Comparative Study of Raman Spectroscopy in Graphene and MoS2-type Transition Metal Dichalcogenides

Abstract

CONSPECTUS: Raman spectroscopy is one of the most powerful experimental tools to study graphene, since it provides much useful information for sample characterization. In this Account, we show that this technique is also convenient to study other bidimensional materials beyond graphene, and we will focus on the semiconducting transition metal dichalcogenides (MX2), specifically on MoS2 and WS2. We start by comparing the atomic structure of graphene and 2H-MX2 as a function of the number of layers in the sample. The first-order Raman active modes of each material can be predicted on the basis of their corresponding point-group symmetries. We show the analogies between graphene and 2H-MX2 in their Raman spectra. Using several excitation wavelengths in the visible range, we analyze the first- and second-order features presented by each material. These are the E2g and 2TO(K) bands in graphene (also known as the G and 2D bands, respectively) and the A1', E', and 2LA(M) bands in 2H MX2. The double-resonance processes that originate the second-order bands are different for both systems, and we will discuss them in terms of the different electronic structure and phonon dispersion curves presented by each compound. According to the electronic structure of graphene, which is a zero band gap semiconductor, the Raman spectrum is resonant for all the excitation wavelengths. Moreover, due to the linear behavior of the electronic dispersion near the K point, the double-resonance bands of graphene are dispersive, since their frequencies vary when we change the laser energy used for the sample excitation. In contrast, the semiconducting MX2 materials present an excitonic resonance at the direct gap, and consequently, the double-resonance Raman bands of MX2 are not dispersive, and only their intensities depend on the laser energy. In this sense, resonant Raman scattering experiments performed in transition metal dichalcogenides using a wide range of excitation energies can provide information about the electronic structure of these materials, which is complementary to other optical spectroscopies, such as absorption or photoluminescence. Raman spectroscopy can also be useful to address disorder in MX2 samples in a similar way as it is used in graphene. Both materials exhibit additional Raman features associated with phonons within the interior of the Brillouin zone that are activated by the presence of defects and that are not observed in pristine samples. Such is the case of the well-known D band of graphene. MX2 samples present analogous features that are clearly observed at specific excitation energies. The origins of these double-resonance Raman bands in MX2 are still subjects of current research. Finally, we discuss the suitability of Raman spectroscopy as a strain or doping sensor. Such applications of Raman spectroscopy are being extensively studied in the case of graphene, and considering its structural analogies with MX2 systems, we show how this technique can also be used to provide strain/doping information for transition metal dichalcogenides.