Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

2D atomic crystal molecular superlattices by soft plasma intercalation

Abstract

Two-dimensional (2D) atomic crystal superlattices integrate diverse 2D layered materials enabling adjustable electronic and optical properties. However, tunability of the interlayer gap and interactions remain challenging. Here we report a solution based on soft oxygen plasma intercalation. 2D atomic crystal molecular superlattices (ACMSs) are produced by intercalating O2+ ions into the interlayer space using the plasma electric field. Stable molecular oxygen layer is formed by van der Waals interactions with adjacent transition metal dichalcogenide (TMD) monolayers. The resulting interlayer gap expansion can effectively isolate TMD monolayers and impart exotic properties to homo-(MoS2[O2]x) and hetero-(MoS2[O2]x/WS2[O2]x) stacked ACMSs beyond typical capacities of monolayer TMDs, such as 100 times stronger photoluminescence and 100 times higher photocurrent. Our potentially universal approach to tune interlayer stacking and interactions in 2D ACMSs may lead to exotic superlattice properties intrinsic to monolayer materials such as direct bandgap pursued for future optoelectronics.

Introduction

Two-dimensional (2D) atomic crystal superlattices allow effective manipulation of stacking and coupling of atomic layers of diverse materials without restricting lattice matching due to the van der Waals forces between the adjacent layers. Therefore, such artificial 2D atomic crystal superlattices possess a wide range of adjustable electronic properties, offering technological opportunities, and applications beyond the reach of existing materials1,2,3,4,5. The most common fabrication methods based on layer-by-layer exfoliation and transfer are complex, laborious, and time-consuming, with limited yield and reproducibility6,7,8,9,10,11. Chemical vapor deposition (CVD) has been successfully applied to produce high-quality 2D heterostructures, bilayer, and multilayer transition metal dichalcogenides (TMDs), as well as lateral surface superlattices. However, CVD lacks growth and stacking precision and is thus impractical for high-order vertical superlattices12,13,14,15,16,17.

Recently, a new approach for the formation of vertical 2D superlattices has been demonstrated. Instead of growing or stacking layers of different 2D materials on top of each other, superlattice structures can be produced by the intercalation of selected 2D atomic crystals with alkali metal ions via electrochemical reactions18,19,20. Molecular intercalation is now regarded as a new and promising way to create the new class of stable superlattices in which monolayer atomic crystals alternate with ammonium bromide molecular layers4. However, the current state-of-the-art electrochemical approach is a wet process that often uses a whole 2D crystal as the source material and thus suffers from excessive material consumption. Furthermore, the alkali metal intercalation method usually requires the protection of inert gas to avoid the degradation of the properties of these superlattices, while the ammonium bromide molecule used for electrochemical intercalation is a toxic material.

To address the above issues, here we propose a soft oxygen plasma intercalation concept and demonstrate the 2D atomic crystal molecular superlattices where monolayer TMDs alternate with oxygen molecular layers. This dry method is effective for both mechanically exfoliated or CVD-grown TMD flakes (including MoS2, WS2, MoSe2, WSe2, and ReS2, etc.) with thicknesses ranging from 2 to 8 layers.

By using MoS2 as a model system, we demonstrate that plasma intercalation with oxygen molecular layers produces MoS2[O2]x superlattices in which the interlayer distance increases from 0.6 to 0.9 nm compared to pure MoS2, thereby effectively decoupling the MoS2 monolayers. As such, the MoS2[O2]x superlattices display extremely strong photoluminescence (PL) with an intensity approximately 100 times higher compared to pristine MoS2. We confirm the superlattice structure by means of PL spectroscopy, Raman spectroscopy, atomic force microscopy (AFM), X-ray photoelectron spectroscopy (XPS), transmission electron microscopy (TEM) as well as first-principle atomistic numerical simulations. The bilayer MoS2[O2]x/WS2[O2]x superlattice lateral heterostructures show much better photoelectric performance than the pristine bilayer MoS2/WS2 lateral heterostructures.

Results

The schematic and chemical reaction of soft plasma intercalation

Figure 1 shows the schematic of fabricating monolayer-MoS2/O2-molecule superlattices by the soft oxygen plasma intercalation. The soft oxygen plasma was excited in the capacitive discharge mode (E-mode) of a planar low-frequency (2 MHz) inductively coupled plasma system, which is sketched in Supplementary Fig. 1. The capacitive coupling originating from the radial potential drop across the two ends of the planar induction coil generates the radial electrostatic field parallel to the substrate surface. This electrostatic field drives positive oxygen ions along the substrate surface and facilitates oxygen intercalation into the interlayer space between every two adjacent TMD layers. Furthermore, the E-mode discharge usually operates at very low input power (5–30 W) so that the ion density is too low to induce damage to the treated flakes. The chemical reaction of the soft oxygen plasma intercalation process consists of two half-reactions:

$${\mathrm{3O}}_2 + {\mathrm{e}}\mathop { \to }\limits^{{\mathrm{ionization}}} {\mathrm{2O}}_2^ + + {\mathrm{2O}}^ - + {\mathrm{e}},$$
(1)
$${\mathrm{MoS}}_{2} + x{\mathrm{O}}_{2}^ + + x{\mathrm{e}}\mathop { \to }\limits^{{\mathrm{intercalation}}} {\mathrm{MoS}}_{2}\left[ {\mathrm{O}}_{2} \right]_x.$$
(2)
Fig. 1: Soft oxygen plasma intercalation creates 2D ACMSs.
figure1

a Schematic of the soft oxygen plasma treatment of few-layer MoS2 flakes, where the plasma-induced radial electrostatic field is parallel to the interlayer space of MoS2 flakes, and the oxygen plasma contains O2+ ions, O ions, O2 molecules, and electrons. b MoS2[O2]x superlattices arise through the plasma-enabled oxygen intercalation and formation of the oxygen molecular layer within the interlayer spaces of the few-layer MoS2 flakes.

The oxygen molecules formed in the interlayer space after oxygen ion intercalation may be stabilized via the van der Waals interaction with the adjacent MoS2 monolayers thereby forming the MoS2[O2]x superlattices.

Microscopic and optical characterization of MoS2[O2]x superlattices

Figure 2a, b displays the AFM images together with the corresponding optical images of a mechanically exfoliated MoS2 flake taken before and after 3 min of the oxygen plasma intercalation, respectively. AFM images reveal an apparent increase in the cross-sectional thickness from 2.54 to 3.72 nm. The slight change of the root-mean-square (RMS) roughness from 0.46 ± 0.14 nm to 0.74 ± 0.10 nm suggests that the surface of the sample was less affected by the 3 min oxygen plasma intercalation. The 1.18 nm increase in thickness, distributed across the 2.54 nm MoS2 flake consisting of 4 monolayers by assuming 0.65 nm as the thickness of a single S–Mo–S layer, corresponds to an average increase in each van der Waals gap of 3.93 Å. Further studies using TEM (Fig. 2c, d) show the distinct difference in microstructure between the pristine and treated flakes, giving a clearly resolved interlayer distance expansion ranging from 6.05 Å in the pristine sample (Fig. 2c) to 9.27 Å in the treated one (Fig. 2d). The deduced interlayer distance expansion (3.22 Å) is consistent with that (3.93 Å) obtained by the AFM measurements within the error margins, further validating the formation of MoS2[O2]x superlattices.

Fig. 2: Structure and property evolution from MoS2 flake to MoS2[O2]x superlattice.
figure2

a, b AFM images of a pristine mechanically exfoliated four-layer MoS2 flake and the corresponding MoS2[O2]x superlattice obtained by the 3 min long oxygen plasma intercalation. Insets are the corresponding optical images and height profiles along the dashed white lines. Scale bars: 1 μm. c, d Cross-sectional TEM images of a pristine MoS2 flake and a MoS2[O2]x superlattice. Scale bars: 2 nm. e, f PL and Raman spectra of the MoS2 flake and MoS2[O2]x superlattice in a and b, respectively, as well as the corresponding spectra (after multiplying by four) of monolayer MoS2. g High-resolution XPS spectra of S 2p, Mo 3d, and O 1s of the pristine MoS2 flake and MoS2[O2]x superlattice.

More interestingly, as shown in Fig. 2e, the PL intensity of the MoS2 flake can be strongly enhanced by 100-fold after the oxygen plasma intercalation, along with a ~32 meV decrease of the full width at half-maximum (FWHM). For the pristine few-layer MoS2, the PL peak is well fitted to two peaks by using Lorentzian functions, which are assigned to A exciton (~1.83 eV) and B exciton (~1.96 eV), respectively21,22. Specifically, peak A is due to the direct electron–hole recombination in a neutral exciton at the K point, while peak B with higher energy is ascribed to the indirect electron–hole recombination in a neutral exciton at a lower valence band because of the spin–orbit coupling.

For the MoS2[O2]x superlattice, however, the intensity of peak A is greatly enhanced by 100-fold while peak B vanishes. It is well-known that the layer number reduction from multilayer to monolayer can also lead to this variation in PL properties due to the indirect-to-direct bandgap transition23. The frequency differences between \(E^1_{{\mathrm{2g}}}\) (in-plane vibrational mode) and A1g (out-plane vibrational mode) peak of Raman spectra provide a reliable means to determine the number of layers24,25. The Raman spectra in Fig. 2f show a slight change of the frequency difference between \(E^1_{{\mathrm{2g}}}\) and A1g modes from 23.6 to 23.0 cm−1, indicating that the number of layers in the 4-layer (4 L) flake does not change. Both the decrease in Raman intensity of \(E^1_{{\mathrm{2g}}}\) mode and the disorder between some of the layers in the TEM image (Fig. 2d) may be attributed to the disorder introduced in the MoS2 lattice by the incorporation of a small amount of oxygen-containing species as validated by the following XPS and first-principle calculation analysis. Another detailed experiment on the 6-layer MoS2 flake (Supplementary Fig. 2) reveals a thickness increase of 1.8 nm (corresponding to an average increase in each van der Waals gap of 3 Å), a PL enhancement of 60-fold (with peak A greatly enhanced and peak B disappearing) as well as similar Raman spectra behaviors.

We emphasize that both the astonishing enhancement of peak A and the vanishment of peak B are not the result of the indirect-to-direct bandgap transition from multilayer to monolayer due to a reduction in the layer numbers as proved by AFM, TEM, and Raman measurements. Instead, the observed indirect-to-direct bandgap transition owes to the formation of MoS2[O2]x superlattices which can effectively isolate the MoS2 monolayers26,27.

In order to testify the universality of our soft oxygen plasma intercalation on MoS2 flakes, we performed similar experiments on 32 MoS2 flakes with different layer numbers ranging from 2 to 8 layers and list PL (including peak A position shift, FWHM, and intensity enhancement) and Raman (including .. position and A1g position, FWHM) properties in Supplementary Tables 1 and 2, respectively. On average, the PL intensity (peak A) is increased by 53.1 times, and the FWHM of peak A decreases by 31 meV. Furthermore, the Raman statistical data shows that \(E^1_{{\mathrm{2g}}}\) position hardly changes while A1g position has a slight redshift of 0.7 cm−1 on average, revealing that the intralayer coupling is unaffected by the plasma intercalation, while the interlayer van der Waals coupling (out-of-plane vibration mode) becomes significantly weaker due to the isolation of every two adjacent MoS2 monolayers by the intercalated oxygen molecular layer. However, the van der Waals coupling between adjacent layers still exists and the corresponding Raman signals (A1g) can reflect the multilayer property as observed. On the other hand, the largely reduced van der Waals coupling plays a negligible effect on the PL effect as proved by the following calculated energy bands and therefore such MoS2[O2]x superlattices exhibit an extremely strong PL behavior similar to that of MoS2 monolayer.

We also used energy-dependent XPS to probe the oxygen intercalation effect by comparing high-resolution XPS spectra of S 2p, Mo 3d, and O 1s between the pristine MoS2 flake and the MoS2[O2]x superlattice, as shown in Fig. 2g. The strong peaks at 161.5 and 162.4 eV corresponding to S 2p3/2 and S 2p1/2 states, respectively, hardly change without any new peak emerging, indicating no S–O bonds formed during the oxygen plasma intercalation28. For O 1s and Mo 3d, however, the emergence of a strong peak at around 532 eV and a very small peak at 233.3 eV (apart from prominent Mo4+ 3d3/2 and Mo4+ 3d5/2 at 213.8 and 228.5 eV), respectively, reveal the fundamental change in microstructure after the oxygen plasma intercalation. The newly emerging O 1s peak can be decomposed into a peak at 531.7 eV (MoO3) and another peak at 532.9 eV (O2 molecule)29, while the small Mo 3d peak at 233.3 eV can be ascribed to Mo6+ (MoO3)30. In addition, doublet Mo 3d peaks also present a shift toward lower binding energy which indicates a shift in the Femi level toward the valence band and further validates the p-type doping by oxygen substitution at sulfur vacancies31,32. For the MoS2[O2]x superlattice, the atomic percentage of O in the chemical form of MoO3 and O2 is estimated to be 0.348:0.652, meaning that most oxygen elements exist in O2 molecules and interact with MoS2 monolayer via van der Waals coupling to construct the superlattice in which MoS2 monolayers alternate with oxygen molecular layers.

Mechanism of the soft oxygen plasma intercalation process

In order to exclude potentially competing for oxygen bonding mechanisms that may expand the interlayer space, we carried out density functional theory (DFT) calculations on pristine MoS2 bilayer and five oxygen-incorporated MoS2 bilayer systems, namely MoS2 bilayer with substitutional O at S site (Os), with substitutional 2O at 2S sites (2Os), with oxygen molecule layer intercalated (MoS2[O2]x), with both oxygen molecule layer intercalated and substitutional O site at S site (MoS2[O2]x + OS), with both oxygen molecule layer intercalated and substitutional 2O site at 2S sites (MoS2[O2]x + 2OS), as shown in Fig. 3. For the MoS2[O2]x + OS structure, the atomic percentage of O in the chemical form of MoO3 and O2 is 1:2, which is very close to the XPS deduced value (0.348:0.652). For the MoS2[O2]x + 2OS structure, the atomic percentage of O in the chemical form of MoO3 and O2 is 1:1. Therefore, the MoS2[O2]x + OS model is the nearest to the real MoS2[O2]x superlattices we obtained by such an optimum plasma intercalation process based on the above XPS results. The calculated band structures of three representative structures including MoS2 bilayer, MoS2[O2]x and MoS2[O2]x + OS are displayed in Fig. 3g–i, respectively, while those of other structures including MoS2 monolayer, MoS2 bilayer with substitutional O at S site (Os), MoS2 bilayer with substitutional 2O at 2S sites (2Os), and MoS2 bilayer with both oxygen molecule layer intercalated and substitutional 2O site at 2S sites (MoS2[O2]x + 2OS), are shown in Supplementary Fig. 3.

Fig. 3: Calculated atomic configurations and energy band structures confirm interlayer space expansion and transition to direct optical bandgap.
figure3

ae Calculated atomic configurations of a the pristine MoS2 bilayer and four oxygen-involved MoS2 bilayer systems, namely MoS2 bilayer b with substitutional O site at S site (OS), c with substitutional 2O at 2S sites (2OS), d with oxygen molecule layer intercalated (MoS2[O2]x), e with both oxygen molecule layer intercalated and substitutional O site at S site (MoS2[O2]x + OS), f with both oxygen molecule layer intercalated and substitutional 2O site at 2S sites (MoS2[O2]x + 2OS). gi Calculated band structures of the pristine MoS2 bilayer (f), the MoS2[O2]x structure (g), and the MoS2[O2]x + OS structure, in which the atomic percentage of O in the chemical form of MoO3 and O2 is 1:2 (close to the XPS deduced ratio 0.348:0.652). The red dotted line in g corresponds to the energy state of the oxygen molecule.

The pristine MoS2 bilayer exhibits the interlayer distance of 6.12 Å (Fig. 3a), while the interlayer distance increases slightly to 6.18 and 6.24 Å for the first two cases (Fig. 3b–c) with chemically bonded oxygen atoms (Mo–O bonds). Such an increase of the interlayer distance is not only inconsistent with the above experimental results but is also insufficient to cause an indirect-to-direct bandgap transition leading to the strong enhancement in PL intensity. Indeed, the calculated band structures of the first three cases (pristine MoS2 bilayer, MoS2 bilayer with OS, and MoS2 bilayer with 2OS) exhibit an indirect bandgap as shown in Fig. 3g, Supplementary Fig. 3b, c. In contrast, for the MoS2[O2]x, MoS2[O2]x + OS, and even MoS2[O2]x + 2OS modes, the interlayer distance can increase up to 9.23, 9.12, and 8.96 Å, respectively, due to the intercalation of oxygen molecular layer, fully complying with the above experimental results. Moreover, the calculated energy band structures of all the three modes display a direct bandgap (Fig. 3h, i and Supplementary Fig. 3d), which is comparable to that of MoS2 monolayer (Supplementary Fig. 3a), further supporting the conclusion that our soft oxygen plasma intercalation indeed can produce MoS2[O2]x superlattice in which MoS2 monolayers alternate with oxygen molecular layers.

Figure 4a, b presents the time-dependent PL and Raman properties of the 4 L MoS2 flake with the oxygen plasma treatment. One can deduce that the optimum treatment time for oxygen intercalation is 3 min. Once beyond this optimum treatment time, the crystal quality degrades significantly as reflected by both PL and Raman spectra. Similarly, the 6 L MoS2 flake displays the same evolution with an optimum treatment time of 4 min as reflected by the time-dependent PL and Raman properties in Supplementary Fig. 2. In order to have a more detailed understanding of the oxygen intercalation behavior, we chose one of the most common TMD materials, CVD-grown MoS2 bilayer, and recorded the time-dependent PL intensity mapping of peak A in Fig. 4c–h. The PL intensity first decreases at the first 30 s and then increases drastically and reaches a maximum (20 times as high as the pristine intensity) at 60 s. However, with the further increase in treatment time, the PL intensity decreases gradually.

Fig. 4: Treatment time-dependent PL and Raman properties and mechanism of the oxygen plasma intercalation.
figure4

a, b Time-dependent PL (a) and Raman (b) spectra of the plasma-treated 4 L MoS2 flake. ch Time-dependent PL intensity mapping of peak A of the plasma-treated CVD-grown MoS2 bilayer flake: c 0 s, d 30 s, e 60 s, f 90 s, g 120 s, and h 150 s. i Schematic physical picture of the oxygen plasma intercalation process.

By combining both the above experimental results and the DFT simulation, we obtain a clear physical picture of the oxygen plasma intercalation process as schematically illustrated in Fig. 4i. At the very beginning of the plasma treatment, the plasma irradiation can produce a small number of sulfur vacancies as well as the resultant formation of a small amount of Mo–O bonds via oxygen substitution at sulfur vacancies, thus suppressing the PL33. Thereafter, the O2+ ions generated by the plasma ionization (Eq. (1)) enter into the interlayer space driven by the force of the electrostatic field parallel to the interlayer space and subsequently form stable O2 molecules (Eq. (2)) there via the van der Waals interactions with the adjacent MoS2 monolayers. Consequently, the interlayer distance is expanding, effectively isolating the MoS2 and leading to the formation of MoS2[O2]x superlattice with intriguing PL comparable with that of the corresponding intrinsic monolayer. Obviously, the interspace expansion effect prevails over the influence of plasma-induced Mo–O bonds at the optimum treatment time. However, once the treatment time is beyond this critical value, the influence of the plasma-induced Mo–O bonds may become more and more prominent, finally giving rise to the degradation of the crystal quality as reflected by both PL and Raman spectra. The time-dependent PL intensity mapping studies on vertically oriented MoS2 bilayer (Supplementary Fig. 4) shows no clue of any intercalation effect but the PL degradation, further proving the determining role of the parallel electrostatic field as well as the influence of plasma-induced Mo–O bonds. As shown in Supplementary Fig. 5, the MoS2[O2]x superlattices based on bilayer, three-layer, four-layer, and six-layer MoS2 flakes exhibit good stability by maintaining its intriguing PL properties over 30 days.

Photoelectric properties of MoS2[O2]x/WS2[O2]x superlattices

We have also studied the photoelectrical properties of the same MoS2/WS2 bilayer heterostructure and MoS2[O2]x/WS2[O2]x superlattice heterostructure before and after the plasma intercalation process. Figure 5a, b displays the schematic illustration and the output characteristic (Ids − Vds curves) of the MoS2[O2]x/WS2[O2]x superlattice heterostructure under the 532 nm laser illumination with different power intensities, respectively. The photocurrent (Ids) increases with the increase of the incident power, indicating that the number of photo-induced carriers increases as a result of the increase in the number of absorbed photons. The energy band diagram of such superlattice heterostructure shown in the inset of Fig. 5b exhibits an extremely small build-in electric field directing from MoS2[O2]x to WS2[O2]x due to their similar bandgap (1.87 eV for MoS2[O2]x and 1.97 eV for WS2[O2]x) and identical n-type characteristics. This extremely small build-in electric field can be clearly reflected by the fact that the photocurrent (Ids) exhibits a faster increase with the increase in laser intensity under Vds > 0 V (forward bias) compared to the case under Vds < 0 V (reverse bias). Therefore, we choose two representative cases, the reverse bias (Vds = −1 V) and the forward bias (Vds = 1 V) to compare the photoelectrical properties between the same MoS2/WS2 bilayer heterostructure (before the plasma intercalation) and MoS2[O2]x/WS2[O2]x superlattice heterostructure (after the optimum plasma intercalation), as shown in Fig. 5c, d, respectively. For the former case (Vds = −1 V), the on/off ratio increases from ~10 to ~70 and the photocurrent increases by 70 times from 0.04 to 2.83 nA after the plasma intercalation process. For the latter case (Vds = 1 V), the photocurrent increases by more than 100 times from 0.46 to 48.7 nA after the plasma intercalation process although the on/off ratio is hard to estimate due to the longer decaying time. This strikingly enhanced photoresponse can be attributed to the indirect-to-direct energy band transition after the plasma intercalation process34,35. We also fabricated a vertical 3L-MoS2/multilayer-WSe2 p–n heterojunction by the mechanical exfoliation and transfer method. The optimum plasma intercalation process can translate the top 3 L MoS2 flake with an indirect bandgap into MoS2[O2]x superlattice with a direct bandgap and thus make the resultant self-powered photocurrent at Vds = 0 V largely increased by 20 times (Supplementary Fig. 6), suggesting that our soft plasma intercalation technique has good stability and reproducibility in controlling the optical and electronic properties of these superlattices. The photoelectric performance can be further improved by using thicker superlattice lateral heterostructures with higher absorption.

Fig. 5: Evolution of photoelectric properties from MoS2/WS2 bilayer heterostructure to MoS2[O2]x/WS2[O2]x superlattice heterostructure.
figure5

a Schematic illustration and typical optical images of the MoS2[O2]x/WS2[O2]x superlattice heterostructure. b Ids − Vds curves of the MoS2[O2]x/WS2[O2]x superlattice heterostructure photodetector under 532 nm laser illumination with different power intensities. Inset is the energy band diagram of the MoS2[O2]x/WS2[O2]x superlattice heterostructure. c, d The time-resolved photoresponse of both MoS2/WS2 bilayer heterostructure (before the plasma intercalation) and corresponding MoS2[O2]x/WS2[O2]x superlattice heterostructure (after the optimum plasma intercalation) under c Vds = −1 V (reverse bias) and d Vds = 1 V (forward bias) with 532 nm (1.74 mW) illumination.

Our approach may be expanded to intercalate diverse 2D mechanically exfoliated or CVD-grown TMD flakes, including WS2, MoSe2, and ReS2, etc. with thicknesses ranging from 2 to 8 layers, as evidenced by Supplementary Figs. 79, respectively. Similarly to the MoS2[O2]x superlattice, WS2[O2]x and MoSe2[O2]x superlattices exhibit the same variation trend in the structural and optical properties compared to their pristine counterparts. However, for ReS2[O2]x superlattice, no bandgap transition can be observed apart from the interlayer expansion. This result can be ascribed to the fact that both ReS2 monolayer and multilayer are direct bandgap semiconductors36,37,38. By considering the determining role of the electric field parallel to the interlayer space of TMDs, we expect that proper alignment of the plasma generated electric fields may lead to the observation of similar oxygen intercalation effects in other types of low-temperature plasmas.

Discussion

In summary, the presented soft oxygen plasma intercalation generates 2D ACMSs where TMD monolayers alternate with oxygen molecular layers. This dry method is suitable for mechanically exfoliated or CVD-grown TMD flakes (including MoS2, WS2, MoSe2, and ReS2, etc.) with thicknesses ranging from 2 to 8 layers. The physical mechanism can be ascribed to the O2+ ions entering the interlayer space driven by the parallel electric field and then forming stable O2 molecules via van der Waals interactions with the adjacent TMD monolayers. The interlayer distance can be largely expanded and becomes sufficient to effectively isolate the TMD monolayers as well as to almost suppress the interlayer coupling. These effects make superlattices such as MoS2[O2]x, WS2[O2]x, MoSe2[O2]x, and ReS2[O2]x display monolayer characteristics. The bilayer MoS2[O2]x/WS2[O2]x superlattice lateral heterostructures show much better photoelectric performance (100 times increased photocurrent) than the pristine bilayer MoS2/WS2 lateral heterostructures because of the indirect-to-direct energy band transition. Our studies thus provide a potentially universal approach to create such 2D ACMSs from pristine 2D nanomaterials and provide a generic platform for fundamental physics research and potential technological applications. Moreover, the 2D ACMSs with intrinsic monolayer characteristics such as a direct optical bandgap set an important milestone for future optoelectronics.

Methods

Sample preparation

Few-layer MoS2, WS2, MoSe2, and ReS2 flakes were exfoliated mechanically from corresponding bulk single crystals and deposited onto 300 nm SiO2/Si substrates. Bilayer MoS2 flakes were synthesized on 300 nm SiO2/Si substrates by chemical vapor epitaxy. Before mechanical exfoliation or CVD, all the substrates were first ultrasonically cleaned in acetone and alcohol, and then rinsed in deionized water and finally dried by nitrogen stream.

Oxygen plasma intercalation

As shown in Supplementary Fig. 1, a home-made planar low-frequency (2 MHz) inductively coupled plasma system was applied to produce the MoS2[O2]x, WS2[O2]x, MoSe2[O2]x, and ReS2[O2]x superlattice. The input RF power was kept at as low as 20 W so that the plasma was excited in the E-mode of ICP, producing a radial electrostatic field parallel to the substrate surface. The working pressure was kept at 38 Pa by introducing O2 as the precursor gas with a flow rate of 10 sccm. The sample stage was kept rotating during the plasma treatment process to ensure uniform intercalation into the interlayer space of TMD flakes. Unless other specified, the samples were placed horizontally on the sample stage.

Characterizations

Optical images were obtained by Optical Microscopy (Leica 4000M or Leica 2700M). AFM was carried out by using a Bruker Dimension ICON system in the tapping mode. Raman and PL measurements were recorded using a Renishaw Invia micro-Raman spectrometer with a 532 nm excitation laser. The laser power was lower than 0.2 mW to avoid any laser-induced heating. Raman spectra were measured using an 1800 l/mm grating to disperse the signal while PL ones were measured using a 600 l/mm grating. XPS (Thermo Scientific Esca lab 250Xi) with an Al-Kα (1486.6 eV) source was used to determine the chemical configurations of the TMD flake before and after oxygen plasma intercalation. The cross-sectional high-resolution transmission electron microscopy images were measured by using a Themis z TEM system with an accelerating voltage of 200 kV. All the measurements were performed at room temperature under ambient conditions.

Device fabrication and photoelectrical measurements

Bilayer MoS2/WS2 lateral heterojunction flakes were synthesized on 300 nm SiO2/Si substrates by adjusting both the temperature and the carrier gas flow direction. Then, the exfoliated BN flakes were transferred onto the surface of bilayer MoS2/WS2 heterojunction flakes serving as an insulating layer via a dry transfer method with polydimethylsiloxane (PDMS). Subsequently, we fabricated Cr/Au (5 nm/50 nm) contact electrodes by using e-beam lithography and electron beam evaporation. After that, the bilayer MoS2/WS2 heterojunction devices were subjected to soft oxygen plasma intercalation and transformed into MoS2[O2]x/WS2[O2]x superlattice heterostructures. The electrical and photoresponse characteristics of the same MoS2/WS2 bilayer heterostructure and MoS2[O2]x/WS2[O2]x superlattice heterostructure before and after the plasma intercalation process were measured using a Keithley 2643B analyzer under dark and illuminated conditions in an atmospheric environment. Renishaw Invia micro-Raman spectrometer with a 532 nm excitation laser was employed to attain the spectral photocurrent response. For all the photocurrent measurements, the lasers were focused on the sample with a 50× objective (NA = 0.5) and the spot size of the light is about 1 μm, much smaller than the device channel length. Optical attenuators were used to change the power of the illuminated laser and a chopper with a frequency of 1 Hz was used to record the time-dependent photoresponse.

First-principles calculations

The first-principles calculations were performed by the Vienna ab initio simulation package39 and the projected augmented-wave potential40,41. The exchange-correlation functional introduced by Perdew, Burke, and Ernzerhof42 within the generalized gradient approximation was applied in the calculations. The p semi-core states of Mo were described as valence electrons. For bilayer MoS2 we constructed a slab geometry with the insertion of a vacuum layer of 15 Å. The k-space mesh used was 12 × 12 × 1 for the slab structure. The energy cutoff for the plane-wave basis was set as 520 eV and the forces are relaxed less than 0.01 eV/Å. The positions of atoms were allowed to relax while the lattice constants of the unit cells were fixed to the experimental values.

Data availability

The data that support the findings of this study are available from the corresponding author on reasonable request.

References

  1. 1.

    Novoselov, K. S., Mishchenko, A., Carvalho, A. & Castro Neto, A. H. 2D materials and van der Waals heterostructures. Science 353, 9439 (2016).

    Article  CAS  Google Scholar 

  2. 2.

    Liu, Y. et al. Van der Waals heterostructures and devices. Nat. Rev. Mater. 1, 16042 (2016).

    ADS  CAS  Article  Google Scholar 

  3. 3.

    Jariwala, D., Marks, T. J. & Hersam, M. C. Mixed-dimensional van der Waals heterostructures. Nat. Mater. 16, 170–181 (2016).

    ADS  PubMed  Article  CAS  PubMed Central  Google Scholar 

  4. 4.

    Wang, C. et al. Monolayer atomic crystal molecular superlattices. Nature 555, 231–236 (2018).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  5. 5.

    Ryu, Y. K., Frisenda, R. & Castellanos-Gomez, A. Superlattices based on van der Waals 2D materials. Chem. Commun. 55, 11498–11510 (2019).

    CAS  Article  Google Scholar 

  6. 6.

    Dean, C. R. et al. Boron nitride substrates for high-quality graphene electronics. Nat. Nanotech. 5, 722–726 (2010).

    ADS  CAS  Article  Google Scholar 

  7. 7.

    Withers, F. et al. Light-emitting diodes by band-structure engineering in van der Waals heterostructures. Nat. Mater. 14, 301–306 (2015).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  8. 8.

    Haigh, S. J. et al. Cross-sectional imaging of individual layers and buried interfaces of graphene-based heterostructures and superlattices. Nat. Mater. 11, 764–767 (2012).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  9. 9.

    Yu, W. J. et al. Vertically stacked multi-heterostructures of layered materials for logic transistors and complementary inverters. Nat. Mater. 12, 246–252 (2013).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  10. 10.

    Cheng, R. et al. Electroluminescence and photocurrent generation from atomically sharp WSe2/MoS2 heterojunction p-n diodes. Nano Lett. 14, 5590–5597 (2014).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  11. 11.

    Lee, C. H. et al. Atomically thin p-n junctions with van der Waals heterointerfaces. Nat. Nanotech. 9, 676–681 (2014).

    ADS  CAS  Article  Google Scholar 

  12. 12.

    Duan, X. D. et al. Lateral epitaxial growth of two-dimensional layered semiconductor heterojunctions. Nat. Nanotech. 9, 1024–1030 (2014).

    ADS  CAS  Article  Google Scholar 

  13. 13.

    Gong, Y. J. et al. Vertical and in-plane heterostructures from WS2/MoS2 monolayers. Nat. Mater. 13, 1135–1142 (2014).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  14. 14.

    Yan, A. M. et al. Direct growth of single- and few-layer MoS2 on h-BN with preferred relative rotation angles. Nano Lett. 15, 6324–6331 (2015).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  15. 15.

    Zhang, Z. W. et al. Robust epitaxial growth of two-dimensional heterostructures, multiheterostructures, and superlattices. Science 357, 788–792 (2017).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  16. 16.

    Li, M. Y. et al. Epitaxial growth of a monolayer WSe2-MoS2 lateral p-n junctionwith an atomically sharp interface. Science 349, 524–528 (2015).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  17. 17.

    Zhang, X. M. et al. Transition metal dichalcogenides bilayer single crystals by reverse-flow chemical vapor epitaxy. Nat. Commun. 10, 598 (2019).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  18. 18.

    Bao, W. Z. et al. Approaching the limits of transparency and conductivity in graphitic materials through lithium intercalation. Nat. Commun. 5, 4224 (2014).

    ADS  CAS  PubMed  Article  Google Scholar 

  19. 19.

    Xiong, F. et al. Li intercalation in MoS2: in situ observation of its dynamics and tuning optical and electrical properties. Nano Lett. 15, 6777–6784 (2015).

    ADS  CAS  PubMed  Article  Google Scholar 

  20. 20.

    Yu, Y. J. et al. Gate-tunable phase transitions in thin flakes of 1T-TaS2. Nat. Nanotech. 10, 270–276 (2015).

    ADS  CAS  Article  Google Scholar 

  21. 21.

    Lin, Y. et al. Dielectric screening of excitons and trions in single-layer MoS2. Nano Lett. 14, 5569–5576 (2014).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  22. 22.

    Xie, Y. et al. Controllable growth of monolayer MoS2 by chemical vapor deposition via close MoO2 precursor for electrical and optical applications. Nanotechnology 28, 084001 (2017).

    ADS  PubMed  Article  CAS  PubMed Central  Google Scholar 

  23. 23.

    Splendiani, A. et al. Emerging photoluminescence in monolayer MoS2. Nano Lett. 10, 1271–1275 (2010).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  24. 24.

    Li, H. et al. From bulk to monolayer MoS2: evolution of raman scattering. Adv. Funct. Mater. 22, 1385–1390 (2012).

    ADS  CAS  Article  Google Scholar 

  25. 25.

    Lee, C. et al. Anomalous lattice vibrations of single- and few-layer MoS2. ACS Nano 4, 2695–2700 (2010).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  26. 26.

    Dhall, R. et al. Direct bandgap transition in many-layer MoS2 by plasma-induced layer decoupling. Adv. Mater. 27, 1573–1578 (2015).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  27. 27.

    Mak, K. F., Lee, C., Hone, J., Shan, J. & Heinz, T. F. Atomically thin MoS2: a new direct-gap semiconductor. Phys. Rev. Lett. 105, 136805 (2010).

    ADS  PubMed  Article  CAS  PubMed Central  Google Scholar 

  28. 28.

    Kim, S. et al. Effects of plasma treatment on surface properties of ultrathin layered MoS2. 2D Mater. 3, 035002 (2016).

    Article  CAS  Google Scholar 

  29. 29.

    Petersson, L. G. & Karlsson, S. E. Clean and oxygen exposed potassium studied by photoelectron spectroscopy. Phys. Scr. 16, 425–431 (1977).

    ADS  CAS  Article  Google Scholar 

  30. 30.

    Brown, N. M. D., Cui, N. Y. & Mckinley, A. An XPS study of the surface modification of natural MoS2 following treatment in an RF-oxygen plasma. Appl. Surf. Sci. 134, 11–21 (1998).

    ADS  CAS  Article  Google Scholar 

  31. 31.

    Wu, S. X. et al. High-performance p-type MoS2 field-effect transistor by toroidal-magnetic-field controlled oxygen plasma doping. 2D Mater. 6, 025007 (2019).

    CAS  Article  Google Scholar 

  32. 32.

    Tang, B. et al. Direct n- to p-type channel conversion in monolayer/few-Layer WS2 field-effect transistors by atomic nitrogen treatment. ACS Nano 12, 2506–2513 (2018).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  33. 33.

    Kang, N., Paudel, H. P., Leuenberger, M. N., Tetard, L. & Khondaker, S. I. Photoluminescence quenching in single-layer MoS2 via oxygen plasma treatment. J. Phys. Chem. C 118, 21258–21263 (2014).

    CAS  Article  Google Scholar 

  34. 34.

    Wang, Q. H., Kalantar-Zadeh, K., Kis, A., Coleman, J. N. & Strano, M. S. Electronics and optoelectronics of two-dimensional transition metal dichalcogenides. Nat. Nanotechnol. 7, 699–712 (2012).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  35. 35.

    Lei, J. et al. Layer-dependent photoresponse of 2D MoS2 films prepared by pulsed laser deposition. J. Mater. Chem. C 7, 2522–2529 (2019).

    Article  Google Scholar 

  36. 36.

    Shim, J. et al. Thin-film transistors: high-performance 2D rhenium disulfide (ReS2) transistors and photodetectors by oxygen plasma treatment. Adv. Mater. 28, 6984–6984 (2016).

    CAS  Article  Google Scholar 

  37. 37.

    Rahman, M., Davey, K. & Qiao, S. Z. Advent of 2D rhenium disulfide (ReS2): fundamentals to applications. Adv. Funct. Mater. 27, 1606129 (2017).

    Article  CAS  Google Scholar 

  38. 38.

    Hafeez, M., Gan, L., Li, H. Q., Ma, Y. & Zhai, T. Y. Large-area bilayer ReS2 film/multilayer ReS2 flakes synthesized by chemical vapor deposition for high performance photodetectors. Adv. Funct. Mater. 26, 4551–4560 (2016).

    CAS  Article  Google Scholar 

  39. 39.

    Huang, Q. et al. First-principles study of O-BN: A sp3-bonding boron nitride allotrope. J. Appl. Phys. 112, 053518 (2012).

    ADS  Article  CAS  Google Scholar 

  40. 40.

    Lee, B. & Rudd, R. E. First-principles calculation of mechanical properties of Si001 nanowires and comparison to nanomechanical theory. Phys. Rev. B 75, 195328 (2007).

    ADS  Article  CAS  Google Scholar 

  41. 41.

    Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B 59, 1758–1775 (2007).

    ADS  Article  Google Scholar 

  42. 42.

    Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996).

    ADS  CAS  Article  Google Scholar 

Download references

Acknowledgements

This work is partially supported by the National Nature Science Foundation under Grants 62074070, 11704159, and 11574215, the Natural Science Foundation of Jiangsu Province, China under Grants BK20170167 and BK20190576, the Fundamental Research Funds for the Central Universities of China under Grants JUSRP51726B, the 111 Project under Grant B12018, Postgraduate Research & Practice Innovation Program of Jiangsu Province under Grant KYCX18_1860. A.J.D. and K.O. acknowledge partial support by the Australian Research Council (ARC).

Author information

Affiliations

Authors

Contributions

S.Q.X. conceived the project, designed the experiment, and jointly with K.O. conceptualized the ideas. L.F.Z. and H.Y.N. performed the experiment. X.M.Z. performed the CVD synthesis of TMD flakes and heterojunctions. Q.F.L. and A.J.D. provided theoretical support. L.F.Z. supervised by H.Y.N. and Z.H.N. measured the photoresponse. S.Q.X., L.F.Z., H.Y.N., Q.F.L., X.F.G., and K.O. jointly analyzed the data and interpreted the results. S.Q.X., L.F.Z., H.Y.N., K.O., and X.F.G. co-wrote the paper. All co-authors discussed the results and commented on the paper at all stages.

Corresponding author

Correspondence to Shaoqing Xiao.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zhang, L., Nan, H., Zhang, X. et al. 2D atomic crystal molecular superlattices by soft plasma intercalation. Nat Commun 11, 5960 (2020). https://doi.org/10.1038/s41467-020-19766-x

Download citation

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing