Selective Engineering of Chalcogen Defects in MoS2 by Low-Energy

Research Article Cite This: ACS Appl. Mater. Interfaces 2019, 11, 24404−24411

www.acsami.org

Selective Engineering of Chalcogen Defects in MoS2 by Low-Energy Helium Plasma Binjie Huang,†,‡ Feng Tian,§ Youde Shen,† Minrui Zheng,† Yunshan Zhao,† Jing Wu,∥ Yi Liu,† Stephen J. Pennycook,§,⊥ and John T. L. Thong*,† †

Department of Electrical and Computer Engineering, National University of Singapore, 117583, Singapore NUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, 119077, Singapore § Center for Advanced 2D Materials, National University of Singapore, 117542, Singapore ∥ Institute of Materials Research and Engineering, Agency for Science Technology and Research, 138634, Singapore ⊥ Department of Materials Science and Engineering, National University of Singapore, 117575, Singapore

Downloaded via NOTTINGHAM TRENT UNIV on August 23, 2019 at 16:03:34 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

‡

S Supporting Information *

ABSTRACT: Structural defects in two-dimensional transition-metal dichalcogenides can significantly modify the material properties. Previous studies have shown that chalcogen defects can be created by physical sputtering, but the energetic ions can potentially displace transition-metal atoms at the same time, leading to ambiguous results and in some cases, degradation of material quality. In this work, selective sputtering of S atoms in monolayer MoS2 without damaging the Mo sublattice is demonstrated with low-energy helium plasma treatment. Based on X-ray photoelectron spectroscopy analysis, wide-range tuning of S defect concentration is achieved by controlling the ion energy and sputtering time. Furthermore, characterization with scanning transmission electron microscopy confirms that by keeping the ion energy low, the Mo sublattice remains intact. The properties of MoS2 at different defect concentrations are also characterized. In situ device measurement shows that the flake can be tuned from a semiconducting to metallic-like behavior by introducing S defects due to the creation of mid-gap states. When the defective MoS2 is exposed to air, the S defects are soon passivated, with oxygen atoms filling the defect sites. KEYWORDS: two-dimensional transition-metal dichalcogenides, monolayer MoS2, defect engineering, plasma sputtering, electrical tuning

■

INTRODUCTION Two-dimensional semiconducting transition-metal dichalcogenides (TMDs), such as MoS2 MoSe2, WS2, and WSe2, have received considerable research interest as prospective candidates in future electronic and photonic device applications.1−4 These atomically thin TMDs exhibit distinctive properties compared to their bulk forms, including large mobility, excellent mechanical flexibility, and band gap which varies with thickness.5 In particular, MoS2 shows strong photoluminescence (PL) and electroluminescence,6,7 while fieldeffect transistors (FETs) with a MoS2 channel can achieve excellent on−off ratio, rapid switching, and large transconductance.8 Structural defects play a crucial role in determining the characteristics of TMDs, either deteriorating or enhancing their performance.9 These defects can be either intrinsic or generated during material growth, transfer, or other treatments.10−14 Among the defects, chalcogen defects (e.g., S vacancies in MoS2) make up the dominant species.10 They act as active centers for trapping molecules and chemical reactions in the crystal lattice of TMDs.15 Indeed, it has been found that the catalytic performance of MoS2 in hydrogen evolution can be enhanced by deliberate introduction of S vacancies.16,17 On © 2019 American Chemical Society

the other hand, defect engineering can add new functionalities to TMDs by tuning their optical, mechanical, and magnetic properties.18−20 It can also facilitate substitutional doping by incorporating new elements at the chalcogen defect sites.21 In addition, tuning of electronic properties, including conductivity, threshold voltage, transport characteristics, and contact resistance, has been experimentally demonstrated as well by creating chalcogen vacancies in devices fabricated with TMDs.22−24 The role of chalcogen defects has been studied by creating defects in TMDs by tuning growth conditions,25,26 thermal annealing,18,27 and sputtering. Among these approaches, sputtering, including helium ion microscope (HIM) milling,24,28−30 ion bombardment, 31−35 and plasma treatment,21,36,37 is capable of generating a variety of defects ex situ with good controllability. Physical sputtering processes predominantly generate chalcogen defects, but the energetic ions (high-energy ions such as those used in HIM milling or low-energy heavy ions found in an Ar plasma) could possibly Received: April 2, 2019 Accepted: June 14, 2019 Published: June 14, 2019 24404

DOI: 10.1021/acsami.9b05507 ACS Appl. Mater. Interfaces 2019, 11, 24404−24411

Research Article

ACS Applied Materials & Interfaces

Figure 1. Selective creation of sulfur defects in MoS2. (a) XPS spectra of Mo 3d and S 2s core levels for MoS2 sputtered with different helium ion energies (Epeak) for 10 min. (b) Stoichiometry for MoS2 sputtered with different helium ion energies (Epeak) and time determined from the Mo 3d and S 2p core peaks. The S/Mo ratio of pristine MoS2 is normalized to 2. (c) Ratio of Mo6+ to Mo4+ peaks in the Mo 3d core in sputtered MoS2 with respect to helium ion energy (Epeak) for 10 min sputtering. The ratio of Mo6+ to Mo4+ in pristine MoS2 is set to 0. The error bar represents the uncertainty from the fitting process.

are 17.4 and 130.2 eV, respectively. In comparison, the minimum ion energies (Eion) for heavier Ar ions to remove S and Mo are 7 and 24.1 eV, respectively. The calculated energies are detailed in Table S1. In practical experiments, in which the plasma ions usually have an energy spread of tens of eV, the key to achieving selective sputtering is to choose an ion source with low atomic mass (i.e., He), so that there is a large window between the minimum ion energies to displace Mo S and S (i.e., EMo ion − Eion). In this work, we demonstrate selective sputtering of S atoms and wide-range tuning of S defect concentration in monolayer MoS2 with low-energy helium plasma treatment. We map the defect concentration to the He ion energy and sputtering time and investigate the electrical and optical properties of the film at different defect concentrations. With scanning transmission electron microscopy (STEM) images, we show that sputtering of up to 34% S atoms without damaging the Mo sublattice can be achieved with this method. Consistent trends in Raman and PL characterizations are also observed as the density of S defects increases. To eliminate the influence of adsorbed ambient species at the active defect sites, we conducted in situ device measurement in the same vacuum chamber where the plasma treatment takes place. Transition from a semiconducting to metallic-like state is observed, as a small amount of S defects is created with the low-energy plasma treatment.

displace Mo in these studies. As a result, degradation due to amorphization or even etching of the entire film can easily occur as the dose increases,24,38 presenting challenges in controlling the defect concentration precisely or introducing more chalcogen defects without markedly altering the quality of the film. Furthermore, differing observations on the effects of defects on the conductivity have been reported in previous studies. While the transition from semiconducting to metalliclike behavior has been reported in some studies when a small amount of chalcogen defects were introduced,24,28,30 other studies only observed degraded electrical performance in terms of larger resistance or lower mobility.31 A possible explanation for such inconsistencies is that with high-energy ion bombardment, the Mo sublattice is more vulnerable to Ar ions31 compared to lighter He ions,24,28,30 leading to faster degradation. Stripping of S from MoS2 has also been demonstrated with hydrogen plasma treatment,21,37 and Mo atoms may be etched by the hydrogen radicals as well, leading to destruction of the crystal structure.39 To address these issues, a methodology to selectively create chalcogen defects while keeping the transition-metal sublattice intact is highly desirable. The displacement threshold (Td), defined as the minimum kinetic energy transferred to the target atom that is required for it to leave the lattice, is larger for the heavy metal atom (∼20 eV for Mo in MoS2) than that for the typically lighter chalcogen atom (∼6.9 eV for S in MoS2).12 Hence, selective sputtering of chalcogen atoms is possible with the use of a low-energy plasma source, in which the ion is only energetic enough to sputter away the chalcogen atom. During sputtering, the ion energy required to impart sufficient energy to a target atom through momentum transfer to displace it (Td = 6.9 eV for S) depends on the mass of both the ion and the sputtered atom. For example, in a binary collision, a 17.4 eV He ion transfers 6.9 eV to the S atom in the lattice, which is just sufficient to displace the S. However, to create a Mo vacancy, the He ion needs to have at least 130.2 eV, such that 20 eV is transferred to the Mo atom. Therefore, the minimum ion energies (Eion) for He to displace S and Mo from the lattice

■

EXPERIMENTAL SECTION

Helium Plasma Treatment. The treatment was carried out in an ultrahigh vacuum (UHV) chamber with a base pressure of 5 × 10−8 Torr. The helium plasma was generated with a Tectra Gen plasma source. The chamber pressure was maintained at 1 × 10−4 Torr, and the flux was 1.75 × 1012 ions/cm2/s. The defect concentration of the treated sample was tuned by varying the exposure time or adding additional negative bias ranging from 0 to −162 V to the substrate. The additional bias generated an electric field which increased the energy for the He ions. The schematics of the plasma system and the measurement of ion energy are shown in Figure S1 and S2 in detail. During plasma sputtering, the MoS2 flakes prepared for annular dark field (ADF)−STEM imaging were suspended on a transmission 24405


Research Article


Figure 2. Quantification of defects in monolayer MoS2. (a,b) ADF−STEM images of monolayer MoS2 sputtered with (a) Epeak = 11 eV and (b) Epeak = 56 eV helium ions for 10 min. The yellow and red circles denote monosulfur vacancies (VS) and disulfur vacancies (VS2), respectively. (c,d) Site-assigned image intensity histogram for MoS2 sputtered with (c) Epeak = 11 eV and (d) Epeak = 56 eV helium ions for 10 min. (e−g) Comparison of line intensity profiles for S2, V2 defect and VS2 defects at the sulfur sites.

■

electron microscopy (TEM) grid, and the MoS2 flakes prepared for other measurements were supported on silicon wafer substrates. Sample Characterization. The X-ray photoelectron spectroscopy (XPS) characterization was performed with a Kratos AXIS UltraDLD XPS system using a monochromatic Al Kα source (hν = 1486.7 eV). Continuous monolayer MoS2 films on the Si substrate prepared by CVD growth (2D Semiconductors Inc.) were used in the XPS characterization. Raman and PL measurements were performed with a WITec alpha 300R confocal Raman system with a 532 nm laser excitation source. The measurements were conducted in air with a laser spot size of ∼320 nm. The laser power was kept below 1 mW to avoid heating damage to the sample. Optical microscopy images were taken with a Leica DM750M microscope. Atomic force microscopy (AFM) images were taken with a Bruker Dimension FastScan AFM in tapping mode. Exfoliated monolayer MoS2 flakes were used in the Raman, PL, and AFM characterization. For ADF−STEM imaging, exfoliated MoS2 flakes were transferred from the SiO2/Si substrate to holey silicon nitride TEM grids by wet transfer. After spin-coating with polymethyl methacrylate (PMMA), the sample was floated on 30% KOH solution to detach the wafer substrate and then cleaned in deionized water for several times before transferring to the TEM grid. Finally, the PMMA was removed by immersion in acetone and isopropanol, respectively. ADF−STEM imaging was performed on a probe-corrected JEOL-ARM200F STEM equipped with an ASCOR aberration corrector and a cold field emission gun. The microscope was operated at 60 keV with a probe current of 10 pA. The convergence angle was 40 mrad, and the inner detector angle was 65 mrad. The ADF−STEM images were recorded by 1024 × 1024 pixels with 10 μs dwell time. The high-resolution ADF−STEM images were low-pass filtered to increase the contrast. Device Fabrication and in Situ Characterization. MoS2 flakes were mechanically exfoliated from the molybdenite crystal (SPI Supplies) onto thermally oxidized degenerately p-doped silicon substrates. Monolayer MoS2 samples were identified with optical microscopy, Raman, and PL. Back-gated FETs were fabricated with electron-beam lithography on a layer of PMMA, followed by thermal evaporation of Cr/Au with a thickness of 3 and 60 nm. After lift-off in acetone, the sample was annealed for 2 h at 400 K in the UHV chamber used for helium plasma treatment to remove the atmospheric adsorbates on the surface. After helium plasma sputtering or other treatment, in situ device measurement was immediately carried out without exposing the sample to air. The channel length and width of the MoS2 FETs are 1 and 2 μm, respectively.

RESULTS AND DISCUSSION The generation of defects after the plasma treatment was investigated with XPS. Figure 1a shows the effects of sputtering a monolayer MoS2 film with different He ion energies for 10 min. A typical energy profile for He ions in the plasma in which the peak of the energy distribution (Epeak) is at ∼11 eV with a full-width-at-half-maximum of ∼8 eV, is shown in Figure S2. With low ion energies (Epeak = 11 eV, 56 eV), only a small amount of S defects are created, and therefore, the Mo 3d and S 2s core levels remain similar to those of a pristine sample. Although Epeak = 11 eV is below the theoretical minimum ion energy to displace S, S atoms can be sputtered away by the fraction of more energetic ions with more than 17.4 eV. Nevertheless, 99% of the ions are below 38 eV in the energy distribution. Practically all He ions are far below the minimum ion energy to displace Mo, and thus S defects are created selectively. A higher concentration of S defects can be achieved by increasing the ion energy or sputtering time. With a higher ion energy (Epeak = 92 eV), the intensity of the S 2s peak has been significantly reduced because of the increase in the number of S defects. The S/Mo ratio in monolayer MoS2 treated with different ion energies and sputtering time is summarized in Figure 1b. A consistent trend for the decreasing S/Mo ratio as the ion energy increases is observed within the same sputtering time. For the ion collisions in the low-energy regime, the scattering cross section for S atoms increases with the He ion energy,29,40 leading to an enhanced sputtering yield. As a result, more S is displaced as the ion energy increases. For each ion energy, the S/Mo ratio also reduces as the sputtering time increases, revealing the dependence of the concentration of S defects on the sputtering time. For defect-free MoS2, substitutional oxidation is thermodynamically favorable but faces a kinetic barrier of ∼1 eV, and therefore, this process is predicted to take place over a timescale of a month at room temperature.41 Nevertheless, in the presence of S defects, this kinetic barrier is reduced by half, making the process both thermodynamically and kinetically favorable at the reactive defect sites.42 In the sputtered samples, the intensity of the Mo6+ peaks in the Mo 3d core level significantly increases, indicating the formation of 24406


Research Article


Figure 3. Characterization of MoS2 sputtered with different helium ion energies for 10 min. (a) Raman, (b) PL spectra, and (c) AFM images of monolayer MoS2 with its corresponding height profile. The insets of (c) show the optical microscopy images.

MoO3,43 as these defects are filled by oxygen atoms when the sample was exposed to air before XPS characterization. Both oxygen and water vapor in the ambient air contributes to the oxidation.44 The formation of MoO3 was also observed in the XPS spectra of MoS2 treated with Ar or O plasma in previous studies.18,36 At a very large He ion energy of Epeak = 173 eV, the S 2s peak almost disappears, while the Mo6+ peak has increased. The ratio for the Mo6+ and Mo4+ peaks with respect to He ion energy is plotted in Figure 1c, showing passivation of S defect sites by O. As Epeak = 173 eV exceeds the theoretical minimum ion energy to displace Mo and S, both atoms can be sputtered away from the lattice. In this case, the energetic ions would still preferentially sputter away S atoms,24,34 because of their larger scattering cross section. The ratio of Mo to Si substrate remains almost constant in the XPS spectra, regardless of the ion energy (see Figure S3). Nevertheless, though the amount of S defects outnumbers that of Mo defects, the quality of the samples is still degraded, which is demonstrated below. To quantify the concentration of the chalcogen defects at the atomic scale and understand the structural changes of the lattice after vacancies are created following the plasma treatment, we characterized the monolayer MoS2 samples sputtered under different conditions with atomic resolution ADF imaging on an aberration-corrected STEM. Figure 2a shows the ADF−STEM image of monolayer MoS2 sputtered with a low ion energy of Epeak = 11 eV for 10 min. In this process, only a small amount of S defects can be created based on the momentum transfer model. After the plasma treatment, the sample retains the hexagonal lattice structure of the 2H phase. At the S sites, S2 column, monosulfur vacancies (VS), and disulfur vacancies (VS2) can be identified from the line intensity profile (Figure 2e−g). While the Mo atoms are intact, the quantification of defects at the S sites in Figure 2c demonstrates that around 9% of S atoms are missing from the lattice. In comparison, the amount of S defects is below 3% in pristine MoS2 samples.45 Most defects are found to be VS

vacancies and only a few VS2 vacancies are observed. Based on density functional theory calculation of monolayer MoS2 under electron irradiation,12 the formation energy for a S vacancy in the top layer facing the beam (8.1 eV) is slightly higher than that in the bottom layer (6.9 eV), because of the support from the other atomic layers below. Therefore, we expect more VS vacancies to be located at the bottom layer. Figure 2b shows the ADF−STEM image of a sample sputtered for 10 min with an ion energy of Epeak = 56 eV, which is also larger than the minimum ion energy to displace S atoms but smaller than that required to displace Mo. More S vacancies can be found because of the increase in ion energy. Quantifying from Figure 2d, we observe that 34% of S has been sputtered away. Most S defects are VS vacancies (85%), and the amount of VS2 vacancies also increases noticeably to 15%. The ADF−STEM image also shows that the hexagonal lattice structure remains unchanged, and all Mo atoms are preserved. This demonstrates our ability to create a considerable amount of S defects in MoS2 without damaging the crystal lattice by keeping the ion energy low. In comparison, though preferential S sputtering is also possible with a much larger ion energy (e.g., in HIM milling), the irradiated lattice may be partially damaged or even amorphized.24,28−30 The disordered regions act as structural defects, degrading the sample’s electrical and optical properties. If the sample is treated in plasma for a much longer time, as more VS2 vacancies are created in the lattice, the formation energy of Mo defects will eventually reduce,10 because the neighboring S atoms that the Mo had previously been bonded with have been sputtered away. As a result, the formation of Mo defects is possible for a long sputtering time. Mo atoms would also be sputtered away from the lattice if the ion energy is large enough (e.g., Epeak = 173 eV). Figure S6 shows the ADF−STEM image of monolayer MoS2 sputtered with Epeak = 173 eV ions for 3 min. While most parts of the Mo sublattice are preserved, structural pores consisting of both Mo and S defects are formed by the energetic ions. If the sample is sputtered for a longer time (e.g., 10 min) at this energy, the 24407


Research Article


Figure 4. Electrical properties of MoS2 with S defects. (a) Drain current versus gate voltage transfer curve for MoS2 sputtered with different helium ion energies for 10 min. The drain voltage is fixed at 0.5 V during the gate voltage sweep. (b) On−off ratio and electrical resistance for MoS2 sputtered with different helium ion energy. (c) Electrical resistance versus plasma treatment time for MoS2 sputtered with Epeak = 11 and 29 eV He ions. (d) Drain current vs gate voltage transfer curve for MoS2 after being plasma treated (Epeak = 11 eV for 1 h), left in vacuum for 3 h, and then exposed to oxygen for 1 h and water vapor for 1 h. The error bars represent the uncertainty in repeated measurements.

vibration after the removal of S atoms. This results in a much smaller change in A1g modes, which can be compensated for by mass reduction.49 Therefore, in contrast to the notable shift in E12g mode, the position of A1g is less sensitive to defects. A small LA(M) peak at ∼227 cm−1 due to lattice disorder32 is observed as more defects are created with Epeak = 56 eV ions (see Figure S7). The emergence of LA(M) peak was also reported in MoS2 treated with ion irradiation32,33 and chemical etching.50 For Epeak = 92 eV and above, the Raman modes are too weak to be identified. As shown in Figure 3b, the loss of symmetry also leads to PL quenching with more defects generated, similar to previous results with ion irradiation.33 The effect of plasma sputtering on the surface morphology was studied with optical microscopy imaging and AFM. Figure 3c shows the surface of monolayer MoS2 with different S defect concentrations. A typical pristine exfoliated sample has a uniform surface, and the step height measured for monolayer flakes is around 0.8 nm. The surfaces of samples sputtered using ions with Epeak from 11 to 92 eV are very similar to that of a pristine sample, with a monolayer flake showing dark blue color in the optical microscope and a step height of ∼0.8 nm in AFM images. This indicates that S defects are created without damaging the lattice when the ion energy is not high enough to displace Mo. As we continue to increase the ion energy beyond 164 eV, the morphology of the sputtered flakes changes significantly. For an ion energy of Epeak = 173 eV, the step height of the monolayer is reduced to around 0.4−0.5 nm. Interestingly, the monolayer MoS2 flake also becomes transparent in the optical microscope image (shown in the insets of Figure 3c) compared to pristine samples, with its color approaching that of the bare oxide wafer underneath. At this energy, both Mo and S defects can be created, and the S/ Mo ratio is reduced to 0.42 based on XPS analysis. At such a high S deficiency, the previous crystalline structure would not be maintained. Based on the observations from previous HIM studies,19,24 the sputtered MoS2 flake eventually becomes amorphous as the formation of disordered areas increases. This

monolayer MoS2 which is suspended on a TEM grid will disintegrate due to the lack of a supporting substrate beneath. When sputtered with He ions, defects are induced by direct impacts in freestanding MoS2, whereas backscattered ions and atoms sputtered from the substrate (e.g., Si or O from SiO2) can also contribute to the total sputtering yield in a supported sample.46,47 On the other hand, the substrate reduces the sputtering yield of direct impacts by impeding sputtering of S atoms in the bottom layer. These two opposite effects of the substrate can sometimes cancel each other.29 At ion energies below 200 eV, defect creation in the supported sample is dominated by direct impacts,46 and therefore, we expect the sputtering yield in supported and freestanding samples to be relatively close. In our experiment, the S/Mo ratio in MoS2 sputtered for 10 min obtained from XPS (supported) and STEM (freestanding) analysis are comparable (2.5 and 8% difference for Epeak = 11 and 56 eV, respectively). Selectively introducing S defects in MoS2 can significantly alter its optical properties. Figure 3a illustrates the evolution of the Raman spectra as different concentrations of S defects are introduced to monolayer MoS2 with different ion energies. As the defect concentration increases, consistent broadening and intensity attenuation are observed for both E12g and A1g Raman modes because of the breaking of crystal symmetry.26,48 While the position of the A1g Raman mode remains largely unchanged, the E12g mode red-shifts with more defects. The red shift of E12g mode has also been reported in various studies of MoS2 sputtered with plasma,36 ion beam,32,33 and electron beam.49 This red shift in the E12g mode (in-plane vibration of Mo−S bonds) is attributed to reduced restoring force constant due to the reduction in Mo−S bonds, eventually leading to lower E12g photon frequency.49 On the other hand, differing observations on the position of the A1g mode (red shift36 or blue shift32,33,49) were reported in the literature. This inconsistency is attributed to the district vibration pattern of A1g mode (out-of-plane vibration of Mo−S bonds), where the previously static Mo atoms are involved in the out-of-plane 24408


Research Article

ACS Applied Materials & Interfaces marked change of crystallinity can alter the thickness of the flake.51 Here, we attribute the reduced thickness of the amorphous flake to the considerable loss of atoms. As the monolayer flake continued to be sputtered for a longer time (e.g., for 1 h) at this energy, cracks could be found across the sample surface (Figure S8). The depth of the cracks is equivalent to the monolayer thickness (∼0.5 nm), suggesting that the monolayer flake has been fragmented. The formation of cracks can be attributed to lateral shrinkage caused by the loss of atoms. As the amount of vacancy sites (mostly S) increases, contraction accumulates in the lattice, eventually leading to line defects and even cracks.12,52 Similar effects have been seen with electron beam sputtering.53 In comparison with supported samples, 1 L MoS2 suspended on a TEM grid disintegrates within 10 min when it undergoes the same sputtering process with Epeak = 173 eV ions, indicating that the heavily damaged film could not remain freestanding. Apart from optical properties and morphology, the electrical properties of MoS2 can be dramatically tuned by defect engineering as well. To characterize this effect, in situ device measurement was conducted on back-gated FETs in the same UHV chamber following plasma treatment. The measurement was carried out immediately after the plasma treatment, avoiding passivation effects when the samples were exposed to air. The IDS − VGS transfer curve for MoS2 sputtered with different ion energies for 10 min is shown in Figure 4a, and their on−off ratio and electrical resistance are plotted in Figure 4b. Where the He ion energy is not high enough to displace Mo, a phase transition from semiconducting to metallic-like behavior is observed as S vacancies are created. As the ion energy Epeak increases from 11 to 56 eV, the overall current in the IDS − VGS transfer curve increases, demonstrating a consistent drop of resistance. Notably, while the sample still shows n-type characteristics, the off-state current at negative VGS dramatically increases once the sample is sputtered. The on−off ratio dropped from ∼104 for typical pristine MoS2 to as low as 1.5 after S defects are introduced, indicating very little gate dependence, and the resistance can be reduced by more than 2 orders of magnitude. The transition to metallic-like phase is consistent with the theoretical prediction of mid-gap states induced by defects in MoS254,55 and has also been reported in previous studies on defects with HIM milling.24,28,30 Furthermore, S defects also induce n-type doping effects, as evidenced by the reduction of threshold voltage. The evolution of MoS2 resistance with sputtering time at ion energies of Epeak = 11 and 29 eV is shown in Figure 4c. The resistance reduced dramatically within the first minute, indicating a quick transition from the semiconducting phase to metallic-like behavior as soon as the defect states are formed. Afterward, the resistance continued to reduce as more defects are created by sputtering. In comparison, a faster drop of resistance is observed with Epeak = 29 eV, as more S defects are created within the same period of time. Nevertheless, as the ion energy approaches the minimum ion energy to displace Mo atoms, the electrical properties start to degrade. It was observed that the on−off ratio and resistance of the MoS2 FET increases instead of continuing to drop when the ion energy is increased to Epeak = 92 eV. For an ion energy of Epeak = 173 eV, the sample becomes an insulator (resistance > 1010 Ω). At this energy, most S atoms are removed from the lattice, and thinning is observed in both optical microscopy and AFM images. Similar degraded electrical performance has been observed by others with high-dose ion irradiation.24,31

In situ measurement was also performed to investigate the passivation of S defects. The on−off ratio for sputtered samples, which had been reduced to less than an order of magnitude, increased back to >104 soon after they were exposed to air. The threshold voltage also increased at the same time. The metallic-like to semiconducting transition indicates that passivation of defects could take place at room temperature when exposed to oxygen and water vapor in air, agreeing with XPS characterization results, where the formation of Mo−O bonds was observed in sputtered MoS2 samples. The device characterization shows that by filling the S vacancies, oxygen atoms play a similar role as S, reverting to semiconducting behavior and p-doping the flake at the same time. Incorporation of oxygen atoms removes the mid-gap states previously created by the S defects, leading to a similar electronic structure as with pristine MoS2.56,57 To study the individual roles of oxygen and water vapor in the passivation process, oxygen and water vapor were separately introduced into the UHV chamber after S defects had been created. As shown in Figure 4d, the MoS2 FET became metallic-like after sputtering in helium plasma. This metal-like state remained relatively stable in vacuum. Only a slight increase in the on−off ratio (from 7.5 to 10) was observed after 3 h because of the adsorption of remnant oxygen and water vapor in the chamber. When oxygen was introduced into the chamber and maintained at 5 × 10−3 Torr, the on−off ratio doubled to ∼20 in 1 h. In comparison, when water vapor was introduced, the on−off ratio increased significantly to ∼184 in 1 h, indicating that water vapor plays the more significant role in passivation compared to pure oxygen. If the exposure continues, the flake will continue to become more semiconducting with the increase in resistance and on−off ratio. In this process, water vapor diffuses to the defect sites and subsequently dissociates, forming Mo−O bonds.58 After exposure to oxygen and water vapor, the sample will not return to the metallic-like phase (low on−off ratio) even if it is annealed at 100 °C for hours in high vacuum, suggesting formation of chemical bonds instead of merely physical adsorption.

■

CONCLUSIONS In summary, we have demonstrated tuning of chalcogen defect concentration in monolayer MoS2 by controlling ion energy and sputtering time in a helium plasma. Based on XPS and STEM analysis, we show that selective sputtering of a significant amount of S atoms while keeping the Mo sublattice intact is achievable by carefully maintaining the He ions at low energy. The optical and electrical properties of MoS2 have been studied at different concentrations of S defects. The Raman and PL spectra show consistent trends in red shift in the E12g Raman mode and quenched intensity, respectively, as the defect concentration increases. Dramatic change in surface morphology is demonstrated when the ion energy is large enough to displace Mo atoms. Our in situ device measurement demonstrates that the flake can be tuned from semiconducting to metallic-like behavior with low-energy He ions. Moreover, we also show the S defects can be easily passivated with oxygen and water vapor in ambient environment. Our studies on selective defect engineering of MoS2 will be useful for various applications that exploit chalcogen defects in TMDs, such as lowering the contact resistance in transistor applications and reducing the energy barrier for atomic doping. 24409


Research Article


■

Defects in Monolayer Molybdenum Disulfide. Nano Lett. 2013, 13, 2615−2622. (11) Hong, J.; Hu, Z.; Probert, M.; Li, K.; Lv, D.; Yang, X.; Gu, L.; Mao, N.; Feng, Q.; Xie, L. Exploring Atomic Defects in Molybdenum Disulphide Monolayers. Nat. Commun. 2015, 6, 6293. (12) Komsa, H.-P.; Kotakoski, J.; Kurasch, S.; Lehtinen, O.; Kaiser, U.; Krasheninnikov, A. V. Two-Dimensional Transition Metal Dichalcogenides under Electron Irradiation: Defect Production and Doping. Phys. Rev. Lett. 2012, 109, 035503. (13) Donarelli, M.; Bisti, F.; Perrozzi, F.; Ottaviano, L. Tunable Sulfur Desorption in Exfoliated MoS2 by Means of Thermal Annealing in Ultra-High Vacuum. Chem. Phys. Lett. 2013, 588, 198−202. (14) Sim, D. M.; Kim, M.; Yim, S.; Choi, M.-J.; Choi, J.; Yoo, S.; Jung, Y. S. Controlled Doping of Vacancy-Containing Few-Layer MoS2 via Highly Stable Thiol-Based Molecular Chemisorption. ACS Nano 2015, 9, 12115−12123. (15) Chen, B.; Sahin, H.; Suslu, A.; Ding, L.; Bertoni, M. I.; Peeters, F. M.; Tongay, S. Environmental Changes in MoTe2 Excitonic Dynamics by Defects-Activated Molecular Interaction. ACS Nano 2015, 9, 5326−5332. (16) Li, G.; Zhang, D.; Qiao, Q.; Yu, Y.; Peterson, D.; Zafar, A.; Kumar, R.; Curtarolo, S.; Hunte, F.; Shannon, S. All the Catalytic Active Sites of MoS2 for Hydrogen Evolution. J. Am. Chem. Soc. 2016, 138, 16632−16638. (17) Le, D.; Rawal, T. B.; Rahman, T. S. Single-Layer MoS2 with Sulfur Vacancies: Structure and Catalytic Application. J. Phys. Chem. C 2014, 118, 5346−5351. (18) Nan, H.; Wang, Z.; Wang, W.; Liang, Z.; Lu, Y.; Chen, Q.; He, D.; Tan, P.; Miao, F.; Wang, X. Strong Photoluminescence Enhancement of MoS2 through Defect Engineering and Oxygen Bonding. ACS Nano 2014, 8, 5738−5745. (19) Iberi, V.; Liang, L.; Ievlev, A. V.; Stanford, M. G.; Lin, M.-W.; Li, X.; Mahjouri-Samani, M.; Jesse, S.; Sumpter, B. G.; Kalinin, S. V. Nanoforging Single Layer MoSe2 through Defect Engineering with Focused Helium Ion Beams. Sci. Rep. 2016, 6, 30481. (20) Mathew, S.; Gopinadhan, K.; Chan, T. K.; Yu, X. J.; Zhan, D.; Cao, L.; Rusydi, A.; Breese, M. B. H.; Dhar, S.; Shen, Z. X. Magnetism in MoS2 Induced by Proton Irradiation. Appl. Phys. Lett. 2012, 101, 102103. (21) Lu, A.-Y.; Zhu, H.; Xiao, J.; Chuu, C.-P.; Han, Y.; Chiu, M.-H.; Cheng, C.-C.; Yang, C.-W.; Wei, K.-H.; Yang, Y. Janus Monolayers of Transition Metal Dichalcogenides. Nat. Nanotechnol. 2017, 12, 744. (22) Leong, W. S.; Li, Y.; Luo, X.; Nai, C. T.; Quek, S. Y.; Thong, J. T. L. Tuning the Threshold Voltage of MoS2 Field-Effect Transistors via Surface Treatment. Nanoscale 2015, 7, 10823−10831. (23) Tosun, M.; Chan, L.; Amani, M.; Roy, T.; Ahn, G. H.; Taheri, P.; Carraro, C.; Ager, J. W.; Maboudian, R.; Javey, A. Air-Stable nDoping of WSe2 by Anion Vacancy Formation with Mild Plasma Treatment. ACS Nano 2016, 10, 6853−6860. (24) Fox, D. S.; Zhou, Y.; Maguire, P.; O’Neill, A.; Ó ’Coileáin, C.; Gatensby, R.; Glushenkov, A. M.; Tao, T.; Duesberg, G. S.; Shvets, I. V. Nanopatterning and Electrical Tuning of MoS2 Layers with a Subnanometer Helium Ion Beam. Nano Lett. 2015, 15, 5307−5313. (25) Wu, K.; Li, Z.; Tang, J.; Lv, X.; Wang, H.; Luo, R.; Liu, P.; Qian, L.; Zhang, S.; Yuan, S. Controllable Defects Implantation in MoS2 Grown by Chemical Vapor Deposition for Photoluminescence Enhancement. Nano Res. 2018, 11, 4123−4132. (26) Mahjouri-Samani, M.; Liang, L.; Oyedele, A.; Kim, Y.-S.; Tian, M.; Cross, N.; Wang, K.; Lin, M.-W.; Boulesbaa, A.; Rouleau, C. M. Tailoring Vacancies Far Beyond Intrinsic Levels Changes the Carrier Type and Optical Response in Monolayer MoSe2−X Crystals. Nano Lett. 2016, 16, 5213−5220. (27) Tongay, S.; Suh, J.; Ataca, C.; Fan, W.; Luce, A.; Kang, J. S.; Liu, J.; Ko, C.; Raghunathanan, R.; Zhou, J. Defects Activated Photoluminescence in Two-Dimensional Semiconductors: Interplay between Bound, Charged, and Free Excitons. Sci. Rep. 2013, 3, 2657. (28) Stanford, M. G.; Pudasaini, P. R.; Belianinov, A.; Cross, N.; Noh, J. H.; Koehler, M. R.; Mandrus, D. G.; Duscher, G.; Rondinone,

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.9b05507. Minimum ion energy required to displace S and Mo in MoS2; schematics of the helium plasma sputtering system; measurement of ion energy of helium plasma; XPS analysis for MoS2 sputtered with different helium ion energies; ADF−STEM images; Raman spectra for MoS2 sputtered with different helium ion energies; and characterization of MoS2 sputtered with Epeak = 173 eV helium ions for 1 h (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

John T. L. Thong: 0000-0001-6954-9331 Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS This research is supported by the A*STAR Science and Engineering Research Council Grant (No. 152-70-00013). REFERENCES

(1) Bhimanapati, G. R.; Lin, Z.; Meunier, V.; Jung, Y.; Cha, J.; Das, S.; Xiao, D.; Son, Y.; Strano, M. S.; Cooper, V. R. Recent Advances in Two-Dimensional Materials Beyond Graphene. ACS Nano 2015, 9, 11509−11539. (2) Das, S.; Robinson, J. A.; Dubey, M.; Terrones, H.; Terrones, M. Beyond Graphene: Progress in Novel Two-Dimensional Materials and Van Der Waals Solids. Annu. Rev. Mater. Res. 2015, 45, 1−27. (3) Jariwala, D.; Sangwan, V. K.; Lauhon, L. J.; Marks, T. J.; Hersam, M. C. Emerging Device Applications for Semiconducting TwoDimensional Transition Metal Dichalcogenides. ACS Nano 2014, 8, 1102−1120. (4) Lv, R.; Robinson, J. A.; Schaak, R. E.; Sun, D.; Sun, Y.; Mallouk, T. E.; Terrones, M. Transition Metal Dichalcogenides and Beyond: Synthesis, Properties, and Applications of Single-and Few-Layer Nanosheets. Acc. Chem. Res. 2014, 48, 56−64. (5) Manzeli, S.; Ovchinnikov, D.; Pasquier, D.; Yazyev, O. V.; Kis, A. 2D Transition Metal Dichalcogenides. Nat. Rev. Mater. 2017, 2, 17033. (6) Splendiani, A.; Sun, L.; Zhang, Y.; Li, T.; Kim, J.; Chim, C.-Y.; Galli, G.; Wang, F. Emerging Photoluminescence in Monolayer MoS2. Nano Lett. 2010, 10, 1271−1275. (7) Sundaram, R. S.; Engel, M.; Lombardo, A.; Krupke, R.; Ferrari, A. C.; Avouris, P.; Steiner, M. Electroluminescence in Single Layer MoS2. Nano Lett. 2013, 13, 1416−1421. (8) Wang, Q. H.; Kalantar-Zadeh, K.; Kis, A.; Coleman, J. N.; Strano, M. S. Electronics and Optoelectronics of Two-Dimensional Transition Metal Dichalcogenides. Nat. Nanotechnol. 2012, 7, 699. (9) Patra, T. K.; Zhang, F.; Schulman, D. S.; Chan, H.; Cherukara, M. J.; Terrones, M.; Das, S.; Narayanan, B.; Sankaranarayanan, S. K. R. S. Defect Dynamics in 2-D MoS2 Probed by Using Machine Learning, Atomistic Simulations, and High-Resolution Microscopy. ACS Nano 2018, 12, 8006−8016. (10) Zhou, W.; Zou, X.; Najmaei, S.; Liu, Z.; Shi, Y.; Kong, J.; Lou, J.; Ajayan, P. M.; Yakobson, B. I.; Idrobo, J.-C. Intrinsic Structural 24410


Research Article

ACS Applied Materials & Interfaces A. J.; Ivanov, I. N. Focused Helium-Ion Beam Irradiation Effects on Electrical Transport Properties of Few-Layer WSe2: Enabling Nanoscale Direct Write Homo-Junctions. Sci. Rep. 2016, 6, 27276. (29) Ma, L.; Tan, Y.; Ghorbani-Asl, M.; Boettger, R.; Kretschmer, S.; Zhou, S.; Huang, Z.; Krasheninnikov, A. V.; Chen, F. Tailoring the Optical Properties of Atomically-Thin WS2 via Ion Irradiation. Nanoscale 2017, 9, 11027−11034. (30) Stanford, M. G.; Pudasaini, P. R.; Gallmeier, E. T.; Cross, N.; Liang, L.; Oyedele, A.; Duscher, G.; Mahjouri-Samani, M.; Wang, K.; Xiao, K. High Conduction Hopping Behavior Induced in Transition Metal Dichalcogenides by Percolating Defect Networks: Toward Atomically Thin Circuits. Adv. Funct. Mater. 2017, 27, 1702829. (31) Bertolazzi, S.; Bonacchi, S.; Nan, G.; Pershin, A.; Beljonne, D.; Samorì, P. Engineering Chemically Active Defects in Monolayer MoS2 Transistors via Ion-Beam Irradiation and Their Healing via Vapor Deposition of Alkanethiols. Adv. Mater. 2017, 29, 1606760. (32) Mignuzzi, S.; Pollard, A. J.; Bonini, N.; Brennan, B.; Gilmore, I. S.; Pimenta, M. A.; Richards, D.; Roy, D. Effect of Disorder on Raman Scattering of Single-Layer MoS2. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 91, 195411. (33) Chen, Y.; Huang, S.; Ji, X.; Adepalli, K.; Yin, K.; Ling, X.; Wang, X.; Xue, J.; Dresselhaus, M.; Kong, J. Tuning Electronic Structure of Single Layer MoS2 through Defect and Interface Engineering. ACS Nano 2018, 12, 2569−2579. (34) Ma, Q.; Odenthal, P. M.; Mann, J.; Le, D.; Wang, C. S.; Zhu, Y.; Chen, T.; Sun, D.; Yamaguchi, K.; Tran, T. Controlled Argon Beam-Induced Desulfurization of Monolayer Molybdenum Disulfide. J. Phys.: Condens. Matter 2013, 25, 252201. (35) He, Z.; Zhao, R.; Chen, X.; Chen, H.; Zhu, Y.; Su, H.; Huang, S.; Xue, J.; Dai, J.; Cheng, S. Defect Engineering in Single-Layer MoS2 Using Heavy Ion Irradiation. ACS Appl. Mater. Interfaces 2018, 10, 42524−42533. (36) Tao, L.; Duan, X.; Wang, C.; Duan, X.; Wang, S. PlasmaEngineered MoS2 Thin-Film as an Efficient Electrocatalyst for Hydrogen Evolution Reaction. Chem. Commun. 2015, 51, 7470− 7473. (37) Cheng, C.-C.; Lu, A.-Y.; Tseng, C.-C.; Yang, X.; Hedhili, M. N.; Chen, M.-C.; Wei, K.-H.; Li, L.-J. Activating Basal-Plane Catalytic Activity of Two-Dimensional MoS2 Monolayer with Remote Hydrogen Plasma. Nano Energy 2016, 30, 846−852. (38) Liu, Y.; Nan, H.; Wu, X.; Pan, W.; Wang, W.; Bai, J.; Zhao, W.; Sun, L.; Wang, X.; Ni, Z. Layer-by-Layer Thinning of MoS2 by Plasma. ACS Nano 2013, 7, 4202−4209. (39) Cai, D.; Xiao, S.; Nan, H.; Gu, X.; Ostrikov, K. K. Robust Fabrication of Quantum Dots on Few-Layer MoS2 by Soft Hydrogen Plasma and Post-Annealing. Part. Part. Syst. Charact. 2018, 35, 1800060. (40) Ghorbani-Asl, M.; Kretschmer, S.; Spearot, D. E.; Krasheninnikov, A. V. Two-Dimensional MoS2 under Ion Irradiation: From Controlled Defect Production to Electronic Structure Engineering. 2D Mater. 2017, 4, 025078. (41) Pető , J.; Ollár, T.; Vancsó, P.; Popov, Z. I.; Magda, G. Z.; Dobrik, G.; Hwang, C.; Sorokin, P. B.; Tapasztó, L. Spontaneous Doping of the Basal Plane of MoS2 Single Layers through Oxygen Substitution under Ambient Conditions. Nat. Chem. 2018, 10, 1246. (42) Santhosh, K. C.; Longo, R. C.; Wallace, R. M.; Cho, K. Surface Oxidation Energetics and Kinetics on MoS2 Monolayer. J. Appl. Phys. 2015, 117, 135301. (43) Li, X.; Li, X.; Zang, X.; Zhu, M.; He, Y.; Wang, K.; Xie, D.; Zhu, H. Role of Hydrogen in the Chemical Vapor Deposition Growth of MoS2 Atomic Layers. Nanoscale 2015, 7, 8398−8404. (44) Tongay, S.; Zhou, J.; Ataca, C.; Liu, J.; Kang, J. S.; Matthews, T. S.; You, L.; Li, J.; Grossman, J. C.; Wu, J. Broad-Range Modulation of Light Emission in Two-Dimensional Semiconductors by Molecular Physisorption Gating. Nano Lett. 2013, 13, 2831−2836. (45) Roy, S.; Choi, W.; Jeon, S.; Kim, D.-H.; Kim, H.; Yun, S. J.; Lee, Y.; Lee, J.; Kim, Y.-M.; Kim, J. Atomic Observation of Filling Vacancies in Monolayer Transition Metal Sulfides by Chemically Sourced Sulfur Atoms. Nano Lett. 2018, 18, 4523.

(46) Kretschmer, S.; Maslov, M.; Ghaderzadeh, S.; Ghorbani-Asl, M.; Hlawacek, G.; Krasheninnikov, A. V. Supported Two-Dimensional Materials under Ion Irradiation: The Substrate Governs Defect Production. ACS Appl. Mater. Interfaces 2018, 10, 30827−30836. (47) Li, W.; Wang, X.; Zhang, X.; Zhao, S.; Duan, H.; Xue, J. Mechanism of the Defect Formation in Supported Graphene by Energetic Heavy Ion Irradiation: The Substrate Effect. Sci. Rep. 2015, 5, 9935. (48) Tang, B.; Yu, Z. G.; Huang, L.; Chai, J.; Wong, S. L.; Deng, J.; Yang, W.; Gong, H.; Wang, S.; Ang, K.-W. Direct n- to p-Type Channel Conversion in Monolayer/Few-Layer WS2 Field-Effect Transistors by Atomic Nitrogen Treatment. ACS Nano 2018, 12, 2506−2513. (49) Parkin, W. M.; Balan, A.; Liang, L.; Das, P. M.; Lamparski, M.; Naylor, C. H.; Rodríguez-Manzo, J. A.; Johnson, A. T. C.; Meunier, V.; Drndić, M. Raman Shifts in Electron-Irradiated Monolayer MoS2. ACS Nano 2016, 10, 4134−4142. (50) Mercado, E.; Goodyear, A.; Moffat, J.; Cooke, M.; Sundaram, R. S. A Raman Metrology Approach to Quality Control of 2D MoS2 Film Fabrication. J. Phys. D: Appl. Phys. 2017, 50, 184005. (51) Moser, J.; Lévy, F.; Bussy, F. Composition and Growth Mode of MoSx Sputtered Films. J. Vac. Sci. Technol., A 1994, 12, 494−500. (52) Zhou, S.; Wang, S.; Li, H.; Xu, W.; Gong, C.; Grossman, J. C.; Warner, J. H. Atomic Structure and Dynamics of Defects in 2D MoS2 Bilayers. ACS Omega 2017, 2, 3315−3324. (53) Zhao, X.; Dan, J.; Chen, J.; Ding, Z.; Zhou, W.; Loh, K. P.; Pennycook, S. J. Atom-by-Atom Fabrication of Monolayer Molybdenum Membranes. Adv. Mater. 2018, 30, 1707281. (54) Kodama, N.; Hasegawa, T.; Okawa, Y.; Tsuruoka, T.; Joachim, C.; Aono, M. Electronic States of Sulfur Vacancies Formed on a MoS2 Surface. Jpn. J. Appl. Phys. 2010, 49, 08LB01. (55) Kim, B. H.; Park, M.; Lee, M.; Baek, S. J.; Jeong, H. Y.; Choi, M.; Chang, S. J.; Hong, W. G.; Kim, T. K.; Moon, H. R. Effect of Sulphur Vacancy on Geometric and Electronic Structure of MoS2 Induced by Molecular Hydrogen Treatment at Room Temperature. RSC Adv. 2013, 3, 18424−18429. (56) Liu, Y.; Stradins, P.; Wei, S.-H. Air Passivation of Chalcogen Vacancies in Two-Dimensional Semiconductors. Angew. Chem. 2016, 128, 977−980. (57) Komsa, H.-P.; Kurasch, S.; Lehtinen, O.; Kaiser, U.; Krasheninnikov, A. V. From Point to Extended Defects in TwoDimensional MoS2: Evolution of Atomic Structure under Electron Irradiation. Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 88, 035301. (58) Ghuman, K. K.; Yadav, S.; Singh, C. V. Adsorption and Dissociation of H2O on Monolayered MoS2 Edges: Energetics and Mechanism from Ab Initio Simulations. J. Phys. Chem. C 2015, 119, 6518−6529.

24411


Selective Engineering of Chalcogen Defects in MoS2 by Low-Energy

Recommend Documents