92 IEEE TRANSACTIONS ON MAGNETICS, VOL. 46, NO. 1, JANUARY 2010 Media Corrosion: Not Just an Overcoat Problem Q. Dai , B. Marchon , H. Do , K. Takano , and J. L. Wang San Jose Research Center, Hitachi Global Storage Technologies Inc., San Jose, CA 95135 USA Media Development, Hitachi Global Storage Technologies, San Jose, CA 95135 USA This paper demonstrates that electrochemical impedance spectroscopy combined with atomic force microscopy analysis can success- fully characterize the coverage ability of an overcoat on perpendicular magnetic recording media. Rougher media, brought about by lower surface energy oxide segregants, can adversely impact the overcoat integrity. The role of the capping layer, and its ability to some- what planarize the overall structure, is also discussed. Index Terms—EIS, overcoat, PMR corrosion, roughness. I. INTRODUCTION S INCE the hard disk drive (HDD) industry shifted to perpen- dicular magnetic recording (PMR), media corrosion has been a constant challenge [1]. We have previously shown that the increased media corrosion susceptibility of PMR media is due to its higher roughness compared with longitudinal mag- netic recording (LMR) media [2]. To offset the roughness in- crease, we have also shown that as much as an additional 10 of overcoat may be necessary in order to attain adequate coverage, compared to LMR [2]. However, increase in overcoat thickness leads to larger head-to-media spacing (HMS), impacting mag- netic performance negatively. In this paper, we present further studies on PMR media roughness and overcoat coverage. We will demonstrate that although the corrosion problem is generally alleviated by the use of thicker overcoat, a more fruitful approach that is often neglected is to optimize the metallurgy and structural design, such as the segregant types. The resulting media roughness can be greatly reduced, allowing the use of thinner overcoats. Electrochemical impedance spectroscopy (EIS) has been em- ployed for the investigation of overcoat coverage. EIS is a pow- erful tool in the study of protective coatings [3]. For the sake of simplicity, we focused on SiN overcoat which is a well known dielectric overcoat with superior barrier properties [4]. The di- electric nature of this overcoat simplifies the interpretation of the results. II. EXPERIMENTAL SETUP PMR media samples used in the studies were standard 65 mm disks coated with various metallic layers pertinent to PMR recording. Soft underlayer (SUL, nm) and exchange break layer (EBL, m) were deposited first, followed by a magnetic layer, nm thick, composed of two half-layers using two different segregants, TaO and SiO in this order. The process condition of the two oxide layers are similar, with process pressure – mTorr. Oxygen partial pressure is identical. The resulting films have 2% (mole) Ta O and 8% Manuscript received March 06, 2009; revised May 21, 2009. Current ver- sion published December 23, 2009. Corresponding author: Q. Dai (e-mail: qing. dai@hgst.com). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TMAG.2009.2034473 SiO for each of the layer, respectively. The different amount of oxygen was a result of target composition. The structure is then topped with a 7 nm capping layer made of CoPtCr alloy. To understand how each layer of the stack is impacting the roughness, partial stacks were made, where all the layers up to a particular metallurgy of interest are deposited, and everything above it is omitted, except for a 3 nm thick SiN overcoat. These samples are referred to as “SUL,” “TaO ” or “SiO ,” and “Cap” samples. Electrochemical impedance spectra in this study were ob- tained with Princeton Applied Research 283 Potentiostat cou- pled with EG&G 1025 Frequency Analyzer. The frequencies ranged from 0.01 Hz to 10 KHz. An AgCl/KCl microelectrode was used as reference electrode, and DI water was used as elec- trolyte. Due to the close proximity of the electrodes and sam- ples, the IR drop due to electrolyte is negligible [5]. 1 m 1 m AFM images were obtained in order to mea- sure sample roughness. The parameter of interest was chosen as , or maximum valley depth relative to the arithmetic av- erage of the 2-D profile. It should be mentioned that the mea- surement of is not very accurate, due to the very small grain boundary compared to the relatively large AFM tip diameter. For this reason, a replica was produced from an as sputtered disk, using a standard UV-cured photo-resist film. This scheme turned the sharp grain boundaries of the PMR media into peaks, allowing a more accurate depth measurement. III. RESULTS AND DISCUSSIONS To illustrate the principle of EIS applied to overcoat coverage, and the use of equivalent circuits diagrams, let us first limit ourselves to the two limiting cases of the bare magnetic metal (Fig. 1(a)), and the perfect SiN film [Fig. 1(b)]. denotes the resistance of the electrolyte, and is generally small. and are the double layer capacitance and polarization resistance of the electrolyte-metal interface, respectively, and refers to the capacitance of the silicon nitride film which we assume has no conductivity (infinite resistance). is generally much lower than , and therefore can be neglected when the SiN is present. Fig. 2(c) also assumes infinite resistance of the SiN film. On bare metal surfaces without coating, the circuit is dom- inated at low frequencies by the sum of , the resistance of the electrolyte, and , its polarization resistance. At high fre- quencies, the impedance asymptotes to , while the crossover 0018-9464/$26.00 © 2009 IEEE Authorized licensed use limited to: Hitachi San Jose. Downloaded on February 1, 2010 at 20:22 from IEEE Xplore. Restrictions apply.