**Abstract**

The discovery and control of new phases of matter are a central endeavor in materials research. Phase transition in two-dimensional (2D) materials has been achieved through laser irradiation, strain engineering, electrostatic doping, and controlled chemical vapor deposition growth, and laser irradiation is considered as a fast and clean technique for triggering phase transition. By using first-principles calculations, we predict that the monolayer MoTe2 exhibits a photo-induced phase transition (PIPT) from the semiconducting 2H phase to the topological 1T<sup>0</sup> phase. The purely electronic excitations by photon soften multiple lattice vibrational modes and lead to structural symmetry breaking within sub-picosecond timescales, which is shorter than the timescale of a thermally driven phase transition, enabling a controllable phase transition by means of photons. This finding provides deep insight into the underlying physics of the phase transition in 2D transition-metal ditellurides and show an ultrafast phase-transition mechanism for manipulation of the topological properties of 2D systems. More importantly, our finding opens a new avenue to discover the new families of PIPT materials that are very limited at present but are essential to design the next generation of devices operated at ultrafast speed.

**Keywords:** photo induced phase transition, two-dimensional material, ab-initio calculations, electronic excitation, landau theory

### **1. Introduction**

#### **1.1 Current status of photo-induced phase transition (PIPT)**

Phase transition is a process by which matter changes from one phase to another. The discovery, characterization, and control of materials at different phase are a central endeavor in condensed matter physics and materials science [1]. The traditional methods using theoretical and experimental methods including thermal

annealing, strain engineering, charge doping, adsorption, alloying and electron injection have been adopted to understand the transformation of phase transition [2]. However, the abovementioned phase transformation states are all stabilized in the socalled equilibrium phases, and some require a complex environment to generate the phase transition. Meanwhile, with the increasing importance of optical modulation in the aspect of modulating materials structures and properties in recent years, PIPT has become an important tool in adjusting materials in the aspect that the abovementioned traditional methods are ineffective [2].

PIPT is a phase transition directly by light that can be precisely controlled toward target structures with desired physical properties at transient speeds, which is much more convenient than using heat or electric field to control the phase transition. The process of phase transition is basically a non-equilibrium process, and it is especially important to understand the evolution process of this non-equilibrium phase transition. Therefore, it is necessary and extremely important to obtain basic information such as timescale and driving force of PIPT. Using light to control ultrafine transformation between different forms of matter to realize ultrafast functional devices is the goal that scientific researchers have been pursuing since the discovery of PIPT in the 1990s, which takes place in one-dimensional (1D) corrected organic charge-transfer molecular system tetrathiafulvalene-p chloranil (TTF-CA) [3]. Actually, because light-induced charge transfer excitation and/or carrier generation can excite the instability of 1D properties inherent in electronic states through strong electron (spin)-electron (spin) interactions, researchers treat the PIPT of 1D correlated electron systems as their targets.

Although PIPT has the advantages of ultrafast dynamics and simple operation, the areas where PIPT phase transition can be successfully implemented are limited, such as: (1) time-resolved spectroscopy of the dynamics of photo-induced ionic-to-neutral or semiconductor-to-metal phase transition [4]: mixed-stack organic charge transfer crystal, color-phase transitions in polydiacetylene and polythiophene, for example. In such PIPT system, it has strong electron-lattice coupling during the cooperative relaxation interactions; (2) low-spin to high-spin state conversion related magnetic materials or light-induced full topological physics in magnetic semiconductors: [Fe(2pic)3]Cl3 EtOH and RbMn [Fe(CN)6], for example [5, 6]. Phase transition in magnetic materials under the action of light is caused by the coupling between volume striction effect and acoustic modes, and the abrupt step transition observed in the experiment is attributed to the depinning effect of volume shrinkage; (3) femtosecond/picosecond dynamics of the photo-induced lattice rearrangements: [Pt(en)2] [Pt(en)2X2](ClO4)4 with X = Cl, Br and I, for example [4]. The formation and decay dynamics of self trapped excitons are the starting point of various metastable formation processes accompanied by lattice rearrangement as the excited states of electrons induced by photo-excitation change chemical bonds or generate metastable states; (4) PIPT also presented in ferroelectric materials: barium titanate (BaTiO3) and lead titanate (PbTiO3) in perovskite oxides system, for example [7]. With the typical property of polar distortions in ABO3 perovskite oxides, the polarization within it nearly disappears under the stimulation of light, nonpolar phases such as antiferroelectric come into being with the tilt of the oxygen octahedra in ABO3 perovskite oxides.

The emergence of 2D and the layered van der waals materials gives the research a new platform of the diffusive, displacive, and quantum phase transitions with distinct physical properties arised from the dimensionality confinement, chemistry, electrostatics, elasticity, and defects [8]. Recently, experimental evidence has shown that the laser irradiation induced phase transition observed in MoTe2 and WTe2 often along

*Photo-Induced Displacive Phase Transition in Two-dimensional MoTe2 from First… DOI: http://dx.doi.org/10.5772/intechopen.108460*

takes place with the abrupt changes of electronic structure, especially with the spring up of some new topological phases [9–11]. Taking few-layer MoTe2 under laser irradiation, for example, it experiences an irreversible transition from the semiconductor hexagonal 2H phase to the topologically distorted octahedral 1T<sup>0</sup> phase. And it can be used in the desired area for the fabrication of an ohmic heterophase homojunction with the accurate command of micrometer patterning. Despite the intensive investigations, the microscopic mechanism of the laser irradiation induced phase transition in these 2D materials is still not very clear. These experimental findings also suggest that 2D materials may provide a new platform to discover the PIPT materials that is the focus of this theoretical study.

#### **1.2 Scientific importance of PIPT**

Recently, experimental observation of ultrafast disordering of atomic motions in PIPTs has challenged the convertional knowledge of phase transition that the atomic motion in PIPTs is always treated as a coherent process [12]. Actually, ultrashort laser pulses manipulate the structure and function of materials, transforming the electronic or magnetic properties with a timescale in the limit of atomic motion and providing a powerful approach under photo-excitation. Active optical control or interplay between optical and other external influences impact matter or materials are desirable in the field of scientific disciplines, which may display fresh fundamental theoretical and device concepts by optically induced electron-electron correlations, electron-lattice coupling, altering band structures and topology, and so on [4]. Photoexcitation is not only widely used in one-dimensional material systems over the past three decades [13], but also emerging in other fields in recent years, such as memory devices, twistronics field, insensor computing field, ultrafast dynamics, and ultra-high frequency acoustic devices and so on. It's all due to the rich physics effects of PIPT.

Light can introduce the charge density wave as the phase competition in far-fromequilibrium system of LaTe3 [14], with the presentation of topological defects under the influence of slower re-establishment of phase coherence [15]. This result paves the way for understanding the mechanical nature of topological defects with nonequilibrium defect-mediated transitions and providing a framework for the production, control, and manipulation of optically introduced other ordered phases. PIPT can also be used in the emerging in-sensor computing field [16, 17]. Under oxygen stoichiometry engineering, VO2 films can show non-volatile multi-level control under ultraviolet irradiation, with integrated processing, sensing, and memory functions at 300 K. It can significantly improve the recognition accuracy from 24 to 93% because this artificial neural network can exact UV information from ambient environment consisting of the abovementioned neuromorphic sensor. This PIPT observed in the application of neuromorphic ultraviolet sensors not only provides an avenue of the design of neuromorphic sensors but also facilitates its potential applications in artificial vision systems.

Furthermore, in various field such as complex liquids, biological and sensing nanostructures systems, ultrafast light-generated giga–terahertz (GHz–THz) frequencies acoustic phonons have attracted much attention [18–20]. For example, ferroelectric BiFeO3 has photogeneration/photodetection of coherent phonons [21], with the generation of strain under giant ultrafast light because of the generation of stress by inverse piezoelectric effect under the screening of the internal electric fields by light-induced charges. This giant opto-acoustic response opens up a new prospect for the application of ferroelectric oxides in ultra-high-frequency acoustic devices and

promoting the development of some new GHz-THz sound sources. Meanwhile, the corresponding ultrafast dynamics of materials at far from equilibrium states under laser irradiation are one of the ultimate problems in modern science and technology [22, 23]. Actually, PIPT is always treated as a coherent process, but ultrafast disordering of atomic motions in PIPT is observed in recent experiments [24]. It is very important to understand the evolution process of the non-equilibrium phase transition and obtain the basic information such as timescale and driving force of the structural phase transition.

In fact, there are various opinions on the basic information of timescale and driving force in PIPT. The competitive relationship of all driving forces such as photoexcitation, thermal phonon vibration, strain, and so on, has been the focus of scientific attention. With low laser fluence, the timescale of phase transition is always long under the combination of the atomic driving forces caused by thermal phonon vibration and photo-excitation, due to influence of photo-excited hot carriers cooling. In VO2, at high laser fluence, the smallest timescale of phase transition is indeed saturated to a minimum value of about 55 fs, which is due to the saturation of the population of the V–V bonding states by the photo-excited holes [24]. In this work, the structural transition takes place within 1 picosecond, shorter than the timescale for the photo-excited electrons to transfer their energy to the lattice. We also rule out a thermally driven or a strain-induced phase transition under laser irradiation by comparing the thermodynamic stability.

### **1.3 Novel two-dimensional PIPT material family predicted from ab-initio calculations**

Although PIPT has some applications, its physical mechanism has not been clearly defined, and there is still no unified method to explain this physical phenomenon. The microscopic nature of optically driven phase transition is still unclear. With the changes of strain, lattice vibrational modes, chemical state, electronic excitation, and temperature in the studied system excited by light, the corresponding phase transition may be triggered by one of these factors or on a combination of them. It is of great physical importance to understand the dynamic process of PIPT-induced evolution. However, there are still many technical difficulties to understand the evolution of the ultrafast phase transition both experimentally and theoretically. In this work, we adopt a mature computational method: first-principles calculation, combining ultrafast dynamics and Landau theory to predict the emergence of PIPT in 2D materials for the first time, which opens a new direction for the future search of PIPT materials.

Several models have been proposed to explain this phase transition mechanism: (1) local phase transition induced by Te vacancies created by irradiation [25]: in the presence of nearby vacancies, the contraction of the migration energy makes the higher rate of larger defects accumulation than that expected from the isolated migration. Therefore, it acts as a seed for gathering more vacancies around the photoinduced Te vacancy region, producing nucleus arrangement, increasing the growth rate of 1T<sup>0</sup> phase; (2) accumulated heat is a main driving force for the phase transition [26]: the local instantaneous heating from a laser leads to a certain number of Te vacancies, which significantly reduces the potential energy barrier between 1T<sup>0</sup> - and 2H-MoTe2; (3) laser-induced thermal strain contributes to phase transition [27]: strain effect originating from thermal expansion can tune the electronic properties and induce a phase transition between 1T<sup>0</sup> - and 2H-MoTe2; (4) the electronic excitation plays a critical role in the phase transition [2, 28, 29]: with the increase of excited

*Photo-Induced Displacive Phase Transition in Two-dimensional MoTe2 from First… DOI: http://dx.doi.org/10.5772/intechopen.108460*

charge carriers density, the energy barrier between 1T0 crystal and electronically excited 2H phase decreases monotonically and decreases to the thermal energy fluctuation range under the carrier concentration.

Due to electronically excited population inversion, a new distorted trigonal prismatic phase of 1T<sup>0</sup> MoTe2 was generated in previous work [29], which will significantly broaden the application spectrum of these materials. The discovery and control of this phase transition are expressively important to electron regulation science of 2D materials, and the characterization of PIPT is quite less explored with its uncertainty. It is necessary to correctly explain the phase transformation mechanism or the relationship between the magnitude of activation barriers and the potential energy surface in the excited state. Here, based on the first-principle calculations, the microscopic nature of PIPT was studied by a pure photo-induced excitation processes, not by thermal accumulation, defects, or thermal strain as previously assumed. In our calculations, available electrons are stimulated by photo-excitation energies of 1.58 eV (785 nm), 1.96 eV (633 m), 2.34 eV (532 nm), and 2.63 eV (473 nm) respectively. On the premise of unchanging the chemical composition of 2D materials, electrons are excited by light, and the electronic distribution state and lattice vibration mode in the crystal are fundamentally changed in the process of excitonic excitation, which leads to the significant change of chemical bond. Meanwhile, the appearance of ultrafast phase transition to 1T<sup>0</sup> MoTe2 driven by lattice vibration mode softening when the photon energy is larger than 1.96 eV for 2H MoTe2, and this phase transition by electron excited state is very similar to Peierls phase transition.

In this chapter, we not only demonstrate that the phase transition of monolayer MoTe2 can be triggered by photo-excitation of carriers alone, but also successfully using Landau theory for the first time to reveal the difference between the physical mechanism of PIPT and the traditional equilibrium thermodynamic phase transition caused by temperature or pressure. The starting point of Landau's theory is that the symmetry of thermodynamic functions should be the same as the symmetry of the crystal structure of the system. The key point of Landau theory is that symmetry is broken by mean-field approximation or saddle-point approximation and the phase transition with non-zero order sign is obtained. The theory of Landau phase transition is based on the construction of Landau Free Energy after selecting the state variables corresponding to the order parameters of phase transition [30]. For a soft-mode displacive phase transition, the order parameter can be chosen as the amplitude of the distortion of the soft-mode eigenvector u [31]. In this work, we applied the Landau phase transition theory to the explanation of PIPT, the lattice vibration barrier of the E "and A2" uses Landau expansion, which gives the relationship between vibration mode frequency and excitation photon energy. Finally, the vibration mode frequency will drop to zero with the lattice vibration mode completely softened when the excitation energy reached 1.96 eV, which leads to the emergence of the structural phase transition. Our findings not only reveal the microscopic origin of PIPT 2D transition metal tellurides but also provide hope to motivate both fundamental and applied studies of ultrafast phase transitions in these new class of materials for topological switching and neuromorphic computing.

#### **2. Numerical method: First-principle calculations**

We use Vienna ab-initio simulation package (VASP) to calculate the ground electron states and phonon properties and stimulate the photo-excitation processes [32].

For the pseudopotentials, we use the PBE potentials of Mo(4*s* 24*p*65*s* 14*d*<sup>6</sup> ) and Te (5*s* 25*p*4) atoms [33]. The structural optimization, self-consistent field calculation are conducted under energy cutoff of 500 eV, kmesh of 19 � 19 � 1, energy threshold of 10�6*eV*, Hellman-Feynman force threshold of 10�4*eV=*̊A.

In the stimulation of optical excited electrons, we neglect the electron-phonon scattering process and all the thermal effects. We approximately assume that the distribution function of electron states is by quasi-Fermi-Dirac function with infinite smearing, thus the distribution becomes stepwise, as follows:

$$f(\varepsilon) = \begin{cases} \mathbf{1}, \ \varepsilon < \varepsilon\_{\rm FV} \\ \mathbf{0}, \ \varepsilon > \varepsilon\_{\rm FC} \end{cases} \tag{1}$$

During the excited state simulation, we manually lift and lower the quasi-Fermi levels by,

$$f(\varepsilon) = \begin{cases} \mathbf{1}, \ \varepsilon < \varepsilon\_{FV} \\ \mathbf{0}, \ \varepsilon\_{FV} < \varepsilon < \varepsilon\_F \\ \mathbf{1}, \ \varepsilon\_F < \varepsilon < \varepsilon\_{FC} \\ \mathbf{0}, \ \varepsilon > \varepsilon\_{FC} \end{cases} \tag{2}$$

The energy difference between the quasi-Fermi level *εFC* for electrons and the quasi-Fermi level *εFV* for holes is used as a merit of photo-excitation, which takes five discrete values, 1.58, 1.96, 2.34, and 2.63 eV respectively.
