300 (300 MHz), Varian 400-MR (400 MHz), and Varian INOVA-600 (600 MHz) spectrometer (supplementary material). High-resolution mass spectrometry (HRMS) was carried out on a JEOL JMS-700 MStation (double-focusing mass spectrometer). Elemental analyses were carried out with a PerkinElmer 2400 CHN elemental analyzer at Okayama University. Infrared spectra were recorded on a Shimadzu IRPrestige-21 spectrophotometer and reported in wave numbers (cm −1 ). UV-vis absorption spectra were measured using a Shimadzu UV-2450 UV-vis spectrometer. Differential scanning calorimetry (DSC) measurement was performed at the rate of 10 °C/min from 25 °C to 270 °C for both heating and cooling steps under a nitrogen flow using a SSC5200H (Seiko Instruments). Thermogravimetric analysis (TGA) was carried out at a heating rate of 10 °C/min from 25 °C to 600 °C under a nitrogen flow rate of 20 mL/min using a TG4000 (Perkin Elmer). Dynamic force-mode atomic force microscopy (AFM) was carried out using an SPA 400-DFM (SII Nano Technologies). Grazing incidence wide-angle X-ray scattering (GIWAXS) analysis was performed at the SPring-8 on beamline BL46XU. The sample was irradiated at a fixed angle on the order of 0.12° through a Huber diffractometer with an X-ray energy of 12.39 keV (λ = 1 Å), and the GIWAXS patterns were recorded with a 2D image detector (Pilatus 300K). The thin films of C6-DNADT were fabricated by vapor deposition on OTS or ODTS-treated n + -Si/SiO2 substrate. The FET properties were measured at room temperature in air on a Keithley 6430 subfemtoampere remote source meter combined with a Keithley 2400 measure-source unit. Geometry optimizations and normal-mode calculations were performed at the B3LYP/6-311G(d) level using the Gaussian 09, Revision D. 01, program package.

Instrumentation
All the reactions were carried out under an Ar atmosphere using standard Schlenk techniques. Glassware was dried in an oven (130 • C) and heated under reduced pressure prior to use. For thin layer chromatography (TLC) analyses throughout this work, Merck pre-coated TLC plates (silica gel 60 GF 254 , 0.25 mm) were used. Silica gel column chromatography was carried out using silica gel 60 N (spherical, neutral, 40−100 µm) from Kanto Chemicals Co., Ltd. The 1 H and 13 C{ 1 H} NMR spectra were recorded on a Varian Mercury-300 (300 MHz), Varian 400-MR (400 MHz), and Varian INOVA-600 (600 MHz) spectrometer (Supplementary Materials). High-resolution mass spectrometry (HRMS) was carried out on a JEOL JMS-700 MStation (double-focusing mass spectrometer). Elemental analyses were carried out with a PerkinElmer 2400 CHN elemental analyzer at Okayama University. Infrared spectra were recorded on a Shimadzu IRPrestige-21 spectrophotometer and reported in wave numbers (cm −1 ). UV-vis absorption spectra were measured using a Shimadzu UV-2450 UV-vis spectrometer. Differential scanning calorimetry (DSC) measurement was performed at the rate of 10 • C/min from 25 • C to 270 • C for both heating and cooling steps under a nitrogen flow using a SSC5200H (Seiko Instruments). Thermogravimetric analysis (TGA) was carried out at a heating rate of 10 • C/min from 25 • C to 600 • C under a nitrogen flow rate of 20 mL/min using a TG4000 (Perkin Elmer). Dynamic force-mode atomic force microscopy (AFM) was carried out using an SPA 400-DFM (SII Nano Technologies). Grazing incidence wide-angle X-ray scattering (GIWAXS) analysis was performed at the SPring-8 on beamline BL46XU. The sample was irradiated at a fixed angle on the order of 0.12 • through a Huber diffractometer with an X-ray energy of 12.39 keV (λ = 1 Å), and the GIWAXS patterns were recorded with a 2D image detector (Pilatus 300K). The thin films of C6-DNADT were fabricated by vapor deposition on OTS or ODTS-treated n + -Si/SiO 2 substrate. The FET properties were measured at room temperature in air on a Keithley 6430 subfemtoampere remote source meter combined with a Keithley 2400 measure-source unit. Geometry optimizations and normal-mode calculations were performed at the B3LYP/6-311G(d) level using the Gaussian 09, Revision D. 01, program package.

Fabrication of Vapor-Deposited OFET Devices
Typical bottom-gate top-contact OFET devices were fabricated as follows: All processes were performed under a nitrogen atmosphere except for substrate cleaning. A heavily doped n-Si wafer with a 200 nm-thick thermally grown SiO 2 (C i = 17.3 nF cm −2 ) as the dielectric layer was used as the substrate. The n + -Si/SiO 2 substrates were carefully cleaned by ultrasonication with acetone and isopropanol for 10 min, respectively. After drying, the substrates were irradiated with UV−O 3 for 20 min and then treated with a solution of 0.1 M n-octyltrichlorosilane (OTS) or n-octadecyltrichlorosilane (ODTS) in anhydrous toluene to form the self-assembled monolayer (SAM). The active layers were deposited on the SAM treated substrate by vapor deposition at a rate of 0.5 Å s −1 under reduced pressure (5 × 10 −5 Pa). The substrate temperature was room temperature. Thermal annealing was performed at 50, 100, and 150 • C for 30 min on the hotplate in the glovebox. After treatment, gold electrodes (67 nm-thick) were deposited through a shadow mask on top of the active layer under reduced pressure (5 × 10 −5 Pa). The current−voltage characteristics of the OFETs were measured at room temperature in air on a Keithley 6430 sub-femto ampere remote source meter combined with a Keithley 2400 measure-source unit. Field effect mobilities were calculated in the saturation regime of I D using the following equation (1), where C i is the capacitance of the SiO 2 insulator; I D is the source−drain current; and V D , V G , and V th are the source−drain, gate, and threshold voltages, respectively. The current on/off ratio (I on /I off ) was determined from a minimum I D at around V G = 0−10 V and maximum I D at V G = −60 V. 2 (1)

Conclusions
In summary, we have designed the molecule by DFT calculation and found that HOMO and NHOMO levels can be controlled by the installation of alkyl chains onto the framework. Hence, the DNADT derivative, C6-DNADT, bearing two hexyl chains along the longitudinal molecular axis have successfully been synthesized. C6-DNADT has similar optical energy gap of 2.48 eV to that of DNADT and sufficiently deep HOMO energy level of −5.29 eV, which is a close value to the work function of gold, implying the high air-stability and the smooth hole injection in OFETs. Furthermore, C6-DNADT also has the high thermal stability even in the existence of flexible alkyl chains. From AFM and GIWAXS analyses, although the introduction of two hexyl groups along the molecular long-axis direction can improve the molecular orientation, the crystallinity of C6-DNADT in thin film was much poorer than that of DNADT. This may be due to the shorter length of alkyl side chains than that of the central DNADT framework, which may suppress a fastener effect. As the result, the fabricated devices based on the C6-DNADT polycrystalline film exhibited the maximum hole mobility of up to 2.6 × 10 −2 cm 2 V −1 s −1 , which was much lower than that of our previously reported DNADT. From these results, the introduction of optimal alkyl chains is highly important to develop the high-performance materials for FETs. Currently, the synthesis and characterization of DNADT derivatives by installing longer alkyl groups are elucidated for improving OFET properties, expecting a more suitable packing structure in the solid state due to tunable intermolecular hydrophobic interactions. This study provides a potential avenue to be explored in the design of organic molecules suitable for FET materials.