materials Letter
Effects of the F4TCNQ-Doped Pentacene Interlayers on Performance Improvement of Top-Contact Pentacene-Based Organic Thin-Film Transistors Ching-Lin Fan 1,2, *, Wei-Chun Lin 2 , Hsiang-Sheng Chang 1 , Yu-Zuo Lin 1 and Bohr-Ran Huang 2 Received: 10 November 2015; Accepted: 6 January 2016; Published: 13 January 2016 Academic Editors: Federico Bella and Claudio Gerbaldi 1
2
*
Department of Electronic and Computer Engineering, National Taiwan University of Science and Technology, No. 43, Section 4, Keelung Road, Da’an District, Taipei City 106, Taiwan;
[email protected] (H.-S.C.);
[email protected] (Y.-Z.L.) Graduate Institute of Electro-Optical Engineering, National Taiwan University of Science and Technology, No. 43, Section 4, Keelung Road, Da’an District, Taipei City 106, Taiwan;
[email protected] (W.-C.L.);
[email protected] (B.-R.H.) Correspondence:
[email protected]; Tel.: +886-2-2737-6374
Abstract: In this paper, the top-contact (TC) pentacene-based organic thin-film transistor (OTFT) with a tetrafluorotetracyanoquinodimethane (F4 TCNQ)-doped pentacene interlayer between the source/drain electrodes and the pentacene channel layer were fabricated using the co-evaporation method. Compared with a pentacene-based OTFT without an interlayer, OTFTs with an F4 TCNQ:pentacene ratio of 1:1 showed considerably improved electrical characteristics. In addition, the dependence of the OTFT performance on the thickness of the F4 TCNQ-doped pentacene interlayer is weaker than that on a Teflon interlayer. Therefore, a molecular doping-type F4 TCNQ-doped pentacene interlayer is a suitable carrier injection layer that can improve the TC-OTFT performance and facilitate obtaining a stable process window. Keywords: pentacene; organic thin-film transistors (OTFTs); F4 TCNQ; Teflon; carrier injection layer
1. Introduction Over the past few decades, pentacene-based organic thin-film transistors (OTFTs) have attracted considerable research interest as promising candidates for smart cards, sensors, flexible displays, and radio frequency identification circuitry, because of their unique characteristics, which include their light weight, low-temperature processing, and high mechanical flexibility [1–4]. Au is usually used as the source/drain (S/D) electrode material in pentacene-based OTFTs since its work function matches the highest occupied molecular orbital level of pentacene; this matching benefits hole injection between the S/D electrodes and the pentacene channel [5]. In a typical top-contact (TC) pentacene-based OTFT structure, the S/D metal electrodes are formed by depositing Au onto the pentacene channel layer. However, when Au is directly deposited onto the pentacene channel layer, it diffuses into the upper layer of pentacene to form a mixture layer of metal and pentacene. The formation of the mixture layer is associated with the formation of hole injection barriers between Au and the pentacene channel layer [6–8]. The formation of the hole injection barriers leads to a low carrier injection efficiency and a high contact resistance (RC ), which degrade the performance of TC pentacene-based OTFTs. In organic light-emitting diodes, the carrier injection efficiency can be enhanced by introducing a carrier injection layer between the organic material and the metal electrode. Introducing a carrier injection layer reduces the barrier height between the organic layer and the metal electrode, increasing the luminance efficiency and current density [9,10]. Thus, in TC pentacene-based OTFTs, enhancement
Materials 2016, 9, 46; doi:10.3390/ma9010046
www.mdpi.com/journal/materials
Materials 2016, 9, 46
2 of 9
of carrier injection efficiency, which is crucial, can be achieved by introducing a carrier injection layer between the S/D metal electrodes and the pentacene channel layer. The carrier injection layer is believed to play a critical role in improving the electrical characteristics of the TC pentacene-based OTFTs. Carrier injection layers used in TC pentacene-based OTFTs can be divided into two categories: the insulating type of layers, which enables carriers from the S/D metal electrodes to tunnel to the pentacene channel layer easily, and the molecular doping type of layers, which forms a step hole transport configuration and is characterized by high local conductivity, thereby enhancing the injection of holes from the S/D metal electrodes to the pentacene channel layer [7,8,11,12]. The present author have previously reported TC pentacene-based OTFTs fabricated by introducing an ultrathin insulating Teflon carrier injection layer, which enhanced the drain current and field-effect mobility as a result of a decrease in the hole injection barrier and increased tunneling at the Au electrode/pentacene channel interface [7]. However, when the thickness of the Teflon layer exceeds a critical value, the carrier tunneling probability decreases because of an increase in the local resistance at the Au electrode/pentacene channel interface. It was found that the critical thickness of an insulating type of carrier injection layer is less than or equal to 5 nm. This critical thickness value is too low to accurately control the evaporation process, and therefore, it results in an unstable and sensitive process window. In the case of the molecular doping type, the organic materials employed for the buffer layer should be highly conductive for pentacene-based OTFTs, which be attributed to the boost of carrier density. Therefore, the use of the molecular doping materials might be an efficient way to tune carrier injection from metal electrode to organic material, resulting in a conductivity enhancement and thereby improved device performance. Kentaro et al. have confirmed that the Fermi energy (EF ) of F4 TCNQ-doped pentacene layer shifts toward the valence states [13]. Therefore, the molecular doping type of layers can be attributed to the improved conductivity of the p-doped layer with the doping of F4 TCNQ into the pentacene. Li et al. proposed a molecular doping carrier injection layer scheme in which the pentacene carrier injection layer of TC pentacene-based OTFTs is doped with different concentrations of tetrafluorotetracyanoquinodimethane (F4 TCNQ) [12]. The scheme involves the evaporation of a mixture solution of F4 TCNQ and pentacene in chloroform with a specific F4 TCNQ:pentacene ratio. However, the melting point of organic molecules varies with their molecular weight. The melting points of F4 TCNQ and pentacene are 290 ˝ C and above 300 ˝ C, respectively. The direct evaporation of a mixture solution of F4 TCNQ and pentacene with a specific F4 TCNQ:pentacene ratio, rather than the evaporation of these two chemicals individually (i.e., simultaneous “co-evaporation”), results in a discrepancy between the practical and designed F4 TCNQ:pentacene ratios during the evaporation process. The published report indicated that the co-evaporation method also had been used for interlayer deposition between the pentacene channel layer and Au electrodes, such as m-MTDATA-doped V2 O5 [11], MoOx -doped pentacene [14], and MoO3 and m-MTDATA-doped pentacene [15]. However, the co-evaporation scheme of the F4 TCNQ and pentacene for forming an F4 TCNQ-doped pentacene interlayer had been not investigated. In this study, for the first time, the co-evaporation scheme is used for evaporating F4 TCNQ and pentacene simultaneously to form an F4 TCNQ-doped pentacene interlayer between the Au S/D electrodes and the pentacene channel layer of TC pentacene-based OTFTs and, thus, improve the OTFT performance. The field-effect mobility was enhanced by a factor of 1.6 and the on/off current ratio was enhanced by a factor of 1.5 compared with a control device. In addition, the influence of the F4 TCNQ-doped pentacene interlayer thickness on the electrical characteristics of the OTFTs was investigated. The dependence of the device performance on the F4 TCNQ-doped pentacene interlayer (molecular doping type) thickness was not stronger than that for a Teflon interlayer (insulating type). Therefore, the F4 TCNQ-doped pentacene layer is a suitable carrier injection layer between the Au electrodes and the pentacene channel layer because it can not only improve the device performance but also facilitate obtaining a larger and stable process window.
Materials 2016, 9, 46
3 of 9
2. Device Fabrication Materials 2016, 9, 46
Pentacene-based OTFTs with a bottom-gate and top-contact structure were fabricated. A glass substrate with an indium-tin-oxide (ITO) layer was used as the substrate and as the bottom-gate 2. Device Fabrication electrode. A 350-nm-thick cross-linked poly (4-vinylphenol) (PVP) coating was provided on the Pentacene‐based OTFTs with a bottom‐gate and top‐contact structure were fabricated. A glass ITO layer for use as the gate dielectric, and it was cured in a vacuum oven at 200 ˝ C for 1.5 h. A substrate with an indium‐tin‐oxide (ITO) layer was used as the substrate and as the bottom‐gate 50-nm-thick pentacene channel layer was then deposited on the PVP layer through a shadow mask electrode. A 350‐nm‐thick cross‐linked poly (4‐vinylphenol) (PVP) coating was provided on the ITO by using a thermal evaporator (base pressure: 2.0 ˆ 10´6 Torr) at a deposition rate of 0.2 Å/s and layer for use as the gate dielectric, and it was cured in a vacuum oven at 200 °C for 1.5 h. A 50‐nm‐thick withpentacene channel layer was then deposited on the PVP layer through a shadow mask by using a the substrate temperature maintained at room temperature. Before the deposition of the S/D evaporator (base pressure: 2.0 × 10−6 Torr) at a deposition rate of 0.2 Å/s and was with the metalthermal electrodes, a 1.5-nm-thick F4 TCNQ-doped pentacene carrier injection interlayer deposited substrate temperature maintained at room temperature. Before the deposition of the S/D metal on the pentacene channel layer through a metal mask by simultaneously co-evaporating F4 TCNQ 1.5‐nm‐thick F4TCNQ‐doped carrier injection process, interlayer was deposited on of and electrodes, pentacenea from a thermal evaporator. pentacene In the co-evaporation the source materials the pentacene channel layer through a metal mask by simultaneously co‐evaporating F 4TCNQ and pentacene and F4 TCNQ were independently controlled, and were heated by linear thermal boat when pentacene from a thermal evaporator. In the co‐evaporation process, the source materials of the materials are deposited, as shown in Figure 1a. The deposition rate of F4 TCNQ was fixed at pentacene and F4TCNQ were independently controlled, and were heated by linear thermal boat 0.1 Å/s, while that of pentacene was set at 0.1, 0.3 and 1.0 Å/s; the designed F4 TCNQ:pentacene when the materials are deposited, as shown in Figure 1a. The deposition rate of F4TCNQ was fixed ratios corresponding to the pentacene deposition rates were 1:1, 1:3 and 1:10 respectively. Finally, Au at 0.1 Å/s, while that of pentacene was set at 0.1, 0.3 and 1.0 Å/s; the designed F 4TCNQ:pentacene (50 nm) S/D electrodes were formed through a shadow mask by using a thermal evaporator. Therefore, ratios corresponding to the pentacene deposition rates were 1:1, 1:3 and 1:10 respectively. Finally, the FAu TCNQ-doped pentacene can be uniformly deposited to form the interlayer between pentacene 4 (50 nm) S/D electrodes were formed through a shadow mask by using a thermal evaporator. Therefore, the F TCNQ‐doped pentacene can be uniformly deposited to form the interlayer between channel layer and 4Au metal electrodes. The width and length of the electrode channels were 500 pentacene channel layer and Au metal electrodes. The width and length of the electrode channels and 100 µm, respectively. Figure 1b shows a schematic structure of the pentacene-based OTFTs with were 500 and 100 μm, respectively. Figure 1b shows a schematic structure of the pentacene‐based F4 TCNQ-doped pentacene interlayer, and Figure 1c,d show the molecular structures of F4 TCNQ and OTFTs with F 4TCNQ‐doped pentacene interlayer, and Figure 1c,d show the molecular structures of pentacene, respectively. For convenience, the devices with F4 TCNQ:pentacene ratios of 1:1, 1:3 and F4TCNQ and pentacene, respectively. For convenience, the devices with F4TCNQ:pentacene ratios of 1:10 are hereafter designated as Devices 1, 2 and 3, respectively. A device without an F4 TCNQ-doped 1:1, 1:3 and 1:10 are hereafter designated as Devices 1, 2 and 3, respectively. A device without an pentacene interlayer was considered the control sample. All measurements were performed under F4TCNQ‐doped pentacene interlayer was considered the control sample. All measurements were darkperformed conditions and dark at anconditions ambient and temperature by using a semiconductor analyzer under at an ambient temperature by using a parameter semiconductor (HP parameter analyzer (HP 4145B, HP Inc., Palo Alto, CA, USA). 4145B, HP Inc., Palo Alto, CA, USA).
(a)
(b)
(c)
(d)
Figure 1. (a) a schematic diagram of the co‐evaporation process; (b) a schematic structure of the
Figure 1. (a) a schematic diagram of the co-evaporation process; (b) a schematic structure of the pentacene‐based organic thin‐film transistors (OTFTs) with F4TCNQ‐doped pentacene injection layer, pentacene-based organic thin-film transistors (OTFTs) with F4 TCNQ-doped pentacene injection layer, and the molecular structures of (c) F4TCNQ and (d) pentacene, respectively. and the molecular structures of (c) F4 TCNQ and (d) pentacene, respectively.
3
Materials 2016, 9, 46
4 of 9
Materials 2016, 9, 46
3. Results and Discussion 3. Results and Discussion Figure 2a shows the output characteristics (IDS -V DS ) of the devices with different DS‐Vsaturation DS) of the devices with different F 4TCNQ:pentacene Figure 2a shows the output characteristics (I F4 TCNQ:pentacene ratios for a V GS = ´20 V. The current of Device 1 was ´3.02 ˆ 10´6 ratios for a V GS = −20 V. The saturation current of Device 1 was −3.02 × 10 A for Vwas DS and V GS values A for V DS and V GS values of ´30 and ´20 V, respectively; this saturation−6current much higher of −30 and −20 V, respectively; this saturation current was much higher than those of Devices 2 and than those of Devices 2 and 3 and the control sample. The IDS value of Device 1 was approximately 3 and the control sample. The I DS value of Device 1 was approximately 193% higher than that of the 193% higher than that of the control sample. Figure 2b shows the transfer characteristics (IDS -V GS ) of control sample. 2b Fshows the transfer ratios. characteristics DS‐VGSof ) of the devices with different the devices withFigure different The µFE (Ivalue Device 1 was enhanced by a 4 TCNQ:pentacene F4TCNQ:pentacene ratios. The μ FE value of Device 1 was enhanced by a factor of 1.6 and the on/off factor of 1.6 and the on/off current ratio was enhanced by a factor of 1.5 compared with the control current ratio was enhanced by a factor of 1.5 compared with the control sample. This appreciable sample. This appreciable improvement in the electrical performance of Device 1 could be attributed to improvement in the electrical performance of Device 1 could be attributed to the F the F4 TCNQ:pentacene ratio being appropriate and to the charge density at interface4TCNQ:pentacene between the S/D ratio being appropriate and channel to the charge density at interface between the S/D electrodes electrodes and the pentacene being high, improving charge injection [12,16]. However,and the µthe FE pentacene channel being high, improving charge injection [12,16]. However, the μ FE value of the devices value of the devices decreased with a decrease in the F4 TCNQ concentration of the F4 TCNQ-doped decreased with a decrease in the F 4TCNQ concentration of the F4TCNQ‐doped pentacene interlayer. pentacene interlayer. -5
10
Control Device 1 Device 2 Device 3
-3.0 -2.5
Drain current (A)
Drain current (A)
-3.5
W/L=500/100 (m) VGS= -20V
-2.0 -1.5 -1.0 -0.5 0.0
0
-5
-10
-15
-20
Drain voltage (V)
-25
-30
(a)
Control Device 1
-6
10
Device 2 Device 3
-7
10
-8
10
W/L=500/100 (m)
VDS= -20V
-9
10
-10
10
10
5
0
-5 -10 -15 -20 -25 -30
Gate voltage (V) (b)
Figure 2. 2. Drain pentacene‐based OTFTs with different different Figure Drain current current characteristics characteristics of of top‐contact top-contact (TC) (TC) pentacene-based OTFTs with F 4TCNQ:pentacene ratios (Device 1 = 1:1, Device 2 = 1:3, and Device 3 = 1:10) (a) Output curves F4 TCNQ:pentacene ratios (Device 1 = 1:1, Device 2 = 1:3, and Device 3 = 1:10) (a) Output curves DS). (b) Transfer curves (IDS‐VGS). (IDS‐V (I DS -V DS ). (b) Transfer curves (IDS -V GS ).
To quantitate the effect of the F4TCNQ:pentacene ratio of the F4TCNQ‐doped pentacene To quantitate the effect of the F4 TCNQ:pentacene ratio of the F4 TCNQ-doped pentacene interlayer interlayer on the electrical contact between the pentacene channel layer and the S/D metal electrodes, on the electrical contact between the pentacene channel layer and the S/D metal electrodes, the channel the channel length dependence of the total resistance (RTotal) was determined according to the linear length dependence of the total resistance (RTotal ) was determined according to the linear regions regions of the output characteristics of the fabricated devices. The channel lengths were 100, 150, 200 of the output characteristics of the fabricated devices. The channel lengths were 100, 150, 200 and and 250 μm. The RC value was obtained from the transmission line model (TLM) by extrapolating 250 µm. The RC value was obtained from the transmission line model (TLM) by extrapolating the the RTotal curve to obtain the y‐intercept for zero channel length [17]. The RC values of the four RTotal curve to obtain the y-intercept for zero channel length [17]. The RC values of the four devices, devices, determined from the TLM, are shown in Figure 3. The RC value of Device 1 at a gate voltage determined from the TLM, are shown in Figure 3. The RC value of Device 1 at a gate voltage of 6 Ω, which is approximately 1.5 times lower than that of the control sample of −20 V was 0.617 × 10 ´20 V was6 0.617 ˆ 106 Ω, which is approximately 1.5 times lower than that of the control sample (0.935 × 10 Ω). As shown in Figure 3, RC first decreases and then increases with a decrease in the (0.935 ˆ 106 Ω). As shown in Figure 3, RC first decreases and then increases with a decrease in F4TCNQ concentration of the F4TCNQ‐doped pentacene interlayer. Apparently, the F4TCNQ the F4 TCNQ concentration of the F4 TCNQ-doped pentacene interlayer. Apparently, the F4 TCNQ concentration affects hole transport across the interface between the S/D metal electrodes and the concentration affects hole transport across the interface between the S/D metal electrodes and the pentacene channel layer. In the presence of an F4TCNQ‐doped pentacene interlayer, the hole carrier pentacene channel layer. In the presence of an F4 TCNQ-doped pentacene interlayer, the hole carrier could easily move through the F4TCNQ‐doped pentacene layer into the pentacene channel because could easily move through the F4 TCNQ-doped pentacene layer into the pentacene channel because of the improvement in the local conductivity and the formation of a multiple step barrier at the of the improvement in the local conductivity and the formation of a multiple step barrier at the interlayer [12,16]. F4TCNQ is an acceptor dopant, and therefore, in the interlayer, it leads to which interlayer [12,16]. F4 TCNQ is an acceptor dopant, and therefore, in the interlayer, it leads to which hole doping involving the electron transfer of energy level between pentacene and F4TCNQ, hole doping involving the electron transfer of energy level between pentacene and F4 TCNQ, resulting resulting in enhancing conductivity. Furthermore, the F4TCNQ‐doped pentacene interlayer changes in enhancing conductivity. Furthermore, the F4 TCNQ-doped pentacene interlayer changes the Fermi the Fermi energy (EF) position of the intrinsic pentacene, which leads to the formation of subdivided energy (EF ) position of the intrinsic pentacene, which leads to the formation of subdivided barriers barriers instead of a single barrier in the interlayer, and enhances the injection of hole carriers from instead of a single barrier in the interlayer, and enhances the injection of hole carriers from the the Au electrodes to the pentacene channel. Consequently, the F4TCNQ dopant amount can be easily fine‐tuned to achieve the optimal device performance, which corresponds to a reduction in RC, as shown in Figure 3. 4
Materials 2016, 9, 46
5 of 9
Au electrodes to the pentacene channel. Consequently, the F4 TCNQ dopant amount can be easily fine-tuned to achieve the optimal device performance, which corresponds to a reduction in RC , as shown in Figure 3. Materials 2016, 9, 46 Materials 2016, 9, 46 1.0
1.0
RC(M RC (M ) )
0.9
0.9
0.8
0.8 0.7 0.7 0.6 0.6 Control Device 1 Device 2 Device 3
Figure 3. The R value as a function of F 4TCNQ:pentacene ratio of TC pentacene‐based OTFTs for Figure 3. The RC value Cas a function of F4Control TCNQ:pentacene ratio TC pentacene-based OTFTs for V DS Device 3 2 of Device 1 Device VDS and VGS values of −2 and −20 V, respectively. and V GS values of ´2 and ´20 V, respectively.
Figure 3. The RC value as a function of F4TCNQ:pentacene ratio of TC pentacene‐based OTFTs for The dependence of the device performance on the thickness of the molecular‐doping‐type VDS and VGS values of −2 and −20 V, respectively. F4TCNQ‐doped pentacene layer was investigated. In a previous report of the present authors, the The dependence of the device performance on the thickness of the molecular-doping-type performance of pentacene‐based OTFTs was found to be strongly dependent on the thickness of the F4 TCNQ-doped pentacene of layer investigated. previous report of the present authors, the The dependence the was device performance In on a the thickness of the molecular‐doping‐type insulating‐type Teflon interlayer [7]. To investigate the effect of the thickness of F4TCNQ‐doped F4TCNQ‐doped pentacene layer was investigated. In a previous report of the present authors, the performance of pentacene-based OTFTs was found to be strongly dependent on of the pentacene and Teflon interlayers on the device performance, the F4TCNQ:pentacene the ratio thickness of the performance of pentacene‐based OTFTs was found to be strongly dependent on the thickness of the insulating-type Teflon interlayer [7]. To investigate the effect of the thickness of F4 TCNQ-doped F4TCNQ‐doped pentacene interlayer was fixed at 1:1, and the thicknesses of both carrier injection insulating‐type Teflon investigate the shows effect the of thickness‐dependent the thickness of F4transfer TCNQ‐doped interlayers were set at interlayer 0, 1.5, 3, and 10 nm. Figure 4a,b pentacene and Teflon interlayers on5 [7]. the To device performance, the F4 TCNQ:pentacene ratio of the property of the device with Teflon and F 4TCNQ‐doped pentacene interlayers of various thicknesses. pentacene and Teflon interlayers on the device performance, the F 4TCNQ:pentacene ratio of the F4 TCNQ-doped pentacene interlayer was fixed at 1:1, and the thicknesses of both carrier injection The device with Teflon interlayer showed remarkable thickness dependence, while the one with F4TCNQ‐doped pentacene interlayer was fixed at 1:1, and the thicknesses of both carrier injection interlayersF4were set at pentacene 0, 1.5, 3, interlayer 5 and 10showed nm. Figure 4a,b showsIn the thickness-dependent transfer TCNQ‐doped a slight dependence. addition, Figure 4c,d shows transfer interlayers were set at 0, 1.5, 3, 5 and 10 nm. Figure 4a,b shows the thickness‐dependent the AFM micrographs of the pentacene layers without and with a 5 nm F 4TCNQ‐doped pentacene property of the device with Teflon and F4 TCNQ-doped pentacene interlayers of various thicknesses. property of the device with Teflon and F 4TCNQ‐doped pentacene interlayers of various thicknesses. interlayer deposited on the PVP dielectric surface. The root mean square (rms) roughness for both The device with Teflon interlayer showed remarkable thickness dependence, whilethe the one with The devices device without with Teflon interlayer showed remarkable thickness dependence, while one with and with 5 nm F4TCNQ‐doped pentacene interlayer is 5.62 and 4.97 nm, F4 TCNQ-doped pentacene interlayer showed a slight dependence. In addition, addition,Figure 4c,d shows Figure 4c,d shows F4TCNQ‐doped pentacene interlayer showed a slight dependence. In respectively. It had been clearly seen that the surface become more smooth for devices with 5 nm the AFM micrographs of the pentacene layers without and with a 5 nm F TCNQ-doped the AFM micrographs of the pentacene layers without and with a 5 nm F 4 TCNQ‐doped pentacene 4 F4TCNQ‐doped pentacene interlayer, which is necessary for providing a better contact with metal pentacene interlayer deposited on the PVP dielectric surface. The root mean square (rms) roughness for both electrodes. This is the mechanism for lowering the barrier height and increasing the charge injection, interlayer deposited on the PVP dielectric surface. The root mean square (rms) roughness for both and then improving The pentacene field‐effect mobility (μ FE) was calculated a Vrespectively. GS devices without and with 5 nm F4TCNQ‐doped pentacene interlayer is 5.62 and 4.97 nm, devices without and with 5 device nm F4performance. TCNQ-doped interlayer is 5.62 and 4.97for nm, value of −30 V in the saturation region, and the threshold voltage (V TH) was determined according respectively. It had been clearly seen that the surface become more smooth for devices with 5 nm It had been clearly seen that the surface become more smooth for devices with 5 nm F4 TCNQ-doped to the saturation region of the plot of |IDS|1/2 versus VGS [18]. The sub‐threshold swing (SS) was F4TCNQ‐doped pentacene interlayer, which is necessary for providing a better contact with metal pentacene interlayer, which is necessary for providing a better contact with metal electrodes. This is calculated using the equation SS = dVGS/d(log10|IDS|). The maximum and minimum values of I DS at a electrodes. This is the mechanism for lowering the barrier height and increasing the charge injection, −20 V were the designated ION (on current) and Ithe OFF (off current), respectively [19]. improving VDS of for the mechanism lowering barrieras height and increasing charge injection, and then and The then improving device performance. The field‐effect mobility (μFE) and was F4calculated for a VGS thickness‐dependent properties of pentacene‐based with Teflon TCNQ‐doped device performance. The field-effect mobility (µFE ) was OTFTs calculated for a V value of ´30 V in the GS value of −30 V in the saturation region, and the threshold voltage (V pentacene interlayers of various thicknesses were summarized in Table 1. TH) was determined according
Drain current (A)
saturation region, and the threshold (V was determined according to theswing saturation region to the saturation region of the voltage plot of |I DSTH |1/2) versus VGS [18]. The sub‐threshold (SS) was of the plot of |IDS |1/2 versus V GS [18]. TheGSsub-threshold swing (SS) was calculated using the equation calculated using the equation SS = dV /d(log10|IDS|). The maximum and minimum values of I DS at a SS = dV |I The maximum values of IDS at a Vrespectively were /d(log of −20 10V were designated as ION and (on minimum current) Unstable and Iprocess OFF (off current), [19]. VDS GS DS |). DS of ´20 V window Stable process window 10 10 The thickness‐dependent pentacene‐based OTFTs with Teflon F4TCNQ‐doped designated as ION (on current)properties and IOFFof (off current), respectively [19]. The and thickness-dependent pentacene interlayers of various thicknesses were summarized in Table 1. 10 properties of pentacene-based OTFTs Control with Teflon and F410TCNQ-doped pentacene interlayers of various Control 1.5nm 1.5nm 3nm 1. thicknesses were 10summarized in Table 3nm 10 -7
-8
5nm 10nm
-9
Stable process window
Drain current (A)
-6
10
F4TCNQ-doped pentacene
10
-10
10
10
5
-7
10
0
(a)
10
-9
1.5nm 3nm 5nm 10nm F4TCNQ-doped pentacene
10
10
5
0
-5 -10 -15 -20 -25 -30
Gate voltage (V)
-8
10
5nm 10nm Teflon
-9
-10
10
-6
10
10
-7
10
-8
10
-9
Figure 4. Cont.
W/L=500/150 (m) VDS= -20V
-10
-7
10
-5 -10 -15 -20 -25 -30
Gate voltage (V) Control
-8
10
W/L=500/150 (m) VDS= -20V
-6
Drain current (A)
Drain current (A)
-6
10
W/L=500/150 m Unstable process window (
5
0
-5 -10 -15 -20 -25 -30
Gate voltage (V) (b)
(a)
10
5
0
Figure 4. Cont.
5
-5 -10 -15 -20 -25 -30
Gate voltage (V) (b)
Figure 4. Cont.
Control 1.5nm 3nm 5nm 10nm Teflon
W/L=500/150 (m) VDS= -20V
-10
5
)
VDS= -20V
Materials 2016, 9, 46 Materials 2016, 9, 46
6 of 9
(c)
(d)
Figure 4. Transfer characteristics of pentacene‐based OTFTs with (a) F TCNQ‐doped pentacene and Figure 4. Transfer characteristics of pentacene-based OTFTs with (a) F44TCNQ-doped pentacene and (b) Teflon interlayers of various thicknesses, and AFM images of pentacene films (c) (c) without without and and (b) Teflon interlayers of various thicknesses, and AFM images of pentacene films 4 TCNQ‐doped pentacene interlayer. (d) with 5 nm F (d) with 5 nm F4 TCNQ-doped pentacene interlayer. Table 1.1. The The electrical electrical characteristics characteristics for for the the devices devices with with FF4TCNQ-doped TCNQ‐doped pentacene pentacene and and Teflon Teflon Table 4 interlayers of various thicknesses. interlayers of various thicknesses. Molecular Doping Type: Insulating Type: Electrical Molecular Doping Type: F4TCNQ‐Doped Pentacene Teflon Insulating Type: Teflon Parameters F4 TCNQ-Doped Pentacene Electrical Parameters Control 1.5 nm 3 nm 5 nm 10 nm Control 1.5 nm 3 nm 5 nm 10 nm 1.5 nm −4.4 3 nm −4.1 5 nm −4.2 10 nm −6 Control 1.5 nm 3−5.5 nm 5−3 nm 10 nm VTH (V) −7.7 Control −5.7 −5.9 −5.3 SS (V/decade) 2.2 2.7 2.5 2.6 2.9 1.77 1.98 1.73 1.93 2.5 V TH (V) ´7.7 ´5.7 ´4.4 ´4.1 ´4.2 ´6 ´5.9 ´5.5 ´3 ´5.3 2 V−1 s−1) 0.12 2.2 0.14 2.7 0.17 2.5 0.21 2.6 0.2 2.9 0.08 0.09 0.12 0.15 0.02 μFE (cm SS (V/decade) 1.77 1.98 1.73 1.93 2.5 2¨ V ´1 ¨ s´1 )4.95 0.12 3) 0.14 0.17 0.21 0.2 0.08 0.09 0.12 0.15 0.02 (cm /I OFF (10 5 5.34 4.3 5.76 10.4 9.86 20.2 15.6 39.7 IµON FE 4.95 5 5.34 4.3 5.76 10.4 9.86 20.2 15.6 39.7 ION /IOFF (103 )
The RTotal of the device with Teflon and F4TCNQ‐doped pentacene interlayers of various thicknesses is shown Figure 5a,b, respective. variation of the RC and μFE values the The RTotal of the in device with Teflon and F4The TCNQ-doped pentacene interlayers of with various carrier‐injection interlayer thickness for the devices with Teflon and TCNQ‐doped thicknesses is shown in Figure 5a,b, respective. The variation of the RCF4and µFE valuespentacene with the interlayers is shown in Figure 5c. The R C value of the device with a molecular‐doping‐type carrier-injection interlayer thickness for the devices with Teflon and F4 TCNQ-doped pentacene F4TCNQ‐doped pentacene interlayer depended only on with the Fa4TCNQ‐doped pentacene interlayers is shown in Figure 5c. The RC value of slightly the device molecular-doping-type FE, increased with the Finterlayer thickness. The performances of both types of devices, according to μ 4 TCNQ-doped pentacene interlayer depended only slightly on the F4 TCNQ-doped pentacene thickness. However, when the interlayer thickness exceeded 5 nm, the μ FE value of the device with interlayer thickness. The performances of both types of devices, according to µFE , increased with the an F4TCNQ‐doped pentacene interlayer decreased slightly; by contrast, the μ FE value of the device thickness. However, when the interlayer thickness exceeded 5 nm, the µFE value of the device with with a Teflon interlayer decreased appreciably. It is presumed that the degradation μFE device can be an F4 TCNQ-doped pentacene interlayer decreased slightly; by contrast, the µFE value of of the directly attributed to the increase in R C . Unlike the R C value of the device with an insulating‐type with a Teflon interlayer decreased appreciably. It is presumed that the degradation of µFE can be Teflon interlayer, the C value in of RCthe device a molecular‐doping‐type 4TCNQ‐doped directly attributed to theRincrease . Unlike thewith RC value of the device with an F insulating-type pentacene interlayer depended only slightly on the F 4 TCNQ‐doped pentacene interlayer thickness. Teflon interlayer, the RC value of the device with a molecular-doping-type F4 TCNQ-doped pentacene 18 Ω∙cm). Therefore, when the Teflon is an insulating material with an extremely high resistivity (10 interlayer depended only slightly on the F4 TCNQ-doped pentacene interlayer thickness. Teflon is 18 Teflon interlayer thickness exceeds a critical value, the insulating properties the the interlayer an insulating material with an extremely high resistivity (10 Ω¨ cm). Therefore, of when Teflon become dominant and reduce the carrier tunneling probability. Consequently, the device interlayer thickness exceeds a critical value, the insulating properties of the interlayer become dominant performance is carrier degraded appreciably. It is to Consequently, say that the molecular doping‐type F4TCNQ‐doped and reduce the tunneling probability. the device performance is degraded pentacene interlayer is a suitable carrier injection layer that can not only improve the TC‐OTFT appreciably. It is to say that the molecular doping-type F4 TCNQ-doped pentacene interlayer is a performance but also facilitate obtaining a stable process window. It is the first time to compare the suitable carrier injection layer that can not only improve the TC-OTFT performance but also facilitate process windows for the interlayers of the insulating type and molecular‐doping type. It is believed obtaining a stable process window. It is the first time to compare the process windows for the interlayers that the results will be interesting and important from the viewpoint of manufacturing as a result of of the insulating type and molecular-doping type. It is believed that the results will be interesting and the stable process window. important from the viewpoint of manufacturing as a result of the stable process window.
6
Materials 2016, 9, 46 Materials 2016, 9, 46
7 of 9
(a)
(b) 0.25
F4TCNQ-doped pentacene Teflon
0.20 0.15 0.10
8.00
0.05
1 0
0
2
4
6
8
10
Mobility (cm2V-1s-1)
R C ( M )
8.25
0.00
Interlayer thickness (nm) (c) Figure 5. 5. The and RC of pentacene‐based OTFTs with the different thickness of the Figure The RRTotal Total and RC of pentacene-based OTFTs with the different thickness of the (a) FF44TCNQ-doped TCNQ‐doped pentacene pentacene and and (b) (b) Teflon Teflon interlayer. interlayer. (c) The variation variation of of the the mobility mobility and and Rc Rc (a) (c) The 4 TCNQ‐doped values with the various carrier‐injection interlayer thickness for the devices with F values with the various carrier-injection interlayer thickness for the devices with F4 TCNQ-doped pentacene or Teflon interlayers. pentacene or Teflon interlayers.
In Figure 6, the dependence of the carrier injection efficiency on the thickness of the Teflon and In Figure 6, the dependence of the carrier injection efficiency on the thickness of the Teflon and F4TCNQ‐doped pentacene is clearly explained. In the control sample, hot Au atoms are directly F4 TCNQ-doped pentacene is clearly explained. In the control sample, hot Au atoms are directly deposited onto the pentacene channel layer, and diffuse into the upper layer of pentacene. A high deposited onto the pentacene channel layer, and diffuse into the upper layer of pentacene. A high energy barrier formed because of the presence of an interface dipole, there is a hole injection barrier energy barrier formed because of the presence of an interface dipole, there is a hole injection barrier of of 0.8–1 eV between the Au electrodes and the pentacene channel layer, resulting in a relatively large 0.8–1 eV between the Au electrodes and the pentacene channel layer, resulting in a relatively large hole injection barrier, as shown in Figure 6a [6,7]. In addition, a high energy barrier associated with hole injection barrier, as shown in Figure 6a [6,7]. In addition, a high energy barrier associated with the formation of a metallic surface at the Au electrode/pentacene channel interface obstructed hole the formation of a metallic surface at the Au electrode/pentacene channel interface obstructed hole carrier transport. In the 5‐nm‐thick Teflon interlayer, the carriers could easily tunnel through to the carrier transport. In the 5-nm-thick Teflon interlayer, the carriers could easily tunnel through to the pentacene channel layer because of suppressed metal penetration and a reduced hole injection pentacene channel layer because of suppressed metal penetration and a reduced hole injection barrier barrier at the Au electrode/pentacene channel interface [7]. However, for a Teflon layer thickness at the Au electrode/pentacene channel interface [7]. However, for a Teflon layer thickness greater greater than 5 nm, the probability of carriers tunneling through clearly decreased because of the than 5 nm, the probability of carriers tunneling through clearly decreased because of the increased increased RC at the Au electrode/pentacene channel interface. In the case of the F4TCNQ‐doped RC at the Au electrode/pentacene channel interface. In the case of the F4 TCNQ-doped pentacene pentacene interlayer, regardless of its thickness (5‐ or 10‐nm thick), the hole carriers could easily interlayer, regardless of its thickness (5- or 10-nm thick), the hole carriers could easily move through move through the interlayer to the Au electrodes because of the formation of a subdivided barrier the interlayer to the Au electrodes because of the formation of a subdivided barrier and high local and high local conductivity at the Au electrode/pentacene channel interface, which effectively conductivity at the Au electrode/pentacene channel interface, which effectively reduced the hole reduced the hole injection barrier. Moreover, in the device containing a 5‐nm‐thick F4TCNQ‐doped injection barrier. Moreover, in the device containing a 5-nm-thick F4 TCNQ-doped pentacene interlayer pentacene interlayer with an F4TCNQ:pentacene ratio of 1:1, the carrier injection efficiency is high with an F4 TCNQ:pentacene ratio of 1:1, the carrier injection efficiency is high and RC is low, resulting and RC is low, resulting in high device performance. Even when the thickness of the F4TCNQ‐doped in high device performance. Even when the thickness of the F4 TCNQ-doped pentacene interlayer pentacene interlayer was 10 nm, the effect of the thickness on the OTFT electrical performance was was 10 nm, the effect of the thickness on the OTFT electrical performance was not apparent, implying not apparent, implying that the thickness dependence of the electrical performance of OTFTs with a that the thickness dependence of the electrical performance of OTFTs with a molecular-doping-type molecular‐doping‐type F4TCNQ‐doped pentacene interlayer is weaker than that of the performance F4 TCNQ-doped pentacene interlayer is weaker than that of the performance of OTFTs with an of OTFTs with an insulating‐type Teflon interlayer. Therefore, it is believed that an F4TCNQ‐doped insulating-type Teflon interlayer. Therefore, it is believed that an F4 TCNQ-doped pentacene interlayer pentacene interlayer is a good candidate as a carrier injection layer for TC pentacene‐based OTFTs since it provides a high carrier injection efficiency and a stable process window. 7
Materials 2016, 9, 46
8 of 9
is a good candidate as a carrier injection layer for TC pentacene-based OTFTs since it provides a high and a stable process window.
Materials 2016, 9, 46 carrier injection efficiency
Figure 6. Band Figure 6. Band diagram diagram of of top‐contact top-contact (TC) (TC) pentacene‐based pentacene-based OTFTs OTFTs with with various various interlayer interlayer thicknesses (a) control sample, (b) 5‐, and (c) 10‐nm‐thick Teflon and (d) 5‐, and (e) 10‐nm‐thick thicknesses (a) control sample, (b) 5-, and (c) 10-nm-thick Teflon and (d) 5-, and (e) 10-nm-thick F44TCNQ‐doped pentacene. TCNQ-doped pentacene.
4. Conclusions 4. Conclusions 4TCNQ‐doped pentacene interlayer was fabricated as a carrier injection In summation, An An FFTCNQ-doped In summation, pentacene interlayer was fabricated as a carrier injection layer 4 layer by using the co‐evaporation method, and the effect of the doping concentration and thickness by using the co-evaporation method, and the effect of the doping concentration and thickness of the of the interlayer on the performance of pentacene‐based OTFTs was investigated. The results show interlayer on the performance of pentacene-based OTFTs was investigated. The results show that the that the device an F4TCNQ:pentacene of considerably 1:1 showed considerably improved electrical device with an F4with TCNQ:pentacene ratio of 1:1 ratio showed improved electrical characteristics characteristics compared with the device without an interlayer, because of a decrease in the contact compared with the device without an interlayer, because of a decrease in the contact resistance resistance the metal and channel. the pentacene channel. Furthermore, observations between thebetween metal electrodes andelectrodes the pentacene Furthermore, observations suggested that the suggested that the dependence of the electrical performance of the OTFTs on the molecular‐doping‐type dependence of the electrical performance of the OTFTs on the molecular-doping-type F4 TCNQ-doped F 4TCNQ‐doped pentacene interlayer thickness was weaker than that of the performance of an pentacene interlayer thickness was weaker than that of the performance of an insulating-type Teflon pentacene insulating‐type Teflon interlayer. Therefore, it is believed that the interlayer F4TCNQ‐doped interlayer. Therefore, it is believed that the F4 TCNQ-doped pentacene (molecular doping interlayer (molecular doping type) is a suitable candidate for the carrier injection layer of TC type) is a suitable candidate for the carrier injection layer of TC pentacene-based OTFTs since it pentacene‐based OTFTs since it facilitates obtaining a stable process window and improves the facilitates obtaining a stable process window and improves the OTFT performance. OTFT performance.
Acknowledgments: The authors would like to acknowledge the financial support of the National Science Council Acknowledgments: The No. authors would like to acknowledge financial support of the National of Taiwan under contract MOST 104-2221-E-011-159, and of the the Taiwan Building Technology CenterScience (TBTC) Council of Taiwan under contract No. MOST 104‐2221‐E‐011‐159, and of the Taiwan Building Technology of National Taiwan University of Science and Technology (NTUST). Center (TBTC) of National Taiwan University of Science and Technology (NTUST). Author Contributions: Ching-Lin Fan advise the study and work; Ching-Lin Fan and Hsiang-Sheng Chang designedContributions: the research; Hsiang-Sheng Chang andthe Wei-Chun Lin work; performed the experimental work; Wei-Chun Lin Author Ching‐Lin Fan advise study and Ching‐Lin Fan and Hsiang‐Sheng Chang wrote the manuscript. All authors discussed, edited and approved the final version. designed the research; Hsiang‐Sheng Chang and Wei‐Chun Lin performed the experimental work; Conflicts of Interest: The authors declare no conflict of interest. Wei‐Chun Lin wrote the manuscript. All authors discussed, edited and approved the final version.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Lin, S.K.; Li, Y.C.; Lin, Y.J.; Wang, Y.H. Effects of tri-layer polymer dielectrics on electrical characteristics in References
1. 2. 2.
3.
pentacene thin-film transistors. ECS Solid State Lett. 2014, 3, N19–N22. [CrossRef] Lin, S.K.; Li, Y.C.; Lin, Y.J.; Wang, Y.H. Effects of tri‐layer polymer dielectrics on electrical characteristics Chen, H.Y.; Wu, I.W.; Chen, C.T.; Liu, S.W.; Wu, C.I. Self-assembled monolayer modification of silver in pentacene thin‐film transistors. ECS Solid State Lett. 2014, 3, N19–N22. source-drain electrodes for high-performance pentacene organic field-effect transistors. Org. Electron. 2012, Chen, H.Y.; Wu, I.W.; Chen, C.T.; Liu, S.W.; Wu, C.I. Self‐assembled monolayer modification of silver 13, 593–598. [CrossRef] source‐drain electrodes for high‐performance pentacene organic field‐effect transistors. Org. Electron. 2012, 13, 593–598. Fan, C.L.; Lin, Y.Z.; Chiu, P.C.; Wang, S.J.; Lee, W.D. Teflon/SiO2 bilayer passivation for improving the electrical reliability of pentacene‐based organic thin‐film transistors. Org. Electron. 2013, 14, 2228–2232.
Materials 2016, 9, 46
3.
4.
5.
6. 7. 8. 9. 10.
11. 12. 13. 14. 15.
16. 17. 18.
19.
9 of 9
Fan, C.L.; Lin, Y.Z.; Chiu, P.C.; Wang, S.J.; Lee, W.D. Teflon/SiO2 bilayer passivation for improving the electrical reliability of pentacene-based organic thin-film transistors. Org. Electron. 2013, 14, 2228–2232. [CrossRef] Fujisaki, Y.; Nakajima, Y.; Takei, T.; Fukagawa, H.; Yamamoto, T.; Fujikake, H. Flexible active-matrix organic light-emitting diode display using air-stable organic semiconductor of Dinaphtho[2, 3-b: 21 , 31 -f]thieno[3, 2-b]-thiophene. IEEE Trans. Electron. Devices 2012, 59, 3442–3449. [CrossRef] Su, Y.; Wang, M.; Xie, F.; Chen, J.; Xie, W.; Zhao, N.; Xu, J. In situ modification of low-cost Cu electrodes for high-performance low-voltage pentacene thin-film transistors (TFTs). Org. Electron. 2013, 14, 775–781. [CrossRef] Xu, Y.; Liu, C.; Sun, H.; Balestra, F.; Ghibaudo, G.; Scheideler, W.; Noh, Y.Y. Metal evaporation dependent charge injection in organic transistors. Org. Electron. 2014, 15, 1738–1744. [CrossRef] Fan, C.L.; Chiu, P.C. Performance improvement of top-contact pentacene-based organic thin-film transistors by inserting an ultrathin Teflon carrier injection layer. Jpn. J. Appl. Phys. 2011, 50. [CrossRef] Alam, M.W.; Wang, Z.; Naka, S.; Okada, H. Mobility enhancement of top contact pentacene based organic thin-film transistor with bi-layer GeO/Au electrodes. Appl. Phys. Lett. 2013, 102. [CrossRef] Wang, H.; Wang, L.; Gao, Y.; Zhang, D.; Qiu, Y. Efficient organic light-emitting diodes with Teflon as buffer layer. Jpn. J. Appl. Phys. 2004, 43, L1353–L1355. [CrossRef] Wu, Z.; Wang, L.; Wang, H.; Gao, Y.; Qiu, Y. Charge tunneling injection through a thin teflon film between the electrodes and organic semiconductor layer: Relation to morphology of the teflon film. Phys. Rev. B 2006, 74, 165307–165315. [CrossRef] Su, S.H.; Wu, C.M.; Kung, S.Y.; Yokoyama, M. Enhancing the performance of organic thin-film transistors using an organic-doped inorganic buffer layer. Thin Solid Films 2013, 536, 229–234. [CrossRef] Li, J.; Zhang, X.W.; Zhang, L.; Haq, K.U.; Jiang, X.Y.; Zhu, W.Q.; Zhang, Z.L. Improving organic transistor performance through contact-area-limited doping. Solid State Commun. 2009, 149, 1826–1830. [CrossRef] Harada, K.; Riede, M.; Leo, K.; Hild, O.R.; Elliott, C.M. Pentacene homojunctions: Electron and hole transport properties and related photovoltaic responses. Phys. Rev. B 2008, 78. [CrossRef] Lee, C.T.; Chen, H.C. Performance improvement mechanisms of organic thin-film transistors using MoOx -doped pentacene as channel layer. Org. Electron. 2011, 12, 1852–1857. [CrossRef] Yan, P.; Liu, Z.; Zhang, S.; Liu, D.; Wang, X.; Yue, S.; Zhao, Y. Observation of hole injection boost via two parallel paths in Pentacene thin-film transistors by employing Pentacene: 4,4”-tris(3-methylphenylphenylamino) triphenylamine: MoO3 buffer layer. APL Mater. 2014, 2. [CrossRef] Wakatsuki, Y.; Noda, K.; Wada, Y.; Toyabe, T.; Matsushige, K. Molecular doping effect in bottom-gate, bottom-contact pentacene thin-film transistors. J. Appl. Phys. 2011, 110. [CrossRef] Jin, S.H.; Jung, K.D.; Shin, H.; Park, B.G.; Lee, J.D. Grain size effects on contact resistance of top-contact pentacene TFTs. Synth. Met. 2006, 156, 196–201. [CrossRef] Fan, C.L.; Yang, T.H.; Chiu, P.C. Performance improvement of bottom-contact pentacene-based organic thin-film transistors by inserting a thin polytetrafluoroethylene buffer layer. Appl. Phys. Lett. 2010, 97. [CrossRef] Bae, J.H.; Choi, Y. Engineered interface using a hydroxyl group-free polymeric buffer layer onto a TiO2 nanocomposite film for improving the electrical properties in a low-voltage operated organic transistor. Surf. Interface Anal. 2012, 44, 445–449. [CrossRef] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).