Subscriber access provided by LAURENTIAN UNIV
Article
Recoil Effect and Photoemission Splitting of Trions in Monolayer MoS2 Qicheng Zhang, Carl H. Naylor, Zhaoli Gao, Ruizhe Wu, Irfan Haider Abidi, Meng-Qiang Zhao, Yao Ding, Aldrine Abenoja Cagang, Minghao Zhuang, Xuewu Ou, and Zhengtang Luo ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.7b03457 • Publication Date (Web): 09 Nov 2017 Downloaded from http://pubs.acs.org on November 10, 2017
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 free 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 accessible to all readers and 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.
ACS Nano 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 23
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
ToC graphic 82x44mm (300 x 300 DPI)
ACS Paragon Plus Environment
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
Recoil Effect and Photoemission Splitting of Trions in Monolayer MoS2 Qicheng ZHANG1, Carl H. NAYLOR2, Zhaoli GAO2, Ruizhe WU1, Irfan Haider ABIDI1, Meng-Qiang Zhao2, Yao DING1, Aldrine Abenoja CAGANG1, Minghao ZHUANG1, Xuewu OU1, Zhengtang LUO1,* 1
Department of Chemical and Biomolecular Engineering, the University of Hong Kong Science and
Technology, Clear Water Bay, Kowloon, Hong Kong 2
Department of Physics and Astronomy, University of Pennsylvania, 209S 33rd Street, Philadelphia,
Pennsylvania 19104 6396, USA *
Email:
[email protected] Abstract The 2D geometry nature and low dielectric constant in transition metal dichalcogenides (TMDCs) lead to easily formed strongly bound excitons and trions. Here, we studied the photoluminescence of van der Waals heterostructures of monolayer MoS2 and graphene at room temperature and observed two photoluminescence peaks that are associated with trion emission. Further study of different heterostructure configurations confirms that these two peaks are intrinsic to MoS2 and originate from a bound state and Fermi level, respectively, of which both accept recoiled electrons from trion recombination. We demonstrate that the recoil effect allows us to electrically control the photon energy of trion emission by adjusting the gate voltage. In addition, significant thermal smearing at room temperature results in capture of recoil electrons by bound states, creating photoemission peak at low doping level whose photon energy is less sensitive to gate voltage tuning. This discovery reveals an unexpected role of bound states for photoemission, where binding of recoil electrons becomes important. 1 ACS Paragon Plus Environment
Page 2 of 23
Page 3 of 23
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
Keywords: molybdenum disulfide, photoluminescence, trion, 2D materials, doping Monolayer semiconducting transition metal dichalcogenides (TMDCs) are a family of semiconductors with direct bandgaps,1-3 in which the strong light-matter interaction and optical properties make them as promising materials for optoelectronics.4-6 The low dielectric constant and the two-dimensional geometry of TMDC monolayers lead to a poorly screened Coulomb potential, which lead to the formation of excitons and trions due to the strong many-body effects.7-10 On the other hand, the direct band gap at K point has valley difference and large spin splitting at valence band due to the breaking of inversion symmetry and spin-orbital coupling.2,11-14 Nevertheless, significant differences exist between excitons and trions. For instance, the excitons are reported to emit coherently while the exchange-splitting in trions destroys such coherence.10 Moreover, Raman signal is enhanced when the emission energy is in resonance with neutral exciton emission but such phenomenon has not been observed in trions.15 To understand unusual optical properties of TMDCs and further development of optoelectronic devices, it is imperative to clarify the physics and methods that can be used to control the interactions between electrons, excitons, phonons and defects. Among the TMDCs family, molybdenum disulfide (MoS2) is the most widely studied, as its large crystals is found in natural minerals16 and more importantly, large monolayer MoS2 is produced by chemical vapor deposition (CVD).17 To use MoS2 for sensing applications, the main method is to control their electrical or optical response with doping, including electrochemical doping or doping with the target molecules.18-20 However, for trion emission, only intensity is found to be tunable at room temperature, but photon energy is also tunable at low temperature.7 Some studies at room temperature thus regard photon energy of trions differs to excitons by a constant binding energy term during the electrical tuning,21 while low temperature study reports that this 2 ACS Paragon Plus Environment
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
energy difference strongly increases with MoS2 Fermi energy due to the recoil effect.7 In this paper, we discovered two trion-like photoemission peaks at room temperature. However, photon energy of one is sensitive to electrical tuning, while the other does not. These two peaks originate from two recoil processes upon trion recombination, where the states accepting recoiled electrons are a bound state and the Fermi level respectively. The presence of bound states leads to an additional final state for recoiled electrons from trions, thus splitting photoemission for trions. This discovery indicates the importance of bound states in MoS2 due to the recoil effect of trion recombination, where the capability of electrical tuning and temperature dependent behavior are significantly influenced.
Results and Discussion In order to study the photoluminescence (PL) of monolayer MoS2, we firstly fabricated the FET devices by using heterostructures of graphene and MoS2. To facilitate a high electron doping level, we resorted to ion gate in liquid phase, where the high capacitance of ion electric double layer makes it possible to achieve high carrier density at relatively low gate voltage. Graphene layer is used as gate electrode to provide a simple gating. Figure 1A illustrates the configuration of the MoS2/graphene heterostructure device. Briefly, this device is a three-electrode electrochemical cell that resembles a scanning electrochemical microscope,22 in which the MoS2/graphene heterostructure acts as the working electrode, directly facing the counter electrode, i.e. the ITO glass. Here, graphene, MoS2 (and hBN) were all grown by chemical vapor deposition (CVD) using established procedure routinely performed in our laboratory.23,24 Secondly, MoS2 and graphene were layer-by-layer transferred by using the standard PMMA-assisted wet transfer method to a 1cm×1cm silicon wafer with 300 nm silicon oxide layer.24 Thirdly, a patterned PDMS channel was placed on 3 ACS Paragon Plus Environment
Page 4 of 23
Page 5 of 23
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 silicon wafer, followed by stacking with a piece of transparent ITO glass, which allows the laser to excite the heterostructure through the ITO window. Lastly, an electrolyte solution was injected into the pre-defined channel, which guided into an electrolyte reservoir installed with a reference electrode. The electrolyte solution connects to the reference electrode situated at the reservoir and flow between the MoS2/graphene and ITO glass. Before the fabrication of devices, a thorough characterization was performed to assure the quality of the 2D materials. The layer number of graphene was determined by its Raman spectra, along with atomic force microscopy (AFM) measurements. The measured full width half maximum (FWHM) of its 2D peak (at ~ 2700 ), is ~30 , indicating the monolayer nature.25 This is corroborated by that fact that its 2D peak intensity is much higher than that of its G peak (~1580 ) (Figure 1B) with 514 nm excitation wavelength, typical for monolayer graphene.25 Similarly, the layer number of MoS2 was confirmed by its Raman spectroscopy and AFM. Majority of the flakes are found to be monolayer, showing the signature 18 in difference between and modes,26,27 as shown in Figure 1B, except in the center part of the MoS2 triangle, the where the and modes apart more than 20 (Figure 1C), indicating the presence of
multilayer. The layer number was further confirmed by AFM measuring, showing a height of ~0.7 nm per layer (Figure 1D). After stacking graphene and MoS2, and modes of MoS2 has
become further apart, and FWHM of 2D peak of graphene are increased to around 40 (Figure 1B) due to strain and interlayer interactions.26,28 Figure 2A illustrates the resultant PL spectra at different doping level of the MoS2/graphene stack with 632.8 nm laser excitation. Electron doping level is tuned by adjusting positive gate voltage ( ) (also refer to Supporting Information Section 5 and Figure S8 for MoS2 Raman 4 ACS Paragon Plus Environment
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
change). At low doping level, those PL spectra of the heterostructure display similarities to that of bare MoS2,
7
and a previously reported MoS2/graphene heterostructure.21 A readily observable
feature in the spectrum is a very large peak appearing between 1.76 - 1.92 eV, centered at 1.85 eV (hereafter referred to as Peak I). Another smaller peak centered ~1.82 eV (Peak II) become prominent when is in the range of 0.26 and 0.31 V. When equals 0.36 V or higher, a third peak (Peak III) appears at much lower photon energy, concurrent with the gradual disappearing of Peak II and I. To further elucidate the nature of the bound states, we measured the linear polarization dependent PL using two linear polarizers at excitation and detection light path respectively. The linearly polarized PL measurement result is summarized in Figure S9B. This figure shows a linearly polarized peak, and non-polarized two humps, consistent with the hypothesis of the existence of three peaks (Supporting Information Section 5). Based on this observation, the PL spectra are de-convoluted accordingly as shown in Figure 2B. As we can see from the de-convoluted peaks, in contrast to Peak I and II, where the peaks are almost fixed, Peak III significantly shifted, and the variation is strongly dependent on the gate voltage. These features are different from that described in the literature7,21 where only two peaks from exciton (formed by an electron and a hole shown in Figure 2C) and trion (formed by two electrons and a hole shown in Figure 2C) are present in this range of photon energy. To better illustrate the variation of photon energies and emission intensities of Peaks I, II and III, we plot the intensity map against of the three separated peaks shown in Figure 2D (combined mapping in Figure S2). The separation of Peak II and III is evident combining Figure 2B and the different trends upon doping. We observe that the highest intensities of Peaks I and II are located at of 0.20 and 0.35 V, respectively. Significant difference is found for Peak III, whose 5 ACS Paragon Plus Environment
Page 6 of 23
Page 7 of 23
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
intensities remain almost unchanged across a wide range of from 0.3 to 0.8 V. If we only consider the range of
