Micromagnetic Studies Of Domain Walls And Skyrmions In Multilayer Film System

With the advent of the era of big data,the urgent need for massive information processing and intelligent information processing has stimulated significant increase in requirements on data storage devices with advanced function.At present,traditional magnetic hard drives has faced bottleneck for sustainable growth of capacity and performance.In the field of spintronics,the dynamic behavior of magnetic domain walls（DW）or magnetic Skyrmions driven by spin transfer torque（STT）or spin-orbit coupled torque（Spin-Orbit Torque,SOT）has potential to resolve the problems.New spintronic information processing devices such as racetrack memory and neuromorphic computing may be advantageous in small size and high information processing speed.Compared with a single-layer ferromagnetic medium,the DW or Skyrmion in a double-layer ferromagnetic composite has the advantages of higher motion velocity and better stability,which is a research hotspot in recent years.However,the properties of the two ferromagnetic layers consisting the composite are factually different,and the influence of this difference on the dynamical behavior of DW or Skyrmion in the double-layer system has not been investigated systematically.In my thesis,through micromagnetic simulations,I have reported my work about dynamical behavior of DW or Skyrmion in a composite ferromagnetic thin film composed by two ferromagnetic layers with distinct parameters.It includes the following research contents:（1）In the second chapter,we have studied the movement of DW driven by STT in a soft/hard magnetic composite nanowire.In previous paper,the magnetostatic coupling between the two ferromagnetic nanowires has been found to be able to increase the speed of DW movement when compared to that in a single ferromagnetic nanowire.However,it also reduces the range of density of the operating current.There is a critical current density（J_a）for triggering the DW motion and another critical value for decoupling the two DWs（J_b）.J_a becomes smaller and J_b increases significantly,when the uniaxial magnetic anisotropy constants of the two ferromagnetic layers are different.By adjusting this difference of uniaxial magnetic anisotropy constant between the two ferromagnetic layers,J_b can be enhanced to the order of 10¹³ A/m²,which is much higher than the maximum working current density of current electronic devices.It lays a foundation for the high-speed racetrack memory operating in a wider range of current density.（2）In Chapter 3,we have studied the physical mechanism of high-speed motion of DW driven by SOT in a synthetic antiferromagnetic composite nanowire which is composed of two ferromagnetic nanowires.By adjusting the physical parameters of the ferromagnetic layer including RKKY-antiferromagnetic coupling strength,DMI,current density and magnetic anisotropy constant,we have studied the contribution of these physical properties on the velocity of DW movement in detail.We have found that the RKKY-antiferromagnetic coupling and DMI have a great impact on the speed of DW motion.Under the combined action of RKKY and DMI,the difference of the azimuthal angle between the magnetic moments in the central of the two DWs is close to 90 degrees,and the DW can be driven at a speed of not less than 1000 m/s.It paves a way to develope racetrack memory with ultra-fast information reading.（3）The research in Chapter 3 shows that the coupled DW can be driven by SOT at a high speed in the composite nanowires with a synthetic antiferromagnetic structure.However,this antiferromagnetic structure also has the problem that the stray magnetic field is weak and the data is not easy to read.In Chapter 4,we have designed a soft/hard magnetic double-layer composite nanowire with a synthetic antiferromagnetic structure,and the upper and lower layer ferromagnetic nanowires exhibit different uniaxial anisotropy constants.The results show that due to the difference in uniaxial anisotropy constants between the upper and lower layers,a stable staggered region with a fixed length can be generated between the two adjacent DWs.In this area,the upper and lower magnetic moments are parallel with an enhanced stray magnetic field.By adjusting the magnetic parameters including RKKY antiferromagnetic coupling strength,DMI constant,and the difference of uniaxial anisotropy constant between the two ferromagnetic layers,the stability and length of the staggered region can be manipulated.During the DW movement,the type of antiferromagnetic coupling for the coupled magnetic domains can be indirectly revealed by detecting the stray magnetic field of the nearby staggered area,so as to determine the digital information 1 and 0.In practice,the digital information can be read by connecting a Schmitt trigger to a magnetic read head.（4）In addition to racetrack memory,the dynamic behavior of DW or Skyrmions is also very promising for neuromorphic computing in the field of artificial intelligence.In Chapter 5,we have proposed a neuromorphic computing device based on electric-field-induced size manipulation of Skyrmion in a soft/hard magnetic double-layer composite with inter-layer antiferromagnetic coupling.This composite is deposited on a piezoelectric substrate.The Skyrmion exists in the soft magnetic layer,while the magnetic moments of the hard magnetic layer are arranged in parallel under a strong uniaxial anisotropy energy.The results show that by adjusting the voltage to cause the change in antiferromagnetic coupling strength,the size of the Skyrmions can be tuned significantly.And a slight change of antiferromagnetic coupling strength leads to very big change of the size of Skyrmions by hundreds of nanometers.Changes in the size of Skyrmions will lead to changes in tunneling resistance,thereby achieving enhancement/inhibition function of an artificial synapse.In addition,the dynamic process based on this size change can mimic the LIF function of an artificial neuron.The size control has an advantage of high speed（ns）and ultra-low power consumption（～0.3 f J）,which lays the foundation for the development of neuromorphic computing devices with ultra-high information processing speed and ultra-low power consumption.