Resonant Raman Scattering Study of Strain and Defects in Chemical Vapor Deposition Grown MoS2 Monolayers

Abstract

The development of bottom-up synthesis routes for semiconducting transition metal dichalcogenides (TMDs) and the assessment of their defects are of paramount importance to achieve their applications. TMD monolayers grown by chemical vapor deposition (CVD) can be subjected to significant strain and, here, Raman and photoluminescence spectroscopies are combined to characterize strain in over one hundred MoS₂ monolayer samples grown by CVD. The frequency changes of phonons as a function of strain are analyzed, and used to extract the Grüneisen parameter of both zone-center and edge phonons. Additionally, the intensity of the defect-induced longitudinal acoustic (LA) and transverse acoustic (TA) Raman bands are discussed in relation to strain and electronic doping. The experimental mode-Grüneisen parameters obtained are compared with those calculated by density functional theory (DFT), to better characterize the type of strain and its resulting effects on Grüneisen parameters. The findings indicate that the use of Raman spectra to determine defect densities in 2D MoS₂ must be always conducted considering strain effects. To the best of the authors' knowledge, this work constitutes the first report on double resonance Raman processes studied as a function of strain in 2D-MoS₂. The new approach to obtain the Grüneisen parameter from zone-edge phonons in MoS₂ can also be extended to other 2D semiconducting TMDs.

Keywords: 2D materials; DFT calculations; MoS2; Raman; chemical vapor deposition; defects; electronic doping; strain.

Figure 1

a) Raman and b) PL intensity color map for all the MoS₂ samples studied in this work. The scale bar on the right shows the normalized intensity of the spectra.

Figure 2

a) MoS₂ Raman spectrum obtained from sample #40 shown in Figure 1. Empty hexagons correspond to the experimental data after a baseline subtraction, green lines are the fitted peaks, and the solid gray line is the cumulative fit. b) $A_1^{\prime }$ band frequency (${\omega }_{A_1^{\prime }}$) plotted against the frequency of the E′ band (ω _E′). Tension‐compression and electronic doping levels are obtained from the literature.^{[ 8} , ^{35 ]}

Figure 3

Raman intensities, normalized by the intensity of the $E^{\prime }$ mode, of the a) $A_1^{\prime }$ and b) LA bands in MoS₂. The black dots are experimental points while the red line corresponds to a smoothed average. The purple, green and blue spheres in (a) correspond to data from refs. [31, 40, 46]. The blue spheres in (b) correspond to data from ref. [14].

Figure 4

Frequency of the MoS₂ Raman bands as a function of the X_A position. The numbers represent the slope from the line fitting in units of cm⁻¹ eV⁻¹.

Figure 5

Peak specific Grüneisen parameter as a function of the frequency of MoS₂ Raman bands. a) Experimental data extracted from results presented in Figure 4. b) Computational values obtained from DFT calculations for the Grüneisen parameter while applying strain uniaxially along the a ₁ vector. Open circles in (b) show phonons with strong probability of participating in DR processes for the LA, TA, 2LA and 2TA bands, while full symbols show specific phonon frequencies at the Γ (full hexagons), K and ${\mathbf{K}}^{\prime }$ (full diamonds) and M (full squares) points.