Recent Progress on Irradiation-Induced Defect Engineering of Two-Dimensional 2H-MoS₂ Few Layers

Abstract

Atom-thick two-dimensional materials usually possess unique properties compared to their bulk counterparts. Their properties are significantly affected by defects, which could be uncontrollably introduced by irradiation. The effects of electromagnetic irradiation and particle irradiation on 2H MoS${}_2$ two-dimensional nanolayers are reviewed in this paper, covering heavy ions, protons, electrons, gamma rays, X-rays, ultraviolet light, terahertz, and infrared irradiation. Various defects in MoS${}_2$ layers were created by the defect engineering. Here we focus on their influence on the structural, electronic, catalytic, and magnetic performance of the 2D materials. Additionally, irradiation-induced doping is discussed and involved.

1. Introduction

With the discoveries of zero-dimensional buckyballs in the 1980s [1] and one-dimensional carbon nanotubes in the 1990s [2,3], various nanomaterials have been synthesized and characterized. Their physical and chemical properties are unique compared with counterpart bulks (graphite, diamond, and amorphous carbon) because of size-induced quantum effects, enabling their great potential in various fields. Two-dimensional (2D) graphene was reported in the 2000s [4] and excellent electrical behaviors were claimed. Since then investigations of 2D layers were initialized and rapidly developed in recent decades. At present, the 2D layer family spans conductors (such as graphene), insulators (such as hexagonal boron nitride), and semiconductors. Transition-metal dichalcogenide (TMDC) materials, including MoS${}_2$, WS${}_2$, MoSe${}_2$, and WSe${}_2$, are layered semiconductors and have been fabricated into 2D semiconducting layers. MoS${}_2$ 2D layers have novel potential applications in novel nano-optoelectronics, optical sensors, catalysts, energy storages, and environments. Additionally, MoS${}_2$ 2D layers show thickness-dependent semiconducting behaviors, being the most interesting and important among TMDC 2D layers. Bulk MoS${}_2$ is a semiconductor with an indirect energy band gap of $1.2\mathrm{eV}$ while mono-layer MoS${}_2$ is a direct band-gap semiconductor with an energy band gap of $1.9\mathrm{eV}$ due to 2D quantum confinement. The MoS${}_2$ layers have also important applications in valleytronics—a new type of electronics where information is encoded by wave quantum number of electrons.

MoS${}_2$ low-dimensional materials have been investigated over the past 50 years. MoS${}_2$ few-layers (4–5 layers) were first prepared by a peeling process from bulk crystals in the 1960s [5]. Tribological properties of MoS${}_2$ films have been studied later in a vacuum or dry air as solid lubricants since the 1990s [6,7]. Compared with bulks, the friction coefficient of MoS${}_2$ films was reduced, and the wear life of the films was enhanced. Catalytic properties of MoS${}_2$ nanoparticles [8] and films [9,10] have been investigated from the 1980s. It is believed that edge sites of MoS${}_2$ nanomaterials play important roles in electrochemical and catalytic behaviors. MoS${}_2$ atomically thin materials, especially mono-layer MoS${}_2$, have been widely investigated in recent years because of their sizable direct band gap, high charge-carrier mobility, and excellent mechanical flexibility [11,12]. The materials have potential applications in the next generation flexible electronics [11,13,14], elastic energy storage [12], field-effect transistors [15,16], electronic switches [13], electronic devices such as chemical biosensors [16,17], and optoelectronic devices [18] such as photodetectors and solar cells [19]. It was demonstrated that chemically active defects, such as sulfur vacancies, significantly control or tune their catalytic activities [20], electronic transports [21,22,23], and optical properties [24].

Various forms of irradiation and particle fluxes modulate the density of sulfur vacancies of chemical-vapor deposition (CVD) grown and mechanically exfoliated MoS${}_2$ sheets, affecting their chemically active defects. Therefore, irradiation effects have been carried out in recent years. Additionally, these MoS${}_2$ devices and sensors have been used in harsh environments, and it is necessary to investigate their tolerance under irradiation. Radiation effects on two-dimensional graphene were recently reviewed [25,26]. Defect engineering and defect-induced properties of other atom-thick mono-layers were summarized as well [26]. The impacts of irradiation on 2H MoS${}_2$ two-dimensional materials are reviewed here.

1.1. MoS${}_2$ 2D Materials

MoS${}_2$ has two main types of natural phases: hexagonal structure (2H type) with the space group of P6${}_3$/mmc (D${}_{6h}^4$, No. 194) and the trigonal structure (3R type) with the space group of R3m (C${}_{3v}^5$, No. 160). The 2H phase contains two layers per unit cell, $a=3.15\mathsf{\AA }$ and $c=12.30\mathsf{\AA }$. Similar to graphite, the 2H type crystals can be easily cleaved to form (0001) layers because of weaker van der Waals forces between (0001) layers. The 3R phase contains three layers per unit cell, stacking in rhombohedral symmetry with trigonal prismatic coordination, $a=3.16\mathsf{\AA }$ and $c=18.33\mathsf{\AA }$. A third metastable phase, 1T-MoS${}_2$ with a tetragonal symmetry, was also artificially produced by intercalating 2H-MoS${}_2$ with alkali metals [27,28,29]. The artificial 1T phase and natural 3R phase are metastable at room temperature and transform to the 2H phase on heating [27] or IR irradiation [30] or microwave irradiation [31]. This work focuses on the 2H MoS${}_2$ phase. 1T and 3R polytypes will not be discussed.

Molybdenum disulfide (2H-MoS${}_2$) is a transition-metal dichalcogenide with a melting point of $2375{}^{\circ }\mathrm{C}$. The bulk material is chemically stable in dilute acids and oxygen and insoluble in water. Figure 1a shows the crystallographic structure of 2H-MoS${}_2$ bulks. There are two S-Mo-S layers per unit cell stacked in hexagonal symmetry. Each S-Mo-S layer consists of two hexagonal planes of sulfur atoms and an intermediate hexagonal plane of molybdenum atoms. The sandwiched molybdenum atom plane is coordinated through ionic-covalent interactions with the sulfur atoms, forming a stable MoS${}_2$ mono-layer as shown in Figure 1b. The MoS${}_2$ mono-layers are held together by weak van der Waals interactions. The forces between the mono-layers are not very strong [32], and the interatomic forces within a mono-layer are sufficient for thermodynamical stability [33]. Therefore, 2H-MoS${}_2$ has been widely used as a dry lubricant under high temperature due to the microscaled chemical properties.

Figure 2 shows the calculated band gap of MoS${}_2$ bulks and that of MoS${}_2$ few-layers. The bulk 2H-MoS${}_2$ crystal is an indirect band-gap semiconductor with a band gap of 1.23–1.29 $\mathrm{e}\mathrm{V}$ [32,34,35,36]. MoS${}_2$ bulks can be employed as cathode materials in batteries [29]. MoS${}_2$ bulks show very poor catalytic activities for hydrogen evolution reactions. The band gap of MoS${}_2$ is a function of layer thickness as shown in Figure 2b–d, and unperturbed by substrate interactions. The mono-layer MoS${}_2$ has a direct band gap of $1.8\mathrm{e}\mathrm{V}$ [37]. The band structure changes from the indirect-gap in bulks to the direct-gap in MoS${}_2$ mono-layers, being confirmed from experiments [38,39]. The thickness-dependent electronic structure change is closely associated with many corresponding changes in chemical and physical properties. For example, the transition for indirect-to-direct band gap is correlated with changes in mechanical strength, spin density, bond energy, electrical conductivity, and the properties of transistor and sensor devices. MoS${}_2$ nanoparticles can adsorb various compounds, including tetrahydrothiophene, thiophene, benzothiophene, dibenzothiophene, and their derivatives [40]. The chemical [29,41] and physical properties [42,43,44] of MoS${}_2$ few-layers were recently reviewed. These novel properties of MoS${}_2$ 2D materials make them suitable 2D candidates in environmental applications [45,46], catalysts (such as electrocatalysts [47,48,49,50,51] and photocatalysts [48,49,50,51,52,53]), energy storage (such as lithium ion batteries [48,49,54], supercapacitors [49,54], elastic energy [12], and solar cells [49]), as well as sensing [55]. MoS${}_2$ mono-layers were also fabricated into ultrasensitive photodetectors [56,57], and integrated circuits to perform the NOR logic operation [58], besides being employed in drug delivery [59], thermoelectrics [60,61], piezotronics [62,63], osmotic nanopower generators [64], and valleytronics [65].

Because of the novel properties and wide applications of 2H-MoS${}_2$ 2D layers, MoS${}_2$ 2D layers have been prepared by various physical or chemical methods [41,44,45,46,66,67,68,69]. Individual MoS${}_2$ layers were usually obtained by the micromechanical cleavage technique [70] because of weak interactions between sulfur layers and strong intralayer interactions of MoS${}_2$ bulks. The mechanically cleaved MoS${}_2$ layers are usually highly crystalline with low defect density. Massive MoS${}_2$ layers have successfully been produced through the liquid-phase exfoliation [71,72]. MoS${}_2$ few-layers can also be fabricated from CVD. Interested readers can refer to the references herein.

The MoS${}_2$-based devices will be used in harsh environments from basic science to industry, such as near nuclear plants/reactors and nuclear medicine imaging. Radiations may damage MoS${}_2$ layers by breaking atomic bonds and ionizing atoms. To use the MoS${}_2$ devices reliably under irradiation, it is desirable to explore the irradiation effects on MoS${}_2$ layers.

1.2. Irradiation Sources

Figure 3 shows the spectrum of electromagnetic irradiation. Cosmic rays are composed of particularly high-energy particles and photons. Primary cosmic rays are composed primarily of 99% nuclei (about 90% protons, 9% alpha particles, and 1% nuclei of heavier elements) and about 1% solitary electrons. The energy of the cosmic rays is high up to ${10}^{20}$$\mathrm{e}\mathrm{V}$ for the primary cosmic rays. When entering the Earth atmosphere, the primary cosmic rays collide with atoms and molecules to produce secondary cosmic particles with lower energy and electromagnetic waves, including gamma-ray, X-ray, neutrons, protons, electrons, and alpha particles. The energy of the generated gamma rays can be 50 $\mathrm{M}$$\mathrm{e}\mathrm{V}$. These primary and secondary cosmic rays should damage microelectronic devices because of sufficient energy, causing soft errors in electronic integrated circuits, especially in small-scale devices. Here, several ion irradiation and electromagnetic irradiation from GeV to meV are reviewed on 2D MoS${}_2$ few-layers. Physical properties of MoS${}_2$ bulks and thick films are also covered for comparison with those of 2D MoS${}_2$ few-layers.

1.3. Irradiated MoS${}_2$ Materials

Defects are created in MoS${}_2$ few-layers when the layers are fabricated by exfoliation technique and vapor deposition growth when layers are transferred from substrate to substrate, and placed under electromagnetic radiation and energetic particle fluxes. The defects typically deteriorate the device quality of MoS${}_2$ layers. On the other hand, defects in MoS${}_2$ also have a beneficial impact on material properties. For instance, defects have been shown to be the dominant dopant in MoS${}_2$, with natural defect variation allowing n-type and p-type regions to coexist across distinct regions of the same sample [73]. The sulfur vacancies can lower Schottky barrier heights of MoS${}_2$ [73]. Therefore, it is crucial to fully understand defects in MoS${}_2$ to develop a reliable MoS${}_2$-based electronics. In this paper, irradiation-related defects of MoS${}_2$ few-layers are focused on.

Up to now, MoS${}_2$ 2D layers have been irradiated under different irradiation sources.

Table 1. Radiation Induced Defects in 2H MoS${}_2$.

MoS${}_2$	Irradiation	Particle Energy	Dose/Fluence	Irradiation Source	Irradiation Conditions	Results	Ref.
$200\mathsf{\mu }\mathrm{m}$ thickness	electron	0.7 MeV	150–600 kGy	electron accelerator	RT	created S vacancies	[74]
$200\mathsf{\mu }\mathrm{m}$ thickness	electron	2.0 MeV	100–250 kGy	electron accelerator	RT	created S vacancies and Mo vacancies, diamagnetic to ferromagnetic phase transition	[74]
20 layer	electron	$5\mathrm{k}\mathrm{e}\mathrm{V}$	50 nA for 15 min	SEM beam	in vacuum	formed crystalline islands	[75]
mono-layer	electron	$60\mathrm{k}\mathrm{e}\mathrm{V}$	${10}^6$–${10}^9$ electron $/\mathrm{n}{\mathrm{m}}^2$	STEM beam	400–700 ${}^{\circ }\mathrm{C}$ in vacuum	induced 2H/1T phase transition	[76]
∼10 layer	electron	3–15 $\mathrm{k}$$\mathrm{e}\mathrm{V}$	n/a	EPMA	in vacuum	broke the inversion symmetry	[77]
mono-layer	electron	$80\mathrm{k}\mathrm{e}\mathrm{V}$	n/a	TEM beam	in vacuum	removed top and bottom S atoms	[78]
mono-layer	electron	$200\mathrm{k}\mathrm{e}\mathrm{V}$	$3000\mathrm{electrons}/\mathrm{n}{\mathrm{m}}^2/\mathrm{s}$	TEM beam	in vacuum	created S vacancies, increased electric resistance	[79]
mono-layer	electron	$15\mathrm{k}\mathrm{e}\mathrm{V}$	$280\mathsf{\mu }\mathrm{C}/\mathrm{c}{\mathrm{m}}^2$	EBL	in vacuum	produced local strain and changed band structure	[80]
mono-layer	electron	$80\mathrm{k}\mathrm{e}\mathrm{V}$	$40\mathrm{A}/\mathrm{c}{\mathrm{m}}^2$	TEM beam	in vacuum	produced holes and Mo${}_5$S${}_3$ nanoribbons	[81]
amorphous 5–7 layer	electron	$1\mathrm{k}\mathrm{e}\mathrm{V}$	1–10 $\min$	EBI	in vacuum	crystallized	[82]
mono-layer	U${}^{238}$	1.14 GeV	$4000\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$	heavy ion accelerator	in vacuum	total damaged	[83]
micron thickness	Ar${}^{+}$	500 eV	$2.26\times {10}^{15}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$	plasma	UHV	produced S vacancies	[84]
bi-layer	Ar${}^{+}$	500 eV	${10}^{14}$–${10}^{15}$ ions$/\mathrm{c}{\mathrm{m}}^2$	plasma	UHV	produced S vacancies and MoS${}_6$ vacancy clusters	[84]
mono-layer	Ar${}^{+}$	500 eV	$2.26\times {10}^{15}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$	plasma	UHV	damaged	[84]
$200\mathsf{\mu }\mathrm{m}$ thickness	proton	3.5 MeV	$5\times {10}^{18}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$	Singletron facility	RT	preserved lattice structure, produced defects, changed magnetic moments	[85]
few-layer	proton	10 MeV	${10}^{12}$–${10}^{14}$ ions$/\mathrm{c}{\mathrm{m}}^2$	MC-50 cyclotron	n/a	decreased electrical conductance	[86]
mono-layer	proton	100 keV	${10}^{12}$–${10}^{15}$ particles$/\mathrm{c}{\mathrm{m}}^2$	LEAF	n/a	created defects	[87]
bi-layer	proton	100 keV	$6\times {10}^{14}\mathrm{particles}/\mathrm{c}{\mathrm{m}}^2$	LEAF	n/a	created defects	[87]
bulk	He${}^{2+}$	1.66 MeV	900 MGy	ion accelerator	n/a	changed Raman scattering slightly	[88]
nanosheet	He${}^{2+}$	1.66 MeV	900 MGy	ion accelerator	n/a	invariant	[88]
few-layer	He${}^{2+}$	30 keV	${10}^{18}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$	FIB beam	in vacuum	milled or damaged	[89]
mono-layer	He${}^{2+}$	3.04 MeV	$8\times {10}^{13}\mathrm{particles}/\mathrm{c}{\mathrm{m}}^2$	PTA	n/a	produced defects	[90]
mono-layer	He${}^{+}$	30 keV	${10}^{12}$–${10}^{16}$ ions$/\mathrm{c}{\mathrm{m}}^2$	HIM	in vacuum	produced S vacancies	[91]
mono-layer	He${}^{+}$	30 keV	${10}^{13}$–${10}^{17}$ ions$/\mathrm{c}{\mathrm{m}}^2$	NFM	in vacuum	generated disorder	[92]
bulk	Ar${}^{+}$	500 eV	$3\times {10}^{-3}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$	STM	UHV	removed S atoms	[93]
micron thickness	Ar${}^{+}$	500 eV	${10}^{14}$–${10}^{15}$ ions$/\mathrm{c}{\mathrm{m}}^2$	n/a	UHV	produced S vacancies at low doses and Mo/MoS${}_6$ vacancies at high doses	[84]
powder	Ar${}^{+}$	3 keV	$0.1\mathrm{A}/\mathrm{c}{\mathrm{m}}^2$	n/a	n/a	removed S atoms	[94]
200 nm	Ar${}^{+}$	180 keV	${10}^{15}$–${10}^{17}$ ions$/\mathrm{c}{\mathrm{m}}^2$	IBAD	$200{}^{\circ }\mathrm{C}$ in vacuum	induced amorphization	[95]
bi-layer	Ar${}^{+}$	500 eV	${10}^{14}$–${10}^{15}$ ions$/\mathrm{c}{\mathrm{m}}^2$	n/a	UHV	produced S vacancies at low doses and Mo/MoS${}_6$ vacancies at high doses	[84]
mono-layer	Ar${}^{+}$	500 eV	${10}^{14}$–${10}^{15}$ ions$/\mathrm{c}{\mathrm{m}}^2$	n/a	UHV	produced S vacancies at low doses and Mo/MoS${}_6$ vacancies at high doses	[84]
mono-layer	Ar${}^{+}$	500 eV	3–11 $\mathsf{\mu }\mathrm{A}/\mathrm{c}{\mathrm{m}}^2$	sputter beam	vacuum	selectively removed S atoms without significantly depleted Mo atoms	[96]
mono-layer	Ar${}^{+}$	13.56 MHz	n/a	RF plasma	RT	produced 2H/1T phase transition	[97]
mono-layer	Ga ion	n/a	n/a	PAMBE	$450{}^{\circ }\mathrm{C}$ in vacuum	doped Ga, reduced binding energy	[98]
mono-layer	Ga ion	30 keV	${10}^{13}\mathrm{ion}/\mathrm{c}{\mathrm{m}}^2$	FIB beam	in vacuum	produced sub-nm pores and vacancies	[99]
bulk	Xe	91 MeV	$10\mathrm{ions}/\mathsf{\mu }{\mathrm{m}}^2$	GANIL	n/a	formed nano-hillocks	[100]
few-layer	Xe	91 MeV	$15\mathrm{ions}/\mathsf{\mu }{\mathrm{m}}^2$	GANIL	n/a	formed nano-incisions	[100]
mono-layer	Xe ion	91 MeV	$15\mathrm{ions}/\mathsf{\mu }{\mathrm{m}}^2$	GANIL	n/a	formed nano-incisions	[100]
mono-layer	Xe ion	25–30 keV	${10}^{10}\mathrm{ion}/\mathrm{c}{\mathrm{m}}^2$	EBIT	n/a	induced pits and hillocks	[101]
1–4 layer	Bi ion	0.45–1.23 $\mathrm{G}\mathrm{e}\mathrm{V}$	${10}^{10}$–${10}^{12}$ ions$/\mathrm{c}{\mathrm{m}}^2$	HIRFL	vacuum, RT	formed hillocks	[102]
mono-layer	Mn ion	25 keV	${10}^{12}$–${10}^{14}$ ions$/\mathrm{c}{\mathrm{m}}^2$	TOF-SIMS	UHV	formed defects	[103]
bulk	C${}_{60}$	20–40 MeV	${10}^9$–${10}^{11}$ ions$/\mathrm{c}{\mathrm{m}}^2$	TA	RT	modified structures	[104]
micron thickness	$\gamma$-ray	n/a	$5\times {10}^9\mathrm{R}$	RGIF	in air	no damaging effects	[105]
50–132 layer	$\gamma$-ray	662 keV	5000 photons	${}^{137}$Cs source	RT in air	unaffected	[106]
5–8 layer	$\gamma$-ray	∼1.2 $\mathrm{M}$$\mathrm{e}\mathrm{V}$	$120\mathrm{M}\mathrm{rad}$	${}^{60}$Co source	RT in air	converted to MoO${}_x$	[107]
1–3 layer	X-ray	10 keV	6 Mrad	XRD	RT in air	no noticeable degradation	[108]
film	UV	n/a	n/a	mercury lamp	in O${}_2$ (ozone)	formed oxygen-sulfur bonds	[109]
mono-layer and multilayer	UV	n/a	${10}^{10}$–${10}^{13}$ photos$/\mathrm{c}{\mathrm{m}}^2$	deuterium lamp	vacuum	no structural damage, no oxidation	[110]
multilayer	laser	$\lambda =514\mathrm{n}\mathrm{m}$	1–20 mW for 1–100 s	laser	RT in air	stable or damage depending on laser power	[111]
few-layer	laser	$\lambda =532\mathrm{n}\mathrm{m}$	$300\mathrm{m}\mathrm{W}$	diode laser	in air or vacuum	patterned and thinned	[112]
mono-layer	laser	$\lambda =800\mathrm{n}\mathrm{m}$	20–50 $\mathrm{m}\mathrm{J}/\mathrm{c}{\mathrm{m}}^2$	Ti:sapphire laser	RT	damaged or unaffected depending on irradiation intensity	[113]

EBI: electron-beam irradiation; EBIT: electron-beam ion trap; EBL: electron-beam lithography system; EPMA: electron probe micro-analyzer; FIB: focused ion beam; GANIL: The Grand Accélérateur National d’Ions Lourds (Large Heavy Ion National Accelerator); HIM: helium-ion microscope; HIRFL: heavy ion Research facility in Lanzhou; IBAD: ion-beam assisted deposition; LEAF: low-energy accelerator facility; NFM: NanoFab microscope; PAMBE: plasma-assisted molecular beam epitaxy; PTA: Pelletron tandem accelerator; RF: radio frequency; RGIF: Reactor Gamma Irradiation Facility; RT: room temperature; S: sulfur; SEM: scanning electron microscope; STEM: scanning transmission electron microscope; STM: scanning tunneling microscope; TA: Tandem accelerator; TEM: transmission electron microscope; TOF-SIMS: time-of-flight secondary ion mass spectrometer; UHV: ultra-high vacuum; XRD: X-ray diffractometer.

lists some typical irradiation effects of MoS${}_2$ materials. The details are summarized in the following sections. The particle irradiation and electromagnetic irradiation will be reviewed first on structural properties and defects, then theoretical explanations of the irradiation mechanism are discussed. The irradiation-induced band structures, electric, catalytic, and magnetic properties will be summarized at the end. To understand the irradiation-induced defects well, the synthesis/preparation details of the as-prepared MoS${}_2$ layers will be described in each section in addition to irradiation.

2. Charged Particle Irradiation

2.1. Swift-Heavy Ion Irradiation

Swift-heavy ions are usually accelerated in particle accelerators to very high energies, typically in the MeV or GeV range. They have sufficient kinetic energy and mass to penetrate solid materials along straight lines. Therefore, heavy ions release sufficient energy displace atoms, induce heating, permanently modify crystal structure, and leave tracks of heavily damaged structure.

2.1.1. Uranium-238 Ion Irradiation

Exfoliated MoS${}_2$ mono-layers were irradiated by $1.14\mathrm{G}\mathrm{e}\mathrm{V}$ uranium (${}^{238}$U) ions [83]. It was found that its electrical properties were significantly changed and the MoS${}_2$-based transistor was destroyed at a fluence of $4\times {10}^{11}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ ($4000\mathrm{ions}/\mathsf{\mu }{\mathrm{m}}^2$).

2.1.2. Gallium Ion Irradiation

Michra et al. [98] deposited p-type mono-layered MoS${}_2$ on c-sapphire substrates using CVD technique and irradiated the layers at $450{}^{\circ }\mathrm{C}$ for 30 s at a low Ga flux beam (equivalent pressure: $6\times {10}^8\mathrm{torr}$) in ultra-high vacuum (UHV) conditions. Ga ions were produced from a plasma-assisted MBE system. Figure 4 shows the Raman spectra and X-ray photoelectron spectroscopy (XPS) spectra of Ga-irradiated MoS${}_2$ mono-layers. Raman peaks, especially A${}_{1g}$ mode, shifted after the Ga-irradiation. Mo-$3d^{5/2}$ peaks shifted towards lower energies for Ga-irradiated samples relative to that of the pristine MoS${}_2$ mono-layer. New Ga-$2p^{1/2}$ and Ga-$2p^{3/2}$ core levels were observed at $1117\mathrm{e}\mathrm{V}$ and $1144\mathrm{e}\mathrm{V}$, respectively, revealing the incorporation of Ga into the MoS${}_2$ layers. The Ga-irradiation reduced the value of binding energy of $0.2\mathrm{e}\mathrm{V}$. Room temperature photoluminescence (PL) spectroscopy indicated that the optical properties of the MoS${}_2$ layers were changed.

Thiruraman et al. [99] synthesized single-layer MoS${}_2$ triangular-flakes via a halide-assisted powder vaporization method and transferred over holes. The suspending MoS${}_2$ mono-layers were irradiated by Ga${}^{+}$ ions in a focused ion beam (FIB). The Ga${}^{+}$ ion dose varied from $6.25\times {10}^{12}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ to $2.50\times {10}^{13}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$. Figure 5 shows high angle annular dark-field (HAADF) images of the MoS${}_2$ mono-layers before and after Ga${}^{+}$ ion irradiation for different doses: 0 (pristine), $6.25\times {10}^{12}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$, $8.16\times {10}^{12}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$, $1.11\times {10}^{13}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$, $1.60\times {10}^{13}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$, and $2.50\times {10}^{13}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$. Statistical analysis showed that Ga${}^{+}$ ion irradiation produced pores with average and maximum diameters of $0.5\mathrm{n}\mathrm{m}$ and $1.0\mathrm{n}\mathrm{m}$. Within the irradiation dose range, the nanopore density increased with increasing doses. With increasing ion doses, nanopores started to merge, resulting in larger and irregularly shaped pores. For the lowest dose ($6.25\times {10}^{12}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$), most of the atomic pores were single-molybdenum-based vacancies, while the amount of missing sulfur varied. With an increasing Ga${}^{+}$ ion dose, the number of double-molybdenum-based vacancies increased, and some triple-molybdenum-based vacancies were also found. When the Ga${}^{+}$ ion dose reached to $2.50\times {10}^{13}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$, the density of larger pores (with diameter size $>0.8\mathrm{n}\mathrm{m}$) increased. Metal atomic vacancies were formed with sulfur vacancies while few topological defects and amorphous regions were observed.

2.1.3. Xenon Ion Irradiation

Madau$\beta$ et al. [100] fabricated ultra-thin mechanically exfoliated MoS${}_2$ sheets on SiO${}_2$ substrates and irradiated them using $91\mathrm{M}\mathrm{e}\mathrm{V}$ Xe ions. Figure 6a shows Atomic force microscopy (AFM) images of the track morphology in MoS${}_2$ with various thicknesses. In the case of bulk-like MoS${}_2$, chains of nanosized hillocks were generated and protruded from the surface. For few-layer MoS${}_2$ sheets (thickness under $10\mathrm{n}\mathrm{m}$), individual hillocks were created and mixed with incisions. In even thinner MoS${}_2$ sheets, i.e., tri-layered, bi-layered, and mono-layered MoS${}_2$ sheets, the major part of the surface track consisted of a continuous incision. The length of the incision varied among tri-layered (3L), bi-layered (2L), and single-layered (1L) sheets. Compared with the surface track length of mono-layers, the length of the incisions was reduced by ∼25% for a bi-layer and by ∼50% for a tri-layer. Additionally, the surface tracks in tri-layered and bi-layered MoS${}_2$ were accompanied by protrusions before and after the central incision. Mono-layered MoS${}_2$ surface tracks generally consisted only of incisions and no protrusions. Figure 6b shows lattice information around an incision (length $100\mathrm{n}\mathrm{m}$) created by a projectile. The extended incisions were oriented along the direction of the incoming ion beam and extremely narrow (less than $10\mathrm{n}\mathrm{m}$). The edges of the incision were relatively straight at an atomic scale. The surrounding lattice remained undisturbed while the violent atomic displacements took place inside the ion track core. Several mechanisms were proposed to explain the ion irradiation damages, such as electrostatic repulsion between the atoms in the ionized region, exciton self-trapping causing a local lattice distortion, non-thermal melting caused by significant changes of the interatomic potentials, and phase transitions such as melting due to a thermal spike. It was assumed that the hillock chains in bulk-like MoS${}_2$ sheets consisted of nonstoichiometric MoS${}_2$ which re-solidified after a phase transition caused by the thermal spike. The nanoscaled incisions should have originated from the material evaporation. It was reported that the threshold for grazing incidence of Xe ions was $2.0\mathrm{k}\mathrm{e}\mathrm{V}/\mathrm{n}\mathrm{m}$.

Hopster et al. [101] deposited MoS${}_2$ mono-layers on KBr(100) substrates and irradiated them with highly charged Xe ions (Xe${}^{35+}$ with kinetic energy of 25.4 keV and Xe${}^{40+}$ with kinetic energy of 38.5 keV) in a vacuum at room temperature. The fluences ranged from $5\times {10}^9$–$2\times {10}^{10}$ ions$/\mathrm{c}{\mathrm{m}}^2$. Hillocks were observed on the MoS${}_2$ mono-layer surfaces. The MoS${}_2$ mono-layers were exfoliated from a single crystal under ambient condition while no details were provided, so it is hard to compare their results with other groups.

2.1.4. Bismuth Ion Irradiation

Guo et al. [102] deposited 1–4-layer-ed MoS${}_2$ on silicon substrates capped with SiO${}_2$. The mechanically exfoliated MoS${}_2$ nanosheets were irradiated by ${}^{209}$Bi ions with energies of 0.45–1.23 $\mathrm{G}$$\mathrm{e}\mathrm{V}$ and fluences of $1\times {10}^{10}$–$3.6\times {10}^{12}$ ions$/\mathrm{c}{\mathrm{m}}^2$ in a vacuum at room temperature under normal incidence. The hillock-like latent tracks were observed on the surface of irradiated MoS${}_2$ few-layers and attributed to the ionization and excitation of energy transfer from ${}^{209}$Bi ions to the electron system. The induced damages shifted the Raman A${}_{1g}$ peak to a higher frequency and increased the intensity ratio between the A${}_{1g}$ and E${}_{2g}^1$ modes.

2.1.5. Manganese Ion Irradiation

Mignuzzi et al. [103] mechanically exfoliated natural bulk MoS${}_2$ and deposited it on Si substrates covered with SiO${}_2$. The exfoliated MoS${}_2$ mono-layers were bombarded with Mn${}^{+}$ (kinetic energy of $25\mathrm{keV}$) with different ion doses (${10}^{12}$–${10}^{14}$ ions$/\mathrm{c}{\mathrm{m}}^2$) in a UHV. The resulting average inter-defect distance ranged from $1\mathrm{n}\mathrm{m}$ to $10\mathrm{n}\mathrm{m}$. The generated defects activated new Raman modes around $227\mathrm{c}{\mathrm{m}}^{-1}$, broadened Raman peaks, and shifted Raman modes.

2.1.6. Gold Ion Irradiation

Zhai et al. [114] bombarded MoS${}_2$ (0001) bulk surfaces using Au ions with $13.4\mathrm{M}\mathrm{e}\mathrm{V}$. Under ion doses of $1.0\times {10}^{13}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$, 1.0–3.5 $\mathrm{n}$$\mathrm{m}$ craters were generated on the surface.

2.1.7. Silver Ion Irradiation

Bhattacharya et al. [115] irradiated magnetron-sputtered amorphous MoS${}_2$ films of thicknesses 50–750 $\mathrm{n}$$\mathrm{m}$ with a $5\times {10}^{15}\mathrm{c}{\mathrm{m}}^{-2}$ dose of $2\mathrm{M}\mathrm{e}\mathrm{V}$ Ag ions. The sliding life of Ag ion-irradiated films increased ten-fold to thousand-fold compared to as-sputtered films. The improvement in wear life was correlated with a significant improvement in adhesion of the films with the substrates and a small increase in the density of the ion-irradiated films.

2.1.8. C${}_{60}$ Ion Irradiation

Henry et al. [104] irradiated MoS${}_2$ bulks under 20–40 $\mathrm{M}$$\mathrm{e}\mathrm{V}$ C${}_{60}$ ions at $300\mathrm{K}$. Irradiation was performed at normal incidence and at high fluences ($3\times {10}^{11}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$), as well at a grazing incidence of $20$° and a low fluence of ${10}^9\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$. The structural modifications occurred in the vicinity of the projectile paths.

2.2. Argon Ion Irradiation

Argon ion beams ($500\mathrm{e}\mathrm{V}$) were employed as early as 1987 [9] to irradiate MoS${}_2$ crystalline films (cleaved from a single crystal) with a thickness of $0.7\mathsf{\mu }\mathrm{m}$. The Ar bombardments created sulfur vacancies and reduced molybdenum of MoS${}_2$ edge surfaces.

Inoue et al. [93] studied defects generated by Ar${}^{+}$ ion irradiation of MoS${}_2$ bulk surface by scanning tunneling microscopy (STM). A clean MoS${}_2$ surface was prepared by cleaving the surface layers of a MoS${}_2$ bulk crystal with Scotch tape. The cleaved clean MoS${}_2$ was then degassed in a UHV chamber by electrical heating at $538\mathrm{K}$ for half an hour to provide a clean and contamination-free surface. The sample was irradiated with Ar${}^{+}$ ions at an energy of $500\mathrm{e}\mathrm{V}$ and an irradiation density of $75\times {10}^{11}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$. The incident angle of Ar${}^{+}$ ion was ±2–3${}^{\circ }$ from the normal to the MoS${}_2$ surface. The irradiated MoS${}_2$ sample was heated at $583\mathrm{K}$ for an hour by direct current heating through the substrate to remove residual ions on the sample surface after irradiation. Low-density individual defects were observed. Some surface defects were formed by removal of sulfur atoms from the top MoS${}_2$ surface. Some defects may originate from the hybridized dangling bond composed of the Mo-$4d$ orbital and removal of MoS${}_2$ layer fragments.

Bae et al. [84] MoS${}_2$ prepared slabs with a thickness of several microns from MoS${}_2$ crystals using the mechanical exfoliation method and then irradiated them with Ar${}^{+}$ ions with an energy of $500\mathrm{e}\mathrm{V}$ in an UHV, with fluences of $5.65\times {10}^{14}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ and $2.26\times {10}^{15}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$. The incident angle of Ar${}^{+}$ was almost normal to the surface of the samples. Residual argon on sample surfaces was removed by heating at $583\mathrm{K}$ for one hour in an UHV after the irradiation process. Figure 7 shows the Raman spectra of Ar${}^{+}$ irradiated samples. After Ar${}^{+}$ irradiation, additional broad satellite peaks appeared in the lower-frequency side of both the $E_{2g}^1$ and $A_{1g}$ peaks, located at 377–381 $\mathrm{c}{\mathrm{m}}^{-1}$ and 404–406 $\mathrm{c}{\mathrm{m}}^{-1}$ respectively.

The Raman satellite peaks shifted to lower wave-numbers while their intensity increased with increasing Ar${}^{+}$ doses. First-principles calculation showed that new satellite modes were related to molybdenum vacancies, sulfur vacancies, or MoS${}_6$ vacancy clusters. The lower shift of the satellites of the $E_{2g}^1$ mode should have come from the large cluster vacancy. The $A_{1g}$ and $E_{2g}^1$ mode of MoS${}_2$ materials remained unchanged before and after the irradiation. It is believed that the new peaks were induced by lattice defects introduced by Ar${}^{+}$ irradiation. Additionally, the intensity of the satellite peaks increased with the increasing irradiation dose and dominated over the $E_{2g}^1$ and $A_{1g}$ modes of pristine MoS${}_2$, indicating the formation of molybdenum vacancies or MoS${}_6$ vacancies.

Baker et al. [94] irradiated MoS${}_2$ powders under a $3\mathrm{keV}$ Ar${}^{+}$ ion beam with a current density of $0.1\mathsf{\mu }\mathrm{A}/\mathrm{c}{\mathrm{m}}^2$. The Ar${}^{+}$ ion-beam bombardment caused the preferential sputtering of sulfur and produced an amorphous MoS${}_x$ phase of depth of $3.8\mathrm{n}\mathrm{m}$. The Mo/S ratio increased with increasing bombardment time, as shown in Figure 8.

Wahl et al. [95] deposited MoS${}_2$ layers (roughly $200\mathrm{n}\mathrm{m}$) on steel and Si substrates via ion-beam assisted deposition. The produced MoS${}_2$ layers were fully dense with basal-oriented microstructures. The deposited MoS${}_2$ layers were then irradiated with $180\mathrm{k}\mathrm{e}\mathrm{V}$ Ar${}^{2+}$ ions at doses of $1\times {10}^{15}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$, $1\times {10}^{16}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$, and $5\times {10}^{16}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ at room temperature in a vacuum. Figure 9a shows X-ray diffraction (XRD) patterns of the as-deposited and ion-irradiated coatings. The intensity of the (002) main peak at $13.5{}^{\circ }$ decreased with increasing irradiation doses. The basal plane peak ($2\theta =11{}^{\circ }$) disappeared after the irradiation. Figure 9b shows Raman spectra of the samples. The main peaks of the Raman spectra were kept after the irradiation, while their intensity decreased with increasing doses. To explore the irradiation effects, the microstructure of the MoS${}_2$ layers was characterized by high-resolution transmission electron microscopy (HRTEM) as shown in Figure 9c. The as-deposited MoS${}_2$ consisted of basal-oriented and nanocrystalline MoS${}_2$ with horizontal (002) fringes. After an irradiation dose of $1\times {10}^{15}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$, some regions became amorphous, and the surviving crystalline regions (still with the basal orientation) were embedded in the amorphous matrix. After a high irradiation dose of $5\times {10}^{16}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$, the MoS${}_2$ regions were nearly amorphous, although few crystalline regions existed. The surviving crystalline regions oriented randomly after the high dose irradiation. The irradiation-reduced microstructural change degraded tribological behaviors of MoS${}_2$ coatings, such as accelerated wear and eliminated lubrication.

Murray et al. [116] irradiated mechanically exfoliated MoS${}_2$ few-layers (5–15 layers) under low-energy Ar ($200\mathrm{e}\mathrm{V}$ and 1 $\times {10}^{13}$–${10}^{15}$ ions $/\mathrm{c}{\mathrm{m}}^2$). The electric resistance increased 100 times after a high dose of irradiation because of induced defects.

Bae et al. [84] also prepared MoS${}_2$ bi-layers by chemical exfoliation from MoS${}_2$ micron powders in a hexane solution of butyllithium. The prepared bi-layers were then irradiated by Ar${}^{+}$ ions with an energy of $500\mathrm{e}\mathrm{V}$ in an UHV, with fluences of $5.65\times {10}^{14}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ and $2.26\times {10}^{15}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$. Figure 10 shows Raman spectra of the MoS${}_2$ bi-layers before and after Ar${}^{+}$ irradiation. The $A_{1g}$ and $E_{2g}^1$ modes were broadened. It was claimed that a new broad peak emerged at approximately $373\mathrm{c}{\mathrm{m}}^{-1}$, which was $10\mathrm{c}{\mathrm{m}}^{-1}$ away from the $E_{2g}^1$ peak. There was a broad feature at around $435\mathrm{c}{\mathrm{m}}^{-1}$. The mode was relatively enhanced after the irradiation. The new vibration mode was identified as a unique signature of the sulfur vacancies.

Bae et al. [84] claimed that MoS${}_2$ mono-layers were destroyed by Ar${}^{+}$ irradiation with $500\mathrm{e}\mathrm{V}$. Chen et al. [117] sputtered MoS${}_2$ mono-layers with Ar${}^{+}$ ions ($500\mathrm{e}\mathrm{V}$ and emission current of $20\mathrm{m}\mathrm{A}$), defect peaks of Raman scattering were observed and its hydrogen evolution performance was enhanced. However, no details were reported in the literature.

Zhu et al. [97] treated CVD-grown MoS${}_2$ mono-layers under radio-frequency (RF) argon plasma with $13.56\mathrm{M}\mathrm{Hz}$ at room temperature for 40 s. The very weak Ar plasma had enough kinetic energy to trench the S-Mo bond while no enough energy to etch molybdenum and sulfur atoms. Thus, the plasma treatments induced the lateral sliding of the top S layer and the $2H$ phase transited to the $1T$ phase.

Ma et al. [96] grew MoS${}_2$ mono-layers on SiO${}_2$ substrates and treated them in a vacuum under $500\mathrm{e}\mathrm{V}$ Ar${}^{+}$ ions at room temperature. The beam current density was 3.1–11.2 $\mathsf{\mu }\mathrm{A}/\mathrm{c}{\mathrm{m}}^2$. The molybdenum content of the MoS${}_2$ layers remained essentially constant while the amount of sulfur decreased significantly during irradiation. The sulfur content of the layers could be reduced to 50%. The average sputter yield was 0.03 sulfur atom per Ar${}^{+}$ ion. During the irradiation procedure, MoS${}_2$ mono-layers were selectively de-sulfurized while the basic physical structure of the MoS${}_2$ remained largely intact. The photoluminescence (PL) yield decreased as sulfur atoms were removed by the Ar${}^{+}$ irradiation.

Ar${}^{+}$ plasma was also employed to thin MoS${}_2$ sheets at room temperature [118]. It was reported that the top MoS${}_2$ layers were entirely removed by the Ar${}^{+}$ plasma while the bottom MoS${}_2$ layers remained largely unaffected. Thus, MoS${}_2$ multi-layers with controllable thickness, even MoS${}_2$ single-layers, could be prepared reliably (with almost 100% success rate).

Chen et al. [117] prepared MoS${}_2$ mono-layers using a liquid exfoliation method, and annealed them at 300–450 ${}^{\circ }\mathrm{C}$ in a vacuum. The treated layers were then deposited on Au substrates and exposed to $0.5\mathrm{k}\mathrm{e}\mathrm{V}$ Ar${}^{+}$ ions for $1\min$. Defects were introduced into the MoS${}_2$ layers, influencing the electronic structures of MoS${}_2$. Figure 11 shows XPS spectra of irradiated MoS${}_2$ mono-layers. Peaks shifted because of induced defects.

2.3. Neon Ion Irradiation

Maguire et al. [92] also irradiated CVD-grown MoS${}_2$ mono-layers with Ne${}^{+}$ ions with an energy of $30\mathrm{k}\mathrm{e}\mathrm{V}$. With increasing Ne${}^{+}$ ion doses, the $A_{1g}$ and $E_{2g}$ modes were quenched and broadened, indicating the growing disorder induced by the Ne${}^{+}$ ions.

2.4. Alpha-Particle/Helium-Ion Irradiation

Alpha particles consist of two protons and two neutrons which are tightly bound together bound together, which is identical to a helium ion (He${}^{2+}$). They are produced either from particle accelerators in the form of helium-ion beams or from the process of alpha decay of alpha-particle-emitting radionuclides. Alpha particles have been widely studied in radiopharmaceutical therapy that is a promising treatment approach under active pre-clinical and clinical investigation [119,120]. Decayed alpha particles generally have a kinetic energy of about $5\mathrm{M}\mathrm{e}\mathrm{V}$, which induce defects and even `cut’ the MoS${}_2$ nanosheets.

Isherwood et al. [88] studied the effects of alpha-particle (helium nuclei) irradiation ($1.66\mathrm{M}\mathrm{e}\mathrm{V}$) on both bulk and liquid-phase exfoliated MoS${}_2$ nanosheets using Raman spectroscopy and energy-dispersive X-ray spectroscopy. Liquid-phase exfoliated MoS${}_2$ layers were often more defective than mechanically exfoliated MoS${}_2$ and CVD-grown MoS${}_2$ layers. Besides water, the solvent used during liquid exfoliation could be retained between the MoS${}_2$ layers during ultrasonication and annealing/cleaning post-procedures. Raman spectroscopy showed a small blueshift of the E${}_{2g}^1$ and $A_{1g}$ modes in the MoS${}_2$ bulk under a high total absorbed dose (∼900 MGy), accompanied by a small broadening of both peaks. The exfoliated MoS${}_2$ layers were less radiation tolerant than the bulk material. Both spectroscopies proved the presence of amorphous carbon in the exfoliated MoS${}_2$ membranes, which could be formed by radiolytic amorphization of residual solvent and retained within the exfoliated nanosheets.

Fox et al. [89] irradiated freestanding mechanically exfoliated MoS${}_2$ few-layers in a helium-ion microscope. Figure 12 shows HRTEM images of a MoS${}_2$ few-layer. The hexagonal structure of the pristine MoS${}_2$ was destroyed to an amorphous state. Energy-dispersive X-ray spectroscopy (EDX) indicated that preferential sputtering of sulfur occurred in the mechanically exfoliated MoS${}_2$ few-layers. Its stoichiometry was modified by the preferential sputtering of sulfur at nanometer scales. It was believed that He${}^{2+}$ ions transfer more energy to sulfur atoms than to molybdenum atoms because of the lighter mass of sulfur atoms. Electric properties of the MoS${}_2$ layers were altered to be semiconducting, metallic-like, or insulating depending on doses. When the dose was above $(2.56\pm 0.05)\times {10}^{18}\mathrm{ion}/\mathrm{c}{\mathrm{m}}^2$, complete removal of material (milling) was observed. Figure 12c shows He${}^{2+}$-ion fabricated freestanding nanoribbons in MoS${}_2$ few-layers. The edges of the milled regions may be amorphous or crystalline, depending on He${}^{2+}$ beam sizes. MoS${}_2$ layers were milled when the dose was higher than $(1.30\pm 0.03)\times {10}^{18}\mathrm{ion}/\mathrm{c}{\mathrm{m}}^2$. When the dose was below $(1.00\pm 0.02)\times {10}^{17}\mathrm{ion}/\mathrm{c}{\mathrm{m}}^2$, MoS${}_2$ layers were extensively damaged but not completely etched away.

Fox et al. [89] also irradiated MBE-grown pristine MoS${}_2$ 6-layers supported on an MgO substrate. With increasing He${}^{2+}$ doses, the full width at half maximum (FWHM) of the E${}_{2g}^1$ (the in-plane Mo-S vibration) increased, indicating more in-plane defects during irradiation. The FWHM of A${}_{1g}$ peaks decreased with increasing doses, inferring material removal with high He${}^{2+}$ doses.

Tongay et al. [90] irradiated mechanically exfoliated MoS${}_2$ mono-layers with a high-energy He${}^{2+}$ beam ($3.04\mathrm{M}\mathrm{e}\mathrm{V}$) with doses up to $8\times {10}^{13}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$. Upon the He${}^{2+}$ particle irradiation at different doses, a new PL peak appeared at $1.78\mathrm{e}\mathrm{V}$ and the integrated intensity of this new peak increased with the irradiation dose. The new peak became stronger and broader after thermal annealing at $500{}^{\circ }\mathrm{C}$, which thermally introduced sulfur vacancies. The integrated intensity of the main PL peak at $1.90\mathrm{e}\mathrm{V}$ increased three-fold while the PL peak position shifted to a higher energy by $20\mathrm{m}\mathrm{e}\mathrm{V}$ after the irradiation. Calculations estimated that approximately one defect per 100 unit cells was generated under the $8\times {10}^{13}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ irradiation dose.

Klein et al. [91] employed $30\mathrm{k}\mathrm{e}\mathrm{V}$ He${}^{2+}$ beams to irradiate MoS${}_2$ mono-layers on SiO${}_2$/Si substrates under various doses of ${10}^{12}$–${10}^{16}$ ions$/\mathrm{c}{\mathrm{m}}^2$ in a scanning helium-ion microscope. Helium-ion bombardments affected the intrinsic vibrational, luminescent, and valleytronic properties of the atomically thin MoS${}_2$ sheets.

Maguire et al. [92] grew MoS${}_2$ mono-layers on SiO${}_2$ substrates by a CVD technique and irradiated them with He${}^{2+}$ at an energy of 30 keV and an angle of incidence of $0{}^{\circ }$. These irradiation doses ranged from $1\times {10}^{13}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ to $1\times {10}^{17}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$. With increasing He${}^{2+}$ ion doses, the two characteristic Raman peaks quenched and broadened, reflecting the growing disorder with increasing irradiation. A new peak appeared at $227\mathrm{c}{\mathrm{m}}^{-1}$ after the irradiation and its intensity increased with increasing doses.

Helium-ion beams were also employed to mill MoS${}_2$ layers [121]. Some regions of MoS${}_2$ few-layers were thinned to mono-layers under $30\mathrm{k}\mathrm{e}\mathrm{V}$ He${}^{2+}$ sub-nanometer ion beams.

2.5. Proton Irradiation

Mathew et al. [85] prepared MoS${}_2$ flakes with a thickness of $200\mathsf{\mu }\mathrm{m}$ and irradiated them at room temperature using a $3.5\mathrm{MeV}$ proton ion beam. The Raman spectra of the pristine and irradiated samples at a fluence of $5\times {10}^{18}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ are shown in Figure 13. A new peak at $483\mathrm{c}{\mathrm{m}}^{-1}$ was clearly visible in the irradiated samples. The appearances of the mode at $483\mathrm{c}{\mathrm{m}}^{-1}$ along with the broadening of the mode at $452\mathrm{c}{\mathrm{m}}^{-1}$ indicated the presence of lattice defects due to proton irradiation of the samples. The FWHM of the E${}_{2g}^1$ and A${}_{1g}$ modes did not increase in the irradiated MoS${}_2$, indicating that the lattice structure was preserved in the near-surface region of the irradiated samples. The intensity ratio of A${}_{1g}$/E${}_{2g}$ enhanced 16% after the proton irradiation, showing the enhanced interaction of electrons with A${}_{1g}$ phonons. X-ray photoelectron spectroscopy indicated there were zone-edge phonons in the irradiated samples and the molybdenum valence of irradiated samples was higher than $+4$. The irradiation-induced changes of structures and chemical states affected their magnetic moments.

Kim et al. [86] micromechanically exfoliated MoS${}_2$ few-layers from a bulk MoS${}_2$ crystal and fabricated them into MoS${}_2$ field-effect transistor (FET) devices on highly doped Si substrates coated with SiO${}_2$. Source and drain electrodes were made by depositing Au/Ti electrodes. The FET devices were irradiated with a $10\mathrm{M}\mathrm{e}\mathrm{V}$ proton beam under fluences of ${10}^{12}$, ${10}^{13}$, and ${10}^{14}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$. Sufficiently high irradiation fluences decreased the electrical current and conductance of the devices while low dose did not change them significantly. The threshold voltage was shifted towards the positive gate voltage direction under proton irradiation. However, these changes were recovered over a time scale of days. It was believed that proton irradiation changed the SiO${}_2$/MoS${}_2$ interface states, and the interface trap states at the SiO${}_2$/MoS${}_2$ interfaces affected the electrical behaviors of the FET devices.

Wang et al. [87] exfoliated MoS${}_2$ bi-layers, transferred them to silicon nitride membranes, and irradiated them at $100\mathrm{k}\mathrm{e}\mathrm{V}$ protons (H${}^{+}$) with a fluence of $6\times {10}^{14}\mathrm{particles}/\mathrm{c}{\mathrm{m}}^2$. Figure 14a,b shows the PL spectra of the bi-layers. Both the suspended and substrate-supported bi-layers showed almost complete suppression of the indirect emission (1.55–1.60 $\mathrm{e}\mathrm{V}$) after the irradiation and the emergence of a defect-induced sideband peak at around $1.70\mathrm{e}\mathrm{V}$. The direct band emission at $1.83\mathrm{e}\mathrm{V}$ increased 1.6× after irradiation while the substrate-supported direct band emission increased by a factor of 2.7. After being annealed at $300{}^{\circ }\mathrm{C}$ for $1\mathrm{h}$ in argon, the defect-induced sideband peak disappeared for both kinds of bi-layers. The indirect band emission was suppressed after annealing, indicating that the bi-layers underwent an irreversible indirect-to-direct band-gap transition. Wang et al. [87] also examined MoS${}_2$ tri-layers and four-layers. These multi-layers showed similar behaviors as bi-layers.

Wang et al. [87] also exfoliated MoS${}_2$ mono-layers and transferred them to silicon nitride membranes with holes. The freestanding MoS${}_2$ mono-layers were then irradiated at $100\mathrm{k}\mathrm{e}\mathrm{V}$ protons (H${}^{+}$) with fluences ranging from $2\times {10}^{12}\mathrm{particles}/\mathrm{c}{\mathrm{m}}^2$ to $6\times {10}^{14}\mathrm{particles}/\mathrm{c}{\mathrm{m}}^2$. Figure 14c shows PL spectra of suspending MoS${}_2$ mono-layers after proton irradiation with a fluence of $6\times {10}^{14}\mathrm{particles}/\mathrm{c}{\mathrm{m}}^2$ with and without annealing, compared with those taken before the irradiation. The intensity of the direct band-gap emission of suspending mono-layers decreased two-fold after the irradiation due to the irradiation-induced defects. The defects caused non-radiative recombination and shortened the lifetime of the photo-excited carriers. On the contrary, the intensity of the direct band-gap emission at $1.85\mathrm{eV}$ increased after the irradiation for the substrate-supported regions, as shown in Figure 14d. After being annealed at $300{}^{\circ }\mathrm{C}$ for $1\mathrm{h}$ in argon, the defect-induced sideband peak disappeared in the bi-layers and mono-layers, and the direct band emission reverted to its original pre-irradiated intensity for both kinds of MoS${}_2$ layers.

2.6. Electron Irradiation

Han et al. [122] irradiated MoS${}_2$ single crystals with a thickness of about $50\mathsf{\mu }\mathrm{m}$ by using high-energy electrons in ambient conditions at room temperature. The electron dose was $300\mathrm{kGy}$ ($6.70\times {10}^{14}\mathrm{electrons}/\mathrm{c}{\mathrm{m}}^2$) and the acceleration energy of electrons was $0.7\mathrm{M}\mathrm{e}\mathrm{V}$. There was a negligible reduction of sulfur XPS intensity compared to the molybdenum XPS intensity of the samples after the irradiation. However, 1T-like defects were generated in the MoS${}_2$ surface, inducing a weak ferromagnetic state at room temperature and improving transport properties.

Han et al. [74] prepared single-crystalline MoS${}_2$ lamellae with a thickness of $100\mathsf{\mu }\mathrm{m}$. The lamellae were electron-irradiated in ambient conditions at room temperature. Figure 15 shows HRTEM images of the irradiated samples, showing honeycomb lattices of the MoS${}_2$. The vacancies depended on the acceleration energy and dose. Without irradiation, the lattice of crystalline MoS${}_2$ was a honeycomb with few defects, shown in Figure 15a. After being electron-irradiated with $150\mathrm{kilogray}\left(\mathrm{kGy}\right)$ dose at $0.7\mathrm{MeV}$ acceleration energy, additional defects were produced. Double-sulfur vacancies ($V_{S2}$) were more frequently observed than the monosulfur vacancies ($V_S$) after electron irradiation. Slightly displaced molybdenum atoms were occasionally found, as marked by the red arrow in Figure 15b. Under electron irradiation with a $300\mathrm{k}\mathrm{Gy}$ dose at $0.7\mathrm{MeV}$ acceleration energy, the honeycomb lattice was heavily distorted (Figure 15c) and perfect honeycomb lattices were rarely observed. The numerous $V_S$ and $V_{S2}$ vacancies distorted the honeycomb lattice and the irradiation considerably decreased the lattice parameters. $V_{S2}$ defects were more frequently observed under the electron irradiation with a $100\mathrm{kGy}$ dose at $2.0\mathrm{MeV}$. With an increasing dose of $2.0\mathrm{MeV}$ electrons, the $V_S$ and $V_{S2}$ concentration increased significantly. Molybdenum vacancies were produced too. The lattice parameters were also significantly increased. Under electron irradiation at higher energy of $10\mathrm{MeV}$, different types of defects were produced. Under all irradiation, the vacancy densities increased with the electron irradiation dose and electron energy.

Rotunno et al. [75] mechanically exfoliated MoS${}_2$ 20-layers and transferred them to carbon-coated copper grids. Figure 16a shows a transmission electron microscopy (TEM) image of a typic MoS${}_2$ few-layer. The multi-layers were then irradiated under a scanning electron microscopy (SEM) beam with an accelerating voltage of $5\mathrm{keV}$ and an electron-beam current of $50\mathrm{nA}$ for $15\min$. Many crystalline islands were formed throughout the multi-layers after the irradiation, as shown in Figure 16b. The selected-area electron diffraction (SAED) patterns indicated that the islands were metallic molybdenum. Under high-energy ($200\mathrm{keV}$) electron irradiation, the MoS${}_2$ layers were drastically modified, and their sulfur stoichiometry was changed, inducing the formation of the molybdenum nanoislands, as shown in Figure 16c. In both cases, massive surface sulfur depletion was induced together with the consequent formation of molybdenum nanoislands.

Kim et al. [82] deposited five to seven atomic layers of MoS${}_2$ on a SiO${}_2$/Si wafer at room temperature using the RF magnetron sputtering method. The thickness of the deposed amorphous MoS${}_2$ layers was about $4\mathrm{nm}$. The films were then exposed under a collimated electron beam with an energy of $1\mathrm{keV}$ for 1–10 min at room temperature without any additional heating processes. HRTEM images indicated that the as-deposited MoS${}_2$ was amorphous (left panel in Figure 17a). Under the electron beam irradiation (EBI) of $1\min$, the random atoms re-arranged from amorphous to crystalline structure, forming MoS${}_2$ crystal domains with a size about $5\mathrm{nm}$ (middle panel in Figure 17a). However, longer irradiation time would damage crystalline domains (right panel in Figure 17a) because of the Mo-S bond breaking. The amorphous-crystalline transformation was confirmed by Raman scattering, as shown in Figure 17b. The two prominent Raman peaks of MoS${}_2$, the in-plane mode $E_{2g}^1$ and the out-of-plane mode $A_g^1$, did not appear in the as-deposited film because of its amorphous nature. The peak intensities of the $E_{2g}^1$ and $A_g^1$ modes increased dramatically after the electron irradiation at room temperature. Their XPS measurements provided a stronger proof that the amorphous films crystallized under the $1\mathrm{keV}$ electron irradiation. Figure 17c shows the XPS spectra of MoS${}_2$ samples characterized with an Al-K${}_{\alpha }$ ray with an energy of $1486\mathrm{e}\mathrm{V}$. The Mo-$3d$ peak was de-convoluted into three chemical bonding states of Mo-Mo, Mo-S, and Mo-O in the as-deposited sample. After $1\min$ of electron irradiation, the Mo-$3d$ spectrum only consisted of two peaks originating from the states of the Mo-S and Mo-O bonds, with dramatically increased peak intensity of the Mo-S bonds. Similarly, the S-$2p$ peak from the as-deposited sample were de-convoluted into S-S and S-Mo bonding while only the S-Mo bond remained in the $1\min$ electron-irradiated sample. The authors also checked the samples under TEM at $400\mathrm{keV}$ and then measured the Raman scattering with a $532\mathrm{nm}$ laser. It was not clear if the amorphous films were changed under the TEM electron-beam irradiation and laser irradiation.

Karmakar et al. [77] investigated the electron irradiation effect of MoS${}_2$ few-layers through micro-Raman scattering. The few-layered MoS${}_2$ samples were prepared by mechanical exfoliation and irradiated under a vacuum of $10\times {10}^{-7}\mathrm{torr}$. The thickness of the few-layers was about $7\mathrm{n}\mathrm{m}$, which was equivalent to about ten layers of MoS${}_2$. Figure 18 shows the Raman scattering before and after irradiation. In addition to the in-plane $E_{2g}^1$ and out-of-plane $A_{1g}$ modes, extra Raman modes appeared after being irradiated under $5\mathrm{keV}$, $7\mathrm{keV}$, and $10\mathrm{keV}$, showing the maximum change under $10\mathrm{keV}$ irradiation. Obviously, the electron-irradiation-induced defects, resulting in the breaking of inversion symmetry. The $15\mathrm{keV}$ irradiation did not lead to any changes in the spectra because the electron beam passed the few-layers without creating any damage.

Matsunaga et al. [80] irradiated triangular-shaped MoS${}_2$ mono-layers grown on SiO${}_2$/Si substrates under an electron beam with an acceleration voltage of $15\mathrm{kV}$ and an areal dose of $280\mathsf{\mu }\mathrm{C}/\mathrm{c}{\mathrm{m}}^2$ charge. Figure 19a shows an optical image of a mono-layer MoS${}_2$ crystal. The right half region was covered by polymethyl methacrylate (PMMA) and was not electron irradiated. Raman scattering indicated that both regions (with and without irradiation) exhibited the $A_g^1$ and $E_{2g}^1$ peaks of MoS${}_2$. However, the two peaks were blue-shifted in the exposed region, as shown in Figure 19b. It was suggested that the electron-beam exposure introduced a local compressive strain in the MoS${}_2$ crystal. The induced strain would further change the band structure. The PL peak of the exposed left region, ∼1.85 $\mathrm{e}\mathrm{V}$, was blue-shifted 40–50 $\mathrm{m}$$\mathrm{e}\mathrm{V}$ compared with that of unexposed right region ($1.81\mathrm{e}\mathrm{V}$). Figure 19c shows the PL mapping of PL intensity. It was stated that the averaged PL spectrum of the unexposed region of the MoS${}_2$ mono-layer could be characterized by a broad peak at $1.81\mathrm{eV}$ while the exposed region of the same MoS${}_2$ sample showed a narrower feature at $1.85\mathrm{eV}$, corresponding to a blue shift of 40–50 meV.

Liu et al. [81] prepared a MoS${}_2$ mono-layer by micromechanical cleavage, as shown in Figure 20a. The well-crystalline MoS${}_2$ mono-layer was then exposed in a vacuum under a $80\mathrm{k}\mathrm{e}\mathrm{V}$ electron beam with an intensity of $40\mathrm{A}/\mathrm{c}{\mathrm{m}}^2$. Initial defects and holes were induced in the MoS${}_2$ sheet once exposed to the electron beam (Figure 20b). After a $81\mathrm{second}$ exposure under the electron irradiation, the small holes spread rapidly (Figure 20c) and extended into big holes with diameters of 3–6 nm (Figure 20d) after a $103\mathrm{s}$ irradiation. The MoS${}_2$ mono-layer converted to Mo${}_5$S${}_4$ nanoribbons after long-term electron irradiation.

Komsa et al. [78] prepared freestanding single-layer MoS${}_2$ samples by mechanical exfoliation of natural MoS${}_2$ bulk crystals, and observed a MoS${}_2$ sheet under an $80\mathrm{keV}$ electron beam in a vacuum on an aberration-corrected HRTEM. It was experimentally observed that the top and bottom sulfur atoms were removed under the $80\mathrm{keV}$ electron beam and vacancies were generated in the MoS${}_2$ mono-layers under the electron irradiation. The sulfur vacancies agglomerated into line defects due to migration of the defects [123].

Parkin et al. [79] prepared MoS${}_2$ single-crystalline mono-layers by CVD method and irradiated them with $200\mathrm{keV}$ electrons in an aberration-corrected TEM. Most generated defects were single sulfur vacancies. The irradiation-induced defects were quantitatively related to the electron dose with the actual defect concentration. Figure 21 shows the defect-induced stoichiometry as a function of electron dose. The elemental composition within the illuminated region was measured by electron X-ray dispersive spectroscopy (EDS). The EDS measurements indicated that the S/Mo atomic ratio was very close to the stoichiometric 2:1 ratio before electron irradiation. The M/Mo ratio decreased with the irradiation dose because of the sulfur removal under electron irradiation. It should be noted that the EDS intensity of the molybdenum signal remains constant during the irradiation, supporting the fact that the main effect of irradiation was to create sulfur vacancies while molybdenum atoms were much more difficult to be sputtered.

Electron-irradiation-induced phase transformation of MoS${}_2$ layers was also reported at high temperatures, such as 400–700 ${}^{\circ }\mathrm{C}$ in a vacuum [76]. Electron irradiation initiated the phase transition of 2H-MoS${}_2$ at high temperatures, and the transformation ratio increased with increasing electron doses above $40\times {10}^6$–$100\times {10}^6$ electrons $/\mathrm{n}{\mathrm{m}}^2$, depending on the temperature. It was also found that MoS${}_2$ layers were damaged when the total electron dose exceeded $5.0\times {10}^8$–$1.1\times {10}^9$ electrons$/\mathrm{n}{\mathrm{m}}^2$ with the accelerating voltage of $60\mathrm{kV}$.

2.7. Plasma Irradiation

Plasma has been employed to remove impurities and contaminants from surfaces by the collision energy of gas molecules and the chemical action on impurities and contaminants. Various gases (such as argon, oxygen, hydrogen, and nitrogen, as well as their mixtures) have been employed to carry out the cleaning procedures. The plasma is generated by using high-frequency voltages (typically kHz-MHz) to ionize the working gases. The activated species in plasma, including atoms, molecules, ions, electrons, free radicals, metastables, react with contaminants on surfaces. Short-wave ultraviolet (vacuum UV), whose energy is very effective to break most organic bonds of surface contaminants, is produced in plasma too. The technique has been utilized to modify surfaces of MoS${}_2$ few-layers in recent years.

2.7.1. Active Nitrogen N${}_2^{*}$

Michra et al. [98] investigated oxygen-plasma-irradiated MoS${}_2$ mono-layers deposited on sapphires at $450{}^{\circ }\mathrm{C}$. Figure 22 shows Raman scattering of irradiated MoS${}_2$ mono-layers. Raman A${}_{1g}$ mode shifted $1.79\mathrm{c}{\mathrm{m}}^{-1}$ towards a higher wavenumber and E${}_{2g}^1$$1.11\mathrm{c}{\mathrm{m}}^{-1}$ towards a lower wavenumber after 3 min of N${}_2$ plasma irradiation. The peak intensity ratio $I_{A_{1g}}/I_{E_{2g}^1}$ and FWHM of the $E_{2g}^1$ peak increased with plasma irradiation time. XPS investigations indicated that N-Mo bonds formed during the irradiation and the binding energies (calculated from Mo-$3d^{5/2}$, Mo-$3p^{3/2}$, and S-$2p^{3/2}$ peaks) shifted towards lower binding energy after N${}_2^{*}$ irradiation. The valence band maximum (VBM) reduced to $0.9\mathrm{eV}$ after $1\min$ of irradiation and $0.5\mathrm{eV}$ after $3\min$ of irradiation from $1.0\mathrm{eV}$ of pristine MoS${}_2$ mono-layer.

Azcatl et al. [124] doped MoS${}_2$ with nitrogen through a remote N${}_2$ plasma surface treatment. Nitrogen covalently bonded to MoS${}_2$ upon nitrogen plasma exposures and substituted chalcogen sulfur of MoS${}_2$. The nitrogen doping converted MoS${}_2$ to p-type and changed electrical properties. The nitrogen concentration in the doped MoS${}_2$ was controlled through adjusting N${}_2$ plasma exposure time. XPS measurements indicated that binging energies of Mo-3d${}_{5/2}$ and S-2p${}_{3/2}$ decreased with increasing exposure time.

Mono-layer MoS${}_2$ heterojunctions, such as intrinsic GaN/p-type MoS${}_2$ heterojunction, were also irradiated by N${}_2$ plasma under UHV conditions [98]. The values of VBM were reduced to $0.5\mathrm{eV}$ for N${}_2^{*}$ irradiated MoS${}_2$ layers.

2.7.2. Active Oxygen O${}_2^{*}$

Oxygen plasma species include ionized oxygen atoms O${}^{+}$, excited oxygen atoms O${}^{*}$, ionized oxygen molecules O${}_2^{+}$, metastable excited oxygen molecules O${}_2^{*}$, ozone O${}_3$, ionized ozone O${}_3^{+}$, excited ozone O${}_3^{*}$, and free electrons. Therefore, oxygen plasma can effectively clean and etch MoS${}_2$ 2D materials, introducing defects and doping oxygen to MoS${}_2$ 2D layers.

Nan et al. [125] treated MoS${}_2$ layers under oxygen plasma irradiation with $13.56\mathrm{MHz}$ and $5\mathrm{W}$ under $5\mathrm{Pa}$. As shown in Figure 23 and the inset, the PL intensity increased with increasing plasma irradiation time. The PL enhancement could be increased as high as 100 times over. Considered the unchanged Raman scattering and XPS information, it was concluded that oxygen plasma introduced defects and oxygen bonding in MoS${}_2$, enhancing PL intensity.

Chen et al. [126] treated MoS${}_2$ multi-layers under oxygen plasma in a reactive ion etcher. Initial n-type MoS${}_2$ layers were selected-area doped to p-type, forming p-n junctions in MoS${}_2$. The fabricated highly rectifying diodes exhibited high forward/reverse current ratios and a superior long-term stability at ambient conditions.

Kang et al. [127] treated MoS${}_2$ mono-layers with oxygen plasma. The photoluminescence evolved from a higher intense to completely quenched with increasing plasma exposure time because of a direct-to-indirect band-gap transition. The MoS${}_2$ lattice was distorted after oxygen bombardment and MoO${}_3$ disordered regions were generated in the MoS${}_2$ flakes.

Islam et al. [128] deposited MoS${}_2$ mono-layers (mechanically exfoliated from crystals) on Si/SiO${}_2$ wafers and exposed them under oxygen plasma for several seconds. The plasma treatments were carried out at a power of $100\mathrm{W}$ operating at $50\mathrm{kHz}$, using a gas mixture of oxygen (20%) and argon (80%) with a pressure of 250–350 mTorr. Raman spectroscopy and XPS spectroscopy indicated that MoS${}_2$ mono-layers were oxidized and MoO${}_3$ was produced in the reaction, $2{\mathrm{MoS}}_2+7{\mathrm{O}}_2\to 2{\mathrm{MoO}}_3+4{\mathrm{SO}}_2$. The resulting MoO${}_3$-rich domains significantly decreased the mobility and conductivity of the fabricated MoS${}_2$ mono-layer devices.

Ye et al. [129] exposed CVD-grown MoS${}_2$ mono-layers under oxygen plasma at a pressure of $10\mathrm{Torr}$ with oxygen gas. Figure 24 shows the morphologies of MoS${}_2$ exposed to oxygen plasma for $10\mathrm{s}$, $20\mathrm{s}$, and $30\mathrm{s}$. After $10\mathrm{s}$ of oxygen plasma treatments, short and isolated cracks were observed on the continuous basal plane. Angles of those connected cracks were around $120{}^{\circ }$. After $20\mathrm{s}$ of oxygen plasma treatments, cracks became longer and were connected to each other, forming a continuous network. When treated for $30\mathrm{s}$, the widths of the cracks were further enlarged, most angles between the interconnected cracks were $120{}^{\circ }$, and the MoS${}_2$ was decomposed into even smaller fragments with a greater number of exposed edges. The structural change significantly decreased the intensity of A${}_{1g}$ and E${}_{2g}^1$ Raman modes of the MoS${}_2$ mono-layers, and shifted the A${}_{1g}$ mode to shorter wave-numbers and the E${}_{2g}^1$ mode to longer wave-numbers. The exposure also decreased the PL intensity. The changes of the Raman and PL spectra indicated that oxygen plasma can lower the MoS${}_2$ crystal symmetry and increase the lattice distortion, which can be attributed to the defects that may benefit MoS${}_2$ as electrochemical catalyst.

Dhall et al. [130] treated MoS${}_2$ few-layers (15 layers with approximately $12\mathrm{nm}$ thickness) in oxygen plasma for $3\min$. The plasma was generated by flowing air past an electrode supplied with $20\mathrm{W}$ of RF power at $200\mathrm{mTorr}$. The PL efficiency of the few-layers was increased after the plasma treatments.

Kim et al. [131] grew triangular MoS${}_2$ mono-layers on SiO${}_2$/Si substrates by CVD technique. The MoS${}_2$ mono-layers were then subjected to plasma-oxygen treatment for times ranging from $10\mathrm{s}$ to $120\mathrm{s}$. Ultra-high pure oxygen gas (99.9999%) was activated by a RF plasma cell, and the working pressure was $1.3\times {10}^{-3}\mathrm{Pa}$ under an ultra-high pure oxygen gas flow of $2\mathrm{sccm}$. Optical properties of the mono-layers, such as PL and Raman scattering, changed with treatment time. Figure 25a shows optical images of the pristine MoS${}_2$ mono-layer and these MoS${}_2$ mono-layers treated with oxygen plasma for $10\mathrm{s}$, $30\mathrm{s}$, $60\mathrm{s}$ and $120\mathrm{s}$. Figure 25b shows integrated PL intensity mapping images obtained from each of the mono-layers. The integrated PL intensity decreased with the oxygen plasma treatment time. Figure 25c shows the peak position of the main PL peak. The main peak position of the PL spectra gradually red-shifted from $674\mathrm{nm}$ to $692\mathrm{nm}$ with an increasing oxygen plasma treatment duration. It was suggested that shallow defect states were generated by the oxygen plasma treatments.

2.7.3. Active Hydrogen H${}_2^{*}$

Ye et al. [129] exposed CVD-grown MoS${}_2$ mono-layers in hydrogen plasma at different temperatures (400–700 ${}^{\circ }\mathrm{C}$), as shown in Figure 26. The MoS${}_2$ mono-layer were treated to expose more active sites in the basal plane of MoS${}_2$ and to improve catalytic activity. No significant changes were observed on the MoS${}_2$ when the annealing temperature was $400{}^{\circ }\mathrm{C}$. The MoS${}_2$ was then H${}_2$-treated at $500{}^{\circ }\mathrm{C}$ and small triangular holes with sizes around 1–4 $\mathsf{\mu }$$\mathrm{m}$ appeared. High-density holes with sizes around 10–20 $\mathrm{n}$$\mathrm{m}$ were omnipresent in the basal plane of MoS${}_2$. High-density triangular-shaped holes became the prominent part in the original mono-layer MoS${}_2$ at $600{}^{\circ }\mathrm{C}$. $700{}^{\circ }\mathrm{C}$ H${}_2$-treatments led to severe MoS${}_2$ decomposition. The hydrogen plasma treatments decreased the peak intensities of Raman scattering and PL spectra, indicating that hydrogen treatments could increase the defects and edges in mono-layer MoS${}_2$.

2.7.4. Other Molecules

Fluoride

Chen et al. [126] treated MoS${}_2$ multi-layers under fluoride plasma (SF${}_6$, CHF${}_3$, and CF${}_4$) for selected-area p-doping to form p-n MoS${}_2$ junctions and diodes.

Chlorine

Murray et al. [116] irradiated exfoliated MoS${}_2$ few-layers (5–15 layers) under low-energy Cl ($200\mathrm{eV}$ and $1\times {10}^{13}$–${10}^{15}$ ions$/\mathrm{c}{\mathrm{m}}^2$). The natural n-type MoS${}_2$ layers were doped to p-type and the electric conductivity was reduced after irradiation.

Phosphorus

Nipane et al. [132] reported a compatible, controllable, and area selective phosphorus plasma immersion ion implantation process for p-type doping of mechanically exfoliated MoS${}_2$ layers using PH${}_3$–He plasma. Homogeneous p-n MoS${}_2$ junction diodes were fabricated.

Gallium

Mono-layer MoS${}_2$ heterojunctions, such as intrinsic GaN/p-type MoS${}_2$ heterojunction, were also irradiated by Ga plasma under UHV conditions [98]. The values of VBM were reduced to $0.2\mathrm{eV}$ for Ga-irradiated MoS${}_2$ layers.

Combination

Jadwiszczak et al. [133] exposed mechanically exfoliated MoS${}_2$ few-layers to an O${}_2$–Ar plasma (O${}_2$:Ar = 1:3) for 2–28 $\mathrm{s}$. The frequency of the plasma was $13.52\mathrm{MHz}$. An oxide phase was generated under the plasma exposure, changing the electrical conductivity and carrier mobility of the 2D materials.

Mishra et al. [98] irradiated MoS${}_2$ under N${}_2^{*}$ plasma for 1–3 min and then under Ga flux. The VBM was changed.

Bhimanapati et al. [134] treated MoS${}_2$ vertical layers in UV–ozone. The ozone treatment increased super wettability and enhanced hydrogen evolution reaction of the material by changing edge chemistry and surface defects.

Nguyen et al. [19] treated MoS${}_2$ layers in UV-ozone and then fabricated the layers into organic photovoltaic cells. The open-circuit voltage, fill factor, and power conversion efficiency increased significantly compared with un-irradiated MoS${}_2$-based solar cells.

3. Electromagnetic Irradiation

Electromagnetic irradiation refers to the waves of the electromagnetic field, including gamma rays, X-rays, ultraviolet, (visible) light, infrared, microwaves, and radio waves. The frequency and wavelength of the electromagnetic irradiation is shown in Figure 3.

3.1. Gamma-ray Irradiation

It is generally accepted that MoS${}_2$ macroscopic materials are stiff under gamma-ray irradiation. There is little literature on $\gamma$-irradiated bulks. It was reported that MoS${}_2$ powders were resistant to a $\gamma$-ray irradiation with a dose of $5\times {10}^9\mathrm{R}$ ($1.29\times {10}^6\mathrm{C}/\mathrm{k}\mathrm{g}$) [105].

Lee et al. [106] mechanically exfoliated MoS${}_2$ 2D layers (50–132 layers) and exposed them to 5000 $\gamma$-ray photons with $662\mathrm{k}\mathrm{e}\mathrm{V}$. It was found that the resonance frequency of the layers upshifted immediately after $\gamma$-ray exposure and returned to their initial frequency after $60\mathrm{h}$. The procedure was repeatable. It was assumed that $\gamma$-ray photons generated charges on MoS${}_2$ and caused electrostatic forces between MoS${}_2$ and substrates, resulting in electrostatic tension and deflection of MoS${}_2$ layers. The MoS${}_2$ multi-layers were stiff under the $\gamma$-ray and showed no irradiation damage.

Ozden et al. [107] prepared MoS${}_2$ multi-layers (5–8 layers) through the vapor phase sulfurization and exposed the layers to a $\gamma$-ray irradiation of ${}_{27}^{60}$Co source ($\gamma$-ray energy: $1.1732\mathrm{M}\mathrm{e}\mathrm{V}$ and $1.3324\mathrm{M}\mathrm{e}\mathrm{V}$) with a dose of $120\mathrm{M}\mathrm{rad}$. The irradiation was carried out at room temperature in an ambient atmosphere. The X-ray photoelectron spectra indicated that MoS${}_2$ layers were converted to molybdenum oxide (MoO${}_x$) after the irradiation. It is plausible that the $\gamma$-ray displaced or knocked out S atoms while leaving molybdenum atoms unaffected. Maybe the as-produced sulfur vacancies were filled with oxygen atoms to form MoO${}_x$.

3.2. X-ray Irradiation

There are few reports on X-ray irradiation of MoS${}_2$. Only one paper [108] reported that MoS${}_2$ mono-layers were irradiated with $10\mathrm{keV}$ X-rays with varying total ionizing doses. It was reported that the MoS${}_2$ mono-layers were robust to X-ray radiation, withstanding doses of up to $6\mathrm{Mrad}$ doses without any noticeable degradation of optical properties. It is generally accepted that X-rays do not affect MoS${}_2$ few-layers.

Zhang et al. [135] fabricated single-layer MoS${}_2$ FETs and irradiated them under $10\mathrm{k}\mathrm{e}\mathrm{V}$ X-ray exposure. At room temperature in air, the drain current of the devices decreased significantly under X-ray irradiation up to $10\mathrm{Mrad}$. Effective threshold voltage and mobility were degraded with X-ray irradiation dose. It stated that the degradation was consistent with the generation of negatively charged surface states during the X-ray exposure.

3.3. Ultraviolet Light Irradiation

Azcatl et al. [109] treated MoS${}_2$ crystals under ultraviolet–ozone exposures at room temperature. Oxygen-sulfur bonds were formed at the top sulfur layer of the MoS${}_2$ surface without breaking sulfur-molybdenum bonds. Li et al. [136] investigated optical behaviors of azobenzene-functionalized MoS${}_2$ mono-layers and exfoliated multi-layers on Au substrates. It was reported that UV light could tune doping and the Fermi level of the hybrid structures. Lu et al. [137] irradiated liquid-exfoliated MoS${}_2$ nanosheets in aqueous solutions and found that the MoS${}_2$ layers were oxidatively etched because of photon-induced powerful OH${}^{*}$ radicals.

Singh et al. [138] illustrated multilayer MoS${}_2$ FETs under ultraviolet light in N${}_2$ atmospheres. The multi-layers were micromechanically exfoliated from natural MoS${}_2$ crystals and deposited on SiO${}_2$ coated silicon. The wavelength of the UV light was $220\mathrm{nm}$ and the average intensity was $10\mathrm{mW}/{\mathrm{cm}}^2$. Nitrogen gas alone did not affect the electrical properties of MoS${}_2$ nanosheets, but nitrogen gas in the presence of UV light remarkably affected the electrical properties of MoS${}_2$ nanosheets. Charge-carrier mobility, carrier density, and drain current were enhanced after exposure to nitrogen gas under UV light irradiation because of a possible doping effect. Detailed investigations [139] showed that the charge-carrier density of single-layer, bi-layer, and few-layer MoS${}_2$ nanosheets were reversibly tuned with nitrogen and oxygen gas in the presence of ultraviolet light. The device performance was adjusted by exposure to gases in the presence of UV light.

McMorrow et al. [110] fabricated single-layer and multilayer MoS${}_2$ (mechanically exfoliated) on Si substrates into FETs. It was found that the electron mobility increased with UV exposure up to 3.4 $\times {10}^{10}$–2.2 $\times {10}^{13}$ photos$/\mathrm{c}{\mathrm{m}}^2$ in a vacuum. Raman spectroscopy showed no significant crystalline radiation damage or oxidation degradation under UV exposure. Irradiation-hard MoS${}_2$ FET devices are expected.

3.4. Visible Light Irradiation

Liu et al. [140] tested photocatalytic performance of N-doped MoS${}_2$ nanoflowers under visible light. It was reported that the N-doped MoS${}_2$ nanoflowers showed excellent photocatalytic activities and durability on the elimination of the organic pollutants under visible light irradiation.

Laser has been widely employed in Raman scattering of MoS${}_2$ few-layers. Laser has also been used to thin multilayered MoS${}_2$ down to a single-layer two-dimensional crystal [141,142] (power density up to 80–140 mW$/\mathsf{\mu }{\mathrm{m}}^2$) and generate ripples of MoS${}_2$ [142,143]. It is reported that the upper atoms of MoS${}_2$ layers can be removed by high-power lasers because of laser ablation [112,141].

Early investigation indicated that natural MoS${}_2$ crystals do not undergo any kind of significant laser-assisted oxidation when exposed to high laser power (up to 32,000 $\mathrm{W}/\mathrm{c}{\mathrm{m}}^2$) (wavelength: $632.8\mathrm{nm}$) [144] while their Raman mode intensity changes slightly with laser power. For microcrystalline MoS${}_2$ powders, Raman scattering indicated that MoS${}_2$ oxidized to MoO${}_2$ under a high-power laser. Raman mode intensity and position were also significantly affected by the laser power.

Paradisanos et al. [113] mechanically exfoliated MoS${}_2$ few-layers from a bulk natural crystal and subsequently deposited them on Si/SiO${}_2$ wafers, irradiated the few-layers under a pulsed laser with $800\mathrm{nm}$ wavelength and $1\mathrm{kHz}$ repetition rate. No modification of MoS${}_2$ mono-layers was observed up to a certain single-pulse fluence of $50\mathrm{mJ}/\mathrm{c}{\mathrm{m}}^2$ ($2.5\mathrm{mW}$) while damage occurred beyond the fluence via the material ablation. Further work indicated that MoS${}_2$ mono-layers were practically unaffected by a low-power ($600\mathsf{\mu }\mathrm{W}$) pulsed laser irradiation when exposed to ${10}^3$ pulses at a fluence of $20\mathrm{mJ}/\mathrm{c}{\mathrm{m}}^2$ (lower than the damage threshold). However, the MoS${}_2$ mono-layers were damaged upon ${10}^5$ pulses at $20\mathrm{mJ}/\mathrm{c}{\mathrm{m}}^2$. The $A_{1g}$ and $E_{2g}$ intensities were almost constant with pulse times, then rapidly decreased at a critical exposure time. The abrupt decrease possibly came from ablation and eventual sublimation of the MoS${}_2$ atoms.

Gu et al. [111] in situ studied Raman scattering of MoS${}_2$ layers during the laser thinning of MoS${}_2$. Due to the high surface-area-to-volume ratio, thinner MoS${}_2$ layers are less stable and easier to decompose under high-power laser irradiation because of laser-induced thermal effects and sublimation.

Tran Khac et al. [145] irradiated MoS${}_2$ layers under laser power of $1\mathrm{mW}$, $5\mathrm{mW}$, and $10\mathrm{mW}$ for an exposure time of $60\mathrm{s}$. No significant changes were observed in topographic images of the irradiated regions under the $1\mathrm{mW}$ laser. However, significant amounts of particles or adsorbates were formed on the MoS${}_2$ surface after irradiated under $5\mathrm{mW}$ and $10\mathrm{mW}$ lasers.

To minimize thermal effects on atomically thin MoS${}_2$ during Raman spectral measurements, low laser powers ($0.14\mathrm{mW}$ to $2\mathrm{mW}$) are usually employed [145] to avoid potential laser-induced local surface temperatures.

Lu et al. [112] employed a focused laser beam to directly pattern MoS${}_2$ mono-layers and few-layers. Focused laser beam irradiation modified and thinned MoS${}_2$ layers to create well-defined structures and controllable thickness.

Alrasheed et al. [146] used a $532\mathrm{nm}$ laser to irradiate mechanically exfoliated MoS${}_2$ few-layers (less than $5\mathrm{layers}$) on SiO${}_2$/Si substrates. The power ranged between 0.93 and 8.3 $\mathrm{m}$$\mathrm{W}$ and the treatment time was 0.01–180 $\mathrm{s}$. In ambient conditions, MoS${}_2$ nanosheets were etched and amorphous MoS${}_2$ redeposited on the nanosheets at low laser powers while the few-layers were oxidized and MoS${}_2$ nanoparticles formed at high laser powers. The nanoparticle formation and oxidation were dependent on the number of layers and laser exposure time. The Raman intensity and Raman peak width changed with laser treatments, as shown in Figure 27.

3.5. Infrared Light Irradiation

Fan et al. [30] irradiated lithium intercalated 1T MoS${}_2$ layers with thickness of a few microns under a near-IR laser ($780\mathrm{n}\mathrm{m}$) in air at ambient temperature. The MoS${}_2$ layers were completely converted to 2H counterpart under the irradiation.

2H MoS${}_2$ multi-layers were grown through pulsed mid-infrared laser deposition [147] (wavelength of 7.0–8.2 $\mathsf{\mu }$$\mathrm{m}$). No any IR damage was reported.

3.6. Terahertz Wave Irradiation

MoS${}_2$ multi-layers were fabricated into terahertz modulators [148] applied in high-speed communications. Terahertz conductivities of MoS${}_2$ few-layers were experimentally measured [149,150] from THz spectroscopy. However, there was no report on terahertz damage on MoS${}_2$. Theoretical simulations indicated that the THz absorption of mono-layer MoS${}_2$ was very low and the maximum THz absorption of mono-layer MoS${}_2$ was approximately 5% [151]. The sum of reflection and absorption losses of mono-layer MoS${}_2$ was lower than that of graphene by one to three orders of magnitude.

3.7. Microwave Irradiation

Few-layer MoS${}_2$ were fabricated into two-dimensional nanoelectromechanical systems (NEMS) as ultralow-power, high-frequency tunable oscillators and ultrasensitive resonant transducers [152]. These devices can operate in the very high frequency band (up to ∼120 MHz). Ultra-thin MoS${}_2$ was also fabricated to MoS${}_2$ transistors operating at gigahertz frequencies [153,154]. All these devices worked at microwave range. Therefore, it is necessary to characterize the microwave damages to MoS${}_2$ few-layers.

It was reported that microwave can accelerate catalytic reactions [155] because of the formation of hot-spots within catalysts. However, no significant differences were detected between microwave irradiated MoS${}_2$ catalysts (size in microns) and untreated MoS${}_2$ during the sulfating reaction [155].

Zhao et al. [156] irradiated MoS${}_2$ powders under a microwave ($1\mathrm{kW}$ power, $2450\mathrm{MHz}$) at $200{}^{\circ }\mathrm{C}$. It was reported that the (001) basal planes of the MoS${}_2$ crystal structure were cracked into (100) edge planes under microwave irradiation, creating additional active edge sites. The Raman intensity of microwave irradiated materials increased significantly after microwave treatments. The irradiated MoS${}_2$ could capture Hg${}^0$ efficiently because of the microwave-induced cracks.

Xu et al. [31] treated chemically exfoliated MoS${}_2$ nanosheets in an inert nonpolar solvent of 1,2-dichlorobenzene under $2.45\mathrm{GHz}$ microwave irradiation at $130{}^{\circ }\mathrm{C}$ in an inert atmosphere, with an output power of $500\mathrm{W}$. 2H MoS${}_2$ nanosheets converted to the 1T phase in minutes.

4. Other Irradiation

4.1. Ultrasonic Wave Irradiation

MoS${}_2$ nanosheets have been prepared by ultrasonication through liquid exfoliation technique [19,41,46,71]. The physical properties were affected by ultrasonication conditions. For example, Gao et al. [157] sonicated MoS${}_2$ powders in N,N-dimethylformamide for 2–10 $\mathrm{h}$ to chemically exfoliate MoS${}_2$ nanosheets. Crystalline MoS${}_2$ few-layers were produced after $2\mathrm{h}$ sonication and the size of the nanosheets decreased gradually with increasing sonication time. The exfoliated MoS${}_2$ nanosheets showed ferromagnetic properties at room temperature, in contrast to pristine MoS${}_2$ bulks which showed diamagnetism only. The saturation magnetizations of the ultrasonicated nanosheets increased with the ultrasonication time and the decreasing crystalline size.

MoS${}_2$ nanomaterials were also prepared by ultrasonicating slurrys of molybdenum hexacarbonyl and sulfur in 1,2,3,5-tetramethylbenzene (isodurene) with a high-intensity ultrasound ($20\mathrm{kHz}$) under Ar atmosphere [158]. The nanostructured MoS${}_2$ had a higher surface area and showed higher catalytic activities for thiophene hydrodesulfurization.

4.2. Thermal Irradiation

MoS${}_2$ bulks sublimate at $450{}^{\circ }\mathrm{C}$ and melt at $2375{}^{\circ }\mathrm{C}$. MoS${}_2$ crystals are not stable in the presence of oxygen, resulting in MoO${}_3$. It was reported that the material oxidized at 315–375 ${}^{\circ }\mathrm{C}$ in air and converts to MoO${}_3$ at $400{}^{\circ }\mathrm{C}$ [144].

MoS${}_2$ mono-layers are not very stable at ambient conditions either. Peto et al. [159] examined atomic-resolution STM images of the basal plane of mechanically exfoliated MoS${}_2$ mono-layers after one month and one year of ambient exposure (in air, at room temperature and ambient light). STM measurements clearly revealed modifications in the atomic structure of the MoS${}_2$ basal plane during the ambient exposure. New point-defects were formed after one month of the ambient exposure and the defect concentration increased 20–30 times. The point-defects increased 100–500 times after one year of exposure. The oxidation of MoS${}_2$ basal planes yielded 2D MoS${}_{2-x}$O${}_x$ layers.

Chen et al. [117] annealed the liquid-exfoliated MoS${}_2$ mono-layers on various substrates (SiO${}_2$, Au, graphene, BN, and CeO${}_2$) in a vacuum. Thermal annealing did not introduce any noticeable defects into MoS${}_2$ layers up to $450{}^{\circ }\mathrm{C}$ and MoS${}_2$ remained stable, as shown in Figure 11. Other group reported that thermal annealing in a vacuum caused S vacancies in MoS${}_2$ at $500{}^{\circ }\mathrm{C}$ [90].

Donarelli et al. [160] annealed MoS${}_2$ crystals in UHV. Single sulfur vacancies were generated in bulks while the S/Mo ratio in the bulk did not change with the annealing temperature up to $400{}^{\circ }\mathrm{C}$. They also examined the liquid-exfoliated MoS${}_2$ few-layers. The S/Mo ratio of MoS${}_2$ few-layers gradually decreased once the annealing temperature increased up to $400{}^{\circ }\mathrm{C}$. Two kinds of sulfur vacancies (single and double) were introduced in the molybdenite layers. The threshold for the formation of double vacancies typically occurred upon thermal annealing at $200{}^{\circ }\mathrm{C}$.

Nan et al. [125] mechanically exfoliated MoS${}_2$ mono-layers from bulk crystals and transferred them to Si wafers with SiO${}_2$ capping layer. The mono-layers were then annealed at temperature of $350{}^{\circ }\mathrm{C}$ or $500{}^{\circ }\mathrm{C}$ for $1\mathrm{h}$ in a vacuum ($0.1\mathrm{Pa}$). Figure 28a shows PL intensity image of an as-prepared MoS${}_2$ mono-layer, which was uniform across the whole sample. After being annealed for $1\mathrm{h}$ at $350{}^{\circ }\mathrm{C}$ in a vacuum, the PL intensity was enhanced by 6-fold (Figure 28b). At the same time, the PL peak blue-shifted after the annealing. Figure 28c,d show another MoS${}_2$ mono-layer before and after annealing at $500{}^{\circ }\mathrm{C}$ for $1\mathrm{h}$ in a vacuum. The PL image becomes highly inhomogeneous after the annealing. More detailed investigation indicated that the PL enhancement and inhomogeneity should come from the reaction of oxygen (due to not very high vacuum condition) with MoS${}_2$ at high temperature and followed chemical bonding of oxygen molecules to MoS${}_2$.

Yamamoto et al. [161] mechanically exfoliated MoS${}_2$ single-layers and few-layers from MoS${}_2$ bulk crystals, deposited them onto $300\mathrm{nm}$ thick SiO${}_2$/Si substrates. The MoS${}_2$ layers were then exposed to an Ar/O${}_2$ mixture at temperatures ranging from $27{}^{\circ }\mathrm{C}$ to $400{}^{\circ }\mathrm{C}$. The flow rates of Ar and O${}_2$ were $1.0\mathrm{L}/\min$ and $0.7\mathrm{L}/\min$, respectively. Raman spectroscopy measurements indicated that the oxygen treatments led to triangular etch pits on the surfaces of the atomically thin MoS${}_2$. MoS${}_2$ was etched preferentially along the crystallographic directions of the zigzag edges with a preferential termination, resulting in uniform orientations of the pits, as shown in Figure 29a–e. The pit size increased with increasing oxidation exposure (Figure 29f), and the growth rate was larger at higher temperature. However, the numbers of etch pits per unit area was uncorrelated with oxidation time, oxidation temperature, and MoS${}_2$ thickness but varied significantly from sample to sample. It was assumed that the oxidative etching in MoS${}_2$ layers on SiO${}_2$ was initiated at intrinsic defect sites. Additionally, oxygen exposure above $200{}^{\circ }\mathrm{C}$ significantly diminished the electron density in MoS${}_2$ single-layers and oxygen treatments at $400{}^{\circ }\mathrm{C}$ resulted in conversion of MoS${}_2$ layers to MoO${}_3$ platelets. No molybdenum oxide (MoO${}_3$) was detected after oxygen treatments below $340{}^{\circ }\mathrm{C}$.

Tongay et al. [162] annealed MoS${}_2$ single-layers in a vacuum. Figure 30 shows room temperature PL spectra of a MoS${}_2$ mono-layer annealed at $450{}^{\circ }\mathrm{C}$ in a vacuum for different annealing time. A $40\min$ annealing at $450{}^{\circ }\mathrm{C}$ enhanced the PL intensity by over 50 times. The FWHM of the PL peak decreased and the PL peak position shifted slightly with annealing time. The thermal annealing did not degrade the crystalline quality of the material.

4.3. Hybrid Irradiation: UV–Ozone Treatments

Ultraviolet–ozone cleaning technique is a dry-cleaning method to clean material surfaces by the decomposition of contaminants through ultraviolet irradiation and the chemical action of oxidation by ozone. During cleaning, surface organic compounds decompose to volatile substances by ultraviolet lights and by strong oxidation of ozone. The technique has been employed to irradiate 2D MoS${}_2$ mono-layers [163,164] and few-layers [109,165,166,167].

Wang et al. [164] rapidly treated MoS${}_2$ 1–3 layers that were mechanically exfoliated from crystals. These was no any change before and after the UV–ozone treatments. It was assumed that the irradiation effect was below detection because of the rapid treatment ($30\mathrm{s}$) and nitrogen protection that reduced high oxidation and mobility change. Le et al. [165] prepared MoS${}_2$ nanosheets using a sonication exfoliation method and UV–ozone treated the MoS${}_2$ multi-layers for 15–30 min. The MoS${}_2$ layers were converted to MoO${}_x$. Burman et al. [167] treated liquid-exfoliated MoS${}_2$ few-layers (15–16 layers) in UV–ozone atmosphere for $1.5\mathrm{h}$. It was found that the Mo/S atomic ratio decreased, and atomic percentage of oxygen increased with the expose time. Azcatl et al. [109] exposed the mechanically exfoliated MoS${}_2$ layers under ultraviolet-ozone at room temperature. They found that oxygen-sulfur bonds were formed at the top sulfur layer while the sulfur-molybdenum bonds were kept besides removal of adsorbed carbon contamination from the MoS${}_2$ surface.

The ultroviolet–ozone technique was also used to mill MoS${}_2$ few-layers to mono-layers [166]. Compared to mechanically exfoliated MoS${}_2$ single-layers, the PL intensity of the UV–ozone milled MoS${}_2$ mono-layers was enhanced by 20–30 times.

5. Irradiation Mechanism and Theoretical Simulations

Besides the experimental work reviewed in the previous sections, there are lots of theoretical approaches to investigate the irradiation mechanism of MoS${}_2$ 2D nanosheets.

Komsa et al. [78] calculated the displacement threshold energy and the minimum initial kinetic energy of the recoil atom to kick an atom from MoS${}_2$ mono-layers, using the density-functional theory with the Perdew-Burke-Ernzerhof exchange-correlation function. The sulfur displacement threshold energy of MoS${}_2$ was $6.9\mathrm{eV}$, corresponding to incident-electron energies of about $90\mathrm{keV}$. About $20\mathrm{eV}$ was required to displace one molybdenum atom from its site in MoS${}_2$ mono-layer lattices, corresponding to an electron energy of $560\mathrm{keV}$. Therefore, the sulfur sub-layers were easier destroyed while it was unlikely to generate molybdenum vacancies in molybdenum sub-layers under TEM observations (where $200\mathrm{keV}$ electron beams were employed). Figure 31 shows the calculated cross-section of sulfur atoms. Lattice vibrations were taken into account, assuming a Maxwell-Boltzmann velocity distribution. The cross denotes the experimentally determined cross-section for MoS${}_2$. The inset shows the same data for a larger range of electron energies. The atom displacement cross-section was estimated from the electron threshold energy by using the McKinley-Feshbach formalism.

The calculated displacement threshold energy was in agreements with experimental observations. Single sulfur vacancies (VS’s) were frequently observed under transmission electron microscopes with an $200\mathrm{kV}$ acceleration voltage (the electron energy, $200\mathrm{keV}$ is higher than the displacement energy of sulfur atoms ($90\mathrm{keV}$) while lower than that of molybdenum atoms, $560\mathrm{keV}$).

Table 2. Comparison of formation thresholds in eV for various vacancy configurations of MoS${}_2$ bulks and mono-layers [77].

Comfiguration	Bulk (eV)	Mono-layer (eV)
single S-vacancy	14.4	5.7
single Mo-vacancy	24.8	18.8
single-line S-vacancy	39.2	30.5
double-line S-vacancy	66.2	62.3
DIV	28.8	23.3

DIV: di-vacancy comprised of nearest neighbor single sulfur vacancies and single molybdenum vacancies.

compares formation thresholds of five kinds of defect configurations for MoS${}_2$ bulks and mono-layers. The thresholds for single S vacancies were very low, both in bulk and mono-layers. Single-line S-vacancy and double-line S-vacancy configurations are generated with increasing single S-vacancy concentration by prolonged electron irradiation. In addition, the formation thresholds for Mo-vacancy and DIV are not very high (18–29 eV).

Kretschmer et al. [168] combined analytical potential molecular dynamics with Monte Carlo simulations to simulate helium and neon ion irradiation of MoS${}_2$ nanosheets deposited on SiO${}_2$ substrates. Their simulation indicated that substrates governed the defect production of irradiated MoS${}_2$ layers. The irradiation-induced defect production was dominated by backscattered ions and sputtered substrate atoms, rather than by the direct helium and neon ion impacts.

Molecular dynamics simulations were also employed to study the production of defects in MoS${}_2$ mono-layers under noble gas ion irradiation, such as He, Ne, Ar, Kr and Xe [169]. Sulfur atoms were sputtered away predominantly from the top or bottom layers by the ion irradiation, depending on the incident angle, ion type, and ion energy.

6. Irradiation-Induced Properties

Pure 2H-MoS${}_2$ crystalline bulks are indirect band-gap semiconductors and show diamagnetic from $10\mathrm{K}$ to room temperature. Their physical properties are sensitive to structures and defects. Irradiation produces defects in MoS${}_2$ few-layers and affects their properties significantly. Below, the optical properties, electronic properties, catalytic properties, and magnetic properties are summarized.

6.1. Band Structures

Bulk MoS${}_2$ is a semiconductor with an indirect band gap of $1.23\mathrm{eV}$ [36,170]. The band gap of the material depends on thickness [171], as shown in Figure 2. Figure 32a shows the band structure of MoS${}_2$ mono-layers without vacancies. MoS${}_2$ mono-layers have a direct band gap of $1.8\mathrm{eV}$, resulting in strong absorption, PL bands [38,39], and electroluminescence [172] near $680\mathrm{nm}$.

Defects affect the band gap of MoS${}_2$ layers significantly. Figure 32b,c shows calculated band structures of MoS${}_2$ mono-layers with defects. Negative (S vacancies) and positive (Mo vacancies) charges transferred to the orbitals of nearest neighbor Mo or S atoms around the vacancies, resulting in additional states at the bonding level and at mid-gap of the band structures. The K-point direct band gap remained intact for MoS${}_2$ mono-layers with S vacancies, while the gap value was reduced. The direct band gap at K-point was reduced too for MoS${}_2$ mono-layers with Mo vacancies.

6.2. Electric Properties

The density-functional method and the Green’s function approach indicated that structural defects and grain boundaries were principal contributors to electric transport properties of MoS${}_2$ mono-layers [173]. Atomic vacancies could significantly reduce the conductance of the 2D layers. In experiments, MoS${}_2$ 2D materials were deposited on various substrates and connected to metal electrodes to fabricate FETs to detect electric properties under irradiation.

Durand et al. [174] studied electrical transport properties of CVD-grown MoS${}_2$ mono-layers on SiO${}_2$/Si substrates. The carrier density was significantly increased, and the mobility was reduced under a low-energy electron irradiation ($5.0\mathrm{k}\mathrm{e}\mathrm{V}$) in an UHV environment. It was believed that the electron-irradiation generated defects in MoS${}_2$ layers and caused Coulomb potentials at the MoS${}_2$/SiO${}_2$ interfaces.

Lu et al. [175] grew MoS${}_2$ single-crystalline few-layers on SiO${}_2$/Si substrates at $700{}^{\circ }\mathrm{C}$ through CVD method, and then transferred them to pre-patterned SiO${}_2$/Si substrates with Au electrodes by the poly(methyl methacrylate)-assisted method. The fabricated MoS${}_2$ devices were then electron-irradiated in a scanning electron microscope with a $30\mathrm{keV}$ electron beam. The electron dose was 100–1800 $\mathsf{\mu }\mathrm{C}/\mathrm{c}{\mathrm{m}}^2$. It was found that the threshold voltage shifted to the negative side and the mobility increased upon the increasing electron doses.

Parkin et al. [79] deposited single-crystalline MoS${}_2$ mono-layers onto SiN${}_x$ membranes, contacted them to Au electrodes. The fabricated device was then irradiated under a TEM electron beam with $200\mathrm{keV}$. It was found that the electrical resistance increased with the electron irradiation dose.

Kim et al. [86] prepared few-layer MoS${}_2$ FETs, and investigated the effect of irradiation under $10\mathrm{MeV}$ proton exposures. The electrical properties of the devices were nearly unchanged in response to low fluence proton irradiation (${10}^{12}\mathrm{protons}/\mathrm{c}{\mathrm{m}}^2$). The current and conductance of the devices significantly decreased after high fluence irradiation (${10}^{13}$–${10}^{14}$ protons$/\mathrm{c}{\mathrm{m}}^2$). The electrical changes contributed to proton-irradiation-induced traps in the SiO${}_2$ substrate layers and at the interfaces between the MoS${}_2$ layers and the SiO${}_2$ layers.

Islam et al. [128] measured electrical properties of MoS${}_2$ mono-layers exposed under oxygen plasma. The plasma was generated from a gas mixture of oxygen (20%) and argon (80%) under a pressure of 250–350 mTorr at $50\mathrm{kHz}$. Figure 33a shows the $I_D$-$V_{DS}$ curves of the device at a back-gate voltage $V_G=40\mathrm{V}$ for various plasma exposure time. The $I_D$-$V_{DS}$ curves were linear around the zero bias, showing Ohmic behaviors. Figure 33b demonstrates the exposure-time dependence of electric resistance. The resistance increased exponentially with increasing plasma exposure time. The drain current and mobility decreased exponentially with plasma exposure time, dropping more than four orders of magnitude after only a total of $6\mathrm{s}$ plasma exposure. It was claimed that the significant degradation of electronic properties was caused by the creation of insulating MoO${}_3$-rich disordered domains in the MoS${}_2$ sheet upon oxygen plasma exposure.

6.3. Catalytic Properties

Molybdenum is relatively abundant and cheaper than noble metals. Thus, molybdenum compounds have been used as electrocatalysts and photoelectrocatalysts to supersede noble metal catalysts. It was found that MoS${}_2$ edges were catalytically active for hydrogen evolution reaction (HER) while the basal surfaces were catalytically inert. So MoS${}_2$ nanomaterials have been employed as catalysts to generate hydrogen [69,176,177,178,179] because of more catalytic edges. MoS${}_2$ nanoparticles were first used as photocatalysts [180] and electrocatalysts [181] to generate hydrogen in the 2000s. The catalytic properties were controlled by edge sites of the MoS${}_2$ materials. MoS${}_2$ 2D films were more active and employed as photocatalysts [182] late in 2016 for water purification,

Irradiation can introduce defects in MoS${}_2$ layers, which create more HER active sites. Sulfur vacancies can activate and optimize hydrogen evolution [20]. Additionally, electrical conductivity can be tuned by the introduced defects. Therefore, catalytic activities of MoS${}_2$ HER can be significantly enhanced by irradiation defect engineering. Tao et al. [183] treated the CVD-grown MoS${}_2$ thin films by Ar plasma. The active site density of MoS${}_2$ thin films increased five times to $7.74\times {10}^{16}\mathrm{c}{\mathrm{m}}^2$ after the Ar plasma treatments. HER performance was enhanced significantly too. However, long-time Ar treatments etched MoS${}_2$, not benefited the HER behaviors. The MoS${}_2$ sheets were also treated by O${}_2$ plasma [183]. The O${}_2$ plasma treatments led to the formation of both Mo-O and S-O bonds, generating more active sites with the increasing plasma time, enhancing the HER activity significantly.

6.4. Magnetic Properties

2H-MoS${}_2$ bulks are temperature-dependent diamagnetic [184]. It was reported that irradiation improved magnetic ordering in single crystals of MoS${}_2$ [85]. Karmakar et al. [77] investigated electron-irradiation-induced magnetic behaviors of MoS${}_2$ crystals. The irradiation was carried out under $30\mathrm{keV}$ electrons with exposure time of 30 min. The effective magnetic moment of single crystals increased from $0.42{\mathsf{\mu }}_{\mathrm{B}}$ per Mo-ion to $1.11{\mathsf{\mu }}_{\mathrm{B}}$ per Mo-ion on account of irradiation.

Han et al. [74] investigated magnetism of electron-irradiated MoS${}_2$ single crystals. The diamagnetic MoS${}_2$ single crystals transformed into ferromagnetic state after irradiation up to room temperature, as shown in Figure 34. The irradiation-induced magnetic phase transition was largely attributed to the strain around the irradiation-induced vacancies.

Mathew et al. [85] irradiated MoS${}_2$ sheets with a thickness of $200\mathsf{\mu }\mathrm{m}$ at room temperature using a $3.5\mathrm{MeV}$ proton ion beam. The pristine sample was diamagnetic in nature. After irradiated at a fluence of $1\times {10}^{18}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$, ferromagnetic ordering was induced in the MoS${}_2$, as shown in Figure 35a. More detailed work indicated that the induced magnetization depended on the irradiation dose. Figure 35b shows the magnetization of irradiated MoS${}_2$ crystalline flakes that were measured at $300\mathrm{K}$ and $5\mathrm{k}\mathrm{Oe}$ as a function of proton ion fluence from $1\times {10}^{17}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ to $5\times {10}^{18}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$. The magnetization of the pristine flakes was negative as well as in irradiated flakes at a fluence of $1\times {10}^{17}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$. The magnetization became positive after the sample was irradiated at a fluence of $2\times {10}^{18}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ and increased with the irradiation doses. However, the magnetization decreased when the flakes were irradiated at a fluence of $5\times {10}^{18}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$.

Mathew et al. [85] also exposed MoS${}_2$ crystals ($200\mathsf{\mu }\mathrm{m}$ thickness) under $0.5\mathrm{MeV}$ proton irradiation at an ion fluence of $1\times {10}^{18}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ (same fluence at which magnetic ordering was observed using $2\mathrm{MeV}$ protons) while ferromagnetism was not observed in the low-energy proton irradiation. After annealing at $350{}^{\circ }\mathrm{C}$ for $1\mathrm{h}$ in Ar flow, a weak magnetic hysteresis loop was observed. The weak magnetic signal was also observed in the irradiated samples at a lower fluence of $2\times {10}^{17}\mathrm{ions}/\mathrm{c}{\mathrm{m}}^2$ and a lower energy of $0.5\mathrm{MeV}$ proton irradiation. Based on these phenomena, the appearance of observed magnetism in proton-irradiated MoS${}_2$ flakes were due to a combination of defect moments arising from vacancies, interstitials, deformation and partial destruction of the lattice structure, such as the formation of edge states and reconstructions of the lattice.

Zhang et al. [185] reported weak ferromagnetism phenomenon in MoS${}_2$ nanosheets and attributed the magnetic properties to the presence of unsaturated edge atoms. Later, first-principles computations and density-functional theory calculations were carried out to predict ferromagnetism in zigzag nanoribbons of MoS${}_2$ [186,187,188], magnetism in MoS${}_2$ clusters [189], and magnetic edge state of hydrodesulfurizated MoS${}_2$ particles [190]. It was generally accepted that [191] zigzag MoS${}_2$ edges were ferromagnetic and metallic whereas armchair MoS${}_2$ edges were nonmagnetic and semiconducting. The predication was experimentally approved in MoS${}_2$ few-layers [192]. It stated that grain boundaries or defects in the nanosheets were responsible for the ferromagnetism of MoS${}_2$ nanosheets.

Recent work stated that zigzag edges of MoS${}_2$ few-layers can induce ferromagnetism. Therefore, defects, such as sulfur vacancies, can convert diamagnetic 2H-MoS${}_2$ nanosheets ferromagnetism [193]. Irradiation could create numerous defects in MoS${}_2$ 2D materials as discussed in the previous sections. So irradiation should affect magnetic properties of MoS${}_2$ few-layers significantly. More fruitful outputs are expected in irradiated MoS${}_2$ few-layers.

7. Conclusions and Outlook

Semiconducting MoS${}_2$ 2D layers have unique transport properties and have been applied in various devices. Their band-gap, doping, photonic, electric, electronic, magnetic, chemical, and bio-properties can be tuned effectively and simply by defect engineering. Various irradiation exposures, including swift-heavy ions, argon ions, alpha particles, protons, electrons, electromagnetic waves, can significantly induce defects in MoS${}_2$ few-layers to manipulate various properties essentially. Various defects, mainly sulfur vacancies and molybdenum vacancies, have been generated/created in MoS${}_2$ layers. MoS${}_2$ layers can be significantly activated, functionalized, and modified. The irradiation-induced defects are beneficial for several applications including: solar cells [19,194,195,196], batteries [54,197,198], supercapacitors [199], thin film transistors [15,16,80,86,110,138,200,201,202,203,204,205], sensors [17,55,56,106,206,207,208,209,210], hydrogen generators [8,47,50,69,134,176,177,183,211,212], and applications in thermoelectrics [213,214,215,216], piezotronics [63], valleytronics [65], and environments [45,49]. MoS${}_2$ 2D layers were damaged simultaneously under the irradiation and lost their unique properties. Therefore, it is critical to control the irradiation exposure to tune MoS${}_2$ properties by the irradiation-induced defect engineering while avoid potential damages.

Besides MoS${}_2$ 2D layers, the irradiation techniques can be employed to tune physical properties of other disulfide 2D materials, such as NiS${}_2$ urea-electrocatalysts [217], WS${}_2$ hybrid catalysts [218], as well as selenide water-electrocatalysts [219], perovskite electrocatalytic nanoparticles [220], C${}_3$N${}_4$ catalytic materials [221], hydroxide nanosheet battery-electrodes [222], oxide photocatalysts [223], low-dimensional thermoelectric materials [224,225], graphene and carbon nanotubes [226,227,228], and even porous electrodes [229,230].