Adding Solvent into Ionic Liquid-Gated Transistor - ACS Publications

Subscriber access provided by LUNDS UNIV

Surfaces, Interfaces, and Applications

Adding Solvent into Ionic-Liquid-Gated Transistor: The Anatomy of Enhanced Gating Performance Wei Zhao, Sheng Bi, Cheng Zhang, Philip D. Rack, and Guang Feng ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b03433 • Publication Date (Web): 15 Mar 2019 Downloaded from http://pubs.acs.org on March 18, 2019

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 33 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 Applied Materials & Interfaces

Adding Solvent into Ionic-Liquid-Gated Transistor: The Anatomy of Enhanced Gating Performance Wei Zhao1, Sheng Bi1, Cheng Zhang2, Philip D. Rack2,3, Guang Feng1,4* 1State

Key Laboratory of Coal Combustion, School of Energy and Power Engineering, Huazhong

University of Science and Technology, Wuhan 430074, China 2Department

of Materials Science and Engineering, University of Tennessee, Knoxville, Tennessee 37996,

USA 3Center

for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831,

USA 4Shenzhen

Research Institute of Huazhong University of Science and Technology, Shenzhen, 518057,

China

1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 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

ABSTRACT: Most studies of ionic liquid (IL) gated field effect transistors (FETs) focus on the extremely large electric field and capacitance induced in liquid/solid interfaces and correspondingly the significantly enhanced carrier density in semiconductors, which can appreciably improve the gating performance. However, how to boost the switching speed, another key property of gating performance of FETs, has been rarely explored. In this work, the gating performance of molybdenum disulfide (MoS2) FETs, gated by the mixtures of IL/organic solvent (1-butyl-3-methylimidazolium tetrafluoroborate/acetonitrile, [Bmim][BF4]/ACN) at different ion concentrations, is investigated for both dynamic and static properties by a combination of molecular dynamics simulation and resistance network analysis. Results reveal that organic solvent can speed up the IL response time by a factor of about 40x at the optimal ion concentration of 1.94 M, which is mainly attributed to the increased ionic conductivity of IL via the addition of organic solvent. Meanwhile, the surface charge distribution of MoS2 becomes more homogenous after the addition of organic solvent, which increases the conductivity of MoS2 by up to 2.4x. Surprisingly, the optimal ion concentration for increased switching speed is nearly the same as that for achieving highest MoS2 conductivity. Thus, our findings provide a strategy to simultaneously improve the dynamics and static gating performance of IL-gated FETs as well as a modeling technique to screen out the ideal ion concentration.

KEYWORDS: ionic liquid gated transistor, switching speed, charging dynamics, organic solvent, molybdenum disulfide


Page 2 of 33

Page 3 of 33 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


INTRODUCTION Electrolyte-gated field effect transistors (FETs), owing to their ultrathin electric double layers (EDLs) inducing extremely large carrier density at semiconductor surfaces,1-3 have become a new class of low power-consuming FETs, attracting increasing interests in recent years.1-7 Different types of electrolytes, including aqueous/organic salt solutions, ionic liquids (ILs), ion gels, polyelectrolytes and polymer electrolytes, have been widely used to enhance the metal-insulator transition in various materials compared to conventional high-k dielectric materials (such as HfO2, Al2O3, and ZrO2).1-3,

7-9

Among these electrolytes, ILs with superb

properties, such as high polarization, wide electrochemical window, and high thermal and chemical stability,10-12 have demonstrated the capability of inducing sufficiently high carrier density (up to 1015 cm-2) that can trigger the metal-insulator transition in semiconductor but could not be achieved by other electrolytes.1, 3, 5, 6, 9, 13, 14 Meanwhile, recent experiments have reported that ILs could not only improve the gating performance, including ON/OFF ratio, sub-threshold swing and carrier mobility of FETs,15-17 but also are capable of suppressing the metal-insulator transition by inducing oxygen vacancies in semiconductors via extreme high electric fields.6, 18 Thus, comparing with other electrolytes, ILs could enrich the material science of FETs as well as be a versatile tool to explore high carrier density physics. Despite many studies of IL-gated FETs focusing on improving the aforementioned static characteristics, dynamic properties, specifically the switching speed, of FETs are vital for their application. In electrolyte-gated FETs, the switching speed mainly depends on the polarization 3 ACS Paragon Plus Environment


time of the electrolyte rather than the carrier mobility of the semiconductor,4,

Page 4 of 33

19

since the

polarization of electrolyte makes the switching speed of electrolyte-gated FETs extremely slower than that of the transistor gated by solid dielectrics. Among different electrolytes, although ILs with higher ionic conductivity could polarize faster than polymer-based electrolytes,20 the intrinsic high viscosity makes the double-layer formation in ILs still much longer than in aqueous/organic salt solutions which have been widely used as gate dielectrics.21-26 This limits the operation frequency of IL-gated FETs and, ultimately, their potential applications. By far, to inherit the high capacitance of ILs and hasten its response, great efforts have been made to increase the switching speed of IL-gated FETs. For example, binary IL mixtures, which are fabricated by adding inorganic salts to dissociate organic ions, are used as dielectrics to speed up the response of FETs through accelerating the formation of EDL.27 Furthermore, adding organic solvents into ILs, which could boost the ion diffusion in ILs and consequently increase the ionic conductivity of electrolytes,28,

29

has been adopted to speed up the dynamics process in many

EDL-related applications, in particular for supercapacitors.30-32 Therefore, the addition of organic solvent can potentially accelerate the response of IL-gated FETs,33 but there is still a lack of work to explore why and how the organic solvent could play such a role in IL gating (i.e., the underlying mechanism of gating for IL with organic solvent, such as how fast the EDL could form and how EDL impacts the surface charge and its distribution). In addition, unlike the supercapacitor community in which organic solvent added into IL has to sacrifice the wide electrochemical window of IL, IL gating would not suffer from such negative impact, since the 4 ACS Paragon Plus Environment

Page 5 of 33 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


gate voltage could be less than the safe operating voltages of organic solvents (e.g., a cell voltage up to 2.7 V for organic electrolytes with acetonitrile as solvent)9, 16, 34-37, especially in IL-gated MoS2 transistors,15, 38, 39 which suggests that using organic solvent for IL gating could be a very promising strategy. In this work, the mixtures of IL 1-butyl-3-methylimidazolium tetrafluoroborate ([Bmim][BF4]) and organic solvent acetonitrile (ACN) with different ion concentrations are used to gate semiconducting 2H-MoS2 that is an emerging 2D semiconductor widely used in FETs and produces excellent gating performance.15,

40-43

The gating performance is explored by a

hierarchical simulation approach coupling molecule dynamics (MD) simulation with resistance network analysis. Herein, the effects of ion concentration on the polarization response time of IL-gated MoS2 FETs to a pulsed square-shaped gate voltage are systematically analyzed. The mechanism of improving the switching speed is uncovered by characterizing the interfacial properties of both electrolyte and semiconductor at a nanoscale perspective. Subsequently, utilizing resistance network analysis, the influence of ion concentration on the conductivity of MoS2 is quantitatively examined.

RESULTS AND DISCUSSION To investigate the effect of organic solvent on the dynamic property of IL-gated FETs, a typical channel-like simulation cell is constructed (Figure 1), consisting of slabs of MoS2, gold and electrolyte of IL mixed with organic solvent (used as semiconductor, gate electrode and dielectric, respectively). Four different ion concentrations are selected to study the influence of 5 ACS Paragon Plus Environment


Page 6 of 33

ion concentration on the gating performance of IL-gated MoS2 FETs from the perspective of both the dynamic and static properties. The gate voltage (𝑉𝑔) of 0.5 V is applied between MoS2 and gold, which is compatible with the threshold voltage of IL-gated MoS2 FETs obtained by Perera et al.15 Since the switching speed of IL-gated FETs is mainly limited by the EDL forming at the liquid/solid interface,4, 19 it can be characterized by the process of charge accumulation on the semiconductor surface. We begin our work by directly analyzing the gating process of IL-gated MoS2 FETs. As shown in Figure 2, the time evolution of the accumulated charge on MoS2 has been computed from MD simulations, which makes it possible to compare different gating processes by extracting the polarization response time for charging MoS2. Figure 2a shows that the charging processes of MoS2 FET gated by IL/solvent mixtures could be finished within 10 ns; however, when using pure IL as the gate dielectric, the complete gating process requires a charging time of at least 30 ns (see Figure S1). These observations indicate that adding organic solvent into IL could significantly accelerate the EDL forming and consequently the response of semiconductor surface charge. In order to quantitatively measure the switching speed of MoS2 FET, the polarization response time, 𝜏, is evaluated by fitting the charging process curve with a biexponential function as follows,44

[

𝑄𝑡 = 𝑄𝑚𝑎𝑥 1 ― 𝑎𝑒

𝑡

―𝑏

𝑡

― (1 ― 𝑎)𝑒

―𝑐

]

(1)

where 𝑄𝑡 is the charge accumulated on MoS2 at time 𝑡, and 𝑄𝑚𝑎𝑥 is the total charge induced on MoS2 with EDLs fully formed and is obtained directly by averaging the flat region of the 6 ACS Paragon Plus Environment

Page 7 of 33 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


charging curve in Figure 2a. The fitting has been processed based on the overall charging duration (Figure S2) and the fitting parameters are listed in Table S1, which indicate that using Eq. 1, the charging processes could be well reproduced by monotonically increasing exponential functions. The polarization response time of each process is estimated as the total surface charge reaches the 95% charge state (i.e., 𝑄𝑡 = 0.95𝑄𝑚𝑎𝑥). It shows that the switching speed of semiconductor could be enlarged by about 40 times via adding organic solvent (Table S2). Even if 𝑄𝑚𝑎𝑥 is treated as a fitting parameter, it is nearly the same with that directly calculated from MD simulation (see Tables S1 and S3), which indicates that the fitting model used here is reliable and the extracted polarization response time is capable of quantitatively assessing the charging process of a MoS2 FET. Another two sets of polarization response times based on different MoS2 charging states (0.90𝑄𝑚𝑎𝑥 and 0.98𝑄𝑚𝑎𝑥) are also calculated and similar trends can be observed (Table S2), which confirm that adding organic solvent could improve the switching speed of IL-gated FETs. Interestingly, through the comparison of polarization response times under different ion concentrations, it is noted that the relationship between switching speed of MoS2 FET and ion concentration is not linear, and the polarization response time has a minimum value at the ion concentration of 1.94 M. To better understand the optimum response at 1.94 M, bulk simulations of IL and its mixtures with various ion concentrations were performed. The inset of Figure 2a exhibits that the ionic conductivity shows a bell-shaped relationship with ion concentration, which is consistent with previous results both in trend and magnitude28, 45 and could be illustrated 7 ACS Paragon Plus Environment


Page 8 of 33

by Nernst-Einstein relation46. That is, the ionic conductivity is proportional to the product of ion concentration and ion diffusion coefficient.46 Although both the ion concentration and ion diffusion coefficient are changed monotonically with the addition of organic solvent (see Figure S5), because the decrease in ion concentration and the increase in ion diffusion coefficient have different change rates at different amount of organic solvent, the product of ion concentration and ion diffusion coefficient finally presents a bell-shaped relationship with the ACN mole fraction

(see

the

inset

of

Figure

S5),

which

is

in

consistent

with

the

ionic

conductivity-concentration relationship shown in the inset of Figure 2a. By comparing the trends of ionic conductivity and polarization response time, it shows that the IL/solvent mixture with higher ionic conductivity leads to a more rapid EDL formation, which reduces the polarization response time of MoS2 FET more. This analysis of ionic conductivity further proves that the switching speed of IL-gated FETs is controlled by the polarization of EDL, which is dominated by the ionic conductivity of IL and could be significantly improved by adding organic solvent. Thus, one can obtain the optimal ion concentration for fast electrolyte gating, mainly via evaluating the ionic conductivity of IL/solvent mixture. Beyond the effect of ion concentration on switching speed of IL-gated FETs shown by a single charging process (Figure 2a), as displayed in Figure 2b, a pulsed square-shaped gate voltage (top panel of Figure 2b) was applied to all the studied systems, which has been widely used in experiments to test the operation frequency of IL-gated FETs. Two and half cycles of charging and discharging processes were used to capture the response of charge on MoS2 and the 8 ACS Paragon Plus Environment

Page 9 of 33 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


results between different cycles are in highly consistent (bottom panel of Figure 2b). As the gate voltage turns on, charges start to accumulate on MoS2, and gradually vanish when turning off the gate voltage. In a half cycle period (0 – 1.5 ns) of the charging process, when gated by IL/solvent mixture with lower ion concentrations (1.94 M and 0.83 M), the charges accumulated on MoS2 could reached 95% charged state (i.e., 0.95𝑄𝑚𝑎𝑥). However, when gated by high concentrated IL/solvent mixture (4.28 M) and pure IL (5.36 M), especially by pure IL, the MoS2 FET cannot respond effectively to the evolving gate voltage, which is consistent with charging process shown in Figure 2a. In addition, MoS2 FET gated by IL/solvent mixture can respond to the change of gate voltage faster during the half cycle period for discharging processes. Briefly, by applying a pulsed gate voltage, it is further verified that the addition of solvent will significantly improve the response of the IL-gated FETs. Near the charged surface, ILs are well known to have distinct ion distributions and orientations compared with their bulk states,11, effect on gating performance.50,

51

47-49

which have been shown to have a strong

On the basis of the EDL structures obtained by MD

simulations, we examined how organic solvent affects the gating performance of IL-gated FETs. Firstly, the 1D EDL structures in terms of number densities and orientation distributions are analyzed. As shown in Figure 3a, with adding organic solvent, the ion number densities for both cations and anions in the first ion layer gradually decrease, and more solvent molecules accumulate at MoS2 surface. Interestingly, the peak location of the first solvent layer is closer to the MoS2 surface than that of the first ion layer, which indicates that ILs are squeezed out from 9 ACS Paragon Plus Environment


the MoS2 surface by organic solvent since the first atom layer of MoS2 near the electrolyte mainly contacts with solvent. It should be noted that 𝑄𝑚𝑎𝑥 is not determined by the peak height of ion density but by the synergistic actions of cations and anions in EDL. Moreover 𝑄𝑚𝑎𝑥 is found to vary weakly with the addition of solvent, which is similar to the observation in previous work of supercapacitors with IL mixture.52 Since ILs are diluted by organic solvent, the interaction between ions and electrode would be changed and could be reflected by orientation distribution of molecules in the first layer. As shown in Figure 3b, the cation tends to be parallel to MoS2 surface and its longer side chain prefers lying on the MoS2 surface in pure IL; however, the orientation ordering of cation approaches a more random distribution with the addition of organic solvent, which indicates that the ion-electrode interaction is weakened by solvent located in-between. The orientations of solvent molecules reveal that solvent molecules prefer to lie on the MoS2 surface with weak influence from the ion concentration. The 1D EDL structures imply that in the first layer of EDL, organic solvent gradually plays a more dominant role with adding more solvent (i.e., reducing ion concentration). The relatively random orientation distribution of ions after adding organic solvent indicates that the packing pattern of the first ion layer is altered and then the surface charge distribution at semiconductor surface is changed as well. To delve deeper into the effect of organic solvent on the first ion layer and corresponding surface charge induced, a series of 2D interfacial structures were analyzed with a focus of electrolyte and semiconductor (Figure 4). The MD snapshots of first layer in EDL show that solvent molecules gradually squeeze out ions with adding organic solvent (Figure 4a). The dense 10 ACS Paragon Plus Environment

Page 10 of 33

Page 11 of 33 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


packing formed between cation and anion is gradually broken with lowering the ion concentration, which agrees well with the conclusion drawn from 1D EDL structures that organic solvent plays a key role in the first layer of EDL and could lessen the ion-electrode interaction. In line with the ion packing, due to the lower viscosity of IL/solvent mixture

28, 53, 54

and more

random orientation distribution of IL, the distribution of charge density in EDL tends to be more homogeneous with addition of organic solvent (Figure 4b). Induced by the differently distributed ions of EDL, the homogeneity of charge distribution on MoS2 surface is also significantly altered with the addition of organic solvent. Figure 4c shows the MD-obtained surface charge distribution on the first sulfur layer at MoS2 surface which contacts with the first layer of EDL. Correspondingly, the charge distribution on MoS2 surface becomes more homogeneous after adding organic solvents, which is compatible with the ion distribution in EDL. The homogeneity of the surface charge distributions was further quantitatively validated by their histograms and variances (Figure S3 and Table S4). It has been reported that the more homogeneous the surface charge distribution is, the higher the carrier mobility of semiconductor would achieve.55,

56

Therefore, by adding the organic solvent, the homogeneity of surface charge distribution is significantly improved, which could enhance the carrier mobility. Although the organic solvent which carries zero net charge plays a dominate role in the first layer of EDL (i.e., squeezing out ions and changing ion orientations), the total carrier density of MoS2 electrode does not vary much with changing ion concentration (Figure S4), which is in line with the previous molecular simulation on the influence of solvent on IL/electrode interface in 11 ACS Paragon Plus Environment


Page 12 of 33

supercapacitor.52 Therefore, the 2D analyses for both EDL and semiconductor suggest that organic solvent can improve the carrier mobility without sacrificing the carrier density at semiconductor surface. This gives a potential strategy that the addition of organic solvent could simultaneously accelerate the switching process and increase the conductivity of semiconductor. Herein, to evaluate the effect of organic solvent on the conductivity of semiconductor, the resistance network analysis was taken to calculate the source-drain current (𝐼𝐷𝑆) of MoS2 FET with different ion concentrations and then the conductivity of semiconductor was characterized quantitatively. The MoS2 surface is sliced into a 38 × 22 grids network (with a grid size of 0.16 × 0.27 nm2 in xy-plane), by making the center of surface sulfur atom coincide with the center of the grid (Figure 5a). Actually, the MD-obtained partial charge on semiconductor surface atom is a Gaussian charge, due to the employed CMP technique in our MD simulation. Therefore, the contribution of atom charge on the charge accumulated (𝑞𝑗) in the grid 𝑗 of the surface atom could be obtained by integrating the Gaussian distribution of partial charge on each surface atom (detailed information could be found in Method part). The electrical conductance of grid 𝑗 (𝐺𝑗) is obtained by, (2)

𝐺𝑗 = 𝜎𝑗𝜇 where 𝜎𝑗 is the surface charge density of grid 𝑗, which is obtained by

𝑞𝑗 𝐷𝑥𝐷𝑦

(𝐷𝑥 and 𝐷𝑦 is the

grid size in x and y direction, respectively), and 𝜇 is the carrier mobility of MoS2, which is taken from the previous experiment work on IL-gated MoS2 FETs.15 As shown in Figure 5b, a more homogeneous distribution of electrical conductance 12 ACS Paragon Plus Environment

Page 13 of 33 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


corresponding to different grids on MoS2 surface is generated by adding more solvent, and the distribution of electrical conductance is in accordance with the charge distribution in the first layer of EDL and surface charge distribution on MoS2 shown respectively in Figure 4b and 4c. However, when the IL/solvent mixture is diluted from 1.94 to 0.83 M, it is hard to distinguish the difference of homogeneity of carrier distribution just by the conductance distributions. To quantitatively compare the conductance of MoS2 gated by IL/solvent mixtures with different ion concentrations, a source-drain voltage (𝑉𝐷𝑆 = 0.1 V) was applied to the resistance network and 𝐼𝐷𝑆 was obtained by the line integration of current density. As shown in Figure 5c, comparing to the concentrated IL solutions (4.28 M) and pure IL (5.36 M), higher 𝐼𝐷𝑆 is achieved in lower ion concentrations which could render more homogenous surface charge distribution, and the highest 𝐼𝐷𝑆, occurring at the ion concentration of 1.94 M, is nearly 2.4 times as large as that of pure IL. Surprisingly, the optimal ion concentrations in terms of faster switching speed and higher conductivity of MoS2 are the same, which implies that the conductivity and the response of IL-gated FETs could be improved simultaneously by adding organic solvent. Furthermore to validate the hierarchical simulation method used above, 𝐼𝐷𝑆 of the MoS2 FET was scaled up by adopting the practical size of semiconductor in experiment to generate the equivalent source-drain current 𝐼′𝐷𝑆 (details can be seen in Part 5 of Supporting Information). 𝐼′𝐷𝑆 for MoS2 gated by pure IL is calculated as 29.3 µA, which is in good agreement with the source-drain current obtained by experiment (~ 25 µA) under same gate voltage and source-drain voltage.15 The difference may be attributed to the fact that MoS2 was gated by a different IL 13 ACS Paragon Plus Environment


(N,N-diethyl-N-methyl-N-(2-methoxyethyl)ammonium bis(trifluoromethylsulphonyl)imide) in the experiment.15

CONCLUSIONS In summary, we have studied the effects of organic solvent on the polarization response time and the homogeneity of surface charge distribution on both electrolyte side and semiconductor sides of the IL-gated MoS2 FETs by MD simulation, and quantitatively evaluated the conductivity of MoS2 by calculating the source-drain current using resistance network analysis. By adding organic solvent, the switching speed of MoS2 FET is enhanced and could be optimized at the ion concentration of 1.94 M. The relationship of ionic conductivity with respect to the ion concentration also shows that the highest ionic conductivity is achieved at 1.94 M. These suggest that the response of semiconductor in IL-gated FETs, which is dominated by the polarization of ions in the EDL, could be accelerated by increasing the ionic conductivity via adding organic solvent. The 1D EDL structures combined with 2D interfacial structures for both electrolyte and semiconductor demonstrate that the organic solvent gradually takes the place of IL in the first layer of EDL and weaken the ion-electrode interaction. The 2D surface charge density on MoS2 surface shows that as organic solvent is added, the total charge on semiconductor is weakly changed, meanwhile the surface charge distribution tends to be more homogeneous, which consequently increases the conductivity of semiconductor. Through the resistance network


Page 14 of 33

Page 15 of 33 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


analysis, the source-drain current was quantitatively evaluated, confirming that the addition of organic solvent could increase the conductivity of MoS2. Interestingly, the optimal ion concentration for the source-drain current is the same as that for the response of MoS2, which suggests that organic solvent could not only accelerate the dynamic process of IL-gated FETs, but also enhance the static performance at the same time. Therefore, our MD simulations combined with resistance network method have revealed the effect of organic solvent on the gating performance of IL-gated FETs and the mechanism underlying from a nanoscopic view, which is helpful to understand the ion behavior in practical IL-gated FETs and to enhance the gating performance from both dynamic and static properties. This hierarchical simulation approach also provides an efficient way to screen out the electrolyte for better gating performance.

METHODS Molecular Dynamics Simulation. As shown in Figure 1, the MD simulation system comprises a slab of IL/solvent mixture with four different ion concentrations sandwiched between Au(111) and MoS2. The sheets of gold and MoS2 both have an area of 6.0 × 6.0 nm2 in xy-plane and are separated by 12.0 nm in z-direction which is wide enough to prevent overlapping of the EDLs formed at both electrodes and produce a bulk-like electrolyte region in the middle of system. In order to simplify the system construction, we set up nine different systems based on the IL-solvent mole ratios (𝑁𝐼𝐿/𝑁𝑠𝑜𝑙 = 0.01, 0.03, 0.05, 0.10, 0.15, 0.25, 0.50, 1.00, and pure IL,



Page 16 of 33

where 𝑁𝐼𝐿 represents the number of ion pairs and 𝑁𝑠𝑜𝑙 is the number of solvent molecules) to analyze the ionic conductivity, and the number ratios are finally converted into the form of ion concentrations through the whole manuscript (see Figure 1). Furthermore, four different ion concentrations are selected for analyzing the dynamic properties of IL-gated FETs (see the inset of Figure 2a). The constant potential method (CPM) was used to generate the gate voltage applied between the gold and MoS2 during the MD simulations,57, 58 which has been widely used to take into account the electronic polarizability of the electrode surfaces.59,

60

Through this

method, the partial charge on the solid atom which is applied with electrical potential could be extracted directly. More importantly, studies have shown that only CPM could correctly simulate the charging dynamics of the EDL formation process,44 while it is much more computationally expensive than the general constant charge method in MD simulation, which is actually the reason that even in supercapacitor modeling there are quite few studies on charging process,44, 61 despite numerous studies on molecular modeling of static properties of EDL-related applications. The IL [Bmim][BF4] and organic solvent ACN are both represented by coarse-grained models and the corresponded force fields are taken from Merlet et al.,52 which could greatly save the computational resource, compared with all-atom models. Both the cation and ACN molecule are described by three sites (imidazolium ring, butyl group and methyl group of cation and nitrogen atom, carbon atom and methyl group of ACN are treated as single spheres, respectively) and the anions are treated as one-bead spheres. The accuracy of the coarse-grained model has been verified by comparing the simulated physical properties with experiments and the data obtained 16 ACS Paragon Plus Environment

Page 17 of 33 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


by all-atom model.54, 62 Additionally, these coarse-grained models have well predicted the EDL properties coupled with various charged surfaces.44, 52, 54, 58, 63 The interaction between IL/solvent mixture and solid atom is characterized by Lennard-Jones (LJ) interaction, and the LJ interaction parameters for MoS2 and gold are taken from Tao et al.64 and Verde et al.,65 respectively. MD simulations were performed via a customized GROMACS code,66 with a time step of 2 fs. All the chemical bonds in IL and solvent are constrained by LINCS algorithm.67 The temperature of electrolyte was maintained at room temperature (298 K) using the velocity rescaling thermostat with a time constant of 1 ps.68 The number of electrolyte molecules inside the channel was adjusted to make the IL/solvent mixture showed a bulk-like state in central part of the channel. The electrostatic interactions were computed using PME method69 with an FFT grid spacing of 0.1 nm and cubic interpolation for charge distribution in the reciprocal space. The cut-offs for both Coulombic and van der Waals interactions in real space were set to 1.2 nm. The slab correction was used to remove the interactions between periodic images in z-direction.70 Each simulation was run for 10 ns to reach equilibrium before applying gate voltage, and then simulations started to run with a gate voltage applied at 𝑡 = 0 to obtain the charging process. The information of EDL structure was analyzed by MD trajectories after the charging process is completed. Twenty independent simulations were performed for each ion concentration to obtain accurate averaged charging and discharging curves shown in Figure 2. Ionic Conductivity Analysis. A series of bulk simulations with nine different ion concentrations were applied to analyze the effect of ion concentration on ionic conductivity of ILs. The time 17 ACS Paragon Plus Environment


Page 18 of 33

integral of electrical current autocorrelation function (ECACF) was used to analyze the ionic conductivity which has been proved that it can predict the ionic conductivity accurately, especially for the situation that the interaction between ions of electrolyte is weak.53 Details could be found in Part 6 of Supporting Information. Integration of Gauss Distribution of Surface Atom Charge. The contribution of each sulfur atom partial charge on the charge accumulated on the whole MoS2 surface could be calculated as, 3

𝛼2

( )𝑒

𝑞𝑟 = 𝑞𝑖

2

― 𝛼2|𝑟 ― 𝑟𝑖|2

𝜋

(3)

where 𝑞𝑖 is the specific atom charge of semiconductor at the position 𝑟𝑖 in space, 𝑞𝑟 is the charge accumulated at the position 𝑟 in space and 𝛼 denotes the width of Gaussian charge distribution. The in-plane charge distribution can be obtained by integrating the Gaussian distribution of atom charge in z-direction, and the contribution of sulfur atom 𝑖 at the position

(𝑥𝑖, 𝑦𝑖) on the charge accumulation at the specific position (𝑥, 𝑦) could be simplified as 𝑞(𝑥,

𝑦), 𝑖

shown in Eq. 4. 𝑞(𝑥,

𝛼2

𝑦), 𝑖

2

(

)2 + (𝑦 ― 𝑦𝑖)2]

= 𝜋 𝑞𝑖𝑒 ― 𝛼 [ 𝑥 ― 𝑥𝑖

(4)

Then, the total charge on each grid could be calculated by summation of the charge accumulated inside each grid as, 𝑥 + 𝐷𝑥 𝑦𝑗 + 𝐷𝑦 ∫𝑦 𝑞(𝑥,𝑦),𝑖𝑑𝑥𝑑𝑦 𝑗 𝑗

𝑞𝑗 = ∑𝑖∫𝑥𝑗

(5)

where 𝑞𝑗 denotes the charge accumulated in grid 𝑗 and the 𝑥𝑗, 𝑦𝑗 represents the start point of 18 ACS Paragon Plus Environment

Page 19 of 33 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


the grid 𝑗, respectively. Owing to the extreme narrow Gaussian charge distribution and the staggered pattern of the first sulfur layer on MoS2 surface, compared to the grid centered on the sulfur atom, the grid whose center do not coincide with the sulfur atom have much lower electrical conductance. As a result, it would generate an ideal interlaced arrangement of electrical conductance on MoS2 surface gated by IL/solvent mixture with lower ion concentration (0.83 and 1.94 M) (Figure 4a). Resistance Network Analysis. The MoS2 surface was considered as a periodical material with a unit cell consisting of 836 grids (38 × 22), the electrical conductance of each grid is derived from the surface charge density. A source-drain voltage with 0.1 V was applied along y-axis to generate current. An equipotential boundary condition was applied at two sides of MoS2 surface (x = 0 and x = 6.0 nm). The source-drain current can be calculated by integrating the current density along the vector perpendicular to the direction of the electric field. All these calculations were carried out by finite element method (FEM). The detailed information can be found in our previous work.51

AUTHOR INFORMATION Supporting Information The Supporting Information is available free of charge on the ACS Publications website at … Corresponding Author Email: [email protected] Notes The authors declare no competing financial interests. 19 ACS Paragon Plus Environment


ACKNOWLEDGMENTS This work was financially supported by the National Natural Science Foundation of China (51876072) and Shenzhen Basic Research Project (JCYJ20170307171511292). C.Z. acknowledges support by U.S. Department of Energy (DOE) under Grant DE-SC0002136. PDR acknowledges support from the Center for Nanophase Materials Sciences, which is a DOE Office of Science User Facility. All simulations were performed at the National Supercomputing Centers in Guangzhou (Tianhe II) and Tianjin (Tianhe-1A).


Page 20 of 33

Page 21 of 33 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


REFERENCES (1)

Ye, J. T.; Inoue, S.; Kobayashi, K.; Kasahara, Y.; Yuan, H. T.; Shimotani, H.; Iwasa, Y. Liquid-Gated Interface Superconductivity on an Atomically Flat Film. Nat. Mater. 2010, 9, 125-128.

(2)

Ueno, K.; Nakamura, S.; Shimotani, H.; Ohtomo, A.; Kimura, N.; Nojima, T.; Aoki, H.; Iwasa, Y.; Kawasaki, M. Electric-Field-Induced Superconductivity in an Insulator. Nat. Mater. 2008, 7, 855-858.

(3)

Bollinger, A. T.; Dubuis, G.; Yoon, J.; Pavuna, D.; Misewich, J.; Božović, I. Superconductor–Insulator Transition in La2-XSrxCuO4 at the Pair Quantum Resistance. Nature 2011, 472, 458-460.

(4)

Cho, J. H.; Lee, J.; Xia, Y.; Kim, B.; He, Y.; Renn, M. J.; Lodge, T. P.; Daniel Frisbie, C. Printable Ion-Gel Gate Dielectrics for Low-Voltage Polymer Thin-Film Transistors on Plastic. Nat. Mater. 2008, 7, 900-906.

(5)

Xia, Y.; Cho, J. H.; Lee, J.; Ruden, P. P.; Frisbie, C. D. Comparison of the Mobility-Carrier Density Relation in Polymer and Single-Crystal Organic Transistors Employing Vacuum and Liquid Gate Dielectrics. Adv. Mater. 2009, 21, 2174-2179.

(6)

Jeong, J.; Aetukuri, N.; Graf, T.; Schladt, T. D.; Samant, M. G.; Parkin, S. S. P. Suppression of Metal-Insulator Transition in VO2 by Electric Field-Induced Oxygen Vacancy Formation. Science 2013, 339, 1402-1405.

(7)

Panzer, M. J.; Frisbie, C. D. Polymer Electrolyte Gate Dielectric Reveals Finite Windows of High Conductivity in Organic Thin Film Transistors at High Charge Carrier Densities Polymer Electrolyte Gate Dielectric Reveals Finite Windows of High. J. Am. Chem. Soc. 2005, 127, 6960-6961.

(8)

Panzer, M. J.; Frisbie, C. D. High Carrier Density and Metallic Conductivity in Poly(3-Hexylthiophene) Achieved by Electrostatic Charge Injection. Adv. Funct. Mater. 2006, 16, 1051-1056.

(9)

Nakano, M.; Shibuya, K.; Okuyama, D.; Hatano, T.; Ono, S.; Kawasaki, M.; Iwasa, Y.; Tokura, Y. Collective Bulk Carrier Delocalization Driven by Electrostatic Surface Charge Accumulation. Nature 2012, 487, 459-462.

(10) Maton, C.; De Vos, N.; Stevens, C. V. Ionic Liquid Thermal Stabilities: Decomposition 21 ACS Paragon Plus Environment


Mechanisms and Analysis Tools. Chem. Soc. Rev. 2013, 42, 5963-5977. (11) Fedorov, M. V.; Kornyshev, A. A. Ionic Liquids at Electrified Interfaces. Chem. Rev. 2014, 114, 2978-3036. (12) Welton, T. Room-Temperature Ionic Liquids. Solvents for Synthesis and Catalysis. Chem. Rev. 1999, 99, 2071-2084. (13) Yamada, Y.; Ueno, K.; Fukumura, T.; Shimotani, H.; Iwasa, Y.; Gu, L.; Tsukimoto, S.; Ikuhara, Y.; Kawasaki, M. Electrically Induced Ferromagnetism Titanium Dioxide. Science 2011, 332, 1065-1067. (14) Ueno, K.; Nakamura, S.; Shimotani, H.; Yuan, H. T.; Kimura, N.; Nojima, T.; Aoki, H.; Iwasa, Y.; Kawasaki, M. Discovery of Superconductivity in KTaO3 by Electrostatic Carrier Doping. Nat. Nanotechnol. 2011, 6, 408-412. (15) Perera, M. M.; Lin, M.-W.; Chuang, H.-J.; Chamlagain, B. P.; Wang, C.; Tan, X.; Cheng, M. M.-C.; Tománek, D.; Zhou, Z. Improved Carrier Mobility in Few-Layer MoS2 Field-Effect Transistors with Ionic-Liquid Gating. ACS Nano 2013, 7, 4449-4458. (16) Chuang, H. J.; Tan, X.; Ghimire, N. J.; Perera, M. M.; Chamlagain, B.; Cheng, M. M. C.; Yan, J.; Mandrus, D.; Tománek, D.; Zhou, Z. High Mobility WSe2 p- and n- Field-Effect Transistors Contacted by Highly Doped Graphene for Low-Resistance Contacts. Nano Lett. 2014, 14, 3594-3601. (17) Kim, B. J.; Jang, H.; Lee, S. K.; Hong, B. H.; Ahn, J. H.; Cho, J. H. High-Performance Flexible Graphene Field Effect Transistors with Ion Gel Gate Dielectrics. Nano Lett. 2010, 10, 3464-3466. (18) Li, M.; Han, W.; Jiang, X.; Jeong, J.; Samant, M. G.; Parkin, S. S. P. Suppression of Ionic Liquid Gate-Induced Metallization of SrTiO3 (001) by Oxygen. Nano Lett. 2013, 13, 4675-4678. (19) Lee, J.; Panzer, M. J.; He, Y.; Lodge, T. P.; Frisbie, C. D. Ion Gel Gated Polymer Thin-Film Transistors. J. Am. Chem. Soc. 2007, 129, 4532-4533. (20) Yuan, H.; Shimotani, H.; Tsukazaki, A.; Ohtomo, A.; Kawasaki, M.; Iwasa, Y. High-Density Carrier Accumulation in ZnO Field-Effect Transistors Gated by Electric Double Layers of Ionic Liquids. Adv. Funct. Mater. 2009, 19, 1046-1053. (21) Roest, A. L.; Kelly, J. J.; Vanmaekelbergh, D.; Meulenkamp, E. A. Staircase in the 22 ACS Paragon Plus Environment

Page 22 of 33

Page 23 of 33 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


Electron Mobility of a ZnO Quantum Dot Assembly Due to Shell Filling. Phys. Rev. Lett. 2002, 89, 036801. (22) Rosenblatt, S.; Yaish, Y.; Park, J.; Gore, J.; Sazonova, V.; McEuen, P. L. High Performance Electrolyte Gated Carbon Nanotube Transistors. Nano Lett. 2002, 2, 869-872. (23) Yoon, H.; Lee, S. H.; Kwon, O. S.; Song, H. S.; Oh, E. H.; Park, T. H.; Jang, J. Polypyrrole Nanotubes Conjugated with Human Olfactory Receptors: High-Performance Transducers for FET-Type Bioelectronic Noses. Angew. Chem. 2009, 121, 2793-2796. (24) Kwon, O. S.; Park, S. J.; Hong, J.-Y.; Han, A. R.; Lee, J. S.; Lee, J. S.; Oh, J. H.; Jang, J. Flexible FET-Type VEGF Aptasensor Based on Nitrogen-Doped Graphene Converted from Conducting Polymer. ACS Nano 2012, 6, 1486-1493. (25) Sarkar, D.; Liu, W.; Xie, X.; Anselmo, A. C.; Mitragotri, S.; Banerjee, K. MoS2 Field-Effect Transistor for Next-Generation Label-Free Biosensors. ACS Nano 2014, 8, 3992-4003. (26) Kutovyi, Y.; Zadorozhnyi, I.; Handziuk, V.; Hlukhova, H.; Boichuk, N.; Petrychuk, M.; Vitusevich, S. Liquid-Gated Two-Layer Silicon Nanowire FETs: Evidence of Controlling Single-Trap Dynamic Processes. Nano Lett. 2018, 18, 7305-7313. (27) Ono, S.; Miwa, K.; Seki, S.; Takeya, J. High-Performance Organic Field-Effect Transistors with Binary Ionic Liquids. Org. Electron. 2009, 10, 1579-1582. (28) Chaban, V. V.; Voroshylova, I. V.; Kalugin, O. N.; Prezhdo, O. V. Acetonitrile Boosts Conductivity of Imidazolium Ionic Liquids. J. Phys. Chem. B 2012, 116, 7719-7727. (29) Li, S.; Zhang, P.; Fulvio, P. F.; Hillesheim, P. C.; Feng, G; Dai, S.; Cummings, P. T. Enhanced Performance of Dicationic Ionic Liquid Electrolytes by Organic Solvents. J. Phys.: Condens. Matter 2014, 26, 284105. (30) Chmiola, J.; Yushin, G.; Gogotsi, Y.; Portet, C.; Simon, P.; Taberna, P. L. Anomalous Increase in Carbon Capacitance at Pore Sizes Less Than 1 Nanometer. Science 2006, 313, 1760. (31) Béguin, F.; Presser, V.; Balducci, A.; Frackowiak, E. Carbons and Electrolytes for Advanced Supercapacitors. Adv. Mater. 2014, 26, 2219-2251. (32) Zhong, C.; Deng, Y.; Hu, W.; Qiao, J.; Zhang, L.; Zhang, J. A Review of Electrolyte Materials and Compositions for Electrochemical Supercapacitors. Chem. Soc. Rev. 2015, 23 ACS Paragon Plus Environment


44, 7484-7539. (33) Sayago, J.; Meng, X.; Quenneville, F.; Liang, S.; Bourbeau, É.; Soavi, F.; Cicoira, F.; Santato, C. Electrolyte-Gated Polymer Thin Film Transistors Making Use of Ionic Liquids and Ionic Liquid-Solvent Mixtures. J. Appl. Phys. 2015, 117, 112809. (34) Ono, S.; Seki, S.; Hirahara, R.; Tominari, Y.; Takeya, J. High-Mobility, Low-Power, and Fast-Switching Organic Field-Effect Transistors with Ionic Liquids. Appl. Phys. Lett. 2008, 92, 2006-2009. (35) Xia, Y.; Cho, J. H.; Lee, J.; Ruden, P. P.; Frisbie, C. D. Comparison of the Mobility–Carrier Density Relation in Polymer and Single-Crystal Organic Transistors Employing Vacuum and Liquid Gate Dielectrics. Adv. Mater. 2009, 21, 2174-2179. (36) Balducci, A.; Dugas, R.; Taberna, P. L.; Simon, P.; Plée, D.; Mastragostino, M.; Passerini, S. High Temperature Carbon–Carbon Supercapacitor Using Ionic Liquid as Electrolyte. J. Power Sources 2007, 165, 922-927. (37) Frackowiak, E. Carbon Materials for Supercapacitor Application. Phys. Chem. Chem. Phys. 2007, 9, 1774-1785. (38) Zan, W.; Zhang, Q.; Xu, H.; Liao, F.; Guo, Z.; Deng, J.; Wan, J.; Zhu, H.; Chen, L.; Sun, Q.; Ding, S.; Zhou, P.; Bao, W.; Zhang, D. W. Large Capacitance and Fast Polarization Response of Thin Electrolyte Dielectrics by Spin Coating for Two-Dimensional MoS2 Devices. Nano Res. 2018, 11, 3739-3745. (39) Lee, Y.; Lee, J.; Kim, S.; Park, H. S. Rendering High Charge Density of States in Ionic Liquid-Gated MoS2 Transistors. J. Phys. Chem. C 2014, 118, 18278-18282. (40) Lu, J. M.; Zheliuk, O.; Leermakers, I.; Yuan, N. F. Q.; Zeitler, U.; Law, K. T.; Ye, J. T. Evidence for Two-Dimensional Ising Superconductivity in Gated MoS2. Science 2015, 350, 1353-1357. (41) Lee, J.; Mak, K. F.; Shan, J. Electrical Control of the Valley Hall Effect in Bilayer MoS2 Transistors. Nat. Nanotechnol. 2016, 11, 421. (42) Wang, F.; Stepanov, P.; Gray, M.; Lau, C. N.; Itkis, M. E.; Haddon, R. C. Ionic Liquid Gating of Suspended MoS2 Field Effect Transistor Devices. Nano Lett. 2015, 15, 5284-5288. (43) Yu, Z.; Ong, Z.-Y.; Pan, Y.; Cui, Y.; Xin, R.; Shi, Y.; Wang, B.; Wu, Y.; Chen, T.; Zhang, 24 ACS Paragon Plus Environment

Page 24 of 33

Page 25 of 33 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


Y.-W.; Zhang, G.; Wang, X. Realization of Room-Temperature Phonon-Limited Carrier Transport in Monolayer MoS2 by Dielectric and Carrier Screening. Adv. Mater. 2016, 28, 547-552. (44) Péan, C.; Merlet, C.; Rotenberg, B.; Madden, P. A.; Taberna, P. L.; Daffos, B.; Salanne, M.; Simon, P. On the Dynamics of Charging in Nanoporous Carbon-Based Supercapacitors. ACS Nano 2014, 8, 1576-1583. (45) Liang, M.; Zhang, X.-X.; Kaintz, A.; Ernsting, N. P.; Maroncelli, M. Solvation Dynamics in

a

Prototypical

Ionic

Liquid

+

Dipolar

Aprotic

Liquid

Mixture:

1-Butyl-3-Methylimidazolium Tetrafluoroborate + Acetonitrile. J. Phys. Chem. B 2014, 118, 1340-1352. (46) Frömling, T.; Kunze, M.; Schönhoff, M.; Sundermeyer, J.; Roling, B. Enhanced Lithium Transference Numbers in Ionic Liquid Electrolytes. J. Phys. Chem. B 2008, 112, 12985-12990. (47) Rotenberg, B.; Salanne, M. Structural Transitions at Ionic Liquid Interfaces. J. Phys. Chem. Lett. 2015, 6, 4978-4985. (48) Kornyshev, A. A.; Qiao, R. Three-Dimensional Double Layers. J. Phys. Chem. C 2014, 118, 18285-18290. (49) Feng, G.; Zhang, J.; Qiao, R. Microstructure and Capacitance of the Electrical Double Layers at the Interface of Ionic Liquids and Planar Electrodes. J. Phys. Chem. C 2009, 113, 4549-4559. (50) Black, J. M.; Come, J.; Bi, S.; Zhu, M.; Zhao, W.; Wong, A. T.; Noh, J. H.; Pudasaini, P. R.; Zhang, P.; Okatan, M. B.; Dai, S.; Kalinin, S. V.; Rack, P. D.; Ward, T. Z.; Feng, G.; Balke, N. Role of Electrical Double Layer Structure in Ionic Liquid Gated Devices. ACS Appl. Mater. Interfaces 2017, 9, 40949-40958. (51) Zhao, W.; Bi, S.; Balke, N.; Rack, P. D.; Ward, T. Z.; Kalinin, S. V.; Dai, S.; Feng, G. Understanding Electric Double-Layer Gating Based on Ionic Liquids: From Nanoscale to Macroscale. ACS Appl. Mater. Interfaces 2018, 10, 43211-43218. (52) Merlet, C.; Salanne, M.; Rotenberg, B.; Madden, P. A. Influence of Solvation on the Structural and Capacitive Properties of Electrical Double Layer Capacitors. Electrochim. Acta 2013, 101, 262-271. 25 ACS Paragon Plus Environment


(53) Sha, M.; Dong, H.; Luo, F.; Tang, Z.; Zhu, G.; Wu, G. Dilute or Concentrated Electrolyte Solutions? Insight from Ionic Liquid/Water Electrolytes. J. Phys. Chem. Lett. 2015, 6, 3713-3720. (54) Burt, R.; Breitsprecher, K.; Daffos, B.; Taberna, P.-L.; Simon, P.; Birkett, G.; Zhao, X. S.; Holm, C.; Salanne, M. Capacitance of Nanoporous Carbon-Based Supercapacitors Is a Trade-Off between the Concentration and the Separability of the Ions. J. Phys. Chem. Lett. 2016, 7, 4015-4021. (55) Xia, Y.; Xie, W.; Ruden, P. P.; Frisbie, C. D. Carrier Localization on Surfaces of Organic Semiconductors Gated with Electrolytes. Phys. Rev. Lett. 2010, 105, 1-4. (56) Li, M.; Graf, T.; Schladt, T. D.; Jiang, X.; Parkin, S. S. P. Role of Percolation in the Conductance of Electrolyte-Gated SrTiO3. Phys. Rev. Lett. 2012, 109, 1-5. (57) Vatamanu, J.; Vatamanu, M.; Bedrov, D. Non-Faradaic Energy Storage by Room Temperature Ionic Liquids in Nanoporous Electrodes. ACS Nano 2015, 9, 5999-6017. (58) Merlet, C.; Rotenberg, B.; Madden, P. A.; Taberna, P.-L.; Simon, P.; Gogotsi, Y.; Salanne, M. On the Molecular Origin of Supercapacitance in Nanoporous Carbon Electrodes. Nat. Mater. 2012, 11, 306. (59) Wu, P.; Huang, J.; Meunier, V.; Sumpter, B. G.; Qiao, R. Complex Capacitance Scaling in Ionic Liquids-Filled Nanopores. ACS Nano 2011, 5, 9044-9051. (60) Kondrat, S.; Wu, P.; Qiao, R.; Kornyshev, A. A. Accelerating Charging Dynamics in Subnanometre Pores. Nat. Mater. 2014, 13, 387-393. (61) Zhan, C.; Lian, C.; Zhang, Y.; Thompson, M. W.; Xie, Y.; Wu, J.; Kent, P. R. C.; Cummings, P. T.; Jiang, D.-e.; Wesolowski, D. J. Computational Insights into Materials and Interfaces for Capacitive Energy Storage. Adv. Sci. 2017, 4, 1700059. (62) Merlet, C.; Salanne, M.; Rotenberg, B. New Coarse-Grained Models of Imidazolium Ionic Liquids for Bulk and Interfacial Molecular Simulations. J. Phys. Chem. C 2012, 116, 7687-7693. (63) Merlet, C.; Péan, C.; Rotenberg, B.; Madden, P. A.; Daffos, B.; Taberna, P. L.; Simon, P.; Salanne, M. Highly Confined Ions Store Charge More Efficiently in Supercapacitors. Nat. Commun. 2013, 4, 2701. (64) Liang, T.; Phillpot, S. R.; Sinnott, S. B. Parametrization of a Reactive Many-Body 26 ACS Paragon Plus Environment

Page 26 of 33

Page 27 of 33 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


Potential for Mo-S Systems. Phys. Rev. B 2009, 79, 245110. (65) Verde, A. V.; Acres, J. M.; Maranas, J. K. Investigating the Specificity of Peptide Adsorption on Gold Using Molecular Dynamics Simulations. Biomacromolecules 2009, 10, 2118-2128. (66) Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E. GROMACS 4: Algorithms for Highly Efficient, Load-Balanced, and Scalable Molecular Simulation. J. Chem. Theory Comput. 2008, 4, 435-447. (67) Hess, B.; Bekker, H.; Berendsen, H. J.; Fraaije, J. G. LINCS: A Linear Constraint Solver for Molecular Simulations. J. Comput. Chem. 1997, 18, 1463-1472. (68) Bussi, G.; Donadio, D.; Parrinello, M. Canonical Sampling through Velocity Rescaling. J. Chem. Phys. 2007, 126, 014101. (69) Raghunathan, A. V.; Aluru, N. R. Self-Consistent Molecular Dynamics Formulation for Electric-Field-Mediated Electrolyte Transport through Nanochannels. Phys. Rev. E 2007, 76, 011202-011202. (70) Yeh, I.-C.; Berkowitz, M. L. Ewald Summation for Systems with Slab Geometry. J. Chem. Phys. 1999, 111, 3155-3162.



Figures

Figure 1. Schematics of simulation system. In the snapshots of MD simulation, the red, blue spheres and green continuum between Au and MoS2 denote the cation, anion of IL and solvent, respectively. Four different ion concentrations are scrutinized in this work, which are made by dissolving IL in organic solvents. The ratios of number of solvent molecules (𝑁𝑠𝑜𝑙) to the number of ion pairs (𝑁𝐼𝐿) and corresponding concentrations are listed at the bottom. In particular, the pure IL has an ion concentration of 5.36 M.


Page 28 of 33

Page 29 of 33 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


Figure 2. Dependence of ion concentration on the charging dynamics. (a) Time evolution of total charge (𝑄) on MoS2 under a gate voltage of 0.5 V and the ionic conductivity of different IL/solvent mixtures. The fitting of charging processes are shown in dashed lines, respectively. The dots in the inset exhibit the ionic conductivity as a function of the ion concentration, and the dots in larger size correspond to the ion concentrations used to analyze the charging processes. (b) Cycles of charging/discharging under different concentrations. Top panel of (b) shows a pulsed square-shaped gate voltage was applied to all the studied systems, and the bottom panel exhibits the time evolution of the voltage and corresponded total charge of MoS2. 𝑞𝑒 stands for the electric charge of an electron.



Figure 3. 1D EDL structure of interfacial electrolyte. (a) Number density profiles of cations, anions and solvent molecules as a function of distance to the MoS2 surface. (b) Orientation profiles of angles formed by different vectors shown in cation/solvent with the normal of MoS2 surface in the first ion/solvent layer (< 0.55 nm). IM, BT and ME denote the imidazolium ring, butyl group and methyl group of Bmim+. N, C and MEA stand for nitrogen, carbon atom of cyano group and methyl group of ACN. 𝛼, 𝛽, and 𝜂 denote the angle formed by IM-BT vector, IM-ME vector and the normal of cation plane with the normal of MoS2 surface, respectively. 𝛾 stands for the angle formed between N-MEA vector of solvent molecule and the normal of MoS2 surface.


Page 30 of 33

Page 31 of 33 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


Figure 4. 2D interfacial structure of the first ion/solvent layer. (a-b) Snapshots (a) and 2D charge density profiles (b) of the first ion/solvent layer for electrolytes with different ion concentration. The red, blue and green continuums in (a) denote the cation, anion of IL and solvent, respectively. (c) Surface charge distributions on MoS2 surface, circles with different colors denote the surface sulfur atoms carrying different partial charges.



Figure 5. The conductance of MoS2. (a) Grids drawn by slicing the sulfur atoms at MoS2 surface (the yellow dots represent the positions of sulfur atoms). (b) The conductance distributions obtained by integrating the Gaussian distribution of atom charge located in the grid shown in (a) for different ion concentrations. (c) Source-drain current on semiconductor as a function of ion concentration.


Page 32 of 33

Page 33 of 33 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 Paragon Plus Environment

Adding Solvent into Ionic Liquid-Gated Transistor - ACS Publications

Recommend Documents