Interfacial effects on solution-sheared thin-film transistors

Dong-Yue Guo , Yi-bei Tsai , Ting-Feng Yu and Wen-Ya Lee *
Department of Chemical Engineering and Biotechnology, National Taipei University of Technology, Taipei 106, Taiwan, Republic of China. E-mail: wenyalee@mail.ntut.edu.tw

Received 27th March 2018 , Accepted 10th May 2018

First published on 10th May 2018


Meniscus-guided solution-sheared fabrication processes are considered a promising approach to develop high-performance polymer-based transistors owing to their enhancement of electrical performance and improvement of cystalline structure. However, the effect of the interface on the molecular packing and charge transport in the solution-sheared devices remains unclear. Therefore, in this study, we investigated interfacial effects on the electrical properties and crystalline morphology of solution-sheared polymer films based on a high-mobility donor–acceptor copolymer, poly(diketopyrrolo[3,4-c]pyrrole-co-thieno[3,2-b]thiophene) (PBDT-co-TT). We employed solution-shearing processes for making substrates modified with octadecyl trimethoxylsilane (OTS) and phenylbutyltrimethoxysilane, and bare substrates for device fabrication. The highest mobility obtained with OTS devices was 1.77 cm2 V−1 s−1, which is much higher than that obtained with the bare structure. In addition, OTS-treated devices exhibited the highest polymer alignment. According to the analysis of thicknesses and shearing speed, we conclude that OTS-modified substrates provide an ideal interface, fulfilled by simple mass balance near the meniscus, in contrast to the bare substrate. Grazing incidence X-ray diffraction analysis revealed that the OTS-based film showed the longest coherence length and a tunable lamellar d-spacing distance with shearing speeds. In contrast, the film made on the bare subtrate exhibited the smallest coherence length and a negligible change of lamellar d-spacing distance with shearing speeds. Thus, this study demonstrates the importance of the interface on polymer alignment, charge transport and meta-stable molecular packing for the solution shearing process. This may enhance the application of solution processes in the electronics industry.


Introduction

Solution deposition of organic or polymeric semiconductors provides a facile approach to prepare large-scale roll-to-roll flexible integrated circuits.1,2 Furthermore, solution deposition can manipulate the molecular packing and crystalline orientation structure of polymer chains by optimizing process parameters.3–8 Low-temperature solution deposition enables the fabrication of organic electronics on a flexible plastic substrate. Therefore, solution deposition is considered as a promising method for the large-scale production of flexible electronics. To date, several solution process methods have been developed for organic electronics, including spin coating, die-slot coating,9 dip coating,10 drop casting,11 inkjet printing,12,13 doctor blading,14–17 zone casting,18 solution shearing,19–24 and others.25

Among these solution deposition methods, solution shearing is a facile approach to deposit high-quality organic and polymer thin films.26–30 Solution shearing requires only a small amount of an organic solution (<30 μL) to deposit a film with a size of several square centimeters. The organic solution is sandwiched between two preheated substrates. By moving the top substrate, a highly crystalline organic film can be produced. Compared with spin-coated films, solution-sheared films, in general, have higher crystallinity and charge mobility because of the meniscus-guided growth of organic crystals during solution shearing. Solution-sheared polymer semicrystalline films can be produced with a solution meniscus, which is an unstable air–liquid interface on the polymer solution. With the evaporation of the used organic solvent, the meniscus moves with the movement of the top blade. Once the solution achieves supersaturation, the solute forms a thin film. Because of the linear motion of the solution meniscus, organic molecules tend to form a metastable crystalline structure and exhibit improved charge transport properties. This method was developed by Hector et al. in 2008. They deposited the organic crystals of substituted quarterthiophene devices with a mobility greater than 1 cm2 V−1 s−1.26 Subsequently, they obtained a high mobility value of more than 4.6 cm2 V−1 s−1 from the solution-sheared 6,13-bis(triisopropylsilylethynyl)pentacene (TIPS-pentacene) film.31 Moreover, the same group reported that the shearing process can alter the tilted angles of TIPS-pentacene crystals and result in close π–π stacking. Diao et al. employed a top blade with the micropillar structure for solution shearing and obtained a single crystal of TIPS-pentacene with an extremely high mobility of 11 cm2 V−1 s−1,32 which is one of the highest values reported from TIPS-pentacene in the literature.

Although several studies have focused on small-molecule-based devices using solution shearing, only a few studies have reported on polymer-based devices using solution shearing.23,33,34 Conjugated copolymers have attracted considerable attention for organic thin-film transistors (OTFTs) because of their excellent flexibility, unique electronic properties, and solution processablity.35–38 Compared with common spin-coating processes, solution shearing has several advantages for the fabrication of polymer-based devices. For example, during solution shearing, the polymer solution is treated at an elevated temperature. The high temperature can reduce the entanglement of the polymer chains, resulting in improved molecular packing.23 Furthermore, the convection flow induced by the movement of the top blade can potentially control crystal growth through the parameter optimization of solution shearing.

Giri et al. first reported the influence of shearing parameters on the charge transport properties of conjugated polymers. They demonstrated the structural rearrangement of a diketopyrrolopyrrole (DPP)-based conjugated polymer and poly(2,5-bis(3-alkylthiophene-2-yl)thieno[3,2-b]thiophenes) (PBTTT) through solution shearing.33 They deposited polymer films on phenyltrichlorosilane (PTS)-modified SiO2/Si substrates through solution shearing at various shearing speeds. They found that a high shearing speed facilitates the formation of a metastable molecular packing structure. However, no significant dependence was observed between charge mobilities and shearing speeds. They obtained the maximum mobilities of approximately 0.07 and 0.01 cm2 V−1 s−1 for the DPP-based polymer and PBTTT, respectively. They suggested that the low mobility and insignificant mobility dependence may be attributed to no significant changes in the polymer backbone tilt and π–π stacking distance, which are critical for charge transport in organic semiconductors.

In addition to solution deposition parameters, the semiconductor/dielectric interface plays a crucial role in charge transport in OTFTs. The surface parameters of the interface, including roughness, surface energy, and chemical functionality, may lead to significant variations in device performance.39–42 To optimize the interface, modifying self-assembly monolayers (SAMs) is a popular approach for OTFT fabrication. The Iwasa group reported that a dipole of SAMs can manipulate Schottky barriers and control the charge density in the channel.43 Ajay et al. demonstrated that crystalline, dense monolayers prepared from octadecyl trimethoxylsilane (OTS) by using the Langmuir–Blodgett technique can promote the two-dimensional crystal growth of organic semiconductors.40 Kim et al. reported a significant improvement in the mobility of pentacene-based transistors by a factor of 3 by using OTS surface modification.44

Although SAM modification is widely used in organic or polymer-based transistors, interfacial effects on solution-sheared OTFT performance have rarely been investigated. Moreover, how the metastable crystalline structure varies with different interfaces during solution shearing remains unclear. Therefore, in this study, we selected two SAMs for device fabrication, including n-octadecyltrimethoxysilane (OTS) and 4-phenylbutyltrimethoxysilane (PBTS) for surface modifications. Both the SAMs possess methoxylsilane groups, which can cause a slower reaction with hydroxyl groups on SiO2/Si substrates compared with molecules with trichlorosilane groups. The octadecyl tail group of the OTS molecule tends to form an ordered hexagonal structure due to the strong van der Waals force.39 The formation of crystalline OTS SAMs provides high-quality monolayers for the deposition of organic molecules and polymers.23,39 By contrast, PBTS showed a different self-assembly behavior. No crystalline packing formed on the PBTS layer. Additionally, the phenyl end group of PBTS provided better wetting capability with common organic solvents, which is a major benefit for solution processes. Additionally, a bare wafer without any specific modification was employed as a control device for comparison. In this article, we investigate interfacial effects on the solution shearing parameters, metastable crystalline structure, and charge transport properties.

Experimental

Materials

The conjugated polymer comprising diketopyrrolopyrrole and thieno[3,2-b]thiophene moieties, PBDT-co-TT, was purchased from Lumtech, Taiwan (molecular weight = approximately 36[thin space (1/6-em)]000). Anhydrous toluene and o-dichlorobenzene were purchased from Sigma-Aldrich and used without any further purification.

Transistor fabrication

Highly doped n-type Si(100) wafers (<0.004 Ω cm) with a thermally grown 300 nm-thick SiO2 dielectric layer (capacitance per unit area Ci = 10 nF cm−2) were employed as a gate dielectric. To grow SAMs on the wafers, the wafers were first activated by oxygen plasma at a power of 120 W under an atmosphere pressure of 300 Torr for 5 min. A crystalline OTS-treated surface on an SiO2/Si substrate was obtained using the following procedure: a clean SiO2/Si substrate was spin coated with 3 mM OTS solution in anhydrous toluene, and the substrate was then treated overnight with ammonia vapor.45,46 The OTS-treated surface was rinsed with toluene, acetone, and isopropyl alcohol and then dried with nitrogen. Organic semiconductor thin films were deposited on SiO2/Si substrates through solution-shearing (SS) or spin-coating methods. Solution-sheared thin films were prepared as reported previously.26 The solution concentration for spin coating and SS deposition was 10 mg mL−1. The spin-coated films were deposited at a spin rate of 1000 rpm for 60 s, and the SS films were prepared at a shearing rate of 0.1–2 mm s−1. After solution deposition, these samples were annealed at 150 °C for 10 min in ambient conditions. Gold contacts (40 nm) were evaporated onto the SS and spin-coated thin films with a channel length (L) of 50 μm and a channel width (W) of 1000 μm.

Characterization

The thickness of the polymer film was measured using a Microfigure Measuring Instrument (Surfcorder ET3000, Kosaka Laboratory Ltd). Changes in the morphology of the polymer semiconductor were measured using tapping-mode Atomic Force Microscopy (AFM) (Innova, Bruker). The force constant of the used AFM tip (NANOSENSORS™ PointProbe® Plus) is 42 N m−1. The electrical performance of the device was recorded in a N2-filled glovebox by using a Keithley 2634B semiconductor parametric analyzer. Grazing incidence X-ray diffraction (GIXD) patterns were measured at the National Synchrotron Radiation Research Center on beamline 13A and 17A in Hsinchu. The angle of incidence was fixed at 0.12° to enhance the diffraction intensity and reduce substrate scattering.

Results and discussion

Surface energy characterization

We first evaluated the surface properties of these interfaces by measuring the contact angles of the SiO2/Si substrates modified with OTS and PBTS, as shown in Fig. 1(a). Furthermore, bare SiO2/Si substrates were evaluated for a control experiment. According to Young's equation, the surface energy of the modified and bare substrates can be estimated from the water contact angles as follows:
 
γs = γl[thin space (1/6-em)]cos[thin space (1/6-em)]θ + γsl(1)
where γs is the gas–solid interfacial energy, γl is the gas–liquid interfacial energy, and γsl is the solid–liquid interfacial energy. γsl can be replaced by the surface free energy of the solid (γs) and the liquid (γl) by using dispersion and polar components. To measure the surface free energy of these substrates, we employed water and CH2I2 to estimate dispersion and polar components. The surface energy and contact angles of the OTS- and PBTS-modified SiO2/Si substrates and bare wafers are summarized in Table 1. The OTS, PBTS, and bare substrates had water contact angles of 112°, 104°, and 36°, respectively. Moreover, the OTS, PBTS, and bare substrates had diiodomethane contact angles of 57.93°, 54.38°, and 39.07°, respectively. According to eqn (1), the surface energies of the OTS, PBTS, and bare substrates were 29.91, 31.83, and 65.77 mJ m−2. Among these substrates, the OTS monolayer exhibited the lowest surface energy and a high dewetting ability.

image file: c8tc01439f-f1.tif
Fig. 1 (a) Chemical structures of the SAM-modified and bare substrates, and images of their water contact angles. (b) Schematic structure of the setup of solution shearing on the substrate. The inset figure represents the chemical structure of PBDT-co-TT.
Table 1 Summary of the surface properties of the substrates
Interface Water contact angle (°) Diidomethane contact angle (°) Surface energy (mJ m−2)
OTS 112 57.93 29.91
PBTS 104 54.38 31.83
Bare 36 39.07 65.77


Organic thin-film transistor characterization

By using the SS technique, polymer thin films were deposited on the prepared substrates for device fabrication. Fig. 2 and Fig. S1, S2 (ESI) show the characteristics of OTFTs prepared on the OTS, PBTS, and bare SiO2/Si substrates. The OTFTs are fabricated using the bottom-gate top-contact structure. The devices exhibited typical transfer curves and well-defined output characteristics. The saturation regime mobility of the devices was estimated using the following equation:
 
image file: c8tc01439f-t1.tif(2)
where IDS is the drain current, VG is the gate voltage, VTH is the threshold voltage, μ is the linear mobility, W is the channel width, L is the channel length, and C is the capacitance per unit area of the polymer dielectric layer (10 nF cm−2). The detailed information of OTFTs prepared at various shearing speeds is listed in Table 2. The typical transfer curves we obtained showed a gate-voltage-dependent behavior. The dependence of mobilities on gate voltages likely arises from field-related contact resistance and ambipolarity (hole and electron injection).47–51 Therefore, we calculated apparent mobilities at low (0 to −20 V) and high gate voltages (−40 to −60 V). The details of the mobility estimation of the solution-sheared devices with different speeds are shown in Fig. S1 and Table S1 in the ESI. From the mobility-VG curves, the devices manifest decreased mobilities with increased gate voltages. The OTS-based devices made at the shearing speeds of 0.1, 0.4 and 1 mm s−1 showed charge mobilities of 0.50, 0.80 and 0.48 cm2 V−1 s−1 at small gate voltages, respectively, and different charge mobilities of 0.04, 0.14 and 0.08 cm2 V−1 s−1 at high gate voltages, respectively. Regardless of the gate voltage range, the devices made at the shearing speeds of 0.3–0.5 mm s−1 commonly exhibited the highest apparent mobilities. Additionally, we found that the PBTS-based devices exhibited less mobility dependence on gate voltage, as compared to the bare and OTS-based devices (Fig. S3, ESI). The reason for the greater stability and reliability of the PBTS-based devices is still not clear. Further studies are required in order to investigate the relationship between the interfaces and the gate voltage dependence. Note that the mobility values discussed in this work are mainly based on the values estimated from the small gate voltages.

image file: c8tc01439f-f2.tif
Fig. 2 (a–c) Transfer curves (VDS = −80 V) and (d–f) output characteristics of solution-sheared PBDT-co-TT field-effect transistors based on different interfaces, where the shearing speeds ranged from 0.1 to 2 mm s−1.
Table 2 Summary of solution-sheared PBDT-co-TT field-effect transistors based on different interfaces. Note that the mobilities were estimated from the region of small gate voltages (0 to −20 V)
Bare SiO2 substrate PBTS-modified SiO2 substrate OTS-modified SiO2 substrate
Speed (mm s−1) μ avg (cm2 V−1 s−1) On/off V avgth (V) Speed (mm s−1) μ avg (cm2 V−1 s−1) On/off V avgth (V) Speed (mm s−1) μ avg (cm2 V−1 s−1) On/off V avgth (V)
0.1 (4.74 ± 1.4) × 10−2 3 × 104 1.59 ± 2.9 0.1 (2.14 ± 0.6) × 10−1 9 × 104 −0.72 ± 6.4 0.1 (5.17 ± 1.1) × 10−1 2 × 104 9.23 ± 2.3
0.2 (3.46 ± 0.2) × 10−2 7 × 104 3.58 ± 1.7 0.2 (1.84 ± 0.4) × 10−1 3 × 105 3.21 ± 2.9 0.2 (6.99 ± 0.8) × 10−1 6 × 104 8.37 ± 1.5
0.3 (5.49 ± 1.2) × 10−2 1 × 105 1.15 ± 2.5 0.3 (2.55 ± 0.3) × 10−1 4 × 105 2.54 ± 2.8 0.3 (7.20 ± 1.4) × 10−1 1 × 104 11.4 ± 5.7
0.4 (6.44 ± 1.8) × 10−2 6 × 104 −0.74 ± 4.8 0.4 (2.29 ± 0.3) × 10−1 5 × 105 2.86 ± 3.0 0.4 (8.62 ± 1.9) × 10−1 4 × 104 8.88 ± 4.4
0.5 (4.40 ± 1.0) × 10−2 3 × 105 −1.84 ± 5.6 0.5 (3.37 ± 0.7) × 10−1 5 × 104 3.33 ± 5.4 0.5 (7.77 ± 1.9) × 10−1 6 × 104 4.58 ± 2.6
1 (6.20 ± 1.6) × 10−2 6 × 105 0.78 ± 5.2 1 (1.65 ± 0.5) × 10−1 3 × 105 3.39 ± 5.0 1 (4.82 ± 1.6) × 10−1 2 × 105 3.16 ± 6.6
1.5 (1.90 ± 0.7) × 10−2 1 × 105 −2.97 ± 8.1 1.5 (1.77 ± 0.7) × 10−1 3 × 105 1.43 ± 4.4 1.5 (2.09 ± 0.9) × 10−1 4 × 105 −1.72 ± 4.8
2 (3.71 ± 0.8) × 10−2 1 × 105 −4.05 ± 7.6 2 (7.66 ± 2.6) × 10−2 1 × 106 −6.75 ± 3.0 2 N/A N/A N/A


To investigate the effect of shearing speed, we prepared solution-sheared films by using different shearing speeds. Fig. 3 shows the relationship between the shearing speed and mobility for the bare, PBTS-modified, and OTS-modified SiO2/Si substrates. The PBTS and OTS substrates exhibited a similar behavior. The hole mobilities first increased at low shearing speed (0.1–0.4 mm s−1) and then decreased at high shearing speed (0.5–2 mm s−1). The highest performance was achieved at a shearing speed of approximately 0.4–0.5 mm s−1. The highest carrier mobility was 0.862 cm2 V−1 s−1. The Ion/Ioff ratio was 104–105. However, at high shearing speed (1.5–2 mm s−1), the devices showed significant mobility degradation. The lowest mobility of the sheared devices prepared on OTS substrates was 0.209 cm2 V−1 s−1, which is close to the values obtained from spin-coated devices.


image file: c8tc01439f-f3.tif
Fig. 3 Relationship between the average mobilities and shearing speeds for the bare, PBTS-modified, and OTS-modified SiO2/Si substrates. The mobility values were averaged from at least 6 devices in two batches.

The aforementioned polymer devices were produced using toluene as the solvent. When o-dichlorobenzene (DCB) was used as the solvent, the highest mobility of the solution-sheared devices at a small gate voltage was up to 1.77 cm2 V−1 s−1 (Fig. S4 in ESI). However, we obtained a low device yield when using a high-boiling-point DCB (180 °C), especially for a high shearing speed. Therefore, to investigate the effect of solution shearing speed on charge transport, we used toluene (boiling point = 111 °C) as the solvent. Although toluene-based devices exhibited a relatively lower mobility (maximum mobility = approximately 1 cm2 V−1 s−1), it is much easier to obtain uniform films from the toluene solution.

The best average mobility of the OTS, PBTS and bare devices was 0.862, 0.373 and 0.06 cm2 V−1 s−1, respectively. The higher performance on the OTS surface may originate from its low surface energy (29.91 mJ m−2). Typically, a surface with lower surface energy provides high diffusivity and a low nucleation density of organic molecules.40 Therefore, the low surface energy may be more beneficial to the crystal growth of PBDT-co-TT on the interface. The OTS-based devices showed the highest performance, which may be assisted by its lowest surface energy. Furthermore, this enhanced performance may be attributed to the crystalline structure of the OTS molecules as well. The spin-coated OTS monolayer could form a highly ordered hexagonal lattice.39 Furthermore, crystalline SAMs result in less defects and the high crystal growth of organic semiconductors. However, the PBTS monolayer is amorphous SAM, because no significant crystalline signals were observed from the PBTS monolayer.

The interface affects not only the charge carrier mobility but also the yield of solution-deposited devices. Although the OTS surface can exhibit high performance, the highly dewetting OTS surface usually causes difficulty in polymer film formation. The yield of OTS devices is typically lower than 50%, even at a low shearing rate (0.3–0.5 mm s−1). To increase the yield, we etched the edge of the OTS substrates by using oxygen plasma to define the OTS area. The detailed patterning processes are described in Fig. S5 (ESI). We covered with a PDMS stamp to protect the OTS layer. The uncovered OTS area will be etched by plasma. The toluene and o-dichlorobenzene contact angles of OTS were 42° and 37°, respectively. After the plasma treatment, the toluene and o-dichlorobenzene contact angles of the uncovered area were 15 and 13°, respectively. Due to the reduced contact angles, the solvent can fully cover the surface of the substrates after patterning the OTS substrates (Fig. S5 in the ESI). The OTS device yield can increase up to 90% at the low shearing rate by using the patterning method. Although the patterning method can reduce the difficulty of polymer film formation at a low shearing rate, once the shearing speed is greater than 1.5 mm s−1, the yield of OTS devices decreases considerably. Compared with the OTS surface, it is much easier to shear a uniform film on the PBTS and bare surfaces, especially for the PBTS surface. This may be because of the phenyl end group of the PBTS molecule, which provides great affinity to the used organic solvent toluene. A toluene contact angle of less than 10° was observed for the PBTS surface. Therefore, the PBTS surface can be potentially used for high-speed shearing, which is beneficial for mass production.

The devices fabricated on the bare substrate exhibited the lowest performance. Despite the high wetting capability and uniform film quality, the mobility of the devices fabricated on the bare substrates exhibited a low mobility of 0.06 cm2 V−1 s−1, which is 10 times lower than that of the OTS devices. The low performance may be because of hydroxyl groups on the bare SiO2 surface, which can act as charge-trapping sites, leading to decreased charge transport properties. In addition, we did not observe significant influence of shearing speeds on charge carrier mobility in the bare SiO2 devices, indicating that interfacial modification is necessary to obtain high-performance devices.

The transfer and output characteristics of the spin-coated devices made on the OTS substrate are shown in Fig. S6 in the ESI. The devices showed an average mobility of 0.34 cm2 V−1 s−1, which is comparable to the values reported in the literature. In the literature, mobilities of the spin-coated PDBT-co-TT devices have been reported in the range from 0.2 to 10.5 cm2 V−1 s−1 depending on process parameters, molecular weight and device structure. Li et al. first demonstrated PDBT-co-TT-based OTFTs annealed at 200 °C showing a mobility of 0.94 cm2 V−1 s−1.52 DeLongchamp, McCulloch and coworkers have investigated the crystalline structure and charge mobility of the same polymer. The linear and saturation mobilities obtained from PDBT-co-TT were 0.2 and 0.38 cm2 V−1 s−1.53 Sirringhaus and coworkers used a top-gate bottom-contact structure combined with electrode treatment and high temperature annealing (200 to 320 °C) to obtain enhanced hole mobilities of up to 1.36 cm2 V−1 s−1.54 Subsequently, Ong, Liu and coworkers further optimized device performance and employed the polymer with an ultrahigh molecular weight of around 500[thin space (1/6-em)]000. They improved the mobility to more than 10 cm2 V−1 s−1.55 Although several approaches have been reported to improve the charge transport of PDBT-co-TT, these approaches mainly focus on device structure, molecular weight and electrode treatment. In this work, we investigate the effect of interface modification on the charge transport in solution-sheared polymer devices, which may be beneficial for the development of continuous roll-to-roll solution deposited devices.

Additionally, we found that the Vth and turn-on voltages of the bare, PBTS- and OTS-modified devices showed significant negative shifts with increasing shearing speeds. From Table 2, the Vth of the bare devices was shifted from 1.59 V to −4.05 V, the Vth of the PBTS-based devices was shifted from −0.72 V to −6.75 V, and the Vth of the OTS-based devices was shifted from 9.23 V to −1.72 V. The shift of the Vth in the negative direction may be related to the decrease of the film thicknesses. During solution shearing, the increased shearing speed can significantly reduce film thickness. At slow shearing speed, the polymer tends to form a thick film. However, the thick film may tend to form aggregates which can act as potential defects for charges. This may occur unintentionally in the thick film, resulting in the increase of free charge density in the bulk semiconductor layer. Furthermore, the on/off current ratios increased with increased speeds. The improved on/off current may be also attributed to the reduced defects in the thinner polymer semiconductor layer. A similar thickness effect has been observed in poly(3-hexylthiophene) (P3HT).56,57 Deen et al. have found that the film thickness has significant influence on the electrical properties of P3HT. The thicker P3HT film showed more positive threshold voltages and smaller on/off current ratios.

The optimized shearing speed (0.4–0.5 mm s−1) was similar for the OTS- and PBTS-modified surfaces. To deeply understand the effect of shearing speeds on these interfaces, we employed a surface profiler, UV-vis spectra, the atomic force image technique, and GIXD for further investigations.

Thickness analysis of solution-sheared films

To characterize the formation mechanism of film growth during solution deposition, we employed a surface profiler to analyze the thickness curves of the substrates. For the solution meniscus-guided deposition, there are two regimes of film deposition: the evaporation regime and the Landau–Levich regime.24,58 The typical thickness curve of the evaporation regime shows a decreased thickness with an increased shearing speed. In this regime, the solute tends to accumulate near the meniscus at the air–liquid interface and forms a film behind the meniscus. The deposition of polymer films is directly affected by the movement of the shearing blade. Therefore, the shear effect may facilitate the orientation of the polymer chains. By contrast, the Landau–Levich regime is characterized by the increased film thickness with an increase in the shearing speed. Typically, the Landau–Levich regime is predominated by the viscous force. In this regime, a liquid layer is first deposited on a substrate at a high shearing speed and then precipitated afterward. The crystal growth mechanism of this film is similar to that of a drop-cast film. In the Landau–Levich regime, the shear effect is not expected. Polymer chains may tend to release their chain orientation after drying.

From the thickness curves (Fig. 4), we observed different behaviors between the bare and SAM-modified substrates. On SAM-modified substrates, the thickness of the film showed a power law dependence hvα, where h is the film thickness, v is the shearing speed, and α is an exponent of the thickness dependence. Typically, negative (approximately −1) and positive (0.76) exponents represent the evaporation and Landau–Levich regime, respectively.24,58 The exponents of the OTS- and PBTS-based samples fitted to the shearing rate from 0.1 to 2 mm s−1 are −0.94 (R2 = 0.93) and −0.88 (R2 = 0.99). The detailed fitting curves of these interfaces are shown in Fig. S7 in the ESI. This indicates that both the PBTS and OTS films were grown in the evaporation regime. The exponent of the sample prepared on the bare substrate was −0.86 (R2 = 0.97) at shearing speeds ranging from 0.1 to 0.4 mm s−1. At a shearing speed of more than 0.5 mm s−1, the thickness curve becomes flat, thereby resulting in inaccurate fitting. This flattened curve may be affected by the transition between the evaporation and Landau–Levich regimes.58 On the bare substrate, the films grew thicker than expected, which may be attributed to the hydroxyl groups of the bare substrate with high surface energy, thus causing increased viscous forces near the interface. According to the aforementioned thickness analysis, we can conclude that the OTS-modified substrate provides an ideal interface, which has the exponent of −1 derived from the assumption of simple mass balance near the meniscus. In contrast, the bare substrate results in a less slipped surface for film growth around the meniscus.


image file: c8tc01439f-f4.tif
Fig. 4 Thicknesses of the solution-sheared films on the OTS-modified, PBTS-modified, and bare substrates at various shearing speeds.

UV-vis spectra and polymer alignment characterization

UV-vis spectra were employed to investigate the electronic band structures of the solution-sheared films. Fig. 5 shows the UV-vis spectra of polymer films prepared on the OTS substrate. The information of the UV-vis spectra of the polymer films on the PBTS-based and bare substrates is summarized in Fig. S8 in the ESI. We observed that the peaks of the UV-vis absorption spectra were almost identical, even at various shearing speeds. However, the peaks of the solution-sheared polymer films became narrower at a high shearing speed. With an increase in the shearing speed, the band gaps also increased. On the bare surface, as the shearing speed increased from 0.1 to 2 mm s−1, the band gaps increased from 1.31 to 1.39 eV. The increased band gaps and narrow bands in solution-sheared films may be related to the thickness of the solution-sheared films. At a low shearing speed of 0.1 mm s−1, the thicknesses of the films on the bare, PBTS, and OTS substrates were 836, 562, and 550 nm (Fig. 5), respectively. At a high shearing speed of 2 mm s−1, the average thicknesses of the films on the bare, PBTS, and OTS substrates decreased to 114, 46, and 45 nm, respectively. The small thicknesses may significantly reduce interaction within the polymer chains, leading to narrow bands and larger band gaps.
image file: c8tc01439f-f5.tif
Fig. 5 (a) UV-vis spectra of solution-sheared PBDT-co-TT films prepared on the OTS substrate at various shearing speeds. (b) Polarized UV-vis spectra of solution-sheared PBDT-co-TT films on OTS-modified glass substrates in parallel and perpendicular directions.

To investigate the effect of shearing speeds on the orientation of the polymer chains, we measured polarized UV-vis spectra in parallel and perpendicular directions. Fig. 5(b) shows the polarized UV-vis spectra of solution-sheared films under various speeds. We found that the UV-vis spectra of solution-sheared films in the direction parallel and perpendicular to the shearing direction showed anisotropic absorption levels compared with isotropic spin-coated films. In solution-sheared films, absorption in the parallel direction was constantly stronger than that in the perpendicular direction. This indicates that the polymer chains of PBDT-co-TT tend to be aligned with the shearing direction.

To estimate the alignment of the polymer chains, we used the dichroic ratio, which is calculated from the ratio of the parallel and perpendicular absorption (dichroic ratio (R) = A/A). Fig. 6 shows dichroic ratios at various speeds on different interfaces. At a low shearing speed of 0.1 mm s−1, no significant difference was observed in the dichroic ratio among the bare, PBTS, and OTS surfaces. Once the shearing speed was increased to 0.4 mm s−1, the dichroic ratio showed a considerable difference among these three interfaces. The OTS surface showed the highest increase in the dichroic ratio. The highest dichroic ratio of solution-sheared PBDT-co-TT films could reach 1.83 for the OTS surface. The bare substrate showed small changes in the dichroic ratio. The dichroic ratio of samples prepared on the bare surface was approximately 1.2. These results indicate that the OTS surface is a more favorable interface to improve polymer chain orientation. This may be attributed to its low surface energy, which can enhance polymer chain mobility and improve polymer alignment. In contrast, the bare surface has the largest surface energy and the greater affinity for polymer chains, leading to a low alignment of the polymer chains. In the high-speed regime (1–2 mm s−1), all samples showed a decrease in the dichroic ratio, regardless of interface. This may be due to the fact that a fast shearing speed induces a longer meniscus. The longer meniscus may induce more nuclei for crystalline growth and cause a higher generation of disoriented nuclei.24 Furthermore, the influence of the shear flow gradually decreased when the nuclei were far from the meniscus. Thus, the alignment of the polymer chains decreases.


image file: c8tc01439f-f6.tif
Fig. 6 Dichroic ratios of solution-sheared PBDT-co-TT films at various shearing speeds.

The PBTS surface exhibited the highest dichroic ratio at the speed of 1 mm s−1, which is higher than the optimized speed of 0.4 mm s−1 on the OTS surface. This is likely because of the affinity of the phenyl end groups, which have a better interaction with the aromatic polymer chains. Thus, the PBTS interface requires a higher shearing speed to align the polymer chains.

Crystalline structure and morphology characterization

Fig. 7 shows the atomic force microscopy phase images of solution-sheared PBDT-co-TT films on bare, PBTS, and OTS substrates at shearing speeds of 0.1, 0.4, and 1 mm s−1. The films based on bare and PBTS substrates showed small spherical aggregates on the surface, whereas those based on the OTS substrate exhibited a fiber-like morphology, which may be because of its higher crystallinity. For OTS-based films, the morphology produced at a low speed (0.1 mm s−1) did not exhibit clear fiber-like grains. When the shearing speed was as high as 0.4 mm s−1, polymer chains formed long fiber-like aggregates. However, at a high speed, the fiber-like aggregates became smaller and exhibited several grain boundaries, which may act as defects to suppress charge transport.
image file: c8tc01439f-f7.tif
Fig. 7 Atomic force microscopy (AFM) phase images of solution-sheared PBDT-co-TT films on the (a, c and e) OTS substrate at shearing speeds of 0.1, 0.4, and 1 mm s−1, respectively. The AFM images of solution-sheared films prepared on the bare (b) and PBTS (d) substrates. (f) AFM image of the spin-coated film made on the bare substrate. The size of the images is 1 × 1 μm. The arrow represents the direction of solution shearing.

Fig. 8 shows the GIXD patterns of solution-sheared films prepared for various speeds or substrates. The polymer tended to show a well-defined out-of-plane lamellar spacing. The (100) peak of the GIXD patterns showed a slight shift to the low q regime, indicating a longer spacing distance (Fig. 8a and b) as the shearing speed increased. When the shearing speeds increased from 0.1 to 2 mm s−1, the estimated lamellar spacing of the polymer films showed a significant increase from 19.11 to 20.19 Å, which is slightly higher than that of the spin-coated film (20.07 Å; Fig. 8c). According to the Scherrer equation, the coherence length can be estimated from the full width at half maximum of the GIXD out-of-plane peaks (Fig. 8d). At a shearing speed of 0.4 mm s−1, the coherence lengths of the polymer showed the highest values in all substrates (Fig. 8e). Furthermore, the polymer formed the largest crystal size (18 nm) on the OTS surface, higher than that on the PBTS surface (16.5 nm) and the bare substrate (15.1 nm). In addition, lamellar spacing changes were more significant in the parallel direction than in the perpendicular direction. The π–π stacking distance (approximately 3.7 Å) did not show any significant change with increasing shearing speeds.


image file: c8tc01439f-f8.tif
Fig. 8 (a) GIXD 2D patterns of solution-sheared PBDT-co-TT films on OTS-modified silicon substrates under various shearing speeds. The red dashed line indicates the (100) Bragg peak position of the film sheared at 0.1 mm s−1; (b) integrated GIXD 1D peaks of solution-sheared films on OTS-modified silicon substrates under various shearing speeds; (c) (100) out-of-plane lamellar spacing of solution-sheared polymer films prepared at different shearing speeds (solid circle) and the lamellar spacing of the spin-coated polymer film (hollow circle), where all the films were coated on OTS substrates; (d) one-dimensional integrated GIXD patterns of solution-sheared PBDT-co-TT films on OTS-modified, PBTS, and bare substrates prepared at shearing speeds of 0.4–0.5 mm s−1; (e) coherence lengths of solution-sheared PBDT-co-TT films on OTS-modified, PBTS-modified, and bare substrates under different shearing speeds.

The solution shearing process could tune a metastable state of the crystalline structure of organic semiconductors.31 Giri et al. reported that the lamellar spacing of the interdigitated polymer PBTTT also showed a significant increase with an increasing shearing speed. They found that this is mainly because of changes in side-chain interdigitation instead of changes in the tilt of the polymer backbone.33 The increase in shearing speed results in increased disorder in the side chains and larger spacing. The distance of the longest side chains fully extended from the backbone of PBDT-co-TT was approximately 18.5 Å.59 The lamellar spacing without interdigitation should be more than 3 nm.60 However, our lamellar spacing ranged from 19 to 20.19 Å. Hence, the PBDT-co-TT films should be partially interdigitated. During solution shearing, the meniscus-guided flow promoted the disentanglement of the polymer chains. This induced reorganization and alignment of the polymer chains toward a metastable structure, resulting in an enhanced crystalline coherence length. However, with further increase in shearing speeds, the increase in the disorder of the side chains suppressed the molecular packing, leading to a decreased coherence length. Notably, the trend of decreasing coherence length is similar to the dependence of the dichroic ratio and charge mobility on shearing speeds. This indicates that shearing speeds not only influence metastable molecular packing but also the orientation and charge transport of the polymer chains. However, the dependence of the dichroic ratio and charge transport on shearing speeds was not clearly observed for the bare-SiO2-based devices. This indicates that the low-surface-energy interface may be essential for shearing effects on charge transport.

Conclusions

In this study, we demonstrated the effects of solution shearing on the charge transport, polymer orientation, and molecular packing in different interfaces, including the bare SiO2 wafer, PBTS-modified, and OTS-modified substrates for OTFT applications. The OTS-treated surface with a low surface energy can manipulate the metastable molecular packing of polymer chains. The charge transport and polymer chain orientation could be further improved through solution shearing on OTS interfaces. By contrast, the effects of shearing speeds on bare SiO2-based devices are not significant. From the thickness analysis, the OTS interface is more favorable compared with the bare SiO2 interface. The evaporation regime can be clearly observed from polymer films prepared on OTS substrates. Among these interfaces, the PBTS surface exhibited higher wetting ability for the polymer solution. This indicates that the PBTS surface can potentially be employed for high-speed processes because of its affinity for organic solvents, which is beneficial for solution deposition. For the production of next-generation flexible electronics, the solution shearing process is a promising method for the low-cost mass production of polymer-based devices. Thus, this study investigated interfacial effects on solution shearing processes, which may be applied for the optimization of polymer-based electronics with highly aligned and superior charge transport properties.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

W.-Y. L. acknowledges funding support from the Ministry of Science and Technology of the Republic of China (Grant No. 104-2218-E-027-007-MY3) and X-ray scattering equipment support from National Synchrotron Radiation Research Center on beamline 13A and 17A in Hsinchu. This manuscript was edited by Wallace Academic Editing.

References

  1. F. C. Krebs, T. Tromholt and M. Jorgensen, Nanoscale, 2010, 2, 873 RSC.
  2. M. Helgesen, R. Sondergaard and F. C. Krebs, J. Mater. Chem., 2010, 20, 36 RSC.
  3. P. J. Diemer, C. R. Lyle, Y. Mei, C. Sutton, M. M. Payne, J. E. Anthony, V. Coropceanu, J.-L. Brédas and O. D. Jurchescu, Adv. Mater., 2013, 25, 6956 CrossRef PubMed.
  4. D. H. Kim, J. T. Han, Y. D. Park, Y. Jang, J. H. Cho, M. Hwang and K. Cho, Adv. Mater., 2006, 18, 719 CrossRef.
  5. S. Wang, M. Kappl, I. Liebewirth, M. Müller, K. Kirchhoff, W. Pisula and K. Müllen, Adv. Mater. Processes, 2012, 24, 417 CrossRef PubMed.
  6. L. L. Chua, P. K. H. Ho, H. Sirringhaus and R. H. Friend, Adv. Mater., 2004, 16, 1609 CrossRef.
  7. L. Qiu, J. A. Lim, X. Wang, W. H. Lee, M. Hwang and K. Cho, Adv. Mater., 2008, 20, 1141 CrossRef.
  8. C. Liu, Y. Li, M. V. Lee, A. Kumatani and K. Tsukagoshi, Phys. Chem. Chem. Phys., 2013, 15, 7917 RSC.
  9. X. Ding, J. Liu and T. A. L. Harris, AIChE J., 2016, 62, 2508 CrossRef.
  10. H. N. Tsao, D. Cho, J. W. Andreasen, A. Rouhanipour, D. W. Breiby, W. Pisula and K. Müllen, Adv. Mater., 2009, 21, 209 CrossRef.
  11. A. Salleo, M. L. Chabinyc, M. S. Yang and R. A. Street, Appl. Phys. Lett., 2002, 81, 4383 CrossRef.
  12. J. Z. Wang, Z. H. Zheng, H. W. Li, W. T. S. Huck and H. Sirringhaus, Nat. Mater., 2004, 3, 171 CrossRef PubMed.
  13. H. Minemawari, T. Yamada, H. Matsui, J. y. Tsutsumi, S. Haas, R. Chiba, R. Kumai and T. Hasegawa, Nature, 2011, 475, 364 CrossRef PubMed.
  14. C.-Y. Chen, H.-W. Chang, Y.-F. Chang, B.-J. Chang, Y.-S. Lin, P.-S. Jian, H.-C. Yeh, H.-T. Chien, E.-C. Chen, Y.-C. Chao, H.-F. Meng, H.-W. Zan, H.-W. Lin, S.-F. Horng, Y.-J. Cheng, F.-W. Yen, I.-F. Lin, H.-Y. Yang, K.-J. Huang and M.-R. Tseng, J. Appl. Phys., 2011, 110, 094501 CrossRef.
  15. D. Khim, H. Han, K.-J. Baeg, J. Kim, S.-W. Kwak, D.-Y. Kim and Y.-Y. Noh, Adv. Mater., 2013, 25, 4302 CrossRef PubMed.
  16. W.-T. Park and Y.-Y. Noh, J. Mater. Chem. C, 2017, 5, 6467 RSC.
  17. Z. Zhang, B. Peng, X. Ji, K. Pei and P. K. L. Chan, Adv. Funct. Mater., 2017, 27, 1703443 CrossRef.
  18. C.-Y. Liu and A. J. Bard, Chem. Mater., 2000, 12, 2353 CrossRef.
  19. H. A. Becerril, M. E. Roberts, Z. H. Liu, J. Locklin and Z. N. Bao, Adv. Mater., 2008, 20, 2588 CrossRef.
  20. W.-Y. Lee, J. H. Oh, S.-L. Suraru, W.-C. Chen, F. Würthner and Z. Bao, Adv. Funct. Mater., 2011, 21, 4173 CrossRef.
  21. Y. Diao, B. C. K. Tee, G. Giri, J. Xu, D. H. Kim, H. A. Becerril, R. M. Stoltenberg, T. H. Lee, G. Xue, S. C. B. Mannsfeld and Z. Bao, Nat. Mater., 2013, 12, 665 CrossRef PubMed.
  22. B. D. Naab, S. Himmelberger, Y. Diao, K. Vandewal, P. Wei, B. Lussem, A. Salleo and Z. Bao, Adv. Mater., 2013, 25, 4663 CrossRef PubMed.
  23. W.-Y. Lee, G. Giri, Y. Diao, C. J. Tassone, J. R. Matthews, M. L. Sorensen, S. C. B. Mannsfeld, W.-C. Chen, H. H. Fong, J. B. H. Tok, M. F. Toney, M. He and Z. Bao, Adv. Funct. Mater., 2014, 24, 3524 CrossRef.
  24. L. Shaw, P. Hayoz, Y. Diao, J. A. Reinspach, J. W. To, M. F. Toney, R. T. Weitz and Z. Bao, ACS Appl. Mater. Interfaces, 2016, 8, 9285 Search PubMed.
  25. Y. Diao, L. Shaw, Z. Bao and S. C. B. Mannsfeld, Energy Environ. Sci., 2014, 7, 2145 Search PubMed.
  26. H. A. Becerril, M. E. Roberts, Z. Liu, J. Locklin and Z. Bao, Adv. Mater., 2008, 20, 2588 CrossRef.
  27. M. Niwa, H. Kawakami, T. Kanamori, T. Shinbo, A. Kaito and S. Nagaoka, Macromolecules, 2001, 34, 9039 CrossRef.
  28. P. Schilinsky, C. Waldauf and C. J. Brabec, Adv. Funct. Mater., 2006, 16, 1669 CrossRef.
  29. S. Ito, Y. Makari, T. Kitamura, Y. Wada and S. Yanagida, J. Mater. Chem., 2004, 14, 385 RSC.
  30. L. J. Lanticse, Y. Tanabe, K. Matsui, Y. Kaburagi, K. Suda, M. Hoteida, M. Endo and E. Yasuda, Carbon, 2006, 44, 3078 CrossRef.
  31. G. Giri, E. Verploegen, S. C. B. Mannsfeld, S. Atahan-Evrenk, D. H. Kim, S. Y. Lee, H. A. Becerril, A. Aspuru-Guzik, M. F. Toney and Z. Bao, Nature, 2011, 480, 504 CrossRef PubMed.
  32. Y. Diao, B. C. Tee, G. Giri, J. Xu, D. H. Kim, H. A. Becerril, R. M. Stoltenberg, T. H. Lee, G. Xue, S. C. Mannsfeld and Z. Bao, Nat. Mater., 2013, 12, 665 CrossRef PubMed.
  33. G. Giri, D. M. DeLongchamp, J. Reinspach, D. A. Fischer, L. J. Richter, J. Xu, S. Benight, A. Ayzner, M. He, L. Fang, G. Xue, M. F. Toney and Z. Bao, Chem. Mater., 2015, 27, 2350 CrossRef.
  34. J. Lee, A. R. Han, J. Kim, Y. Kim, J. H. Oh and C. Yang, J. Am. Chem. Soc., 2012, 134, 20713 CrossRef PubMed.
  35. D. Ji, X. Xu, L. Jiang, S. Amirjalayer, L. Jiang, Y. Zhen, Y. Zou, Y. Yao, H. Dong, J. Yu, H. Fuchs and W. Hu, J. Am. Chem. Soc., 2017, 139, 2734 CrossRef PubMed.
  36. G. Kim, S.-J. Kang, G. K. Dutta, Y.-K. Han, T. J. Shin, Y.-Y. Noh and C. Yang, J. Am. Chem. Soc., 2014, 136, 9477 CrossRef PubMed.
  37. Y. Jiang, Y. Gao, H. Tian, J. Ding, Y. Geng and F. Wang, Acta Polym. Sin., 2017, 7, 1141 Search PubMed.
  38. Y. Jiang, Y. Gao, H. Tian, J. Ding, D. Yan, Y. Geng and F. Wang, Macromolecules, 2016, 49, 2135 CrossRef.
  39. Y. Ito, A. A. Virkar, S. Mannsfeld, J. H. Oh, M. Toney, J. Locklin and Z. A. Bao, J. Am. Chem. Soc., 2009, 131, 9396 CrossRef PubMed.
  40. A. Virkar, S. Mannsfeld, J. H. Oh, M. F. Toney, Y. H. Tan, G. Y. Liu, J. C. Scott, R. Miller and Z. Bao, Adv. Funct. Mater., 2009, 19, 1962 CrossRef.
  41. M. C. LeMieux, M. Roberts, S. Barman, Y. W. Jin, J. M. Kim and Z. Bao, Science, 2008, 321, 101 CrossRef PubMed.
  42. E. Mohammadi, C. Zhao, Y. Meng, G. Qu, F. Zhang, X. Zhao, J. Mei, J.-M. Zuo, D. Shukla and Y. Diao, Nat. Commun., 2017, 8, 16070 CrossRef PubMed.
  43. S. Kobayashi, T. Nishikawa, T. Takenobu, S. Mori, T. Shimoda, T. Mitani, H. Shimotani, N. Yoshimoto, S. Ogawa and Y. Iwasa, Nat. Mater., 2004, 3, 317 CrossRef PubMed.
  44. D. H. Kim, H. S. Lee, H. Yang, L. Yang and K. Cho, Adv. Funct. Mater., 2008, 18, 1360 Search PubMed.
  45. Y. Ito, A. A. Virkar, S. Mannsfeld, J. H. Oh, M. Toney, J. Locklin and Z. Bao, J. Am. Chem. Soc., 2009, 131, 9396 CrossRef PubMed.
  46. A. Virkar, S. Mannsfeld, J. H. Oh, M. F. Toney, Y. H. Tan, G.-Y. Liu, J. C. Scott, R. Miller and Z. Bao, Adv. Funct. Mater., 2009, 19, 1962 CrossRef.
  47. Y. Yu, Y. Wu, A. Zhang, C. Li, Z. Tang, W. Ma, Y. Wu and W. Li, ACS Appl. Mater. Interfaces, 2016, 8, 30328 Search PubMed.
  48. J. Du, C. Bulumulla, I. Mejia, G. T. McCandless, M. C. Biewer and M. C. Stefan, Polym. Chem., 2017, 8, 6181 RSC.
  49. I. McCulloch, A. Salleo and M. Chabinyc, Science, 2016, 352, 1521 CrossRef PubMed.
  50. E. G. Bittle, J. I. Basham, T. N. Jackson, O. D. Jurchescu and D. J. Gundlach, Nat. Commun., 2016, 7, 10908 CrossRef PubMed.
  51. H. Sirringhaus, Adv. Mater., 2014, 26, 1319 CrossRef PubMed.
  52. Y. Li, S. P. Singh and P. Sonar, Adv. Mater., 2010, 22, 4862 CrossRef PubMed.
  53. X. Zhang, L. J. Richter, D. M. DeLongchamp, R. J. Kline, M. R. Hammond, I. McCulloch, M. Heeney, R. S. Ashraf, J. N. Smith, T. D. Anthopoulos, B. Schroeder, Y. H. Geerts, D. A. Fischer and M. F. Toney, J. Am. Chem. Soc., 2011, 133, 15073 CrossRef PubMed.
  54. Z. Chen, M. J. Lee, R. S. Ashraf, Y. Gu, S. Albert-Seifried, M. M. Nielsen, B. Schroeder, T. D. Anthopoulos, M. Heeney, I. McCulloch and H. Sirringhaus, Adv. Mater., 2012, 24, 647 CrossRef PubMed.
  55. J. Li, Y. Zhao, H. S. Tan, Y. Guo, C.-A. Di, G. Yu, Y. Liu, M. Lin, S. H. Lim, Y. Zhou, H. Su and B. S. Ong, Sci. Rep., 2012, 2, 754 CrossRef PubMed.
  56. M. J. Deen, M. H. Kazemeini, Y. M. Haddara, Y. Jianfei, G. Vamvounis, S. Holdcroft and W. Woods, IEEE Trans. Electron Devices, 2004, 51, 1892 CrossRef.
  57. H. Jia, S. Gowrisanker, G. K. Pant, R. M. Wallace and B. E. Gnade, J. Vac. Sci. Technol., A, 2006, 24, 1228 Search PubMed.
  58. M. Le Berre, Y. Chen and D. Baigl, Langmuir, 2009, 25, 2554 CrossRef PubMed.
  59. J. Li, Y. Zhao, H. S. Tan, Y. Guo, C.-A. Di, G. Yu, Y. Liu, M. Lin, S. H. Lim, Y. Zhou, H. Su and B. S. Ong, Sci. Rep., 2012, 2, 754 CrossRef PubMed.
  60. R. J. Kline, D. M. DeLongchamp, D. A. Fischer, E. K. Lin, L. J. Richter, M. L. Chabinyc, M. F. Toney, M. Heeney and I. McCulloch, Macromolecules, 2007, 40, 7960 CrossRef.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c8tc01439f

This journal is © The Royal Society of Chemistry 2018