Second-harmonic generation and photoluminescence properties of colloidal WS₂ monolayers deposited from solution

Abstract

Monolayered WS₂ exhibits a direct band gap, strongly bound excitons, trions and biexcitons as well as no inversion symmetry in the 2D crystal structure (D_3h), allowing frequency doubling by second-harmonic generation. The recent progress in colloidal synthesis of transition metal dichalcogenides enables the scalable production of phase-pure monolayer-dominated WS₂. In this work, we use parabolic mirror-assisted microscopy to investigate the linear and nonlinear optical properties of deposited colloidal WS₂. Spatial and spectral resolution of one- and two-photon photoluminescence and second-harmonic generation correlates structural motifs of the WS₂ layers including orientation and agglomeration to the local optical response. We find exciton and trion emission at 613 nm and 639 nm respectively, with significantly varying contribution to the overall intensity between one- and two-photon excitation. This is attributed to the increased carrier density under high fluences needed for two-photon excitation. By evaluating the second-harmonic generation intensity, the nonlinear susceptibility of colloidally synthesized WS₂ nanosheets is calculated to range from 4.28–7.55 nm V⁻¹, averaging higher than a commercial WS₂ monolayer reference sample. Our results show that colloidal WS₂ is suitable for large area coverage creating an isotropic WS₂ film, while retaining the highly sought after linear and nonlinear optical properties of WS₂ monolayers, even under surface ligand separated aggregation.

---

New concepts

Research on transition metal dichalcogenides (TMDCs) is dominated by individual monolayers (MLs) of typically exfoliated and/or CVD-grown WS₂ flakes. Building on our recent advances in the wet-chemical synthesis of TMDCs, organic-ligand separated WS₂ nanosheet (NS) MLs are synthesized. We then use parabolic mirror-assisted microscopy to characterize the photoluminescence (PL), two-photon photoluminescence and second-harmonic generation (SHG) of deposited colloidal WS₂ NSs with spatial resolution, allowing us to correlate the local structure with spectral and temporal information of colloidal TMDCs for the first time. Depending on the polarized excitation, the orientation of individual WS₂ NSs within the ensemble is revealed. By phasor plot analysis of PL lifetimes, we disentangle emission stemming from WS₂ NSs and residual ligands. WS₂ NSs are characterized by HR-TEM, Raman-, absorption and PL spectroscopy, underpinning that they retain ML properties in solution and when deposited. Highly efficient second-harmonic generation (SHG, 4.28–7.55 nm V⁻¹) from colloidally synthesized WS₂ NSs exceeds benchmark CVD samples and shows that they are solution-processable and scalable, enabling their future integration into fiber optics or biological systems as contrast agents for enhanced resolution SHG-microscopy. Characterizing the properties, applicability and limitations of wet-chemically synthesized WS₂ NSs is the first step for colloidal TMDCs as innovative nonlinear materials.

Introduction

WS₂ has been studied for its wide range of possible applications from catalysis over sensing to a variety of different electronic and optoelectronic applications including transistors and lasers.^2–5 WS₂ shows strong excitonic absorption and emission at room temperature with high binding energies (up to 0.7 eV) and an indirect (bulk) to direct band gap (2.02 eV) transition, when monolayer (ML) thickness is reached.^6–9 The optical features of WS₂ MLs in the ultraviolet and visible (UV-Vis) spectral range exhibit four excitonic transitions. The direct transitions A and B located at 614 and 517 nm emerge from split valence bands at the K point in the Brillouin zone, while at higher energies the indirect C and D transitions can be observed at 436 and 364 nm, resp.^10–12 Photoluminescence (PL) spectra of WS₂ MLs are complex with several contributions from A and B excitons, trions and/or biexcitons.^13–16 While previous studies found the neutral A excitons to recombine mainly radiatively and give rise to one-photon PL, non-radiative recombination is commonly observed from excitonic complexes.¹⁷ Sulfur vacancies are ubiquitous defects in WS₂ resulting in negative doping and therefore facilitate the formation of negative trions as well as Auger recombination, which suppresses PL and thereby limits the application of colloidal WS₂ nanosheets (NSs).^18–20 Nonlinear optical properties of WS₂ such as two-photon photoluminescence (2PPL) have also been explored.^21–23 Applying 2PPL, Ye et al. revealed excitonic dark states in WS₂ MLs, which cannot be probed by one photon PL.²⁴ Second-harmonic generation (SHG) is a higher-order light–matter interaction process where an incoming fundamental electromagnetic wave with a frequency ω is converted via a virtual state to light of double the frequency 2ω (see Fig. S1 in the SI). The SHG intensity depends on how well the phase matching condition is met, thus requiring a material without inversion symmetry.^10,25–27 When reduced from bulk (D_6h symmetry) to ML thickness (D_3h symmetry), WS₂ shows a remarkably high SHG, varying greatly between odd and even layered samples.^10,28 Laterally translated or defected crystal structures as well as strain and twist angles break the symmetry of even-layers in 2H-phase TMDCs, hence also allowing SHG to emerge from an originally non-SHG active system.^29,30

So far, 2D TMDC mono- and few layers are typically obtained by top-down processes, previously established for graphene by exfoliating the monolayers using the scotch tape method or liquid phase exfoliation.^31–33 Bottom-up methods include chemical vapor deposition (CVD) and wet-chemical synthesis of WS₂ MLs.³⁴ The latter enables a scalable production of solution processed colloidal NSs, while maintaining high control over the crystal phase, size and layer number as well as the tunability of optical properties and the elemental composition.^35–40 Recently, the linear optical properties of colloidal TMDCs (2D WS₂ and MoS₂ NSs) and laterally smaller nanoplatelets have been characterized at room- and cryogenic temperatures, underpinning the ability to wet-chemically synthesize photoluminescent TMDC NSs.^38,41,42 However, the nonlinear optical properties of colloidal WS₂ NSs lack investigation up to now.

Here, we describe the 2PPL and SHG properties of wet chemically synthesized WS₂ NSs in detail and evaluate the relationship between local structure motifs and the optical response under radial-, azimuthal- and linear-excitation polarization. We determine the nonlinear susceptibility (NLS) of colloidal WS₂ NSs based on their power-dependent SHG intensity and compare it to a CVD grown WS₂ flake. By applying phasor plot analysis to fluorescence lifetime data, PL contributions of excess ligands are separated from the PL of WS₂. We find that colloidal WS₂ NSs exhibit optical properties comparable to CVD grown flakes, with added scalability and flexible application to different substrates/host-matrices. Our results paint a promising picture for future implementation of colloidal WS₂ NSs in nonlinear optoelectronics such as optical modulators, optical switching and high-density storage devices.^43–46

Results and discussion

The colloidal WS₂ NSs characterized here are synthesized wet-chemically by the dropwise addition of the tungsten-precursor consisting of WCl₆ dissolved in oleylamine (OlAm) to a pre-heated solution of elemental sulfur/OlAm under inert conditions as has been described in our previous work (detailed information on the synthesis and sample preparation as well as the chemicals used can be found in the SI, Section A and B).^39,42

To confirm the synthesis of predominantly monolayered 2H-phase WS₂ NSs, we use a combination of (high-resolution) transmission electron microscopy (HR-TEM), scanning electron microscopy (SEM) and optical spectroscopy (UV-vis absorption, PL and Raman), covering the microscopic to the ensemble level range. HR-TEM in Fig. 1a shows an example of a flat lying ML as well as different twisted bilayers (Moiré pattern). The fast Fourier transform (FFT) insets in the left and right corners show the hexagonal pattern of the semiconducting WS₂ 2H-phase around the (001) crystal plane axis. Due to agglomeration during the drying process on the substrate, so-called nanoflowers are formed on the μm scale as is shown by SEM in the inset in Fig. 1a. These porous structures consist of a multitude of individual NSs each sized 18.4 ± 6 nm as is evident from the HR-TEM images in Fig. 1a and Fig. S3b (the size distribution is given in Fig. S2). The solvent (OlAm) can also act as a L-type ligand and helps to keep individual WS₂ NSs from stacking into bulk material in solution and in the deposited film. Examples of ligand-separated WS₂ NSs are shown in Fig. S3a. Here, the inter layer distance within the nanoflower varies. In contrast to the bulk distance of WS₂ bilayers of 0.3 nm,⁴⁷ typically closely aligned NSs are 1–3 nm apart, underpinning efficient ligand separation.

Fig. 1 (a) HR-TEM of drop-casted colloidal WS₂ NSs with SEM inset showing agglomerated NSs on the substrate in lower magnification. The FFT insets show a hexagonal WS₂ structure for the ML area as well as the Moiré area. (b) Absorption of colloidal WS₂ NSs in 1,2-dichlorobenzene showing A, B, C and D excitonic shoulders (black) and second derivative of the absorption plotted for better visibility of the C and D excitonic features (blue). (c) Fitted Raman spectrum of drop-casted WS₂ NSs measured under linear laser polarization with E¹_2g and A_1g modes found at WS₂ ML wavenumbers.¹

Fig. 1b displays the absorbance spectrum of colloidal WS₂ NSs in solution (see Fig. S4 for deposited sample). The A, B, C and D excitonic transitions (614 nm (2.02 eV); 517 nm (2.40 eV); 436 nm (2.84 eV) and 364 nm (3.41 eV)) are observed. The second derivative of the absorbance spectrum is plotted for better visibility and spectral separation of C and D excitonic absorption. Raman spectroscopic measurements confirm the ML thickness of the colloidal WS₂ NSs to be retained in the dried state as well as the associated 2H-phase purity shown in Fig. 1c. The spectrum is dominated by first-order modes including the longitudinal acoustic mode LA(M) at 179.4 cm⁻¹ at the M point in the Brillouin zone, which is related to the in-plane collective movements of the atoms in the lattice, while the A_1g feature at 415.4 cm⁻¹ corresponds to the out-of-plane phonon mode for two adjacent layers.^1,48 The strongest peak at 352.3 cm⁻¹ consists of two subpeaks, which are identified by multipeak Gaussian fitting to correspond to the in-plane vibrational E¹_2g(M) mode, and the second-order longitudinal acoustic phonon 2LA(M) mode. The subpeaks between ∼290 and 330 cm⁻¹ consist of several unassigned peaks beneath the spectral resolution limit of our setup (and are not vital for the following discussion). The frequencies of the E¹_2g and A_1g modes match WS₂ ML values,¹ while the intensity at ∼352.0 cm⁻¹ is roughly twice that of the A_1g mode. The high intensity at ∼352.0 cm⁻¹ is due to the 2LA(M) acoustic mode, which only arises in ML WS₂ and is matching the frequency of the E¹_2g, thereby adding onto the peak at ∼352.0 cm⁻¹.⁴⁸ This shows a substantial ML content in the colloidal WS₂ NS ensemble and compliments the microscopic evaluation from individual NSs by TEM. The removal of excess OlAm ligand (as is described in the SI, Section B), is assessed by phasor plot analysis of the lifetime distribution from the as-synthesized and subsequently washed WS₂ NSs (see Fig. S5 for a schematic description). The presences of excess OlAm, shown as green areas in Fig. S6a, are clearly identified in the fluorescence lifetime phasor plot analysis as a long lifetime component. The successful removal of excess OlAm ligand from the WS₂ NSs after repeated rinsing with hexane and isopropanol is confirmed in Fig. S6b by the general absence of long lifetime components. The shorter lifetime of WS₂ NSs is highlighted red.

To investigate the relationship between the one-photon PL and local morphology of WS₂ nanoflowers, polarized excitation experiments were performed using the home-built parabolic mirror-assisted optical setup shown in Fig. S7. Additionally to the commonly used linear polarization, radially polarized donut mode (RPDM) and azimuthally polarized donut mode (APDM) were also applied, as they are well-suited for studying materials with preferred dipole orientations due to their distinct electric field distributions in the focal plane (Fig. S8).^49,50 In the focus of a linearly polarized beam (Fig. S8a) both transverse components and longitudinal components are present. RPDM on the other hand, exhibits a dominant longitudinal component E_z perpendicular to the focal plane (Fig. S8b), hence it can be used to efficiently excite the out-of-plane dipole moment. In contrast, APDM produces only a transverse field parallel to the focal plane E_x,y (Fig. S8c), which is ideal for exciting in-plane oriented dipole moments. The dark-field scattering image in Fig. 2a shows the unordered agglomeration of many single drop-casted WS₂ NSs into nanoflowers. We correlate this dark-field optical image with the PL maps shown in Fig. 2c and d in which the PL intensity distribution is inhomogeneous under excitation with RPDM and APDM. Though the electric field distribution in the focus of RPDM and APDM is distinctively different, the PL intensity of WS₂ NS nanoflowers is rather comparable. We hence infer an isotropic arrangement of WS₂ NSs in the agglomerated structure. However, some regions indeed show excitation-dependent PL (marked by the white arrows in Fig. 2d). These areas excited by the APDM show higher PL intensity than the corresponding regions excited by RPDM in Fig. 2c, which indicates the presence of many or exceptionally bright flat-lying WS₂ NSs traced by the E_x,y electric component present in the focus in APDM.

Fig. 2 (a) Dark-field optical image of a WS₂ NSs agglomerate. (b) PL spectrum of WS₂ nanoflowers collected under APDM with fitted subpeaks, which can be attributed to X_A, X_A⁻, and X_D. (c) and (d) PL intensity maps of WS₂ NS agglomerate shown in (a) under RPDM and APDM excitation, resp. White arrows highlight spots that are particularly bright under APDM excitation and indicate possibly flat-lying NSs with increased polarization-dependent PL intensity in APDM.

In our previous work, we described the first μ-PL of wet chemically synthesized WS₂ MLs.⁴² Here, we use spatially, temporally and spectroscopically resolved methods to further disentangle the single PL contributions. By exciting the A and B excitonic transitions with a 532 nm continuous wave laser, μ-PL is detected at 613 nm (2.02 eV, see Fig. 2b). This observation is consistent with experimentally observed and calculated optical band gaps for WS₂ MLs.^6,24,51 To assign the origin of the PL signal, we extract the excitonic subpeaks by Gaussian fits.^52,53 The result in Fig. 2b contains three fitted subpeaks at 613 nm (2.02 eV), 639 nm (1.94 eV) and 670 nm (1.85 eV), resp. Previous studies by Cui et al. and Kato et al. on WS₂ allow for the assignment of the peaks to a neutral exciton (X_A), a negatively charged trion (X_A⁻) and defect related excitons (X_D) originating from sulfur vacancies and exposed vacant sites on the edges of the WS₂ NSs.^54–58 To enhance the statistical validity, we compare excitonic peaks in other sample regions (see Fig. S9). In all cases, the emission of X_A dominates the PL spectra with the trion PL as the second most intense contribution. The overall intensity may vary from spot to spot (as is evident from the maps in Fig. 2c and d) due to the heterogeny of the nanoflower structure, but the spectral features of the individual NSs stay consistent.

Further, we excited the same WS₂ NS agglomerate with high intensity femtosecond laser pulses to study its nonlinear optical response including 2PPL and SHG. Fig. 3a shows the 2PPL signal of a WS₂ NS agglomerate. The origin of the PL here is exclusively nonlinear because the excitation energy (779 nm, 1.59 eV) is lower than the bandgap of ML WS₂ (2.02 eV). Compared to linearly polarized and RPDM excitation, the 2PPL image excited by APDM shows a particularly high intensity and a distinctly inhomogeneous intensity distribution, indicating the presence of many or exceptionally bright flat-lying WS₂ NSs selectively excited by the APDM. Fig. 3b shows the SHG of WS₂ NSs under three different excitation polarizations. While the distribution of the SHG intensity under linear polarization (LP) or RPDM excitation is quite homogeneous, the SHG under APDM appears slightly more intense in the regions where the 2PPL is high (see Fig. 3a for comparison). Since SHG is strongly dependent on the local crystal structure and orientation of the WS₂ NSs, we assume an increased SHG signal results from flat-lying WS₂ NSs by APDM excitation. Fig. 3c clearly shows a sharp SHG peak at 389.5 nm, which is the frequency-doubled fundamental excitation laser wavelength (779 nm), as well as the 2PPL signal together with a broad background feature. We fitted the background with a third order polynomial to be able to extract the excitonic contributions of the 2PPL in Fig. 3d. In comparison to the PL spectrum (Fig. 2b), three subpeaks corresponding to 613 nm (X_A), 639 nm (X_A⁻) and 670 nm (X_D) are fitted to the 2PPL. Interestingly, unlike in the one-photon PL (Fig. 2b), where the neutral excitonic feature dominates, the 2PPL spectra are dominated by trion emission for most sample regions (seen in S10). Comparing these findings with literature, a similar result is found also for exfoliated WS₂ MLs.⁵⁹ A reason for this could be the high peak power of the femtosecond pulsed laser used for the two-photon absorption experiments, which may result in a higher charge carrier density, thereby increasing the likelihood of trion formation.^60,61

Fig. 3 (a) 2PPL intensity maps of WS₂ NS an agglomerate under LP, RPDM and APDM, resp. (b) SHG intensity maps of WS₂ NSs an agglomerate under LP, RPDM and APDM, resp. (c) 2PPL and SHG spectrum as collected by the nonlinear setup under APDM excitation. A background polynomial is fitted to allow subsequent fitting of the 2PPL contributions. (d) Fitted 2PPL spectrum (extracted from 3c) with three peaks associated to X, X_A⁻, and X_D.

Finally, to quantify the nonlinear susceptibility (NLS), we followed the theoretical surface SHG formalism of Shen et al. and converted equations from Woodward et al.'s work.^62,63 Previously we have used the same method to calculate the NLS of Nb₃SeI₇ flakes.⁶⁴ The precondition that the light is at normal incidence to the sample and the contribution by the nonlinearity of air or substrate (mainly the effect from dispersion) is neglected are applied.

The specific equations used in the NLS calculation are:

where ε₀ is the permittivity of vacuum, c is the speed of light in vacuum, n₂ is the refractive index (1.45–1.50), ω is the pump angular frequency, I₁(ω) is the focused pump peak intensity in air, |χ_s⁽²⁾| is the magnitude of the sheets' susceptibility for second-order nonlinearity, f is the pump laser repetition rate, S is a shape factor for the Gaussian pulse applied in the setup, t_FWHM is the pulse full width at half maximum, and P₁(ω) is the average pump power.

Since the parameters related to the setup configuration are fixed during all susceptibility measurements, the equations can be reduced to eqn (3) (detailed derivation process and specific parameters are shown in eqn (S9), (S10) and (S11) in the SI):

The ratio of NLS is calculated by the power-dependent SHG intensity measurements on the WS₂ NSs and CVD grown WS₂ flakes (optical characterization of the CVD grown WS₂ flakes can be found in Fig. S11). According to work by Janisch et al.,⁶⁵ CVD grown WS₂ on a SiO₂/Si substrate exhibit intense NLS, as high as 4.50 nm V⁻¹ at 832 nm, which we use as reference value.

For the power-dependent SHG measurements, we employed excitation powers ranging from 0.8 to 8 mW corresponding to 52.1 to 521 GW cm⁻² of in-focus peak-power, considering a Gaussian beam distribution of the femtosecond pulsed laser system under the Rayleigh diffraction limit condition (calculated in the SI, Section B). Fig. 4a shows that the SHG intensity of all WS₂ samples studied is proportional to the squared excitation power, agreeing well with eqn (2). For each power, five spots are measured and averaged. Their deviation is indicated by the error bars. The magnitude of the slopes ranges from 116 to 141 for our WS₂ NSs and from 84 to 120 for the CVD sample, showing the significantly higher nonlinear response of the colloidally synthesized WS₂ NS sample (Fig. 4b). We observed the NSs to be subject to damage (as indicated by the muted color datapoints in Fig. 4a) at high power above 40 mW². Since the CVD-grown reference sample did not show signs of damage, we suspect the damage to originate from burning (carbon build-up) or evaporation of the OlAm ligands that stabilize the NSs due to local heating by the high intensity laser. Note that the observed damage threshold at >36 mW² (391 GW cm⁻²) is beyond usual operating power, as is evident from similar reports,^66,67 which mitigates applicability risks. Following eqn (3), we can derive the relative NLS of the deposited colloidal WS₂ NSs. Comparing the slope values and taking the reference NLS value of CVD-grown WS₂ MLs (4.50 nm V⁻¹) into account,⁶⁵ the NLS lies in the range of 4.28 to 7.55 nm V⁻¹.

Fig. 4 (a) Power-dependent SHG intensity collected on deposited WS₂ NSs. At high intensity, the WS₂ NSs exhibit signs of photodamage (b) calculated slopes of WS₂ NSs and CVD grown WS₂ flakes shown in (a) with deposited WS₂ NSs exhibiting a higher response on average.

Conclusions

We have characterized the linear and nonlinear optical properties of colloidal WS₂ NSs samples, which are ML-dominated as is shown by HR-TEM, absorption, Raman and PL spectroscopy. Polarization-dependent PL measurements indicate that the emission of WS₂ NSs is dominated by neutral excitons (X_A) with an intensity distribution which is non-uniform but comparable under RPDM and APDM excitation, which suggests the WS₂ NSs to be isotropically oriented. The same concept applies to the intense SHG of WS₂ NSs, which is only weakly influenced by polarized excitation. However, differences in the local crystal structure and dipole orientation still cause minor polarization-related changes. The 2PPL signal of colloidal WS₂ NSs, in contrast to the one-photon PL exhibits an increased amount of trion emission (X_A⁻), which could be caused by the high intensity femtosecond pulsed laser excitation. Other contributions stemming from excess OlAm ligand adsorbed to the nanoflower surface were identified and removed using phasor plot analysis of the PL lifetime data.

Drop-casted colloidal WS₂ MLs exhibit high NLS (4.28 to 7.55 nm V⁻¹), thereby achieving an equal or higher NLS response than comparable CVD grown single WS₂ MLs and rendering them a scalable solution-processed SHG material alternative.

Author contributions

Synthesis, characterization (UV-Vis Abs-, Raman-, SEM-, TEM-evaluation) and sample preparation was done by M. F. with initial experiments by O. S. and both supervised by J. L. Low res. TEM images were obtained by A. N. under supervision by J. L. High res. TEM images were obtained by M. K. under supervision by J. C. M. Y. Z. performed the PL, 2PPL, FLIM, SHG and Raman experiments and the related data analysis. T. P. and F. S. assisted the linear and nonlinear optical measurements. D. Z. supervised the PL, 2PPL FLIM, and SHG experiments and the data analysis. Y. Z., M. F., D. Z. and J. L. conceptualized the manuscript. All authors commented on the draft. A. J. M., J. M., D. Z. and J. L. obtained the funding.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article has been included as part of the supplementary information (SI). Supplementary information: additional figures and data. See DOI: https://doi.org/10.1039/d5nh00515a.

All other relevant data are available from the corresponding authors upon reasonable request.

Acknowledgements

Y. Z. thanks the Chinese Scholarship Council for financial support. D. Z. acknowledges funding by the German Research Foundation (ZH279/13-1). J. L. gratefully acknowledges funding by the German Research Foundation under the Excellence Strategy of the Cluster of Excellence PhoenixD (EXC 2122, Project ID 390833453). J. L. is also grateful for additional funding by an Athene Grant of the University of Tübingen (by the Federal Ministry of Education and Research (BMBF) and the Baden-Württemberg Ministry of Science as part of the Excellence Strategy of the German Federal and State Governments).

Notes and references

1. S. Qiao, H. Yang, Z. Bai, G. Peng and X. Zhang, Adv. Eng. Res., 2017, 141, 1408–1413.
2. K. P. J. Thomas, F. Jaramillo, J. Bonde, J. H. Nielsen, S. Horch and I. Chorkendorff, Science, 2007, 317, 100–102.
3. R. Kumar, W. Zheng, X. Liu, J. Zhang and M. Kumar, Adv. Mater. Technol., 2020, 5, 1901062.
4. B. Radisavljevic, A. Radenovic, J. Brivio, V. Giacometti and A. Kis, Nat. Nanotechnol., 2011, 6, 147–150.
5. Y. Ye, Z. J. Wong, X. Lu, X. Ni, H. Zhu, X. Chen, Y. Wang and X. Zhang, Nat. Photonics, 2015, 9, 733–737.
6. B. Zhu, X. Chen and X. Cui, Sci. Rep., 2015, 5, 9218.
7. K. F. Mak, C. Lee, J. Hone, J. Shan and T. F. Heinz, Phys. Rev. Lett., 2010, 105, 136805.
8. Y. Ma, Y. Dai, M. Guo, C. Niu, J. Lu and B. Huang, Phys. Chem. Chem. Phys., 2011, 13, 15546–15553.
9. A. Splendiani, L. Sun, Y. Zhang, T. Li, J. Kim, C. Y. Chim, G. Galli and F. Wang, Nano Lett., 2010, 10, 1271–1275.
10. H. Zeng, G. B. Liu, J. Dai, Y. Yan, B. Zhu, R. He, L. Xie, S. Xu, X. Chen, W. Yao and X. Cui, Sci. Rep., 2013, 3, 1608.
11. A. Carvalho, R. M. Ribeiro and A. H. Castro Neto, Phys. Rev. B: Condens. Matter Mater. Phys., 2013, 88, 115205.
12. T. Goswami, H. Bhatt, K. J. Babu, G. Kaur, N. Ghorai and H. N. Ghosh, J. Phys. Chem. Lett., 2021, 12, 6526–6534.
13. G. Plechinger, P. Nagler, J. Kraus, N. Paradiso, C. Strunk, C. Schüller and T. Korn, Phys. Status Solidi RRL, 2015, 9, 457–461.
14. S. Golovynskyi, O. I. Datsenko, D. Dong, Y. Lin, I. Irfan, B. Li, D. Lin and J. Qu, J. Phys. Chem. C, 2021, 125, 17806–17819.
15. A. Boulesbaa, B. Huang, K. Wang, M.-W. Lin, M. Mahjouri-Samani, C. Rouleau, K. Xiao, M. Yoon, B. Sumpter, A. Puretzky and D. Geohegan, Phys. Rev. B: Condens. Matter Mater. Phys., 2015, 92, 115443.
16. P. Miao, Y. T. Chen, L. Pan, A. Horneber, K. Greulich, T. Chasse, H. Peisert, P. M. Adam, P. Xu, A. Meixner and D. Zhang, J. Chem. Phys., 2022, 156, 034702.
17. D.-H. Lien, S. Z. Uddin, M. Yeh, M. Amani, H. Kim, J. W. Ager III, E. Yablonovitch and A. Javey, Science, 2019, 364, 468–471.
18. H. Qiu, T. Xu, Z. Wang, W. Ren, H. Nan, Z. Ni, Q. Chen, S. Yuan, F. Miao, F. Song, G. Long, Y. Shi, L. Sun, J. Wang and X. Wang, Nat. Commun., 2013, 4, 2642.
19. L. Li, M. F. Lin, X. Zhang, A. Britz, A. Krishnamoorthy, R. Ma, R. K. Kalia, A. Nakano, P. Vashishta, P. Ajayan, M. C. Hoffmann, D. M. Fritz, U. Bergmann and O. V. Prezhdo, Nano Lett., 2019, 19, 6078–6086.
20. M. R. Rosenberger, H. J. Chuang, K. M. McCreary, C. H. Li and B. T. Jonker, ACS Nano, 2018, 12, 1793–1800.
21. Y.-Z. Lan and X.-H. Bao, Phys. Rev. B: Condens. Matter Mater. Phys., 2020, 101, 195437.
22. S. Karmakar, U. Chatterjee and P. Kumbhakar, Opt. Laser Technol., 2021, 136, 106696.
23. H. Long, L. Tao, C. Y. Tang, B. Zhou, Y. Zhao, L. Zeng, S. F. Yu, S. P. Lau, Y. Chai and Y. H. Tsang, Nanoscale, 2015, 7, 17771–17777.
24. Z. Ye, T. Cao, K. O'Brien, H. Zhu, X. Yin, Y. Wang, S. G. Louie and X. Zhang, Nature, 2014, 513, 214–218.
25. W. J. Schutte, J. L. D. Boer and F. Jellinek, J. Solid State Chem., 1987, 70, 207–209.
26. A. Klein, S. Tiefenbacher, V. Eyert, C. Pettenkofer and W. Jaegermann, Phys. Rev. B: Condens. Matter Mater. Phys., 2001, 64, 205416.
27. N. Kumar, S. Najmaei, Q. Cui, F. Ceballos, P. M. Ajayan, J. Lou and H. Zhao, Phys. Rev. B: Condens. Matter Mater. Phys., 2013, 87, 125427.
28. Y. Li, Y. Rao, K. F. Mak, Y. You, S. Wang, C. R. Dean and T. F. Heinz, Nano Lett., 2013, 13, 3329–3333.
29. A. R. Khan, B. Liu, T. Lu, L. Zhang, A. Sharma, Y. Zhu, W. Ma and Y. Lu, ACS Nano, 2020, 14, 15806–15815.
30. J. Liang, J. Zhang, Z. Li, H. Hong, J. Wang, Z. Zhang, X. Zhou, R. Qiao, J. Xu, P. Gao, Z. Liu, Z. Liu, Z. Sun, S. Meng, K. Liu and D. Yu, Nano Lett., 2017, 17, 7539–7543.
31. K. S. Novoselov, F. S. D. Jiang, T. J. Booth, V. V. Khotkevich, S. V. Morozov and A. K. Geim, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 10451–10453.
32. J. N. Coleman, M. Lotya, A. O’Neill, S. D. Bergin, P. J. King, U. Khan, K. Young, A. Gaucher, S. De, R. J. Smith, I. V. Shvets, S. K. Arora, G. Stanton, H.-Y. Kim, K. Lee, G. T. Kim, G. S. Duesberg, T. Hallam, J. J. Boland, J. J. Wang, J. F. Donegan, J. C. Grunlan, G. Moriarty, A. Shmeliov, R. J. Nicholls, J. M. Perkins, E. M. Grieveson, K. Theuwissen, D. W. McComb, P. D. Nellist and V. Nicolosi, Science, 2011, 331, 568–571.
33. P. Joensen, R. F. Frindt and S. R. Morrison, Mater. Res. Bull., 1986, 21, 457–461.
34. J. R. Brent, N. Savjani and P. O'Brien, Prog. Mater. Sci., 2017, 89, 411–478.
35. B. Mahler, V. Hoepfner, K. Liao and G. A. Ozin, J. Am. Chem. Soc., 2014, 136, 14121–14127.
36. G. Pippia, D. Van Hamme, B. Martin-Garcia, M. Prato and I. Moreels, Nanoscale, 2022, 14, 15859–15868.
37. N. Kapuria, N. N. Patil, A. Sankaran, F. Laffir, H. Geaney, E. Magner, M. Scanlon, K. M. Ryan and S. Singh, J. Mater. Chem. A, 2023, 11, 11341–11353.
38. A. Niebur, A. Soll, P. Haizmann, O. Strolka, D. Rudolph, K. Tran, F. Renz, A. P. Frauendorf, J. Hubner, H. Peisert, M. Scheele and J. Lauth, Nanoscale, 2023, 15, 5679–5688.
39. M. Frohlich, M. Kogel, J. Hiller, L. Kahlmeyer, A. J. Meixner, M. Scheele, J. C. Meyer and J. Lauth, Phys. Chem. Chem. Phys., 2024, 26, 13271–13278.
40. Y. Sun, K. Fujisawa, Z. Lin, Y. Lei, J. S. Mondschein, M. Terrones and R. E. Schaak, J. Am. Chem. Soc., 2017, 139, 11096–11105.
41. G. Pippia, A. Rousaki, M. Barbone, J. Billet, R. Brescia, A. Polovitsyn, B. Martín-García, M. Prato, F. De Boni, M. M. Petrić, A. Ben Mhenni, I. Van Driessche, P. Vandenabeele, K. Müller and I. Moreels, ACS Appl. Nano Mater., 2022, 5, 10311–10320.
42. A. P. Frauendorf, A. Niebur, L. Harms, S. Shree, B. Urbaszek, M. Oestreich, J. Hübner and J. Lauth, J. Phys. Chem. C, 2021, 125, 18841–18848.
43. A. Dasgupta, X. Yang and J. Gao, ACS Photonics, 2020, 7, 2506–2514.
44. S. Klimmer, O. Ghaebi, Z. Gan, A. George, A. Turchanin, G. Cerullo and G. Soavi, Nat. Photonics, 2021, 15, 837–842.
45. M. Stavrou, N. Chazapis, E. Nikoli, R. Arenal, N. Tagmatarchis and S. Couris, ACS Appl. Nano Mater., 2022, 5, 16674–16686.
46. V. Pelgrin, H. H. Yoon, E. Cassan and Z. Sun, Light: Adv. Manuf., 2023, 4(3), 311–333.
47. W. J. Schutte, J. L. De Boer and F. Jellinek, J. Solid State Chem., 1987, 70, 207–209.
48. A. Berkdemir, H. R. Gutiérrez, A. R. Botello-Méndez, N. Perea-López, A. L. Elías, C.-I. Chia, B. Wang, V. H. Crespi, F. López-Urías, J.-C. Charlier, H. Terrones and M. Terrones, Sci. Rep., 2013, 3, 1–8.
49. L. Pan, P. Miao, A. Horneber, A. J. Meixner, P. M. Adam and D. Zhang, Beilstein J. Nanotechnol., 2022, 13, 572–581.
50. X. Wang, K. Broch, R. Scholz, F. Schreiber, A. J. Meixner and D. Zhang, J. Phys. Chem. Lett., 2014, 5, 1048–1054.
51. A. L. Elías, N. Perea-López, A. Castro-Beltrán, A. Berkdemir, R. Lv, S. Feng, A. D. Long, T. Hayashi, Y. A. Kim, M. Endo, H. R. Gutiérrez, N. R. Pradhan, L. Balicas, T. E. Mallouk, F. López-Urías, H. Terrones and M. Terrones, ACS Nano, 2013, 7, 5235–5242.
52. Y. Liu, X. Hu, T. Wang and D. Liu, ACS Nano, 2019, 13, 14416–14425.
53. X. Xu, L. Li, M. Yang, Q. Guo, Y. Wang, X. Li, X. Zhuang and B. Liang, Opt. Express, 2021, 29, 5856–5866.
54. Q. Cui, Z. Luo, Q. Cui, W. Zhu, H. Shou, C. Wu, Z. Liu, Y. Lin, P. Zhang, S. Wei, H. Yang, S. Chen, A. Pan and L. Song, Adv. Funct. Mater., 2021, 31.
55. T. Kato and T. Kaneko, ACS Nano, 2016, 10, 9687–9694.
56. P. K. Chow, R. B. Jacobs-Gedrim, J. Gao, T.-M. Lu, B. Yu, H. Terrones and N. Koratkar, ACS Nano, 2015, 9, 1520–1527.
57. V. Carozo, Y. Wang, K. Fujisawa, B. R. Carvalho, A. McCreary, S. Feng, Z. Lin, C. Zhou, N. Perea-López, A. L. Elías, B. Kabius, V. H. Crespi and M. Terrones, Sci. Adv., 2017, 3, 1–9.
58. H. Sun, Q. Yang, J. Wang, M. Ding, M. Cheng, L. Liao, C. Cai, Z. Chen, X. Huang, Z. Wang, Z. Xu, W. Wang, K. Liu, L. Liu, X. Bai, J. Chen, S. Meng and L. Wang, Nat. Commun., 2024, 15, 9476.
59. Z. Ye, T. Cao, K. O’Brien, H. Zhu, X. Yin, Y. Wang, S. G. Louie and X. Zhang, Nature, 2014, 513, 214–218.
60. C. Poellmann, P. Steinleitner, U. Leierseder, P. Nagler, G. Plechinger, M. Porer, R. Bratschitsch, C. Schüller, T. Korn and R. Huber, Nat. Mater., 2015, 14, 889–893.
61. S. Zhang, N. Dong, N. McEvoy, M. O’Brien, S. Winters, N. C. Berner, C. Yim, Y. Li, X. Zhang, Z. Chen, L. Zhang, G. S. Duesberg and J. Wang, ACS Nano, 2015, 9, 7142–7150.
62. Y. R. Shen, Annu. Rev. Phys. Chem., 1989, 40, 327–350.
63. R. I. Woodward, R. T. Murray, C. F. Phelan, R. E. P. D. Oliveira, T. H. Runcorn, E. J. R. Kelleher, S. Li, E. C. D. Oliveira, G. J. M. Fechine, G. Eda and C. J. S. D. Matos, 2D Mater., 2016, 4.
64. J.-P. Wang, Y.-Q. Fang, W. He, Q. Liu, J.-R. Fu, X.-Y. Li, Y. Liu, B. Gao, L. Zhen, C.-Y. Xu, F.-Q. Huang, A. J. Meixner, D. Zhang and Y. Li, Adv. Opt. Mater., 2023, 11, 2300031.
65. C. Janisch, Y. Wang, D. Ma, N. Mehta, A. L. Elias, N. Perea-Lopez, M. Terrones, V. Crespi and Z. Liu, Sci. Rep., 2014, 4, 5530.
66. L. Pan, P. Miao, A. Horneber, A. J. Meixner, P.-M. Adam and D. Zhang, ACS Appl. Nano Mater., 2023, 6, 6467–6473.
67. J. Hernandez-Rueda, M. L. Noordam, I. Komen and L. Kuipers, ACS Photonics, 2021, 8, 550–556.

---

Footnote

† Yang Zhao and Markus Fröhlich contributed equally.

---