Defect Coupling and Sub-Angstrom Structural ... - ACS Publications

Subscriber access provided by UNIV OF TASMANIA

Communication

Defect Coupling and Sub-Angstrom Structural Distortions in W MoS Monolayers 1–x

x

2

Amin Azizi, Yuanxi Wang, Greg Stone, Ana Laura Elías, Zhong Lin, Mauricio Terrones, Vincent H. Crespi, and Nasim Alem Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.6b05045 • Publication Date (Web): 13 Mar 2017 Downloaded from http://pubs.acs.org on March 14, 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.

Nano Letters 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 25

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

Defect Coupling and Sub-Angstrom Structural Distortions in W1–xMoxS2 Monolayers Amin Azizi1,2,3, Yuanxi Wang2,3,4, Greg Stone1, Ana Laura Elias2,4, Zhong Lin2,4, Mauricio Terrones1,2,3,4,5, Vincent H. Crespi1,2,3,4,5, and Nasim Alem1,2,3 1

Department of Materials Science and Engineering, The Pennsylvania State University, University Park, Pennsylvania 16802, USA 2 Center for 2-Dimensional and Layered Materials, The Pennsylvania State University, University Park, Pennsylvania 16802, USA 3 Materials Research Institute, The Pennsylvania State University, University Park, Pennsylvania 16802, USA 4 Department of Physics, The Pennsylvania State University, University Park, Pennsylvania 16802, USA 5 Department of Chemistry, The Pennsylvania State University, University Park, Pennsylvania 16802, USA ABSTRACT: Two-dimensional materials offer a remarkably rich materials platform to study the origin of different material behaviors at the atomic level, and doping provides a key means of tailoring such materials’ functional properties. The local atomic structure around such dopants can be critically important in determining the material behavior as it could modulate scattering, catalytic activity, electronic and magnetic properties, etc. Here, using aberration-corrected scanning transmission electron microscopy with subÅngstrom resolution in conjunction with density functional theory calculations, we demonstrate a strong coupling between Mo dopants and two types of defects in WS2 monolayers: sulfur mono-vacancies and grain boundaries. Although Mo does occupy a transition metal lattice site, it is not an ideal substitutional dopant: ~80% of the S vacancies identified by STEM co-localize with Mo dopants, an affinity that appears to be enhanced by symmetry breaking of a partially occupied mid-gap defect state. Although a Mo dopant by itself does not considerably distort the WS2 lattice, it induces substantial lattice deformation by apparently facilitating the charging of a sulfur mono-vacancy paired with it, which is consistent with the results of first-principles calculations. This coupling of foreign substitutional dopants with vacancies could potentially be exploited to control the distribution and location of chalcogenide vacancies within transition metal dichalcogenides, by segregating vacancies into regions of high Mo concentration that are purposely placed away from active regions of TMD-based devices. KEYWORDS: Doped monolayer transition metal dichalcogenides, atomic and chemical structures, defect coupling, sub-Ångstrom structural distortion

ACS Paragon Plus Environment

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

Doping can have a profound effect on the physical, chemical and mechanical properties of materials, due to both new chemistry at the dopant sites and the structural distortions of crystalline lattice in the dopant’s vicinity; these effects can modify electronic structure1, tune band gaps2,3, improve catalytic properties4, and introduce magnetism5,6. For example, while two-dimensional (2D) transition metal dichalcogenides (TMDs) have high catalytic efficiency for the hydrogen evolution reaction7,8 and hydrodesulfurization9, doped few-layer MoS2 show higher catalytic activity for hydrogen evolution than does their pristine state4. To better nanoengineer the functional properties of doped crystals, it is crucial to directly probe dopant atoms10–12, lattice defects13–15, and their interactions. While great progress has been made recently in the theory, synthesis, and optical and electronic properties of doped 2D TMDs2,5,16–18, little is known about the atomic structure and local chemistry at the dopant sites. This study explores the relaxation effects in the atomic and electronic structure of doped 2D TMDs using aberration-corrected scanning transmission electron microscopy (AC-STEM) and first-principles density functional theory (DFT) calculations. We discover strong interactions between dopants and lattice defects in doped WS2 monolayers and find significant sub-Angstrom distortions around individual Mo dopant atoms when they are coupled with sulfur vacancies. The observed structural distortions and dopant/defect affinity can modulate the electronic structure of monolayer TMDs. Such modulations in the atomic and electronic structure could lead, for example, to higher local chemical reactivity at the Mo sites. While an affinity between sulfur vacancies in MoS2 or WS2 has been previously reported19,20, as well as interaction between Re dopants and other point defects such as C adatoms and sulfur vacancies21,22, here we describe an affinity between two distinct defect types: sulfur vacancies and


Page 2 of 25

Page 3 of 25

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

isoelectronic substitutional dopants (such as Mo in WS2). Such an interaction could potentially be used to remove undesirable S vacancies from the active region of a device by collecting them in an inactive, Mo-rich region of the crystal.

Figure 1. (a) Atomic-resolution ADF-STEM image of a doped WS2 monolayer showing several Mo dopant atoms in the lattice. (b) Magnified image of the Mo dopant marked by the yellow arrow. (c) Displacement map generated by calculating the deviations of metal atom distances in (b) from the ideal nearest metal-metal neighbor distance. (d,e) Local structure near the Mo defect obtained from DFT calculations for neutral and charged (–2|e| charge per defect) systems. Figure 1a shows an atomic-resolution ADF-STEM image of a substitutionally Mo-doped WS2 monolayer, in which Mo dopant atoms are embedded in the WS2 lattice. Optical characterization of the doped monolayer crystals (i.e. optical images and


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

photoluminescence and Raman spectroscopy analysis) can be found in supplementary material Figure S1. A primary beam energy of 80 kV and a beam current of 49 pA were used for imaging to minimize damage to the crystal23. The collected signal in ADFSTEM images is proportional to the atomic number (Z) of the atoms, through which we can identify the chemical identity of all atoms in the lattice24. The atom marked by the yellow arrow in Figure 1a has a distinct intensity: it is a Mo atom (Z=42) within a lattice of tungsten (Z=74) and sulfur (Z=32 for the “stack” of two sulfur atoms) as indicated by the intensity line profile. Looking more broadly across the ADF-STEM image, it is evident that W and Mo consistently occupy the transition metal sublattice, while S occupies the chalcogen sublattice. The bandpass filtered ADF-STEM image of Figure S2 accentuates the distribution of Mo atoms within the WS2 lattice. In order to further probe local structural relaxations introduced by the dopants, the metalmetal bond lengths around the Mo atoms are quantified. Figure 1(b-c) provides a magnified ADF-STEM image of an isolated Mo substitutional dopant and its displacement map through direct image analysis (see Methods). Figure 1c depicts the deviation of each metal-metal nearest-neighbor distance from the ideal metal-metal separation in pristine WS2. The changes in metal-metal separation around this Mo dopant are negligible, i.e. smaller than the measurement error of ~8 pm. DFT calculations for a substitutional Mo dopant in WS2 are consistent with this result: the maximum calculated change in metal-metal separation near the Mo dopant is only 1.5 pm, for both neutral (Fig. 1d) and charged defects (Fig. 1e, for a defect charged at q = –2, in units of the elementary charge |e|). Here, the vanishing strain is due to the nearly identical lattice constants for MoS2 and WS2, consistent with the calculated and observed lattice


Page 4 of 25

Page 5 of 25

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

structures in WxMo1-xS2 alloys, which have uniform metal-metal separation3,12,25,26.

Figure 2. (a) ADF-STEM image of a doped WS2 monolayer with a substitutional Mo dopant and a line profile through the row of S atoms marked by white arrows. Orange arrows reveal a sulfur vacancy next to the Mo site. Band structures of monolayer WS2 with (b) an isolated sulfur vacancy and (c) a sulfur vacancy that is next to a Mo dopant. The blue (red) components of the band colors indicate the projection of the calculated wave-function onto the atomic orbitals of the Mo (W) atom(s) next to the sulfur vacancy. (d) Sulfur vacancy formation energy on a sulfur site neighboring 0,1,2, and 3 Mo atoms are plotted in red and light, medium, dark blue, as a function of the Fermi energy (relative to the conduction band minimum). While analyzing ADF-STEM images, we noticed sulfur vacancies next to Mo dopant atoms. For example, Figure 2a shows a substitutional Mo dopant plus an intensity line profile along the indicated direction that reveals a sulfur vacancy immediately adjacent next to the Mo atom (see supplementary material Fig. S3 for the S-vacancy analysis). In total we detected 70 S vacancies, 55 of which are next to a Mo dopant (see


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

supplementary material Fig. S3-S4). Sulfur atoms can be ejected from the TMD lattice during continuous imaging using an 80 keV electron beam23. To ensure that sulfur vacancies are not created during imaging under electron irradiation, the beam current was reduced and multiple images were acquired from the same region and examined in order to determine sulfur vacancy formation and its dynamics under the electron beam. This analysis did not show any sulfur vacancy formation or dynamics in the acquired image series and confirmed stability of the sample under the imaging conditions. By calculating the energies of a sulfur mono-vacancy as a function of distance from a Mo dopant within density functional theory, we find an affinity between the two defects of only 0.02 eV in the charge-neutral case, but rising to 0.15 and 0.23 eV for systems with a charge of q = – 1 and –2 per S vacancy. The strong affinity between Mo and S when they are negatively charged is consistent with the experimental observations, given that the sample is under electron beam irradiation during imaging and that CVD-grown WS2 monolayers generally display n-type behavior16,27. The origin of this affinity can be explained in terms of the splitting of partially filled defect bands due to symmetry breaking. An isolated S vacancy introduces two degenerate mid-gap states in the WS2 band structure with dxy+dx2−y2 orbital character, as shown in Figure 2b for the charge-neutral case. When one of the W atoms is next to a S vacancy and it is replaced by Mo, the in-plane threefold symmetry is broken and the defect band splits into two levels that are 0.15 eV apart (as shown in Fig. 2c). The lower (upper) level has strong Mo (W) d orbital character, as indicated by the blue (red) components of the band colors. Since the first two added electrons preferentially occupy the lower level, the charged system prefers to locate sulfur vacancies near Mo dopants. Similar calculations for the complementary system


Page 6 of 25

Page 7 of 25

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

(i.e. W dopants in MoS2) show the reverse: W dopants repel negatively charged sulfur mono-vacancies. Similar effects have been proposed previously28 when tuning the defect level for Se vacancies in charge-neutral WxMo1-xSe2 alloys: in contrast to sulfur vacancies, selenium vacancies have an affinity of 0.5 eV for W defects under even charge-neutral conditions. The affinity between a Mo defect and a S vacancy relies on the vacancy being charged. Does the presence of Mo defects actively facilitate the charging of S vacancies? To investigate this question, we calculated the formation energies of a vacancy charged at q = 0,–1,–2, with and without neighboring Mo defects, as a function of the Fermi energy, as shown in Fig. 2d (the Methods for details). The Fermi energy above which a state of charge q can be stabilized against the q+1 state, called the thermodynamic charge transition energyε(q+1/q), can be determined by the energy where the crossover from q+1 to q occurs29. Since charge transition energies are independent of the sulfur chemical potential µS, we chose a µS for the calculation but leave the vertical axis unmarked. For pure WS2, ε(0/–1) and ε(–1/–2) occur at –0.27 and 0.36 eV relative to the conduction band edge, consistent with previous studies on MoS222. When one, two, and all three of the W sites neighboring the sulfur vacancy are replaced by Mo, ε(0/–1) lowers by 0.10, 0.18, and 0.22 eV and ε(–1/–2) drops by 0.06, 0.19, and 0.22 eV, as shown by the shift of the light, medium, and dark blue lines from the red line in Fig. 2d. That is, the Fermi level required to stabilize sulfur vacancies becomes lower as more Mo defects become available near the sulfur vacancies. Hybrid functional calculations (which presumably more accurately handles the energies of the occupied mid-gap states) are expected to down-shift ε(–1/–2) for vacancies in WS2 by ~0.3 eV (based on a study on vacancies in


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

MoS222). Thus the q = –2 state may be stabilized with a Fermi energy within the band gap. The variety of charge transition levels induced by S-Mo defect pairs discussed here may leave distinct optical signatures (that are related to optical transitions to and from the charged states) compared with that of simple S vacancies, such as red-shifted or broadened peaks in absorption or photoluminescence spectroscopy. The ability of sulfur vacancies to find their optimal locations (i.e. adjacent to Mo dopants) may vary with the kinetics of sulfur incorporation, defect concentration, and sulfur vacancy mobility during growth and post-growth treatments. These are thus key parameters to consider in optimizing the spatial distribution of SV in Mo/W alloy samples produced by different methods30,31.


Page 8 of 25

Page 9 of 25

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

Figure 3. (a) Atomic-resolution ADF-STEM image of Mo-doped WS2 and intensity line profile for S atoms along the zigzag direction indicated by white arrows, confirming the presence of a sulfur vacancy next to the middle Mo atom. (b) Measured displacement map for the region marked by the red box in (a). Displacements near the Mo defect obtained by DFT calculations when the unpaired sulfur atom at the sulfur vacancy site is: (c) in the upper chalcogen plane or (d) in the same plane as the transition metal atoms, as highlighted by black open circles. (e) Calculated energy profile for the unpaired sulfur atom moving through the TMD mid-plane. We similarly measured lattice distortions when the Mo dopant is paired with a sulfur vacancy, with results shown in Figure 3. Three Mo dopants sit within the marked box, each surrounded by W atoms. Detailed image analysis (see Fig. S5) reveals a sulfur vacancy next to only the middle Mo atom. Unlike the case of an isolated Mo dopant, we now see substantial structural distortions, with metal-metal separations expanding by ~60 pm on one side of this Mo atom and shrinking by ~54 pm on the other side, as compared to the pristine W–W distance of 3.18 Å. Note further that earlier experimental observations showed only very small (~3 pm) distortions around isolated sulfur mono-


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

vacancies in MoS232, and we did not see any noticeable distortion around an isolated Mo dopant with the absence of S vacancy (Figure 1). Substantial lattice distortion is observed only when the vacancy and the dopant atom are coupled. Although our DFT calculations show increased local distortions near a Mo dopant with a neighboring sulfur vacancy (Fig. 3c), they show bond contraction next to the sulfur vacancy, which is contrary to the experimental observations. We explored computationally multiple possible defect configurations, in terms of the number of sulfur vacancies neighboring a Mo defect (1,2, or 3 vacancies), extra charge (from q = –1 to –4), and uniform tensile strain (2% and 4%); however, neither of these structures yielded local lattice dilations close to ~60 pm, nor did other defect structures reported previously14,33–35. The sulfur mono-vacancy has one additional degree of freedom: the z coordinate of the remaining unpaired sulfur atom. By nudging the unpaired S atom (highlighted by open black circles in Fig. 3c-e) into the TMD midplane, we found a metastable structure emerging at q = –2, as shown in Figure 3d. The energy profile of a sulfur atom moving through the mid-plane was calculated using the Nudged Elastic Band (NEB) method36–38 as shown in Figure 3e. Although the metastable state is 2.62 eV less favorable than having the S atom buckled out of the midplane, this energy difference drops to 0.96 eV when a 4% uniform tensile strain is applied, and to 0.79 eV when the strained case was examined using the HSE06 hybrid functional39. Moreover, the energy barrier needed to leave this metastable state increases to 0.29 eV with this strain level. A strain of this magnitude yields a change in the lattice constant of only 13 pm, which lies within the experimental measurement uncertainty of ±8 pm. Long-wavelength strain at this level could potentially be present in the alloy due to gradients in composition, vacancy density, or charge doping level. The requirement of


Page 10 of 25

Page 11 of 25

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

a charged level of q = –2 is consistent with its relative stability discussed earlier: sulfur vacancies that are next to Mo defects are more easily stabilized at q = –2 (to be more accurate, stabilized earlier under increasing n-doping). An out-of-plane structural distortion has been previously observed in other 2D crystals, where the edge atoms at defect sites and edges in h-BN undergo a relaxation in the z direction40,41. When the lone sulfur atom occupies the mid-plane site, the metal-metal separations around the Mo dopant expand by amounts that are consistent with the experimental observations. Keep in mind that the dilation map of Figure 3d assumes 100% occupation of the metastable midplane site: hopping of the S atom between midplane and off-midplane sites on a timescale rapid compared to the TEM exposure time would reduce the measured dilations. In addition, variations in tensile strain or local charge states will also modulate the strength of the local lattice dilations. See supplementary materials for a case with smaller displacements (Fig. S6) and also the calculated displacement maps for different conditions (Fig. S7). It is also important to note that the energy deposited by the electron beam may be a significant contributor to the occupancy of this mid-plane site – its potential occupancy under more assuredly equilibrium conditions remains a topic of further study.


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

Figure 4. (a) ADF-STEM image of a tilt grain boundary in Mo-doped WS2 monolayers. The corresponding fast Fourier transform (inset in upper right) indicates a 26° misorientation between grains. (b) Atomic resolution ADF-STEM image from the grain boundary revealing the 5|7 dislocation structure. (c) A periodic WS2 superlattice containing two grain boundaries consisting of 5|7 dislocations with either S-S or W-W homopolar bonds, and the corresponding energy profile for a single substitutional Mo atom as a function of its distance from the grain boundary. Besides their affinity for isolated point defects (i.e. sulfur mono-vacancies) in the WS2 lattice, Mo dopant atoms were also observed to segregate at line defects (i.e. grain boundaries), where a large population of sulfur mono-vacancies were also found (See


Page 12 of 25

Page 13 of 25

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

Fig. S4d). Figure 4a shows an atomic-resolution ADF-STEM image of a tilt grain boundary in a Mo-doped WS2 monolayer at a misorientation angle of 26°. Multiple Mo atoms sit on the grain boundary, with additional phase-segregated MoS2 regions located a few unit cells away from the grain boundary. The grain boundary contains 5|7 dislocations with S-S homopolar bonds (Fig. 4b), where some of the W atoms have been replaced by Mo atoms. DFT calculations were performed to understand the origin of this segregation of Mo near the grain boundary and inside the dislocation cores. A periodic WS2 structure, containing two grain boundaries with 5|7 dislocations (highlighted by red stripes in Fig. 4c) was examined. Note that the two grain boundaries in the simulation cell are not structurally equivalent: the one at the cell center has S-S homopolar bonds while the one at the cell boundary has W-W bonds42,43. The total energy of the system was calculated as a function of the distance between the grain boundary and a single Mo atom substituting for W, as shown in the lower panel (each energy is obtained after structural relaxation). When compared to the interior of the grain far away from the grain boundaries, Mo atoms show a preference of 0.05–0.15 eV to segregate at either type of grain boundary. The preference extends ~6 Å from the grain boundaries into the interior region. This is consistent with the observation of the Mo atoms at the dislocation cores (Fig. 4b) and segregating near the grain boundary (Fig. 4a).

In summary, aberration-corrected STEM combined with first-principles calculations reveals an affinity between Mo dopant atoms and S vacancies in WS2 monolayers, driving 79% of the observed S vacancies to co-localize with Mo defects. While neither a Mo atom nor a sulfur vacancy by themselves noticeably change the ideal metal-metal


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

separation in the WS2 lattice, significant lattice distortions were observed when the two defects were coupled, suggesting that the coupled dopant+vacancy defect may actually be the active species that controls materials properties that are sensitive to Mo doping level. First-principles calculations agree with the experimentally observed structural distortions and suggest that the lone sulfur atom in this coupled Mo-S defect may actually occupy a mid-plane site. Mo dopant atoms also tend to segregate near the WS2 grain boundaries and sit within 5|7 dislocation cores. The control of chalcogen vacancies appears to be a key issue in obtaining high-quality TMD materials. Our results on defect-defect interactions suggest intriguing strategies to manage the distribution of these chalcogen vacancies: isoelectronic transitional metal substitutional dopants (such as Mo-for-W) located away from key functional areas of a device could, for example, act as “vacancy collectors” that attract chalcogen vacancies to locations where they do not degrade device properties. In addition, the knowledge that sulfur vacancies and isoelectronic metal substitutions have significant mutual affinity will motivate detailed investigations of the catalytic, optical, chemical, and electronic properties of these novel co-defects, and could further motivate exploration of whether similar “co-defect” strategies could be exploited to introduce e.g. magnetically active dopants into 2D chalcogenides.

Materials and Methods: Synthesis and transfer of the Mo-doped WS2 monolayers. One nanometer of MoO3 was initially deposited on a Si/SiO2 substrate through thermal evaporation, followed by deposition of one nanometer of WO3 as a second layer. The sample was then placed in a tube furnace at 800°C and sulfurized for 30 min at atmospheric pressure. For sulfurization, sulfur powders were placed upstream in a region with a local temperature of ~250 ° C with Ar gas flowing at 200 sccm as the carrier gas. The samples were then


Page 14 of 25

Page 15 of 25

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

transferred onto a TEM grid with 2 µm holes using poly(methyl methacrylate) (PMMA) as follows. The samples were spin-coated with PMMA at 3000 rpm for 30 s and then kept overnight at room temperature to cure. After cleaning PMMA off the edges of the sample using a razor blade, the sample was immersed in a 15M KOH solution. Once the polymer/WS2 film was separated from the Si substrate, we fished it out using the TEM grid, washed it with deionized water several times, and dissolved the PMMA layer using acetone. Aberration Corrected Scanning Transmission Electron Microscopy (STEM). Annular Dark Field (ADF) scanning transmission electron microscopy was performed in an aberration-corrected FEI Titan3 60-300 S/TEM to image the atomic structure of the Mo-doped WS2 monolayers. The image acquisition was carried out at 80 kV, using a beam current of 49 pA and a camera length of 115 mm. Bandpass filtering was performed in Digital Micrograph to enhance the contrast of Mo dopant atoms in the lattice more clearly and improve the visibility. Displacement Measurements. To minimize measurement error, we acquired multiple images from the same region near Mo dopants at different scan directions (e.g. 0, 90º). We then corrected for potential STEM drifts in the recorded images using Matlab codes described in the literature44. Next, using Matlab codes developed at Penn State, we determined the position of metal atoms on the drift-corrected images by Gaussian fitting around each peak maximum. We subsequently calculated distances between metal atoms along three zigzag directions by measuring the distance between Gaussian centers. Displacement maps were generated by calculating the deviation of metal atom distances from that in pristine WS2. Theoretical calculations. Density functional theory calculations were performed using the exchange-correlation functional within the generalized gradient approximation (Perdew-Burke-Ernzerhof parametrization, GGA-PBE)45,46. The interaction between valence electrons and ions is described using the projector-augmented wave method47,48. All calculations were performed using the Vienna Ab-initio Simulation Package (VASP)49,50, with an energy cutoff of 400 eV, an electronic loop convergence of 10–5 eV, and force convergence of 0.02 eV/Å. The energy differences between states were converged at a supercell size of 9×5√3×1 (270 atoms), and differed from energy


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

differences using a 5×3√3×1 (90 atoms) supercell by 0.05 eV; therefore the latter dimension was used for all nudged elastic band (NEB) calculations and calculations with the HSE06 hybrid functional due to the more demanding computational resources required. For the NEB calculations, the energy interpolation between images was performed using both the energies and the forces at the images36. Total energies of charged defects were calculated from 6×6×1 supercell geometries containing isolated defects. To handle the error from the interaction of a charged defect with its periodic images and with the positive jellium background, the “special vacuum” method developed for 2D materials is used22,51.

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS The authors would like to acknowledge the Center for 2-Dimensional and Layered Materials (2DLM) at the Pennsylvania State University and support from the National Science Foundation under EFRI 2-DARE awards 1433378, 1433311, and 1542707 and the Materials Innovation Platform program under award DMR-1539916. Y.W., A.L.E., M.T., and V.H.C. acknowledge support from the U.S. Army Research Office under MURI grant W911NF-11-1-0362. The authors also would like to thank Colin Ophus (Berkeley Lab) for helpful discussions. Supporting Information. Details on optical characterization of the doped WS2 monolayers, distribution of Mo dopant atoms in the WS2 lattice, sulfur vacancy analysis around the Mo sites in the WS2 lattice, displacement map for another region of Mo-doped WS2 with coupled defects, and displacement maps from DFT under varying conditions. This material is available free of charge via the Internet at http://pubs.acs.org REFERENCES (1)

Ramasse, Q. M.; Seabourne, C. R.; Kepaptsoglou, D.-M.; Zan, R.; Bangert, U.; Scott, A. J. Nano Lett. 2013, 13 (10), 4989–4995.

(2)

Gong, Y.; Liu, Z.; Lupini, A. R.; Shi, G.; Lin, J.; Najmaei, S.; Lin, Z.; Elías, A. L.; Berkdemir, A.; You, G.; Terrones, H.; Terrones, M.; Vajtai, R.; Pantelides, S. T.; Pennycook, S. J.; Lou, J.; Zhou, W.; Ajayan, P. M. Nano Lett. 2014, 14 (2), 442– 449.


Page 16 of 25

Page 17 of 25

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

(3)

Chen, Y.; Xi, J.; Dumcenco, D. O.; Liu, Z.; Suenaga, K.; Wang, D.; Shuai, Z.; Huang, Y.-S.; Xie, L. ACS Nano 2013, 7 (5), 4610–4616.

(4)

Deng, J.; Li, H.; Xiao, J.; Tu, Y.; Deng, D.; Yang, H.; Tian, H.; Li, J.; Ren, P.; Bao, X. Energy Environ. Sci. 2015, 8 (5), 1594–1601.

(5)

Cheng, Y. C.; Zhu, Z. Y.; Mi, W. B.; Guo, Z. B.; Schwingenschlögl, U. Phys. Rev. B 2013, 87 (10), 100401.

(6)

Mishra, R.; Zhou, W.; Pennycook, S. J.; Pantelides, S. T.; Idrobo, J.-C. Phys. Rev. B 2013, 88 (14), 144409.

(7)

Voiry, D.; Yamaguchi, H.; Li, J.; Silva, R.; Alves, D. C. B.; Fujita, T.; Chen, M.; Asefa, T.; Shenoy, V. B.; Eda, G.; Chhowalla, M. Nat. Mater. 2013, 12 (9), 850– 855.

(8)

Yu, Y.; Huang, S.-Y.; Li, Y.; Steinmann, S. N.; Yang, W.; Cao, L. Nano Lett. 2014, 14 (2), 553–558.

(9)

Tye, C. T.; Smith, K. J. Catal. Today 2006, 116 (4), 461–468.

(10)

Azizi, A.; Wang, Y.; Lin, Z.; Wang, K.; Elias, A. L.; Terrones, M.; Crespi, V. H.; Alem, N. Nano Lett. 2016, 16 (11), 6982–6987.

(11)

Deepak, F. L.; Esparza, R.; Borges, B.; Lopez-Lozano, X.; Jose-Yacaman, M. ACS Catal. 2011, 1 (5), 537–543.

(12)

Dumcenco, D. O.; Kobayashi, H.; Liu, Z.; Huang, Y.-S.; Suenaga, K. Nat. Commun. 2013, 4, 1351.

(13)

Azizi, A.; Zou, X.; Ercius, P.; Zhang, Z.; Elías, A. L.; Perea-López, N.; Stone, G.; Terrones, M.; Yakobson, B. I.; Alem, N. Nat. Commun. 2014, 5, 4867.

(14)

Zhou, W.; Zou, X.; Najmaei, S.; Liu, Z.; Shi, Y.; Kong, J.; Lou, J.; Ajayan, P. M.; Yakobson, B. I.; Idrobo, J.-C. Nano Lett. 2013, 13 (6), 2615–2622.

(15)

Azizi, A.; Eichfeld, S.; Geschwind, G.; Zhang, K.; Jiang, B.; Mukherjee, D.; Hossain, L.; Piasecki, A. F.; Kabius, B.; Robinson, J. A.; Alem, N. ACS Nano 2015, 9 (5), 4882–4890.

(16)

Duan, X.; Wang, C.; Fan, Z.; Hao, G.; Kou, L.; Halim, U.; Li, H.; Wu, X.; Wang, Y.; Jiang, J.; Pan, A.; Huang, Y.; Yu, R.; Duan, X. Nano Lett. 2016, 16 (1), 264– 269.

(17)

Zhang, K.; Feng, S.; Wang, J.; Azcatl, A.; Lu, N.; Addou, R.; Wang, N.; Zhou, C.;


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

Lerach, J.; Bojan, V.; Kim, M. J.; Chen, L.-Q.; Wallace, R. M.; Terrones, M.; Zhu, J.; Robinson, J. A. Nano Lett. 2015. (18)

Zhang, M.; Wu, J.; Zhu, Y.; Dumcenco, D. O.; Hong, J.; Mao, N.; Deng, S.; Chen, Y.; Yang, Y.; Jin, C.; Chaki, S. H.; Huang, Y.-S.; Zhang, J.; Xie, L. ACS Nano 2014, 8 (7), 7130–7137.

(19)

Le, D.; Rawal, T. B.; Rahman, T. S. J. Phys. Chem. C 2014, 118 (10), 5346–5351.

(20)

Komsa, H.-P.; Kurasch, S.; Lehtinen, O.; Kaiser, U.; Krasheninnikov, A. V. Phys. Rev. B 2013, 88 (3), 035301.

(21)

Lin, Y.-C.; Dumcenco, D. O.; Komsa, H.-P.; Niimi, Y.; Krasheninnikov, A. V.; Huang, Y.-S.; Suenaga, K. Adv. Mater. 2014, 26 (18), 2857–2861.

(22)

Komsa, H.-P.; Krasheninnikov, A. V. Phys. Rev. B 2015, 91 (12), 125304.

(23)

Komsa, H.-P.; Kotakoski, J.; Kurasch, S.; Lehtinen, O.; Kaiser, U.; Krasheninnikov, A. V. Phys. Rev. Lett. 2012, 109 (3), 035503.

(24)

Krivanek, O. L.; Chisholm, M. F.; Nicolosi, V.; Pennycook, T. J.; Corbin, G. J.; Dellby, N.; Murfitt, M. F.; Own, C. S.; Szilagyi, Z. S.; Oxley, M. P.; Pantelides, S. T.; Pennycook, S. J. Nature 2010, 464 (7288), 571–574.

(25)

Gan, L.-Y.; Zhang, Q.; Zhao, Y.-J.; Cheng, Y.; Schwingenschlögl, U. Sci. Rep. 2014, 4, 6691.

(26)

Song, J.-G.; Ryu, G. H.; Lee, S. J.; Sim, S.; Lee, C. W.; Choi, T.; Jung, H.; Kim, Y.; Lee, Z.; Myoung, J.-M.; Dussarrat, C.; Lansalot-Matras, C.; Park, J.; Choi, H.; Kim, H. Nat. Commun. 2015, 6, 7817.

(27)

Duan, X.; Wang, C.; Shaw, J. C.; Cheng, R.; Chen, Y.; Li, H.; Wu, X.; Tang, Y.; Zhang, Q.; Pan, A.; Jiang, J.; Yu, R.; Huang, Y.; Duan, X. Nat. Nanotechnol. 2014, 9 (12), 1024–1030.

(28)

Huang, B.; Yoon, M.; Sumpter, B. G.; Wei, S.-H.; Liu, F. Phys. Rev. Lett. 2015, 115 (12), 126806.

(29)

Freysoldt, C.; Grabowski, B.; Hickel, T.; Neugebauer, J.; Kresse, G.; Janotti, A.; Van de Walle, C. G. Rev. Mod. Phys. 2014, 86 (1), 253–305.

(30)

McCreary, A.; Berkdemir, A.; Wang, J.; Nguyen, M. A.; Elías, A. L.; PereaLópez, N.; Fujisawa, K.; Kabius, B.; Carozo, V.; Cullen, D. A.; Mallouk, T. E.; Zhu, J.; Terrones, M. J. Mater. Res. 2016, 31 (7), 931–944.


Page 18 of 25

Page 19 of 25

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

(31)

Terrones, Z. L. and B. R. C. and E. K. and R. L. and R. R. and H. T. and M. A. P. and M.; Lin, Z.; Mccreary, A.; Briggs, N.; Wang, F.; Wang, Z.; Wang, Q.; Lin, Z.; Carvalho, B. R.; Kahn, E.; Lv, R.; Rao, R.; Terrones, H. 2D Mater. 2016, 3 (2), 22002.

(32)

Wang, S.; Lee, G.-D.; Lee, S.; Yoon, E.; Warner, J. H. ACS Nano 2016, 10 (5), 5419–5430.

(33)

Hong, J.; Hu, Z.; Probert, M.; Li, K.; Lv, D.; Yang, X.; Gu, L.; Mao, N.; Feng, Q.; Xie, L.; Zhang, J.; Wu, D.; Zhang, Z.; Jin, C.; Ji, W.; Zhang, X.; Yuan, J.; Zhang, Z. Nat. Commun. 2015, 6, 6293.

(34)

Noh, J.-Y.; Kim, H.; Kim, Y.-S. Phys. Rev. B 2014, 89 (20), 205417.

(35)

Liu, D.; Guo, Y.; Fang, L.; Robertson, J. Appl. Phys. Lett. 2013, 103 (18), 183113.

(36)

Henkelman, G.; Jónsson, H. J. Chem. Phys. 2000, 113 (22), 9978–9985.

(37)

Henkelman, G.; Uberuaga, B. P.; Jónsson, H. J. Chem. Phys. 2000, 113 (22), 9901.

(38)

Mills, G.; Jónsson, H.; Schenter, G. K. Surf. Sci. 1995, 324 (2–3), 305–337.

(39)

Krukau, A. V.; Vydrov, O. A.; Izmaylov, A. F.; Scuseria, G. E. J. Chem. Phys. 2006, 125 (22), 224106.

(40)

Alem, N.; Ramasse, Q. M.; Seabourne, C. R.; Yazyev, O. V; Erickson, K.; Sarahan, M. C.; Kisielowski, C.; Scott, A. J.; Louie, S. G.; Zettl, A. Phys. Rev. Lett. 2012, 109 (20), 205502.

(41)

Alem, N.; Yazyev, O. V.; Kisielowski, C.; Denes, P.; Dahmen, U.; Hartel, P.; Haider, M.; Bischoff, M.; Jiang, B.; Louie, S. G.; Zettl, a. Phys. Rev. Lett. 2011, 106 (12), 126102.

(42)

Najmaei, S.; Liu, Z.; Zhou, W.; Zou, X.; Shi, G.; Lei, S.; Yakobson, B. I.; Idrobo, J.-C.; Ajayan, P. M.; Lou, J. Nat. Mater. 2013, 12 (8), 754–759.

(43)

Zou, X.; Liu, Y.; Yakobson, B. I. Nano Lett. 2013, 13 (1), 253–258.

(44)

Ophus, C.; Ciston, J.; Nelson, C. T. Ultramicroscopy 2016, 162, 1–9.

(45)

Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77 (18), 3865–3868.

(46)

Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1997, 78 (7), 1396–1396.

(47)

Blöchl, P. E. Phys. Rev. B 1994, 50 (24), 17953–17979.

(48)

Kresse, G.; Joubert, D. Phys. Rev. B 1999, 59 (3), 1758–1775.

(49)

Kresse, G.; Furthmüller, J. Phys. Rev. B 1996, 54 (16), 11169–11186.


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

(50)

Kresse, G.; Furthmüller, J. Comput. Mater. Sci. 1996, 6 (1), 15–50.

(51)

Komsa, H.-P.; Berseneva, N.; Krasheninnikov, A. V.; Nieminen, R. M. Phys. Rev. X 2014, 4 (3), 031044.


Page 20 of 25

Page 21 of 25

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

149x187mm (150 x 150 DPI)


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

215x164mm (150 x 150 DPI)


Page 22 of 25

Page 23 of 25

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

248x182mm (150 x 150 DPI)


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

142x171mm (150 x 150 DPI)


Page 24 of 25

Defect Coupling and Sub-Angstrom Structural ... - ACS Publications

Recommend Documents