Search

With the requirement of increasing data areal density in hard disk drive (HDD) recording, one of the main applied approaches that effectively reduce recording bit size is to incorporate a spin transfer torque (STT) device in the magnetic writer, either to reverse a magnetic layer in the device in order to guide the magnetic flux of the writer pole more concentrated towards the into and out-of air bearing surface (ABS) directions [1] , or to drive a magnetic layer into large angle oscillations for a sufficiently large radiofrequency (rf) field in the media to realize the assist [2] . Micromagnetic simulations of the magnetic device inside the write gap (WG) between the main writer pole and its magnetic shield with an alternating write gap field, were conducted incorporating STT interactions between the magnetic layers. Modeling results for understanding assist performance optimized by STT device design from these two main approaches are discussed in this talk.

Heat-assisted magnetic recording operates close to the Curie temperature of the magnetic recording layer. This allows one to neglect grain-to-grain interactions and to develop a recording model based on distributions. The two main ingredients are the anisotropy distribution at every temperature and the Stoner–Wohlfarth grain state distribution at the recording point. From these two distributions a recording spot can be derived, which is the basis for various embodiments of the recording process. After determining characteristic properties of the two distributions for a given recording medium, this approach requires only point-wise operations on arrays representing physical properties in the plane of the recording layer. This allows straightforward investigation of how different parameters impact recording performance. In addition, the calculation time is solely proportional to the spatial resolution in the recording layer.

Among key technologies that support areal density growth, shingled magnetic recording (SMR) can be applied with a conventional magnetic recording head and does not require significant head design or structure changes compared to heat-assisted magnetic recording and microwave-assisted magnetic recording. Shingled writing is also essential for 2-D magnetic recording. In the application of SMR with a conventional head which has a single reader and a symmetrical writer, one would expect equal performance between writing with the inner side and the outer side when the head is at 0° skew angle. However, it is found statistically, with a large quantity of samples, that there is a preference toward writing with either inner or outer side depending on whether the head is an up or down facing head in a drive. Two types of readers with opposite wafer level free layer (FL) and reference layer (RL) canting directions also show opposite preference for the data written by the same heads. Experiments performed suggest that the reader may induce some “skew” bias in SMR application on top of the writer contribution. A hypothesis combining reader FL direction and reader RL canting is proposed to explain the observed phenomena.

Heat-assisted magnetic recording (HAMR) has become one of the leading approaches to increase the areal density in hard disk drives to several Tb/in2. Laser-induced protrusion significantly impacts the HAMR heads’ performance and lifetime. Nonetheless, the pointy local protrusion amount is hard to accurately quantify by touchdown sensors. Based on the previous indirect thermal-mechanical studies with different sensors and techniques, a new methodology is developed to measure the laser-induced protrusion by burnishing the head during writing with laser powered ON and by utilizing atomic force microscopy scans. This burnish method requires the air-bearing surface (ABS) to physically contact the rotating disk without an excessive overdrive, which relies on the precise settings of the touchdown sensors and the detection methods. The time dependence of the burnish behavior is studied. The results show that its time constant is much longer than the pure thermal and mechanical time constants. The repeatability of the experiment is validated. The local protrusion induced by the laser is in the range of 2–5 nm based on current head design. Accordingly, the detailed studies about heater power effects and protrusion amount as a function of laser power have been conducted. In addition, the impact of varying degrees of stiffness of the ABS is discussed.

In thermally assisted magnetic recording, one of the most important issues is the reliability of the near-field transducer and the magnetic recording head. When the near-field transducer transfers the optical energy to the media through coupling from the waveguide, it inevitably heats itself and its surrounding area, such as the magnetic recording head, which tends to cause long-term performance degradation. To mitigate the problem, one possible solution is to apply a pulsed laser beam instead of a constant one. In this paper, the performance characteristics of the pulsed thermally assisted magnetic recording are analyzed by modeling. Simulation results show that similar recording performance can be achieved through a much reduced laser beam duty cycle (~ 30%).

We developed a dynamic thermal and micromagnetic model for thermally assisted magnetic recording, which employed Landau-Lifshitz-Bloch equation for the thermo-dynamic magnetization motion. It has shown completely different switching mechanism at high temperature compared to that from the Landau-Lifshitz-Gilbert equation. This results, in part, from the assumption in the Landau-Lifshitz-Gilbert equation that the magnetization vector length is conserved at any temperature. In view of the fact that the Garanin's Landau-Lifshitz-Bloch model does not have this constraint, it can estimate the switching field and switching speed in including around Curie temperature. Since thermally assisted magnetic recording is near Curie temperature recording, Landau-Lifshitz-Bloch equation must be suitable for the model of this recording process.

Micromagnetic models of side shielded perpendicular magnetic recording heads show detailed magnetization configuration of the trailing and side shield during the dynamic writing process. The calculation result indicates possible origins of three kinds. The leakage field at the side shield edge, the side shield saturation, and trailing and side shield domain switching. The side shield edge and the saturation induced fields are based on the geometric boundary and they are limited to just around the side shield edge. However the shield switching field can spread to far track position from the side shield to the trailing shield, and it originates from magnetic boundary of the domains and wall formed during the dynamic writing process. As a result, it produces bump field at far track positions in some trailing and side shields.

Wrap-around shielded writers combining both trailing shield and side shield have been studied both by modeling and by experiments. WAS design is demonstrated to have superior performance over its predecessor trailing shielded writer in perpendicular magnetic recording systems. While maintaining the down-track performance, the side shield significantly reduces the side fringing field, thus enabling high track densities. In this paper, various aspects of WAS design are investigated. Issues unique to side shielded design such as near and far track erasure are also discussed in detail. An outlook of future PMR writer and recording systems design is presented at the end.

We developed three home-made modeling programs to design thermally assisted magnetic recording heads: optical beam propagation method for waveguides, optical finite difference time domain method for plasmon generators, and thermal/micromagnetic finite difference method for the recording media. These models lead to the following results. To get higher throughput efficiency of the waveguide, the periodic wavy thickness of the inlet can provide better inlet coupling with laser diode light, and the wavy taper shape can improve the propagation efficiency. As for plasmon generators, the model requires waveguide, recording media and main-pole to estimate the correct performance, because the optimized design depends on all of these parts. Our proposed sharp pointed plasmon generator can provide tiny near field spot, and it must have good scalability for narrow track recording. In addition to these optics models, we performed recording process simulation. As a result, depending on the condition of the thermal spot and head field alignment, either thermal or magnetic field can be dominant in creating the final magnetic transition in the media. The signal to noise ratio and the transition curvature are greatly affected by the recording process.

Side track erasure (STE), either skip or far track, has been studied for a PMR writer with side shielded (SS) design. Both bit error rate (BER) and noise amplitude based STE measurements indicate the side writing fields are strong at SS bottom corners and/or inner edges. With DC background low-frequency noise amplitude measurement, the root cause of the STE is characterized as flux from main pole tip going into media soft under-layer (SUL) and returning to SS bottom corners and/or inner edges. Moreover, the return flux path is identified with a unique one-sided signature depending on the pole polarity and shield initialization direction and can switch side when pole polarity or shield initialization direction is changed. By employing magnetic force microscope (MFM) analysis of a SS head under two shield initialization directions, we can explain this unique one-sided flux path by the observed SS domain structure. Furthermore, a micro-magnetic modeling of the SS design is constructed to understand the observation qualitatively.

We developed auxiliary differential equation finite-difference time-domain (ADE-FDTD) method for dispersive Drude medium, and simulated the optical part of the plasmon antenna (PA) and waveguide (WG) system for thermally assisted magnetic recording (TAMR) heads. To achieve higher efficiency without sacrificing the small spot size, our newly developed automatic computational shape optimization program was carried out for the PA designs with a variety of configurations: no recording media, with-media, with magnetic write pole, and only-PA models. The optimized designs for these cases show an enhancement of the electrical field at the PA tip compared to the initial configuration. In addition, both the PA body as well as the tip affects the performance of the guiding and focusing of the plasmon. However, the optimized detail shape and the sensitivity of the dimension are dependent upon the configuration.

As the magnetic moment of soft under layer (SUL) decreases, the magnetic saturation during the write process becomes severe. It induces some partial erasures in the media write pattern and results in poor write performance such as an output roll-off and signal-to-noise ratio degradation. This erasure is due to the head field longitudinal component at the write gap after the transition writing in the trailing shielded pole perpendicular recording head. In case the SUL has low magnetic moment and high write current is inputted, erasure bubble appears next to the write bubble because the head field perpendicular component switches from positive to negative in the polarity. But the erasure appears even without the negative perpendicular field, and the longitudinal head field of the write gap is dominant for the erasure.

With the continuous push for high areal density, the conventional trailing shield perpendicular writer design faces an increasing challenge in maintaining sufficient writability while scaling down the dimensions of the write pole and the shield. In this work, we investigate an improved trailing shield design with the tapered write gap (TWG). It has been found that the TWG design has two unique recording characteristics that enable better extendibility: 1) flux concentration and 2) self-compensation. Both modeling and experimental results show that the TWG design has superior dynamic performance (DP) than conventional trailing shield design, while the sigma of DP parameters induced by the wafer/back-end variations is reduced.

This paper presents micromagnetic modeling results in a perpendicular magnetic recording, which consists of a single-pole head and double-layered media. To know the inplane head field contribution, the relationship between applied angular field and media remanence is studied, and we propose a perpendicular corresponding field for the head field assessment, which can explain media remanence as well such as the main-pole footprint on the media. The shielded-pole/single-coil head shows the best signal-to-noise ratio but the overwrite performance is somewhat poorer than monopole heads, with a large positive to negative head field, which implies a large field gradient. Although the shielded-pole head has smaller erase bands with straight transitions, it can cause trailing-shield-related erasure by the negative field under the high write current input in the single-coil design

Heat assisted magnetic recording (HAMR) requires a sufficiently small heat spot, which is much below the diffraction limit of the wavelength of the used light. This can be achieved with an optical near field source consisting of a small metallic wedge which supports edge plasmons. The power transfer between a dielectric rectangular waveguide and this metallic wedge is investigated in simulations and experiments. Beating of two eigenmodes of this system leads to power oscillations between the waveguide core and the edge plasmon along their overlap length. This was confirmed in near field experiments which are based on the evaporation of phase change material with the absorbed optical near fields as heat source. Devices with weak and strong edge plasmon excitation could be clearly distinguished in a simple far field experiment.

Internal shunting is introduced on tunneling magnetoresistive heads to enhance device anti-static robustness and external high-frequency noise pickup immunity. The details of the shunting scheme and the mechanism leading to both anti-static robustness and reduced high-frequency noise pickup are discussed.

The tolerance of the expected read-back signal of spin valve giant magnetoresistance based structures to varying deposition and process conditions are described. We determine if spin valves can be produced reliably, and evaluate which thicknesses and properties are most critical. First, the dependence of spin valve properties on layer thickness are experimentally determined. Next, the variation of read-back signal and transfer curve characteristics with spin valve properties is calculated from micromagnetic modeling. Finally, these are convolved with the expected reproducibility of layer thickness to obtain an effective “yield” of structures within 10% of the mean amplitude. We find that spin valves can be reliably deposited, with “yields” well in excess of 90% likely.

We describe the design, fabrication, and performance of a "spin valve" magnetic field sensor based on the giant magnetoresistive effect. The sensor is a balanced, four resistor, fully biased, Wheatstone bridge network with bipolar output. The devices described here show magnetoresistance ratios /spl Delta/V/V=/spl Delta/R/R=6%, saturation fields of 25 Oe, and a Johnson limited noise floor of 2.6 /spl mu/Oe/(Hz)/sup 1/2/. The linearity of the device is +/-2% of the full scale amplitude, with a hysteresis of 1% over the linear range. Fabrication of this device requires a novel approach to setting the directions of the antiferromagnetic exchange layers that bias the sensor. As compared to bridges based on the anisotropic magnetoresistance effect, these devices offer superior signal amplitude and linearity. To our knowledge this is the first report of such a device.

RF vortex spin-transfer oscillators based on low RA magnetic tunnel junctions were investigated. A very high power of excitations has been obtained characterized by a power spectral density containing a very sharp peak at the fundamental frequency and a series of harmonics. The observed behaviour is attributed to the combined effect of Oersted-Ampere field generated by the large applied dc-current and of the spin transfer torque. We furthermore show the synchronization of a vortex oscillation by applying a RF bias microwave which frequency is twice the oscillator fundamental frequency.

We have studied the magnetization configuration of various pole-tip design perpendicular magnetic recording heads by using finite element method (FEM) based micromagnetic simulation. Taper pole has larger head field of 13%–25% than nontaper pole and delivers better gradient because the magnetization of the main pole becomes more perpendicular to the air bearing surface (ABS) and an increase in magnetic charge is observed not only at the ABS but also at the slope region. Side shield has less fringing field in the cross-track direction but the field profile is not as simple as the conventional FEM result. The side shield has a complex domain configuration around the edge, which is difficult to switch in polarity via write current excitation, differing from the trailing-shield situation. The magnetic charge of the side-shield wall induces off-track peaks of the fringing field, and so it must be minimized to optimize the side-shield structure.