CSPM EDP Sciences logo
Submit your paper
Search
Menu
All issues Volume 4 / No 2 (2025) Natl Sci Open, 4 2 (2025) 20240043 Full HTML
Open Access
Issue
Natl Sci Open
Volume 4, Number 2, 2025
Article Number 20240043
Number of page(s) 14
Section Materials Science
DOI https://doi.org/10.1360/nso/20240043
Published online 09 January 2025

© The Author(s) 2025. Published by Science Press and EDP Sciences.

Licence Creative CommonsThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

INTRODUCTION

The Raman enhancement effect has attracted increasing attention in both fundamental studies of light-matter interaction and microanalytical applications toward trace amounts of substances with chemical specificity [1,2]. As a prototypical Raman enhancement technique, surface-enhanced Raman scattering enables the ultrasensitive detection of a single molecule [2], and reveals great promise for practical detection due to its highly efficient, non-invasive, and versatile nature [3]. Generally, the electromagnetic mechanism plays a dominant role in metallic surface-enhanced Raman spectroscopy (SERS) substrates with an enhancement factor of 106–1010 [4,5]. In this process, SERS substrates primarily exploit surface plasmons induced by the incident electromagnetic field in metallic nanoparticles or roughened surfaces to significantly increase the Raman intensity [6]. The implementation of the SERS effect is largely limited by the cost of SERS-active substrates, which involves the use of noble metals [5,7] and the fabrication of delicate nanostructure to enhance the electromagnetic ‘hot-spots’ and highly concentrated charge density. Additionally, most SERS-active systems studied today are based on random nanostructures, whose properties vary from experiment to experiment making quantitative correlation between theory and experiment difficult.

Compared with noble metals, two-dimensional (2D) materials possess several intrinsic advantages involving ultra-large specific surface areas [8], structural tailorability in atomic precision [9], ease of assembly into desirable heterostructure [10], excellent biocompatibility [11,12] and flexibility [13,14], and others, rendering them promising candidates for superior SERS substrates. So far, significant progress has been made regarding 2D materials-based SERS, like graphene [15], transitional metal dichalcogenide (TMDC) semiconductors [16], as well as their heterostructures [17]. Due to the abundant density of states (DOS) at the Fermi level, the recently emerging metallic 2D materials [18,19] offer another avenue to circumvent the match requirement of the frontier orbitals of the molecules and valance/conduction band energies in many widely researched 2D semiconductors [20,21]. Considering the rich combination of elemental components in a 2D plane [22], such a material family provides a versatile platform to develop inexpensive SERS substrates free from complicated structure engineering [23,24] like surface roughing, decoration of nanoparticle, fabrication of delicate heterostructure, which shall provide an efficient solution to the long-standing reproducibility issue in conventional SERS substrates [4,25,26].

The quality of crystallinity in 2D materials has been proven to be a pivotal parameter not only determining the specific light-matter interaction [27], but also laying a firm foundation for their photonic applications [28]. Inspired by our previous research that the absence of oxygen in the 2D material synthesis procedure plays a crucial role in determining its post chemical stability [29], we deem that such controllability shall be a viable alternative in developing 2D SERS substrate with high stability, which is highly desirable in currently reported 2D materials including 1T′-WTe2 [30], NbS2 [31], NbSe2 [32], and TaSe2 [33] that generally suffer from metastability in phase transition or oxidation induced degradation [29,34]. For instance, oxygen-induced vacancies have been widely attested to exist in numerous 2D planes [35,36], triggering crystal degradation within a few minutes after exposure [18,37]. Although the vacancy-induced stability issue can be partially alleviated via surface capping with a molecular or film-based protective layer, this also induces several practical concerns, particularly in surface detection like SERS.

Herein, we present an oxygen-free vapor continuous deposition-growth design, proposing an industry-compatible technology for the batch growth of wafer scale nanocrystalline NbSe2 films with high stability for SERS application. In stark contrast to the instability of conventional metallic nanomaterials, the synthesized NC-NbSe2 film reveals remarkable chemical stability both in the crystalline domains (with an average lateral size of ~8.1 nm) and grain boundaries. Notably, we found that the grain boundaries (GBs) in NC-NbSe2 play an important role in the enhanced SERS effect toward molecular sensing. The abundance of density states around the Fermi level, coupled with GBs-enhanced surface adsorption in NC-NbSe2 film, contribute to its superior SERS performance. As a proof of concept, Rhodamine 6G (R6G) molecules were detectable, with an enhancement factor higher than that of conventional planar substrates, achieving sensitivity down to 1×10−10 M [38,39], and even comparable to noble metal-based SERS substrates [4,23]. Moreover, the NC-NbSe2 films show negligible SERS performance degradation even after exposure to the atmosphere and elevated temperature, due to the low density of degradable vacancy configuration. This work provides new insights into the development of stable 2D SERS substrates, the controllability, scalability, and direct batch growth of NC-NbSe2 film on commercial SiO2/Si substrate would be favorable for its practical application in photonic devices.

RESULTS AND DISCUSSION

Batch growth of wafer-scale NC-NbSe2 films

A schematic illustration of our batch growth method is shown in Figure 1A. The two-step vapor deposition system involves a combined magnetron sputtering and a post-selenylation chamber, enabling the deep degassing and ultra-low oxygen concentration throughout the growth procedure due to the utilization of ultrapure Nb and Se elements as growth precursors. Notably, SiO2/Si was selected in this work due to the following three reasons: (1) the amorphous nature of the SiO2 surface favors the surface adsorption and nucleation of the deposited film [27,40,41], which is favorable for the facile synthesis of NbSe2 film with lower growth temperature; (2) commercially available SiO2/Si wafer shows excellent compatibility with the current semi-conductor industry and device integration techniques, thus revealing versatility in future applications; (3) the direct growth of SERS-active substrate film on SiO2/Si shall eliminate post-transfer procedure, drastically avoiding the adverse impact of polymer residuals towards the surface adsorption of target molecules.

thumbnail Figure 1

The batch growth of wafer-scale NC-NbSe2 films. (A) Schematic illustration of the two-step vapor deposition growth that involves a magnetron sputtering of Nb film and a subsequential selenylation into NC-NbSe2 film in a continuous preparation route. (B, C) Photographs and (D) optical microscopic image of as-grown NC-NbSe2 film on 4-inch SiO2/Si substrates. Notably, the SiO2 substrate was artificially exposed for visual guidance. (E) The representative Raman spectra of NC-NbSe2 film. The inset shows the characteristic Raman stretching modes of A1g and E2g1 of NbSe2. (F) AFM image and the corresponding height profile of NC-NbSe2 films directly grown on SiO2/Si. (G) Atomic structure model and (H) the ADF-STEM image of bilayer NC-NbSe2 film.

Nanocrystalline Nb films with homogeneous surface roughness (±0.15 nm) are firstly sputtered on commercial 4-inch SiO2/Si substrates at room temperature (Figure S1). Subsequently, the Nb films were selenized into continuous NC-NbSe2 film at a mild growth temperature of 550 °C, the lowest value compared with those reported in other NbSe2 films [42,43]. The amorphous surface of SiO2/Si can effectively increase the GB density of the Nb films, as proved by the transport measurements in Figure S2. Figure 1B and C show the as-grown NC-NbSe2 films on 4-inch SiO2/Si wafers, which are highly reproducible from batch to batch. As revealed in the representative optical microscopic image in Figure 1D, NC-NbSe2 shows uniform optical contrast, as well as consistent Raman characteristic peaks located at ~226 and ~247 cm−1, which were assigned to A1g (corresponding to an out-of-plane mode) and E2g1 (corresponding to an in-plane mode) characteristic modes, respectively [44]. The peak spacing between A1g and E2g1 is thickness-dependent, and the peak spacing of 21 cm−1 over the entire film implies the NC-NbSe2 is composed of two NbSe2 atomic layers vertically stacked on SiO2/Si [45]. Atomic force microscopy (AFM) images show that the thickness of NC-NbSe2 films is 2.0 nm (Figure 1F), further attesting the layer number of NC-NbSe2 film is 2, which is consistent with our previous result on sapphire [29] and other works [43,46]. The high-angle annular dark-field scanning transmission electron microscopy (ADF-STEM) image also shows the bilayer number of NC-NbSe2 film at atomic resolution (Figure 1G and H). The Raman spectra of the NC-NbSe2 film with perpendicular polarization are shown in Figure S3. The peak around ~20 cm−1 can be attributed to the shear mode of bilayer NbSe2 [47]. As a comparison, crystalline NbSe2 (C-NbSe2) films were grown on sapphire substrate under similar preparation conditions, as shown in Figure S4. Notably, the Raman characteristic peaks of the NbSe2 films are far away from the Raman-active region of most common dye analytes (500–1800 cm−1), implying its favorable application as a SERS substrate.

Atomic structure characterization of as-grown nanocrystalline NbSe2 film

The as-grown NC-NbSe2 films are further characterized at the atomic level by the transmission electron microscope (TEM) and the high-resolution transmission electron microscope (HR-TEM). With incredible stability, the morphology and atomic structure of NbSe2 films can be well maintained even after conventional etchant-involved wet transfer procedure. Compared to the C-NbSe2 grown on sapphire (Figure 2A–C), the amorphous nature of the SiO2 surface, combined with the relatively mild growth temperature, facilitates the dynamic crystallization of NbSe2 nano-domains into a continuous film (Figure 2D). Owing to the deep degassing advantage in the combined deposition-growth system, the as-grown NC-NbSe2 film reveals a defined stoichiometric ratio of Nb to Se in the energy dispersive spectrum (EDS) in Figure 2E. The oxygen-free lattice is robust against post-exposure to the atmosphere, due to the high energy barrier in the production of reactive oxygen species on crystalline TMDC surface [36]. The diffraction ring of selected area electron diffraction (SAED) (Figure 2F) among a 30 nm×30 nm region attests to the nanocrystallinity in NC-NbSe2 film. To quantitively clarify the crystallinity and lattice orientation in NC-NbSe2, we conducted HR-TEM statistics in the as-grown sample (Figure 2G). As shown in Figure 2H, the false color highlights the nano-domain in NC-NbSe2 film with an average lateral size of ~8.1 nm. Notably, the grain boundaries between two neighboring nano-domains are atomically distinguishable as shown in Figure 2I, indicating its chemical stability and high quality. This is quite different from the conventional recognition that domain boundaries in 2D materials are generally nonstable in the air [48], which further highlights the crucial role of oxygen-elimination in the growth of stable 2D materials.

thumbnail Figure 2

The nanocrystalline domains in NC-NbSe2 film. (A, B) Large-scale and (C) zoom-in high-resolution TEM images, and (inset in (C)) Fourier transform pattern of crystalline NbSe2 (C-NbSe2) film grown on a sapphire substrate as comparison. (D) A large-scale TEM image showing the uniformity of as-grown NC-NbSe2 film on SiO2/Si substrate. (E) The EDS spectrum and (F) SAED pattern of as-grown NC-NbSe2 film. Notably, the Cu Kα signal in panel (E) sources from the TEM grid. (G) A representative high-resolution TEM image with (H) false color and crystalline orientation superimposed to highlight nano-domain in NC-NbSe2 film. (I) A representative grain boundary between nano-domain in NC-NbSe2 film.

SERS effect on nanocrystalline NbSe2 film

Figure 3A shows a schematic diagram of the as-grown NC-NbSe2 films used as the SERS substrates for Raman signal enhancing of analytes. As a proof-of-concept demonstration, a prototypical R6G dye dissolved in ethanol was adopted as an example to evaluate the SERS performance of the as-grown NC-NbSe2. Due to the position of the R6G absorption peak, the 532 nm wavelength laser is selected to yield a high SERS intensity. In a general procedure, 10 μL ethanol solution containing different concentrations of R6G was dropcast onto the as-grown samples and dried for Raman measurement. The intensity comparison of characteristic peaks of R6G on different substrates is shown in Figure 3B. The labeled Raman signatures of R6G on NC-NbSe2 substrate are distinguishable at 1183, 1309, 1361, 1505, 1570, and 1646 cm−1, in good agreement with those reported in previous works [27,30], which evidences that the NC-NbSe2 could detect trace amounts of R6G with better sensitivity compared to C-NbSe2 and bare SiO2 substrates.

thumbnail Figure 3

The SERS activity of NC-NbSe2 film. (A) Schematic diagram of the NC-NbSe2 films used for SERS activity towards R6G molecules. (B) Raman spectra of R6G coated on SiO2/Si, C-NbSe2 films, NC-1-NbSe2 film, and NC-NbSe2 film, respectively, correspond to 10 μL R6G solution (10−6 M) on substrates. Notably, the Raman intensities are normalized to the integration times. (C) The comparison of characteristic vibrational Raman peaks of R6G on NC-NbSe2 and C-NbSe2 film. Notably, the overall intensity was normalized to the intensity of peak at 1361 cm−1 on C-NbSe2 film. (D) The ADF-STEM images of NC-NbSe2 film. Notably, the amorphous domain (highlighted in purple) implies the GBs-enhanced surface adsorption of organic species on NC-NbSe2 film. (E) The corresponding SAED pattern of the bottom two domains in NC-NbSe2 film. (F) Diagram of energy level diagram and charge transfer transitions between R6G and NbSe2 film. Notably, μi-CT and μk-CT denote the charge transfers from the HOMO of the R6G molecules to the Fermi level of NbSe2, and from the Fermi level of NbSe2 to the LUMO of the R6G molecules, respectively. The DOS near the Fermi level is based on the results of previous studies [54,55].

Compared with the reported data, no obvious shift is observed, indicating that target molecules are adsorbed as a single molecule on the SERS substrates [27,49]. We selected NC-NbSe2 (grain size ~8.1  nm), NC-1-NbSe2 (grain size ~50  nm), and C-NbSe2 (grain size >1  μm) to compare the SERS effect of NbSe2 with different grain sizes. As shown in Figure 3B and C, the SERS effect enhances as the grain size of NC-NbSe2 decreases. Specifically, the Raman intensity at 1361  cm−1 of R6G coated on NC-NbSe2 is 1.9 and 3.2 times higher than that of R6G coated on NC-1-NbSe2 and C-NbSe2, respectively. To our knowledge, this level of enhancement is significantly greater than the grain domain effect observed in graphene-based materials, which is generally less than a factor of 2 when compared with crystalline graphene [27,38,39]. The higher SERS effect of the as-grown nanocrystalline NbSe2 films proves the huge potential of metallic 2D TMDCs as SERS substrates. In these two cases, the GB density of NbSe2 films is the only experimental variable. It should be attributed to the higher GB density and higher adsorption energy of NC-NbSe2, which results in more R6G molecules being adsorbed effectively. GBs-enhance surface adsorption has been observed and interpreted in several previous works [50–52], increased DOS in grain boundaries, particularly those pentagon structures and distorted lattice configurations, serve as charge-poor for an enhanced surface adsorption of the positively charged dye molecules. The GBs-enhanced surface adsorption effect in NC-NbSe2 was also proved at the atomic scale as shown in Figure 3D. Considering the non-stability of dye molecules under electron beam irradiation, we adopted polymethyl methacrylate (PMMA) polymers as indicators [51,52], and observed by ADF-STEM. As shown in Figure 3D, the GBs are distinguishable by both atomic orientation in the crystalline domain and the Fourier transform pattern in Figure 3E, due to the relatively light element nature in polymer, its preferable adsorption upon domain boundary is distinguishable and highlighted in Figure 3D, with an average width around 2.8 nm, indicating the strong adsorption of the atomically sharp domain boundary. Additionally, according to the electrical transport measurement (Figure S5A and B) [53], corroborated with the negligible influence of GBs on the DOS near the Fermi level [54,55], we shall conclude that the large DOS of NC-NbSe2 films combined with the GBs-enhanced surface adsorption collectively facilitate the charge transfer between dye molecules and NC-NbSe2 substrates, leading to superior SERS effect (Figure 3F).

The Raman enhancement of R6G on NC-NbSe2 can be attributed to the chemical mechanism (CM), which originates from charge transfer between the target molecules and the substrate. CM is a short-range effect occurring at the molecular scale, where the distance between the substrate and the target molecules is crucial. N atoms in R6G with positive charge tend to be adsorbed on seven-membered rings. Through theoretical analysis, we found that near the GBs, the charge density of the heptagonal rings increases significantly, which is beneficial for adsorbing positively charged R6G molecules, as shown in Figure S6. The adsorption energy of the heptagonal rings for R6G is −2.4 eV, which proves that it does have a strong adsorption effect on R6G (Figure S7). For the case of NC-NbSe2, the high density of GBs can uniformly anchor more positively charged R6G molecules, providing a prerequisite for collecting enhanced Raman signals.

After this effective adsorption, a photon-induced charge transfer (PICT) process occurs. The PICT process includes charge transfer from the highest occupied molecular orbital (HOMO) energy level of R6G to the Fermi level of NbSe2, and from the Fermi level of NbSe2 to the lowest unoccupied molecular orbital (LUMO) energy level of R6G (Figure 3F). This molecule-metal coupling can be described using the vibronic Herzberg-Teller theory, where the polarizability tensor of the probe is greatly amplified in the coupled system [56]. The large number of electrons at the metallic NbSe2 Fermi level can increase the number of allowed energy states for PICT in the dye molecule-NbSe2 coupled complex, thereby enhancing the probability of charge transitions. These factors may collectively lead to the high Raman enhancement efficiency of NC-NbSe2.

We also investigated the SERS effect of NC-NbSe2 for different dyes, including indocyanine green (ICG), methylene blue (MB), 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI), all at a concentration of 10−8  M, and found that NC-NbSe2 exhibits a very good SERS effect for all of them (Figure S8).

The comprehensive SERS performance evaluation of NC-NbSe2 film

Figure 4A shows the Raman spectra of R6G/NC-NbSe2 at various concentrations of R6G. The Raman intensity at 1363 cm−1 as a function of R6G concentration is shown in Figure S9. The labeled R6G Raman signatures located at 1362, 1570, and 1646 cm−1 assigned to aromatic C–C stretching are detectable [57]. The detection limit down to 1×10−10 M is comparable with or even beyond those of noble-metal-based SERS substrates (~10−9 M) [4,23]. Notably, this result is also superior to the previously reported results based on metallic TMDCs [33].

thumbnail Figure 4

The comprehensive SERS performance evaluation of NC-NbSe2 film. (A) Stack plots of Raman spectra for R6G coated onNC-NbSe2 films with various R6G concentrations ranging from 1×10−5 to 1×10−10 M. (B) The time-dependent evolution of Raman spectra of R6G coated on NC-NbSe2 films exposure to air at 50 °C. (C) The time-dependent Raman spectra of R6G coated on NC-NbSe2 films that were pre-immerged in water for 30 min. (D) The robustness demonstration of NC-NbSe2 film with pre-immersion in water and subsequent reliable SERS signal towards RhB (4×10−6 M) detection.

Stability is another key concern regarding SERS for practical application, particularly for 2D materials that generally suffer from intrinsic metaphase transition or oxidation in the atmosphere [29,34]. Then, we mainly focus on the stability evaluation of the SERS effect in NC-NbSe2 via pretreatment in various chemical environments. In order to make a comparison with previous works, the Raman spectra of R6G coated on NC-NbSe2 films after exposure to air at room temperature for 30 d were first tested. As shown in Figure S10, the Raman intensity of NbSe2 and R6G remains unchanged. In contrast, the electrochemically exfoliated NbSe2 shows visible structure defects after exposure to air for 7 d due to its intrinsic defects sourced from bulk crystal (Figure S11). To identify the origin of the instability observed in electrochemically exfoliated NbSe2 flakes, we conducted atomic-resolution characterization of the exfoliated flakes. As shown in the ADF-STEM image in Figure S12A, within an area of 10  nm×10  nm, there are more than 13 Se(Nb) atomic vacancies. These vacancies may originate from defects in bulk NbSe2 or be caused by the environmental sensitivity of the exfoliated NbSe2 flakes. Theoretical results indicate that these atomic vacancies may be the origin of the instability observed in electrochemically exfoliated NbSe2 flakes (Figures S12–S14). As shown in Figure S13, H2O and O2 can not be adsorbed stably on defect-free monolayer NbSe2 with very low adsorption energy (−0.09 and −0.06 eV). While on the single-vacancy surface, H2O and O2 have very strong adsorption energies of −0.84 and −2.03 eV, respectively, which leads to a decrease in the stability of NbSe2 in solution (Figure S14A–D). Especially, in the site of dual defect NbSe2, the adsorption energy of H2O is higher (−0.96 eV) (Figure S14E and F). O2 will directly dissociate and be embedded in NbSe2 (Figure S14G and H). Therefore, more defects make oxygen atoms with stronger electronegativity embedded in the lattice, making the structure unstable after electrochemical exfoliation.

More interestingly, the NbSe2 films show a robust SERS effect even after heating at 50 °C for 10 h (Figure 4B, Figure S15A) without measurable degradation in Raman intensity, this is substantially different from conventional physical adsorption cases that tend to be very sensitive to elevated temperature due to weak adsorption and gradual desorption [31,32]. Given oxidation triggered by H2O or O2 represents one of the main reasons for the instability of 2D metallic TMDCs-based SERS substrates [36], additional harsh treatments were conducted to verify the stability of NC-NbSe2 films. Before dropping the dye solution onto it, the NC-NbSe2 substrate was pre-immersed in deionized water for 30 min (Figure 4C, Figure S15B). Then, the Raman intensity of R6G was detected against time, which was almost identical after exposure to air at 25 °C for 2 d. When the exposure time was extended to 7 d, only a slight reduction (~10% decrease of the Raman intensity) was observed. The stable SERS effect of NC-NbSe2 films was also applicable to other dyes such as Rhodamine B (RhB), as shown in Figure 4D. The SERS performance of representative 2D SERS substrates is shown in Table S1. In our case, the SERS performance of NC-NbSe2 is comparable to the recently reported 2D TMDC SERS substrates. Interestingly, our as-grown NC-NbSe2 films show a stable SERS effect even under harsh chemical conditions. For example, the as-grown NC-NbSe2 shows a stable SERS effect in air for 30 d and even shows a stable SERS effect for 7 d after treatment in water for 30 min (Table S1). The above tests unambiguously confirm the excellent stable SERS effect of NC-NbSe2 films grown on SiO2/Si, which holds significant implications for the future applications of molecular sensing. To further justify the benefit of ‘Batch growth’ and confirm the homogenous thickness of our sample, we supplement 10 additional Raman spectra, which are randomly distributed at the R6G coated as-grown 4-inch NC-NbSe2 film (Figure S16). These results clearly show that Raman intensity is similar at different locations, confirming the homogeneous SERS performance of our batch-grown films, which is promising from lab to practical application.

CONCLUSIONS

In this study, we develop an industry-compatible technology for the batch growth of wafer-scale NC-NbSe2 films on SiO2/Si substrates, exhibiting an exceptionally stable SERS effect. We found that the GBs of NbSe2 can effectively enhance the SERS effect toward molecular sensing. When the density of GBs in NbSe2 films increased (with an average lateral grain size of ~8.1 nm), the DOS near the Fermi level of NbSe2 remained unaffected. However, this increase in GB density resulted in a greater number of adsorption sites on the NbSe2 films and higher adsorption energy for target molecules. Crucially, the absence of oxygen-associated vacancies in the NbSe2 films ensures a stable SERS effect for molecular sensing, even after exposure to harsh conditions such as heating or immersion in solutions. Given their sufficient sensitivity for technical detection limits and their incredibly stable SERS effect, the batch-grown NC-NbSe2 film-based SERS substrates emerge as prime candidates to address the critical challenges of industrial applications for next-generation SERS substrates.

METHODS

Batch growth of nanocrystalline NbSe2 films on silicon (SiO2/Si)

The two-step deposition method was used to grow NbSe2 films. The growth process is as follows: first, nanocrystalline Nb (>99.9%) films are sputtered on four pieces of 4-inch Si/SiO2 wafers within 5 min at room temperature (25 °C); second, Nb films are transferred to a growth chamber (base pressure ​<10−5 Pa) within few minutes to avoid the oxidation and heated to 550 °C for 30 min. High-purity Se (>99.9%) was evaporated from a standard Knudsen cell Se source. A mixture gas of H2/Ar (1:10) was used as a carrier gas to ensure uniform selenization of Nb films. The growth time of a homogenous film is 8 min. Note: four pieces of 4-inch nanocrystalline NbSe2 films can be prepared in batches with continuous sputtering-growth time within 15 min. The grown films are transferred by the traditional wet method [29] to different substrates or TEM grids for other characterizations, noted that the sample is annealed at 350 °C under high vacuum (~10−5 Pa) for 2 h to remove possible surface adsorbates and contaminants.

SERS measurements

Ethanol solutions with different concentrations of dye molecules (R6G, RhB, etc.) were obtained through a continuous dilution process. For each SERS measurement, 10 μL probe solution is dropped on NbSe2 and dried in air. The substrate is then washed with ethanol three times to remove the free molecules. All the Raman measurements are collected by a Witec/alpha 300 R confocal microscope system with a laser spot size of ~1 μm. The excitation wavelength is 532 nm, and the laser power is set at 1 mW unless specified. The SERS spectra are obtained with 10 s acquisition time unless specified. The spectra for comparison are acquired under identical measuring conditions.

Characterizations

TEM and ADF-STEM images are captured by FEI Titan ChemiSTEM probe aberration corrected scanning TEM with a monochromator, which is operated at 80 kV to minimize the knock-on damage. Raman spectra are performed using a Witec/alpha 300 R confocal microscope with a 532 nm laser at ambient conditions. AFM images are taken using the Bruker AXS Dimension Icon in tapping mode. The superconducting and transport properties are carried out in 4He cryostat with a superconducting magnet (Oxford Teslatron 8T), and electrical transport measurements are performed using a standard lock-in amplifier (Stanford SR830) with currents of 1–10 μA. Four-probe contacts are used by an e-beam evaporated 80 nm Cr/Au array with 500 μm dots array spacing.

Density functional theory (DFT) calculations

The projector augmented wave (PAW) potential and generalized gradient approximation with Perdew-Burke-Ernzerh (GGA-PBE) functional were used to describe the exchange-correlation interactions. The van der Waals (vdW) correction DFT-D3 proposed by Grimme was chosen due to the good description of long-range vdW interactions. Moreover, the plane wave cutoff energy was set to 520 eV, and 1×1×1 k-point was applied to the sample. The 20 Å vacuum space was used to avoid the layers’ interactions. The electronic energy was considered self-consistent when the energy change was smaller than 10−5 eV. Geometry optimization was considered convergent when the energy was smaller than 0.02 eV/Å. The adsorption energy (Eads) between the substrate and adsorbate (X) on the support is defined as

E ads = E substrate + X − E substrate − E X .

Acknowledgments

We sincerely thank Y.M. for the valuable assistance with the theoretical analysis.

Funding

This work was supported by the National Natural Science Foundation of China (52402044, 22304087, 22272004), the Major Research Project of the Natural Science Foundation of the Jiangsu Higher Education Institutions (23KJA150006), the Natural Science Foundation of Jiangsu Province (BK20230349), the High-Level Talent Introduction Project of Nanjing University of Posts and Telecommunications (NY222110), the Fundamental Research Funds for the Central Universities (YWF-22-L-1256), the Shenzhen Science and Technology Program (KQTD20221101115627004, JCYJ20240813160206009), and the Guangdong Provincial Key Laboratory of Nano-Micro Materials Research.

Author contributions

H.L., L.W., Z.C. and J.L. conceived and supervised the project, and designed the experiments. H.L. organized the collaborations. H.L., Y.W.L. and Y.X.L. carried out growth experiments and SERS measurements. L.W. performed STEM. H.L. and M.X.L. carried out the theoretical analysis. H.L. wrote the manuscript with input from L.W., Z.C. and J.L. All authors discussed the research and revised the paper.

Conflict of interest

The authors declare no conflict of interest.

Supplementary information

Supplementary file provided by the authors. Access here

The supporting information is available online at https://doi.org/10.1360/nso/20240043. The supporting materials are published as submitted, without typesetting or editing. The responsibility for scientific accuracy and content remains entirely with the authors.

References

  • Moskovits M. Surface-enhanced spectroscopy. Rev Mod Phys 1985; 57: 783-826. [Article] [Google Scholar]
  • Nie S, Emory SR. Probing single molecules and single nanoparticles by surface-enhanced Raman scattering. Science 1997; 275: 1102-1106. [Article] [CrossRef] [Google Scholar]
  • Camden JP, Dieringer JA, Wang Y, et al. Probing the structure of single-molecule surface-enhanced Raman scattering hot spots. J Am Chem Soc 2008; 130: 12616-12617. [Article] [Google Scholar]
  • Li JF, Huang YF, Ding Y, et al. Shell-isolated nanoparticle-enhanced Raman spectroscopy. Nature 2010; 464: 392-395. [Article] [Google Scholar]
  • Matricardi C, Hanske C, Garcia-Pomar JL, et al. Gold nanoparticle plasmonic superlattices as surface-enhanced Raman spectroscopy substrates. ACS Nano 2018; 12: 8531-8539. [Article] [Google Scholar]
  • Schlücker S. Surface-enhanced Raman spectroscopy: Concepts and chemical applications. Angew Chem Int Ed 2014; 53: 4756-4795. [Article] [Google Scholar]
  • Fleischmann M, Hendra PJ, McQuillan AJ. Raman spectra of pyridine adsorbed at a silver electrode. Chem Phys Lett 1974; 26: 163-166. [Article] [NASA ADS] [CrossRef] [Google Scholar]
  • Geim AK, Novoselov KS. The rise of graphene. Nat Mater 2007; 6: 183-191. [Article] [CrossRef] [Google Scholar]
  • Wang H, Yuan H, Sae Hong S, et al. Physical and chemical tuning of two-dimensional transition metal dichalcogenides. Chem Soc Rev 2015; 44: 2664-2680. [Article] [CrossRef] [PubMed] [Google Scholar]
  • Novoselov KS, Mishchenko A, Carvalho A, et al. 2D materials and van der Waals heterostructures. Science 2016; 353: aac9439. [Article] [Google Scholar]
  • Chimene D, Alge DL, Gaharwar AK. Two-dimensional nanomaterials for biomedical applications: Emerging trends and future prospects. Adv Mater 2015; 27: 7261-7284. [Article] [NASA ADS] [PubMed] [Google Scholar]
  • Wu L, Lin H, Cao X, et al. Bioorthogonal Cu single-atom nanozyme for synergistic nanocatalytic therapy, photothermal therapy, cuproptosis and immunotherapy. Angew Chem Int Ed 2024; 63: e202405937. [Article] [Google Scholar]
  • Akinwande D, Petrone N, Hone J. Two-dimensional flexible nanoelectronics. Nat Commun 2014; 5: 5678. [Article] [Google Scholar]
  • Ren H, Xia X, Sun Y, et al. Electrolyte engineering for the mass exfoliation of graphene oxide across wide oxidation degrees. J Mater Chem A 2024; 12: 23416-23424. [Article] [Google Scholar]
  • Xu W, Mao N, Zhang J. Graphene: A platform for surface-enhanced Raman spectroscopy. Small 2013; 9: 1206-1224. [Article] [CrossRef] [PubMed] [Google Scholar]
  • Ling X, Fang W, Lee YH, et al. Raman enhancement effect on two-dimensional layered materials: Graphene, h-BN and MoS2. Nano Lett 2014; 14: 3033-3040. [Article] [Google Scholar]
  • Ling X, Huang S, Deng S, et al. Lighting up the Raman signal of molecules in the vicinity of graphene related materials. Acc Chem Res 2015; 48: 1862-1870. [Article] [Google Scholar]
  • Li J, Song P, Zhao J, et al. Printable two-dimensional superconducting monolayers. Nat Mater 2021; 20: 181-187. [Article] [Google Scholar]
  • Zhao B, Shen D, Zhang Z, et al. 2D metallic transition-metal dichalcogenides: Structures, synthesis, properties, and applications. Adv Funct Mater 2021; 31: 2105132. [Article] [Google Scholar]
  • Demirel G, Usta H, Yilmaz M, et al. Surface-enhanced Raman spectroscopy (SERS): An adventure from plasmonic metals to organic semiconductors as SERS platforms. J Mater Chem C 2018; 6: 5314-5335. [Article] [Google Scholar]
  • Han XX, Ji W, Zhao B, et al. Semiconductor-enhanced Raman scattering: Active nanomaterials and applications. Nanoscale 2017; 9: 4847-4861. [Article] [Google Scholar]
  • Zhou J, Lin J, Huang X, et al. A library of atomically thin metal chalcogenides. Nature 2018; 556: 355-359. [Article] [NASA ADS] [CrossRef] [PubMed] [Google Scholar]
  • Ding SY, Yi J, Li JF, et al. Nanostructure-based plasmon-enhanced Raman spectroscopy for surface analysis of materials. Nat Rev Mater 2016; 1: 16021. [Article] [Google Scholar]
  • Tian ZQ, Ren B, Wu DY. Surface-enhanced Raman scattering: From noble to transition metals and from rough surfaces to ordered nanostructures. J Phys Chem B 2002; 106: 9463-9483. [Article] [Google Scholar]
  • Zhang L, Fang M. Nanomaterials in pollution trace detection and environmental improvement. Nano Today 2010; 5: 128-142. [Article] [Google Scholar]
  • Chen Z, Wu C, Yuan Y, et al. CRISPR-Cas13a-powered electrochemical biosensor for the detection of the L452R mutation in clinical samples of SARS-CoV-2 variants. J Nanobiotechnol 2023; 21: 141. [Article] [CrossRef] [Google Scholar]
  • Liu D, Chen X, Hu Y, et al. Raman enhancement on ultra-clean graphene quantum dots produced by quasi-equilibrium plasma-enhanced chemical vapor deposition. Nat Commun 2018; 9: 193-202. [Article] [Google Scholar]
  • Xia F, Wang H, Xiao D, et al. Two-dimensional material nanophotonics. Nat Photon 2014; 8: 899-907. [Article] [Google Scholar]
  • Lin H, Zhu Q, Shu D, et al. Growth of environmentally stable transition metal selenide films. Nat Mater 2019; 18: 602-607. [Article] [Google Scholar]
  • Tao L, Chen K, Chen Z, et al. 1T′ transition metal telluride atomic layers for plasmon-free SERS at femtomolar levels. J Am Chem Soc 2018; 140: 8696-8704. [Article] [Google Scholar]
  • Song X, Wang Y, Zhao F, et al. Plasmon-free surface-enhanced Raman spectroscopy using metallic 2D materials. ACS Nano 2019; 13: 8312-8319. [Article] [Google Scholar]
  • Lv Q, Wu X, Tan J, et al. Ultrasensitive molecular sensing of few-layer niobium diselenide. J Mater Chem A 2021; 9: 2725-2733. [Article] [Google Scholar]
  • Ge Y, Wang F, Yang Y, et al. Atomically thin TaSe2 film as a high-performance substrate for surface-enhanced Raman scattering. Small 2022; 18: 2107027. [Article] [PubMed] [Google Scholar]
  • Martincová J, Otyepka M, Lazar P. Is single layer MoS2 stable in the air?. Chem Eur J 2017; 23: 13233-13239. [Article] [CrossRef] [PubMed] [Google Scholar]
  • Rhodes D, Chae SH, Ribeiro-Palau R, et al. Disorder in van der Waals heterostructures of 2D materials. Nat Mater 2019; 18: 541-549. [Article] [NASA ADS] [CrossRef] [PubMed] [Google Scholar]
  • Li Q, Zhou Q, Shi L, et al. Recent advances in oxidation and degradation mechanisms of ultrathin 2D materials under ambient conditions and their passivation strategies. J Mater Chem A 2019; 7: 4291-4312. [Article] [Google Scholar]
  • Yu W, Li J, Herng TS, et al. Chemically exfoliated VSe2 monolayers with room-temperature ferromagnetism. Adv Mater 2019; 31: 1903779. [Article] [Google Scholar]
  • Ling X, Xie L, Fang Y, et al. Can graphene be used as a substrate for Raman enhancement?. Nano Lett 2010; 10: 553-561. [Article] [Google Scholar]
  • Cheng H, Zhao Y, Fan Y, et al. Graphene-quantum-dot assembled nanotubes: A new platform for efficient Raman enhancement. ACS Nano 2012; 6: 2237-2244. [Article] [Google Scholar]
  • Liang C, Sun Q, Al-Salihy A, et al. Recent advances in crystal phase induced surface-enhanced Raman scattering. Chin Chem Lett 2024; 35: 109306. [Article] [CrossRef] [Google Scholar]
  • Liang C, Lu ZA, Zheng M, et al. Band structure engineering within two-dimensional borocarbonitride nanosheets for surface-enhanced Raman scattering. Nano Lett 2022; 22: 6590-6598. [Article] [Google Scholar]
  • Lin H, Chang M, Fu X, et al. Tunability of the superconductivity of NbSe2 films grown by two-step vapor deposition. Molecules 2023; 28: 1059. [Article] [PubMed] [Google Scholar]
  • Wang H, Huang X, Lin J, et al. High-quality monolayer superconductor NbSe2 grown by chemical vapour deposition. Nat Commun 2017; 8: 394. [Article] [Google Scholar]
  • Saito R, Tatsumi Y, Huang S, et al. Raman spectroscopy of transition metal dichalcogenides. J Phys-Condens Matter 2016; 28: 353002. [Article] [Google Scholar]
  • Staley NE, Wu J, Eklund P, et al. Electric field effect on superconductivity in atomically thin flakes of NbSe2. Phys Rev B 2009; 80: 184505. [Article] [Google Scholar]
  • Duan X, Wang C, Shaw JC, et al. Lateral epitaxial growth of two-dimensional layered semiconductor heterojunctions. Nat Nanotech 2014; 9: 1024-1030. [Article] [Google Scholar]
  • Xi X, Zhao L, Wang Z, et al. Strongly enhanced charge-density-wave order in monolayer NbSe2. Nat Nanotech 2015; 10: 765-769. [Article] [Google Scholar]
  • Yazyev OV, Chen YP. Polycrystalline graphene and other two-dimensional materials. Nat Nanotech 2014; 9: 755-767. [Article] [Google Scholar]
  • Thrall ES, Crowther AC, Yu Z, et al. R6G on graphene: High Raman detection sensitivity, yet decreased Raman cross-section. Nano Lett 2012; 12: 1571-1577. [Article] [Google Scholar]
  • Yang B, Xu H, Lu J, et al. Periodic grain boundaries formed by thermal reconstruction of polycrystalline graphene film. J Am Chem Soc 2014; 136: 12041-12046. [Article] [Google Scholar]
  • Huang PY, Ruiz-Vargas CS, van der Zande AM, et al. Grains and grain boundaries in single-layer graphene atomic patchwork quilts. Nature 2011; 469: 389-392. [Article] [Google Scholar]
  • Zhao T, Xu C, Ma W, et al. Ultrafast growth of nanocrystalline graphene films by quenching and grain-size-dependent strength and bandgap opening. Nat Commun 2019; 10: 4854. [Article] [Google Scholar]
  • Calandra M, Mazin II, Mauri F. Effect of dimensionality on the charge-density wave in few-layer 2H-NbSe2. Phys Rev B 2009; 80: 241108. [Article] [Google Scholar]
  • Wang Z, Su Q, Yin GQ, et al. Structure and electronic properties of transition metal dichalcogenide MX2 (M = Mo, W, Nb; X = S, Se) monolayers with grain boundaries. Mater Chem Phys 2014; 147: 1068-1073. [Article] [Google Scholar]
  • Malliakas CD, Kanatzidis MG. Nb-Nb interactions define the charge density wave structure of 2H-NbSe2. J Am Chem Soc 2013; 135: 1719-1722. [Article] [Google Scholar]
  • Burke K. Perspective on density functional theory. J Chem Phys 2012; 136: 150901. [Article] [CrossRef] [PubMed] [Google Scholar]
  • Soundiraraju B, George BK. Two-dimensional titanium nitride (Ti2N) MXene: Synthesis, characterization, and potential application as surface-enhanced Raman scattering substrate. ACS Nano 2017; 11: 8892-8900. [Article] [Google Scholar]

All Figures

thumbnail Figure 1

The batch growth of wafer-scale NC-NbSe2 films. (A) Schematic illustration of the two-step vapor deposition growth that involves a magnetron sputtering of Nb film and a subsequential selenylation into NC-NbSe2 film in a continuous preparation route. (B, C) Photographs and (D) optical microscopic image of as-grown NC-NbSe2 film on 4-inch SiO2/Si substrates. Notably, the SiO2 substrate was artificially exposed for visual guidance. (E) The representative Raman spectra of NC-NbSe2 film. The inset shows the characteristic Raman stretching modes of A1g and E2g1 of NbSe2. (F) AFM image and the corresponding height profile of NC-NbSe2 films directly grown on SiO2/Si. (G) Atomic structure model and (H) the ADF-STEM image of bilayer NC-NbSe2 film.
In the text
thumbnail Figure 2

The nanocrystalline domains in NC-NbSe2 film. (A, B) Large-scale and (C) zoom-in high-resolution TEM images, and (inset in (C)) Fourier transform pattern of crystalline NbSe2 (C-NbSe2) film grown on a sapphire substrate as comparison. (D) A large-scale TEM image showing the uniformity of as-grown NC-NbSe2 film on SiO2/Si substrate. (E) The EDS spectrum and (F) SAED pattern of as-grown NC-NbSe2 film. Notably, the Cu Kα signal in panel (E) sources from the TEM grid. (G) A representative high-resolution TEM image with (H) false color and crystalline orientation superimposed to highlight nano-domain in NC-NbSe2 film. (I) A representative grain boundary between nano-domain in NC-NbSe2 film.
In the text
thumbnail Figure 3

The SERS activity of NC-NbSe2 film. (A) Schematic diagram of the NC-NbSe2 films used for SERS activity towards R6G molecules. (B) Raman spectra of R6G coated on SiO2/Si, C-NbSe2 films, NC-1-NbSe2 film, and NC-NbSe2 film, respectively, correspond to 10 μL R6G solution (10−6 M) on substrates. Notably, the Raman intensities are normalized to the integration times. (C) The comparison of characteristic vibrational Raman peaks of R6G on NC-NbSe2 and C-NbSe2 film. Notably, the overall intensity was normalized to the intensity of peak at 1361 cm−1 on C-NbSe2 film. (D) The ADF-STEM images of NC-NbSe2 film. Notably, the amorphous domain (highlighted in purple) implies the GBs-enhanced surface adsorption of organic species on NC-NbSe2 film. (E) The corresponding SAED pattern of the bottom two domains in NC-NbSe2 film. (F) Diagram of energy level diagram and charge transfer transitions between R6G and NbSe2 film. Notably, μi-CT and μk-CT denote the charge transfers from the HOMO of the R6G molecules to the Fermi level of NbSe2, and from the Fermi level of NbSe2 to the LUMO of the R6G molecules, respectively. The DOS near the Fermi level is based on the results of previous studies [54,55].
In the text
thumbnail Figure 4

The comprehensive SERS performance evaluation of NC-NbSe2 film. (A) Stack plots of Raman spectra for R6G coated onNC-NbSe2 films with various R6G concentrations ranging from 1×10−5 to 1×10−10 M. (B) The time-dependent evolution of Raman spectra of R6G coated on NC-NbSe2 films exposure to air at 50 °C. (C) The time-dependent Raman spectra of R6G coated on NC-NbSe2 films that were pre-immerged in water for 30 min. (D) The robustness demonstration of NC-NbSe2 film with pre-immersion in water and subsequent reliable SERS signal towards RhB (4×10−6 M) detection.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.