Janus MoSSe Nanotubes on 1D SWCNT‐BNNT van der Waals Heterostructure

Abstract

Two‐dimensional (2D) Janus transition metal dichalcogenide (TMDC) layers with broken mirror symmetry exhibit giant Rashba splitting and unique excitonic behavior. For their one‐dimensional (1D) counterparts, the Janus nanotubes possess curvature, which introduces an additional degree of freedom to break the structural symmetry. This can potentially enhance these effects or even give rise to novel properties. Moreover, Janus MSSe nanotubes (M = W, Mo), with diameters surpassing 40 Å and Se positioned externally consistently demonstrate lower energy states compared to their Janus monolayer counterparts. However, there are limited studies on the preparation of Janus nanotubes, due to the synthesis challenge and limited sample quality. In this study, we first synthesized MoS₂ nanotubes on single‐walled carbon nanotube (SWCNT) and boron nitride nanotube (BNNT) heterostructures and then explored the growth of Janus MoSSe nanotubes from MoS₂ nanotubes at room temperature with the assistance of H₂ plasma. The successful formation of the Janus structure is confirmed by Raman spectroscopy, and atomic structure and elemental distribution of the grown samples are further characterized by advanced electronic microscopy. The synthesis of Janus MoSSe nanotubes based on SWCNT‐BNNT heterostructures paves the way for further exploration of novel properties in Janus TMDC nanotubes.

Janus MoSSe nanotubes exhibit unique properties and demonstrate lower strain energy than their corresponding Janus monolayer counterparts when the diameter exceeds 40 Å. In this work, Janus MoSSe nanotubes are successfully synthesized from MoS₂ nanotubes using H₂ plasma at room temperature. The formation of the Janus structure is confirmed through Raman spectroscopy and microscopy characterization, enabling further exploration of their novel properties.

1. Introduction

Two‐dimensional (2D) Janus transition metal dichalcogenides (TMDCs) represent a promising class of quantum materials, featuring different chalcogen atoms at the top and bottom of the transition metal layer.^{[ 1} , ^{2 ]} This unique atomic configuration breaks the out‐of‐plane mirror symmetry, resulting in an intrinsic electric field due to the electronegativity difference between the two chalcogens.^{[ 3} , ⁴ , ^{5 ]} Investigations to date have revealed that many intriguing physical properties emerge in Janus TMDC monolayers, including Rashba spin‐orbit coupling,^{[ 6} , ^{7 ]} second harmonic generation (SHG),^{[ 8} , ^{9 ]} piezoelectric polarization,^{[ 10} , ^{11 ]} and others.^{[ 2 ]}

Their one‐dimensional (1D) counterparts, Janus TMDC nanotubes, possess curvature, which provides an additional degree of freedom to break the structural symmetry and potentially enhance these effects.^{[ 12 ]} As demonstrated by Hung et al., the SHG intensity in Janus MoSSe/MoS₂ heterobilayers can be significantly improved through strain engineering.^{[ 13 ]} Cai et al. also reported that the optimal power conversion efficiency of 1D MoSSe/WSe₂ heterostructure‐based solar cells reaches up to 6.25%, which is considerably higher than that of their 2D counterparts (1.94%).^{[ 14 ]} Additionally, MSSe (M = W, Mo) nanotubes with diameters exceeding 40 Å (with Se positioned outside the tube wall) are energetically more stable than the corresponding Janus monolayers.^{[ 15 ]} According to theoretical predictions, the strain energy of Janus MoSSe nanotubes reaches a minimum at a diameter of around 80 Å.^{[ 15} , ^{16 ]} The enhanced performance and excellent stability of Janus TMDC nanotubes make them highly promising for practical applications. However, the preparation of Janus TMDC materials remains challenging. Currently, experimental efforts in the synthesis of Janus TMDC materials are still in the exploratory stage and primarily focused on 2D Janus monolayers,^{[ 10} , ¹⁷ , ¹⁸ , ¹⁹ , ^{20 ]} while little attention has been paid to the preparation of high‐crystallinity 1D Janus nanotubes.^{[ 21 ]}

Here, we report the fabrication and characterization of Janus MoSSe nanotubes (MoSSeNTs) based on 1D van der Waals heterostructures composed of single‐walled carbon nanotubes (SWCNTs) and hexagonal boron nitride nanotubes (BNNTs). First, MoS₂ nanotubes (MoS₂NTs) were synthesized on the SWCNT‐BNNT templates and then converted into Janus MoSSeNTs at room temperature using H₂ plasma. Raman spectroscopy confirms the successful formation of the Janus MoSSe structure. Transmission electron microscopy (TEM) characterization reveals that the structure of the synthesized Janus MoSSe nanotubes remains intact. Elemental distribution and sulfur‐to‐molybdenum (S/Mo) ratio analysis for single‐ and double‐walled TMDC nanotubes provide evidence that only the exposed outer‐layer sulfur atoms in MoS₂ nanotubes are replaced by selenium. Furthermore, cross‐sectional images of the synthesized 1D Janus heteronanotubes illustrate a coaxial structure with a Se‐Mo‐S sandwiched configuration. Additionally, it is challenging to convert MoSe₂ nanotubes into Janus nanotube structures, which is consistent with the theoretical calculations.

2. Results and Discussion

Figure 1a schematically illustrates the fabrication process of Janus MoSSe nanotubes. The SWCNT‐BNNT coaxial heterostructures,^{[ 22} , ^{23 ]} loaded on a ceramic washer, serve as the growth substrate throughout the synthesis process. First, MoS₂NTs were grown on the SWCNT‐BNNT coaxial nanotubes via the low‐pressure chemical vapor deposition (LPCVD) method. Subsequently, the SWCNT‐BNNT‐MoS₂NT sample, along with selenium (Se) particles, was placed in a low‐pressure chamber, where sulfur (S) atoms were selectively replaced by Se under H₂ plasma treatment. For nanotube samples at different stages, the color change in the sample films is visible to the naked eye, as illustrated in Figure 1b. The SWCNT‐BNNT heterostructure is initially transparent but turns yellow after the growth of MoS₂. Following exposure to H₂ plasma, the sample further changes to a brown color. This difference is also evident in the optical absorption spectrum (Figure S1a, Supporting Information).

Figure 1.

a) Schematic illustration of the fabrication process of Janus MoSSe nanotubes based on 1D SWCNT‐BNNT heterostructures. b) Optical image of the suspended SWCNT‐BNNT, SWCNT‐BNNT‐MoS₂, and SWCNT‐BNNT‐MoSSe films mounted on ceramic washers. c) Raman spectra of SWCNT‐BNNT‐MoS₂ and SWCNT‐BNNT‐MoSSe samples measured in the range from 200 to 450 cm⁻¹. d) TEM and HAADF‐STEM images of the SWCNT‐BNNT‐MoSSeNT. e) A magnified HAADF‐STEM image of the SWCNT‐BNNT‐MoSSeNT.

To confirm the formation of the Janus structure, Raman spectroscopy was employed to analyze the samples after both MoS₂ synthesis and the Janus conversion process. The overall Raman spectra, presented in Figure S1b (Supporting Information), exhibit characteristic peaks originating from CNT, BN, and MoS₂/MoSSe within the range of 200 to 1800 cm⁻¹.^{[ 24} , ^{25 ]} A magnified view of these Raman spectra in the 200 to 450 cm⁻¹ region is provided in Figure 1c for clearer depiction. For the fabricated MoS₂ nanotube sample, characteristic in‐plane E_{2 g} and out‐of‐plane A_{1 g} vibration modes are observed at 386 and 406 cm⁻¹, respectively.^{[ 24 ]} After the Janus MoSSe synthesis process, a new peak with strong intensity appears at 287.3 cm⁻¹, corresponding to the out‐of‐plane S‐Mo‐Se vibration mode. Simultaneously, another distinct peak assigned to the in‐plane Janus vibration mode appears at 354.5 cm⁻¹.^{[ 10 ]} However, the A_{1 g} vibration mode from MoS₂, with significantly lower intensity at 406 cm⁻¹, is also observed, indicating that the majority (likely 70–80%), but not all (reasons to be discussed later), of MoS₂ was converted to the Janus MoSSe structure. In addition, the A_{1 g} and E_{2 g} modes for monolayer Janus MoSSe synthesized by a similar method are located at 288 cm⁻¹ and 355 cm⁻¹, respectively.^{[ 19 ]} The peak positions of the MoSSe nanotubes are red‐shifted compared to those of single‐layer Janus MoSSe, indicating a clear effect from the crystal curvature.

A typical TEM image of the SWCNT‐BNNT‐MoSSeNT van der Waals heterostructure is shown in Figure 1d, clearly revealing the coaxial structure. Additional TEM images in Figure S2 (Supporting Information) reveal that the outermost layer of TMDC nanotubes is predominantly monolayer, with a minority of regions exhibiting bilayer TMDC nanotube growth. Furthermore, as evidenced by the high‐angle annular dark‐field scanning transmission electron microscopy (HAADF‐STEM) characterization, the outermost MoSSe nanotubes are uniformly coated onto the SWCNT‐BNNT template, with their crystalline structures remaining intact even after H₂ plasma treatment. A magnified HAADF‐STEM image of a Janus MoSSe nanotube is shown in Figure 1e. As observed in the reference, in MoS₂ nanotube samples, the Mo sites exhibit significantly higher brightness than the S₂ sites.^{[ 21 ]} However, in Janus nanotube samples that have undergone Se substitution, the S₂ sites change into SeS and display an intensity that is comparable to that of the Mo sites, as indicated by the blue rectangle. Additionally, in the sidewall of the nanotube, as indicated by the yellow rectangle, the outer chalcogen atoms appear brighter than the inner ones, which can be attributed to the presence of Se and S atoms, respectively.^{[ 26 ]}

The distribution of elements in the fabricated SWCNT‐BNNT‐MoSSeNT heterostructure is further verified by electron energy‐loss spectroscopy (EELS) and energy‐dispersive X‐ray spectroscopy (EDS) mapping. The extensive elemental distribution across the sample is visually demonstrated in Figure S2g (Supporting Information), which confirms the large‐area homogeneity of the sample. Figure 2a illustrates the distribution of elements within a single‐walled Janus MoSSe nanotube. Notably, the innermost carbon (C) signal originates from the SWCNT, while the boron (B) and nitrogen (N) signals correspond to the BNNT in the middle layer. At the outermost layer, molybdenum (Mo), sulfur (S), and selenium (Se) are all detected. Compared to pure MoS₂ and MoSe₂ nanotubes grown on SWCNT‐BNNT templates (Figure 2b,c), the S and Se signal intensities in Janus nanotubes are significantly lower than those in MoS₂ and MoSe₂ nanotubes.

Figure 2.

HAADF‐STEM images and EELS/EDS chemical maps for single‐walled a) MoSSe, b) MoS₂, and c) MoSe₂ nanotubes grown on SWCNT‐BNNT coaxial heterostructures. Mapping of C/B/N/Mo by EELS and Se by EDS.

In the synthesized Janus nanotube, which features both single‐ and double‐walled structures, the distribution of elements exhibits notable variations. As depicted in Figure 3a, within the upper single‐walled segment, the presence of S, Se, and Mo elements is evident, with their signal intensities appearing relatively comparable. Conversely, in the double‐walled region, while all three elements (S, Se, and Mo) are still present, their signal intensities differ significantly. Specifically, when transitioning from the monolayer to the bilayer area, the intensities of S and Mo increase markedly, whereas the signal intensity of Se remains consistently low. This observation underscores the distinct elemental arrangements and interactions within the different wall configurations of the coaxial nanotube.

Figure 3.

a) HAADF‐STEM image and EELS/EDS chemical maps for single/double‐walled Janus MoSSe heteronanotubes. Mapping of C/B/N/Mo by EELS and Se by EDS. b) Schematic of MoS₂, MoSSe, and MoS₂@MoSSe nanotubes, along with the corresponding S/Mo ratio. c) HAADF‐STEM images of synthesized single/double walled heteronanotubes, the distribution of S and Mo elements, and the estimated S/Mo ratio.

According to the study by Trivedi et al.,^{[ 18 ]} during the transformation of WSe₂ into Janus WSSe, sulfur substitution is limited to the outermost selenium atoms. This selective substitution is attributed to the lower energy barrier and higher reactivity of the surface layer, facilitating the preferential incorporation of sulfur at the surface. Based on this, we propose that the disparate signal intensities of the three elements, S, Mo, and Se, within the single‐ and double‐walled parts can be attributed to the selective replacement of the outermost exposed S atoms on MoS₂ nanotubes. This hypothesis is also corroborated by the estimation of the S/Mo ratio by EELS. As shown in Figure 3b, the S/Mo ratio in the MoS₂ nanotube is used as a reference and corrected to a value of 2.0. In the monolayer Janus MoSSe nanotube, the S/Mo ratio is expected to decrease to 1, given that the outermost layer of S atoms has been replaced by Se atoms. In the double‐walled part, the S/Mo ratio should be approximately 1.5, as only the outermost S atoms have been substituted by Se, forming a MoS₂@MoSSe heteronanotube structure. In the synthesized samples, as shown in Figure 3c, the estimated S/Mo ratio in the single‐walled part is close to 1 (1.11 and 1.06, respectively), and the value in the double‐walled part is close to 1.5 (1.47 and 1.51, respectively). These results provide indirect evidence that only the exposed outermost layer of S atoms in MoS₂ nanotubes is replaced by Se. This can also explain why some MoS₂ remains even after a relatively long‐time H₂ plasma treatment, as indicated by the Raman spectroscopy (Figure 1c, as previously mentioned). Currently, the growth of MoS₂ nanotubes is influenced by multiple factors. By optimizing the growth parameters of MoS₂, it is possible to achieve precise control over the wall number of MoS₂ nanotubes and further enable better regulation of Janus MoSSe nanotube synthesis. The elemental distributions of the sidewalls of Janus MoSSe nanotubes are also analyzed by EELS mapping (Figure S3, Supporting Information), but S and Se cannot be clearly distinguished from this (001) basal‐plane bird's‐eye view. This is because the side‐walled crystals are geometrically highly curved in this 1D structure, resulting in significant overlap of atoms at the sidewalls.

Cross‐section characterization is an efficient approach for determining the feature of the Janus structure.^{[ 10} , ²⁷ , ^{28 ]} In order to gain a deeper understanding of the substitution of S atoms by Se within the Janus MoSSe nanotubes, cross‐section samples of the heteronanotubes were prepared using focused ion beam (FIB), and the characterization results are presented in Figure 4 . The original nanotube film on the washer was first transferred onto a SiO₂/Si substrate, and it can be seen that the nanotubes are randomly distributed, as shown in Figure 4a. After being prepared into TEM cross‐section samples, various nanotube morphologies could be observed, as depicted in Figure 4b. When the randomly distributed nanotubes are perpendicular to the cutting plane, a circular cross‐section of the heteronanotubes becomes evident. Figure 4c illustrates the coaxial structure of SWCN‐BNNT, featuring a SWCNT as the innermost nanotube, wrapped by several layers of BNNTs. Additionally, the coaxial structure of SWCNT‐BNNT‐TMDCNT is clearly observed in Figure 4d,e. The contrast levels of the two outermost layers are markedly different, indicating the formation of TMDC nanotubes. Based on the S/Mo ratio, it can be concluded that the inner layer is still MoS₂, while the outer layer is Janus MoSSe. However, the MoSSe structure is only partially observed (likely due to the initially discontinuous formation of the MoS₂ tube or damage to the crystals during FIB preparation), as shown in Figure 4f,g. Meanwhile, when the FIB cutting plane is parallel to the heteronanotubes, an alternative perspective for observing the Se substitution is obtained, as depicted in Figure S4 (Supporting Information).

Figure 4.

a) Scanning electron microscope image of randomly distributed heteronanotubes. b) TEM images of the sample cross‐section. c) TEM images of the SWCNT‐BNNT coaxial structure. d,e) Cross‐section annular bright field (ABF)‐STEM and HAADF‐STEM images of SWCNT‐BNNT‐MoS₂‐MoSSe heteronanotubes. f,g) Cross‐section images showing the Janus MoSSe structure with S (yellow) inside and Se (orange) outside.

Finally, we compared the synthesis of 1D Janus MoSSe (Se outer) and MoSeS (S outer) nanotubes. In the case of 2D structures, selenizing monolayer MoS₂ or sulfurizing monolayer MoSe₂ yields identical planar Janus TMDC monolayers.^{[ 18} , ^{19 ]} However, the resulting structures of tubular Janus nanotubes differ significantly depending on whether selenization or sulfurization is performed. Given that only the outermost exposed chalcogen atoms are substituted, the presence of Se atoms on the outside of the nanotube (MoSSe) or S atoms on the exterior (MoSeS) can exert a considerable influence on the stability of the Janus nanotube structure. The stability of the Janus nanotubes can be measured by the strain energy per formula unit of the nanotube with respect to the monolayer. As shown in Figure 5a, the strain energy per atom of Janus nanotubes with S on the outer layer is higher than that with Se on the outer layer, indicating that Janus MoSSe nanotubes are energetically more favorable. Moreover, when the diameter exceeds approximately 22 Å, Janus MoSSe nanotubes exhibit lower strain energy compared to their corresponding Janus monolayer counterparts, as indicated by Figure 5a. The stability of Janus WSSe and Janus WSeS nanotubes was also evaluated by theoretical calculation and is shown in Figure S5 (Supporting Information). It is worth noting that while the SWCNT‐BNNT coaxial nanotube substrate plays a crucial role in the preparation of Janus nanotubes, the aforementioned theoretical calculations for Janus MoSSe and MoSeS did not account for substrate effects due to the computational complexity and the large unit cell size of the system. Furthermore, previous calculations also revealed that the strain energies of Janus MSSe (M = Mo, W) nanotubes are smaller than those of the corresponding Janus MSeS nanotubes, and the minimum strain energy is reached at a diameter of approximately 80 Å for Janus MoSSe nanotubes.^{[ 15} , ¹⁶ , ^{29 ]} In Figure 5b, we analyze the diameters of the Janus MoSSe nanotubes synthesized in this work. The diameters vary from 5 to 12 nm, with the majority falling in the range of 6 to 8 nm, which seems to be consistent with these previous predictions.

Figure 5.

a) Strain energies of Janus nanotubes as a function of tube radius. The filled blue squares and orange circles correspond to the calculated strain energies of MoSSe (Se outer) and MoSeS (S outer) nanotubes, respectively. The strain energies are measured with respect to monolayer Janus MoSSe and fitted using the equation E _strain (R) =  aR ⁻² + bR ⁻¹, where R is the tube radius.^{[ 30} , ^{31 ]} b) Histogram of the observed diameter of the Janus MoSSe nanotubes. c) Atomic model of the transformation from MoS₂ nanotube to Janus MoSSe nanotube, and from MoSe₂ nanotube to Janus MoSeS nanotube, along with corresponding HAADF‐STEM images of MoS₂, MoSSe, MoSe₂, and MoSeS nanotubes.

At present, Trivedi and Guo et al. have successfully synthesized 2D Janus TMDC monolayers at room temperature through diverse experimental strategies, specifically the selenization of MoS₂ monolayers and the sulfurization of MoSe₂ monolayers.^{[ 18} , ^{19 ]} In these 2D cases, both selenizing of MoS₂ and sulfurizing of MoSe₂ appear similarly successful. The sulfurization process is believed to be even easier and facilitate the synthesis of Janus monolayers.^{[ 18 ]} However, in the 1D experiments, the experimental results obtained from Janus nanotubes prepared through the selenization and sulfurization processes are entirely distinct. Although samples obtained through sulfurization also exhibited the characteristic Raman peak positions of the Janus, i.e., MoSeS, structure (Figure S6, Supporting Information), the conversion ratio is much lower than that of the other approach. Furthermore, the microscopic characterization results indicate that the Janus MoSSe nanotube has high crystallinity, whereas the Janus MoSeS structure is fragmented and unable to maintain the nanotube morphology, as shown in Figure 5c (upper and lower panels). These observations clearly solidify that, kinetically, the formation of the 1D MoSeS nanotube is slightly unfavorable, but energetically, the structure of the 1D MoSeS is significantly unstable. This is a distinct difference between 1D and 2D Janus crystals.

In contrast to conventional TMDC nanotubes such as MoS₂ nanotubes, the strain energy (compared to infinite monolayer) of Janus MoSSe nanotubes is lower than that of MoS₂ nanotubes.^{[ 15 ]} The negative strain energy of the Janus MoSSe nanotubes suggests that they are energetically stable relative to the flat Janus layer, while the strain energy of MoS₂ nanotubes is always positive, indicating that they are metastable compared to an infinite MoS₂ monolayer. However, MoS₂ nanotubes can be synthesized with high yield and scalability through various established methods,^{[ 32} , ³³ , ³⁴ , ³⁵ , ^{36 ]} the fabrication of Janus MoSSe nanotubes remains challenging due to their substrate‐dependent growth mechanisms and lower production efficiency. This discrepancy may originate from the distinct formation energies (defined as the difference between the energy of the layer and its atoms) of the two systems. The formation energy of Janus MoSSe monolayers is higher than that of MoS₂ counterparts,^{[ 15 ]} which imposes inherent thermodynamic challenges in the synthesis process. The elevated formation energy likely arises from the asymmetric atomic structure and interfacial strain in Janus systems, further complicating their scalable production under conventional conditions.

3. Conclusion

In summary, MoS₂ nanotubes grown on SWCNT‐BNNT templates were converted into Janus MoSSe structures at room temperature with the assistance of H₂ plasma. The formation of Janus MoSSe nanotubes was characterized by Raman spectroscopy. The microstructure and element distribution of SWCNT‐BNNT‐Janus MoSSeNT heteronanotubes were analyzed by TEM and EELS/EDS mapping results. The TEM results demonstrated the formation of monolayer and bilayer TMDC nanotubes on SWCNT‐BNNT heterostructures. EELS/EDS mapping and the S/Mo ratio for single‐ and double‐walled heteronanotubes indirectly demonstrate that only the exposed outermost S atoms in MoS₂ nanotubes are replaced by Se. From the cross‐sectional images, coaxial nanotube structures can be distinguished, and Se‐Mo‐S Janus structures are partially observed. Furthermore, it is advisable to prepare Janus MoSSe nanotubes derived from MoS₂ nanotubes, given the superior feasibility and stability of this preparation strategy. The successful fabrication of Janus MoSSe nanotubes opens new avenues for exploring their unique chirality‐dependent electronic and optical properties, paving the way for next‐generation nanoscale devices.

4. Experimental Section

Preparation of SWCNT‐BNNT Growth Template

The SWCNT film was first transferred onto a ceramic washer, and then the washer loaded with SWCNT was placed at the center of the CVD furnace. Ammonia borane (H₃NBH₃) was used as the BN precursor and positioned upstream of the SWCNT. During the synthesis process, the BN precursor was heated to 70–90 °C, while the SWCNT was heated to 1000–1100 °C. Ar (with 3% H₂) was used as the carrier gas, and the growth pressure was maintained at 300 Pa. The BNNT growth was maintained for 1 h. After the reaction, the furnace was allowed to cool naturally to room temperature, and the SWCNT‐BNNT sample was retrieved. This sample served as the substrate for the subsequent growth of MoS₂ or MoSe₂ nanotubes.

Synthesis of MoS₂ Nanotubes Based on SWCNT‐BNNT

MoS₂ nanotubes need to be prepared first for the fabrication of Janus MoSSe nanotubes. The MoS₂ nanotubes were synthesized via the LPCVD method, as shown in Figure 1a. MoO₃ (30 mg) precursor was placed in a quartz boat at the first heating zone. The SWCNT‐BNNT template was placed in another quartz boat at the second heating zone. The distance between MoO₃ and SWCNT‐BNNT was around 8 cm. Sulfur powder (6 g) was placed upstream and heated to 130 °C using a heating belt. The furnace (heating zones 1 and 2) was ramped up at a rate of 30 °C min⁻¹ to 530 °C and held for 50 min before naturally cooling to room temperature. The Ar flow was maintained at 50 sccm throughout the experiment.

Synthesis of Janus MoSSe Nanotubes Based on SWCNT‐BNNT

The as‐grown SWCNT‐BNNT‐MoS₂ sample was further used in the preparation of Janus MoSSe nanotubes. The schematic diagram of the related experiment is shown in Figure 1a, where SWCNT‐BNNT‐MoS₂ was placed in the center of the quartz tube, selenium (Se) particles were placed in the upstream region, and the downstream region was equipped with a home‐built ICP setup. At the beginning of the experiment, the system was evacuated to a base pressure of 2 Pa. Subsequently, 20 sccm H₂ was introduced as the carrier gas, maintaining the pressure around 27 Pa. The entire plasma treatment process lasted for 20 min. At the end of the experiment, H₂ flow was stopped, the system was refilled with Ar to atmospheric pressure, and then the sample was retrieved.

Synthesis of MoSe₂ Nanotubes Based on SWCNT‐BNNT

The synthesis of MoSe₂ nanotubes is different from that of MoS₂ nanotubes. A thick sapphire substrate was designed to enable the SWCNT‐BNNT film to be suspended on the MoO₂ precursor. The sapphire substrate, as shown in Figure S7a (Supporting Information), has a counterbore in the center area, designed to accommodate the ceramic washer bearing the SWCNT‐BNNT film. Figure S7b (Supporting Information) shows the schematic of the MoSe₂ nanotube growth. The Se particles are placed upstream of the molybdenum precursors and heated to 220 °C by a heating belt. This temperature ensures sufficient evaporation of Se during the growth process. Additionally, the MoO₂ and sapphire substrate are heated to 630 °C by the CVD furnace. A mixture of Ar and H₂ is used as the carrier gas at a flow rate of 50 sccm (Ar: 45 sccm; H₂: 5 sccm). Throughout the experiment, the growth pressure is maintained at approximately 50 Pa.

Synthesis of Janus MoSeS Nanotubes Based on SWCNT‐BNNT

The preparation process of Janus MoSeS nanotubes was similar to that of Janus MoSSe nanotubes. The SWCNT‐BNNT‐MoSe₂ heteronanotube sample and S powder were placed in the low‐pressure chamber, and the experimental parameters, including the carrier gas, flow rate, and treatment time, were kept consistent with those used in the synthesis of Janus MoSSe nanotubes.

SWCNT‐BNNT‐MoSSe Cross‐Section Specimen Preparation

The ultra‐thin STEM specimen of heteronanotube cross‐section was prepared by combining a dual beam FIB precision manufacturing instrument (FEI Scios 2 Dual Beam) and Nano Mill system (Fischione 1040). Before the nanotube specimen was lifted out, a 2.0 µm thick carbon layer (0.5 µm e‐beam deposition and 1.5 µm ion‐beam deposition) was deposited on the nanotube surface to avoid Ga ion beam damage. After lift‐out, the specimen was thinned to approximately 200–300 nm with a 1.0 µm carbon layer remaining by using 16–5 kV Ga ion beam with different currents. After the milling procedure in FIB, the lamella of nanotube cross‐section was further milled by low‐energy argon ions in the Nano Mill system, to remove the damaged surface caused by the FIB polishing and to optimize the specimen thickness to below 50 nm. The focused argon ion beam ranged from 500 to 200 eV with a beam current of 180 µA. Once the cross‐section specimen preparation process by Nano Mill was finished, the samples were immediately stored in a glove box to protect them from prolonged exposure to air.

Characterizations

Raman spectra of the samples were acquired in a backscattering configuration using a Raman spectrometer (Horiba, LabRAM Odyssey) with 532 nm excitation. A 100× microscope objective was employed to focus the laser beam and collect the scattered light. Absorption spectra were measured using a UV–vis–NIR spectrophotometer (Shimadzu, UV‐3600i Plus). All measurements were carried out at room temperature. High‐resolution TEM images were acquired using a JEOL JEM F200 operated at an electron accelerating voltage of 200 kV. HAADF‐STEM imaging combined with EELS/EDS chemical mapping was performed using a JEOL JEM‐2100F microscope equipped with double JEOL Delta correctors, a Gatan Quantum electron spectrometer, and double JEOL Centurio detectors at an accelerating voltage of 60 kV. The atomic‐scale characterization of the heteronanotube cross‐section with Janus structure was conducted in the spherical aberration‐corrected STEM (FEI Titan G2 80–200 ChemiSTEM), which is equipped with four Bruker energy‐dispersive X‐ray spectrometers (super‐EDS) for element analysis. For HAADF‐STEM imaging, the inner and outer collection angles of the annular dark‐field detector were set at 55 and 220 mrad, respectively.

Density Functional Theory Calculations

The calculations of the geometric structure of Janus nanotubes were based on density functional theory,^{[ 37} , ^{38 ]} employing the simulation tool for atom technology (STATE) program package.^{[ 39 ]} The exchange‐correlation potential between interacting electrons was expressed using the generalized gradient approximation (GGA) with the Perdew–Burke–Ernzerhof (PBE) functional.^{[ 40 ]} Ultrasoft pseudopotentials generated by the Vanderbilt scheme were employed as the interaction between nuclei and electrons.^{[ 41 ]} Valence wave functions and the deficit charge density were expanded in terms of plane wave basis sets with cutoff energies of 25 and 225 Ry, respectively. The atomic structures of the Janus nanotube were refined until the force exerted on each atom was minimized to below 1.33 × 10⁻³ Hartree/bohr. For the lattice parameter aligned with the tube axis, a cell parameter of c = 0.3191 nm was adopted, derived from the optimum lattice parameter of an isolated MoSSe monolayer. The Brillouin zone integration was carried out with the use of 5 k points along the tube axis.

Conflict of Interest

The authors declare no conflict of interest.

Supporting information

Supporting Information

Yang C., Lin Q., Sato Y., Gao Y., Zheng Y., Wang T., Ma Y., Dai W., Li W., Maruyama M., Okada S., Suenaga K., Maruyama S., Xiang R., Janus MoSSe Nanotubes on 1D SWCNT‐BNNT van der Waals Heterostructure. Small 2025, 21, 2412454. 10.1002/smll.202412454

Data Availability Statement

The data that support the findings of this study are available in the supplementary material of this article.