Disk Lubricant Research Articles

Slider-disk clearance characterisation and optimisation for high-reliability magnetic hard disk drives

This paper deals with the tribological reliability and durability issues in modern magnetic recording hard disk drives where the slider flies over the disk at distances typically less than 10 nm and at linear velocities of dozens meters per second. Specifically, we address the slider-disk clearance, which is known to have a major impact on HDD reliability. We discuss clearance-measuring techniques, the theoretical relationship between the clearance and reliability, and the relative contributions of different factors such as disk surface roughness, slider dynamic pitch, and disk lubricant properties to the overall drive performance.

Confinement Effects of Hard Disk Lubricant Investigation by Atomic Force Microscopy

AbstractAs the critical design requirements decrease, we need to understand how confinement affects the performance of the materials we use. One example of this occurs with the lubricant used on magnetic recording disks to reduce wear between the disk and head reader during unexpected intermittent contact. Lubricants for use in hard drives need to have sufficient reflow and redistribution properties which are stable in a wide range of temperatures experienced in today's hard drive. Since the material properties of the confined system often differ from the bulk properties, the tools needed to properly investigate these systems need to be developed and understood. In this study, we utilized atomic force microscopy (AFM) techniques, to study the confinement effects on lubricant stability with temperature. Our studies have demonstrated distinct changes in adhesion force with changes in the thickness of the confining substrate layer. As adhesion is a component of friction, these changes in adhesion are directly related to lubricant performance and its ability to reduce wear between disk and reader.

Analytical determination of the surface energy of sub-5nm head-disk interfaces accounting for multilayer effects

In this work, a five-layer HDI configuration was employed to derive a simple way to relate the multilayer Hamaker constant to an effective surface energy of the HDI. Specifically, the five layers of the HDI include two layers on the slider [diamondlike carbon (DLC) overcoat, Al2O3–TiC] and two layers on the disk (lubricant, DLC overcoat) as well as the air gap (or flying height). Following the formulation for two equivalent homogeneous surfaces, one can derive the effective surface energy of the five-layer model. To investigate the effect of the topmost layer thicknesses of the slider and disk on the surface energy, they were first varied from 2 to 5nm and from 0.5 to 2nm, respectively. It was found that the surface energy depends largely on the disk lubricant thickness while it has little dependence on the slider DLC thickness. Lubricant pickup by the slider and its effect on the surface energy was also investigated and it was found that the overall surface energy of the interface is lower when there is lubricant pickup.

Voltage pulsing for localized clearance measurement

Mechanical clearance is a critical parameter of the head/disk interface (HDI) for sliders flying in close proximity in hard disk drives. It depends on a variety of HDI parameters, such as the air-bearing design, disk roughness, and the disk lubricant. An experimental setup was developed to determine the mechanical clearance locally at any location on the disk. Voltage pulsing is used to add an additional force component to the complex HDI equilibrium. The applied force is attractive in character and therefore reduces the mechanical spacing between the slider and the disk. If the magnitude of the applied voltage is changed, a slight contact between the slider and the disk can be realized. Simultaneous measurement of the induced slider displacement using a laser vibrometer allows the quantification of the mechanical clearance. The effect of disk roughness and microwaviness on the mechanical clearance is investigated.

Pole-tip protrusion effect on head-disk interface at low flying clearance

With the increase in the areal recording density on hard disk drives, head flying height is quickly dropping below 10 nm. The actual flying clearance of the head above the media surface during read-write operation is farther reduced by write-current induced pole-tip protrusion (PTP). At such a low flying height, the amount of PTP become a critical factor affecting head-disk interface tribology, especially in the areas of interface fly-ability and lubricant transfer from disk to head. This paper presents a new approach utilizing controlled PTP with acoustic emission sensing to study the head-disk interaction at close proximity flying. Actual flying clearance can be measured and controlled through this technique. The PTP contact hysteresis of surfaces with different is compared. With fixed flying clearance, the effects of PTP protrusion to disk lubricant redistribution and lubricant transfer from disk to head are reported.

Partial-contact head-disk interface approach for high-density recording

A modern hard disk drive (HDD) generally requires a head-disk spacing of less than 10 nm. This paper focuses on studying slider vibration and designed several novel air bearing shapes to prevent slider vibration. An isolated pad slider and a narrow trailing edge slider were designed and fabricated for the feasibility study of the partial-contact interface. From the study, we found out that: 1) a slider with a smaller isolated pad had a smaller off-track vibration; 2) one with a smaller isolated pad had less aggressive disk lubricant depletion; and 3) a narrower trailing edge led to an easy transition from contact to flying.

Aging of a fluorinated lubricant on bare and DLC-coated silicon-based MEMS

The aging of Fomblin Zdol® on a silicon-based MEMS device was investigated. The MEMS device used was an electrostatic lateral output motor. Devices freshly coated with Zdol performed exceedingly well compared to uncoated devices, but their performance deteriorated on storage in air. This is apparently due to the decomposition of the fluorinated lubricant in contact with the polysilicon surface for an extended period. XPS and AFM studies show that the composition and morphology of the Zdol-coated polysilicon surface change on aging. A stable, cohesive barrier layer deposited to prevent direct contact of the fluorinated lubricant with polysilicon enhanced the stability of the lubricant. As widely used in hard disk lubrication, a thin film of diamond-like carbon (DLC) acted as an excellent barrier layer extending the life of unlubricated devices and preventing degradation of those lubricated by Zdol.

The physics of disk lubricant in the continuum picture

Molecular dynamics simulation would be well suited to predict the physics of molecularly-thick lubricants on magnetic disk surfaces. The difficulty in defining suitable interatomic potentials, as well as the requirement for large computing power, makes this type of study difficult. Treating the lubricant film in the framework of fluid dynamics has proven a much better way to model its behavior. This paper describes a full numerical model of lubricant moguls and ripples formation using finite difference analysis. We demonstrate that both moguls and ripples result from slider-induced air shear. Ripples tend to form at higher disk speed and narrower slider width, whereas moguls are seen at lower speed and larger slider width. Both moguls and ripples are enhanced for thicker lubricants, higher waviness, and lower flying heights. Ripple instabilities that were predicted earlier using a stability analysis are also quantitatively confirmed using this numerical model.

Interaction Between Disk Lubricants and Solvents in the Dip-Coating Process

It has been reported earlier that, keeping all other factors fixed, the disk performance could vary significantly depending on the choice of solvent used in the dip-coating process. We surmised that the difference in the ability of solvent to disrupt the intermolecular hydrogen bonding between hydroxyl units of Z-dol or Z-tetraol would lead to a different structure or arrangement of the lubricant molecules as the solvent evaporated in the dip-coating process. Using proton NMR and IR spectroscopy, four solvents, hexafluorobenzene, Vertrel-XF, HFE7100, and FC72, were examined for their ability to disrupt the hydrogen bonding of Z-dol and Z-tetraol. The study has revealed that the said ability increases in the order of FC72 < HFE7100 < hexaflurobenzene < Vertrel-XF. Z-dol molecules, for example, are completely separated from each other in Vertrel-XF, but maintain their intermolecular hydrogen bonding in FC72. It has also revealed an extraordinary formation of complexoid between HFE molecules and Z-tetraol with MW n of ∼2000. The complexoid has a specific stoichiometry of (Z-tetraol) n (HFE) ∼15 n and is not miscible with HFE. The formation of such complexoid was not observed for Z-tetraol of larger MW n (3000).

Application of Principal Component Analysis to the FTIR Spectra of Disk Lubricant to Study Lube–Carbon Interactions

Lubricant molecules on carbon surfaces are typically present in two distinctive chemical-physical environments, conventionally referred to as bonded and free lubricant. We observe and analyze the spectral shift in the lubricant FTIR spectrum using principal component analysis (PCA) to quantify and understand the bonded and free lubricant states for the lubricant systems Zdol, AM, and ZTetraol. In addition, we performed extended PCA to elucidate lubricant interactions with phosphazene additives. PCA of FTIR spectra is found to be a viable method in quantifying bonded lube ratio as well as in studying the mechanism of lubricant and phosphazene additive-carbon interactions.

Pole-Tip Protrusion Effect on High Data Rate Recording Performance

Write pole-tip protrusion (PTP) effect on the overwrite performance at high recording data rate is discussed. It is shown that the head thermal expansion during the write process may lead to the degradation of overwrite value at high recording data rate. Overwrite degradation is caused by head-disk interactions when a high-frequency pattern is written. Appropriate selection of the disk lubricant can reduce the effect of the write PTP and increase the write current setting margin. At elevated drive temperatures, excessive PTP can cause head-disk interface failure. A write timing method allowing for the recording head to cool between the write cycles is proposed. A method of head-disk interface testing and a method of mapping the head-disk interactions are further described.

Time-Frequency Analysis of Slider Dynamics and the Effect of Lubricant

The dynamics of pico sliders is investigated during scratch impact using laser Doppler vibrometry. Air-bearing frequencies and damping of the pitch-2 frequency are studied as a function of disk velocity for lubricant thickness values of 1.6 nm, 1.0 nm and 2.0 nm. Time-frequency analysis is performed to study the instantaneous air-bearing frequencies. System identification methods are used to model the response of the slider after a scratch impact. The results indicate that the slider dynamics and the damping coefficients are affected by the disk velocity and lubricant thickness and also possibly by disk micro-waviness.

Liquid Nanodroplets on Thin Film Magnetic Recording Disks©

As magnetic recording slider size and flying height evolve to smaller dimensions, previously insignificant levels of contamination begin to play a role in slider media tribology. This article describes a new type of contamination. Liquid nanodroplets on disks originate with electrostatic deposition of hygroscopic ultrafine particles, also referred to as cloud condensation nuclei (CCN). On lubricated disks, the CCN equilibrate with atmospheric moisture and become partially overcoated with disk lubricant, which acts as a fluorocarbon surfactant. Dark-field microscopy measured deposition rates of 0.001 to 0.006 #/mm 2 /sec on initially clean disks exposed to ambient air in nonconductive cassettes. Tapping mode atomic force microscopy determined that the sites were deformable nanodroplets 70 to 300 nm in diameter and up to 150 nm high. From AFM profiles, the contact angle of the spherical capped nanodroplets with the disk was between 40 and 90 degrees. Nanodroplet contamination is characterized, and its effect on friction, acoustic emission, and slider smears is demonstrated. A surface chemical thermodynamic model is developed and employed to estimate that the average initial dry nucleus diameter is 110 nm. The estimated size range and composition of the initial nuclei are consistent with those well known in the atmospheric sciences. Nanodroplets were absent from disks that were stored in a CCN-free environment, and the deposition rate was reduced 10× by air ionizer or conductive cassette.

Tribology challenges for head-disk interface toward 1 Tb/in/sup 2/

Achieving 1-Tb/in/sup 2/ magnetic recording areal density requires head-media spacing (HMS) to be no greater than 6.5 nm. It begs the question whether this is a reality (or physics)-based target. This paper examines the main functions and requirements of each HMS component, overcoats on media and head, lubricant film on media, and head-disk mechanical separation, to explore the feasibility of this 6.5-nm HMS budget. From the analysis one may conclude that, although extremely challenging and involving major paradigm shifts, this target does not appear to be completely out of bounds of physical limits. One HMS budget scenario has been proposed which allocates 1 nm each for disk overcoat, head overcoat, and disk lubricant film, leaving 3.5 nm for head-disk mechanical separation. Two fundamental strategies to achieve 3.5-nm head-disk mechanical separation were also discussed.

Investigation of slider dynamics under electro-static force

With the continual increase in the areal density of hard disk drives, the head/disk interface engineers are continuously challenged in designing an interface with a lower mechanical clearance between the slider and disk. Although, this has traditionally been achieved by making the disk smoother, some novel ideas have been discussed, where the slider generally flies high, and when necessary, a part of the slider is moved closer to the disk surface. Another method that has been discussed is applying a voltage across the slider and disk to bring the whole slider closer to the disk. However, rapid application of voltages across the interface may lead to undesirable slider oscillations. In this paper, we study the dynamics of the slider when a voltage is applied across the slider and disk. We show that one can easily excite the air-bearing frequency and if the voltage is not well controlled, the slider may also damage the disk. We also identify the most probable mechanisms of current flow and show that the disk lubricant plays a critical role in initiating current flow.

Degradation of Disk Lubricant: Ramification in Disk Drives and Direct Detection by TOF-SIMS

The degradation of Z-dol catalyzed by Lewis acid centers on the slider surface leads to chain scission forming one type of fragment terminated with a fluorocarbonyl end-group and the other with a trifluoromethoxy end-group. The former, in contact with humid air, converts to a fluorinated carboxylic acid Z-COOH. Z-COOH is an excellent scavenger for alkali or alkaline earth metal ions. Z-COO-M+ thus formed is a strong surfactant, and, in a humid environment, forms microdroplets embodying water in the core. Metal ions thus scavenged on a disk surface can be readily detected by TOF-SIMS, and the microdroplets by optical microscopy in the dark-field mode. The presence of fragments having a trifluoromethoxy end-group on the disk surface can also be established by TOM-SIMS. A careful intensity analysis of peaks due to anions having a trifluoromethoxy end-group permits a semi-quantitative assessment of the extent of degradation. The study has also shown that degradation is caused by such production processes as tape-polishing and by such disk drive operations as the head flying over a single track or over a band in a seek-mode.

Lubricant thickness modulation induced by head-disk dynamic interactions

The behavior of a thin lubricant film under a flying head has been studied by examining the lubricant redistribution on the disk after flying, using an HDI-SRA instrument. Lubricant depletion tracks were observed on the disk surfaces and, more interestingly, the lubricant film was found to exhibit a periodic thickness modulation in the downtrack direction. The wavelength of the lubricant thickness modulation is found to increase linearly with the disk linear velocity, and depends weakly on the lube type and lube bonding ratio. The amplitude of the modulation grows slowly with flying time. Both negative-pressure air bearing pico sliders and catamaran-style positive-pressure nano sliders generate similar patterns on lubricant films. The frequency of the thickness modulation is in the range of 40-50 kHz, and is attributed to interactions of the disk lubricant with the slider roll mode. In addition, for highly-bonded lubricant films, a much finer lubricant modulation pattern can be seen with a frequency of 196 kHz, which is very close to the frequency of the pitch mode of the slider. These results indicate that the lubricant thickness modulations on-the disk are generated by slider-disk dynamic interactions, and are due to slider body motions.

Effect of adsorbed water on perfluoropolyether-lubricated magnetic recording disks

The impact of water on the hydroxyl-terminated perfluoropolyether, Zdol, and the amorphous carbon overcoats commonly used on magnetic recording disks was investigated via computer modeling, kinetic experiments, and surface energy measurements. Water is shown to interact via hydrogen bonding with the hydroxyl end groups of Zdol and the polar, carbon–oxygen functionalities on the carbon overcoat. The interaction between water and the hydrophobic perfluorinated backbone is, however, repulsive. The corrosion susceptibility of Zdol lubricated magnetic recording disks was also studied. Reduced levels of corrosion were observed when the total Zdol thickness, and/or the bonded thickness, were increased. The results are consistent with a heterogeneous corrosion mechanism in which electron transfer from the polar functionalities on the carbon overcoat to the surface adsorbed water is the rate limiting step. The disk lubricant inhibits the electrochemical conductivity of the disk surface by; (a) a screening of the electrochemically active surface sites by the hydrophobic perfluoropolyether backbone, and (b) a neutralization of the surface active sites by hydrogen bonding of the hydroxyl end groups of Zdol.

Alkane Contamination Effects on PFPE Lubricant Bonding to a-CHx Overcoats

Increasing operating temperatures within hard disk drives promote increased outgassing from drive components and thus can adversely influence the durability and reliability of hard disk drives. This study investigates the effects of hydrocarbon contamination on the bonding of perfluoropolyalkyl ether (PFPE) lubricants to amorphous hydrogenated carbon (a-CHx) overcoats. This work has used decane (C10H22 )a s a model for the hydrocarbon contamination present in drives and perfluorodiethyl ether ((CF3CF2)2O) and 2,2,2-trifluoroethanol (CF3CH2OH) as models for the backbone and endgroups, respectively, of the common disk lubricant Fomblin Zdol. Temperature-programmed desorption experiments were conducted using CF3CH2OH coadsorbed with C10H22 on an a-CHx overcoat and (CF3CF2)2O coadsorbed with C10H22 on an a-CHx overcoat. The results indicate that the presence of C10H22 decreases the adsorption energies of both CF3CH2OH and (CF3CF2)2O on the a-CHx overcoats. The implication of these results is that the mobility of PFPE lubricant films can be increased by the presence of hydrocarbon contaminants on the disk surface.

Contribution of lubricant thickness to head-media spacing

In this work, we have studied the effect of lubricant thickness on magnetic recording, specifically, its effect on head-media spacing (HMS). In the experiment, a series of disks were first lubricated to various thicknesses ranging from 20 to 60 /spl Aring/. Then, read/write recording tests were performed for each disk on one track near the disk OD with variable linear densities. Next, the lubricant was washed off from MD to OD using a solvent, so the thickness of lubricant at the OD track was reduced. The read/write tests showed that the read-back signal was enhanced after lubricant wash-off. Using Wallace's equation, the HMS change was calculated and found to be proportional to the change of the lubricant thickness. This result conclusively establishes that disk lubricant indeed takes up HMS and must be accordingly accounted for in HMS budgeting.