Skip to main content
NCBI home page
JACS Au logoLink to JACS Au
. 2023 Feb 28;3(3):775–784. doi: 10.1021/jacsau.2c00536

Synthesis and Characterization of Transition Metal Dichalcogenide Nanoribbons Based on a Controllable O2 Etching

Ruben Canton-Vitoria †,‡,*, Takato Hotta †, Mengsong Xue †, Shaochun Zhang †, Ryo Kitaura †,§
PMCID: PMC10052231  PMID: 37006761

Abstract

graphic file with name au2c00536_0011.jpg

Although the synthesis of monolayer transition metal dichalcogenides has been established in the last decade, synthesizing nanoribbons remains challenging. In this study, we have developed a straightforward method to obtain nanoribbons with controllable widths (25–8000 nm) and lengths (1–50 μm) by O2 etching of the metallic phase in metallic/semiconducting in-plane heterostructures of monolayer MoS2. We also successfully applied this process for synthesizing WS2, MoSe2, and WSe2 nanoribbons. Furthermore, field-effect transistors of the nanoribbons show an on/off ratio of larger than 1000, photoresponses of 1000%, and time responses of 5 s. The nanoribbons were compared with monolayer MoS2, highlighting a substantial difference in the photoluminescence emission and photoresponses. Additionally, the nanoribbons were used as a template to build one-dimensional (1D)–1D or 1D–2D heterostructures with various transition metal dichalcogenides. The process developed in this study offers simple production of nanoribbons with applications in several fields of nanotechnology and chemistry.

Keywords: MoS2, nanoribbons, 2D−1D, chemical etching, CVD, phototransistor, optoelectronic, single layer

1. Introduction

Nanoscale ribbon-like structures, nanoribbons, are one-dimensional (1D) nanomaterials that have attracted significant attention. In contrast to three-dimensional (3D) structures, nanoribbons show quantum confinement effects and are expected to exhibit exotic optoelectronic and magnetic properties.1,2 For synthesizing nanoribbons, monolayer transition metal dichalcogenides (TMDs) with subnanometer thicknesses and room-temperature carrier mobility exceeding 50 cm2 V–1 s–1 are versatile precursors.3 TMDs can be represented as MX2, where M = Mo, W, Nb, or Ti, and X = S, Se, or Te.4 Furthermore, TMDs have several crystal phases: 1T (octahedral), 1T’ (distorted octahedral), and 2H (trigonal prismatic). Usually, 1T and 1T’ phases are mixed and have metallic character, whereas TMDs in the 2H phase have a sizable band gap, showing semiconducting character. 2H-TMDs are one of the most studied two-dimensional (2D) materials, in addition to graphene5 and hexagonal boron nitride (h-BN).6

Currently, the fabrication of subnanometer-thick TMD nanoribbons is challenging, and only a few examples are described in the literature.7 First, nanoribbons can be obtained by various ablation techniques.8−10 Other methods include liquid exfoliation of bulk TMDs11 or chemical etching.12 In these cases, yield and reproducibility are low and isolating nanoribbons is still challenging. For instance, O2 etching on semiconducting MoS2 yields several nanoribbons with a width of 100 nm. In this case, poor yield, crack formation, and poor reproducibility are significant problems. Currently, only chemical vapor deposition (CVD) processes can grow nanoribbons with sufficient yield and quality.13

Regarding the O2 etching process, substantial efforts have been made to comprehend the oxidation mechanism in semiconducting TMDs.14 It is well established that dry or wet air can oxidize single-layer semiconducting MoS2 at temperatures higher than 320 °C,15−17 forming triangular holes and shortening the edges of the layers.15−19 The oxidation starts at sulfur vacancies, where O2 can directly interact with Mo atoms,17,20,21 yielding MoO3 and SO2. Then, the oxidation continues, resulting in the most stable zig-zag edges and exposing Mo atoms (ZZ-Mo) at the edges.15,17,22 Metallic phases follow the same oxidation process as semiconducting but are more sensitive to oxidation. For example, at room temperature, O2 attacks the edges of 1T MoS2, oxidizing the surface in a few days.18,23−27 Regarding the metallic phases, efforts have been devoted to enhancing the stability,24,26−30 and the next challenge would be the utilization of the facile O2 etching of metallic phases for nanoscale structural control.

In this study, we developed a fabrication method for stable semiconducting MoS2 nanoribbons and investigated their potential applications. The nanoribbons have been obtained by selective O2 etching, erasing the metallic phase of metallic/semiconducting heterostructures. Atomic force microscopy (AFM) and transmission electron microscopy (TEM) observations have shown a crystalline ribbon-like structure with monolayer thickness; the width and length of the ribbons range from 25 to 8000 nm and 1 to 50 μm, respectively. Electron-dispersive X-ray spectroscopy (EDX) shows that the nanoribbons do not have noticeable contaminants or oxidized species. The semiconducting character of nanoribbons was confirmed by Raman, X-ray photoelectron spectroscopy (XPS), and photoluminescence (PL) spectroscopy. In addition, nanoribbon-based devices show a semiconducting character with a maximum photoresponse at 407 nm. The fabrication method developed in this study can be applied to WS2, MoSe2, and WSe2, and resulting nanoribbons can be employed as a template to perform new lateral or vertical heterostructures in combination with other TMDs.

2. Results and Discussion

2.1. Optical and Spectroscopic Characterization

Metallic/semiconducting in-plane heterostructures of MoS2 were synthesized by employing a method described in the bibliography.26 Briefly, elemental sulfur and a mixture of 75% MoO3 and 15% NaCl were heated in argon flow, resulting in metallic/semiconducting in-plane heterostructures. The metallic phase can be distinguished in optical images since it is slightly darker in optical contrasts (see Figures 1a and S1a).26 Triangular micro- or nanoribbons were obtained by the oxidation of MoS2 under dry airflow at 270 °C (Figures 1b and S1b). For additional information, see methods in the Supporting Information and Section 2.3.

Figure 1.

Figure 1

Open in a new tab

Optical microscopy of (a) monolayers of MoS2 and (b) triangular nanoribbons after O2 oxidation reaction. Note that darker areas inside triangular flakes (a) are removed during the O2 etching reaction.

In order to prove that nanoribbons are formed by the oxidation of the metallic phase in in-plane MoS2 heterostructures, we performed structural characterization by several spectroscopic techniques. First, Raman spectra with 633 nm excitation show differences between metallic and semiconducting phases. E12g, A1g, A1g’, and LA(M) modes have been recorded at 381, 406, 417, and 462 cm–1, respectively. In contrast, the basal plane shows appreciable intensities of J1, J2, and J3 peaks associated with the metallic phase at 151, 187, and 228 cm–1 (Figure 2a).26,32 Moreover, the edges of the material have a pure semiconducting phase, and the basal plane can be described as a mixture of phases with a predominant metallic polytype. Such distribution was further confirmed by Raman mapping (Figure 2b); however, compared with single layers, microribbons show a substantial reduction of metallic modes (Figure 2a,c). Concerning the semiconducting modes, the microribbons show a 20% enhancement of the 2LA(M) after oxidation, indicating a moderate increase in defects. In microribbons, the A1g’ mode of in-plane heterostructures was reduced significantly. Next, only E12g and A1g Raman signatures were observed in nanoribbons. The lack of A1g shift in nanoribbons indicates that the oxidation etching temperature is insufficient to attack the semiconducting phase, implying that p-doping from chemi- or physisorbed species is absent.16,31 Conclusively, the metallic polytype is more sensitive to oxygen than the semiconducting polytype at high temperatures, in concordance with the bibliography.25−28 Raman spectra taken with 533 nm excitation, on the other hand, enable the detailed study of the 700–1000 cm–1 region, revealing no evidence of MoOx species on nanoribbons14,25 (Figure S2).

Figure 2.

Figure 2

Open in a new tab

(a) Raman spectra at 633 nm for monolayers of metallic/semiconducting MoS2 at the basal plane (red) and at the edges (black), microribbons (olive), nanoribbons (blue), and substrate (gray). Mappings highlighting the metallic (red) and semiconducting (blue) phases (b) before and (c) after O2 etching of the remaining semiconducting triangular nanoribbons (blue).

XPS studies of MoS2 monolayers and nanoribbons are depicted in Figure 3. Before the O2 etching process, the S2p1/2, S2p3/2, Mo3d3/2, and Mo3d5/2 signals are shown at 162.2, 163.3, 232.5, and 229.3 eV, respectively. However, after oxidation, all peaks are blue-shifted by 10, 10, 20, and 10 meV, appearing at 162.3, 163.4, 232.7, and 229.4 eV, respectively. Such blue shifts are directly related to a reduction of the metallic character.22 Furthermore, the lack of bands at 232.8, 231.9, and 235.0 eV suggests the absence of Mo4+ species from MoOx or MoOx-Sy.33 Oxidation species of Mo6+ at 235.9 eV after O2 etching are negligible, corresponding to 5% of the overall area (MO3 vs MO3, Mo3d3/2, and Mo3d5/2) after adjusting the Scofield sensitivity factors. Consequently, the majority of α-MoO3 was indeed removed during the washing of the sample.22,34

Figure 3.

Figure 3

Open in a new tab

XPS measurement of (a) molybdenum and (b) sulfur of MoS2 (red) and triangular nanoribbons (blue). Semitransparent green lines represent the fitted areas of molybdenum and sulfur.

PL is a method sensitive to single-layer semiconducting MoS2. The edges of MoS2 monolayers have a strong PL emission at 1.82 eV (Figure 4a). In contrast, the PL intensity at the basal plane (metallic phase) is 30 times smaller than those at the edges, and the PL peak position shows a 7 meV redshift. The intensity at the basal plane completely disappears after oxidation (Figure 4a,b), and the resulting nanoribbons show PL emission at 1.89 eV, close to the emission energy of mechanical exfoliated MoS2. Additionally, there was no evidence of the defective chalcogen band in MoS2 between 1.7 and 1.8 eV. Furthermore, annealing at 270 °C in a vacuum or 270–500 °C under a sulfur supply does not enhance the PL emission of the nanoribbons. These results are consistent with high-quality nanoribbon samples with less amount of defects.

Figure 4.

Figure 4

Open in a new tab

(a) PL spectra measured with 488 nm laser excitation for monolayers of metallic/semiconducting MoS2 at the basal plane (red) and edges (black), microribbons (olive), nanoribbons (blue), and substrate (gray). PL images for (b) monolayers of MoS2 and (c) triangular nanoribbons. The excitation wavelength <550 nm and the emission wavelength >600 nm were used in PL images. The crossectional line-intensity profile for (d) monolayers of MoS2 and (e) triangular nanoribbons.

Regarding PL intensity in Figure 4a–d, we should point out that micro- and especially nanoribbons cannot be compared with the metallic/semiconducting heterostructures since the spot size of the laser (2 μm) is larger than the transverse widths of the micro- (0.5 μm) and nanoribbons (<50 nm). On the other hand, the PL images show triangular shape emissions with a higher intensity at the edges in metallic/semiconducting monolayers (Figure 4b,d). PL images of nanoribbons (Figure 4c,e) show well-defined triangular shapes, which are consistent with Raman (Figure 2c) and optical microscopy (Figure 1b). However, weak PL emission and significant A1g and E21g modes in Raman suggest that the basal plane is not a pure metallic phase but a mixture of several phases. Because the mixture of phases tends to have defects, O2 etching probably does not occur exclusively by the nature of the metallic phase but also by many defects.

2.2. Morphological Characterization

AFM images show triangular microribbons with longitudinal lengths larger than 50 μm and controllable transverse widths ranging between 5 μm (Figure 5a) and 50 nm (Figure 5b,c). Figure 5a unequivocally shows the desired sub-nanometric nanoribbons (0.7 nm) with an empty basal plane, in agreement with the color and shape of the optical microscope (Figure 1). According to previous studies, the triangular holes observed during the O2 etching indicate ZZ-Mo edges.15−19 In addition, the transverse width of the nanoribbon in Figure 5a has three different regions: (1°) 3 μm of continuous non-defective nanoribbons, which are directly correlated with the edges of the original MoS2 monolayers. (2°) Small islands with a length lower than 0.5 μm, and (3°) a huge space of 10 μm. Obviously, control of the region (2°) is essential for further applications. Concretely, nanoribbons with a smooth and well-defined (2°) region (Figure 5c) can be interesting for constructing devices or conducting fundamental studies. In contrast, the long-edge perimeter (Figure 5b) is attractive for chemical modification35−37 or catalytic purposes,38−40 especially by the direct interaction with Mo atoms.15−19 The procedures to obtain smooth or long edges will be explained later in Section 2.3.

Figure 5.

Figure 5

Open in a new tab

2D and 3D AFM topography images as well as the corresponding crossectional line-height profile of (a) microribbons, (b) low level of edge definition nanoribbons, and (c) high level of edge definition of the thin nanoribbon.

TEM observations agree with AFM results and confirm the presence of nanoribbons smaller than 15 nm (Figure 6ai). At the center of nanoribbons (Figure 6aii), MoS2 preserves all the atomic structures without holes or appreciable defects. Additionally, the fast Fourier transform (FFT) pattern (Figure 6aiii) exhibits the hexagonal pattern expected in single-layer semiconducting MoS2 but not in the metallic phase or MoO3. In sharp contrast, microribbons with long transverse widths have triangular-shaped holes near the edges. Triangular holes imply ZZ-Mo edges15,17,22 (Figure 6iv). The presence of holes on the basal plane of MoS2 suggests that the original layer has regions with a mixtures of semiconducting and metallic phases. Probably, the triangular holes were formed when the metallic/defective regions were removed during the O2 etching process. Well-defined transitions between metallic and semiconducting phases abolish such holes (Figure 6aii) and were obtained by reducing the time of the cooldown at the end of CVD synthesis (see more in detail in Section 2.3). EDX spectra of nanoribbons show the superposition of molybdenum and sulfur peaks at 2.32 keV and two additional Mo peaks at 17.45 and 19.6 keV (Figure 6b). The oxygen on the sample is well correlated with silicon, suggesting the existence of SiO2 impurities without a significant amount of MOx species. It is important to note that those samples were stored in the air and with light for 5 months. Therefore, suspended nanoribbons remained structurally stable for long periods, in agreement with other semiconducting nanoribbons.22

Figure 6.

Figure 6

Open in a new tab

TEM of (a) MoS2 nanoribbons. Inset of (a): (i) TEM of the free-of-hole nanoribbon, (ii) TEM ampliation of the center of (a), (iii) diffraction pattern of the nanoribbon, and (iv) basal plane of nanoribbons with low edge-definition. (b) EDX for nanoribbons (blue), CVD MoS2 (red), blank for the TEM copper grid (dark gray), and SiO2 (gray).

2.3. Optimization

First, we pointed out that O2 etching only works with single layers; bi- to multi-layers remain unaffected. The best temperature to remove the metallic phase without damaging the semiconducting polytype was 270–280 °C; the metallic polytype is not oxidized at 265 °C or lower, and degradation of the semiconducting polytype starts at 310 °C (Figure S3). In 30 min, the metallic phase is completely oxidized, and the longer reaction time does not substantially reduce the transverse width of the nanoribbon (see Figure S4). Next, airflow plays an important role. Under the absence of flow, monolayers remained stable for 36 h or longer, whereas micro- or nanoribbons were successfully obtained in 30 min when the flow was 400 mL/min.

The most relevant parameter during CVD growth is the cooling speed once the synthesis is finished. If cooling from 700 to 300 °C finishes in a few minutes, a large amount of a well-defined metallic phase remains, but cooling in a few hours yields only the semiconducting phase. Moreover, fast cooling is necessary to get a sharp interface of phases to produce thin and well-defined nanoribbons by O2 etching. The NaCl loading on TMDs is another critical parameter.26 Concretely, two different concentrations of NaCl onto MoO3 have been studied, at 5 and 33% (see Figure S5). Low concentrations of NaCl result in a small number of in-plane heterostructures, reducing the yield of microribbons, but well-defined edges on triangular nanoribbons were obtained. On the other hand, 33% of NaCl shows significantly sharper nanoribbons but with residual MoO3 or nanoparticles after the O2 etching. In this study, 15% of NaCl is the best for growing nanoribbons or microribbons.

Under these optimal conditions, the nanoribbons were annealed first in argon (300 °C), second in a vacuum (300 °C), and third in argon with a supply of sulfur (500 °C). The first annealing process aims to transform the residue of the metallic phase into the semiconducting phase. The second process removes the physisorbed O2 on the basal plane of MoS2, and the third heals sulfur vacancies and converts any MoOx species into amorphous MoS2. However, the annealing processes result in no changes in PL and Raman of nanoribbons. The absence of spectroscopic changes indicates that all metallic phases were removed during the O2 etching process, and the number of defects on MoS2 nanoribbons is depreciable.

Next, short transverse width nanoribbons were produced at lower temperatures during the CVD growth (616 or 650 °C compared to 725 °C). In addition, the change in CVD synthesis conditions also modifies the morphology of the nanoribbons, finding stars, ultra-small nanoribbons, or fall-show shapes (see Figure S6). Additionally, we discovered that the O2 etching reaction of the metallic phase could also occur in a hot plate with a flow of moist air (relative humidity over 70%) using a commercial fan in a natural atmosphere under various temperatures, resulting in 100 nm-thick nanoribbons (see Figure S7). Moreover, the O2 etching process employed to create nanoribbons is simple enough to be replicated in many laboratories.

In addition, we also extended the present O2 etching technique to other TMDs such as WS2, MoSe2, and WSe2. The Supporting Information(41) displays the synthesis and characterization of all the materials using techniques like optical microscopy, PL intensity maps, AFM, XPS, and Raman (Figures S8–S10).

2.4. Evaluation of the Defects on Nanoribbons

The defects/disorders in TMDs can be classified as (1) holes, (2) cracks, (3) boundaries, (4) folds, (5) vacancies, (6) adatoms, (7) carbon impurities, and (8) mixture of phases. In the followings, each defect/disorder in our samples is discussed. (1) Holes: when the transition between metallic and semiconducting phases is sharp, TEM and AFM show a lack of holes around the material (Figure 5b,c and Figure 6a). (2) Cracks: cracks are common during O2 etching, but their location can be easily predicted; boundaries or no-flat regions have a high possibility of cracks (Figure 1 and Figure 5a). (3) Boundaries: on the other hand, triangular TMD layers are typically monocrystalline, reducing the percentage of boundaries compared to other shapes as stars. (4) Folds: folds are typically performed on CVD samples when any layer is transferred from one substrate to another. Because our TMDs are synthesized and oxidized on the same substrate, the number of folds is kept to an absolute minimum. (5) Vacancies: chalcogen vacancies are expected in CVD-TMDs26 and can be enhanced after O2 etching. Since the nanoribbons have strong PL emission (see Figure 4b,c), Raman spectroscopy shows a depreciable increment in the 2LA(M) band (see Figure 2a), and TEM shows excellent crystallinity (see Figure 6). Therefore, we drew the conclusion that the O2 etching method did not increase the number of vacancies or degrade the nanoribbons’ properties. (6) Adatoms: NaCl is used in the CVD process, and chalcogen vacancies may be filled with either Na+ or Cl– ions.42 Nevertheless, neither EDX (see Figure 6b) nor XPS was able to detect any adatoms in layers of MoS2. (7) Carbon-related impurities: defects regarding polymer impurities are completely absent because only the ultrapure clean surface of SiO2 allows the synthesis of single layers of MoS2. However, external contamination, such as polymer coating during device fabrication, may increase the amount of contaminants. At 275 °C, O2 etching will surely oxidize common polymers such as polypropylene carbonate, decomposing to COx and guaranteeing a clean surface. (8) Mixture of phases: finally, we assume that all metallic phases have been eliminated, leaving exclusively semiconducting polytypes. A reaction mechanism is shown in Figure 7.

Figure 7.

Figure 7

Open in a new tab

Proposed model O2 etching reaction process.

2.5. Applications

2.5.1. Devices

First, we use electron beam lithography to construct devices based on metallic/semiconducting MoS2 in-plane heterostructures (please see Section 5.1, Supporting Information). A total of 23 of 25 field-effect transistor devices work, showing metallic character, and after the O2 etching, 20 nanoribbon-based devices remain operational. We want to point out that such devices have exclusively zig-zag Mo edges since they follow the lines of the triangular flakes.15,17,22,43 The critical benefit of our procedure over O2 etching of a pure semiconducting phase43 is that we can anticipate which flakes will end up as nanoribbons with a 90% success rate, allowing us to investigate the transition from single layers to nanoribbons.

Next, the transfer curve (Figure 8a) successfully shows the semiconducting character in three of the devices, with an on/off ratio of larger than 1000, comparable to the pure semiconducting CVD-MoS2 layer (see the Supporting Information, Figure S12a) and other studies.43 The second etching does not seem to improve the semiconducting character significantly. However, this is mainly attributed to the detection limit of our equipment, which is unable to record values at currents lower than 10–3 nA. Figure 8a displays the transfer curve and photoresponse of a representative device, which shows two nanoribbons with transverse widths of 40 and 25 nm and longitudinal lengths of 3.5 and 6.5 μm, respectively. Furthermore, during the O2 etching, the area of the device decreases by around 300-fold. Figure S11 depicts further device characterization techniques such as optical microscopy, Raman, PL emission, and AFM. Once we corrected the transverse widths of the device, we concluded that the mobility decreased at around 100-fold, indicating that the new nanoribbons have mobility lower than that of the 2D metallic/semiconducting MoS2. We tentatively assume that the metallic phase is more conductive than semiconducting. Moreover, the metallic phase governs the properties of the device before the O2 etching.

Figure 8.

Figure 8

Open in a new tab

(a) Transfer curve in the dark of metallic/semiconducting MoS2 (black line), microribbons (gray line), and nanoribbons (blue line). Dot lines correspond to metallic/semiconducting MoS2, microribbons, or nanoribbons under light irradiation at the wavelength specified in the legend. (b) Photoresponse under pulse light irradiation at 0.5 V of the in-plane heterostructure metallic/semiconducting MoS2 layer (black), semiconducting MoS2 layer (gray), microribbons (olive), and nanoribbons (blue). The power density of incident light is specified in the Supporting Information, Section 1.

White light increases the conductivity of the devices by 10 and 1000 fold after the first and second O2 etchings, respectively (see Figure 8b). Concretely, Figure 8b depicts a remarkable 50-fold increase in conductivity with a total energy irradiation of 1.5 pW on the area of the nanoribbons. In addition, the transfer curves of nanoribbons also show substantial changes under dark and light conditions (see Figure 8a). Shortly, the photoresponse of nanoribbons is independent of the gate voltage, but negative potentials are beneficial at the single layer of MoS2 (Figure 8a vs Figure S12a). For instance, the photoresponse at 0 Vg is lower in 2D layers than in the nanoribbons (30% vs 1000%). Unfortunately, the time response and the recovery response are longer in nanoribbons (around 0.5 s vs 5 s), as seen in Figure 8b. The enhanced photoresponse in nanoribbons can be explained by reducing the exciton diffusion perpendicular to the electrodes due to the low dimensions, leaving only the parallel diffusion available.

Similarly, we believe that exciton collisions at the edges (and defects) are mainly responsible for the longer time responses. In addition, nanoribbons also exhibit photoresponse differences from 2D materials at specific wavelengths. For example, 407 nm laser excitation shows double photoresponse than 532 nm in single-layer MoS2, while this value is inverted in nanoribbons (Figures 8b vs S12b). Such differences are in line with the changes previously described in PL emission (Figure 4a), wherein the band gap is a blue shift from 1.840 eV in monolayers to 1.889 eV in nanoribbons, suggesting excitonic differences caused by switching from 2D-to-1D confinement.

2.5.2. Heterostructures

Additionally, as a proof of concept, the new nanoribbons can be employed as a template to synthesize new lateral or vertical heterostructures. We have developed a set of MoS2/WS2 heterostructure derivatives employing triangular nanoribbons via CVD. Lateral heterostructures grown at the edges of the nanoribbons and the resulting MoS2/WS2 entirely show the semiconducting character, probably because the initial nanoribbon controls the crystallinity of WS2. On the other hand, a minor metallic phase was observed in vertical 1D MoS2/WS2 heterostructures, wherein the nanoribbon exclusively acts as a platform and is coupled with the new layer of WS2 through van der Waals forces. For example, optical microscope observations show how MoS2 nanoribbons were healed, forming in-plane heterostructures of MoS2/WS2 (Figure 9a). PL images, on the other hand, show stronger PL emissions in MoS2 than in WS2 regions (Figure 9b). Other techniques such as XPS, Raman, TEM, or EDX and the corresponding explanation can be found in the Supporting Information, Section 5.2, and Figures S13–14.

Figure 9.

Figure 9

Open in a new tab

(a) Optical microscopy and (b) PL intensity maps for in-plane heterostructures of MoS2/WS2.

3. Conclusions

In summary, we have developed a fabrication technique for semiconducting nanoribbons based on the etching of the metallic phase in metallic/semiconducting heterostructures of TMDs (MoS2, MoS2, WSe2, and WS2). We have optimized various parameters in this study to control the shape and transverse width of the nanoribbons (for a few tens to hundreds of nanometers). Also, we found that nanoribbons showed excellent photoresponse at 532 nm and chemical stability at high temperatures. Furthermore, we proved that nanoribbons could be employed as a template to perform new lateral or vertical heterostructures, opening new synthetic routes. The latter strategy should be strongly relevant for confining unstable TMDs such as HfSe244 in lateral heterostructures of semiconducting MoS2 to enhance its stability19,26 or synthesize new 1D heterostructures. Finally, the considerable ratio of perimeter/area of nanoribbons promises the existence of active edges, which allow the covalent functionalization with organic dyes for tailoring their optoelectronic properties.24,26,34−36,39,45

4. Experimental Section

4.1. Substrate Treatment

SiO2 substrates were washed in a Piranha solution for 15 min under sonication. Then, it was washed with ultrapure H2O and IPA (for devices). The surfaces were dried with N2 gun blow. Finally, substrates are heated at 735 °C for 30 min in an airflow of 400 mL/min.

4.2. Metallic/Semiconducting MoS2

First, 14.45 mg of MoO3 was mixed with 2.55 mg of NaCl and deposited in the center of an alumina boat (10 × 1.5 cm2) with a height of 1.1 cm–1. The SiO2 substrate is placed on the alumina boat over MoO3. Boats were introduced in a quartz tube with a 3 cm diameter. Two additional alumina boats of 3 × 0.5 × 0.3 cm3 filled with sulfur were placed at 45 cm from the SiO2 substrate. Argon gas flowed at a rate of 500 mL/min, and MoO3 was gradually heated until 725 °C (microribbons) or 650 °C (nanoribbons) in 13 min. Cool down was carried out for 15 min or shorter between 650 and 150 °C. Please see Table S1. Similar conditions were used for metallic/semiconducting WS2, MoSe2, and WSe2. Conditions were summarized in the below table. Please see Table S1.

4.3. MoS2 (WS2, MoS2, WSe2) Triangular Nanoribbons

First, 2 × 0.5 cm substrates with metallic/semiconducting TMD single layer flakes were heated at 270 °C (240, 250, and 250 °C); 10 min later, an air flux of 400 mL/min is added. Finally, the substrate was washed with MeOH to remove inner α-MoO3 (α-WO3) nanoparticles inside of nanoribbons. Please see Table S1.

4.4. Lateral MoS2/WS2 Heterostructure and Laterial MoS2/WS2 Heterostructure with a Hole

A 2 × 0.5 cm substrate of SiO2 with MoS2 is placed face-down on a boat with a height of 0.5 cm–1. The center of the boat (with a MoS2 substrate) contains 1 mg of WO3 containing 17% of NaCl. Two additional alumina boats of 3 × 0.5 × 0.3 cm3 filled with sulfur were placed at 45 cm from the SiO2 substrate. WO3 is heated for 13 min at 740 °C and then for 30 min at 740 °C. After the first 5 min, the sulfur boat was heated to 210 °C and kept constant for 35 min. For lateral MoS2/WS2 heterostructure, argon flow was constant in all syntheses (300 mL/min), and for lateral MoS2/WS2 heterostructure with a hole at 500 mL/min, please see Table S2.

4.5. Vertical MoS2/WS2 Heterostructure with a Hole

a 2 × 0.5 cm SiO2 substrate with MoS2 is located at 1.5 cm from the center at which 1 mg of WO3 (17% of NaCl) powder is placed. The nanoribbons are placed at the wall of the boat (with an inclination of 30–45°, the contact area of the boat and substrate should be minimized). The boat was heated for 13 min at 725 °C and then for 30 min at 780 °C. At the same time, the sulfur boat was heated to 210 °C in 5 min, and the temperature was kept constant for 30 additional minutes. Argon flow was constant (113 mL/min) in all experiments. For details about conditions, please see Table S2.

See additional information in the Supporting Information, Section 1.

Acknowledgments

This work was also supported by the Japan Society for the Promotion of Science (JSPS) postdoctoral fellowship grant agreement No P19368. This work was supported by JSPS KAKENHI Grant Numbers JP21K18930, JP20H02566, JP22H05458, and JP20H05664 and JST CREST Grant Numbers PMJCR16FJ3 and JPMJCR19H4, and JST PRESTO Grant Number JPMJPR20A2. Finally, we want to thank all the technical staff at the Department of Chemistry at Nagoya University, especially Chen Zhao for the help with CVD synthesis and Kaori Asai for her proofreading.

Glossary

Abbreviations

CVD

chemical vapor deposition

TMDs

transition-metal dichalcogenides

0D

0 dimensional

2D

2 dimensional

CB

conduction band

VB

valence band

BOC

tert-butoxycarbony ethylene glycol

ZnP

zinc-porphyrin

FWHM

full width at half maximum

AA

biexciton

A-

trion

A0

neutral exciton

AB

and B-exciton

h-BN

hexagonal boron nitride

HOMO

highest occupied molecular orbital

LUMO

lowest energy orbital

EDX

electron-dispersive X-ray

TEM

transmission electron microscopy

AFM

atomic force microscopy

R

responsivity

ΔVTh

variation of threshold voltage with and without light

ZZ-Mo

zig-zag Mo edges

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacsau.2c00536.

  • Experimental sections and figures that support spectroscopic characterization, such as EDX, PL, or Raman, and morphological characterization, like TEM and AFM; description of the devices; and short study regarding the synthesis of WS2, MoSe2, and WSe2 nanoribbons, as well as 1D–2D in-plane, 1D–1D in-plane, and 1D–1D vertical heterostructures (PDF)

Author Contributions

CRediT: Ruben Canton-Vitoria conceptualization, data curation, formal analysis, investigation, methodology, project administration, resources, software, supervision, validation, visualization, writing-original draft, writing-review & editing; Ryo Kitaura conceptualization, funding acquisition, supervision, visualization, writing-review & editing.

The authors declare no competing financial interest.

Supplementary Material

au2c00536_si_001.pdf (3.5MB, pdf)

References

  1. Pan H.; Zhang Y.-W. Edge-dependent structural, electronic and magnetic properties of MoS2 nanoribbons. J. Mater. Chem. 2012, 22, 7280. 10.1039/c2jm15906f. [DOI] [Google Scholar]
  2. López-Urías F.; Laura Elías A.; Perea-López N.; Gutiérrez H.; Terrones M.; Terrones H. Electronic, magnetic, optical, and edge-reactivity properties of semiconducting and metallic WS2 nanoribbons. 2D Mater. 2015, 2, 015002 10.1088/2053-1583/2/1/015002. [DOI] [Google Scholar]
  3. Shen P.; Su C.; Lin Y.; Chou A.; Cheng C.; Park J.; Chiu M.; Lu A.; Tang H.; Tavakoli; Pitner G.; Ji X.; Cai Z.; Mao N.; Wang J.; Tung V.; Li J.; Bokor J.; Zettl A.; Wu C. I.; Palacios T.; Li L. J.; Kong J. Ultralow contact resistance between semimetal and monolayer semiconductors. Nature 2021, 593, 211–217. 10.1038/s41586-021-03472-9. [DOI] [PubMed] [Google Scholar]
  4. Sharma R.; Laishram R.; Gupta B. K.; Srivastva R.; Sinha O. P. A review on MX2 (M = Mo, W and X = S, Se) layered material for opto-electronic devices. Adv. Nat. Sci.: Nanosci. Nanotechnol. 2022, 13, 023001 10.1088/2043-6262/ac5cb6. [DOI] [Google Scholar]
  5. Saraswat V.; Jacobberger R. M.; Arnold M. S. Materials science challenges to graphene nanoribbon Electronics. ACS Nano 2021, 15, 3674–3708. 10.1021/acsnano.0c07835. [DOI] [PubMed] [Google Scholar]
  6. Zhang K.; Feng Y.; Wang F.; Yang Z.; Wang J. Two-dimensional hexagonal boron nitride (2D-hBN): Synthesis, properties and applications. J. Mater. Chem. C 2017, 5, 11992–12022. 10.1039/C7TC04300G. [DOI] [Google Scholar]
  7. Chowdhury T.; Sadler E. C.; Kempa T. J. Progress and prospects in transition-metal dichalcogenide research beyond 2D. Chem. Rev. 2020, 120, 12563–12591. 10.1021/acs.chemrev.0c00505. [DOI] [PubMed] [Google Scholar]
  8. Vasu K.; Yamijala S. S. R. K. C.; Zak A.; Gopalakrishnan K.; Pati S.; Rao C. Clean WS2 and MoS2 nanoribbons generated by laser-induced unzipping of the nanotubes. Small 2015, 11, 3916–3920. 10.1002/smll.201500350. [DOI] [PubMed] [Google Scholar]
  9. Munkhbat B.; Yankovich A. B.; Baranov D. G.; Verre R.; Olsson E.; Shegai T. O. Transition metal dichalcogenide metamaterials with atomic precision. Nat. Commun. 2020, 11, 4604. 10.1038/s41467-020-18428-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chen S.; Kim S.; Chen W.; Yuan J.; Bashir R.; Lou J.; van der Zande A.; King W. Monolayer MoS2 nanoribbon transistors fabricated by scanning probe lithography. Nano Lett. 2019, 19, 2092–2098. 10.1021/acs.nanolett.9b00271. [DOI] [PubMed] [Google Scholar]
  11. Nethravathi C.; Jeffery A.; Rajamathi M.; Kawamoto N.; Tenne R.; Golberg D.; Bando Y. Chemical unzipping of WS2 Nanotubes. ACS Nano 2013, 7, 7311–7317. 10.1021/nn4029635. [DOI] [PubMed] [Google Scholar]
  12. Ionescu R.; George A.; Ruiz I.; Favors Z.; Mutlu Z.; Liu C.; Ahmed K.; Wu R.; Jeong J.; Zavala L.; Mkhoyan L.; Ozkan A.; Ozkan C. S. Oxygen Etching of thick MoS2 films. Chem. Commun. 2014, 50, 11226–11229. 10.1039/C4CC03911D. [DOI] [PubMed] [Google Scholar]
  13. Li S.; Lin Y.; Zhao W.; Wu J.; Wang Z.; Hu Z.; Shen Y.; Tang D.; Wang J.; Zhang Q.; Zhu H.; Chu L.; Zhao W.; Liu C.; Sun Z.; Taniguchi T.; Osada M.; Chen W.; Xu Q.; Wee A. T. S.; Suenaga K.; Ding F.; Eda G. Vapour–liquid–solid growth of monolayer MoS2 nanoribbons. Nat. Mater. 2018, 17, 535–542. 10.1038/s41563-018-0055-z. [DOI] [PubMed] [Google Scholar]
  14. Szoszkiewicz R. Local Interactions of Atmospheric Oxygen with MoS2 Crystals. Materials 2021, 14, 5979. 10.3390/ma14205979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Zhou H.; Yu F.; Liu Y.; Zou X.; Cong C.; Qiu C.; Yu T.; Yan Z.; Shen X.; Sun L.; et al. Thickness-dependent patterning of MoS2 sheets with well-oriented triangular pits by heating in air. Nano Res. 2013, 6, 703–711. 10.1007/s12274-013-0346-2. [DOI] [Google Scholar]
  16. Yamamoto M.; Einstein T. L.; Fuhrer M. S.; Cullen W. G. Anisotropic etching of atomically thin MoS2. J. Phys. Chem. C 2013, 117, 25643–25649. 10.1021/jp410893e. [DOI] [Google Scholar]
  17. Lv D.; Wang H.; Zhu D.; Lin J.; Yin G.; Lin F.; Zhang Z.; Jin C. Atomic process of oxidative etching in monolayer molybdenum disulfide. Sci. Bull. 2017, 62, 846–851. 10.1016/j.scib.2017.05.016. [DOI] [PubMed] [Google Scholar]
  18. Nurdiwijayanto L.; Ma R.; Sakai N.; Sasaki T. Insight into the structural and electronic nature of chemically exfoliated molybdenum disulfide nanosheets in aqueous dispersions. Dalton Trans. 2018, 47, 3014–3021. 10.1039/C7DT03706F. [DOI] [PubMed] [Google Scholar]
  19. Ukegbu U.; Szoszkiewicz R. Microscopic kinetics of heat-induced oxidative etching of thick MoS2 crystals. J. Phys. Chem. C 2019, 123, 22123–22129. 10.1021/acs.jpcc.9b02739. [DOI] [Google Scholar]
  20. Hong J.; Hu Z.; Probert M.; Li K.; Lv D.; Yang X.; Gu L.; Mao N.; Feng Q.; Xie L.; Zhang J.; Wu D.; Zhang Z.; Jin C.; Ji W.; Zhang X.; Yuan J.; Zhang Z. Exploring atomic defects in molybdenum disulphide monolayers. Nat. Commun. 2015, 6, 6293. 10.1038/ncomms7293. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. 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 defects in monolayer molybdenum disulfide. Nano Lett. 2013, 13, 2615–2622. 10.1021/nl4007479. [DOI] [PubMed] [Google Scholar]
  22. De Castro I. A.; Datta R. S.; Ou J. Z.; Castellanos-Gomez A.; Sriram S.; Daeneke T.; Kalan-tar-Zadeh K. Molybdenum oxides—from fundamentals to functionality. Adv. Mater. 2017, 29, 1–31. 10.1002/adma.201701619. [DOI] [PubMed] [Google Scholar]
  23. Nurdiwijayanto L.; Ma R.; Sakai N.; Sasaki T. Stability and nature of chemically exfoliated MoS2 in aqueous suspensions. Inorg. Chem. 2017, 56, 7620–7623. 10.1021/acs.inorgchem.7b01354. [DOI] [PubMed] [Google Scholar]
  24. Canton-Vitoria R.; Sayed-Ahmad-Baraza Y.; Humbert B.; Arenal R.; Ewels C.; Tagmatarchis N. Pyrene coating transition metal disulfides as protection from photooxidation and environmental aging. J. Nanomater. 2020, 10, 363. 10.3390/nano10020363. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Longo R. C.; Addou R.; Noh J.-Y.; Smyth C. M.; Barrera D.; Zhang C.; Hsu J.; Wallace R.; Cho K. Intrinsic air stability mechanisms of two-dimensional transition metal dichalcogenide surfaces: Basal versus edge oxidation. 2D Mater. 2017, 4, 025050 10.1088/2053-1583/aa636c. [DOI] [Google Scholar]
  26. Canton-Vitoria R.; Hotta T.; Liu Z.; Inoue T.; Kitaura R. Stabilization of metallic phases through formation of metallic/semiconducting lateral heterostructures. J. Chem. Phys. 2020, 153, 084702 10.1063/5.0012782. [DOI] [PubMed] [Google Scholar]
  27. Yang C.; Wang B.; Xie Y.; Zheng Y.; Jin C. Deriving MoS2 nanoribbons from their flakes by chemical vapor deposition. Nanotechnology 2019, 30, 255602. 10.1088/1361-6528/ab0a1d. [DOI] [PubMed] [Google Scholar]
  28. Enyashin A. N.; Yadgarov L.; Houben L.; Popov I.; Weidenbach M.; Tenne R.; Bar-Sadan M.; Seifert G. New route for stabilization of 1T-WS2 and MoS2 phases. J. Phys. Chem. C 2011, 115, 24586–24591. 10.1021/jp2076325. [DOI] [Google Scholar]
  29. Lee G.-H.; Cui X.; Kim Y. D.; Arefe G.; Zhang X.; Lee C.-H.; Ye F.; Watanabe K.; Taniguchi T.; Kim P.; Hone J. Highly stable, dual-gated MoS2 transistors encapsulated by hexagonal boron nitride with gate-controllable contact, resistance, and threshold voltage. ACS Nano 2015, 9, 7019. 10.1021/acsnano.5b01341. [DOI] [PubMed] [Google Scholar]
  30. Kan M.; Wang J.; Li S.; Zhang H.; Li Y. W.; Kawazoe Y.; Sun Q.; Jena P. Structures and phase transition of a MoS2 monolayer. J. Phys. Chem. C 2014, 118, 1515. 10.1021/jp4076355. [DOI] [Google Scholar]
  31. Chakraborty B.; Bera A.; Muthu D. V. S.; Bhowmick S.; Waghmare U. V.; Sood A. K. Symmetry-dependent phonon renormalization in monolayer MoS2 transistor. Phys. Rev. B 2012, 85, 161403 10.1103/PhysRevB.85.161403. [DOI] [Google Scholar]
  32. Jiang L.; Zhang S.; Kulinich S.; Song X.; Zhu J.; Wang X.; Zeng H. Optimizing hybridization of 1T and 2H phases in MoS2 monolayers to improve capacitances of supercapacitors. Mater. Res. Lett. 2015, 3, 177–183. 10.1080/21663831.2015.1057654. [DOI] [Google Scholar]
  33. Lamkaouane H.; Ftouhi H.; Richard-Plouet M.; Gautier N.; Stephant N.; Zazoui M.; Addou M.; Cattin L.; Christian B. J.; Mir Y.; Louarn G. Efficient and facile synthetic route of MoO3:MoS2 hybrid thin layer via oxidative reaction of MoS2 nanoflakes. Nanomaterials 2022, 12, 3171. 10.3390/nano12183171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Szoszkiewicz R.; Rogala M.; Dabrowski P. Surface-bound and volatile Mo oxides produced during oxidation of single MoS2 crystals in air and high relative humidity. Materials (Basel) 2020, 13, 3067. 10.3390/ma13143067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Canton-Vitoria R.; Scharl T.; Stergiou A.; Cadranel A.; Arenal R.; Guldi D.; Tagmatarchis N. Ping-pong energy transfer in covalently linked porphyrin-MoS2 architectures. Angew. Chem., Int. Ed. 2020, 59, 3976–3981. 10.1002/anie.201914494. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Canton-Vitoria R.; Sayed-Ahmad-Baraza Y.; Pelaez-Fernandez M.; Arenal R.; Bittencourt C.; Ewels C.; Tagmatarchis N. Functionalization of MoS2 with 1,2-dithiolanes: toward donor-acceptor nanohybrids for energy conversion. npj 2D Mater. Appl. 2017, 1, 13. 10.1038/s41699-017-0012-8. [DOI] [Google Scholar]
  37. Kagkoura A.; Canton-Vitoria R.; Vallan L.; Hernandez-Ferrer J.; Benito A.; Maser W.; Arenal R.; Tagmatarchis N. Bottom-up synthesized MoS2 interfacing polymer carbon nanodots with electrocatalytic activity for hydrogen evolution. Chem. – Eur. J. 2020, 26, 6635–6642. 10.1002/chem.202000125. [DOI] [PubMed] [Google Scholar]
  38. He H.; Lin J.; Fu W.; Wang X.; Wang H.; Zeng Q.; Gu Q.; Li Y.; Yan C.; Tay B.; Xue K.; Hu C.; Pantelides X.; Zhou S.; Liu Z. MoS2/TiO2 edge-on heterostructure for efficient photocatalytic hydrogen evolution. Adv. Energy Mater. 2016, 6, 1600464 10.1002/aenm.201600464. [DOI] [Google Scholar]
  39. Parzinger E.; Miller B.; Blaschke B.; Garrido J.; Ager J.; Holleitner A.; Wurstbauer U. Photocatalytic stability of single- and few-layer MoS2. ACS Nano 2015, 9, 11302–11309. 10.1021/acsnano.5b04979. [DOI] [PubMed] [Google Scholar]
  40. Canton-Vitoria R.; Vallan L.; Urriolabeitia E.; Benito A.; Maser W.; Tagmatarchis N. Electronic interactions in illuminated carbon dot/MoS2 ensembles and electrocatalytic activity towards hydrogen evolution. Chem. – Eur. J. 2018, 24, 10468–10474. 10.1002/chem.201801425. [DOI] [PubMed] [Google Scholar]
  41. Zhou J.; Lin J.; Huang X.; Zhou Y.; Chen Y.; Xia J.; Wang H.; Xie Y.; Yu H.; Lei J.; et al. A Library Of Atomically Thin Metal Chalcogenides. Nature 2018, 556, 355–359. 10.1038/s41586-018-0008-3. [DOI] [PubMed] [Google Scholar]
  42. Wang W.; Shu H.; Wang J.; Cheng Y.; Liang P.; Chen X. Defect passivation and photoluminescence enhancement of monolayer MoS2 crystals through sodium halide-assisted chemical vapor deposition growth. ACS Appl. Mater. Interfaces 2020, 12, 9563–9571. 10.1021/acsami.9b19224. [DOI] [PubMed] [Google Scholar]
  43. Aslam M. A.; Tran T. H.; Supina A.; Siri O.; Meunier V.; Watable K.; Taniguchi T.; Kralj M.; Teichert C.; Sheremet E.; Rodrigez R. D.; Matkovic A. Single-crystalline nanoribbon network field effect transistors from arbitrary two-dimensional materials. npj 2D Mater. Appl. 2022, 6, 76. 10.1038/s41699-022-00356-y. [DOI] [Google Scholar]
  44. Mirabelli G.; McGeough C.; Schmidt M.; McCarthy E.; Monaghan S.; Povey I.; McCarthy M.; Gity F.; Nagle R.; Hughes G.; Cafolla A.; Hurley P. K.; Duffy R. Air Sensitivity of MoS2, MoSe2, MoTe2, HfS2, and HfSe2. J. Appl. Phys. 2016, 120, 125102–125111. 10.1063/1.4963290. [DOI] [Google Scholar]
  45. Canton-Vitoria R.; Gobeze H. B.; Blas-Ferrando V. M.; Ortiz J.; Jang Y.; Fernández-Lázaro F.; Sastre-Santos Á.; Nakanishi Y.; Shinohara H.; D’Souza F.; Tagmatarchis N. Excited-state charge transfer in covalently functionalized MoS2 with a Zinc phthalocyanine donor–acceptor hybrid. Angew. Chem., Int. Ed. 2019, 58, 5712–5717. 10.1002/anie.201900101. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

au2c00536_si_001.pdf (3.5MB, pdf)

Articles from JACS Au are provided here courtesy of American Chemical Society

RESOURCES