Two-dimensional (2D) magnets are a promising area in condensed matter physics. The successful synthesis of 2D magnetic semiconductors like CrI₃ and Cr₂Ge₂Te₆ has expanded the field to include materials such as CrX₃ (X = Cl, Br), Fe₃GeTe₂, VSe₂, MnSe₂, CrSBr, and MPS₃ (M = Fe, Mn, Ni). These 2D magnets offer a platform for manipulating spin at the 2D limit, advancing spintronics, quantum computing, and optical communications. Controlling spin structure is actively pursued using electric fields, mechanical strain, defect doping, molecule adsorption, and magnetic fields. Ultrafast lasers offer a fast, non-contact, and energy-efficient method for spin manipulation. Light control of magnetic order via spin transfer or demagnetization is of particular interest. Studies on ferrimagnetic MXenes (e.g., Cr₂VC₂F₂) show laser-induced transitions from ferrimagnetic to transient ferromagnetic states within femtoseconds, attributed to optically induced intersite spin transfer (OSITR). Optical doping stabilizes the ferromagnetic phase in 2D spin-liquid RuCl₃. In CrI₃, ultrafast laser pulses induce magnetic domain formation after thermal demagnetization and enable helicity-dependent all-optical switches. Similar tunability is observed in few-layer Fe₃GeTe₂. While ferromagnets are well-studied, spin dynamics in 2D antiferromagnetic materials remain largely unexplored. This paper investigates the experimentally synthesized MnPS₃ monolayer, an antiferromagnetic semiconductor, to understand its spin dynamics under light illumination using time-domain ab initio nonadiabatic molecular dynamics (NAMD) with spin-orbit coupling (SOC).

The literature review section focuses on previous research on the optical control of magnetism in 2D materials, highlighting the use of ultrafast lasers to manipulate spin in ferromagnetic materials like CrI3 and Fe3GeTe2. It also emphasizes the lack of research on similar phenomena in antiferromagnetic 2D materials, specifically mentioning the potential advantages of studying antiferromagnets for faster timescales and stronger magneto-transport effects. Several studies on the optical control of magnetism in ferromagnetic MXenes and RuCl3 are reviewed to establish the existing knowledge and contrast it with the novelty of investigating MnPS3, an antiferromagnetic semiconductor.

The study employed density functional theory (DFT) calculations using the Vienna ab initio simulation package (VASP) with the projector augmented wave (PAW) method and a cut-off energy of 500 eV. The generalized gradient approximation with the PBE functional and a Hubbard U term (5 eV for Mn) were used to account for strong electronic correlation effects. A 15 Å vacuum space was used to suppress interlayer interactions. Atomic coordinates were fully relaxed. The magnetic ground state was evaluated by comparing the energies of ferromagnetic (FM), Néel antiferromagnetic (AFM), zigzag AFM, and stripy AFM phases using a 2 × 1 × 1 supercell. Exchange coupling between Mn ions was investigated using a Heisenberg Hamiltonian. To simulate optical doping, the occupation numbers of valence and conduction bands were manually changed to represent photoexcited electron-hole pairs. The temperature increase due to photoexcitation was estimated using a model that combines Debye and Einstein heat capacities. Time-domain ab initio nonadiabatic molecular dynamics (NAMD) simulations, incorporating spin-orbit coupling (SOC) within the spin-diabatic representation, were performed using the Hefei-NAMD code. The fewest-switches surface hopping (FSSH) method was used along with time-dependent density functional theory (TDDFT) to simulate the spin dynamics. The decoherence-induced surface hopping (DISH) method was used to model electron-hole recombination. A 2 ps adiabatic molecular dynamics trajectory was generated, and 150 initial configurations were randomly selected for NAMD simulations.

The calculations revealed that the Néel AFM phase is the ground state of the pristine MnPS₃ monolayer. However, optical doping with an experimentally achievable electron-hole pair density of 1.11 × 10¹⁴ cm⁻² induced a reversible phase transition to a stable ferromagnetic state. This transition is attributed to the creation of mid-gap states, primarily composed of S-p orbitals, which enhance the ferromagnetic superexchange interaction between Mn-d orbitals and reduce the direct antiferromagnetic interaction. Analysis of the exchange parameters (J₁, J₂, J₃) confirmed this transition. The NAMD simulations revealed the spin dynamics during the photoexcitation process. Excited S-p electrons first relax to the mid-gap states via electron-phonon coupling (EPC), and then undergo a spin-flip to the spin-down Mn-d orbitals through SOC. The spin-flip process takes approximately 648 fs, leading to a relatively long-lived ferromagnetic state. The interplay between EPC and SOC in the carrier relaxation process, particularly the role of mid-gap states, is crucial for understanding the observed phase transition and long carrier lifetime. The estimated temperature increase during the transition (approximately 35 K) remains below the magnetic ordering temperature (78 K).

The findings demonstrate that optical excitation can effectively control the magnetic order in MnPS₃ monolayer, switching it from an antiferromagnetic to a ferromagnetic state. This is achieved at experimentally realistic electron-hole pair densities, making the results highly relevant for practical applications in spintronics and optoelectronics. The mechanism, involving the creation of mid-gap states and the interplay between EPC and SOC, provides valuable insights into the spin dynamics of 2D antiferromagnetic materials. The observed long carrier lifetime due to the magnetic phase transition suggests potential applications of MnPS₃ as a photovoltaic material. Further research could explore the effects of different excitation wavelengths and intensities on the phase transition and carrier dynamics.

This study shows that the antiferromagnetic MnPS₃ monolayer can be switched to a ferromagnetic state using optical excitation, at experimentally feasible electron-hole pair densities. The mechanism involves light-induced mid-gap states enhancing the ferromagnetic superexchange and reducing the direct antiferromagnetic interaction, alongside the interplay between electron-phonon coupling and spin-orbit coupling. The long carrier lifetime resulting from this magnetic phase transition suggests potential applications in spintronics and photovoltaics. Future work should explore the influence of various excitation parameters and the potential for other antiferromagnetic 2D materials.

The temperature estimation during the photoinduced phase transition is an approximation, based on a model combining Debye and Einstein heat capacities. This approximation might not capture all complexities of the heating process. The NAMD simulations considered a limited number of initial configurations. While 5000 trajectories per initial configuration were sampled, additional simulations with a larger dataset could further strengthen the results. The effect of reduced carrier density on the lifetime of the photoinduced FM phase was not explicitly considered in the simulations. Finally, the impact of potential defects in real MnPS3 samples might affect the observed behavior.