**1. Introduction**

The information technology revolution is driven by the continuous improvement of electronic devices and their constant scaling down. However, this process is rapidly approaching its limit. In order to continue the innovation process, introduction of new concepts and new materials is required. Some solutions have been identified in (i) the use of strain engineering and high-k gate dielectrics, (ii) silicon–germanium (SiGe) alloys and germanium, because of their higher electron mobility and lower need for power, and (iii) nanostructures and 2D materials, for the possibility of introducing brand-new device designs and concepts.

For the latter, graphene and similar 2D materials recently emerged as promising candidates because they exhibit interesting mechanical, electronic, optical, and transport properties with peculiar differences from those of their bulk counterparts. Graphene, despite its exceptional physical properties, has the major drawback of lacking an electronic bandgap, a limit that poses problems in realizing logic circuits and transistors. Bandgap engineering in graphene, although possible, has several drawbacks, such as increased complexity of the process and degradation of material quality, so it is not usually considered a viable choice for certain applications.

Transition metal dichalcogenides (TMD), silicon, germanium, and boron nitride can be thinned down to monolayers similarly to graphene, exhibiting weak interplane interaction and strong in-plane bonds. MoS<sup>2</sup> is one of the most promising materials of this family because of the relative easiness of its synthesis and its interesting physical properties.

**Citation:** Seravalli, L.; Bosi, M. A Review on Chemical Vapour Deposition of Two-Dimensional MoS<sup>2</sup> Flakes. *Materials* **2021**, *14*, 7590. https://doi.org/10.3390/ma14247590

Academic Editor: Abbes Tahraoui

Received: 28 October 2021 Accepted: 7 December 2021 Published: 10 December 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

It exhibits a hexagonal lattice structure consisting of a layer of transition metal atoms embedded in two chalcogen layers, with strong covalent bonds in each 2D plane and a weaker interaction between different planes. The properties of MoS2, similarly to those of other 2D materials, depend mainly on the vertical thickness of MoS<sup>2</sup> rather than its lateral size. The most important difference with respect to graphene is that 2D MoS<sup>2</sup> shows a direct bandgap of about 1.8 eV, making it an ideal candidate for applications in electronics, photonics, photovoltaics, energy storage, and catalysis [1].

Several proof-of-concept MoS<sup>2</sup> devices have been already demonstrated. FETs with mobility up to 320 cm<sup>2</sup> V −1 s <sup>−</sup><sup>1</sup> and on/off ratio of 10<sup>8</sup> at room temperature were realized [2,3], as well as inverters with gain up to 16 [4]. Several kinds of phototransistors, photosensors [5], and gas and biological sensors [6] were demonstrated, and the mechanical properties of MoS<sup>2</sup> were exploited in the realization of strain sensors [7]. Moreover, the light emission properties of this material are also attracting considerable interest [8–11].

The future development of research is trending towards the integration of different building blocks and different 2D materials into a single device, aiming for the fabrication of high-performance CMOS-based circuits, sensors, and efficient photocatalytic systems [12,13].

The interest in developing new devices from 2D structures is also driven by several novel properties observed in these materials, such as valley polarization, in which conduction and valence bands of MoS<sup>2</sup> monolayers present two inequivalent valleys at points K and K0, giving a new degree of freedom to carriers. This would permit radical new device concepts based on valleytronics and on spin–valley coupling, permitting control of information using circular polarized light [14].

Despite the results obtained so far on MoS<sup>2</sup> devices and transistors in particular, it was observed that their performance was still far from the best theoretical predictions. The transport properties of MoS<sup>2</sup> are limited by defects such as vacancies, antisites, charge traps, grain boundaries, and Coulomb impurities at the flake interface [15,16]. For example, the maximum predicted electron mobility of MoS<sup>2</sup> at room temperature is about 410 cm<sup>2</sup> V −1 s −1 , leaving room for improvement. Understanding these limits and how to overcome them is a major issue on the MoS<sup>2</sup> roadmap.

Reliable and controlled doping in MoS<sup>2</sup> flakes is also a fundamental requirement to achieve precise control of their electrical and optical properties. Usually, substitutional doping is the dominating process as compared to interstitial doping. On this topic, fundamental contributions were the very recent reviews of Lin et al. [17] and Rai et al. [18], who considered the issues of Schottky barriers and contact resistance at the interface between MoS<sup>2</sup> and a metal and the problems related to the doping of a wide range of 2D TMD materials, respectively. N-type doping of MoS<sup>2</sup> could be obtained by adding ReO<sup>3</sup> to the solid precursors [19] or by using ZnS as a sulfur precursor [20], while chloride permitted decreasing the resistivity of MoS<sup>2</sup> flakes from some KΩ to about 0.5 KΩ, allowing for the realization of Schottky barriers with enhanced properties [21]. p-doping could be obtained by adding Nb2O<sup>5</sup> to NaCl promoter when using solid precursors [22].

Another topic of particular interest is the possibility of including elements such as manganese (Mn) and iron (Fe), aiming to add new functionalities related to magnetism and spin to MoS2. First reports indicated the possibility of incorporating Mn in MoS2, but with some issues relative to the substrate used, as Mn was detected only in structures deposited on graphene and not on those deposited on SiO2/Si [23].

Despite their peculiar properties and the achievements so far obtained, the controllable synthesis of large-area flakes and the reproducibility of the process are still challenging issues; these are two essential, key factors required to sustain the future development of 2D materials. Mechanical exfoliation has so far granted a very high material quality, leading to the realization of devices and to an in-depth study of their properties. However, this method is very impractical and poses many limits on large-scale production. Industrial application of TMD is directly linked to the development of a high-throughput and reliable technique to obtain 2D layers of different materials on large area substrates, with a simple, reproducible, and scalable method. Chemical vapour deposition (CVD) is a good candidate

for this task, since it has already been widely adopted for the mass production of many kinds of devices based on III-V and III-N materials and has thus the potential to permit the integration of 2D materials with "standard" semiconductors or other compounds. Many excellent reviews on MoS<sup>2</sup> are already available, giving a broad scenario of all the work done on this material [24,25].

Although MoS<sup>2</sup> flakes with large size (>500 µm) have been obtained with relevant success from several groups [26–28], in our opinion, research is still needed on many points that are usually overlooked: (i) the effect of different parameters (such as carriers' flows or sulfur partial pressure), (ii) the possibility of engineering layers by surface treatments—this is crucial for the realization of heterostructures and possibly hybrid nanostructures, (iii) the growth reproducibility of these structures. It is a common experience to not be able to obtain the same 2D layers in different growth runs, and this is still a consistent problem for the technology's transfer towards industrial applications. A reliable growth protocol that can be transferred to different reactors with minimal setup time is a strong requirement to allow 2D MoS<sup>2</sup> to make the jump to mass production.

The preparation of this material is still challenging, and there is a lot of room for improving both the process and the material quality. Despite its relative simplicity, CVD remains one of the most widely used techniques to synthesize TMD flakes. Recently, new protocols and precursors have been introduced to optimize the deposition, to obtain better quality material, and to improve the reproducibility of the process.

The aim of this work was to collect and critically present the knowledge acquired so far in this field, providing the reader with a comprehensive review of the state of the art of the CVD growth processes used to prepare MoS2.
