Tuning Transition-Metal Dichalcogenide Field-Effect-Transistor by

Subscriber access provided by Kaohsiung Medical University

Article

Tuning Transition-Metal Dichalcogenide Field-Effect-Transistor by Spontaneous Pattern Formation of an Ultrathin Molecular Dopant Film Hisashi Ichimiya, Masahiro Takinoue, Akito Fukui, Kohei Miura, Takeshi Yoshimura, Atsushi Ashida, Norifumi Fujimura, and Daisuke Kiriya ACS Nano, Just Accepted Manuscript • Publication Date (Web): 14 Sep 2018 Downloaded from http://pubs.acs.org on September 14, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Tuning Transition-Metal Dichalcogenide FieldEffect-Transistor by Spontaneous Pattern Formation of an Ultrathin Molecular Dopant Film Hisashi Ichimiya1, Masahiro Takinoue2, Akito Fukui1, Kohei Miura1, Takeshi Yoshimura1, Atsushi Ashida1, Norifumi Fujimura1 and Daisuke Kiriya1,3*

1

Department of Physics and Electronics, Osaka Prefecture University, 1-1 Gakuen-cho, Naka-

ku, Sakai-shi, Osaka 599-8531, Japan 2

Department of Computer Science, Tokyo Institute of Technology, Yokohama, Japan

3

PRESTO, Japan Science and Technology Agency (JST), 4-1-8 Honcho, Kawaguchi, Saitama

332-0012, Saitama, Japan

Corresponding author: Daisuke Kiriya, *[email protected]

KEYWORDS: transition-metal dichalcogenides, metal-oxide-semiconductor field-effecttransistor, spinodal dewetting, spontaneous pattern formation, molecular doping, charge transfer

ACS Paragon Plus Environment

1

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 22

Abstract

Spontaneous pattern formation is an energetically favorable process and is shown in nature in molecular-scale assembly, biological association, and soft material organizations. The opposite regime, the artificial process, which is widely applied to the fabrication of semiconducting devices, such as lithographic techniques, requires enormous amounts of energy. Here we propose a concept of tuning the properties of semiconducting MoS2 and WSe2 devices using the spontaneous pattern formation of adjacent molecular films. The film used was a 10 nm-thick ultrathin film of a molecular electron dopant, which exhibited spontaneous pattern formation and which dynamically transformed the morphology of tiny holes, network, maze, and dots on substrates including SiO2, MoS2 and WSe2. These patterns were exhibited only when the film came in contact with water and tuned with temperature and time. The specific lengths of the patterns were less than 200 nm, which is sufficiently smaller than the exfoliated ~10 µm semiconducting MoS2 and WSe2 flakes. The properties of the field-effect devices of MoS2 and WSe2 were found to be modified according to the pattern formation process of the ultrathin molecular film on the device. This concept aims to applying the spontaneous patterning phenomena shown in nature to the fabrication and optimization of electronic devices by using molecular films and their responses to the external environment.


2

Page 3 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Spontaneous pattern formation is found in nature, from molecular assembled patterns1-3 and fine-cell aggregated patterns forming tissues4, 5 to skin patterns on organisms such as zebra on a macroscopic scale.6-8 In addition to the biological patterns, non-living phenomena such as assemblies of colloidal particles9, 10 and fluids11-13 also exhibit spontaneous patterns. The essence of spontaneous pattern formation is the dynamic conversion of the system under perturbation continually to form a spatial regularity. The entire process is carried out in the direction of decreasing the free energy of the system.9, 14 With the spontaneous pattern formation having been extensively studied as a natural system, the next challenging step is to adapt this bottom-up process to engineering applications.9, 15-17 An example is electronics, which have been developed based on an artificial pattern formation process using top-down technology. Semiconducting materials are an essential element of electronic devices; the key to the architecture of these devices is the junctions consisting of heterogeneous components. Junctions between semiconductors of different polarities forming a pn junction or semiconductors of different carrier concentrations, such as n+/n or p+/p, are important components of transistors and optical devices.18 These junctions are fabricated via lithographic techniques using artificial top-down technology requiring an enormous amount of energy.19

The challenges of combining spontaneous pattern formation in the fabrication of electronics (semiconducting materials/devices) are as follows: 1) the process is limited to which material systems can exhibit an effective combination of the two formation processes; 2) the patterns should be less than several hundred nm in length to be applicable to electronic devices; and 3) the patterns must be stable under ambient conditions but also easily tuned under accessible perturbation. In this study, we propose a spontaneous pattern formation system that optimizes semiconducting materials and devices (Fig. 1). We demonstrate an ultrathin film of an


3

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 22

organic molecular dopant, benzyl viologen (BV), which is a hydrophobic molecule and a strong electron dopant,20,

21

and which reveals various pattern formations on substrates (SiO2/Si

substrate or semiconducting materials) by forming an interface with water. The patterns are variously transformed from tiny holes, a network, a maze and isolated dots. The size of each pattern can be tuned on the micrometer scale to less than 200 nm by changing the initial thickness of the BV molecular film. By applying the spontaneous pattern formation to semiconductor surfaces, we demonstrated that the pattern dynamics directly convert the device characteristics. This work proposes a concept of a spontaneous device-architecture formation system that mimics the process of spontaneous active systems in nature.

RESULTS AND DISCUSSION The target of the semiconducting materials was transition-metal dichalcogenides (TMDCs) which are a class of layered materials.22,

23

The semiconducting TMDCs of

molybdenum disulfide (MoS2) and tungsten diselenide (WSe2) have a honeycomb layer (Fig. 1) that stacks one on top of the other via van der Waals interaction to form bulk crystals. Each layer has a three-atom thickness, and atomically thin optoelectronic devices have been demonstrated.23, 24

Despite this, optimizing the characteristics of the devices, such as carrier concentration and

threshold voltage, remains significant challenge.24,

25

One successful tuning methodology is

surface charge transfer doping by forming an interface between a TMDC and molecules.25 In particular the BV molecule changes the TMDC characteristics dramatically from a semiconducting to a metallic state.21 So far, molecular dopants are deposited entirely onto the TMDC, and the surface is fully covered to uniformly dope the TMDC device.


4

Page 5 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

To achieve a spontaneous pattern formation using a well controllable method, we focused on spinodal dewetting of thin films.13,

26-28

Spinodal dewetting of polymer thin films was

observed, with dewetting in a substrate surface showing patterns ranging in scales from a few hundred of nanometers to microns across a large area.15,

26, 28

The key mechanism is the

instability of the polymer thin films on the surface, an instability which can be controlled by temperature and choice of interface materials.27-29 We first considered the macroscopic behavior of an instability of a BV film on a SiO2/Si substrate. To extend the BV molecules as a film on a substrate, a BV molecular solution (see EXPERIMENTAL SECTION) in toluene was cast (~ 150 µL cm-2) on a SiO2/Si substrate, and a ~670 nm film was obtained (Fig. S1). As shown in Fig. S2, no pattern formation was observed under ambient conditions by annealing at either 50 and 120

. However, we did find that an upper interface with water can destabilize the film and

form a pattern spontaneously on the substrate at 50

(Figs. 2a and S1). Fig. S1 shows optical

microscope images for macroscopic (> 10 µm) patterns of BV molecular films on the SiO2/Si substrate. The initial BV molecular film (as cast, before contact with water) had a flat surface, and the surface roughened in 1 min after immersion in water. Furthermore, the process showed a network pattern after 10 min, with the BV films eventually separating by fractions of dots with spacing of 10-50 µm in scale. This behavior can be observed in time-lapse images under a microscope. Fig. S3 and Movie S1 show the formation and coalescence of tiny holes to form cellular and network patterns from a BV film on the substrate. As a result of the pattern formation, the light-scattering from the substrate was changed, and a uniform pattern formation at a scale of 1 cm2 was observed. Previous studies have demonstrated the instability to form patterns of a hydrophobic polymer film by spinodal dewetting at the interface with water.29 In


5

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 22

our case also, the instability at the water/BV molecular interface would be the driving force to form the patterns. The characteristic length (undulation length, λ) of the pattern due to spinodal dewetting depends highly on the thickness of the film with a relationship of λ ~ h2, where h is the thickness of the film.28 To extend the characteristic length at the smaller nanoscale, we examined the process of thinner BV films on a SiO2/Si substrate. Figs. 2b and S4 show atomic force microscope (AFM) images of sub 20 nm ultrathin BV films obtained by spin-casting the solution between 1000 rpm (for thicker film) and 5000 rpm (for thinner film). Just after spin-casting, the BV films showed a rough surface with a root-mean-square (RMS) of 0.33 nm. After contacting with water for 0.5 min, tiny holes (diameter φ of ~10 nm) appeared on the surface of all films, and isolated dots were observed after 5 min. At 1 min, a network pattern (for 1000 rpm samples) and discontinuous maze patterns (for greater than 2000 rpm samples) were observed according to thickness (Figs. 2b and S5). Decreasing the thickness resulted in conversion of the network pattern to the maze pattern at the same point in water immersion time (the RMS value at 1 min indicates the relative thickness of the films, which decreases monotonically from 4.3 nm at 2000 rpm to 3.1 nm at 5000 rpm as shown in Table S1). To extract the characteristic length λ of the patterns, a 2D fast Fourier transform (FFT) of the AFM images was examined and wavenumber q was evaluated according to the following equation (1),28 Z(x) = h + δh exp(iqx)

(1)

where, Z(x), h, δh, and x are surface undulations, film thickness, fluctuation amplitude, and coordinate parallel to the surface, respectively. From the analysis of the FFT images (Figs. 2c and S5), we extracted the wavenumber q and the characteristic length λ from an equation λ =


6

Page 7 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

2π/q. The extracted λ is 204 nm at 2000 rpm, with the λ value decreasing to 165 nm (at 5000 rpm) as spin-casting speed increases (Fig. 2d), indicating a thickness-dependent patterning process at several hundred nanometers. With the above results, it was confirmed that the BV molecular ultrathin film exhibited spontaneous pattern formation on the nanoscale of tiny holes, network, maze, and dots as summarized in Fig. 2e.

Fig. 3 shows the spontaneous nanoscale pattern formation of an ultrathin BV film on a semiconducting material, MoS2. MoS2 flakes (lateral size of ~10 µm) were mechanically exfoliated from bulk crystals on a Si substrate covered with 260 nm-thick SiO2. Fig. 3a shows a schematic image of the process. A spin-cast BV ultrathin film (~10 nm in thickness) on the MoS2 was immersed in Milli-Q water at 50

. For the original BV film before contacting with water

(at 0 min), the surface roughness was RMS = 1.6 nm (Figs. 3b and c). After contacting with water for 5 min, initial pattern formation was observed as tiny holes of ~20 nm in diameter and 1 nm in depth. The film was flatter than it had been initially (0 min), and the RMS decreased to 0.4 nm; this smoothing can be attributed to a reduction in the interfacing area between the hydrophobic BV film and water, which corresponds to reducing interfacial free energy. As we proceed the process, the network pattern was observed at 30 min which eventually converted to a discontinuous maze pattern at 200 minutes. An FFT analysis of the network patterns (30 min in Fig. 3b) gives an undulation length (λ value) of 178 nm (q value is 35.3 nm-1). The pattern formation process of the BV film on MoS2 was similar to that on the SiO2/Si substrate, but the characteristic time to form the pattern was longer, indicating a slower pattern formation rate of the BV films on MoS2. On the SiO2 surface, the pattern transforms to a network or maze pattern (λ = 141~204 nm) at 1 min (Fig. 2b). The time scale is longer on the MoS2 surface, which showed a network pattern with a similar characteristic length (λ = 178 nm) at 30 min. To


7

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 22

consider the substrate dependency of the pattern formation rate, it would be plausible to hypothesize that the flow rate of BV molecules in the film is slower. This hypothesis is based on the flow being an incompressible viscous flow in the film according to which the flow rate (v) can be expressed as v

1/η, where η is the dynamic viscosity.13 A rough estimation indicates

that the dynamic viscosity η of the BV films on the MoS2 surface is calculated to be 30 times higher than that of the BV films on the SiO2 surface. The reason for the higher dynamic viscosity would be the interaction between BV molecules and the MoS2 by charge transfer interaction, which eventually slows the rate of pattern formation on the MoS2 surface. Because of the higher dynamic viscosity of BV films on the MoS2 surface, the network patterns containing thin BV wired structures would be sufficiently stable and observable under AFM.

Fig. 3c shows the RMS profile along the spontaneous pattern formation on MoS2. RMS increased from tiny holes at 5 min to network pattern formation at 60 min, and gradually decreased with the maze pattern formation at 200 min. When forming the network pattern, BV molecules on MoS2 flow laterally to form hills on MoS2, resulting in an increase in RMS (Fig. 3d). RMS values’ being reduced by a process lasting longer than 100 min can be explained by the MoS2 flat surface’s being exposed by the maze pattern formation. For further analysis, the area of the BV/water interface during the pattern formation process was evaluated. Fig. 3e shows the area of the BV/water interface by process time. It was found that spontaneous pattern formation progressed as the interfacial area was reduced, indicating that the BV/water repellent interaction is the driving force behind the pattern formation. However, in the interval of network formation between 30 min and 100 min, the interfacial area was maintained, indicating that the interfacial free energy (repulsive stress) between BV/water is maintained. To compensate the free energy, the area of the water/MoS2 interface increased, resulting in a reduction of the total


8

Page 9 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

free energy of the system (Figs. 3e and f). According to AFM observation, a spontaneous pattern formation of less than 200 nm in length was achieved on semiconducting MoS2 and the BV patterns were controllable and sufficiently stable even under the repulsive stress of the interface with water due to the slow flow rate of BV molecules on MoS2.

Fig. 4a shows a schematic image of a metal-oxide-semiconductor field-effect-transistor (MOSFET) device covered with the BV electron-dopant film. The device was fabricated using standard lithography techniques for mechanically exfoliated MoS2 on a heavily doped p+-Si substrate covered with 260 nm-thick SiO2 as a back-gate oxide. The BV thin film was spin-cast on a MoS2 MOSFET device, and the device was immersed in water to induce BV film-pattern formation. Fig. 4b shows the transfer characteristic curves along the pattern formation process. The original MoS2 MOSFET device (before casting the BV solution) shows an n-type behavior similar to that found in previous reports.21 The ON-current (ID) is of ~10 nA at gate voltage (VG) = 30 V. By spin-casting the BV solution to form a film over the entire channel of the original device, the ON current level increased to ~1 µA, indicating electron doping of MoS2 by BV molecules.21 By immersing the BV-coated device in water to form patterns, the ON (at VG = 30 V) current monotonically decreased as the immersion time increased, and eventually the ON current level decreased by an order of magnitude, IDS = 0.1 µA (Fig. 4c). This process can also be extended to a p-type WSe2 MOSFET device. The original WSe2 MOSFET device shows ptype behavior, and ambipolar behavior with a large n-branch by the BV film formation (Fig. 4d). When the device was immersed in water at 50

, the n-branch (at VG = 30 V) decreased

monotonically while the p-branch (VG = - 30 V) increased (Fig. 4e). Therefore, the patterning process can dramatically change the transfer characteristics of TMDC MOSFET devices.


9

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 22

The effect of the water treatment process to the devices revealed a further understanding the pattern formation and the corresponding device characteristics. When treated with water (soaked in water), both original MoS2 and WSe2 devices (without BV molecular film) are doped with a small amount of holes (Fig. S6). Considering that the pattern formation process increases the p-type character, the pattern formation exposes the surface of TMDCs and makes contact with water. In addition, we examined the effect of the water treatment temperature to the MoS2 device covered with the BV film (Fig. S7). The transfer characteristics of a spin-cast BV film on a MoS2 MOSFET device was maintained during a 5 hand, a 50

water treatment for 30 min; on the other

water treatment reproduced the reduction of the ON current level as described in

the above discussion. Since the pattern formation requires lateral flow of BV molecules, the pattern formation rate decreases at a lower temperature. These results further indicate the strong correlation between the pattern formation and the corresponding device characteristics. We further examined the reversibility of the patterns and corresponding device characteristics. Since the BV thin film is a molecular film, the patterns (network and maze patterns) would be diminished by annealing at a glass transition (or melting) temperature. This scenario is illustrated in Fig. 5a; the pattern would be reversibly diminished and reproduced by annealing and re-immersing the film in water. Fig. 5b shows reversibility of the pattern formation observed by AFM for a spin-cast BV ultrathin film on MoS2. By annealing in air at 120

, the network pattern (RMS = 0.65 nm) transformed to a more planar surface, while RMS

reducing to 0.62 nm by 30 sec and 0.44 nm by 5 min. The flattened film became destabilized again by contact with water, and the network pattern could be reproduced. In addition to the reversible pattern process, the device characteristics were also reversible (Fig. 5c). Annealing at 120

increased the ON current (point 4), and further immersion in water decreased it. So far,


10

Page 11 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

the full reversibility of the pattern is difficult; as a result, the ON current level is slightly smaller after the 1st recovery cycle. Here we emphasize that the device was sufficiently annealed at 250

in a total of 4 hours (see EXPERIMENTAL SECTION). Therefore, the 5 min annealing

process at 120

should not change the source/drain contact of the MOSFET. Furthermore, we

examined photoluminescence (PL) spectra of monolayer MoS2 through the reversible pattern process (Fig. S8). The PL intensities of peak A and B are reduced with BV doping, which indicates electron doping from BV molecules.30, 31 On a further process, the PL intensities were enhanced and reduced by immersing the BV-coated MoS2 in water to form patterns and by annealing to erase the patterns, respectively. On the other hand, just annealing an original (pristine) monolayer MoS2 results in enhancement of PL intensity. These results indicate that observed reversible process is not heating induced doping such as substrate induced doping but pattern formation of BV molecules is critical to modify the carrier concentration. According to the above results, the pattern formation can be reversible at the nanoscale from morphology and electronic-structure point of views.

CONCLUSIONS In summary, we demonstrated the proof-of-concept of spontaneous pattern formation on semiconducting materials, MoS2 and WSe2. Ultrathin films of donor BV molecules showed various patterns, including tiny holes, a network, a maze, and dots, that could be achieved at lengths less than 200 nm. The driving force for pattern formation was revealed to be a repulsive force at the interface between the water and the BV film, and the slow flow rate (high dynamic viscosity) of BV molecules on MoS2 was found to stabilize the network and maze patterns. We


11

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 22

also showed that the pattern formation was reversible and able to tune MOSFET device characteristics (drain current and polarity). This concept should be useful for tuning device characteristics by the spontaneous pattern ability of organic molecules, and further extend to the idea of the concept of an autonomous-response device.

EXPERIMENTAL SECTION Sample Preparation Method The toluene solution of BV molecules was prepared according to the literature.21, 32 25 mg of benzyl viologen dichloride (Sigma-Aldrich) was dissolved in Milli-Q water (~15 ml), and a 5 ml toluene layer was placed on the aqueous solution to form a toluene/water liquid interface. A reduction agent, sodium borohydride (NaBH4, 3.7 g, Sigma-Aldrich), was poured into the bilayer solution. Immediately, the solution turned cloudy due to the evolution of a large amount of hydrogen gas. About 18 hours after the reaction, the toluene layer (yellow solution) was extracted and poured onto SiO2/Si substrates or TMDCs on the substrates. In some cases, the extracted toluene layer was annealed on a hotplate at 50

for 10 min to separate water from the

toluene layer. To form patterns shown in Figure 2a, the BV solution in toluene was cast onto substrates either with a pipette (150 µL cm-2, Fig. S1) or a spin coater (MIKASA 1H-D7). All the substrates used were Si wafers covered with 260 nm of thermally grown SiO2 (Silicon Valley Microelectronics, Inc.). The BV cast film on the Si wafer was immersed in a Milli-Q water bath at 50

followed removal from the water bath and drying by blowing N2 to remove the water.

For the pattern formation shown in Figures 3 to 5, MoS2 (SPI Supplies) or WSe2 (HQ Graphene) was exfoliated with a mechanical exfoliation method using adhesive tapes on the SiO2/Si wafer. To obtain AFM images on MoS2, the BV solution in toluene was spin-cast on the exfoliated wafer, then immerse in 50

Milli-Q water bath followed by removal from the water bath and

drying by blowing N2 to remove the water. To evaluate device performances (Figures 4 and 5), a standard lithography technique was used to prepare metal-oxide-semiconductor field-effecttransistor (MOSFET) devices for the exfoliated MoS2 and WSe2. The contact metals were Ti/Au


12

Page 13 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

(~0.5 nm/40 nm) and the channel length of the devices were 10 µm. The fabricated devices were first annealed in an H2/N2 (4%/96%) atmosphere at 250

for 2 hours to clean the surface before

carrying out any measurements. Furthermore, before casting the BV toluene solution on MoS2 and WSe2 MOSFET devices, the devices were again annealed in an H2/N2 (4%/96%) atmosphere at 250

for 2 hours to clean the surface.

Physical measurement and analysis To observe the topographic images of the BV films and patterns on substrates, an atomic force microscope (AFM, SII instruments) with dynamic force mode (DFM) was used. The interface areas between BV and water (Fig. 3e) were evaluated using Spise132 software. Image J software was used to make binary images to evaluate the water projection area (Fig. 3e). These areas were divided by the whole projection area of the AFM image. FFT analyses were also carried out using Image J software. To observe the macroscopic BV patterns, an optical microscope (Olympus B51X) was used. The transfer characteristic curves of the MOSFET devices were obtained from a semiconductor parameter analyzer (Hewlett Packard 4140B and Keysight, B1500a). The PL measurement was carried out with LabRAM HR800 equipped with an EMCCD camera (HORIBA Scientific). The incident laser power was ~2.6 W cm-2.


13

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 22

FIGURES

Fig. 1. Schematic image of spontaneous pattern formation of the donor BV molecules on a substrate. Patterns are transformed from tiny holes, network, and dots as the process progresses. TMDC was used as a substrate and the lateral pattern size was on the nanoscale (~200 nm or less).


14

Page 15 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Fig. 2. Spontaneous pattern formation of BV films on SiO2/Si substrates. (a) Illustrative image of the process of the pattern formation. The BV film is immersed in water bath. (b) Pattern evolution of the ultrathin BV films formed by spin-casting under AFM observation. The water immersion time is from 0 min (as BV cast) to 5 min. (c) Spin-speed dependence of peak wavenumber q of the film at 1 min shown in Figs. 2b and S4. The solid lines are Gaussian fits through the data. (d) Characteristic length of each pattern from Fig.2c. (e) Illustrative images of the spontaneous pattern formation of the ultrathin BV film by contacting with water.


15

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 22

Fig. 3. Spontaneous pattern formation of BV film on MoS2. (a) Illustrative image of the pattern formation. (b) AFM images of the pattern formation by immersing in water and the topographic profile along the white lines on the images. The images are as BV cast (0 min), tiny holes (5 min), network (30 min), and maze (200 min). (c) Immersion-time dependence of RMS relative to the images shown in Fig. 3b. (d) Illustrative image of the lateral flow of BV molecules forming a network pattern. (e) Immersion-time dependency (left axis) of the interface area (BV/water) ratio of the images in Fig. 3b. The white and black regions correspond to BV and MoS2 region, respectively. (100% indicates the surface is wholly covered by BV molecules). The right axis is


16

Page 17 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

the area ratio of water penetration into the BV film. (f) Illustrative image of interfaces exhibiting spontaneous pattern formation.

Fig. 4. (a) Illustrative image of the device and expected doping profile by the pattern formation. (b) Transfer characteristic curves for MoS2 MOSFET. The curves are original (before BV cast), BV spin-cast (As BV cast), and water immersion for 5 min to 60 min. Drain voltage is 0.3 V. (c) Water immersion time dependency of the ON current at VG = 30 V in Fig. 4b. (d) Transfer characteristic curves for WSe2 MOSFET. Curves are original (before BV cast), as BV cast, and water-immersion for 5 min to 10 min. Drain voltage is -0.3 V. (e) Drain current (ID) of the original, doped and water treated WSe2 MOSFET at VG = 30 V and -30 V shown in Fig. 4d.


17

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 22

Fig. 5. Reversible pattern formation of the BV film on MoS2. (a) Illustrative image of the pattern recovery process. (b) AFM images for the reversibility of the patterns by annealing and reimmersion in water. (c) Transfer characteristic curves for MoS2 MOSFET. The curves are 1: original (before BV cast), 2: As BV cast, 3: water treated, 4: annealed, and 5: water re-treated. The graph on the right shows the drain current at VG = 30 V for each sample. Drain voltages are 0.3 V. The inset image is the device used. The scale bar is 10 µm.


18

Page 19 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

ASSOCIATED CONTENT The Supporting Information is available free of charge on the ACS Publications website at DOI: xxxxx. Pattern evolution of the cast BV film, annealing effect of the BV film, AFM images of the pattern evolution of the thin BV film, FFT analyses, MOSFET characteristics of the pattern evolution at low temperature, and the movie of the pattern evolution.

AUTHOR INFORMATION Corresponding Author *[email protected]

Author Contributions The manuscript was written by H.I. and D.K. The design of the experiment was done by H.I. and D.K. All the experiment was done by H.I. Mechanism of the pattern formation was examined by H.I., M.T. and D.K. All the researcher continuously discussed to wrap up the manuscript. All authors have given approval to the final version of the manuscript.

ACKNOWLEDGMENT This work was supported by JST PRESTO Grant Number JPMJPR1663.


19

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 22

REFERENCES

1.

2. 3. 4.

5. 6. 7.

8.

9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

Whitesides, G. M.; Mathias, J. P.; Seto, C. T. Molecular Self-Assembly and Nanochemistry: A Chemical Strategy for The Synthesis of Nanostructures. Science 1991, 254, 1312-1319. Barth, J. V.; Costantini, G.; Kern, K. Engineering Atomic and Molecular Nanostructures at Surfaces. Nature 2005, 437, 671-679. Pawin, G.; Wong, K. L.; Kwon, K. Y.; Bartels, L. A Homomolecular Porous Network at a Cu(111) Surface. Science 2006, 313, 961-962. Shyer, A. E.; Rodrigues, A. R.; Schroeder, G. G.; Kassianidou, E.; Kumar, S.; Harland, R. M. Emergent Cellular Self-Organization and Mechanosensation Initiate Follicle Pattern in the Avian Skin. Science 2017, 357, 811-815. Qi, S. Y.; Groves, J. T.; Chakraborty, A. K. Synaptic Pattern Formation During Cellular Recognition. Proc. Natl. Acad. Sci. USA 2001, 98, 6548-6553. Kondo, S.; Miura, T. Reaction-Diffusion Model as a Framework for Understanding Biological Pattern Formation. Science 2010, 329, 1616-1620. Othmer, H. G.; Painter, K.; Umulis, D.; Xue, C. The Intersection of Theory and Application in Elucidating Pattern Formation in Developmental Biology. Math. Model Nat. Phenom. 2009, 4, 3-82. Blagodatski, A.; Sergeev, A.; Kryuchkov, M.; Lopatina, Y.; Katanaev, V. L. Diverse Set of Turing Nanopatterns Coat Corneae Across Insect Lineages. Proc. Natl. Acad. Sci. USA 2015, 112, 10750-10755. Malescio, G.; Pellicane, G. Stripe Phases from Isotropic Repulsive Interactions. Nat. Mater. 2003, 2, 97-100. Chen, Q.; Bae, S. C.; Granick, S. Directed Self-Assembly of a Colloidal Kagome Lattice. Nature 2011, 469, 381-384. Bowman, C.; Newell, A. C. Natural Patterns and Wavelets. Reviews of Modern Physics 1998, 70, 289-301. Peng, S.; Lohse, D.; Zhang, X. Spontaneous Pattern Formation of Surface Nanodroplets from Competitive Growth. ACS nano 2015, 9, 11916-11923. Vandenbrouck, F.; Valignat, M. P.; Cazabat, A. M. Thin Nematic Films: Metastability and Spinodal Dewetting. Phys. Rev. Lett. 1999, 82, 2693-2696. Samid-Merzel, N.; Lipson, S. G.; Tannhauser, D. S. Pattern Formation in Drying Water Films. Phys. Rev. E 1998, 57, 2906-2913. Gentili, D.; Foschi, G.; Valle, F.; Cavallini, M.; Biscarini, F. Applications of Dewetting in Micro and Nanotechnology. Chem. Soc. Rev. 2012, 41, 4430-4443. Doursat, R.; Sayama, H.; Michel, O. A Review of Morphogenetic Engineering. Nat. Comput. 2013, 12, 517-535. Hajime, T. Formation of Network and Cellular Structures by Viscoelastic Phase Separation. Adv. Mater. 2009, 21, 1872-1880. Pierret, R. F. Semiconductor Device Fundamentals. Addison-Wesley: Reading, MA, 1996. Williams, E. D.; Ayres, R. U.; Heller, M. The 1.7 Kilogram Microchip: Energy and Material Use in the Production of Semiconductor Devices. Environ. Sci. Technol. 2002, 36, 5504-5510.


20

Page 21 of 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

20.

21. 22.

23.

24.

25.

26. 27. 28. 29. 30. 31. 32.

Kim, S. M.; Jang, J. H.; Kim, K. K.; Park, H. K.; Bae, J. J.; Yu, W. J.; Lee, I. H.; Kim, G.; Loc, D. D.; Kim, U. J.; Lee, E. H.; Shin, H. J.; Choi, J. Y.; Lee, Y. H. ReductionControlled Viologen in Bisolvent as an Environmentally Stable n-Type Dopant for Carbon Nanotubes. J. Am. Chem. Soc. 2009, 131, 5010-5010. Kiriya, D.; Tosun, M.; Zhao, P.; Kang, J. S.; Javey, A. Air-Stable Surface Charge Transfer Doping of MoS2 by Benzyl Viologen. J. Am. Chem. Soc. 2014, 136, 7853-7856. Chhowalla, M.; Shin, H. S.; Eda, G.; Li, L. J.; Loh, K. P.; Zhang, H. The Chemistry of Two-Dimensional Layered Transition Metal Dichalcogenide Nanosheets. Nat. Chem. 2013, 5, 263-275. Wang, Q. H.; Kalantar-Zadeh, K.; Kis, A.; Coleman, J. N.; Strano, M. S. Electronics and Optoelectronics of Two-Dimensional Transition Metal Dichalcogenides. Nat. Nanotech. 2012, 7, 699-712. Jariwala, D.; Sangwan, V. K.; Lauhon, L. J.; Marks, T. J.; Hersam, M. C. Emerging Device Applications for Semiconducting Two-Dimensional Transition Metal Dichalcogenides. ACS nano 2014, 8, 1102-1120. Schmidt, H.; Giustiniano, F.; Eda, G. Electronic Transport Properties of Transition Metal Dichalcogenide Field-Effect Devices: Surface and Interface Effects. Chem. Soc. Rev. 2015, 44, 7715-7736. Reiter, G. Dewetting of Thin Polymer Films. Phys. Rev. Lett. 1992, 68, 75-78. Suh, K. Y.; Lee, H. H. Dynamic Instability of Strongly Confined Thin Polymer Films in Spinodal Dewetting. Phys. Rev. Lett. 2001, 87, 135502. Xie, R.; Karim, A.; Douglas, J. F.; Han, C. C.; Weiss, R. A. Spinodal Dewetting of Thin Polymer Films. Phys. Rev. Lett. 1998, 81, 1251-1254. Liu, D.; Wang, T. Instability of Hydrophobic and Viscoelastic Polymer Thin Films in Water at Room Temperature. J. Phys.: Condens. Matter 2013, 25, 415101. Mouri, S.; Miyauchi, Y.; Matsuda, K. Tunable Photoluminescence of Monolayer MoS2 via Chemical Doping. Nano lett. 2013, 13, 5944-5948. Mak, K. F.; He, K.; Lee, C.; Lee, G. H.; Hone, J.; Heinz, T. F.; Shan, J. Tightly Bound Trions in Monolayer MoS2. Nat. Mater. 2013, 12, 207-211. Kim, S. M.; Jang, J. H.; Kim, K. K.; Park, H. K.; Bae, J. J.; Yu, W. J.; Lee, I. H.; Kim, G.; Loc, D. D.; Kim, U. J.; Lee, E.-H.; Shin, H.-J.; Choi, J.-Y.; Lee, Y. H. ReductionControlled Viologen in Bisolvent as an Environmentally Stable n-Type Dopant for Carbon Nanotubes. J. Am. Chem. Soc. 2009, 131, 327-331.


21

ACS Nano 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 22

TOC


22

Tuning Transition-Metal Dichalcogenide Field-Effect-Transistor by

Recommend Documents