ACS Publications. Most Trusted. Most Cited. Most Read
Migration-Enhanced Metal–Organic Chemical Vapor Deposition of Wafer-Scale Fully Coalesced WS2 and WSe2 Monolayers
My Activity

Figure 1Loading Img
  • Open Access
Article

Migration-Enhanced Metal–Organic Chemical Vapor Deposition of Wafer-Scale Fully Coalesced WS2 and WSe2 Monolayers
Click to copy article linkArticle link copied!

Open PDFSupporting Information (1)

Crystal Growth & Design

Cite this: Cryst. Growth Des. 2023, 23, 3, 1547–1558
Click to copy citationCitation copied!
https://doi.org/10.1021/acs.cgd.2c01134
Published January 24, 2023

Copyright © 2023 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY-NC-ND 4.0 .

Abstract

Click to copy section linkSection link copied!

Metal–organic chemical vapor deposition (MOCVD) is widely employed for the wafer-scale synthesis of transition metal dichalcogenide (TMDC) monolayers (MLs). Despite large efforts devoted to understanding the intricate nucleation and lateral growth mechanisms of TMDCs, little attention has been paid to the migration of adatoms on the top of an ML and its influence on parasitic/premature bilayer (BL) nucleation. In this work, using a commercial multi-wafer MOCVD platform, a novel two-stage migration-enhanced MOCVD process is introduced to realize the deposition of wafer-scale fully coalesced tungsten disulfide (WS2) and tungsten diselenide (WSe2) MLs with only sparse BL nucleation in a reasonable deposition time. With the WS2 ML coverage exceeding 99% on 2 in. sapphire substrates within 3 h, BL coverage is suppressed to ∼15%. Following the same migration enhancement approach, WSe2 MLs are synthesized in 90 min with <20% BL coverage. The migration of W adatoms on the already formed stable WS2 (or WSe2) ML domains is promoted by ramping down the delivery of the tungsten precursor. From the qualitative analysis of the nanomorphology, the migration length of W adatoms is estimated to be ≤100 nm. This approach can be seen as a reliable and solid basis for the development of future large-scale TMDC ML deposition techniques.

This publication is licensed under

CC-BY-NC-ND 4.0 .
  • cc licence
  • by licence
  • nc licence
  • nd licence
Copyright © 2023 The Authors. Published by American Chemical Society

Synopsis

The upper half describes two possible cases during growth of individual domains. The triangle outlined in blue highlights lateral monolayer growth when the domain size is small compared to the migration length of W adatoms. The yellow one depicts bilayer nucleation caused by W adatoms confined on a monolayer domain. The lower half shows the mechanism of WCO ramp-down.

Introduction

Click to copy section linkSection link copied!

In the last decade, transition metal dichalcogenides (TMDCs) as an emerging class of 2D (semi)conducting materials have received great interest in various application fields such as (opto-)electronics, (1−3) sensing, (4,5) and catalysis. (6,7) High-quality single crystals, fabricated by either exfoliation or solid-source chemical vapor deposition, have been the subject of overwhelming activities to develop TMDC-based device technology, highlighting the great potential of this material system. However, all future applications require wafer-scale deposition techniques that fulfill the demands of industrialization and reproducible large-scale device fabrication.
Among a couple of methods (e.g., exfoliation, chemical vapor deposition, and atomic layer deposition) reported to prepare TMDC thin films, metal–organic chemical vapor deposition (MOCVD) stands out due to its excellent scalability, uniformity, and controllability. (8,9) Until now, various types of TMDC monolayers (MLs) have been successfully synthesized via MOCVD. Kang et al. (10) reported on the deposition of MoS2 and WS2 MLs on 4 in. fused silica and SiO2 substrates at a low growth temperature (550 °C), in which the layer-by-layer growth mode is achieved by low flow rates of precursors. Similarly, Grundmann et al. (11) employed a commercial MOCVD hot-wall reactor and deposited fully coalesced WS2 MLs with only sparse bilayer (BL) nucleation (∼10%) on 2 in. sapphire (0001) substrates at an elevated temperature (820 °C). However, a common drawback of these approaches is the long growth time (≥20 h). Boosting the lateral growth rate (after nucleation) by simply increasing the injection flux of precursors and thus gas-phase supersaturation will typically lead to the formation of a high density of premature BL nuclei. This process delays the coalescence of the first ML due to excessive consumption of adatoms for the new nuclei formed on the top of the ML. As growth proceeds, multilayers and even vertically standing nanostructures (12) are obtained.
Another technical challenge is to avoid unintentional carbon co-deposition in MOCVD. Based on vapor-phase chemistry, such carbon impurities are presumed to be originating from parasitic thermal decomposition, especially of side groups of organic chalcogen precursors. (13) As disclosed by many researchers, co-deposited carbon not only suppresses the lateral growth of ML but also deteriorates the properties of TMDCs. (13−16) Several countermeasures have been developed to prevent carbon co-deposition. For example, reducing the total pressure in the reactor chamber or adding H2 into the carrier gas has shown promising results. (13,15,17)
Recent studies have suggested using domain ripening and varied precursor fluxes to suppress BL nucleation and accelerate deposition. (18−20) The purpose of ripening is to dissociate unstable non-stoichiometric or sub-critically small clusters (on either the substrate (18) or the existing ML (20)) and to have the dissociated adatoms reassemble at the energetically preferential edge of ML domains, which results in the lateral growth of the existing ML domains. Practically, ripening is performed by stopping the supply of the growth-limiting species. Zhang et al. (18) demonstrated a diffusion-controlled multi-step MOCVD process in which nucleation, ripening, and lateral growth were designed with different precursor flow settings. A low flow rate in the later growth stage was chosen to reduce the gas-phase supersaturation. Finally, a closed WSe2 ML with 36% BL coverage was obtained on a 1 cm × 1 cm sapphire substrate. Such a multi-step MOCVD process was later applied to prepare crystallographically unidirectional WS2 ML with minimal BL coverage (<1%) on 2 in. sapphire. (19) As a step further, Seol et al. (20) introduced vertical Ostwald ripening in a pulsed MOCVD process, in which the injection of precursors is periodically interrupted. The authors claimed that the adatoms and sub-critical clusters are able to diffuse from the top of the MLs to their edges (i.e., vertical motion) during growth interruptions. Using this strategy, MoS2 and WS2 ML can be deposited on 6 in. SiO2 substrates in only 12 min. Recently, Choi et al. (21) employed a similar pulsed MOCVD technique and successfully synthesized wafer-scale MoS2 thin films at a relatively low temperature. By introducing water during idle times, Cohen et al. (22) proposed a growth–etch MOCVD process and realized the deposition of WS2 MLs on 2 in. sapphire with a large domain size and a low degree of carbon contamination.
Most of these studies explain the working principle of ripening/pulsing by the diffusion mechanism of adatoms on the substrate. It has been comprehensively elucidated that the diffusive motion of adatoms on the substrates, as an essential step in the lateral growth of 2D TMDCs, can influence growth kinetics (23−26) and the evolution of the domain morphology. (18,24,27) Adsorption and diffusive motion of adatoms on the surface of ML domains (which will be termed “migration” in this work) as well as their vertical climb-down motion across step edges of individual ML domains are commonly ignored to simplify the above model and discussion. The main justification for such simplification is that adsorption and sticking of adatoms are weaker on the ML surface due to the lower density of dangling bonds than that on the sapphire surface. However, as the ML coverage increases and approaches coalescence, the available free area of the sapphire surface will reduce and the number of adatoms arriving at the ML surface will increase. As a result, the adsorption and migration of adatoms on the ML surface start to play a vital role in premature parasitic BL nucleation on ML domains. Therefore, the migration behavior of adatoms in MOCVD, especially near full coalescence of MLs, requires dedicated experimental studies and discussion.
Previously, wafer-scale fully coalesced MoS2 MLs with ∼15% BL coverage have been successfully synthesized in our group without specifically considering the migration behavior of Mo adatoms. (26) However, for the case of WS2 discussed here, it is found that the BL coverage can easily exceed 20%. W-containing TMDCs appear to be more challenging for ML coalescence and BL suppression. We attribute this to a generally lower adatom mobility of W than that of Mo. (28) It means that W adatoms are prone to be confined on the top of ML domains and form BLs (compared to Mo adatoms), so the migration of W adatoms needs to be promoted.
In this work, wafer-scale fully coalesced WS2 and WSe2 MLs with largely suppressed BL formation are synthesized via two-stage migration-enhanced MOCVD processes. Similar to previous studies, (18,19,26) the nucleation (stage I) and lateral growth (stage II) are designed to take place at two distinct temperatures. Furthermore, the flux of tungsten hexacarbonyl (WCO) is gradually reduced by 75% during lateral growth stage II. Such ramp-down processes can be precisely controlled using mass flow controllers and monitored via in situ techniques in real time.
By studying the nanoscale morphology and structural properties of as-deposited WS2 thin films, the role of adatom migration is discussed, and the impact of WCO ramp-down is revealed. The influence of premature BL nucleation and carbon co-deposition on optical properties is unveiled experimentally. Additionally, the synergy between migration length of adatoms and growth conditions is discussed toward the realization of high-quality fully coalesced WS2 MLs. Such WS2 processes can be directly translated for the successful growth of WSe2 MLs, again evidencing the central role of W adatoms in ML formation, as well as showing the influence of different chalcogen species.

Methods

Click to copy section linkSection link copied!

A commercial AIXTRON cold-wall close-coupled showerhead reactor in 7 × 2″ geometry was employed, which was equipped with LayTec’s EpiTT in situ true temperature measurement and spectroscopic reflectance monitoring. EpiTT employs a high-brightness light-emitting diode as the light source and measures the reflectance of the sample surface at 950, 633, and 405 nm during MOCVD processes. A built-in ARGUS pyrometric profiling system controls the temperature distribution across the susceptor. Unless stated otherwise, the temperatures mentioned in this work refer to the emissivity-corrected true temperature on the wafer surface.
Tungsten hexacarbonyl [W(CO)6 or WCO, 99.99% purity], di-tert-butyl sulfide (DTBS, 99.999% purity), and di-iso-propyl selenide (DiPSe, 99.999% purity), all from Dockweiler Chemicals, were chosen as precursors. H2 was used as the carrier gas and reactor atmosphere. All processes were performed on (0001) oriented 2 in. sapphire substrates with a 0.2° offcut toward the m-plane and started with a H2 substrate desorption procedure for 30 min under 150 hPa at 1050 °C (which is adapted from our previous work (15)). For the deposition, a low reactor pressure of 20 hPa was chosen due to the suppression of Raman-detectable carbon incorporation in the TMDC films (see Figure S1 for details). The susceptor rotation was set to 25 rpm during deposition. After desorption, the measured reflectance values are normalized to the standard reflectance spectrum of sapphire provided by LayTec.
For WS2, the two-stage migration-enhanced MOCVD process consists of (i) nucleation stage I at relatively low temperatures (i.e., 700 and 750 °C) for only 15 min and (ii) lateral growth stage II at a relatively high temperature (820 °C) for 2–3 h. A temperature ramp at the beginning of stage II is introduced at a rate of 10 K/min without growth interruption (Figure 1a). In stage II, the molar flow of WCO is linearly ramped down by −75%, from 28 to 7 nmol/min (in, for example, 144 min, as shown in Figure 1b). With a constant DTBS flow (180 μmol/min), the S/W ratio increases from initially approx. 6,400 to 25,600 at the end of the process.

Figure 1

Figure 1. (a) Temperature profile with two different nucleation temperatures and (b) variation of molar flows and S/W ratio in the two-stage migration-enhanced MOCVD process (shown for sample S_R144, see below).

To evaluate the effect of WCO ramp-down (R), several “constant-flow” processes (C) were also conducted for comparison, in which the molar flow of WCO is kept constant in both stages. The temperature profile in these C processes is the same as that in R processes. Detailed growth parameters for all WS2 samples discussed in this work are summarized in Table 1. Samples S_N700 and S_N750 are reference samples to study nucleation at different temperatures, grown only for the usual duration time of 15 min. The three digits after R/C represent the lateral growth time in minutes. “HT” stands for high nucleation temperature (i.e., 750 °C). Samples S_R144 and S_R160.HT are compared to reveal the influence of nucleation density on BL formation. Samples S_R128, S_R144, S_C090, and S_C100 are used to discuss the impact of WCO ramp-down in detail. Coverage values are also given for the ML and BL (if applicable) and show the sample evolution toward a closed ML film.
Table 1. Detailed Summary of Growth Conditions of WS2 Samples (S_xxxx)a
a

Yellow- and gray-shaded samples are grown with low and high nucleation temperatures, respectively. Red- and green-shaded samples are deposited via the ramped (R) WCO flux and constant (C) WCO flux, respectively. The saturation of color reflects the coverage of the corresponding sample.

The development and details of migration-enhanced MOCVD processes is elucidated in the following using the WS2 samples listed in Table 1. WSe2 deposition shares most of the characteristics of WS2 growth. Therefore, WSe2 results will be presented in a compact form, introducing five samples (Se_N28, Se_N56, Se_R45, Se_R60, and Se_R75 as listed in Table 2) of different coalescence levels. The two digits after Se_N (nucleation-only samples) represent the WCO flux in nanomoles per minute during nucleation. As before, the digits after R represent the lateral growth time in minutes.
Table 2. Detailed Summary of Growth Conditions of WSe2 Samples (Se_xxx)a
a

Gray- and yellow-shaded samples are grown with low WCO flux and high WCO flux, respectively. The saturation of green color reflects the coverage of the corresponding sample.

Scanning electron microscopy (SEM; Zeiss Sigma), Raman spectroscopy (Renishaw inVia), photoluminescence (PL) spectroscopy (WiTec), and X-ray photoelectron spectroscopy (XPS; Physical Electronics) were employed to study the structural and optical properties of the as-deposited thin films. SEM images taken from different positions on the samples were segmented via ImageJ (29) to determine the quantities of nuclei as well as the coverage of the ML and BL. In this work, if ML coverage reaches 99%, the thin films will be considered as “fully coalesced”. For both Raman and PL measurements, the excitation wavelength is 532 nm and the laser power is ≤0.7 mW (to avoid damaging samples). The grating used for Raman measurements has 1800 grooves/mm and for PL measurements has 300 grooves/mm. Lorentzian fitting was applied to process all measured spectra. The source of XPS is the monochromatic Al Kα line at 1486.6 eV. An electron neutralizer with the emission of 20 μA current and a bias of 1.37 eV was used to avoid charging of samples during measurements. High-resolution scans were performed with a pass energy at 10 eV and a step width of 0.1 eV. All XPS spectra were deconvoluted using the Voigt function.

Results and Discussion

Click to copy section linkSection link copied!

Migration and Migration Enhancement

Two fundamental diffusive processes are relevant for W adatoms arriving from the gas phase. One takes place on the bare substrate surface and the other takes place on the surface of the already formed ML. In order to distinguish these two, in the following, the former is referred to as “diffusion”, while the latter is termed “migration”. “Migration” is used in analogy with “migration-enhanced MOCVD” (MEMOCVD) in III-nitride epitaxy, (30) which relates to the diffusive motion of precursor species on the surface of the topmost deposited layer. Similar to the diffusion length on the substrate, the migration length of adatoms on the ML λM is proportional to the product of the square roots of the diffusion coefficient D and mean mobile lifetime τ of adatoms. (18,25)
The contribution of diffusion and migration to the evolution of the WS2 morphology is illustrated in Figure 2. After 15 min of nucleation at 700 °C (sample S_N700), the shape of WS2 domains appears to be similar to a three-point star, as seen in Figure 2a. This curved triangular morphology was reported to be the result of a huge difference in the growth rate between W- and S-zigzag terminations under the S-rich condition. (31,32) Nevertheless, a more straightforward geometric approach to understand this shape evolution is illustrated in Figure 2b. Because growth is determined by the limited diffusion of adatoms from the surrounding free sapphire surface and assuming insignificant diffusion along nuclei edges, the corners of triangular domains will bind more adatoms and stretch out. Therefore, the initially tiny triangles will gradually turn into small three-point stars.

Figure 2

Figure 2. (a) Morphology of WS2 after 15 min nucleation at 700 °C (sample S_N700). (b) Schematic of the WS2 domain deviation from the triangular shape. (c) Morphology of WS2 after 128 min lateral growth (sample S_R128). (d) Schematic of WS2 domain shape recovery from three-point stars to triangles. (e) Schematic explaining parasitic/premature BL nucleation. λM stands for the migration length of adatoms.

However, as growth proceeds toward ML coalescence, the triangular shape is recovered. From the thermodynamics perspective, it can be explained by the fact that the curved concave edges are composed of kinks and are less stable than the straight S-zigzag edges, which results in the transformation from three-point stars to equilateral triangles during growth. (33) In this work, from the perspective of adatom migration, the authors additionally highlight the increasing contribution of adatoms from the surface of the ML domain itself to explain the recovery of the triangular shape. These adatoms can migrate to the nearest edge, overcome the Ehrlich–Schwoebel (E–S) barrier, (34,35) and attach at the concave frontiers of ML domains. As seen in Figure 2c, near the coalescence of MLs in sample S_R128, only equilateral triangles can be recognized.
In addition to the shape evolution of the WS2 domains, adatom migration can also determine parasitic/premature BL nucleation on the ML. As the total ML coverage increases by laterally growing ML domains, an increasing fraction of adatoms will arrive and migrate on the top of the ML. When a certain threshold value related to the ML domain size L exceeds the migration length λM, adatoms will be confined to the ML domains, forming BL clusters (Figure 2e). Once the critical nucleus size is reached, premature BL growth will be triggered. In the typical triangular geometry, this threshold value is equal to the distance from the centroid to the edge of equilateral triangles (i.e., 36L). Due to the existence of the E–S barrier at step edges of ML domains, adatom residence time (or effective adatom density) on ML domains will be even longer (or larger). Furthermore, BL nucleation can be seeded by intrinsic defects on the exposed surface of the ML (24) (which is a dangling-bond-free 2D surface). As a result, the effective migration length of adatoms λM,eff observed in experiments becomes much shorter. The formation of BL nuclei is a competing parasitic process, which can be attributed to insufficient λM (in relation to the ML domain size), the E–S barrier, and intrinsic defects within ML domains.
In order to close an ML while suppressing BL nucleation, the migration of adatoms needs to be enhanced. An intuitive approach is to increase the substrate temperature to improve the mobility of adatoms and to reduce the impact of the E–S barrier. However, this approach has practical limits because it increases the probability of gas-phase pre-reactions, (20) parasitic carbon co-deposition, (13,16) and adatom desorption. (24)
In this work, we propose to gradually ramp down the WCO flux with growing ML domain sizes, which reduces the partial pressure of W species in the gas phase as well as the arrival rate of W adatoms on the surface. This increases the critical nucleus size, thereby hindering the formation of stable/super-critical clusters. Additionally, it extends the adatom migration length, hence promoting lateral growth and coalescence of the ML. The mechanism of ramp-down is similar to the principle of vertical Ostwald ripening used by Seol et al. (20) (interrupted precursor injection).
In addition to migration enhancement via increasing temperature and ramping down the WCO supply, another method to prevent BL nucleation is to increase the initial nucleation density so that the required migration length just before coalescence is reduced. Of course, this comes at the cost of an increased density of grain boundaries. The influence of nucleation density on the quality of the final WS2 ML is discussed in the coming section. As stated above, these considerations apply to the growth of WSe2 as well.

Role of WS2 Nucleation Density in ML Coalescence

Controlling the nucleation density can be easily realized by varying the nucleation temperature. (11) WS2 nuclei after 15 min at 750 and 700 °C are shown in Figure 3a,b, respectively. When the temperature is reduced from 750 to 700 °C, the nucleation density increases from ∼140 to ∼210 nuclei/μm2. It can be estimated that the average distance between nuclei grown at 750 °C in Figure 3a is ∼85 nm, while this value drops to ∼70 nm at 700 °C in Figure 3b. The typical size of triangles at 750 °C is ∼60 nm and at 700 °C is ∼ 50 nm. Although these differences are comparably small, the resulting BL coverage after ML coalescence is clearly different for the two nucleation temperatures. Comparing Figure 3c with Figure 3d, we note that BL coverage decreases from ∼26% for growth after 750 °C nucleation (sample S_R160.HT) to ∼15% for growth after 700 °C nucleation (sample S_R144). It should be noted that the growth time of S_R160.HT needed to be extended to 160 min (Table 1) to reach coalescence. Hence, increasing the initial nucleation density can assist the synthesis of the WS2 ML with minimal BL formation.

Figure 3

Figure 3. (a,b) SEM images of WS2 after 15 min nucleation at 750 °C (S_N750) and 700 °C (S_N700). (c,d) SEM images of samples S_R160.HT and S_R144, where the ML is coalesced and BL nuclei are visible. Blue, green, and red arrows point at sapphire, ML, and BL regions, respectively. (e) Comparison of RT PL intensities of samples S_R160.HT and S_R144. (f) Deconvolution of PL spectra in different excitonic peaks. XD stands for bound excitons; XT stands for trions; and A stands for A excitons.

However, a high initial nucleation density is usually seen critically because it leads to a large density of grain boundaries in the coalesced film. To study their impact, room-temperature (RT) PL spectra were measured on the center of samples S_R160.HT and S_R144 and are shown in Figure 3e. Lower PL intensity is recorded from the WS2 ML with higher BL coverage (S_R160.HT), which is explained by the transition from the direct band gap to indirect band gap. (36,37) On the other hand, the peak position and the full width at half maximum (FWHM) of A-exciton emissions can be used as indicators for the crystal quality of WS2. (38) Interestingly, the deconvoluted A-exciton peaks of both samples are found to be almost identical. The FWHM of S_R160.HT (∼71 meV) is even slightly larger than that of S_R144 (∼66 meV). This observation suggests that the BL nuclei and their edge states might serve as defects in excitonic processes.
To sum up, no significant degradation of PL properties could be found as a result of an increased nucleation/grain boundary density, at least when it is accompanied by a reduction of BL coverage. Recently, some reports claimed that homogeneous and even unidirectional nucleation domains can be achieved by properly treating the sapphire substrates. (19,39) In the future, similar treatments can also be applied to this migration-enhanced MOCVD process in order to further optimize the crystal quality and properties of the WS2 ML. In the following sections, all WS2 MLs are grown using the nucleation process shown for sample S_N700.

Impact of Ramping Down the WCO Flux on WS2 Growth

To study the impact of WCO ramp-down (100% → 25%) on the lateral growth of the WS2 ML, the total amount of the WCO precursor injected was kept constant for samples S_R144 and S_C090. According to Figure 1 and Table 1, the total molar amount of WCO used for sample S_R144 can be calculated as 15 min × 28 nmol/min + 144 min × (28 + 7)/2 nmol/min = 2940 nmol. The same amount was used for sample S_C090: 15 min × 28 nmol/min + 90 min × 28 nmol/min = 2940 nmol.
During deposition, the impact of WCO ramp-down can be clearly recognized via the in situ reflectance transients in Figure 4a, in which the slope of reflectance at 405 nm gradually decreases. After linear fitting (the slope between 15 and 27 min is not considered because it is directly related to the temperature ramp-up), the slope for S_C090 between 27 and 105 min is about 8.6‰ per min, while that for S_R144 between 27 and 51 min is ∼6.4‰ per min and drops to ∼1.5‰ per min between 135 and 159 min (end of deposition). Based on our previous work, (26) there is an approximately linear dependence between the transient reflectance slope and the deposition rate during the MOCVD of the TMDC ML. By analogy, the in situ data comparison here indicates a decelerated growth for sample S_R144 near ML coalescence, which is explained by the WCO ramp-down.

Figure 4

Figure 4. (a) In situ reflectance transients of WS2 samples S_R144 and S_C090. The temperature ramp-up (between 15 and 27 min, marked by the red box) increases the slope. (Unit of slope: ‰ per min.) (b,c) SEM images of samples S_R144 and S_C090, respectively. (d) Raman spectra of both samples. (e) Variation of ML and BL coverages according to the type and duration of processes. The red and green arrows are indicators for increasing tendency of coverages.

The morphologies of samples S_R144 and S_C090 are presented in Figure 4b,c, respectively. After segmenting the SEM images taken from multiple positions on the 2 in. wafer, the average ML and BL coverages can be determined to be 99.0% ML and 14.5% BL for S_R144, and 96.9% ML and 11.5% BL for S_C090, respectively. Both the ML coverage and the BL coverage of WS2 after the ramp-down process (S_R144) are slightly (∼2%) higher than those on sample S_C090. This implies that given a certain molar amount of WCO, the ramp-down process features a higher incorporation efficiency of W, which could be the result of a higher mean S/W ratio (from constant DTBS injection) stabilizing the WS2 growth front during ramp-down.
Concerning BL formation, quantitative analyses of BL nuclei are carried out and listed in Table 3. Two facts are noticed from the comparison: (i) the BL nucleation density on sample S_R144 is slightly lower than that on sample S_C090 and (ii) the number of small clusters (with an area no larger than 100 nm2, probably formed during later stages of growth) on sample S_R144 is only about half of that on S_C090. Hence, it can be concluded that BL nucleation in the ramp-down process is suppressed in comparison with that in the constant process.
Table 3. Morphological Comparison of WS2 Samples S_R144 and S_C090
ProcessML coverage [%]BL coverage [%]BL nucleation density [/μm2]Counts of BL nuclei ≤ 100 nm2BL/ML ratio (coverage)I2LA/IA1g ratio (Raman)
S_R14499.0 ± 0.214.5 ± 0.7111 ± 531 in Figure 4b0.154.0 ± 0.4
S_C09096.9 ± 0.511.5 ± 0.4125 ± 763 in Figure 4c0.124.5 ± 0.2
Raman spectra in Figure 4d are used to identify the structural properties. The dominant ML nature of both samples can be confirmed by the high intensity ratio between deconvoluted 2LA(M) and A1g(Γ) modes. (40) Comparing both spectra, although the positions of individual Raman peaks barely differ, the intensity ratio I2LA/IA1g of S_R144 (∼4.0) is found to be lower than that of S_C090 (∼4.5). Details are listed in Table 3. The decrease of the intensity ratio I2LA/IA1g is often linked to a transition from the ML to BL (or multilayer). (22,40) The lower intensity ratio of the ramp-down sample here could therefore be conclusively related to its higher BL to ML coverage ratio. A similar relationship can be observed when comparing the intensity ratio I2LA/IE2g1 with the BL/ML coverage ratio. A summary about the intensity ratios of all WS2 ML samples can be found in the Supporting Information (Figure S2), which further confirms their dependence on the BL/ML ratio. On the other hand, it is important to notice that this intensity ratio is highly sensitive to the small variation of BL/ML coverage.
To emphasize the impact of WCO ramp-down on WS2 growth, the evolution of ML and BL coverages from different processes described in Table 1 is depicted in Figure 4e. It is clearly visible that in the final stage of coalescence (from 90 to 100%), the lateral growth of the ML slows down. In constant processes (samples S_C090 and S_C100), with the ML coverage increasing from 96.9 to 99.3% (ΔML = 2.4%), the BL coverage increases from 11.5 to 19.9% (ΔBL = 8.4%). Such an “obstacle” is less evident in ramp-down processes (samples S_R128 and S_R144): when the ML coverage grows from 94.3 to 99.0% (ΔML = 4.7%), the BL coverage only increases from 9.5 to 14.5% (ΔBL = 5.0%). This can be regarded as additional evidence for the improved lateral growth of the ML via ramping down WCO.
To study the impact of ramp-down on the optical properties of fully coalesced WS2 MLs, PL spectra of samples S_C100 and S_R144 are compared and correlated with the sample morphology. As seen in Figure 5a, the morphology of S_R144 is quite similar to that of S_C100, except for a 5.4% smaller BL coverage for S_R144. This offset of BL coverage leads to a distinct difference in the PL intensities in Figure 5b. Although PL emission of WS2 is expected to strongly drop from the ML to BL due to the direct-to-indirect band gap transition, a higher PL intensity as well as a slightly narrower FWHM of the deconvoluted A-exciton peak are found for S_C100 despite its higher BL coverage. Since the two processes were based on the same nucleation procedure, the influence of the density of grain boundaries can be excluded. The inferior PL properties of S_R144 are speculated to be related to defects within WS2 MLs themselves. The most common type of defect is sulfur vacancies, which can lead to the quenching of PL in WS2. (41−43) Other studies also point out that carbon co-deposition can reduce PL efficiency of TMDCs due to a charge transfer between TMDCs and graphitic carbon. (13,16) Furthermore, an impact of oxidation due to exposure to the atmosphere cannot be excluded.

Figure 5

Figure 5. (a) SEM image of WS2 with an almost identical ML coverage and an approx. 5% difference in BL coverage (top: S_C100 and bottom: S_R144). (b) PL comparison of two MLs, where the red-colored region stands for the assembly of PL spectra measured from different positions on sample S_R144. Green stands for those from sample S_C100. (c) XPS spectra of C 1s, S 2p, and W 4f core levels. Both axes are normalized to the same scale.

In order to figure out the exact reason for the low PL intensity of S_R144, the XPS spectra of C 1s, S 2p and W 4f core levels for samples S_C100 and S_R144 have been analyzed (Figure 5c). The peak positions of the W 4f doublet (33.2 and 35.4 eV) and S 2p doublet (162.8 and 164.0 eV) are in good agreement with reference values (16,20,41) as well as with XPS data of an exfoliated WS2 single-crystal flake. The additional minor doublet at 36.6 and 38.8 eV can be ascribed to the oxidized W species. (16,43) The S/W compositional ratio of WS2 grown from the two processes is calculated based on area fractions and relative sensitivity factors. After normalization with respect to the WS2 single-crystal flake, the stoichiometry of WS2 grown from both the constant (S_C100) and ramp-down (S_R144) processes is approximately WS2.0 (error margin approx. relative 5%). This implies that the stoichiometry of both MOCVD samples matches with that of the single crystal, which proves a high quality. On the other hand, based on the comparison of C 1s core levels, a stronger trace of carbon can be found in sample S_R144 (the atomic concentration is ∼14%, while that of S_C100 is ∼ 10%). This can be correlated to the higher total amount of DTBS resulting from the longer growth time and the larger S/W precursor ratio. Therefore, the reduction of PL intensity is likely caused by a higher degree of carbon incorporation within the ML. One possible approach to improvement may be to simultaneously ramp down the fluxes of both WCO and DTBS so that the potential source of carbon contamination can be minimized.

Estimate of Migration Length for WS2 Growth

Based on the nanoscale morphology of WS2 thin films, the migration length of W adatoms can be roughly estimated by various methods. The first is by identifying circular areas without the BL nucleus in the contour of merged ML domains. The largest radius of such areas is defined as the nominal migration length of adatoms. Larger coherent ML areas cannot be found on the sample because adatoms would have accumulated in these oversized ML regions, forming small BL clusters. Such an estimate is carried out on the samples with relatively few BL nuclei (S_R128 and S_C090) as the adsorption of adatoms to the edges of existing BL domains could shorten the mean travel distance of adatoms and underrate the extracted value of the migration length. As shown in Figure 6a, the nominal migration length λM measured in a ramp-down process (∼100 nm) is generally larger than that in a constant process (∼75 nm).

Figure 6

Figure 6. (a) Morphology of WS2 samples S_R128 (left) and S_C090 (right), on which an estimate of nominal migration length is extracted by red inscribed circles inside the ML contour. The corresponding BL nucleation density and effective migration length after calculations are noted on the top right of SEM images. (b) Illustration for calculating Lav from the density of BL nuclei, where the ML region is colored in light green and BL nuclei is colored in dark green.

As a second approach, the effective migration length can be approximated as the half of the average distance between BL nuclei Lav, which can be derived from the density of BL nuclei N via λM,eff=12×Lav=12×1/N. A similar method has been reported to calculate the average separation of WSe2 clusters on sapphire during ripening. (18) However, this estimate is based on the essential assumption that the size of individual clusters is negligible. In our case, we simply accept the same assumption because the number of small circular clusters is still dominant in comparison with the number of large domains. The results indicate that the effective migration length λM,eff in a ramp-down process (51 ± 2 nm) is also larger than that in a constant process (44 ± 1 nm).
Due to the above simplifications, the absolute values of the migration lengths reported here need to be treated with care. For instance, grain boundaries (or other morphological defects) in MLs could serve as favorable nucleation sites for BL nuclei, which leads to a significant underestimation of the migration length. Nevertheless, the results imply that the migration of adatoms on the top of the ML is rather restricted, with λM,eff in the range of only tens of nanometers up to 100 nm. Besides, by comparing the estimated migration lengths of different processes, it is confirmed that the WCO ramp-down can enlarge the migration length of W adatoms.

Migration-Enhanced MOCVD of WSe2 MLs

The above growth strategy described for WS2 can be directly applied to synthesize WSe2 MLs on 2 in. sapphire. The development of WSe2 processes also starts with two distinct nucleation tests. As shown in Figure 7a, the nucleation density of sample Se_N28 is only ∼20 nuclei/μm2. Based on the same reasoning regarding the role of nucleation density in ML coalescence for WS2, the nucleation density of WSe2 is increased up to ∼50 nuclei/μm2 (in order to fully close the ML) by increasing the WCO flux as shown for sample Se_N56 (in Figure 7b). This can be attributed to the smaller critical nucleus size (under higher partial pressure of W adatoms in the gas phase in the case of Se_N56) and the resulting higher nucleation rate.

Figure 7

Figure 7. (a,b) SEM image of WSe2 nucleation in samples Se_N28 and Se_N56, respectively. (c) Schematic of temperature and WCO flux variations during three MOCVD processes. The resulting morphologies of samples Se_R45, Se_R60, and Se_R75 are shown in figures (d–f). Blue, green, and red arrows in SEM images point at sapphire, ML, and BL regions, respectively. Trilayers are clearly visible in (f) marked by the red arrows.

Additional attention needs to be paid to the difference in the nucleation behavior between WS2 and WSe2, which is clearly discernible when comparing Figures 3a and 7a. The nucleation density of WSe2 in sample Se_N28 is much lower than that of WS2 in S_N750, even though both samples are grown under identical conditions (i.e., temperature, precursor fluxes, nucleation time, and so on). The average size of individual nuclei of WSe2 is also drastically larger than that of WS2, while the coverage of Se_N28 is lower than that of S_N750. The possible causes include the different critical nucleus sizes (WSe2 > WS2) and the stabilizing effect of different chalcogen precursors (DiPSe < DTBS). Thinking from the atomic perspective, it is indicated that S and Se adatoms interact with W adatoms in different kinetics. Further experiments need to be carried out to unveil the nature of this difference between S and Se adatoms and to discuss the detailed influence of chalcogen species in MOCVD processes.
Based on the nucleation, as shown for sample Se_N56, WCO ramp-down (from 56 to 14 nmol/min, −75%) is performed over different periods of lateral growth time, as seen in Figure 7c. From the morphology evolution from Figure 7d–f, the WSe2 ML is found to be coalesced after 75 min of lateral growth (sample Se_R75). Evaluating the variation of ML and BL coverages, similar to the previous analysis of the WS2 samples, we conclude that near ML coalescence, the growth of the BL is dominating (e.g., from Se_R60 to Se_R75: ΔBL = 8.4% while ΔML = 3.1%). With the help of WCO ramp-down, the BL coverage on the coalesced ML can be controlled to be below 20%. Furthermore, it is estimated that when the ML coverage of WSe2 just hits 99.0%, its BL coverage is around 15%, which coincides with the morphology of S_R144. This indicates that the W species is the decisive element for BL coverage in the MOCVD of both WS2 and WSe2, and the WCO ramp-down has similar impacts in both cases.
Nevertheless, the BL nucleation behavior in WSe2 samples is quite different from that in the WS2 one: BL domains of WSe2 as large as ∼500 nm (Figure 7f) are typically bigger than WS2 BL domains, which do not exceed 100 nm (Figure 4b). Some trilayer and even multilayer growth features can be recognized (marked by the red arrows in Figure 7f), which can also be explained by the strong confinement of adatoms on the top of large BL domains. Similar to the above comparison of ML nucleation behavior on a sapphire substrate, the difference in BL nucleation can also be attributed to the different reactivity of chalcogen species. Additionally, since the ML nucleation density in the case of WSe2 is much lower than that of WS2, it will lead to a lower density of grain boundaries throughout the WSe2 ML. Such grain boundaries can be seen as defects that favor the nucleation on top of them. Therefore, the BL nucleation in the case of WSe2 is less favored due to the suppression of defect-related nucleation sites on the surface of the WSe2 ML.

Conclusions

Click to copy section linkSection link copied!

In this work, a novel two-stage migration-enhanced MOCVD process is designed to prepare wafer-scale fully coalesced WS2 and WSe2 MLs with a BL coverage of <20%. Continuously ramping down the W precursor in the lateral growth stage is a commensurate and suitable strategy to enhance migration while the coverage of ML approaches 100%. The lateral growth of the ML is promoted and the formation of BL nuclei is suppressed. The estimated migration length of adatoms is also found to be increased by the WCO ramp-down. Furthermore, the influence of additional BL formation on the properties of the WS2 MLs is discussed. The PL emission is affected by not only the BL coverage but also by carbon co-deposition during synthesis. Although some growth parameters need further optimization, a generally high quality of the WS2 ML produced by ramp-down processes can be concluded as shown by Raman spectroscopy, PL spectroscopy, and XPS analyses. The extensive discussion about WSe2 allows us to extract information on the impact of the chalcogen species on nucleation and growth. This demonstrates that our model of TMDC ML formation in MOCVD should be expanded with the impact of the group VI species in future studies.
The migration-enhanced MOCVD process itself can be seen as a practical and useful technique supporting the large-scale fabrication of 2D TMDCs. This work can also be treated as a pioneer path to the direct growth of vertical 2D–2D all-planar heterostructures. The discussion about migration of adatoms sheds lights on the whole picture of the growth mechanism of 2D TMDCs.

Supporting Information

Click to copy section linkSection link copied!

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.cgd.2c01134.

  • Raman spectra of WS2 and WSe2 MLs, XPS spectra of MOCVD-grown and exfoliated WS2, reproducibility of WS2 ML growth, and a summary about the Raman intensity ratios and BL/ML coverage ratio (PDF)

Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Author Information

Click to copy section linkSection link copied!

  • Corresponding Author
  • Authors
    • Annika Grundmann - Compound Semiconductor Technology, RWTH Aachen University, Sommerfeldstr. 18, 52074Aachen, Germany
    • Hleb Fiadziushkin - Compound Semiconductor Technology, RWTH Aachen University, Sommerfeldstr. 18, 52074Aachen, GermanyOrcidhttps://orcid.org/0000-0002-0456-7480
    • Zhaodong Wang - PGI-7&10 and JARA-FIT, Forschungszentrum Jülich GmbH, Wilhelm Johnen Straße, 52428Jülich, Germany
    • Susanne Hoffmann-Eifert - PGI-7&10 and JARA-FIT, Forschungszentrum Jülich GmbH, Wilhelm Johnen Straße, 52428Jülich, GermanyOrcidhttps://orcid.org/0000-0003-1682-826X
    • Amir Ghiami - Compound Semiconductor Technology, RWTH Aachen University, Sommerfeldstr. 18, 52074Aachen, GermanyOrcidhttps://orcid.org/0000-0002-2451-1388
    • Arne Debald - AIXTRON SE, Dornkaulstr. 2, 52134Herzogenrath, Germany
    • Michael Heuken - Compound Semiconductor Technology, RWTH Aachen University, Sommerfeldstr. 18, 52074Aachen, GermanyAIXTRON SE, Dornkaulstr. 2, 52134Herzogenrath, Germany
    • Andrei Vescan - Compound Semiconductor Technology, RWTH Aachen University, Sommerfeldstr. 18, 52074Aachen, GermanyOrcidhttps://orcid.org/0000-0001-9465-2621
    • Holger Kalisch - Compound Semiconductor Technology, RWTH Aachen University, Sommerfeldstr. 18, 52074Aachen, Germany
  • Notes
    The authors declare no competing financial interest.

Acknowledgments

Click to copy section linkSection link copied!

This work was in part funded by the Federal Ministry of Education and Research (BMBF, Germany) in the projects NEUROTEC (16ME0399, 16ME0398K, and 16ME0403) and ForLab (16ES0940).

References

Click to copy section linkSection link copied!

This article references 43 other publications.

  1. 1
    Radisavljevic, B.; Radenovic, A.; Brivio, J.; Giacometti, V.; Kis, A. Single-Layer MoS2 Transistors. Nat. Nanotechnol. 2011, 6, 147150,  DOI: 10.1038/nnano.2010.279
  2. 2
    Akama, T.; Okita, W.; Nagai, R.; Li, C.; Kaneko, T.; Kato, T. Schottky Solar Cell Using Few-Layered Transition Metal Dichalcogenides toward Large-Scale Fabrication of Semitransparent and Flexible Power Generator. Sci. Rep. 2017, 7, 11967,  DOI: 10.1038/s41598-017-12287-6
  3. 3
    Andrzejewski, D.; Myja, H.; Heuken, M.; Grundmann, A.; Kalisch, H.; Vescan, A.; Kümmell, T.; Bacher, G. Scalable Large-Area p–i–n Light-Emitting Diodes Based on WS2 Monolayers Grown via MOCVD. ACS Photonics 2019, 6, 18321839,  DOI: 10.1021/acsphotonics.9b00311
  4. 4
    Late, D. J.; Doneux, T.; Bougouma, M. Single-Layer MoSe2 Based NH3 Gas Sensor. Appl. Phys. Lett. 2014, 105, 233103,  DOI: 10.1063/1.4903358
  5. 5
    Fathi-Hafshejani, P.; Azam, N.; Wang, L.; Kuroda, M. A.; Hamilton, M. C.; Hasim, S.; Mahjouri-Samani, M. Two-Dimensional-Material-Based Field-Effect Transistor Biosensor for Detecting COVID-19 Virus (SARS-CoV-2). ACS Nano 2021, 15, 1146111469,  DOI: 10.1021/acsnano.1c01188
  6. 6
    Kang, S.; Koo, J.-J.; Seo, H.; Truong, Q. T.; Park, J. B.; Park, S. C.; Jung, Y.; Cho, S.-P.; Nam, K. T.; Kim, Z. H.; Hong, B. H. Defect-Engineered MoS2 with Extended Photoluminescence Lifetime for High-Performance Hydrogen Evolution. J. Mater. Chem. C 2019, 7, 1017310178,  DOI: 10.1039/c9tc02256b
  7. 7
    Ju, L.; Bie, M.; Shang, J.; Tang, X.; Kou, L. Janus Transition Metal Dichalcogenides: A Superior Platform for Photocatalytic Water Splitting. JPhys Mater. 2020, 3, 022004,  DOI: 10.1088/2515-7639/ab7c57
  8. 8
    Liu, Y.; Gu, F. A Wafer-Scale Synthesis of Monolayer MoS2 and Their Field-Effect Transistors toward Practical Applications. Nanoscale Adv. 2021, 3, 21172138,  DOI: 10.1039/d0na01043j
  9. 9
    Lee, D. H.; Sim, Y.; Wang, J.; Kwon, S.-Y. Metal–Organic Chemical Vapor Deposition of 2D van Der Waals Materials─The Challenges and the Extensive Future Opportunities. APL Mater. 2020, 8, 030901,  DOI: 10.1063/1.5142601
  10. 10
    Kang, K.; Xie, S.; Huang, L.; Han, Y.; Huang, P. Y.; Mak, K. F.; Kim, C.-J.; Muller, D.; Park, J. High-Mobility Three-Atom-Thick Semiconducting Films with Wafer-Scale Homogeneity. Nature 2015, 520, 656660,  DOI: 10.1038/nature14417
  11. 11
    Grundmann, A.; Andrzejewski, D.; Kümmell, T.; Bacher, G.; Heuken, M.; Kalisch, H.; Vescan, A. H2S-Free Metal-Organic Vapor Phase Epitaxy of Coalesced 2D WS2 Layers on Sapphire─ERRATUM. MRS Adv. 2019, 4, e1  DOI: 10.1557/adv.2019.229
  12. 12
    Li, H.; Wu, H.; Yuan, S.; Qian, H. Synthesis and Characterization of Vertically Standing MoS2 Nanosheets. Sci. Rep. 2016, 6, 21171,  DOI: 10.1038/srep21171
  13. 13
    Schaefer, C. M.; Caicedo Roque, J. M.; Sauthier, G.; Bousquet, J.; Hébert, C.; Sperling, J. R.; Pérez-Tomás, A.; Santiso, J.; del Corro, E.; Garrido, J. A. Carbon Incorporation in MOCVD of MoS2 Thin Films Grown from an Organosulfide Precursor. Chem. Mater. 2021, 33, 44744487,  DOI: 10.1021/acs.chemmater.1c00646
  14. 14
    Zhang, X.; Al Balushi, Z. Y.; Zhang, F.; Choudhury, T. H.; Eichfeld, S. M.; Alem, N.; Jackson, T. N.; Robinson, J. A.; Redwing, J. M. Influence of Carbon in Metalorganic Chemical Vapor Deposition of Few-Layer WSe2 Thin Films. J. Electron. Mater. 2016, 45, 62736279,  DOI: 10.1007/s11664-016-5033-0
  15. 15
    Marx, M.; Grundmann, A.; Lin, Y.-R.; Andrzejewski, D.; Kümmell, T.; Bacher, G.; Heuken, M.; Kalisch, H.; Vescan, A. Metalorganic Vapor-Phase Epitaxy Growth Parameters for Two-Dimensional MoS2. J. Electron. Mater. 2018, 47, 910916,  DOI: 10.1007/s11664-017-5937-3
  16. 16
    Choudhury, T. H.; Simchi, H.; Boichot, R.; Chubarov, M.; Mohney, S. E.; Redwing, J. M. Chalcogen Precursor Effect on Cold-Wall Gas-Source Chemical Vapor Deposition Growth of WS2. Cryst. Growth Des. 2018, 18, 43574364,  DOI: 10.1021/acs.cgd.8b00306
  17. 17
    Eichfeld, S. M.; Hossain, L.; Lin, Y.; Piasecki, A. F.; Kupp, B.; Birdwell, A. G.; Burke, R. A.; Lu, N.; Peng, X.; Li, J.; Azcatl, A.; McDonnell, S.; Wallace, R. M.; Kim, M. J.; Mayer, T. S.; Redwing, J. M.; Robinson, J. A. Highly Scalable, Atomically Thin WSe2 Grown via Metal–Organic Chemical Vapor Deposition. ACS Nano 2015, 9, 20802087,  DOI: 10.1021/nn5073286
  18. 18
    Zhang, X.; Choudhury, T. H.; Chubarov, M.; Xiang, Y.; Jariwala, B.; Zhang, F.; Alem, N.; Wang, G.-C.; Robinson, J. A.; Redwing, J. M. Diffusion-Controlled Epitaxy of Large Area Coalesced WSe2 Monolayers on Sapphire. Nano Lett. 2018, 18, 10491056,  DOI: 10.1021/acs.nanolett.7b04521
  19. 19
    Chubarov, M.; Choudhury, T. H.; Hickey, D. R.; Bachu, S.; Zhang, T.; Sebastian, A.; Bansal, A.; Zhu, H.; Trainor, N.; Das, S.; Terrones, M.; Alem, N.; Redwing, J. M. Wafer-Scale Epitaxial Growth of Unidirectional WS2 Monolayers on Sapphire. ACS Nano 2021, 15, 25322541,  DOI: 10.1021/acsnano.0c06750
  20. 20
    Seol, M.; Lee, M.; Kim, H.; Shin, K. W.; Cho, Y.; Jeon, I.; Jeong, M.; Lee, H.; Park, J.; Shin, H. High-Throughput Growth of Wafer-Scale Monolayer Transition Metal Dichalcogenide via Vertical Ostwald Ripening. Adv. Mater. 2020, 32, 2003542,  DOI: 10.1002/adma.202003542
  21. 21
    Choi, J.; Ha, M.; Park, J. C.; Park, T. J.; Kim, W.; Lee, M.; Ahn, J. A Strategy for Wafer-Scale Crystalline MoS2 Thin Films with Controlled Morphology Using Pulsed Metal–Organic Chemical Vapor Deposition at Low Temperature. Adv. Mater. Interfaces 2022, 9, 2101785,  DOI: 10.1002/admi.202101785
  22. 22
    Cohen, A.; Patsha, A.; Mohapatra, P. K.; Kazes, M.; Ranganathan, K.; Houben, L.; Oron, D.; Ismach, A. Growth-Etch Metal–Organic Chemical Vapor Deposition Approach of WS2 Atomic Layers. ACS Nano 2021, 15, 526538,  DOI: 10.1021/acsnano.0c05394
  23. 23
    Shang, S.-L.; Lindwall, G.; Wang, Y.; Redwing, J. M.; Anderson, T.; Liu, Z.-K. Lateral Versus Vertical Growth of Two-Dimensional Layered Transition-Metal Dichalcogenides: Thermodynamic Insight into MoS2. Nano Lett. 2016, 16, 57425750,  DOI: 10.1021/acs.nanolett.6b02443
  24. 24
    Nie, Y.; Liang, C.; Cha, P.-R.; Colombo, L.; Wallace, R. M.; Cho, K. A Kinetic Monte Carlo Simulation Method of van Der Waals Epitaxy for Atomistic Nucleation-Growth Processes of Transition Metal Dichalcogenides. Sci. Rep. 2017, 7, 2977,  DOI: 10.1038/s41598-017-02919-2
  25. 25
    Ye, H.; Zhou, J.; Er, D.; Price, C. C.; Yu, Z.; Liu, Y.; Lowengrub, J.; Lou, J.; Liu, Z.; Shenoy, V. B. Toward a Mechanistic Understanding of Vertical Growth of van Der Waals Stacked 2D Materials: A Multiscale Model and Experiments. ACS Nano 2017, 11, 1278012788,  DOI: 10.1021/acsnano.7b07604
  26. 26
    Tang, S.; Grundmann, A.; Fiadziushkin, H.; Ghiami, A.; Heuken, M.; Vescan, A.; Kalisch, H. Detailed Study on MOCVD of Wafer-Scale MoS2 Monolayers: From Nucleation to Coalescence. MRS Adv. 2022, 7, 751756,  DOI: 10.1557/s43580-022-00312-4
  27. 27
    Xu, J.; Srolovitz, D. J.; Ho, D. The Adatom Concentration Profile: A Paradigm for Understanding Two-Dimensional MoS2 Morphological Evolution in Chemical Vapor Deposition Growth. ACS Nano 2021, 15, 68396848,  DOI: 10.1021/acsnano.0c10474
  28. 28
    Fruchart, O.; Jaren, S.; Rothman, J. Growth Modes of W and Mo Thin Epitaxial (110) Films on (112̅0) Sapphire. Appl. Surf. Sci. 1998, 135, 218232,  DOI: 10.1016/s0169-4332(98)00261-x
  29. 29
    Kapur, J. N.; Sahoo, P. K.; Wong, A. K. C. A New Method for Gray-Level Picture Thresholding Using the Entropy of the Histogram. Comput. Vis. Graph Image Process 1985, 29, 273285,  DOI: 10.1016/0734-189x(85)90125-2
  30. 30
    Fareed, R. S. Q.; Jain, R.; Gaska, R.; Shur, M. S.; Wu, J.; Walukiewicz, W.; Khan, M. A. High Quality InN/GaN Heterostructures Grown by Migration Enhanced Metalorganic Chemical Vapor Deposition. Appl. Phys. Lett. 2004, 84, 18921894,  DOI: 10.1063/1.1686889
  31. 31
    Wang, S.; Rong, Y.; Fan, Y.; Pacios, M.; Bhaskaran, H.; He, K.; Warner, J. H. Shape Evolution of Monolayer MoS2 Crystals Grown by Chemical Vapor Deposition. Chem. Mater. 2014, 26, 63716379,  DOI: 10.1021/cm5025662
  32. 32
    Shinde, N. B.; Francis, B.; Ramachandra Rao, M. S.; Ryu, B. D.; Chandramohan, S.; Eswaran, S. K. Rapid Wafer-Scale Fabrication with Layer-by-Layer Thickness Control of Atomically Thin MoS2 Films Using Gas-Phase Chemical Vapor Deposition. APL Mater. 2019, 7, 081113,  DOI: 10.1063/1.5095451
  33. 33
    Reifsnyder Hickey, D.; Nayir, N.; Chubarov, M.; Choudhury, T. H.; Bachu, S.; Miao, L.; Wang, Y.; Qian, C.; Crespi, V. H.; Redwing, J. M.; van Duin, A. C. T.; Alem, N. Illuminating Invisible Grain Boundaries in Coalesced Single-Orientation WS2 Monolayer Films. Nano Lett. 2021, 21, 64876495,  DOI: 10.1021/acs.nanolett.1c01517
  34. 34
    Liu, S. J.; Huang, H.; Woo, C. H. Schwoebel-Ehrlich Barrier: From Two to Three Dimensions. Appl. Phys. Lett. 2002, 80, 32953297,  DOI: 10.1063/1.1475774
  35. 35
    Rao, D.; Biswas, B.; Acharya, S.; Bhatia, V.; Pillai, A. I. K.; Garbrecht, M.; Saha, B. Effects of Adatom Mobility and Ehrlich–Schwoebel Barrier on Heteroepitaxial Growth of Scandium Nitride (ScN) Thin Films. Appl. Phys. Lett. 2020, 117, 212101,  DOI: 10.1063/5.0027091
  36. 36
    Zhao, W.; Ghorannevis, Z.; Chu, L.; Toh, M.; Kloc, C.; Tan, P.-H.; Eda, G. Evolution of Electronic Structure in Atomically Thin Sheets of WS2 and WSe2. ACS Nano 2013, 7, 791797,  DOI: 10.1021/nn305275h
  37. 37
    Gutiérrez, H. R.; Perea-López, N.; Elías, A. L.; Berkdemir, A.; Wang, B.; Lv, R.; López-Urías, F.; Crespi, V. H.; Terrones, H.; Terrones, M. Extraordinary Room-Temperature Photoluminescence in Triangular WS2 Monolayers. Nano Lett. 2013, 13, 34473454,  DOI: 10.1021/nl3026357
  38. 38
    McCreary, A.; Berkdemir, A.; Wang, J.; Nguyen, M. A.; Elías, A. L.; Perea-López, N.; Fujisawa, K.; Kabius, B.; Carozo, V.; Cullen, D. A.; Mallouk, T. E.; Zhu, J.; Terrones, M. Distinct Photoluminescence and Raman Spectroscopy Signatures for Identifying Highly Crystalline WS2 Monolayers Produced by Different Growth Methods. J. Mater. Res. 2016, 31, 931944,  DOI: 10.1557/jmr.2016.47
  39. 39
    Shi, Y.; Groven, B.; Serron, J.; Wu, X.; Nalin Mehta, A.; Minj, A.; Sergeant, S.; Han, H.; Asselberghs, I.; Lin, D.; Brems, S.; Huyghebaert, C.; Morin, P.; Radu, I.; Caymax, M. Engineering Wafer-Scale Epitaxial Two-Dimensional Materials through Sapphire Template Screening for Advanced High-Performance Nanoelectronics. ACS Nano 2021, 15, 94829494,  DOI: 10.1021/acsnano.0c07761
  40. 40
    Berkdemir, A.; Gutiérrez, H. R.; Botello-Méndez, A. R.; Perea-López, N.; Elías, A. L.; Chia, C.-I.; Wang, B.; Crespi, V. H.; López-Urías, F.; Charlier, J.-C.; Terrones, H.; Terrones, M. Identification of Individual and Few Layers of WS2 Using Raman Spectroscopy. Sci. Rep. 2013, 3, 1755,  DOI: 10.1038/srep01755
  41. 41
    Rosenberger, M. R.; Chuang, H.-J.; McCreary, K. M.; Li, C. H.; Jonker, B. T. Electrical Characterization of Discrete Defects and Impact of Defect Density on Photoluminescence in Monolayer WS2. ACS Nano 2018, 12, 17931800,  DOI: 10.1021/acsnano.7b08566
  42. 42
    Li, J.; Su, W.; Chen, F.; Fu, L.; Ding, S.; Song, K.; Huang, X.; Zhang, L. Atypical Defect-Mediated Photoluminescence and Resonance Raman Spectroscopy of Monolayer WS2. J. Phys. Chem. C 2019, 123, 39003907,  DOI: 10.1021/acs.jpcc.8b11647
  43. 43
    Marinov, D.; de Marneffe, J.-F.; Smets, Q.; Arutchelvan, G.; Bal, K. M.; Voronina, E.; Rakhimova, T.; Mankelevich, Y.; El Kazzi, S.; Nalin Mehta, A.; Wyndaele, P.-J.; Heyne, M. H.; Zhang, J.; With, P. C.; Banerjee, S.; Neyts, E. C.; Asselberghs, I.; Lin, D.; De Gendt, S. Reactive Plasma Cleaning and Restoration of Transition Metal Dichalcogenide Monolayers. npj 2D Mater. Appl. 2021, 5, 17,  DOI: 10.1038/s41699-020-00197-7

Cited By

Click to copy section linkSection link copied!
Citation Statements
Explore this article's citation statements on scite.ai

This article is cited by 6 publications.

  1. Amir Ghiami, Hleb Fiadziushkin, Tianyishan Sun, Songyao Tang, Yibing Wang, Eva Mayer, Jochen M. Schneider, Agata Piacentini, Max C. Lemme, Michael Heuken, Holger Kalisch, Andrei Vescan. Improved Reliability of 2D-TMDC Dry Transfer via PMMA and Target Substrate Treatments. ACS Applied Electronic Materials 2025, 7 (6) , 2394-2403. https://doi.org/10.1021/acsaelm.4c02168
  2. Jinfeng Yang, Ye Fan, Ryo Mizuta, Max Rimmer, Jack Donoghue, Shaoliang Guan, Sarah J. Haigh, Stephan Hofmann. Operando Scanning Electron Microscopy Study of Support Interactions and Mechanisms of Salt-Assisted WS2 Growth. Chemistry of Materials 2025, 37 (3) , 989-1000. https://doi.org/10.1021/acs.chemmater.4c02603
  3. Tae Soo Kim, Gichang Noh, Seongdae Kwon, Ji Yoon Kim, Krishna P. Dhakal, Saeyoung Oh, Hyun‐Jun Chai, Eunpyo Park, In Soo Kim, Eunji Lee, Youngbum Kim, Jaehyun Lee, Min‐kyung Jo, Minsoo Kang, Cheolmin Park, Jeongho Kim, Jeongwon Park, Suhyun Kim, Mingyu Kim, Yuseok Kim, Sung‐Yool Choi, Seungwoo Song, Hu Young Jeong, Jeongyong Kim, Joon Young Kwak, Kibum Kang. Diffusion Control on the Van der Waals Surface of Monolayers for Uniform Bi‐Layer MoS 2 Growth. Advanced Functional Materials 2024, 34 (23) https://doi.org/10.1002/adfm.202312365
  4. Isaiah A. Moses, Chengyin Wu, Wesley F. Reinhart. Crystal growth characterization of WSe2 thin film using machine learning. Materials Today Advances 2024, 22 , 100483. https://doi.org/10.1016/j.mtadv.2024.100483
  5. Amir Ghiami, Tianyishan Sun, Hleb Fiadziushkin, Songyao Tang, Annika Grundmann, Michael Heuken, Holger Kalisch, Andrei Vescan. Optimization of Layer Transfer and Photolithography for Device Integration of 2D-TMDC. Crystals 2023, 13 (10) , 1474. https://doi.org/10.3390/cryst13101474
  6. Xiang Lan, Yingliang Cheng, Xiangdong Yang, Zhengwei Zhang. Wafer-scale engineering of two-dimensional transition metal dichalcogenides. Chip 2023, 2 (3) , 100057. https://doi.org/10.1016/j.chip.2023.100057

Crystal Growth & Design

Cite this: Cryst. Growth Des. 2023, 23, 3, 1547–1558
Click to copy citationCitation copied!
https://doi.org/10.1021/acs.cgd.2c01134
Published January 24, 2023

Copyright © 2023 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY-NC-ND 4.0 .

Article Views

1712

Altmetric

-

Citations

Learn about these metrics

Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.

Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.

The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.

  • Abstract

    Figure 1

    Figure 1. (a) Temperature profile with two different nucleation temperatures and (b) variation of molar flows and S/W ratio in the two-stage migration-enhanced MOCVD process (shown for sample S_R144, see below).

    Figure 2

    Figure 2. (a) Morphology of WS2 after 15 min nucleation at 700 °C (sample S_N700). (b) Schematic of the WS2 domain deviation from the triangular shape. (c) Morphology of WS2 after 128 min lateral growth (sample S_R128). (d) Schematic of WS2 domain shape recovery from three-point stars to triangles. (e) Schematic explaining parasitic/premature BL nucleation. λM stands for the migration length of adatoms.

    Figure 3

    Figure 3. (a,b) SEM images of WS2 after 15 min nucleation at 750 °C (S_N750) and 700 °C (S_N700). (c,d) SEM images of samples S_R160.HT and S_R144, where the ML is coalesced and BL nuclei are visible. Blue, green, and red arrows point at sapphire, ML, and BL regions, respectively. (e) Comparison of RT PL intensities of samples S_R160.HT and S_R144. (f) Deconvolution of PL spectra in different excitonic peaks. XD stands for bound excitons; XT stands for trions; and A stands for A excitons.

    Figure 4

    Figure 4. (a) In situ reflectance transients of WS2 samples S_R144 and S_C090. The temperature ramp-up (between 15 and 27 min, marked by the red box) increases the slope. (Unit of slope: ‰ per min.) (b,c) SEM images of samples S_R144 and S_C090, respectively. (d) Raman spectra of both samples. (e) Variation of ML and BL coverages according to the type and duration of processes. The red and green arrows are indicators for increasing tendency of coverages.

    Figure 5

    Figure 5. (a) SEM image of WS2 with an almost identical ML coverage and an approx. 5% difference in BL coverage (top: S_C100 and bottom: S_R144). (b) PL comparison of two MLs, where the red-colored region stands for the assembly of PL spectra measured from different positions on sample S_R144. Green stands for those from sample S_C100. (c) XPS spectra of C 1s, S 2p, and W 4f core levels. Both axes are normalized to the same scale.

    Figure 6

    Figure 6. (a) Morphology of WS2 samples S_R128 (left) and S_C090 (right), on which an estimate of nominal migration length is extracted by red inscribed circles inside the ML contour. The corresponding BL nucleation density and effective migration length after calculations are noted on the top right of SEM images. (b) Illustration for calculating Lav from the density of BL nuclei, where the ML region is colored in light green and BL nuclei is colored in dark green.

    Figure 7

    Figure 7. (a,b) SEM image of WSe2 nucleation in samples Se_N28 and Se_N56, respectively. (c) Schematic of temperature and WCO flux variations during three MOCVD processes. The resulting morphologies of samples Se_R45, Se_R60, and Se_R75 are shown in figures (d–f). Blue, green, and red arrows in SEM images point at sapphire, ML, and BL regions, respectively. Trilayers are clearly visible in (f) marked by the red arrows.

  • References


    This article references 43 other publications.

    1. 1
      Radisavljevic, B.; Radenovic, A.; Brivio, J.; Giacometti, V.; Kis, A. Single-Layer MoS2 Transistors. Nat. Nanotechnol. 2011, 6, 147150,  DOI: 10.1038/nnano.2010.279
    2. 2
      Akama, T.; Okita, W.; Nagai, R.; Li, C.; Kaneko, T.; Kato, T. Schottky Solar Cell Using Few-Layered Transition Metal Dichalcogenides toward Large-Scale Fabrication of Semitransparent and Flexible Power Generator. Sci. Rep. 2017, 7, 11967,  DOI: 10.1038/s41598-017-12287-6
    3. 3
      Andrzejewski, D.; Myja, H.; Heuken, M.; Grundmann, A.; Kalisch, H.; Vescan, A.; Kümmell, T.; Bacher, G. Scalable Large-Area p–i–n Light-Emitting Diodes Based on WS2 Monolayers Grown via MOCVD. ACS Photonics 2019, 6, 18321839,  DOI: 10.1021/acsphotonics.9b00311
    4. 4
      Late, D. J.; Doneux, T.; Bougouma, M. Single-Layer MoSe2 Based NH3 Gas Sensor. Appl. Phys. Lett. 2014, 105, 233103,  DOI: 10.1063/1.4903358
    5. 5
      Fathi-Hafshejani, P.; Azam, N.; Wang, L.; Kuroda, M. A.; Hamilton, M. C.; Hasim, S.; Mahjouri-Samani, M. Two-Dimensional-Material-Based Field-Effect Transistor Biosensor for Detecting COVID-19 Virus (SARS-CoV-2). ACS Nano 2021, 15, 1146111469,  DOI: 10.1021/acsnano.1c01188
    6. 6
      Kang, S.; Koo, J.-J.; Seo, H.; Truong, Q. T.; Park, J. B.; Park, S. C.; Jung, Y.; Cho, S.-P.; Nam, K. T.; Kim, Z. H.; Hong, B. H. Defect-Engineered MoS2 with Extended Photoluminescence Lifetime for High-Performance Hydrogen Evolution. J. Mater. Chem. C 2019, 7, 1017310178,  DOI: 10.1039/c9tc02256b
    7. 7
      Ju, L.; Bie, M.; Shang, J.; Tang, X.; Kou, L. Janus Transition Metal Dichalcogenides: A Superior Platform for Photocatalytic Water Splitting. JPhys Mater. 2020, 3, 022004,  DOI: 10.1088/2515-7639/ab7c57
    8. 8
      Liu, Y.; Gu, F. A Wafer-Scale Synthesis of Monolayer MoS2 and Their Field-Effect Transistors toward Practical Applications. Nanoscale Adv. 2021, 3, 21172138,  DOI: 10.1039/d0na01043j
    9. 9
      Lee, D. H.; Sim, Y.; Wang, J.; Kwon, S.-Y. Metal–Organic Chemical Vapor Deposition of 2D van Der Waals Materials─The Challenges and the Extensive Future Opportunities. APL Mater. 2020, 8, 030901,  DOI: 10.1063/1.5142601
    10. 10
      Kang, K.; Xie, S.; Huang, L.; Han, Y.; Huang, P. Y.; Mak, K. F.; Kim, C.-J.; Muller, D.; Park, J. High-Mobility Three-Atom-Thick Semiconducting Films with Wafer-Scale Homogeneity. Nature 2015, 520, 656660,  DOI: 10.1038/nature14417
    11. 11
      Grundmann, A.; Andrzejewski, D.; Kümmell, T.; Bacher, G.; Heuken, M.; Kalisch, H.; Vescan, A. H2S-Free Metal-Organic Vapor Phase Epitaxy of Coalesced 2D WS2 Layers on Sapphire─ERRATUM. MRS Adv. 2019, 4, e1  DOI: 10.1557/adv.2019.229
    12. 12
      Li, H.; Wu, H.; Yuan, S.; Qian, H. Synthesis and Characterization of Vertically Standing MoS2 Nanosheets. Sci. Rep. 2016, 6, 21171,  DOI: 10.1038/srep21171
    13. 13
      Schaefer, C. M.; Caicedo Roque, J. M.; Sauthier, G.; Bousquet, J.; Hébert, C.; Sperling, J. R.; Pérez-Tomás, A.; Santiso, J.; del Corro, E.; Garrido, J. A. Carbon Incorporation in MOCVD of MoS2 Thin Films Grown from an Organosulfide Precursor. Chem. Mater. 2021, 33, 44744487,  DOI: 10.1021/acs.chemmater.1c00646
    14. 14
      Zhang, X.; Al Balushi, Z. Y.; Zhang, F.; Choudhury, T. H.; Eichfeld, S. M.; Alem, N.; Jackson, T. N.; Robinson, J. A.; Redwing, J. M. Influence of Carbon in Metalorganic Chemical Vapor Deposition of Few-Layer WSe2 Thin Films. J. Electron. Mater. 2016, 45, 62736279,  DOI: 10.1007/s11664-016-5033-0
    15. 15
      Marx, M.; Grundmann, A.; Lin, Y.-R.; Andrzejewski, D.; Kümmell, T.; Bacher, G.; Heuken, M.; Kalisch, H.; Vescan, A. Metalorganic Vapor-Phase Epitaxy Growth Parameters for Two-Dimensional MoS2. J. Electron. Mater. 2018, 47, 910916,  DOI: 10.1007/s11664-017-5937-3
    16. 16
      Choudhury, T. H.; Simchi, H.; Boichot, R.; Chubarov, M.; Mohney, S. E.; Redwing, J. M. Chalcogen Precursor Effect on Cold-Wall Gas-Source Chemical Vapor Deposition Growth of WS2. Cryst. Growth Des. 2018, 18, 43574364,  DOI: 10.1021/acs.cgd.8b00306
    17. 17
      Eichfeld, S. M.; Hossain, L.; Lin, Y.; Piasecki, A. F.; Kupp, B.; Birdwell, A. G.; Burke, R. A.; Lu, N.; Peng, X.; Li, J.; Azcatl, A.; McDonnell, S.; Wallace, R. M.; Kim, M. J.; Mayer, T. S.; Redwing, J. M.; Robinson, J. A. Highly Scalable, Atomically Thin WSe2 Grown via Metal–Organic Chemical Vapor Deposition. ACS Nano 2015, 9, 20802087,  DOI: 10.1021/nn5073286
    18. 18
      Zhang, X.; Choudhury, T. H.; Chubarov, M.; Xiang, Y.; Jariwala, B.; Zhang, F.; Alem, N.; Wang, G.-C.; Robinson, J. A.; Redwing, J. M. Diffusion-Controlled Epitaxy of Large Area Coalesced WSe2 Monolayers on Sapphire. Nano Lett. 2018, 18, 10491056,  DOI: 10.1021/acs.nanolett.7b04521
    19. 19
      Chubarov, M.; Choudhury, T. H.; Hickey, D. R.; Bachu, S.; Zhang, T.; Sebastian, A.; Bansal, A.; Zhu, H.; Trainor, N.; Das, S.; Terrones, M.; Alem, N.; Redwing, J. M. Wafer-Scale Epitaxial Growth of Unidirectional WS2 Monolayers on Sapphire. ACS Nano 2021, 15, 25322541,  DOI: 10.1021/acsnano.0c06750
    20. 20
      Seol, M.; Lee, M.; Kim, H.; Shin, K. W.; Cho, Y.; Jeon, I.; Jeong, M.; Lee, H.; Park, J.; Shin, H. High-Throughput Growth of Wafer-Scale Monolayer Transition Metal Dichalcogenide via Vertical Ostwald Ripening. Adv. Mater. 2020, 32, 2003542,  DOI: 10.1002/adma.202003542
    21. 21
      Choi, J.; Ha, M.; Park, J. C.; Park, T. J.; Kim, W.; Lee, M.; Ahn, J. A Strategy for Wafer-Scale Crystalline MoS2 Thin Films with Controlled Morphology Using Pulsed Metal–Organic Chemical Vapor Deposition at Low Temperature. Adv. Mater. Interfaces 2022, 9, 2101785,  DOI: 10.1002/admi.202101785
    22. 22
      Cohen, A.; Patsha, A.; Mohapatra, P. K.; Kazes, M.; Ranganathan, K.; Houben, L.; Oron, D.; Ismach, A. Growth-Etch Metal–Organic Chemical Vapor Deposition Approach of WS2 Atomic Layers. ACS Nano 2021, 15, 526538,  DOI: 10.1021/acsnano.0c05394
    23. 23
      Shang, S.-L.; Lindwall, G.; Wang, Y.; Redwing, J. M.; Anderson, T.; Liu, Z.-K. Lateral Versus Vertical Growth of Two-Dimensional Layered Transition-Metal Dichalcogenides: Thermodynamic Insight into MoS2. Nano Lett. 2016, 16, 57425750,  DOI: 10.1021/acs.nanolett.6b02443
    24. 24
      Nie, Y.; Liang, C.; Cha, P.-R.; Colombo, L.; Wallace, R. M.; Cho, K. A Kinetic Monte Carlo Simulation Method of van Der Waals Epitaxy for Atomistic Nucleation-Growth Processes of Transition Metal Dichalcogenides. Sci. Rep. 2017, 7, 2977,  DOI: 10.1038/s41598-017-02919-2
    25. 25
      Ye, H.; Zhou, J.; Er, D.; Price, C. C.; Yu, Z.; Liu, Y.; Lowengrub, J.; Lou, J.; Liu, Z.; Shenoy, V. B. Toward a Mechanistic Understanding of Vertical Growth of van Der Waals Stacked 2D Materials: A Multiscale Model and Experiments. ACS Nano 2017, 11, 1278012788,  DOI: 10.1021/acsnano.7b07604
    26. 26
      Tang, S.; Grundmann, A.; Fiadziushkin, H.; Ghiami, A.; Heuken, M.; Vescan, A.; Kalisch, H. Detailed Study on MOCVD of Wafer-Scale MoS2 Monolayers: From Nucleation to Coalescence. MRS Adv. 2022, 7, 751756,  DOI: 10.1557/s43580-022-00312-4
    27. 27
      Xu, J.; Srolovitz, D. J.; Ho, D. The Adatom Concentration Profile: A Paradigm for Understanding Two-Dimensional MoS2 Morphological Evolution in Chemical Vapor Deposition Growth. ACS Nano 2021, 15, 68396848,  DOI: 10.1021/acsnano.0c10474
    28. 28
      Fruchart, O.; Jaren, S.; Rothman, J. Growth Modes of W and Mo Thin Epitaxial (110) Films on (112̅0) Sapphire. Appl. Surf. Sci. 1998, 135, 218232,  DOI: 10.1016/s0169-4332(98)00261-x
    29. 29
      Kapur, J. N.; Sahoo, P. K.; Wong, A. K. C. A New Method for Gray-Level Picture Thresholding Using the Entropy of the Histogram. Comput. Vis. Graph Image Process 1985, 29, 273285,  DOI: 10.1016/0734-189x(85)90125-2
    30. 30
      Fareed, R. S. Q.; Jain, R.; Gaska, R.; Shur, M. S.; Wu, J.; Walukiewicz, W.; Khan, M. A. High Quality InN/GaN Heterostructures Grown by Migration Enhanced Metalorganic Chemical Vapor Deposition. Appl. Phys. Lett. 2004, 84, 18921894,  DOI: 10.1063/1.1686889
    31. 31
      Wang, S.; Rong, Y.; Fan, Y.; Pacios, M.; Bhaskaran, H.; He, K.; Warner, J. H. Shape Evolution of Monolayer MoS2 Crystals Grown by Chemical Vapor Deposition. Chem. Mater. 2014, 26, 63716379,  DOI: 10.1021/cm5025662
    32. 32
      Shinde, N. B.; Francis, B.; Ramachandra Rao, M. S.; Ryu, B. D.; Chandramohan, S.; Eswaran, S. K. Rapid Wafer-Scale Fabrication with Layer-by-Layer Thickness Control of Atomically Thin MoS2 Films Using Gas-Phase Chemical Vapor Deposition. APL Mater. 2019, 7, 081113,  DOI: 10.1063/1.5095451
    33. 33
      Reifsnyder Hickey, D.; Nayir, N.; Chubarov, M.; Choudhury, T. H.; Bachu, S.; Miao, L.; Wang, Y.; Qian, C.; Crespi, V. H.; Redwing, J. M.; van Duin, A. C. T.; Alem, N. Illuminating Invisible Grain Boundaries in Coalesced Single-Orientation WS2 Monolayer Films. Nano Lett. 2021, 21, 64876495,  DOI: 10.1021/acs.nanolett.1c01517
    34. 34
      Liu, S. J.; Huang, H.; Woo, C. H. Schwoebel-Ehrlich Barrier: From Two to Three Dimensions. Appl. Phys. Lett. 2002, 80, 32953297,  DOI: 10.1063/1.1475774
    35. 35
      Rao, D.; Biswas, B.; Acharya, S.; Bhatia, V.; Pillai, A. I. K.; Garbrecht, M.; Saha, B. Effects of Adatom Mobility and Ehrlich–Schwoebel Barrier on Heteroepitaxial Growth of Scandium Nitride (ScN) Thin Films. Appl. Phys. Lett. 2020, 117, 212101,  DOI: 10.1063/5.0027091
    36. 36
      Zhao, W.; Ghorannevis, Z.; Chu, L.; Toh, M.; Kloc, C.; Tan, P.-H.; Eda, G. Evolution of Electronic Structure in Atomically Thin Sheets of WS2 and WSe2. ACS Nano 2013, 7, 791797,  DOI: 10.1021/nn305275h
    37. 37
      Gutiérrez, H. R.; Perea-López, N.; Elías, A. L.; Berkdemir, A.; Wang, B.; Lv, R.; López-Urías, F.; Crespi, V. H.; Terrones, H.; Terrones, M. Extraordinary Room-Temperature Photoluminescence in Triangular WS2 Monolayers. Nano Lett. 2013, 13, 34473454,  DOI: 10.1021/nl3026357
    38. 38
      McCreary, A.; Berkdemir, A.; Wang, J.; Nguyen, M. A.; Elías, A. L.; Perea-López, N.; Fujisawa, K.; Kabius, B.; Carozo, V.; Cullen, D. A.; Mallouk, T. E.; Zhu, J.; Terrones, M. Distinct Photoluminescence and Raman Spectroscopy Signatures for Identifying Highly Crystalline WS2 Monolayers Produced by Different Growth Methods. J. Mater. Res. 2016, 31, 931944,  DOI: 10.1557/jmr.2016.47
    39. 39
      Shi, Y.; Groven, B.; Serron, J.; Wu, X.; Nalin Mehta, A.; Minj, A.; Sergeant, S.; Han, H.; Asselberghs, I.; Lin, D.; Brems, S.; Huyghebaert, C.; Morin, P.; Radu, I.; Caymax, M. Engineering Wafer-Scale Epitaxial Two-Dimensional Materials through Sapphire Template Screening for Advanced High-Performance Nanoelectronics. ACS Nano 2021, 15, 94829494,  DOI: 10.1021/acsnano.0c07761
    40. 40
      Berkdemir, A.; Gutiérrez, H. R.; Botello-Méndez, A. R.; Perea-López, N.; Elías, A. L.; Chia, C.-I.; Wang, B.; Crespi, V. H.; López-Urías, F.; Charlier, J.-C.; Terrones, H.; Terrones, M. Identification of Individual and Few Layers of WS2 Using Raman Spectroscopy. Sci. Rep. 2013, 3, 1755,  DOI: 10.1038/srep01755
    41. 41
      Rosenberger, M. R.; Chuang, H.-J.; McCreary, K. M.; Li, C. H.; Jonker, B. T. Electrical Characterization of Discrete Defects and Impact of Defect Density on Photoluminescence in Monolayer WS2. ACS Nano 2018, 12, 17931800,  DOI: 10.1021/acsnano.7b08566
    42. 42
      Li, J.; Su, W.; Chen, F.; Fu, L.; Ding, S.; Song, K.; Huang, X.; Zhang, L. Atypical Defect-Mediated Photoluminescence and Resonance Raman Spectroscopy of Monolayer WS2. J. Phys. Chem. C 2019, 123, 39003907,  DOI: 10.1021/acs.jpcc.8b11647
    43. 43
      Marinov, D.; de Marneffe, J.-F.; Smets, Q.; Arutchelvan, G.; Bal, K. M.; Voronina, E.; Rakhimova, T.; Mankelevich, Y.; El Kazzi, S.; Nalin Mehta, A.; Wyndaele, P.-J.; Heyne, M. H.; Zhang, J.; With, P. C.; Banerjee, S.; Neyts, E. C.; Asselberghs, I.; Lin, D.; De Gendt, S. Reactive Plasma Cleaning and Restoration of Transition Metal Dichalcogenide Monolayers. npj 2D Mater. Appl. 2021, 5, 17,  DOI: 10.1038/s41699-020-00197-7
  • Supporting Information

    Supporting Information


    The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.cgd.2c01134.

    • Raman spectra of WS2 and WSe2 MLs, XPS spectra of MOCVD-grown and exfoliated WS2, reproducibility of WS2 ML growth, and a summary about the Raman intensity ratios and BL/ML coverage ratio (PDF)


    Terms & Conditions

    Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.