The lifetime of interlayer breathing modes of few-layer 2H-MoSe₂ membranes

Abstract

Investigation of the decay processes leading to the phonon lifetime of optically excited vibrational modes of few-layer MoSe2 membranes.

Time-Domain Investigations of Coherent Phonons in van der Waals Thin Films

“…They are characterized by an extracted frequency in the GHz range, which no longer depends on the probe wavelength, but increases when the material gets thinner, as shown in Figure 7 c,d. This has been observed in exfoliated few layer crystals of WSe 2 [ 133 ], MoS 2 [ 143 ], MoSe 2 [ 132 ], InSe [ 135 , 136 , 137 ], PtSe 2 [ 134 ], BP [ 126 ], Bi 2 Se 3 [ 243 ], and Bi 2 Te 3 [ 139 ]. In all these examples, the penetration depth of the pump beam gets larger than the thickness of the sample.…”

“…Although this can originate from sample and technique uncertainties, the lower values of v B extracted for thinner slabs might originate from boundary effects, which dominate only for few-layer samples. This behavior is clearly evidenced in MoSe 2 for which the extracted acoustic velocity shows a gradual reduction for decreasing number of layers (starting below 10 layers, see Figure 8 a) from 2820 m/s in bulk crystal down to 2000 m/s in a two-layer film [ 132 ]. Velocity in thin Bi 2 Se 3 is similarly reduced from 3000 m/s in bulk and 40-nm-thick stacks down to 2000 m/s for a thickness of 15 nm [ 243 ].…”

“…This condition is obviously not verified when vdW crystals are reduced to only a few layers. Deviation from the simple linear dependence has thus to be considered to reproduce experimental data, and Equation (2) computed in the rigid-layer framework in Section 2.4 has to be used, as presented in Figure 8 a [ 132 ]. Furthermore, although only the fundamental mode is usually observed in experiments, higher harmonics can also be probed [ 136 ], which reduces even more the range of validity of the continuous medium expressions.…”

“…The assignment to the node/anti-node case for a given experimental configuration consists in a good characterization of the nature of the interface. Though it is obvious that a fully suspended film yields anti-nodes at both ends [ 132 ], the assignment when the crystal is lying on a substrate is not as straightforward. The hypothesis in early studies was to assume a node when the layer is supported.…”

“…To reach more intrinsic quality factor limitations, measurements on fully suspended crystals must be performed. In this view, Soubelet et al [ 132 ] reported the highest , up to around 200, in thin vdW materials by investigating MoSe 2 flakes, from 2 to 54 layers, exfoliated on top of a substrate with micron-size holes. Two distinct regimes depending on the flake thickness are demonstrated (see Figure 8 b) and discussed.…”