Search

The current state of the art technology for ultra-high density magnetic storage is based on heat assisted magnetic recording (HAMR). As the demand for greater storage density increases the optimisation and improvement of HAMR technologies is crucial. The nature of HAMR requires models capable of correctly describing the magnetisation over temperatures up to and exceeding the material's Curie point. Additionally, the ability to study long term behaviours, such as data storage stability, requires that these models are capable of very large timesteps ranging from seconds to years. An open-source code called MARS incorporating three key solvers has been developed. MARS is capable of modelling systems over all required temperatures via a Landau-Lifshitz-Bloch (LLB) solver as well as over long time periods via a kinetic Monte Carlo (kMC) solver. The source code is available at https://bitbucket.org/EwanRannala/mars . Using MARS, investigations into numerous key aspects of HAMR systems have been performed: the presence of Curie point distributions, which limit the effectiveness of HAMR performance by inducing variations in the obtained bit positions; the influence of adjacent track erasure and temporal decay on HAMR performance; and the presence of decreasing ferromagnetic resonance (FMR) linewidth at temperatures close to the Curie point which has been attributed to a potential decrease in the system's damping which can influence HAMR performance significantly. The first investigation has resulted in the development of a semi-analytical model capable of extracting the Curie point distribution from experimental thermoremanence measurements. The second has revealed that adjacent track erasure is significant compared to temporal decay for a system with an energy barrier of 80 kBT with no additional signal degradation when adjacent track writes occur after long periods of data storage. The final investigation has shown that the linewidth reduction is due to in-homogeneous line broadening and that the system damping increases as temperatures approach the Curie point.

Heat-Assisted Magnetic Recording (HAMR) has emerged as a promising next-generation approach for magnetic recording. The functioning of HAMR requires the recording media to be made of magnetic materials with high anisotropy and suitable Curie temperature. Iron Platinum in the L10-phase (L10-FePt) has been found to satisfy these two requirements, thus having been attracting extensive attention and investigation. This Thesis presents a coherent and comprehensive computational research on the impacts of finite-size and surface effects in L10-FePt for potential HAMR applications. Simulations are performed by VAMPIRE an atomistic simulation code developed in the University of York and several key findings have been obtained. First, the existence of a size threshold at 3.5 nm is discovered below which finite-size effects are found to permeate into the centre of the L10-FePt grains leading to the well-known Curie temperature dispersion of the recording medium. A correlation between the Curie temperature dispersion and surface disorder is formulated which can be extended beyond L10-FePt to be applicable to different crystal structures. Second, a novel fourth-order anisotropy component is found in a phase-coupled core-shell structured L10/A1-FePt grain which exhibits strong size and geometry-dependence unseen in previous literature. Additionally, the scaling to magnetisation of this fourth-order anisotropy is found to disobey the classical Callen-Callen power law. These properties can be successfully explained by an analytic model which demonstrates the origin of this novel fourth-order anisotropy to be from the canting of the core and shell magnetisation. The applicability of this analytic model, also, is shown to extend to be valid in a generic soft-hard coupled nanocomposite magnetic system. Finally, the switching efficiency of this phase-coupled core-shell structured L10/A1-FePt grain is investigated. Available data demonstrate the existence of a non-negligible switching error rate in all tested grain configurations. Reducing the grain size and using shorter write pulses are found to induce higher switching error rate. However, it is shown that these detriments can be mitigated by surface engineering. Overall, the studies presented in this Thesis have addressed and provided many helpful insights into outstanding challenges on the path toward the realisation of L10-FePt in HAMR recording media.

The next generation of ultra-high density storage technology will be based on heat-assisted magnetic recording that uses the highly anisotropic L10 phase of FePt. As the areal density increases, the grain size decreases and finite-size effects are becoming crucial. The damping mechanism controls the magnetisation dynamics and the writing speed of the information, hence its behavior for small FePt grain sizes needs to be studied. Using atomistic spin dynam- ics simulations, the variation of damping with temperature and system size is systematically analysed by employing ferromagnetic resonance calculations. The damping of FePt grains is enhanced with increased temperatures, but the linewidth of the system can decrease in the presence of size distributions due to the transition of small grains to the paramagnetic state. Switching is investigated within the heated dot limit, assuming the largest areal density possible in recording media and shows that the numerical calculation involving a dynamical switching of the media leads to larger bit-error rates due to thermal transitions over the energy barrier and subsequent smaller areal densities. Atomistic spin dynamics simulations assume a fixed lattice of atoms, however, as the magnetic material is heated up, both magnetic and mechanical properties will be dynamically affected. To include the lattice contribution during magnetic relaxation (magnon-phonon interactions), a coupled spin-lattice dynamics model has been developed. The fundamental properties of the SLD framework, with application to BCC Fe, have been analysed. The existence of a direct channel of energy and angular momentum transfer between magnons and phonons increases the damping via spin-lattice coupling, which is important for the study of both magnetic insulators and metals. Finally, the magnetisation dynamics is studied under the effect of THz phonon excitation, showing that it is possible to switch the magnetisation via the direct excitation of various phonon modes.

Until the late 1980's, anti-ferromagnets were thought of as theoretically interesting but with no practical application. Since then, this idea has been completely altered, and they are a key element in nearly all magnetic recording devices such as read heads in magnetic hard drives. More recently, the development of antiferromagnetic spintronics has enabled the use of the anti-ferromagnet as the active element and could lead to storage devices with THz speeds, much higher storage densities and a lower power consumption. A current problem in the development of such devices is a lack of understanding of the magnetic properties of antiferromagnets. Atomistic modelling is a powerful tool in understanding these properties as it has the ability to model the materials in atomistic detail on a scale comparable to realistic device sizes. In this thesis, I present an atomistic model of the Iridium Manganese (IrMn) that was created to model the static and dynamic magnetic properties of this complex material. The ground state magnetic structure and thermal stability are calculated including composition effects, disorder effects and finite-size effects. The magnitude and symmetry of the anisotropy in IrMn is calculated, solving a long standing debate between theory and experiment, where the calculations differ orders of magnitude. Finally, an IrMn layer is coupled to a ferromagnet to study the origin of the exchange bias effect, with realistic device sizes, including multiple grains, temperature and interface disorder. The results presented in this thesis determine the properties of IrMn in extraordinary detail, paving the way for a full understanding of this complex and interesting material and its interaction with natural magnets. The work in this thesis has been published in four peer reviewed papers in world leading journals.

In this thesis, I used atomistic and micromagnetic model to study the dynamics behaviour of the domain wall and skyrmion in confined nanostructures driven by applied magnetic field, spin polarized current and voltage control magnetic anisotropy gradient. I also studied the thermal behaviour of the magnetic skyrmion in magnetic ultra-thin film. I study the domain structure and the magnetic switching in the Permalloy (Fe20Ni80) nanoscale magnetic junctions with different thicknesses by using micromagnetic simulations. My work shows that the nanoscale magnetic junction has the potential to be used as a building block for future spin-based data storage or logic computing technologies. The current-driven skyrmion motion in a narrow ferromagnetic nano-track with voltage-controlled magnetic anisotropy (VCMA) is studied numerically. The skyrmionium dynamics in a nano-track with voltage-controlled perpendicular magnetic anisotropy (VCMA) also has been reported. The results provide guidelines for practical realization of the skyrmion-based information channel, diode, and skyrmion-based electronic devices such as racetrack memory. The thermal-induced phase transition to a skyrmion state in IrCoPt has been demonstrated by the atomistic simulations parametrised from ab-initio calculations which include long-range exchange interactions. The simulation results give a clear vision of the thermal-induced behaviour in a chiral magnetic thin film.

Magnetic nano-systems have a wide scope of applications ranging from data storage technologies to biomedicine. In data storage devices the information is encoded in the grain magnetisation corresponding to bit "0" and "1". The recording media industry is driven by increasing the areal density of stored data and decreasing cost, while in the same time maintaining the thermal stability and signal-to-noise ratio. For this, close-packed 2-dimensional granular systems are used, with nanometre grain size. The quality of such magnetic recording media depends on the intrinsic material properties and on the inter-granular coupling via exchange and magnetostatic interaction. The work presented here studies the effects of inter-granular coupling and investigates different approaches to extract intrinsic properties from the bulk measurements. Due to the irregular shape of the grains, the dipole approximation for magnetostatic interaction is inaccurate. For higher accuracy, a 5-fold numerical integral is required for each pair of grains. Analytical integration over the grains height is possible reducing the numerical calculation to a 3-fold integral. The competition between the exchange and magnetostatic interactions leads to complex magnetic structures and correlated behaviour, where groups of grains behave collectively. The effects are observed and studied here based on the magnetic radial correlation function which shows a damped oscillatory form as a function of grain separation. The correlation length increases with increasing exchange interaction. An important consequence of correlated behaviour is that it alters the intrinsic switching field distribution (SFD), leading to an effective SFD. The intrinsic SFD is a fundamental characteristic of granular magnetic materials, defined as the distribution of irreversible switching events of magnetic grains in the absence of inter-granular interactions. Separating the intrinsic SFD from the effective SFD remains a challenge. Two methods that have been widely used to extract the intrinsic SFDs from hysteresis based measurements, the so-called FORC method and the Delta H(M, Delta M)-method are compared. It is shown here that the FORC diagrams contain useful information about the interactions in the system, but the ability to extract the intrinsic SFD is limited to the system in which the magnetic correlations can be neglected. Identifying the SFD from hysteresis loop measurements in the parameter range relevant for applications, requires applying the inverse problem solving techniques such as the Delta H(M, Delta M)-method.

With increasing demand for high density magnetic recording devices a paradigm shift is required to overcome the super-paramagnetic limit. By using a high anisotropy material, such as L1_0 FePt, and heat assisted magnetic recording the areal density can be taken well beyond 1 Tbit/in^2. For FePt the grains on which the data is stored can be reduced to 3nm in size before thermal noise becomes an issue. Therefore, the understanding of how magnetic grains of only a few nanometres in size behave and switch is highly important. Here an atomistic model of magnetic materials, based on the Heisenberg exchange interaction and localised classical atomic moments, is used to investigate the magnetic reversal properties of granular recording media. A detailed model for FePt, parametrised from ab-initio, is presented and this is used to investigate the finite size effects in nanometre grains. Using this model, the finite size effects on the linear reversal regime is investigated. Comparing a model using the Landau-Lifshitz-Bloch macrospin equation to the atomistic results show that multi-scale modelling may be valid down to the 3 nm length scale. The inter-granular exchange interaction between neighbouring grains is investigated using a model invoking the presence of magnetic impurities in the grain inter-layer. Using constrained Monte-Carlo methods the effective exchange is calculated for different impurity density, grain separation and temperature. Following this, helicity dependent all optical switching in granular FePt is investigated. The phase space for switching through the Inverse Faraday effect is explored. Using the Master equation, a model for thermal switching with magnetic circular dichroism is explored which qualitatively explains the induced magnetisation observed experimentally. Finally, the coupling of the spin and lattice system is modelled by combining spin and molecular dynamics. The coupling to the lattice excites magnons which appear to decay implying that some damping processes are occurring.

The emerging technology of spintronics promises to revolutionise computing by allowing considerably more energy efficient computing than is currently possible. Gallium Arsenide is one candidate material for use in semiconductor spintronic devices and as such detailed study of the spintronic properties of Gallium Arsenide is required. In this thesis we develop a semiclassical approach to the simulation of the electron population in Gallium Arsenide. We then use this model to look at the properties of the electron system, in particular the time taken for the electron population to undergo spin depolarisation. Comparison of the results to experimental values for the spin depolarisation suggest that the currently well accepted approach is accurate for low to moderate n type doping densities, but for higher densities there is a significant departure from the experimentally obtained results. We explore a number of improvements to the usual model showing substantial improvement in the range of densities that can be accurately predicted. As Dilute Magnetic Semiconductors have been the topic of substantial research, we also apply the model to investigate the properties of an magnetic impurity in n doped Gallium Arsenide. We compare the results to the Langevin enhanced Landau Lifshitz Gilbert equation, which shows good agreement allowing us to predict an appropriate damping constant as a result of the magnetic interaction with the electron population. We also investigate the nature of the noise felt by the magnetic atom showing that the usual white noise approximation is of limited validity.

There has long been an interest in the exploitation of novel magnetic behaviour for practical applications such as magnetic storage devices. Some of the most interesting dynamical behaviour occurs when a material contains both ferromagnetic (FM) and antiferromagnetic (AF) characteristics. Many such systems have forgone study due to practical difficulties of experimental observation of antiferromagnetic order. In this work several systems of current interest which contain both AF and FM order are studied. These materials and systems are used, or are candidates for, technological applications, especially magnetic storage devices. The forefront in this area is concerned with laser induce magnetisation reversal and there are many unexplained phenomena, especially in ferrimagnetic and metamagnetic materials. A combination of analytical and large scale numerical calculations are used, often with comparison to experimental data where available. The approach used is generally based around so-called atomistic spin dynamics, where the Landau-Lifshitz-Gilbert equation, augmented with a Langevin term, is solved for each atomic moment. This allows the description of magnetic materials at elevated temperatures and through phase transitions. Semi-analytic formalisms are studied, comparing with atomistic spin dynamics and micromagnetics, to inform multiscale modelling techniques. The excitation of a localised mode an antiferromagnetic layer which is coupled to a ferro- magnetic layer is studied. It is shown that this excitation leads to an enhanced damping of the ferromagnet, an important consideration for the design and optimisation of spin valves. The metamagnet FeRh which undergoes an antiferromagnetic-ferromagnetic phase transition is also investigated. There is much debate about the origin of the phase transition and a model is constructed in this work which demonstrates that an all magnetic origin is possible if effective four spin exchange terms are considered. This model is also capable of explaining the observed dynamics in femtosecond laser heating experiments. Finally, the spin wave dynamics of the prototypical amorphous ferrimagnet GdFeCo are considered. The thermally induced switching which has been discovered in this material is explained as the excitation of a two-magnon state.

Magnetic materials used in the magnetic recording industry have been developed for several decades. However, since the areal recording desity is increased to around 1 Tbit/in2,some new physical challenges have become clear: particularly the need of using high magnetic anisotropy materials to overcome the super- paramagnetic effect. The requirement of acceptable signal to noise ratio leads to a reduction in grain size. In order to maintain thermal stability it is necessary to increase the magnetic anisotropy K. This leads to unrealistically high write fields. The main challenge is to reduce the write field, which is being met by advanced design and the potential use of Heat Assisted Magnetic Recording. In the meantime, the research on the reversal processes has also attracted much attention because of the use of ultra-fast technology and also the high speed of writing and reading processes in recording media. From the theoretical point of view, an atomistic model can obtain detailed information which is difficult to obtain from experiments. Also it can break through the limitation of micromagnetic calculations in ultra-fast processes and thermal effects because of using a Heisenberg direct exchange model and at the atomic level, especially studying the interface between two coupled materials. In this thesis, an atomistic model is used to simulate the properties of magnetic recording media studying the reversal processes, as well as the exchange coupling, anisotropy and thermal effects of small magnetic particles. Firstofall,theinter-granular exchange coupling is calculated by doping magnetic atoms into the spacing between magnetic recording grains, and indirectly by adding a capping layer onto the granular recording layer,to control the exchange coupling and further the switching field of the magnetic media. This result shows that an optimised interface exchange parameter is important to reduce the coercivity. Secondly, the reversal processes of many independent single domain particles is studied, determining several important parameters such as field, temperature and time scale,to better understand some critical physical problems in the heat assisted magnetic recording process. Furthermore, a larger scale atomistic model is used to simulate the whole heat assisted magnetic recording process in a granular recording media with grain size distribution, demonstrating the full atomistic model is capable to simulate a realistic system. Finally, an exchange coupled composite media is simulated to develop the potential application of this material in recording media.

A model of spin accumulation (m) is proposed to develop theoretical approaches to calculate the m in any arbitrary magnetic structure. The model is based on generalising the approach of Zhang, Levy and Fert (PRL 88, 236601, 2002). The calculation involves the layer-wise discretisation of the structure and the development of semi-analytical approaches to solve for the equilibrium m throughout the structure. Interestingly, the layer discretisation allows the treatment of diff�use interfaces using a gradual variation of the magnetic and transport properties across the interface. The e�ffect of the interfaces between a ferromagnet and a nonmagnet and between two ferromagnets on spin injection is investigated. The formalism for calculating the m is first generalised by taking m as the di�fference of spin-up and spin-down density of states, which is necessary for treating the interface between diff�erent ferromagnets. Then, the e�ffect of atomic species interdiffusion at the interface is included by using Ficks's law. It is shown that the discontinuity of the m at the interface depends strongly on the degree of interface mixing. Subsequently, current-induced domain wall (DW) motion in a ferromagnetic thin fi�lm driven by a spin-polarised current is investigated using an atomistic model coupled with a standard Landau-Lifshitz-Gilbert equation. The inclusion of the spin-transfer torque is represented as an additional �field. The m is calculated self-consistently and naturally includes the adiabatic and non-adiabatic contributions depending on the rate of change of magnetisation relative to the spin di�ffusion length. In this work, it is importantly found that the constants �x and �x used in the standard micromagnetic model do not provide a good description of the spin torque phenomenon due to the non-physical behaviour. Therefore, it is suggested to describe the spin-transfer torque directly from the m. Finally, the evolution of the magnetisation and m are demonstrated by introducing a spin-polarised current into a material containing a DW whose width is varied by changing the anisotropy constant. It is found that the adiabatic spin torque tends to develop in the direction of the magnetisation whereas the non-adiabatic spin torque arising from the mistracking of conduction electrons and local magnetisation results in out-of-plane magnetisation components. However, the adiabatic spin torque signifi�cantly dominates the dynamics of magnetisation. The total spin torque acting on the magnetisation increases with anisotropy constant due to the increasing magnetisation gradient. This leads to increasing DW displacement.

Magnetisation reversal is a process that is of paramount technological importance, as well as fascinating scientists in the field. Despite its importance, only recently have scientists begun to probe the limits of what can be achieved. Driven by consumer desire for compact devices using large scale storage, the size and time limit of magnetisation reversal have begun to be unravelled by exciting experiments using large scale national facilities and state of the art modelling. In the work presented here we develop a model of a ferrimagnetic material parameterised on experimental observations. The key magnetic features are shown to agree well with experimental measurements and provide a basis for more complex calculations. Using time and element resolved X-ray magnetic circular dichroism experiments, model calculations of laser induced magnetisation dynamics are compared to experimental measurements. We present results of switching on the sub-picosecond timescale and conclude that it is possible to reverse magnetisation using heat alone. The conclusion that magnetisation can be reversed without a directional stimulus is scientifically intriguing and never before predicted. Confirmation of such a reversal mechanism using heat alone is verified experimentally, comparing well with the model predictions.

The most common product of the storage industry is the well known hard disk, based on the magnetic recording and reading of data. The research of magnetic materials is enhanced by the establishment of hard disks as a recording media. Given the current trend toward nanostructured materials and complex materials design, the understanding of the origins of magnetic phenomenology at the atomistic level has become necessary. In this thesis, we focus on the aging effects of magnetic materials. The systems that are investigated are bilayer thin films comprised of two materials with different magnetic order. The so called exchange bias systems are comprised of a ferromagnetic layer in contact with an antiferromagetic layer. Exchange bias systems are part of the current read-heads, which utilise the phenomenon of Giant Magnetoresistance discovered by A. Fert and P. Grünberg, who were awarded a Nobel prize in 2007 for this discovery. The most wellknown characteristic of the exchange bias systems is the shift of the hysteresis loop along the horizontal axis. The reduction of this shift with consecutive hysteresis cycles is called the training effect. The current research focused on the dependence of the training effect on various exchange bias system parameters using the well established domain state model. A novel analysis was developed for the study of the antiferromagnet during consecutive hysteresis loops. A special focus was given to the response of training effect on temperature. Several characteristics of the system were varied to investigate the physics of the training effect, such as the antiferromagnetic thickness and dilution with non-magnetic defects. The reversal modes of the ferromagnet were also investigated varying the anisotropies of the system as well the angle of the magnetic field. New characteristics were added to the domain state model increasing the realism of this model. The interface roughness was introduced in the model, as more representative of realistic exchange bias systems. Furthermore, different crystallographic structures such as body-centered cubic and hexagonal-close packed, were investigated as in these structures the coupling between the ferromagnet and the antiferromagnet increases. In these structures, in addition to the interface roughness, the enhance coupling is shown to give rise to complex trends of the exchange bias and the training effect.