pubs.acs.org/JPCC Article Highly Biaxially Strained Silicene on Au(111) Daniele Nazzari, Jakob Genser, Viktoria Ritter, Ole Bethge, Emmerich Bertagnolli, Georg Ramer, Bernhard Lendl, Kenji Watanabe, Takashi Taniguchi, Riccardo Rurali, Miroslav Kolíbal, and Alois Lugstein* Cite This: J. Phys. Chem. C 2021, 125, 9973−9980 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: Many of graphene’s remarkable properties arise from its linear dispersion of the electronic states, forming a Dirac cone at the K points of the Brillouin zone. Silicene, the 2D allotrope of silicon, is also predicted to show a similar electronic band structure, with the addition of a tunable bandgap, induced by spin−orbit coupling. Because of these outstanding electronic properties, silicene is considered as a promising building block for next- generation electronic devices. Recently, it has been shown that silicene grown on Au(111) still possesses a Dirac cone, despite the interaction with the substrate. Here, to fully characterize the structure of this 2D material, we investigate the vibrational spectrum of a monolayer silicene grown on Au(111) by polarized Raman spectroscopy. To enable a detailed ex situ investigation, we passivated the silicene on Au(111) by encapsulating it under few layers hBN or graphene flakes. The observed spectrum is characterized by vibrational modes that are strongly red-shifted with respect to the ones expected for freestanding silicene. By comparing low-energy electron diffraction (LEED) patterns and Raman results with first-principles calculations, we show that the vibrational modes indicate a highly (>7%) biaxially strained silicene phase. 1. INTRODUCTION by applying a small voltage pulse of 3 V, implying a rather weak Silicene, the graphene-like 2D allotrope of silicon, has attracted interaction with the substrate.17 At the same time, however, great attention over the past few years because of its predicted others reported the formation of a highly biaxially strained outstanding electronic properties.1−3 Theoretical studies of the silicene phase, characterized by a lattice constant of 4.1 Å, band structure of freestanding silicene have described the indicating non-negligible interaction between substrate and presence of a Dirac cone,4 similarly to what can be observed in epitaxial layer,18 stronger than originally assumed. graphene.5 Additionally, spin−orbit coupling induces the opening of a bandgap that can be tuned via the application Here, we report the investigation of the structural properties of an external electric field.6,7 of 1 monolayer (ML) silicene grown on Au(111) by micro-low Experimentally, silicene has been successfully grown by electron energy diffraction (μ-LEED) and polarized Raman molecular beam epitaxy on different crystalline substrates.8−10 spectroscopy, enabled by an in situ passivation using few layers Most of the experimental studies have been realized by using hBN or graphene flakes. On the basis of symmetry Ag(111) as a substrate. Early experimental observations of the silicene/Ag(111) structure have reported signatures of a Dirac considerations, we can clearly assign the observed spectrum cone, despite a non-negligible interaction with the sub- to a 2D silicon structure. By comparing the obtained LEED strate.11,12 Subsequent studies have, however, pointed to an and Raman spectra with first-principles calculations, we show absence of such cone,13−15 showing that the interaction with that silicene grown on Au(111) is highly biaxially strained Ag(111) alters the electronic properties of the 2D silicon layer. (>7%). Recently, Au(111) has been explored as an alternative viable substrate for the growth of silicene with preserved Dirac Fermions. High-resolution angle-resolved photoemission spec- Received: December 10, 2020 troscopy experiments support the presence of a Dirac cone and Revised: April 22, 2021 a bandgap of ≈0.5 eV, with an estimated Fermi velocity of 106 Published: May 4, 2021 ms−1, similar to the one of graphene.16 Moreover, scanning tunneling microscopy experiments demonstrated that the 2D silicon layer grown on top of Au(111) can be easily removed © 2021 The Authors. Published byAmerican Chemical Society https://doi.org/10.1021/acs.jpcc.0c11033 9973 J. Phys. Chem. C 2021, 125, 9973−9980 The Journal of Physical Chemistry C pubs.acs.org/JPCC Article 2. EXPERIMENTAL DETAILS Subsequently, a microdiffraction image was collected at an Silicene on Au(111)/mica (MaTecK) is grown in a custom- electron energy of 30 eV, as shown in Figure 1. built UHV system (TU Vienna) at a base pressure of 5 × 10−11 mbar. The substrate is cleaned through three cycles of Ar+ ion sputtering (1 keV, 5 min), followed by annealing at 770 K for 15 min. One monolayer (ML) of silicon is evaporated from a rod by electron beam evaporation (EBE-1, SPECS) with a deposition rate of ≈0.02 ML/min, while keeping the substrate temperature at 533 K. The temperature is controlled by an infrared pyrometer (DGE-10N, DIAS; spectral range 2−2.6 μm) with a precision of ±2 K. Microdiffraction analysis is performed by using a low-energy electron microscope (LEEM, SPECS FE-LEEM P90), part of a complex UHV system at CEITEC, Brno University of Technology, where the samples were prepared in the same way as described above. To perform diffraction in LEEM, electrons are extracted from the field emission source and accelerated to 15 keV. After passing the electron optics, they are decelerated by a strong electric field in the vicinity of the sample to very low energies (0−40 eV). The diffraction patterns are collected by using 30 eV electrons from areas selected by microdiffraction aperture (spot size ranging from 15 μm down to 185 nm). To enable detailed polarized micro-Raman characterization, silicene was passivated in situ immediately after the growth by mechanical exfoliation of few-layer hBN or graphene flakes Figure 1. Microdiffraction pattern (left half of the image) of the (HQ Graphene)19 in a dedicated UHV chamber, directly grown layer, acquired at an electron energy of 30 eV. Two different connected to the evaporation chamber. diffraction models are presented in the right half of the image. The red dots represent the diffracted beam from the Au(111) substrate. The Ex situ Raman analysis is performed in a backscattering upper right quarter model assumes the presence of two surface silicide geometry by using two different confocal micro-Raman setups. phases (green and purple). Some of the brightest unexplained dots of The first is an Alpha300 (WITec) stage equipped with a 532 the microdiffraction pattern (indicated by the arrows) are only nm (2.33 eV) Nd:YAG laser and a grating monochromator accounted for by a silicene-derived phase (black), as shown in the with 1800 grooves/mm. The laser is focused onto the sample lower right quarter. by a Nikon Epiplan 100× objective, enabling a spot size of ≈700 nm. The second setup (Alpha500, WITec) is capable of A well-reproducible, complex diffraction pattern was detecting vibrational modes with a frequency as low as 75 cm−1 observed alongside the signal associated with the Au(111) and is equipped with a grating monochromator with 1200 substrate (red dots in Figure 1). The pattern cannot be grooves/mm and a 532 nm (2.33 eV) Nd:YAG laser, focused explained by a simple single-superstructure model, implying by a Zeiss EC Epiplan-Neofluar 100× objective down to a spot the presence of several mixed phases. The appearance of a size of ≈800 nm. All measurements were performed under multiphase diffraction pattern is not surprising: in a similar ambient conditions, and moderate laser power, to avoid system1 ML of silicon on Ag(111)the diffraction pattern heating effects. is always reflecting the presence of multiple silicene phases due The samples analyzed by LEEM and Raman spectroscopy to the very similar formation energies.24,25 The very same are grown by using the same process as well as employing the microdiffraction pattern was detected all over the substrate, same substrates and the same evaporation material. To further suggesting that the grown layer fully covers the sample surface. confirm that the samples are identical, the ones grown in We have not observed any changes in the diffraction pattern Vienna are further analyzed by using LEED (ErLEED 1000A, even if with the smallest microdiffraction aperture (≈185 nm), Specs). suggesting that the possible domains of each phase are much Density-functional calculations are performed with the smaller. ABINIT code20 to obtain the ground-state geometry in We have compared the diffraction pattern with a series of different strain configurations to compute the phonon modes models using the software LEEDLab,26 as shown in Figure 1. A at the Γ point and the corresponding Raman susceptibility first model, displayed in the upper right quarter, assumes an tensors.21,22 We used norm-conserving pseudopotentials, a overlayer composed of two phases: the first, represented in plane-wave cutoff of 48 hartrees, the local density approx- green, is characterized by a monoclinic unit cell with a size of imation for the exchange-correlation energy, and a 22 × 22 k- 4.25 Å × 5.41 Å and γ = 95.5° and is rotated by 22° with point grid to sample the Brillouin zone. Further details about respect to the Au(111) substrate, while the second (purple) the Raman calculations can be found, for example, in ref 23. has a monoclinic unit cell of size 2.94 × 3.66 Å2 and γ = 91.1°, with a rotation of 30°. The model also accounts for double- scattering events with the substrate. 3. RESULTS AND DISCUSSION These 2D crystalline phases were identified by Shpyrko et Immediately after the cleaning of the Au(111) surface, 1 ML of al.27 as surface silicides, with a stoichiometry of Au4Si2 and silicon was evaporated at a substrate temperature of 533 K. AuSi2. The authors also observed the formation of a Au4Si8 9974 https://doi.org/10.1021/acs.jpcc.0c11033 J. Phys. Chem. C 2021, 125, 9973−9980 The Journal of Physical Chemistry C pubs.acs.org/JPCC Article silicide, characterized by an orthorhombic 7.386 × 9.386 Å2 Based on this argumentation, the study of the polarization unit cell with a rotation of 19° with respect to the Au(111) dependency of the Raman spectrum is extremely important substrate16,28 which cannot be seen in the pattern of Figure 1. because the identification of a fully polarized Raman mode Interestingly, the formation of these phases was observed at the (invisible in crossed polarization) could not be explainable surface of a molten Au82Si18 ingot at temperatures above 632 considering only the presence of the aforementioned silicides K. In our case, however, the growth temperature is much phases, but it would be, instead, a clear indication of the lower: nevertheless, as evidenced by the diffraction pattern, it presence of a higher symmetry phase. leads to the formation of some of these silicide phases as well. To perform detailed ex situ polarized micro-Raman analysis However, not all the diffraction spots can be explained by of the grown 2D silicon layer, it is necessary to protect it from the current model. The ones indicated by the red arrows in oxidation, which otherwise occurs very rapidly under ambient Figure 1 are characterized by a high intensity, indicating a conditions.34 In our previous work, we have presented a crystalline phase with moderate to high coverage. In the lower- method for passivating silicene layers by encapsulating them by right quarter of Figure 1 we show that these spots can be few-layer graphene (FLG) flakes.19 The passivation, performed explained by considering a phase characterized by a rectangular directly in UHV, is based on the mechanical exfoliation of unit cell (4.141 × 7.101 Å2, rotated 30° with respect to the Au graphite on top of the freshly grown silicene layer. The FLG substrate), which, as we will later show, can be related to a strained silicene layer. flake forms an exceptional barrier against oxidative species and, The appearance of 2D hexagonal silicon phases on Au(111) thanks to its inert surface, preserves silicene structure has been already reported.16,18 They are, however, charac- unaltered. We have recently extended this procedure to use terized by a high number of defects, resulting in deformations also few-layer hBN flakes, resulting in a highly transparent and of the lattice, which could lead to the appearance of additional insulating encapsulation layer. It is important to mention that diffraction spots. This, in combination with the appearance of the samples used for the Raman measurements are not the diffraction spots due to multiscattering processes, could pose same as the one used for LEEM analysis. However, the growth an obstacle to the precise determination of the 2D-Si structure has been performed under the same conditions. To further recurring only to the microdiffraction data. confirm that the samples are identical, the ones used for Polarized micro-Raman spectroscopy is an ideal tool for a Raman analysis are also investigated by using LEED. The precise detection and analysis of a hexagonal 2D-Si phase on obtained pattern (Figure S1) shows that the structure is Au(111). It was shown that a bulk gold−silicide crystalline identical with the one analyzed by using LEEM. The blurriness phase does not give rise to detectable Raman peaks, likely due of the pattern is caused by a slight mosaicity of the Au(111) to the lack of Raman-active vibrational modes,29 and is thus substrates that affects the large-area measurements performed incapable of hindering the signal of a silicene phase. 2D silicide by using a classic LEED apparatus. phases, instead, could show active Raman modes due to the Figure 2a shows an unpolarized Raman spectrum, acquired modified crystalline structure. Recently, it has been shown that in backscattering geometry, 30 min after in situ passivation with after depositing Au on Si(111) and annealing the sample, hBN and subsequent removal from the UHV system. The Raman peaks could be detected, which are associated with the observed spectrum is identical with the one taken from a formation of 2D silicides.30,31 This is where the study of the sample encapsulated under few-layer graphene flakes (Figure polarization dependency makes a difference: the unit cells of S2), suggesting that the observed Raman peaks are not the 2D silicides that can be observed in the LEEM pattern and associated with interlayer vibrational modes, which can be that were previously described in the literature in great observed in some van der Waals stacks.35 detail27,28 possess an extremely low symmetry. The two phases The spectrum shows four main peaks at 83, 100, 435, and that we can detect from the diffraction pattern have a 1360 cm−1, with the last one assigned to the well-known E2g monoclinic unit cell (γ ≠ 90°). Therefore, even if we assume mode of multilayer hBN.36 Interestingly, no peak can be found that all the atoms within the unit cell lay in the same plane (zero buckling condition), the highest achievable symmetry, at 520 cm−1shown in more detail in Figure S3clearly for all of these phases, is represented by point group C2. ruling out the presence of detectable crystalline, sp3-hybridized, The Raman tensors associated with C2 structures are of two silicon. types:32 It must be noted that at some points on the sample two ij a d 0 yz ij 0 0 e yz additional peaks can be recorded between 100 and 435 cm−1, jj zz jj zz j z A(z) = jjjj d b 0 zzzz, B(x , y) = jjj 0 0 f zzz positioned at 181 and 304 cm−1, both characterized by a low jj zz jj z jj e f 0 zzz intensity (Figure S4, upper spectrum, red arrows). These k0 0 c { k { peaks, however, are not always detectable and are showing up only at certain locations. A spectrum with a better signal-to- Raman selection rules dictate that in a backscattering geometry noise ratio, obtained by averaging 30 acquisitions, clearly B-type phonon modes are always invisible, while A-type modes shows that these peaks are not observable, while the 100 and are visible both in parallel and in crossed polarization 435 cm−1 peaks still show up at the usual positions, with configuration. unchanged width and intensity (Figure S4, lower spectrum). Differently, if we consider a crystalline structure with a This clearly proves that the two additional peaks are decoupled higher symmetry, we can observe polarized Raman modes that from the 100 and 435 cm−1 ones and belong to a different are visible in parallel polarization configuration but completely structure, possibly a signature of the observed silicide phases. suppressed in crossed polarization. This is true for a hexagonal The broader shoulders visible in Figure S4 at 260 and 360 lattice, even in the case of buckling, and it also holds upon cm−1 are a possible signature of amorphous structures, as uniaxial deformations, as it happens for black phosphorus, hinted by the much larger FWHM compared to the peaks at whose structure belongs to point group D2h.33 100 and 435 cm−1. 9975 https://doi.org/10.1021/acs.jpcc.0c11033 J. Phys. Chem. C 2021, 125, 9973−9980 The Journal of Physical Chemistry C pubs.acs.org/JPCC Article centered at 429 ± 2 and 437.0 ± 0.5 cm−1, with a FWHM of 11 ± 4 and 7 ± 2 cm−1, respectively. An additional broad peak centered around 410 cm−1 is related either to phonon confinement effects or to a vibrational mode that was observed also for silicene on Ag(111) and, in that case, assigned to a second-order phonon mode.37 The low-frequency modes can be fitted with a single Lorentzian function, centered at 82.8 ± 0.2 and 100.4 ± 0.2 cm−1, with a FWHM of 5.0 ± 0.5 and 6.5 ± 0.5 cm−1. No difference can be observed when fitting the peaks of the spectrum related to silicene capped under few-layer graphene flakes, both regarding their positions and the FWHM values. The intensities of the peaks in the graphene-capped sample are lower as the few-layer graphene flake is less transparent than the hBN one. Clearly, the Raman peaks at 100.4 and at 437 cm−1 show a strong polarization dependency. Particularly the lower frequency mode is completely suppressed in crossed polar- ization configuration, with a reduced intensity within the noise level. For the reasons explained earlier, this fully polarized Raman peak is a clear indication of the existence of a high symmetry phase in the sample, as it cannot be generated by the silicide phases. These peaks appear to be strongly red-shifted, if compared to the Raman peaks predicted for freestanding silicene, where the modes are expected at 175 and 566 cm−1,38−40 or to the ones observed for silicene on Ag(111), where the out-of-plane mode is centered at 216 cm−1 and in- Figure 2. (a) Unpolarized Raman spectrum of the grown layer, plane modes are detected at 515 cm−1.37,41,42 This denotes a protected from oxidation by few-layer hBN. In addition to the well- known vibrational mode for boron nitride at 1360 cm−1, three other substantial structural modification that is likely induced by a modes can be distinguished at 83, 100, and 435 cm−1. The dashed rather strong interaction with the substrate. gray line indicates the position of the well-known 520 cm−1 Raman It must be noted that a softening (i.e., red-shift) of the in- mode of bulk, sp3-hybridized Si. (b, c) Polarized Raman spectra of the plane vibrational modes implies longer Si−Si bonds. Notably, grown layer encapsulated by few-layer hBN, acquired in back- it was recently observed that a 2D hexagonal Si phase on scattering geometry in parallel (top, red line) and cross-polarization Au(111) is characterized by much larger lattice constant (4.1 (bottom, blue line). Low-frequency modes can be fitted by a single Å, compared to 3.81 Å for strain-free silicene).18 Lorentzian function (purple and orange dashed lines), while the high- Thus, to confirm this hypothesis of a strained 2D silicene frequency peak is fitted with a combination of two Lorentzian phase, we performed ab initio calculations of the vibrational functions (green and yellow dashed lines). The two panels (b, c) are separated because the spectra are acquired with different spectrom- modes of a silicene structure with a lattice constant of 4.1 Å. At eters. first, we fully characterized a freestanding layer of silicene, obtaining a lattice parameter of 3.810 Å, a buckling of 0.425 Å, and Γ-point phonons of 175 and 566 cm−1. Next, we To fully characterize the most intense, and always visible reoptimized the atomic positions, constraining the lattice vibrational modes, we performed a polarized Raman analysis of parameter of silicene to the value of 4.1 Å, corresponding to a the sample. Our results are shown in Figure 2b,c. biaxial tensile strain of 7.6%. In these highly strained In both panels, the upper part shows the signal acquired in conditions, the buckling is strongly reduced to 0.237 Å, parallel polarization configuration, where the polarization although, differently from the flat geometry reported in ref 18, direction of the scattered light is parallel to the one of the it does not vanish. The relative phonon dispersion plots are incident light. The lower half, instead, shows the signal in shown in Figure S5. The computed Raman spectra, in both the crossed polarization configuration, where the polarization parallel and crossed polarization geometry, are shown as vectors are normal to each other. The lower frequency region dashed black lines in Figure 3a,b. The simulated structure that can be analyzed only by using the Alpha 500 Witec setup, gives rise to these spectra is shown in Figure 3c, described by a albeit with a much worse signal-to-noise ratio and lower rhomboidal unit cell characterized by an edge length of 4.1 Å spectral resolution. The higher frequency region, instead, is and γ = 60°. acquired with the Alpha 300 Witec setup, which allows us to The positions of the simulated modes are in good agreement obtain a higher spectral resolution and a markedly better with the experimental data, with a calculated out-of-plane signal-to-noise ratio for this spectral region. For this reason, the mode centered at 98.6 cm−1 and the two degenerate in-plane analysis of the higher frequency peak will be performed by modes centered at 447.3 cm−1. The experimentally observed using these data. mode centered at 83 cm−1, instead, does not find any match First, we can note that the 435 cm−1 peak is asymmetric. within the proposed model and could be related to edge- or From the crossed polarization data, it is clear that it is defect-activated vibrational modes. composed of two different, slightly separated modes. A least- Additionally, to verify whether such a structure can be squares fitting process allows us to model the modes with accommodated on gold, we placed the strained silicene layer Lorentzian functions. The results show that the two peaks are on top of an ideally terminated Au(111) surfacesomething 9976 https://doi.org/10.1021/acs.jpcc.0c11033 J. Phys. Chem. C 2021, 125, 9973−9980 The Journal of Physical Chemistry C pubs.acs.org/JPCC Article significantly change, improving the agreement with the experimental data. The out-of-plane mode exhibits a small blue-shift, up to 102.9 cm−1. The in-plane vibrational modes get red-shifted and split up in two distinct peaks, as the degeneration is lifted, at 433.9 and 439.5 cm−1. The simulated spectrum obtained by adding 1% additional strain exhibits also a polarization dependency, with the higher frequency mode showing a more pronounced intensity reduction in crossed polarization configuration as compared to the lower frequency one, similar to what can be observed experimentally. Despite showing a good agreement regarding the frequency of the peaks and the polarization dependence, the simulations predict a much lower intensity for the out-of-plane vibrational mode (which is magnified ∼3000 times in Figure 3a) compared to the in-plane ones. We attribute this discrepancy to the presence of the substrate, not accounted for in the theoretical calculations. Au(111)/mica is a SERS-active substrate,45 and it is well-known that the amplification factors of different vibrational modes can noticeably vary.46 Such an amplification is expected to be particularly effective for an out- of-plane vibration, which could couple well with the metallic substrate. Figure 3d depicts how uniaxial strain modifies the lattice of the biaxially strained silicene shown in Figure 3c. For a better visualization of the structural modifications, the image shows the effects of 20% uniaxial strain. The modified structure can Figure 3. (a, b) Simulated Raman spectra in parallel and cross- be optimally described by choosing a rectangular unit cell, with polarization for biaxially strained (7.6%) silicene (black dotted line) dimensions of 4.141 × 7.1 Å2, which possesses all the and for biaxially strained silicene with a superimposed additional symmetry properties of the deformed lattice. The correspond- uniaxial strain of 1% (solid red and blue lines). The doubly degenerate ing diffraction pattern, represented by black circles in the lower in-plane mode splits into two separated peaks (green and yellow right quarter of Figure 1, explains the intense spots, indicated dashed lines). (c) Simulated structure of biaxially strained silicene. by red arrows, that were not addressed by the silicide-only The rhomboidal unit cell (black solid line) edge length is 4.1 Å, with γ model. = 60°. Red atoms are buckled along the out-of-plane direction by 0.237 Å. (d) Simulated structure of biaxially strained silicene with a Based on these results, Au(111)/mica is clearly a promising superimposed additional uniaxial strain. For clarity and visualization platform to study the effects of strain on silicene. The effects of purpose, the image represents a lattice deformed by a uniaxial strain of biaxial tensile strain on the silicene electronic structure have 20%. The unit cell dimensions (4.141 × 7.1 Å2) are relative to the already been explored in the literature by using first-principles simulated 1% uniaxial strain. calculations. The results of the calculations show that despite the huge mechanical deformation, the Dirac cone is that we could not perform when calculating the Raman preserved.47,48 For high biaxial strain levels, the tip of the response, due to computational limitationsand relaxed the Dirac cone shifts above the Fermi energy, inducing the atomic positions until all forces were lower than 0.02 eV/Å. A formation of hole-doped Dirac states. The level of biaxial strain stable configuration is indeed readily obtained, supporting the that induces this transition varies among different studies, picture that such a biaxially strained silicene structure can be ranging from 6%47 to 8%.48 Au(111) could represent a perfect stabilized by the Au(111) substrate. platform for the investigation of such effect, as the strain level It must be noted that the experimental data show a splitting of the detected silicene phase is very close to those values. of the in-plane modes, with the 437 cm−1 peak characterized Future investigations will focus on probing the electronic by a much stronger polarization dependence. The theoretical structure of silicene grown on Au(111)/mica, searching for a calculations, however, predict that the two in-plane modes are confirmation of its Dirac nature. degenerate, with no change in intensity between parallel and crossed polarization configurations, as expected given the 4. CONCLUSIONS symmetry of the lattice. This loss of degeneracy could be We have presented a study of the vibrational properties of a caused by a small deviation from the perfect hexagonal lattice silicene layer epitaxially grown on Au(111). The diffraction due to an additional uniaxial distortion of the structure, pattern can be explained by a combination of two surface similarly to what can be observed in uniaxially strained silicide phases and a strained silicene-derived phase. graphene.43 This could be induced by the peculiar properties A passivation step, obtained by encapsulating the grown of Au(111), which is known to form a complex surface layer under few-layer hBN or graphene flakes, allowed us to reconstructionthe herringbone reconstructioncharacter- perform polarized Raman spectroscopy. The observed vibra- ized by an uniaxial deformation along one of the ⟨11̅0⟩ tional spectrum is characterized by polarization-dependent directions.44 For this reason, we additionally imposed a small peaks that, for symmetry reasons, cannot be related to the uniaxial strain (1%) to the previously simulated structure along silicide structures. Hence, the detected modes are a clear the zigzag direction (see Figure 3d) and monitored the Raman indication for the presence of a higher symmetry phase. Using response. Under the additional strain the vibrational modes first-principles calculations, we identified this phase as a highly 9977 https://doi.org/10.1021/acs.jpcc.0c11033 J. Phys. Chem. C 2021, 125, 9973−9980 The Journal of Physical Chemistry C pubs.acs.org/JPCC Article (>7%) biaxially strained silicene phase, slightly deformed along Author Contributions one direction. D.N., J.G., and V.R. contributed equally to the growth and The present work gives clear experimental evidence on the passivation of silicene, the Raman measurements, the data existence of a stable, highly strained silicene phase on Au(111), analysis, and the writing of the manuscript; R.R. performed the indicating a favorable platform for the investigation of the ab initio calculations and helped shaping the manuscript; M.K. effects of strain on this promising 2D material. acquired the microdiffraction spectrum and contributed to the ■ discussion of the results; G.R. and B.L. provided the ASSOCIATED CONTENT equipment for the low-frequency Raman measurements and contributed to the design of the Raman experiment; K.W. and * Supporting Information sı T.T. provided the hBN crystals for the silicene passivation and The Supporting Information is available free of charge at contributed to the polishing of the manuscript; E.B. https://pubs.acs.org/doi/10.1021/acs.jpcc.0c11033. contributed to the writing of the manuscript; O.B. and A.L. S1: additional LEED pattern; S2: additional Raman conceived the project and contributed essentially to the spectrum of silicene/Au(111) encapsulated under a few- experimental design and to the writing of the manuscript. layer graphene flake; S3: Raman signal near sp3-silicon Funding peak position; S4: additional Raman spectrum showing This work was funded by the Fonds zur Förderung der low-intensity peaks; S5: phonon dispersion curves Wissenschaftlichen Forschung (FWF), Austria (Project (PDF) P29244-N27). We also acknowledge financial support by the Ministerio de Economia,́ Industria y Competitividad (MINE- ■ AUTHOR INFORMATION Corresponding Author CO) under Grant FEDER-MAT2017-90024-P and the Severo Ochoa Centres of Excellence Program under Grant CEX2019- 000917-S, and by the Generalitat de Catalunya under Grant Alois Lugstein − Institute of Solid State Electronics, 2017 SGR 1506. The project i-LINK action LINKA20047 Technische Universität Wien, 1040 Vienna, Austria; funded by CSIC is also acknowledged for financial support. orcid.org/0000-0001-5693-4775; Email: alois.lugstein@ R.R. acknowledges useful discussions with Mariusz Krawiec. tuwien.ac.at CzechNanoLab Project LM2018110 funded by MEYS CR is gratefully acknowledged for the financial support of the Authors measurements at CEITEC Nano Research Infrastructure. Daniele Nazzari − Institute of Solid State Electronics, K.W. and T.T. acknowledge support from the Elemental Technische Universität Wien, 1040 Vienna, Austria; Strategy Initiative conducted by the MEXT, Japan, Grant orcid.org/0000-0003-4267-3142 JPMXP0112101001, JSPS KAKENHI Grant JP20H00354, and Jakob Genser − Institute of Solid State Electronics, Technische the CREST(JPMJCR15F3), JST. Universität Wien, 1040 Vienna, Austria; orcid.org/0000- 0003-0234-0917 Notes Viktoria Ritter − Institute of Solid State Electronics, The authors declare no competing financial interest. Technische Universität Wien, 1040 Vienna, Austria; orcid.org/0000-0002-7877-0860 Ole Bethge − Infineon Technologies Austria AG, 9500 Villach, ■ ACKNOWLEDGMENTS We thank the Center for Micro and Nanostructures (ZMNS) Austria at the TU Vienna for access to the cleanroom facilities. We also Emmerich Bertagnolli − Institute of Solid State Electronics, thank the Centro de Supercomputación de Galicia (CESGA) Technische Universität Wien, 1040 Vienna, Austria for the use of their computational resources. Georg Ramer − Institute of Chemical Technologies and Analytics, Technische Universität Wien, 1060 Vienna, Austria; orcid.org/0000-0001-8307-5435 ■ REFERENCES (1) Tao, L.; Cinquanta, E.; Chiappe, D.; Grazianetti, C.; Fanciulli, Bernhard Lendl − Institute of Chemical Technologies and M.; Dubey, M.; Molle, A.; Akinwande, D. Silicene Field-Effect Analytics, Technische Universität Wien, 1060 Vienna, Transistors Operating at Room Temperature. Nat. Nanotechnol. 2015, Austria; orcid.org/0000-0003-3838-5842 10 (3), 227−231. Kenji Watanabe − Research Center for Functional Materials, (2) Le Lay, G. Silicene Transistors. Nat. Nanotechnol. 2015, 10 (3), National Institute for Materials Science, Tsukuba 305-0044, 202−203. Japan; orcid.org/0000-0003-3701-8119 (3) Zhao, J.; Liu, H.; Yu, Z.; Quhe, R.; Zhou, S.; Wang, Y.; Liu, C. Takashi Taniguchi − International Center for Materials C.; Zhong, H.; Han, N.; Lu, J.; Yao, Y.; Wu, K. Rise of Silicene: A Competitive 2D Material. Prog. Mater. Sci. 2016, 83, 24−151. Nanoarchitectonics, National Institute for Materials Science, (4) Cahangirov, S.; Topsakal, M.; Aktürk, E.; Ş ahin, H.; Ciraci, S. Tsukuba 305-0044, Japan; orcid.org/0000-0002-1467- Two- and One-Dimensional Honeycomb Structures of Silicon and 3105 Germanium. Phys. Rev. Lett. 2009, 102 (23), 236804. Riccardo Rurali − Institut de Ciència de Materials de (5) Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.; Barcelona, ICMAB-CSIC, 08193 Bellaterra, Spain; Katsnelson, M. I.; Grigorieva, I. V.; Dubonos, S. V.; Firsov, A. A. Two- orcid.org/0000-0002-4086-4191 Dimensional Gas of Massless Dirac Fermions in Graphene. Nature Miroslav Kolíbal − Institute of Physical Engineering, Brno 2005, 438 (7065), 197−200. University of Technology, 616 69 Brno, Czech Republic; (6) Drummond, N. D.; Zólyomi, V.; Fal’ko, V. I. Electrically Tunable CEITEC BUT, Brno University of Technology, 612 00 Brno, Band Gap in Silicene. Phys. Rev. B: Condens. Matter Mater. Phys. 2012, Czech Republic; orcid.org/0000-0002-2751-5608 85 (7), 075423. (7) Ni, Z.; Liu, Q.; Tang, K.; Zheng, J.; Zhou, J.; Qin, R.; Gao, Z.; Complete contact information is available at: Yu, D.; Lu, J. Tunable Bandgap in Silicene and Germanene. Nano Lett. https://pubs.acs.org/10.1021/acs.jpcc.0c11033 2012, 12 (1), 113−118. 9978 https://doi.org/10.1021/acs.jpcc.0c11033 J. Phys. Chem. C 2021, 125, 9973−9980 The Journal of Physical Chemistry C pubs.acs.org/JPCC Article (8) Oughaddou, H.; Enriquez, H.; Tchalala, M. R.; Yildirim, H.; (24) Moras, P.; Mentes, T. O.; Sheverdyaeva, P. M.; Locatelli, A.; Mayne, A. J.; Bendounan, A.; Dujardin, G.; Ait Ali, M.; Kara, A. Carbone, C. Coexistence of Multiple Silicene Phases in Silicon Grown Silicene, a Promising New 2D Material. Prog. Surf. Sci. 2015, 90 (1), on Ag(111). J. Phys.: Condens. Matter 2014, 26 (18), 185001. 46−83. (25) Pflugradt, P.; Matthes, L.; Bechstedt, F. Silicene-Derived Phases (9) Kara, A.; Enriquez, H.; Seitsonen, A. P.; Lew Yan Voon, L. C.; on Ag(111) Substrate versus Coverage: Ab Initio Studies. Phys. Rev. Vizzini, S.; Aufray, B.; Oughaddou, H. A Review on Silicene  New B: Condens. Matter Mater. Phys. 2014, 89 (3), 035403. Candidate for Electronics. Surf. Sci. Rep. 2012, 67 (1), 1−18. (26) Sojka, F.; Meissner, M.; Zwick, C.; Forker, R.; Fritz, T. (10) Lalmi, B.; Oughaddou, H.; Enriquez, H.; Kara, A.; Vizzini, S.; Determination and Correction of Distortions and Systematic Errors in Ealet, B.; Aufray, B. Epitaxial Growth of a Silicene Sheet. Appl. Phys. Low-Energy Electron Diffraction. Rev. Sci. Instrum. 2013, 84 (1), Lett. 2010, 97 (22), 223109. 015111. (11) Vogt, P.; De Padova, P.; Quaresima, C.; Avila, J.; Frantzeskakis, (27) Shpyrko, O. G.; Streitel, R.; Balagurusamy, V. S. K.; Grigoriev, E.; Asensio, M. C.; Resta, A.; Ealet, B.; Le Lay, G. Silicene: A. Yu; Deutsch, M.; Ocko, B. M.; Meron, M.; Lin, B.; Pershan, P. S. Compelling Experimental Evidence for Graphenelike Two-Dimen- Crystalline Surface Phases of the Liquid Au-Si Eutectic Alloy. Phys. sional Silicon. Phys. Rev. Lett. 2012, 108 (15), 155501. Rev. B: Condens. Matter Mater. Phys. 2007, 76 (24), 245436. (12) Chen, L.; Liu, C.-C.; Feng, B.; He, X.; Cheng, P.; Ding, Z.; (28) Shpyrko, O. G.; Streitel, R.; Balagurusamy, V. S. K.; Grigoriev, Meng, S.; Yao, Y.; Wu, K. Evidence for Dirac Fermions in a A. Y.; Deutsch, M.; Ocko, B. M.; Meron, M.; Lin, B.; Pershan, P. S. Honeycomb Lattice Based on Silicon. Phys. Rev. Lett. 2012, 109 (5), Surface Crystallization in a Liquid AuSi Alloy. Science 2006, 313 056804. (5783), 77−80. (13) Lin, C.-L.; Arafune, R.; Kawahara, K.; Kanno, M.; Tsukahara, (29) Ashtikar, M. S.; Sharma, G. L. Structural Investigation of Gold N.; Minamitani, E.; Kim, Y.; Kawai, M.; Takagi, N. Substrate-Induced Induced Crystallization in Hydrogenated Amorphous Silicon Thin Symmetry Breaking in Silicene. Phys. Rev. Lett. 2013, 110 (7), 076801. Films. Jpn. J. Appl. Phys. 1995, 34 (10R), 5520. (14) Wang, Y.-P.; Cheng, H.-P. Absence of a Dirac Cone in Silicene (30) Liebhaber, M.; Halbig, B.; Bass, U.; Geurts, J.; Neufeld, S.; on Ag(111): First-Principles Density Functional Calculations with a Sanna, S.; Schmidt, W. G.; Speiser, E.; Räthel, J.; Chandola, S.; Esser, Modified Effective Band Structure Technique. Phys. Rev. B: Condens. N. Vibration Eigenmodes of the Au-(5 × 2)/Si(111) Surface Studied Matter Mater. Phys. 2013, 87 (24), 245430. by Raman Spectroscopy and First-Principles Calculations. Phys. Rev. (15) Mahatha, S. K.; Moras, P.; Bellini, V.; Sheverdyaeva, P. M.; B: Condens. Matter Mater. Phys. 2016, 94 (23), 235304. Struzzi, C.; Petaccia, L.; Carbone, C. Silicene on Ag(111): A (31) Halbig, B.; Liebhaber, M.; Bass, U.; Geurts, J.; Speiser, E.; Honeycomb Lattice without Dirac Bands. Phys. Rev. B: Condens. Räthel, J.; Chandola, S.; Esser, N.; Krenz, M.; Neufeld, S.; Schmidt, Matter Mater. Phys. 2014, 89 (20), 201416. W. G.; Sanna, S. Vibrational Properties of the Au-(√3x√3)/Si(111) (16) Sadeddine, S.; Enriquez, H.; Bendounan, A.; Kumar Das, P.; Surface Reconstruction. Phys. Rev. B: Condens. Matter Mater. Phys. Vobornik, I.; Kara, A.; Mayne, A. J.; Sirotti, F.; Dujardin, G.; 2018, 97 (3), 035412. Oughaddou, H. Compelling Experimental Evidence of a Dirac Cone (32) Aroyo, M. I.; Perez-Mato, J. M.; Capillas, C.; Kroumova, E.; Ivantchev, S.; Madariaga, G.; Kirov, A.; Wondratschek, H. Bilbao in the Electronic Structure of a 2D Silicon Layer. Sci. Rep. 2017, 7 (1), Crystallographic Server: I. Databases and Crystallographic Computing 44400. (17) Tang, L.; Li, F.; Guo, Q. A Structured Two-Dimensional Au−Si Programs. Z. Kristallogr. - Cryst. Mater. 2006, 221 (1), 15−27. (33) Ribeiro-Soares, J.; Almeida, R. M.; Cançado, L. G.; Dresselhaus, Alloy. Appl. Surf. Sci. 2011, 258 (3), 1109−1114. M. S.; Jorio, A. Group Theory for Structural Analysis and Lattice (18) Stpniak-Dybala, A.; Dyniec, P.; Kopciuszyski, M.; Zdyb, R.; Vibrations in Phosphorene Systems. Phys. Rev. B: Condens. Matter Jałochowski, M.; Krawiec, M. Planar Silicene: A New Silicon Mater. Phys. 2015, 91 (20), 205421. Allotrope Epitaxially Grown by Segregation. Adv. Funct. Mater. (34) Solonenko, D.; Selyshchev, O.; Zahn, D. R. T.; Vogt, P. 2019, 29 (50), 1906053. Oxidation of Epitaxial Silicene on Ag(111). Phys. Status Solidi B 2019, (19) Ritter, V.; Genser, J.; Nazzari, D.; Bethge, O.; Bertagnolli, E.; 256 (2), 1800432. Lugstein, A. Silicene Passivation by Few-Layer Graphene. ACS Appl. (35) Lui, C. H.; Ye, Z.; Ji, C.; Chiu, K.-C.; Chou, C.-T.; Andersen, T. Mater. Interfaces 2019, 11 (13), 12745−12751. I.; Means-Shively, C.; Anderson, H.; Wu, J.-M.; Kidd, T.; Lee, Y.-H.; (20) Gonze, X.; Jollet, F.; Abreu Araujo, F.; Adams, D.; Amadon, B.; He, R. Observation of Interlayer Phonon Modes in van Der Waals Applencourt, T.; Audouze, C.; Beuken, J.-M.; Bieder, J.; Bokhanchuk, Heterostructures. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 91 A.; Bousquet, E.; Bruneval, F.; Caliste, D.; Côté, M.; Dahm, F.; Da (16), 165403. Pieve, F.; Delaveau, M.; Di Gennaro, M.; Dorado, B.; Espejo, C.; (36) Gorbachev, R. V.; Riaz, I.; Nair, R. R.; Jalil, R.; Britnell, L.; Geneste, G.; Genovese, L.; Gerossier, A.; Giantomassi, M.; Gillet, Y.; Belle, B. D.; Hill, E. W.; Novoselov, K. S.; Watanabe, K.; Taniguchi, Hamann, D. R.; He, L.; Jomard, G.; Laflamme Janssen, J.; Le Roux, S.; T.; Geim, A. K.; Blake, P. Hunting for Monolayer Boron Nitride: Levitt, A.; Lherbier, A.; Liu, F.; Lukačević, I.; Martin, A.; Martins, C.; Optical and Raman Signatures. Small 2011, 7 (4), 465−468. Oliveira, M. J. T.; Poncé, S.; Pouillon, Y.; Rangel, T.; Rignanese, G.- (37) Solonenko, D.; Gordan, O. D.; Lay, G. L.; Ş ahin, H.; M.; Romero, A. H.; Rousseau, B.; Rubel, O.; Shukri, A. A.; Stankovski, Cahangirov, S.; Zahn, D. R. T.; Vogt, P. 2D Vibrational Properties M.; Torrent, M.; Van Setten, M. J.; Van Troeye, B.; Verstraete, M. J.; of Epitaxial Silicene on Ag(111). 2D Mater. 2017, 4 (1), 015008. Waroquiers, D.; Wiktor, J.; Xu, B.; Zhou, A.; Zwanziger, J. W. Recent (38) Yan, J.-A.; Stein, R.; Schaefer, D. M.; Wang, X.-Q.; Chou, M. Y. Developments in the ABINIT Software Package. Comput. Phys. Electron-Phonon Coupling in Two-Dimensional Silicene and Commun. 2016, 205, 106−131. Germanene. Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 88 (21) Baroni, S.; de Gironcoli, S.; Dal Corso, A.; Giannozzi, P. (12), 121403. Phonons and Related Crystal Properties from Density-Functional (39) Gori, P.; Pulci, O.; de Lieto Vollaro, R.; Guattari, C. Perturbation Theory. Rev. Mod. Phys. 2001, 73 (2), 515−562. Thermophysical Properties of the Novel 2D Materials Graphene (22) Gonze, X.; Vigneron, J.-P. Density-Functional Approach to and Silicene: Insights from Ab-Initio Calculations. Energy Procedia Nonlinear-Response Coefficients of Solids. Phys. Rev. B: Condens. 2014, 45, 512−517. Matter Mater. Phys. 1989, 39 (18), 13120−13128. (40) Li, X.; Mullen, J. T.; Jin, Z.; Borysenko, K. M.; Buongiorno (23) De Luca, M.; Cartoixà, X.; Martín-Sánchez, J.; López-Suárez, Nardelli, M.; Kim, K. W. Intrinsic Electrical Transport Properties of M.; Trotta, R.; Rurali, R.; Zardo, I. New Insights in the Lattice Monolayer Silicene and MoS2 from First Principles. Phys. Rev. B: Dynamics of Monolayers, Bilayers, and Trilayers of WSe2 and Condens. Matter Mater. Phys. 2013, 87 (11), 115418. Unambiguous Determination of Few-Layer-Flakes’ Thickness. 2D (41) Zhuang, J.; Xu, X.; Du, Y.; Wu, K.; Chen, L.; Hao, W.; Wang, J.; Mater. 2020, 7 (2), 025004. Yeoh, W. K.; Wang, X.; Dou, S. X. Investigation of Electron-Phonon 9979 https://doi.org/10.1021/acs.jpcc.0c11033 J. Phys. Chem. C 2021, 125, 9973−9980 The Journal of Physical Chemistry C pubs.acs.org/JPCC Article Coupling in Epitaxial Silicene by in Situ Raman Spectroscopy. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 91 (16), 161409. (42) Molle, A.; Grazianetti, C.; Chiappe, D.; Cinquanta, E.; Cianci, E.; Tallarida, G.; Fanciulli, M. Hindering the Oxidation of Silicene with Non-Reactive Encapsulation. Adv. Funct. Mater. 2013, 23 (35), 4340−4344. (43) Mohiuddin, T. M. G.; Lombardo, A.; Nair, R. R.; Bonetti, A.; Savini, G.; Jalil, R.; Bonini, N.; Basko, D. M.; Galiotis, C.; Marzari, N.; Novoselov, K. S.; Geim, A. K.; Ferrari, A. C. Uniaxial Strain in Graphene by Raman Spectroscopy: G Peak Splitting, Grüneisen Parameters, and Sample Orientation. Phys. Rev. B: Condens. Matter Mater. Phys. 2009, 79 (20), 205433. (44) Hanke, F.; Björk, J. Structure and Local Reactivity of the Au(111) Surface Reconstruction. Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 87 (23), 235422. (45) Caldwell, W. B.; Chen, K.; Herr, B. R.; Mirkin, C. A.; Hulteen, J. C.; Van Duyne, R. P. Self-Assembled Monolayers of Ferrocenyla- zobenzenes on Au(111)/Mica Films: Surface-Enhanced Raman Scattering Response vs Surface Morphology. Langmuir 1994, 10 (11), 4109−4115. (46) Lombardi, J. R.; Birke, R. L. A Unified Approach to Surface- Enhanced Raman Spectroscopy. J. Phys. Chem. C 2008, 112 (14), 5605−5617. (47) Kaloni, T. P.; Cheng, Y. C.; Schwingenschlögl, U. Hole Doped Dirac States in Silicene by Biaxial Tensile Strain. J. Appl. Phys. 2013, 113 (10), 104305. (48) Umam, K.; Sholihun; Nurwantoro, P.; Ulil Absor, M. A.; Nugraheni, A. D.; Budhi, R. H. S. Biaxial Strain Effects on the Electronic Properties of Silicene: The Density-Functional-Theory- Based Calculations. J. Phys.: Conf. Ser. 2018, 1011, 012074. 9980 https://doi.org/10.1021/acs.jpcc.0c11033 J. Phys. Chem. C 2021, 125, 9973−9980