Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES
Communication
Layer-dependent Chemically Induced Phase Transition of Two-dimensional MoS2 Lifei Sun, Xingxu Yan, Jingying Zheng, Hongde Yu, Zhixing Lu, Shang-Peng Gao, Lina Liu, Xiaoqing Pan, Dong Wang, Zhiguo Wang, Peng Wang, and Liying Jiao Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.8b00452 • Publication Date (Web): 21 May 2018 Downloaded from http://pubs.acs.org on May 21, 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 19 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
Nano Letters
Layer-dependent
Chemically
Induced
Phase
Transition of Two-dimensional MoS2 ∥
Lifei Sun†,#, Xingxu Yan‡,&,#, Jingying Zheng†, Hongde Yu†, Zhixing Lu†, Shang-peng Gao , Lina Liu†, Xiaoqing Pan&, Dong Wang†, Zhiguo Wang*,§, Peng Wang*,‡, and Liying Jiao*,† †
Key Laboratory of Organic Optoelectronics and Molecular Engineering of the Ministry of
Education, Department of Chemistry, Tsinghua University, Beijing 100084, China ‡
National Laboratory of Solid State Microstructures, College of Engineering and Applied
Sciences and Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 210093, China §
School of Electronics Science and Engineering, University of Electronic Science and
Technology of China, Chengdu, 610054, China &
Department of Chemical Engineering and Materials Science, Department of Physics and
Astronomy, University of California – Irvine, Irvine, CA 92697, USA ∥
#
Department of Materials Science, Fudan University, Shanghai 200433, China
These authors contributed equally to this work.
*Correspondence to:
[email protected],
[email protected],
[email protected] ACS Paragon Plus Environment
1
Nano Letters 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 19
ABSTRACT: Two-dimensional (2D) transition metal dichalcogenides (TMDCs) with layered structures provide a unique platform for exploring the effect of number of layers on their fundamental properties. However, the thickness scaling effect on the chemical properties of these materials remains unexplored. Here we explored the chemically induced phase transition of 2D molybdenum disulfide (MoS2) from both experimental and theoretical aspects and observed that the critical electron injection concentration and the duration required for the phase transition of 2D MoS2 increased with decreasing number of layers. We further revealed that the observed dependence originated from the layer-dependent density of states of 2H-MoS2, which results in decreasing phase stability for 2H-MoS2 with increasing number of layers upon electron doping. Also, the much larger energy barrier for phase transition of monolayer MoS2 induces the longer reaction time required for monolayer MoS2 than multilayer MoS2. The layer-dependent phase transition of 2D MoS2 allows for the chemical construction of semiconducting-metallic heterophase junctions and subsequently, the fabrications of rectifying diodes and all 2D field effect transistors and thus opens new avenue for building ultrathin electronic devices. In addition, these new findings elucidate how electronic structures affect the chemical properties of 2D TMDCs and therefore, shed new lights on the controllable chemical modulations of these emerging materials.
KEYWORDS: two-dimensional atomic crystals, phase transition, MoS2, layer-dependent
ACS Paragon Plus Environment
2
Page 3 of 19 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
Nano Letters
Two-dimensional (2D) transition metal dichalcogenides (TMDCs) exhibit abundant exciting properties owing to their diverse compositions, crystalline phases and electronic structures.1 As a result of their distinctive layered structures, the number of layers significantly affects their physical properties due to the quantum confinement effect and interlayer interactions,2 which makes them an ideal system to explore layer-dependent properties. The layer-dependent physical properties of 2D TMDCs have been intensively investigated. Taking the representative 2D TMDC, molybdenum disulfide (MoS2) as an example, the layer-dependent band structure,3, 4 spectroscopic,4,
5
and electrical properties6 of MoS2 have been revealed. However, the layer-
dependent chemical properties of MoS2 have been rarely explored. Chemical modification offers a versatile tool for tailoring the structures and properties of 2D TMDCs,7, 8 and thus can further enrich their functionalities and extend their applications. For instance, the phase conversion of semiconducting 2H phase into metallic 1T (1T') phase of MoS2 by utilizing chemical intercalation agents (such as alkali metals) endows 2D MoS2 with new exciting properties of superconductivity,9 ferromagnetism,10 and quantum spin Hall effect.11 Although the alkali metals have been widely used to induce the phase transition of 2D MoS2, the mechanism of the phase transition has not been fully understood, especially for the intrinsic structural effect of 2D MoS2, such as number of layers, on the chemically induced phase conversion process has not been studied experimentally. Here, we explored the effect of number of layers on the chemical properties of 2D TMDCs using the chemically induced phase transition of MoS2 as a model system. We observed that the critical injected electron concentrations (Cc) required for the phase transition of MoS2 were highly dependent on the number of layers. We further correlated this dependence with the layerdependent electronic structures of 2D MoS2 by performing density functional theory (DFT)
ACS Paragon Plus Environment
3
Nano Letters 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 19
calculations. Based on the dramatically different phase transition tendency between monolayer (1L) and multilayered (ML) MoS2, we constructed ultrathin metallic-semiconducting (M-S) hetero-phase junctions and further fabricated rectifying diodes and all 2D field effect transistors (FETs) based on the M-S junctions, providing ultrathin electronic components for future nanoelectronics. Moreover, this study reveals that the chemical properties of 2D TMDCs are highly related to their electronic structures, offering new possibilities for functionalizing 2D electronic materials with chemical approaches. We first investigated the phase transition of 2H-MoS2 with varied number of layers by treating chemical vapor deposition (CVD)-grown 2H-MoS2 flakes5 with n-butyllithium12 (Figure S1). Although the detailed mechanism for 2H→1T (1T') phase transformation has not been fully understood, it is well accepted that the phase transition of 2H-MoS2 into 1T (1T')-MoS2 is induced by the injected electrons, which occupy the lowest energy states above the Fermi energy and thus lead to a destabilization of the lattice of pristine 2H-MoS2 and a structural transition to the 1T phase, and then to the distorted 1T' phases spontaneously with the energy relaxation (Figure 1a).13 According to this understanding, we tuned electron doping concentration by adjusting the concentration and treatment duration of n-butyllithium to reveal the effect of number of layers on the phase conversion tendency. We treated MoS2 flakes with different number of layers (1-20L) using n-butyllithium with a large variations of concentrations of 0.0011.6 M for 1-168 h. Raman spectroscopy was utilized to determine the critical phase conversion conditions for 2D MoS2 as 2H- and 1T (1T')-MoS2 show dramatically different Raman features. The two representive Raman peaks for 2H-MoS2 are located at 383 cm-1 (E12g) and 402 cm-1 (A1g) (Figure S2a),14 whereas typical Raman peaks for 1T (1T')-MoS2 appear at 156 cm-1 (J1), 226 cm-1 (J2), and 330 cm-1 (J3) (Figure S2c).15 The initiation and completion of the phase
ACS Paragon Plus Environment
4
Page 5 of 19 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
Nano Letters
conversion can be indicated by the emerging of J1 peak at 156 cm-1 (Figure S2b) and absence of E12g mode at 383 cm-1 (Figure S2d), respectively.15 We obtained the dependence of critical phase transition conditions (doping concentration and duration) on the number of layers shown in Figure 1b by measuring the Raman spectra of a large number of n-butyllithium-treated MoS2 flakes with varied number of layers. We found that thicker 2H-MoS2 (> 5L) could be transformed into 1T (1T') phase after being treated with very dilute n-butyllithium solution (0.05 M) for 2 h, but the phase transition for 1L 2H-MoS2 required a very high concentration (1.6 M) for much longer time (120 h) (Figure 1b and S3), revealing that Cc for the phase transition was highly dependent on the number of layers and increased with decreasing number of layers. This layer-dependent phase transition behavior can be visualized more clearly in the Raman mapping image taken on a ML MoS2 flake (Figure 1c) treated with a moderate concentration (0.8 M) of nbutyllithium for 24 h. The exposed 1L region of the ML flake remained as 2H phase while the ML region was fully converted into 1T (1T') phase as shown in both Raman mapping image and spectra (Figure 1c and 1e, see more discussions on the Raman data in supporting information). PL mapping image also clearly indicated the formation of 2H-1T (1T') 2D junction of MoS2 as no PL signal was detected in the ML region (Figure 1d) while the 1L region still exhibited typical PL peaks of 2H-MoS2 with the same optical bandgap (Figure 1f).
ACS Paragon Plus Environment
5
Nano Letters 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 19
Figure 1. Layer-dependent phase transition of MoS2 flakes. (a) Schematic for the number of layers selective phase transition of MoS2. (b) Critical phase transition conditions for MoS2 flakes with varied number of layers. Inset: optical images of the CVD-grown 2H-MoS2 flakes with varied number of layers. Scale bars: 10 µm. (c) Raman mapping image with 402 cm-1 (2H) and 156 cm-1 (1T') peak intensity and (d) PL mapping image with 675 nm peak intensity of the same flake after being soaked in 0.8 M n-butyllithium for 24 h, respectively. Inset of (c), optical image of this flake. Scale bars: 5 µm. Typical (e) Raman and (f) PL spectra of as-made 1L MoS2, 1L MoS2 and ML MoS2 after being soaked in 0.8 M n-butyllithium for 24 h, respectively.
ACS Paragon Plus Environment
6
Page 7 of 19 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
Nano Letters
To directly characterize the structural transition of 2D MoS2 upon electron injection at the atomic scale, we performed high angle annular dark field (HAADF) imaging on the same ML MoS2 flakes before and after being treated with 0.8 M n-butyllithium for 24 h using scanning transmission electron microscope (STEM). The as-grown triangular ML MoS2 flake (Figure 2a) showed honeycomb atomic packing with AA stacking between layers (Figure 2b inset).5 The thickness of MoS2 can be readily distinguished by examining the relative intensity of the lattice sites in the HAADF image (Figure 2b and inset). After the n-butyllithium treatment, the 2L region was found to be converted into 1T' phase with 2a×a superstructure according to the image simulation (Figure 2e and 2h), which is a typical distorted 1T phase with superlattice structure resulting from the clusterization of Mo atoms.16 The STEM-HAADF images of the ML region also exhibited evident atomic structure of the 1T' phase with 2a×a superstructure (Figure 2f and 2i). In contrast, the 1L region still remained as 2H structure, in which the coordination of Mo with S is trigonal prismatic (Figure 2c, 2d and 2g). Besides atomically-resolved imaging, the selected area electron diffraction (SAED) patterns and low-loss electron energy loss spectra (EELS) collected on the corresponding areas also confirmed that the 2H → 1T' structural transition only occurred in 2 and more layers regions (Figure S4 and S5). The occasionally exposed 1L domains in the 2L regions also remained as 2H phase, further confirming the layerselective phase conversion of MoS2 (Figure S6).
ACS Paragon Plus Environment
7
Nano Letters 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 19
Figure 2. STEM characterizations of ML MoS2 flakes before and after n-butyllithium treatment. (a) Low-magnification STEM-HAADF image of a ML flake stacked by multiple concentric triangles with shrinking size. Scale bar: 2 µm. (b) STEM-HAADF image of the 1L-2L boundary as marked in the red square in (a) before n-butyllithium treatment. Inset: the intensity line profile along the blue line. Scale bar: 1 nm. (c) STEM-HAADF image of the same 1L-2L boundary after being treated with 0.8 M n-butyllithium for 24 h. The yellow dashed line indicated the boundary of 1L and 2L regions. Scale bar: 2 nm. (d)-(f) High resolution STEM images of 1L, 2L and ML after n-butyllithium treatment, respectively. All images are filtered through a standard Wiener deconvolution to partially remove the background noise. Scale bars: 1 nm. (g)-(i) Simulated STEM images of 1L, 2L and 5L areas corresponding to (d)-(f). Scale bars: 0.5 nm. In the
ACS Paragon Plus Environment
8
Page 9 of 19 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
Nano Letters
overlapping atomic structures in (b), (d), (g), (h) and (i), the dark blue and yellow spheres indicate Mo and S atoms, respectively. Next, we investigated the changes in electrical properties for MoS2 with different number of layers accompanied with the phase conversion. Back-gated FETs were fabricated on CVD-grown 1L, 2L and 6L MoS2, respectively. The as-made 1L, 2L and 6L MoS2 FETs all behaved as n-type semicondutors with on/off current ratios of > 10-6. After being treated with 0.8 M n-butyllithium solution for 24 h, the conductance of FETs made on 2 and more layered MoS2 lost gate voltage dependence, indicating that they were transformed into metallic 1T' phase (Figure 3b and S7). However, all 1L MoS2 devices still exhibited semiconducting characteristics (Figure 3a) and the mobility of the devices increased from ~7 cm2·V-1·s-1 to ~12 cm2·V-1·s-1 with the on/off current ratio almost unchanged after the treatment. This performance enhancement is likely due to the reduced work function of the contact metals by the formation of surface dipole layer as a result of the charge transfer from the alkali atoms to the metals.17 1L MoS2 devices can only be converted into metallic characteristics at very harsh condition, such as being soaked in 1.6 M nbutyllithium for 5 days (Figure S8). Therefore, electrical measurements also suggested that 1L MoS2 showed very high resistance against 2H→1T' conversion.
ACS Paragon Plus Environment
9
Nano Letters 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 19
Figure 3. Electrical performance of 2D MoS2 devices. Ids–Vgs characteristics of a representative (a) 1L and (b) 6L MoS2 FETs before and after being treated with 0.8 M n-butyllithium for 24 h at Vds = 1 V. Insets: optical images of the measured FETs after treatment. (c) Ids–Vds curve of a rectifying diode fabricated on a 2H (1L)–1T' (9L) MoS2 junction measured at Vgs = 0 V. Inset: Schematic and optical image of the device. (d) Statistics of electron mobility of Cr/Au contacted 1L MoS2 FETs (blue columns) and 1T' (ML)-MoS2 contacted 1L MoS2 FETs (red columns), respectively. Inset: Schematic and optical image of the typical FET. Scale bars: 5 µm. As we can construct 1T'-2H (M-S) junctions on the same MoS2 flake by utilizing the dramatically different phase transition tendency between 1L and more layered MoS2 without any lithographic process, we further fabricated rectifying diodes and all 2D FETs using the
ACS Paragon Plus Environment
10
Page 11 of 19 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
Nano Letters
chemically converted 2D M-S junctions as the key building blocks. Typical Ids-Vds characteristics of the rectifying diode made on a 1T' (9L)-2H (1L) MoS2 junction exhibited a forward/reverse current ratio of ~3×103 (Figure 3c), demonstrating the feasibility of fabricating rectifying diodes based on the 2D hetero-phase junctions. In addition, the 1L MoS2 flakes partially covered with two ML MoS2 islands can be transformed into all-MoS2 FETs by treating them with nbutyllithium to convert the ML MoS2 islands into 1T' phase, which can be served as source and drain electrodes (Figure 3d inset), and the 1L MoS2 worked as the semiconducting channel. This kind of all 2D FETs showed better device performance than the devices with metal contacts due to the reduced contact resistance (Figure 3d),18 holding the promise for constructing high performance ultrathin nanoelectronic devices. To reveal the origins of the layer-dependent chemically induced structural transformation of MoS2, we first conducted spin-polarized DFT calculations on the relative stability of the two phases for MoS2 with varied number of layers upon electron injection.19 We observed that the energy difference for the two phases of MoS2 (△E = E2H - E1T') showed layer-dependent behavior both in neutral and electron doping states. △E of neutral MoS2 increased with the decreasing number of layers, indicating the increasing difficulty for the phase transition of thinner MoS2 (Figure S9). In the negatively doping state, △E decreased with increasing electron concentration for MoS2 with any number of layers and the 1T' phase became thermodynamically more stable than the 2H phase after the excessive electrons reached a critical concentration (Figure 4a). Under low doping electron concentrations, △E of MoS2 regardless of the number of layers decreased at almost the same rate, but when more electrons were injected, △E of 1L MoS2 decreased significantly more slowly than the thicker ones, which resulted in the much higher Cc to trigger the 2H→1T' phase transition in 1L MoS2. The reason of this rate difference can be
ACS Paragon Plus Environment
11
Nano Letters 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 19
explained from the density of states (DOS) of MoS2 with varied number of layers (Figure 4d). The DOS of 1L MoS2 exhibited distinct 2D character, so there were significantly more states available in 1L than in ML MoS2 near the edge of conduction band, into which the injected electrons fill. Consequently, the energy of 1L MoS2 in the 2H phase increased much more slowly than that of the thicker MoS2. A similar trend was also observed in 2L, 3L, and 5L MoS2, suggesting that the Cc to trigger the structural phase transition increased with the decreasing thickness of MoS2 (Figure 4a inset), which was highly consistent with the experimental results. Furthermore, the difference between 1L and 2L was evidently larger than that between 2L and 3L, which was also in accord with the experimental results, further verifying the very low phase transition tendency for 1L MoS2.
Figure 4. Layer-dependent phase transition behavior of MoS2 by DFT calculations. (a) Energy difference between 2H and 1T' phases of MoS2 with varied number of layers as a function of
ACS Paragon Plus Environment
12
Page 13 of 19 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
Nano Letters
injected electron concentration. Inset: Critical electron concentrations for the phase transition of MoS2 with varied number of layers extracted from (a). (b) The calculated energy band diagrams of both 2H and 1T' phases of MoS2 with varied number of layers. (c) Comparison of the phase transition barriers for negatively charged 1L (red) and 2L (blue) MoS2. The total energy of 2H phase MoS2 was shifted to zero. (d) Electronic density of Kohn-Sham states for 1L, 2L, 3L and 5L 2H-MoS2, respectively. The black and red dashed lines indicate the Fermi level positions of the intrinsic 2H-MoS2 and the 2H-MoS2 at the Cc with varied number of layers, respectively. The shaded regions depict the additional states occupied in negatively-charged 2H-MoS2. To provide a more intuitive picture of the layer-dependent phase transition of MoS2, we correlated the chemically induced phase transition behaviour with the band structures of MoS2 by DFT calculations. Figure 4b depicted the alignments of valence-band maximum (VBM) and conduction-band minimum (CBM) for the semiconducting 2H phase along with the Fermi level for the metallic 1T' phase of MoS2 with different number of layers. These alignments clearly depicted that 2H→1T' phase transition can be triggered by electron doping because the injected electrons will occupy the CBM of 2H-MoS2 whose energy is > 1 eV higher than the Fermi level of the metallic 1T' phase for all 2H-MoS2 with different number of layers.20 Figure 4b also showed that the band gap of 2H-MoS2 increased with decreasing number of layers, in accordance with the previous reports.4 Accordingly, as the number of layers decreased, the CBM and VBM of the 2H-MoS2 shifted to higher and lower energy, respectively, whereas the Fermi level positions of 1T'-MoS2 almost stayed the same. It indicated that in the chemical doping of 2HMoS2 with decreasing number of layers, electron transferred from n-butyllithium to the CBM of 2H-MoS2 will be increasingly difficult. For negatively charged 2H-MoS2, the excess electrons will occupy CBM (Figure 4d, shaded regions), inducing the upshift of the Fermi level position.
ACS Paragon Plus Environment
13
Nano Letters 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 19
We extracted the Fermi level positions of negatively charged 2H-MoS2 at Cc (Figure 4d, red dash lines), which can be related to the redox potential of Li to trigger the phase transition (∝ RTlnC*, C* is the effective dopant concentration). Remarkably, the Fermi level position of negatively charged 1L MoS2 was 100 meV higher than that of 2L. The difference between 2L and 3L was 40 meV and that between 3L and 5L was 20 meV, which was highly consistent with the experimentally estimated trends. Furthermore, the adsorption or intercalation of Li also showed layer-dependent behavior. The intercalation of Li into ML MoS2 was energetically more favorable than its adsorption onto the surface of 1L MoS2 (Figure S10 and Table S1, see Supporting Information for details). Besides, intercalated Li would be confined to the inter-layer space of ML MoS2, while the adsorbed Li on 1L MoS2 might desorb from the surface, resulting in less efficient doping for 1L MoS2 than ML MoS2, which was in line with the phase transition process where more electrons were needed to trigger the 2H→1T' transition in 1L MoS2 than in ML MoS2 thermodynamically. So both the layer-dependent Cc and the adsorption or intercalation of Li accounted for the experimentally observed layer-dependent phase transition conditions. In addition, it was shown in Figure 1b that the phase transition for 1L MoS2 not only required more injected electrons but also took longer time. To gain insights into the kinetics of the phase transition process, we further conducted calculations on the energy barriers for 2H→1T' phase transition of 1L and 2L MoS2 at CC (0.75 e/MoS2 for 1L and 0.5 e/MoS2 for 2L), respectively. As shown in Figure 4c, the energy barrier for phase transition of 1L MoS2 (0.95 eV) was much higher than that of 2L MoS2 (0.10 eV), indicating that the kinetic rate of phase transition for 1L MoS2 was much slower than that of 2L MoS2, which was in good agreement with our experimental results. The lowering of energy barriers in 2L MoS2 were attributed to the interlayer van der Waals interactions, which stabilized the transition states. It was also noticed that
ACS Paragon Plus Environment
14
Page 15 of 19 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
Nano Letters
there were two energy barriers in 2L MoS2 corresponding to the structural transformation of the two individual layers, respectively, indicating the layer-by-layer phase transition mechanism in ML MoS2. In summary, we revealed the layer-dependent phase transition behavior of 2D MoS2 by performing both experimental and theoretical studies. MoS2 flakes with different number of layers showed greatly different CC, Li adsorption or intercalation efficiency and doping duration to induce the 2H→1T' structural transformation, decreasing with increasing number of layers. The layer-dependent Fermi levels of negatively charged 2H-MoS2 and energy barriers for phase transition further revealed the physical origins of the layer-dependent phase transition behaviors for MoS2. Based on this layer-dependent 2H→1T' conversion, controllable phase transition can be realized to construct hetero-phase M-S 2D junctions for fabricating ultrathin electronic devices. Our work revealed the correlations between the electronic structures and the chemically induced phase transition of 2D MoS2 with varied number of layers and similar layer-dependent behaviors can be expected in other chemical phenomena involved charge transfer for 2D TMDCs and thus provides new insights into the chemical modulations of 2D atomic crystals to further enrich their functionalities and extend possible applications. ASSOCIATED CONTENT Supporting Information. Experimental details. As-grown 2H-MoS2 flakes with varied number of layers. Raman characterizations of the phase transition of MoS2. More STEM characterizations of MoS2 flakes. More electrical data of 1L and 2L MoS2 devices after nbutyllithium treatment. Layer-dependent energy difference between 2H and 1T' phases of neutral
ACS Paragon Plus Environment
15
Nano Letters 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 19
MoS2. Adsorption of Li on 2H-MoS2 with varied number of layers. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author *
[email protected],
[email protected],
[email protected] Author Contributions L. J. and L. S. conceived and designed the experiments. L. S., J. Z., Z. L, and L. L. performed the experiments. X. Y., S.-p. G., X. P. and P. W. performed the STEM characterizations. Z. W., D. W. and H. Y. conducted the density functional theory calculations. L. J. and L. S. co-wrote the paper. All authors discussed the results and commented on the manuscript.
L. S. and X. Y. contributed equally to this work. Notes The authors declare no competing financial interest. ACKNOWLEDGMENT L. J. acknowledges National Natural Science Foundation of China (No. 51372134, No. 21573125, No. 21322303) and Tsinghua University Initiative Scientific Research Program. P. W. acknowledges funding from the National Natural Science Foundation of China (No. 11474147), the National Basic Research Program of China (Grant No. 2015CB654901) and the Natural Science Foundation of Jiangsu Province (Grant No. BK20151383). Z. W. was financially supported by the National Natural Science Foundation of China (No. 11474047). This work was carried out at National Supercomputer Center in Tianjin, and the calculations were performed on TianHe-1(A).
ACS Paragon Plus Environment
16
Page 17 of 19 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
Nano Letters
REFERENCES 1. Chhowalla, M.; Shin, H. S.; Eda, G.; Li, L. J.; Loh, K. P.; Zhang, H. Nat. Chem. 2013, 5, 263275. 2. Huang, X.; Zeng, Z. Y.; Zhang, H. Chem. Soc. Rev. 2013, 42, 1934-1946. 3. Mak, K. F.; Lee, C.; Hone, J.; Shan, J.; Heinz, T. F. Phys. Rev. Lett. 2010, 105, 136805. 4. Splendiani, A.; Sun, L.; Zhang, Y. B.; Li, T. S.; Kim, J.; Chim, C. Y.; Galli, G.; Wang, F. Nano Lett. 2010, 10, 1271-1275. 5. Zheng, J.; Yan, X.; Lu, Z.; Qiu, H.; Xu, G.; Zhou, X.; Wang, P.; Pan, X.; Liu, K.; Jiao, L. Adv. Mater. 2017, 29, 1604540. 6. Li, S. L.; Komatsu, K.; Nakaharai, S.; Lin, Y. F.; Yamamoto, M.; Duan, X. F.; Tsukagoshi, K. ACS Nano 2014, 8, 12836-12842. 7. Hirsch. A.; Hauke, F. Angew. Chem. Int. Ed. DOI:10.1002/anie.201708211. 8. Sun, L. F.; Chen, C. H.; Zhang, Q. H.; Sohrt, C.; Zhao, T. Q.; Xu, G. C.; Wang, J. H.; Wang, D.; Rossnagel, K.; Gu, L.; Tao, C. G.; Jiao, L. Y. Angew. Chem. Int. Ed. 2017, 56, 8981-8985. 9. Zhang, R. Y.; Tsai, I. L.; Chapman, J.; Khestanova, E.; Waters, J.; Grigorieva, I. V. Nano Lett. 2016, 16, 629-636. 10. Cai, L.; He, J. F.; Liu, Q. H.; Yao, T.; Chen, L.; Yan, W. S.; Hu, F. C.; Jiang, Y.; Zhao, Y. D.; Hu, T. D.; Sun, Z. H.; Wei, S. Q. J. Am. Chem. Soc. 2015, 137, 2622-2627. 11. Qian, X. F.; Liu, J. W.; Fu, L.; Li, J. Science 2014, 346, 1344-1347. 12. Voiry, D.; Mohite, A.; Chhowalla, M. Chem. Soc. Rev. 2015, 44, 2702-2712. 13. Gao, G. P.; Jiao, Y.; Ma, F. X.; Jiao, Y. L.; Waclawik, E.; Du, A. J. J. Phys. Chem. C 2015, 119, 13124-13128. 14. Wang, X. S.; Feng, H. B.; Wu, Y. M.; Jiao, L. Y. J. Am. Chem. Soc. 2013, 135, 5304-5307.
ACS Paragon Plus Environment
17
Nano Letters 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 19
15. Sandoval, S. J.; Yang, D.; Frindt, R. F.; Irwin, J. C. Phys. Rev. B 1991, 44, 3955-3962. 16. Heising, J.; Kanatzidis, M. G. J. Am. Chem. Soc. 1999, 121, 638-643. 17. Lang, N. D. Phys. Rev. B 1971, 4, 4234-4243. 18. Kappera, R.; Voiry, D.; Yalcin, S. E.; Branch, B.; Gupta, G.; Mohite, A. D.; Chhowalla, M. Nat. Mater. 2014, 13, 1128-1134. 19. Sun, X. L.; Wang, Z. G.; Li, Z. J.; Fu, Y. Q. Sci. Rep. 2016, 6, 26666. 20. Li, Y.; Duerloo, K. A. N.; Wauson, K.; Reed, E. J. Nat. Commun. 2016, 7, 10671.
ACS Paragon Plus Environment
18
Page 19 of 19 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
Nano Letters
Table of Contents
49.4 mm × 35 mm (150 × 150 DPI)
ACS Paragon Plus Environment
19