September 07, 2014
C 2014 American Chemical Society
Spectroscopic Signatures for
Interlayer Coupling in MoS
2
WSe
2
van der Waals Stacking
Ming-Hui Chiu,†,XMing-Yang Li,‡,XWengjing Zhang,§,^Wei-Ting Hsu,)Wen-Hao Chang,) Mauricio Terrones,#Humberto Terrones,rand Lain-Jong Li*,†
†Physical Sciences and Engineering, King Abdullah University of Science and Technology, Thuwal 23955-6900, Kingdom of Saudi Arabia,‡Institute of Atomic and
Molecular Sciences, Academia Sinica, Taipei 11529, Taiwan,§Department of Physics, National University of Singapore, 2 Science Drive 3, Singapore 117542,
^SZU-NUS Collaborative Innovation Center for Optoelectronic Science & Technology, Shenzhen University, Shenzhen 518060, China, )Department of Electrophysics,
National Chiao Tung University, Hsinchu 300, Taiwan,#Department of Physics, Chemistry, Materials Science and Engineering and Center for 2-Dimensional and Layered Materials, The Pennsylvania State University, 104 Davey Lab., University Park, Pennsylvania 16802, United States, andrDepartament of Physics, Applied Physics and Astronomy, Rensselaer Polytechnic Institute, 110 Eighth Street, Troy, New York 12180, United States.XThese authors contributed equally.
S
emiconducting transition metal dichal-cogenide (STMD) layered materials ex-hibit unique layer-dependent electronic and optical properties.112For example, mo-lybdenum disulfide (MoS2) exhibits anindi-rect bandgap of 1.2 eV in bulk, but it becomes a direct gap semiconductor (bandgap = 1.8 eV) when thinned to a monolayer.4,8 Hence, monolayer MoS2 transistors have
been fabricated and these devices have showed excellent current on/off ratios.1012
MoS2is known as an n-type semiconductor
due to the presence of S vacancies. Another STMD, tungsten diselenide (WSe2), has just
started to attract the attention of numerous scientists due to its p-type characteristics.2,13 In addition to the mechanical exfoliation method to prepare STMD layers, recent de-velopments in the scalable synthesis of STMD monolayers using chemical vapor deposition (CVD) opened up the possibility to form large area of STMD monolayers that could result in fabrication of flexible
electronic devices and photodetectors.1,2,14 Furthermore, it is also possible to envisage building unprecedented solids by stacking monolayers of different 2-D systems one on top of another (e.g., TMDs, h-BN and graphene).1521 Such stacked heterostruc-tures based on atomically thin 2D layers are fundamentally different since only van der Waals interactions exist at the interfaces. Therefore, these layered materials (termed also van der Waals solids) enable the pre-paration of high-quality heterointerfaces without the need of fulfilling an atomic commensurability,17,18,22 thus making the
structure construction easily achievable. Very recently, MoSe2/WSe2 optical studies were
performed,23 and other optoelectronic de-vices based on WSe2/MoS2 pn junction
were also proposed.2427Interestingly, the gate-tunable diode-like current rectification and a photovoltaic response have been re-cently observed in WSe2/MoS2
heterojunc-tions.2426It should be noted that the optical
* Address correspondence to lance.li@kaust.edu.sa.
Received for review July 30, 2014 and accepted September 7, 2014. Published online
10.1021/nn504229z
ABSTRACT Stacking of MoS2and WSe2monolayers is conducted by transferring triangular
MoS2monolayers on top of WSe2monolayers, all grown by chemical vapor deposition (CVD).
Raman spectroscopy and photoluminescence (PL) studies reveal that these mechanically stacked monolayers are not closely coupled, but after a thermal treatment at 300C, it is possible to produce van der Waals solids consisting of two interacting transition metal dichalcogenide (TMD) monolayers. The layer-number sensitive Raman out-of-plane mode A21gfor WSe2(309 cm1) is
found sensitive to the coupling between two TMD monolayers. The presence of interlayer excitonic emissions and the changes in other intrinsic Raman modes such as E00for MoS2at 286 cm1and
A21gfor MoS2at around 463 cm1confirm the enhancement of the interlayer coupling.
KEYWORDS: transition metal dichalcogenides . tungsten diselenides . molybdenum disulfide . van der Wall stacking . heterojunction . interlayer coupling
and electrical properties of TMD heterojuctions strongly depend on the interaction among layers, and efforts aiming at elucidating their proper charac-terization is currently underway.
In this paper, we demonstrate that it is possible to build van der Waals solids by first stacking CVD-grown MoS2on WSe2monolayers, followed by thermal
annealing. The layer-number sensitive Raman mode A21gfor WSe2at around 309 cm1(out-of-plane mode)
appears to be excellent parameters when evaluating a good coupling between monolayers. The Raman modes for bulk MoS2at ca. 463 cm1(A21g) and at ca.
283 cm1(E00) are also enhanced by interlayer coupling. In addition, the interlayer excitonic peak observed in photoluminescence (PL) confirms the coupling between two TMD monolayers and a band diagram of the MoS2/WSe2 heterojunction and the exciton
binding energies for each composition are proposed.
RESULTS AND DISCUSSION
Preparation and Characterizations of Heterojunction. Trian-gular WSe2 and MoS2 single crystalline monolayers
with the lateral dimension of few tens of micrometers were synthesized on c-plane sapphire substrates by CVD.1,2 In brief, transition metal trioxides (MoO3
or WO3) were vaporized and reacted with the S or
Se vapor in a hot-wall furnace under a controlled gaseous environment to form MoS2 or WSe2
mono-layers (see details in Methods). To build a van der Waals solid (or vertical heterojunction of MoS2on WSe2), as
schematically illustrated in Figure 1a, we first detached as-synthesized MoS2 triangular islands using PMMA.
In particular, a layer of PMMA was spin-coated on MoS2followed by dipping it in NaOH so as to release
the PMMA-supported MoS2from the substrates. The
PMMA supported-MoS2was then mechanically
trans-ferred onto WSe2 flakes followed by the removal/
cleaning of PMMA.3Figure 1b is the optical micrograph (OM) showing the mechanically stacked MoS2/WSe2
monolayers, in which it is possible to distinguish the stacked layers by noticing a color contrast. Supporting Information Figure S1 shows the atomic force micro-scopy (AFM) image of as-transferred MoS2/WSe2, where
the surface of the sample is covered with residuals and adsorbates. In this context, it has been demonstrated by scanning tunneling microscopy (STM) that the majority of the adsorbates and residuals can be removed after thermal annealing at elevated temperatures.28 There-fore, as-prepared MoS2/WSe2 samples were then
an-nealed in an hydrogen/Ar environment (atmosphere pressure; H2:Ar = 1:4) at 300C for 4 h. The detailed
fabrication process is described in Methods.
The AFM image and cross-sectional profile of the MoS2/WSe2 heterojunction are shown in Figure 1c,
demonstrating that both WSe2and MoS2are indeed
monolayers (thickness for each is around 0.60.7 nm). The AFM image also shows that the individualflakes
are clean and without cracks or holes after the H2/Ar
thermal treatment. Interestingly, the corrugated struc-tures are found at the overlapped region, which could be due to a commensurability adjustment of the layers in order to minimize the energy followed the thermal treatment. Alternatively, it is also possible that water, used as the solvent to transfer MoS2onto WSe2, got
trapped between MoS2and WSe2monolayers, and its
evaporation resulted in the corrugation of the top transferred layer during the thermal treatment. The high spikes of the cross-sectional profile in inset of Figure 1c reveal the profile of the corrugated structures.
Raman Features. For the Raman spectra we have used the notation of the monolayer modes using the symbols A01, E0, and the notation of the bilayer for the
higher order out of plane modes such as the A21g, since
these modes are not Raman active in the monolayer. Figure 2ad shows the Raman spectra of the double layer van der Waals solid excited with a 473 nm laser in four energy regions, where the curves from the top to bottom are MoS2region (blue), MoS2/WSe2
hetero-junction before annealing (green;μM/W), MoS2/WSe2
heterojunction after annealing (red; cM/W), and WSe2
region (black), respectively. The characteristic peaks for WSe2at about 250 cm1(E0and A01degenerated
mode) and those for MoS2located at 387 cm1(E0mode;
in-plane vibration) and 402 cm1 (A01 mode;
out-of-plane vibration) are observed in individual monolayers of WSe2and MoS2. In addition, the layer-number
sensi-tive mode A21gfor WSe2at around 309 cm1
(out-of-plane mode) has been reported only observable when the WSe2is a bilayer or thicker.29,30A similar feature A21g
for bulk MoS2 is located at around 463 cm1.3136
Furthermore, there is another characteristic peak for bulk MoS2at around 284 cm1in odd number of layers
Figure 1. Representation of the vertical heterojunction consisting of MoS2monolayers deposited on top of WSe2
monolayers: (a) schematic illustration of the MoS2/WSe2van
der Waals heterojunction; (b) optical micrograph of the mechanically stacked MoS2/WSe2 structure after thermal
annealing, and (c) AFM image of the MoS2/WSe2
heterojunc-tion. The cross-sectional height profile along the dashed line indicates that each layer is about 0.60.7 nm in thickness.
is called E00 and in even number of layers becomes Eg
due to the symmetry involved.3133Our Raman results for either MoS2 or WSe2region do not exhibit these
characteristic peaks, confirming that WSe2 and MoS2
are monolayers. Note that the Raman band at 260 cm1 observed at WSe2 region (Figure 2a) and at about
450 cm1 at MoS2 region (Figure 2d) corresponds to
the second order LA(M) phonon (2LA(M)).3537
It is observed that the Raman peaks for either MoS2
or WSe2 region show no obvious change after the
thermal annealing at 300C as shown in Supporting Information Figure S2a, which implies that the anneal-ing process does not cause significant structural de-fects or modification. For the heterojunction before annealing, wefind that the Raman characteristic peaks are closely coinciding with the individual WSe2
and MoS2 regions, which implies that the layers are
uncoupled (independent from each other). By con-trast, after the thermal annealing process, the Raman spectrum for the heterojunction becomes significantly different from that for the individual WSe2and MoS2
monolayers. First, the layer-sensitive characteristic peak A2
1g for WSe2 at 309 cm1 clearly appear in
Figure 2b. Second, the E0and A01degenerated peak
for the WSe2shows a blue-shift as shown in Figure 2a,
and the A01(E0) for the MoS2shows blue (red)-shifts as
shown in Figure 2c, respectively. Third, the anomaly E00feature for MoS2is also observed in Figure 2b, and
the layer-sensitive peak A21gfor MoS2is enhanced as
found in Figure 2d.
The presence of the layer-sensitive Raman bands after thermal annealing strongly evidence that the
MoS2 and WSe2 interact with each other. It is also
possible that some contaminants such as adsorbates trapped between MoS2and WSe2weaken the coupling
between these two layers. After the thermal treatment, these contaminants are efficiently removed (as indi-cated by AFM), thus resulting in the commensurate stacking and the shifts in the Raman peaks. Figure 2e is the OM for the selected area for Raman mapping studies. Figure 2f shows the spatial mappings for the peak intensity of the Raman bilayer signature at 309 cm1for WSe2A21gmode, and the Raman features
is only observed and distributed uniformly in the stacked region, indicating that the peak is originated from the interlayer coupling in the new MoS2/WSe2
bilayer. It is noteworthy that the peak A21gis Raman
inactive in monolayer but become active in homo-bilayer or thicker layers. Here, the heterojunction stack-ing likely introduces a similar effect as in stacked homolayers, leading to the symmetry change and activation of the Raman features.
To further investigate the main features observed in the experimental Raman spectra of the MoS2/WSe2
junction, density functional theory (DFT) and density functional perturbation theory (DFPT) calculations were carried out using the plane wave code CASTEP as implemented in the Materials Studio package.38A hexagonal unit cell with one layer of MoS2and another
of WSe2arranged in an AB stacking was considered
under the local density approximation (LDA). To pro-vide a reasonable description of the van der Waals interaction, we have considered a dispersion correc-tion for LDA (LDA-D).39After the relaxation, due to the lattice mismatch between the monolayers, the MoS2
lattice parameter suffered an expansion from 3.16 to 3.21 Å and the WSe2lattice contracted from 3.28 to
3.21 Å. The phonon dispersion and Raman scattering modes were calculated using the linear response approach for insulators.40The result of Raman scatter-ing modes is shown in Figure 3, where all the modes, not the resonant second order or combinations of first order modes, are shown in the Figure 3. The characteristic peaks E0 and A01 for WSe2 and MoS2
are exactly reproduced. The MoS2/WSe2bilayer
vibra-tions at 309 cm1(WSe2A21g) is also clearly seen from
Figure 3. Raman scattering modes simulation of the MoS2/
WSe2heterojunction.
Figure 2. Raman spectra of the MoS2/WSe2heterojunction.
(ad) Raman spectra for MoS2(blue), uncoupled MoS2/
WSe2heterojunction (green; μM/W), coupled MoS2/WSe2
heterojunction (red; cM/W), and WSe2(black) from the top
to the bottom plot. The curves are shifted for clarity. (e) Optical microscopic image for the MoS2/WSe2
hetero-junction. (f) The spatial mapping of the Raman intensity for WSe2A
2
1gpeak in the selected area shown in (e).
the simulation. Although the calculated frequencies are not exactly the same as obtained in experiments, the simulation results provide a good approximation of the Raman signals observed. The coincidence be-tween the experimental and simulated results implies that the coupling exists in the MoS2/WSe2 bilayer
heterojunction.
The Raman simulation in Figure 3 also presents the MoS2bulk vibration mode A21gat 458 cm1and
the anomaly E00mode at 279 cm1. Comparing to our experimental results, although there have other Raman active modes near 460 cm1, the MoS2A21gmode still
can be identified in Figure 2d for cW/M sample. The observed E00mode at 284 cm1in Figure 2b for cW/M sample can also be correlated to the simulation result. This peak is not normally observed because of the Raman geometry used (backscattering geometry) and also because it has low intensity. The observation on the E00mode may indicate that the presence of the interlayer coupling could enhance its intensity. The observation on these Raman peaks for bulk MoS2is also an evidence
of the existence of the interlayer coupling between MoS2and WSe2.
It is known that the Raman E0peak of monolayer MoS2is sensitive to strain and strain could induce a
red-shift in the E0modes that have been observed for MoS2
monolayer or multilayers in other reports.33,4143 The red-shift of the MoS2E0peak after the two layers
are coupled (thermally treated) shown in Figure 2c indicates that a tensile strain has been imposed to the MoS2. This also implies that the WSe2 monolayer
experiences a compressive strain and the E0peak of WSe2also shows an opposite blue-shift as shown in
Figure 2a. As stated above, our simulation concludes the expansion of the MoS2and contraction of the WSe2
(details in Methods) which explain in part the E0peak shift in MoS2and WSe2after the two layers are coupled.
Although the strain effect could explain the Raman shifts of the annealed hetero bilayer, we still cannot exclude the possibility of charge transfer between these stacked layers. The Raman A01 peak of MoS2
blue-shifts and its intensity increases (relative to E0), thus indicating that MoS2is less n-doped (or a decrease
in the electron concentration).44,45The blue-shift of the
WSe2 Raman A01 peak implies that the WSe2 may
become less p-doped (Supporting Information Figure S3 shows that in our separate experiment, p-doping from AuCl4causes a red-shift of Raman A01peak in
monolayer WSe2). These phenomena suggest that
the electrons transfer from MoS2 to WSe2, which is
expected for the PN-junction composed by the p-type WSe2and n-type MoS2.2427It is also noted that the
effect of a compressive strain on E0and the n-doping effect on A0
1in WSe2is not easily distinguishable since
A01and E0 frequencies in WSe2 are degenerate and
both effects are expected to cause the blue-shift. However, the fact that the Raman shift of MoS2 E0
mode is larger than the A01mode implies that the strain
effect should play a major role than doping effect. On the other hand, the theoretical simulation including only the strain relaxation can well reproduce our experimental results, which further suggests that the strain effect is more significant. In any case, the shifts indicate the interaction between two stacked hetero-layers. It is worth noting that our results show that the A01 and E0 peak energies of MoS2 do not seem
to depend on the stacking angle between the two layers as shown in Supporting Information Figure S4, which is different from the recent studies on Raman spectra of bilayer MoS2 which shows twisting angle
dependence.46,47 This could be reasoned since the homobilayer might exhibit stronger interlayer cou-pling due to the same lattice constant for two layers (better matching between top and bottom layers).
Band Gaps and Band Alignments. To study the optical and energy band properties of the heterojunction, we performed the PL and absorption measurements. Figure 4a shows the PL spectra excited with a 532 nm laser and Figure 4b displays the absorption spectra for MoS2, uncoupled MoS2/WSe2heterojunction (μM/W),
coupled MoS2/WSe2heterojunction (cM/W), and WSe2
regions. The excitonic PL peaks for WSe2 and MoS2
monolayers at about 1.65 and 1.85 eV are clearly observed in the respective nonstacking region, and the peak energies also hold after annealing process as shown in Supporting Information Figure S2b. For theμM/W region, the PL peak energy of WSe2or MoS2
is very close to that for its corresponding layer, thus confirming that there is no strong layer coupling in the heterojunction. The slight suppression of the PL intensity for both MoS2and WSe2may be due to the
Figure 4. (a) Photoluminescence and (b) absorption spec-troscopy of the MoS2/WSe2 heterojunction. The curves
corresponding to MoS2 (blue), noncoupled MoS2/WSe2
heterojunction (μM/W) (green), coupled MoS2/WSe2
hetero-junction (cM/W) (red), and WSe2(black) from up to down.
The curves are shifted for clarity. The dashed line indicates the interlayer PL peak. The orange lines and arrow mark the anti-Stoke shift.
defects/traps introduced by the stacking. For the cM/W sample, the PL intensity for both WSe2 and MoS2
is significantly lower compared to that of uM/W. In addition to the characteristic PL peaks from MoS2and
WSe2, an extra peak at a lower energy (∼1.59 eV), and
marked by a black dashed line is also observed. This PL peak appeared in previous work and it is attributed to the interlayer radiative recombination of the spatially separated carriers.27 The excitons are excited sepa-rately in WSe2and MoS2by the incident laser. Since
the MoS2/WSe2 junction is a type II heterojunction,
the excited electrons in WSe2tend to accumulate in the
conduction band of MoS2and the holes in MoS2in the
valence band of WSe2at the interface as illustrated in
Figure 5b. Therefore, the recombination of the electron hole pairs at the interface gives lower energy than MoS2
and WSe2itself. The interlayer PL peak reveals the nature
of the coupling in the heterojunction.
The absorption spectrum for individual MoS2
ex-hibits two absorptions peaks, A and B excitonic peaks at 1.87 and 2.01 eV, respectively. The WSe2has only
one peak centered at 1.70 eV. As shown in Figure 4b, the absorption peak of the pristine WSe2is significantly
higher (5060 meV) than that of either μW/M or cW/M, which might be due to that the WSe2is under MoS2
and the dielectric environment is largely changed.
Moreover, a Stoke shift is found in cM/W sample between absorption peak (1.64 eV) and interlayer PL peak (1.59 eV), which further supports the radia-tive recombination between the WSe2 and MoS2.
We should note that we found an anti-Stoke shift in the cM/W sample between the WSe2absorption peak
(1.64 eV) and WSe2PL peak (1.66 eV), as marked with
orange lines in Figure 4. The anti-Stoke shift implies that there might have some hot phonons in the optical level of WSe2 under the heterojunction structure,
which requires further experiments and theoretical work to clarify.
The observation on the interlayer PL peak not only show the strong evidence of the interlayer coupling, but also allow us to establish the energy band align-ment of the MoS2/WSe2heterojunction and the
ex-citon binding energy of the junction as shown in Figure 5b. According to our previous work on STS28 andμ-XPS,48we know that the energy band gaps for MoS2and WSe2are 2.34 and 2.46 eV, respectively, and
the valence band offset is 0.42 eV. For consistency, we measure the PL spectrum at 77K to set the energy band alignment. The PL spectrum at 77 K for each regime is shown in Figure 5a, and the extracted PL peaks are summarized in Figure 5b with solid double arrow lines. On the basis of these, we can determine the bind-ing energy of the MoS2/WSe2 interlayer exciton, as
marked by solid double arrow line with gray back-ground, to clarify the discrepancy raised form previous works.4952 We find that the MoS2/WSe2 interlayer
exciton has a lower exciton binding energy 0.26 eV when compared to MoS2or WSe2. The band alignment
reveals the important role of the interlayer coupling, which is crucial for band engineering. In our work, we confirmed that the annealing process can enhance the interlayer coupling, which might result from reducing of layer distance or exclusion of interface residues in order to build van der Waals solids. Besides, we have also performed the experiments for the opposite stack-ing structure (WSe2on top of MoS2) and this structure
exhibits the similar behaviors in Raman (Supporting Information Figure S5) and PL (Supporting Information Figure S6), further corroborating the conclusions reached.
CONCLUSIONS
We fabricated the vertical MoS2/WSe2
heterojunc-tion by stacking CVD-grown MoS2and WSe2triangular
monolayers. The thermal treatment enhances the cou-pling between the two monolayers, based on the shifts observed in the Raman and PL spectra. Interestingly, the characteristic Raman signature at 309 cm1(WSe2A21g),
463 cm1(MoS2A21g), and 286 cm1(MoS2E00) suggest
that the heterostructural stacking impose similar sym-metry change as homostructural stacking. Together with the reported STS andμ-XPS results, the energy band alignment of the MoS2/WSe2 heterojunction and the
Figure 5. Schematic of the energy band alignment of MoS2/
WSe2heterojunction at 77 K. (a) The PL of cM/W sample
(black) and the Lorentzian peaksfitting (red-dashed) at 77 K. The green lines are the composited peaks by using Lorent-zianfitting. The black dashed lines indicate the emitted energy of MoS2and WSe2and cM/W. (b) The solid lines
represent the conduction band and valence band of MoS2
and WSe2. The solid double arrow lines, dashed double
arrow lines, and dot double arrow lines indicate the PL energy, band gap energy, and the valence band offset between MoS2 and WSe2 extracted from experimental
results, respectively. The solid double arrow lines with gray background indicate the derived exciton binding energy.
exciton binding energy were established. It is antici-pated that the fundamental understanding of interlayer coupling and the band alignments in heterostructures
is crucially important for future applications and engi-neering of the devices based on two-dimensional ma-terials and van der Waals solids.
METHODS
WSe2/MoS2 Monolayer Single Crystal Growth. WSe2 and MoS2
monolayer single crystal were separately growth by chemical vapor deposit method. The WO3/MoO3powders (0.3/0.6 g) were
placed in a quartz boat located in the heating zone center of the furnace. The Se/S powders were placed in a separate quartz boat at the upper stream side of the furnace and the tempera-ture was maintained at 290/170C during the reaction. The sapphire substrates for growing were put at the downstream side, next to quartz boat. The gas flow was brought by an Ar/H2/Ar
flowing gas (Ar = 90 sccm, H2= 10 sccm/Ar = 70 sccm), and the
chamber pressure was controlled at 7/40 Torr. The center heating zone was heated to 925/635C. After reaching growth tempera-ture, the heating zone was kept for 15/30 min and the furnace was then naturally cooled down to room temperature.
Heterojunction Fabrication. MoS2 on sapphire substrate was
spin-coated with PMMA (950 K), and then heated on hot plate at 100C for 20 min. The sample was then tipped into NaOH (2 M) solution and kept at 100C for 1 h. After sapphire-etching by NaOH, MoS2with PMMA film was transferred into DI water
twice, each step for 10 min, to dilute NaOH etching solution. And then MoS2with PMMA was transferred on sapphire with
WSe2and heating at 100C on hot plate for 30 min to evaporate
the water at the interface. Following, the sample was put into acetone at 60C for 30 min to remove the PMMA. Finally, the sample was rinsed with IPA and water, and dried with N2gas.
Before annealing, the samples were examined by AFM, Raman spectroscopy, PL, and absorption spectroscopy. The as-prepared MoS2/WSe2 samples were then annealed in an hydrogen/Ar
environment (atmosphere pressure; H2:Ar = 1:4) at 300C for
4 h. The annealing temperature was selected at 300 C to effectively remove the residuals but not too high to initiate the formation of alloys such as MoSxSeyor WSxSey.53
Characterizations. The AFM images were performed in a Veeco Dimension-Icon system. Raman spectra were collected in a confocal Raman system (NT-MDT). The wavelength of laser is 473 nm (2.63 eV), and the spot size of the laser beam is∼0.5 μm and the spectral resolution is 1 cm1(obtained with a 1800 grooves/mm grating). The PL and differential reflectance spectra were measured in a homemade microscopy system. For room-temperature PL measurements, a 532 nm solid-state laser was focused to a spot size <1μm on the sample by an objective lens (100; N.A. = 0.9). The PL signals were then collected by the same objective lens, analyzed by a 0.75 m monochromator and detected by a liquid-nitrogen-cooled CCD camera. The appara-tus for the differential reflectance measurements are basically the same, except that the light source was replaced by a fiber-coupled tungsten-halogen lamp. For low-temperature PL measurements, the sample was cooled down to T = 10.8 K in a low-vibration cryogen-free cryostat. The objective lens for low-temperature measurements is a long working distance objective lens with N.A. = 0.42.
Phonon Dispersion and Raman Scattering Modes Simulation. The density functional theory (DFT) and density functional perturba-tion theory (DFPT) calculaperturba-tions were carried using the plane wave code CASTEP38as implemented in the Materials Studio
package. A hexagonal unit cell with one layer of MoS2 and
another of WSe2arranged in a AB stacking (similar to the one
observed in bulk crystals of trigonal prismatic transition metal dichalcogenides) was considered under the local density ap-proximation (LDA) using the Ceperly-Alder-Perdew and Zunger (CA-PZ) functional54,55 with 6X6X3 Monkhorst-Pack K-points
and a plane waves cut off of 720 eV with a norm-conserving pseudopotential. The structure was relaxed, including the cell, until the forces became smaller than 0.01 eV/Å and with self-consistent energy tolerances less than 5 107eV/atom. A vacuum of 16 Å between the cells was considered. The electronic
structure results obtained for the individual layers and bulk are in agreement with those reported by other groups within the LDA formalism.56,57To provide a reasonable description of the van der Waals interaction, we have considered a dispersion correction for both LDA (LDA-D)39and for general gradient approximation (GGA-PW91)58finding that LDA provides a better description
for the interlayer distance than LDA-D and corrected GGA-PW91, as described in a previous publications.35,59The phonon
disper-sion and Raman scattering modes were calculated using the linear response approach for insulators.40After the relaxation, due to the lattice mismatch between the monolayers, the MoS2
lattice parameter suffered an expansion from 3.16 to 3.21 Å and the WSe2lattice contracted from 3.28 to 3.21 Å. The expansion
of the MoS2and contraction of the WSe2explain in part the shifts
in some of the modes found in the Raman signal due to strain. Nevertheless, the results obtained here provide a good approx-imation of the Raman signal observed.
Conflict of Interest: The authors declare no competing financial interest.
Acknowledgment. This research was partly supported by National Science Council Taiwan (NSC-102-2119-M-001-005-MY3 and NSC101-2628-M-009-002-(NSC-102-2119-M-001-005-MY3) and AFOSR BRI. W.-H.C. acknowledges the supports from the Center for Interdisciplinary Science under the MOE-ATU project for NCTU. M.-H.C. and L.-J.L. acknowledge the support from KAUST. M.T. acknowledges sup-port by the U.S. Army Research Office under MURI ALNOS project, contract/grant number W911NF-11-1-0362, the Penn State Center for Nanoscale Science (seed grant on 2D Layered Materials -DMR-0820404), and the Center for 2-Dimensional and Layered Materials at The Pennsylvania State University.
Supporting Information Available: AFM image for the stacked area of the as-transferred MoS2/WSe2sample; Raman
and PL spectra for the WSe2 and MoS2 areas before and
after annealing process; schematics and AFM images of the WSe2triangularflake after the doping process of immersion in a
AuCl4solution; extracted peak positions of Raman A01mode
(upper) and E0mode of MoS2with twist angle between MoS2
and WSe2; Raman and PL spectra of the WSe2/MoS2
hetero-junction. This material is available free of charge via the Internet at http://pubs.acs.org.
REFERENCES AND NOTES
1. Lee, Y.-H.; Zhang, X.-Q.; Zhang, W.; Chang, M.-T.; Lin, C.-T.; Chang, K.-D.; Yu, Y.-C.; Wang, J. T.-W.; Chang, C.-S.; Li, L.-J.; et al. Synthesis of Large-Area MoS2Atomic Layers with
Chemical Vapor Deposition. Adv. Mater.2012, 24, 2320– 2325.
2. Huang, J.-K.; Pu, J.; Hsu, C.-L.; Chiu, M.-H.; Juang, Z.-Y.; Chang, Y.-H.; Chang, W.-H.; Iwasa, Y.; Takenobu, T.; Li, L.-J. Large-Area Synthesis of Highly Crystalline WSe2
Mono-layers and Device Applications. ACS Nano2013, 8, 923– 930.
3. Lin, Y.-C.; Zhang, W.; Huang, J.-K.; Liu, K.-K.; Lee, Y.-H.; Liang, C.-T.; Chu, C.-W.; Li, L.-J. Wafer-Scale MoS2 Thin Layers
Prepared by MoO3Sulfurization. Nanoscale2012, 4, 6637–
6641.
4. Mak, K. F.; Lee, C.; Hone, J.; Shan, J.; Heinz, T. F. Atomically Thin MoS2: A New Direct-Gap Semiconductor. Phys. Rev.
Lett.2010, 105, 136805.
5. Ramasubramaniam, A. Large Excitonic Effects in Mono-layers of Molybdenum and Tungsten Dichalcogenides. Phys. Rev. B2012, 86, 115409.
6. Lee, C.; Yan, H.; Brus, L. E.; Heinz, T. F.; Hone, J.; Ryu, S. Anomalous Lattice Vibrations of Single- and Few-Layer MoS2. ACS Nano2010, 4, 2695–2700.
7. Jones, A. M.; Yu, H.; Ghimire, N. J.; Wu, S.; Aivazian, G.; Ross, J. S.; Zhao, B.; Yan, J.; Mandrus, D. G.; Xiao, D.; et al. Optical Generation of Excitonic Valley Coherence in Monolayer WSe2. Nat. Nanotechnol.2013, 8, 634–638.
8. Splendiani, A.; Sun, L.; Zhang, Y.; Li, T.; Kim, J.; Chim, C.-Y.; Galli, G.; Wang, F. Emerging Photoluminescence in Mono-layer MoS2. Nano Lett.2010, 10, 1271–1275.
9. Wang, Q. H.; Kalantar-Zadeh, K.; Kis, A.; Coleman, J. N.; Strano, M. S. Electronics and Optoelectronics of Two-dimensional Transition Metal Dichalcogenides. Nat. Nano-technol.2012, 7, 699–712.
10. Wang, H.; Yu, L.; Lee, Y.-H.; Shi, Y.; Hsu, A.; Chin, M. L.; Li, L.-J.; Dubey, M.; Kong, J.; Palacios, T. Integrated Circuits Based on Bilayer MoS2Transistors. Nano Lett.2012, 12, 4674–
4680.
11. Radisavljevic, B.; Radenovic, A.; Brivio, J.; Giacometti, V.; Kis, A. Single-Layer MoS2Transistors. Nat. Nanotechnol.2011,
6, 147–150.
12. Radisavljevic, B.; Whitwick, M. B.; Kis, A. Small-Signal Amplifier Based on Single-Layer MoS2. Appl. Phys. Lett.
2012, 101, 043103.
13. Fang, H.; Chuang, S.; Chang, T. C.; Takei, K.; Takahashi, T.; Javey, A. High-Performance Single Layered WSe2p-FETs
with Chemically Doped Contacts. Nano Lett.2012, 12, 3788–3792.
14. Lee, Y.-H.; Yu, L.; Wang, H.; Fang, W.; Ling, X.; Shi, Y.; Lin, C.-T.; Huang, J.-K.; Chang, M.-T.; Chang, C.-S.; et al. Synthesis and Transfer of Single-Layer Transition Metal Disulfides on Diverse Surfaces. Nano Lett.2013, 13, 1852–1857. 15. Zhang, W.; Chuu, C.-P.; Huang, J.-K.; Chen, C.-H.; Tsai, M.-L.;
Chang, Y.-H.; Liang, C.-T.; Chen, Y.-Z.; Chueh, Y.-L.; He, J.-H.; et al. Ultrahigh-Gain Photodetectors Based on Atomically Thin Graphene-MoS2Heterostructures. Sci. Rep.2014, 4,
3826.
16. Bertolazzi, S.; Krasnozhon, D.; Kis, A. Nonvolatile Memory Cells Based on MoS2/Graphene Heterostructures. ACS
Nano2013, 7, 3246–3252.
17. Britnell, L.; Gorbachev, R. V.; Jalil, R.; Belle, B. D.; Schedin, F.; Mishchenko, A.; Georgiou, T.; Katsnelson, M. I.; Eaves, L.; Morozov, S. V.; et al. Field-Effect Tunneling Transistor Based on Vertical Graphene Heterostructures. Science 2012, 335, 947–950.
18. Geim, A. K.; Grigorieva, I. V. Van der Waals Heterostruc-tures. Nature2013, 499, 419–425.
19. Sup Choi, M.; Lee, G.-H.; Yu, Y.-J.; Lee, D.-Y.; Hwan Lee, S.; Kim, P.; Hone, J.; Jong Yoo, W. Controlled Charge Trapping by Molybdenum Disulphide and Graphene in Ultrathin Heterostructured Memory Devices. Nat. Commun.2013, 4, 1624.
20. Georgiou, T.; Jalil, R.; Belle, B. D.; Britnell, L.; Gorbachev, R. V.; Morozov, S. V.; Kim, Y.-J.; Gholinia, A.; Haigh, S. J.; Makarovsky, O.; et al. Vertical Field-effect Transistor based on Graphene-WS2Heterostructures for Flexible and
Trans-parent Electronics. Nat. Nanotechnol.2013, 8, 100–103. 21. Britnell, L.; Ribeiro, R. M.; Eckmann, A.; Jalil, R.; Belle, B. D.;
Mishchenko, A.; Kim, Y.-J.; Gorbachev, R. V.; Georgiou, T.; Morozov, S. V.; et al. Strong Light-Matter Interactions in Heterostructures of Atomically Thin Films. Science2013, 340, 1311–1314.
22. Haigh, S. J.; Gholinia, A.; Jalil, R.; Romani, S.; Britnell, L.; Elias, D. C.; Novoselov, K. S.; Ponomarenko, L. A.; Geim, A. K.; Gorbachev, R. Cross-sectional Imaging of Individual Layers and Buried Interfaces of Graphene-based Heterostruc-tures and Superlattices. Nat. Mater.2012, 11, 764–767. 23. Rivera, P.; Schaibley, J. R.; Jones, A. M.; Ross, J. S.; Wu, S.;
Aivazian, G.; Klement, P.; Ghimire, N. J.; Yan, J.; Mandrus, D. G.; et al. Observation of Long-Lived Interlayer Exci-tons in Monolayer MoSe2-WSe2Heterostructures. 2014,
ArXiv:1403.4985. arXiv.org e-Print arXiv. http://arxiv.org/ abs/1403.4985.
24. Cheng, R.; Li, D.; Zhou, H.; Wang, C.; Yin, A.; Jiang, S.; Liu, Y.; Chen, Y.; Huang, Y.; Duan, X. Electroluminescence and Photocurrent Generation from Atomically Sharp WSe2/
MoS2Heterojunction p-n Diodes. 2014, ArXiv:1403.3447.
arXiv.org e-Print arXiv. http://arxiv.org/abs/1403.3447.
25. Furchi, M. M.; Pospischil, A.; Libisch, F.; Burgdorfer, J.; Mueller, T. Photovoltaic Effect in an Electrically Tunable van der Waals Heterojunction. 2014, ArXiv:1403.2652. arXiv.org e-Print arXiv. http://arxiv.org/abs/1403.2652. 26. Lee, C.-H.; Lee, G.-H.; van der Zande, A. M.; Chen, W.; Li, Y.;
Han, M.; Cui, X.; Arefe, G.; Nuckolls, C.; Heinz, T. F.; et al. Atomically Thin p-n Junctions with van der Waals Hetero-interfaces. Nat. Nanotechnol.2014, 9, 676–681. 27. Fang, H.; Battaglia, C.; Carraro, C.; Nemsak, S.; Ozdol, B.;
Kang, J. S.; Bechtel, H. A.; Desai, S. B.; Kronast, F.; Unal, A. A.; et al. Strong Interlayer Coupling in van der Waals Hetero-structures Built from Single-Layer Chalcogenides. Proc. Natl. Acad. Sci. U.S.A.2014, 111, 6198–6202.
28. Zhang, C.; Johnson, A.; Hsu, C. L.; Li, L. J.; Shih, C. K. Direct Imaging of Band Profile in Single Layer MoS2on Graphite:
Quasiparticle Energy Gap, Metallic Edge States, and Edge Band Bending. Nano Lett.2014, 14, 2443–2447. 29. Luo, X.; Zhao, Y.; Zhang, J.; Toh, M.; Kloc, C.; Xiong, Q.;
Quek, S. Y. Effects of lower symmetry and dimensionality on Raman spectra in two-dimensional WSe2. Phys. Rev. B
2013, 88, 195313.
30. Tonndorf, P.; Schmidt, R.; Böttger, P.; Zhang, X.; Börner, J.; Liebig, A.; Albrecht, M.; Kloc, C.; Gordan, O.; Zahn, D. R. T.; et al. Photoluminescence Emission and Raman Response of Monolayer MoS2, MoSe2, and WSe2. Opt. Express2013,
21, 4908–4916.
31. Molina-Sánchez, A.; Wirtz, L. Phonons in Single-Layer and Few-Layer MoS2and WS2. Phys. Rev. B2011, 84, 155413.
32. Wieting, T. J.; Verble, J. L. Infrared and Raman Studies of Long-Wavelength Optical Phonons in Hexagonal MoS2.
Phys. Rev. B1971, 3, 4286–4292.
33. Rice, C.; Young, R. J.; Zan, R.; Bangert, U.; Wolverson, D.; Georgiou, T.; Jalil, R.; Novoselov, K. S. Raman-Scattering Measurements and First-Principles Calculations of Strain-induced Phonon Shifts in Monolayer MoS2. Phys. Rev. B
2013, 87, 081307.
34. Li, H.; Zhang, Q.; Yap, C. C. R.; Tay, B. K.; Edwin, T. H. T.; Olivier, A.; Baillargeat, D. From Bulk to Monolayer MoS2:
Evolution of Raman Scattering. Adv. Funct. Mater.2012, 22, 1385–1390.
35. Terrones, H.; Corro, E. D.; Feng, S.; Poumirol, J. M.; Rhodes, D.; Smirnov, D.; Pradhan, N. R.; Lin, Z.; Nguyen, M. A. T.; Elias, A. L.; et al. New First Order Raman-active Modes in Few Layered Transition Metal Dichalcogenides. Sci. Rep.2014, 4, 4215.
36. Zhang, X.; Han, W. P.; Wu, J. B.; Milana, S.; Lu, Y.; Li, Q. Q.; Ferrari, A. C.; Tan, P. H. Raman Spectroscopy of Shear and Layer Breathing Modes in Multilayer MoS2. Phys. Rev. B
2013, 87, 115413.
37. Zhao, W.; Ghorannevis, Z.; Amara, K. K.; Pang, J. R.; Toh, M.; Zhang, X.; Kloc, C.; Tan, P. H.; Eda, G. Lattice Dynamics in Mono- and Few-layer Sheets of WS2and WSe2. Nanoscale
2013, 5, 9677–9683.
38. Clark, S. J.; Segall, M. D.; Pickard, C. J.; Hasnip, P. J.; Probert, M. J.; Refson, K.; Payne, M. C. First Principles Methods using CASTEP. Z. Kristallogr.2005, 220, 567–570.
39. Ortmann, F.; Bechstedt, F.; Schmidt, W. G. Semiempirical van der Waals Correction to the Density Functional De-scription of Solids and Molecular Structures. Phys. Rev. B 2006, 73, 205101.
40. Refson, K.; Tulip, P. R.; Clark, S. J. Variational Density-Functional Perturbation Theory for Dielectrics and Lattice Dynamics. Phys. Rev. B2006, 73, 155114.
41. Wang, Y.; Cong, C.; Qiu, C.; Yu, T. Raman Spectroscopy Study of Lattice Vibration and Crystallographic Orienta-tion of Monolayer MoS2under Uniaxial Strain. Small2013,
9, 2857–2861.
42. Castellanos-Gomez, A.; Roldán, R.; Cappelluti, E.; Buscema, M.; Guinea, F.; van der Zant, H. S. J.; Steele, G. A. Local Strain Engineering in Atomically Thin MoS2. Nano Lett.2013, 13,
5361–5366.
43. Zhu, C. R.; Wang, G.; Liu, B. L.; Marie, X.; Qiao, X. F.; Zhang, X.; Wu, X. X.; Fan, H.; Tan, P. H.; Amand, T.; et al. Strain Tuning of Optical Emission Energy and Polarization in Monolayer and Bilayer MoS2. Phys. Rev. B2013, 88, 121301.
44. Shi, Y.; Huang, J.-K.; Jin, L.; Hsu, Y.-T.; Yu, S. F.; Li, L.-J.; Yang, H. Y. Selective Decoration of Au Nanoparticles on Mono-layer MoS2Single Crystals. Sci. Rep.2013, 3, 1839.
45. Chakraborty, B.; Bera, A.; Muthu, D. V. S.; Bhowmick, S.; Waghmare, U. V.; Sood, A. K. Symmetry-dependent Pho-non Renormalization in Monolayer MoS2transistor. Phys.
Rev. B2012, 85, 161403.
46. Liu, K.; Zhang, L.; Cao, T.; Jin, C.; Qiu, D.; Zhou, Q.; Zettl, A.; Yang, P.; Louie, S. G.; Wang, F. Evolution of Interlayer Coupling in Twisted MoS2Bilayers. ArXiv2014, 1406, 6487.
47. van der Zande, A. M.; Kunstmann, J.; Chernikov, A.; Chenet, D. A.; You, Y.; Zhang, X.; Huang, P. Y.; Berkelbach, T. C.; Wang, L.; Zhang, F.; et al. Tailoring the Electronic Structure in Bilayer Molybdenum Disulfide via Interlayer Twist. Nano Lett.2014, 14, 3869–3875.
48. Chiu, M.-H.; Zhang, C.; Shiu, H. W.; Chuu, C.-P.; Chen, C.-H.; Chang, C.-Y. S.; Chen, C.-H.; Chou, M.-Y.; et al. Determina-tion of Band Alignment in TransiDetermina-tion Metal Dichalcogen-ides Heterojunctions. ArXiv2014, 1406, 5137.
49. Ramasubramaniam, A. Large Excitonic effect in Monolayer of Molybdenum and Tungsten Dichalcogenides. Phys. Rev. B2012, 86, 115409.
50. Klots, A. R.; Newaz, A. K. M.; Wang, B.; Prasai, D.; Krzyzanowska, H.; Caudel, D.; Ghimire, N. J.; Yan, J.; Ivanov, B. L.; Velizhanin, K. A.; et al. Probing Excitonic States in Ultraclean Suspended Two-dimensional Semiconductors by Photocurrent Spectroscopy. 2014, ArXiv:1403.6455. arXiv.org e-Print arXiv. http://arxiv-web3.library.cornell. edu/abs/1403.6455v1.
51. Shi, H.; Pan, H.; Zhang, Y.-W.; Yakobson, B. I. Quasiparticle Band Structures and Optical Properties of Strained Mono-layer MoS2and WS2. Phys. Rev. B2013, 87, 155304.
52. Liang, Y.; Huang, S.; Soklaski, R.; Yang, L. Quasiparticle Band-edge Energy and Band Offsets of Monolayer of Molybdenum and Tungsten Chalcogenides. Appl. Phys. Lett.2013, 103, 042106.
53. Su, S.-H.; Hsu, W.-T.; Hsu, C.-L.; Chen, C.-H.; Chiu, M.-H.; Lin, Y.-C.; Chang, W.-H.; Suenaga, K.; He, J.-H.; Li, L.-J. Control-lable Synthesis of Band Gap-Tunable and Monolayer Transition Metal Dichalcogenide Alloys. Front. Energy Res.2014, 2, 27.
54. Ceperley, D. M.; Alder, B. J. Ground-State of The Electron-Gas by a Stochastic Method. Phys. Rev. Lett.1980, 45, 566– 569.
55. Perdew, J. P.; Zunger, A. Self-Interaction Correction to Density-Functional Approximations for Many-Electron Systems. Phys. Rev. B1981, 23, 5048–5079.
56. Ding, Y.; Wang, Y.; Ni, J.; Shi, L.; Shi, S.; Tang, W. First Principles Study of Structural, Vibrational and Electronic Properties of Graphene-like MX2(M = Mo, Nb, W, Ta; X = S,
Se, Te) Monolayers. Physica B2011, 406, 2254–2260. 57. Mahatha, S. K.; Patel, K. D.; Menon, K. S. R. Electronic
Structure Investigation of MoS2and MoSe2using
Angle-resolved Photoemission Spectroscopy and ab Initio Band Structure Studies. J. Phys.: Condens. Matter 2012, 24, 475504.
58. Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.; Pederson, M. R.; Singh, D. J.; Fiolhais, C. Atoms, Molecules, Solids, And Surfaces: Applications of the Generalized Gradient Approximation for the Exchange and Correla-tion. Phys. Rev. B1992, 46, 6671–6687.
59. Terrones, H.; Lopez-Urias, F.; Terrones, M. Novel Hetero-Layered Materials with Tunable Direct Band Gaps by Sandwiching Different Metal Disulfides and Diselenides. Sci. Rep.2013, 3, 1549.
