NANO SURFACE CONTROL OF MAGNETIC TAPE MEDIA FOR THE FUTURE HIGH RECORDING DENSTITY

Transcription

1 NANO SURFACE CONTROL OF MAGNETIC TAPE MEDIA FOR THE FUTURE HIGH RECORDING DENSTITY Y. FUJINO, S. SATO, M. INOUE, W. OKAWA, H. KONDO Magnetic Media R&D Dept., Core Tech. & Network Co., Sony Corp., Sakuragi, Tagajo, , JAPAN; SUMMARY In order to establish a high capacity tape storage system, the morphology control technique of the tape plays a key role. We have developed a series of metal evaporated (ME) recording tapes, which is used for Hi-8 video systems, Digital- Video cassette and so on, with improving their tribological issues. For the next generation tape, aiming at over 1TB per one cassette, further high state of the solution is required, which is not yet established enough. In this paper, we first review the ME tapes from the viewpoint of their base films, and described our study to understand the effect of surface roughness, that is, protrusions on its properties. The tapes with uniform, various areal densities of fine protrusions could be formed by coating minute particles onto smooth Aramid film. The result clearly shows the relationship between the base morphology and the tape characteristics, in other words, durability and electro-magnetic characteristics. Furthermore, we discuss new tape formation, expecting low noise and high durability, by adding surface asperity onto the magnetic layer by means of sputtering and filler coating. Keywords: ME tape, Aramid, base film, surface, filler, sputter texturing 1 INTRODUCTION Along with coming of the so-called Broad Band Era, demands for the high capacity storage solution greatly increasing. Forecasting has made on the future digital contents that more than 1 PB of information will be handled per year in the next several years, and to provide storage solution for such situation, we are now developing a helical recording tape system which aims at as much as 1 TB recording per one Advanced Intelligent Tape (AIT) or 8mm cassette scale. In order to achieve this capacity, more than 5 Gb / in 2 of areal recording density is estimated to be required, and therefore, it is indispensable to apply the metal evaporation or the sputtering technique for the formation of the magnetic layer. Although this thin film magnetic layer possesses high electromagnetic transfer characteristics, its tribological matter have a pile of difficulties comparing with the conventional magnetic particle tape, in which such problems can usually be solved by the paint compositions. In our development on the variety of metal evaporated tapes, enabling Hi-8, DVC and AIT recording, the surface morphology control of the base film plays an key role, combined with the carbon protective layer and lubricant. The surface of the magnetic layer is defined with tracing its bases, hence, only can be controlled by the surface structure of the base film. Tape morphologies must satisfy dilemma of durability and electromagnetic charcteristics. We need to adopt a smoother base to reduce the spacing losses, which usually have a side effect to threaten its durability. Therefore we have developed a series of base film with controlled asperity for each application. And in the high recording density such as 5Gb/in 2, further accurate surface control is required compared with the conventional case of DVC etc. The other way to increase the volume density is to adopt thinner films. As the base films must possess enough 'stiffness', high Young's Modulus film materials are required. Among the candidates such as PBO, polyamide and so on, only Aramid is succeeded in providing a film, worth applying to the magnetic tapes, at this stage. However, the surface controlling does not reach to the state of PET and PEN polyester film, because of the difference in film production process. In this study, we are trying to make a highly controlled Aramid tape surface to satisfy both durability and the electro-magnetic characteristics to achieve high volume density. 2 REVIEW FOR THE ME BASE FILMS Table 1 shows the base film surface morphology and recording density for metal evaporation (ME) 8mm tapes till now, i.e., Hi-8, AIT and AIT-2 application. It is clearly understood that the base films with smoother surface are developed for every high recording density format. Here we briefly review the successive base films for ME. First ME tape was introduced in 1989 for the Hi-8 tape application, and it contributed to provide the high quality home movie. At that time, 9.8 µm polyethylenetelephtalate (PET) film was used as the base. For this application, smoother film was applied to the ME tape compared with that of Audio, VHS etc [1]. The reason is to make use of its superior electromagnetic characteristics in short wave length of.5 µm, which is strongly suppressed by the spacing loss. On the other hand, however, the ME tapes requires asperity because of its severe rubbing condition between metallic head and metallic tape. Therefore, from the point of durability, sufficient roughness should be added. In Ref. 1, 9 nm base intermixed SiO 2 filler and controlled areal density of 25 nm filler coating was adopted on the base film, to adjust the trade-off properties of electro-magnetic characteristics and still

2 lifetime. The surface asperity reduces the real contact area between head and tape, to decrease the shear force per unit tape area. This effect is known to prevent the ME layer from peeling off, to provide durability [2] [3]. Furthermore, the lubricant was applied on the magnetic layer [4]. The area density of 83 Mb/in 2 (5.8 kbpi by 1.64 ktpi) and volume density of 25 GB / in 3 has been achieved with using this base film. The next generation of ME tapes for the consumer electronics was Digital Video Cassette (DVC), released at The electro-magnetic characteristic of this tape was improved as much as 6 db compared with that of Hi-8. In order to achieve this high specification, newly developed Advanced Metal Evaporation technique (AME) [5][6] was applied and offers 4 db of improvement, and the other 2 db was attributed by the base film designing. To realise the small DVC size cassette, 6.3 µm of thin base film was required, which means the disadvantage in tape stiffness. The stiffness is defined by Eδ 3, where E is young's modulus and δ is the thickness, and that strongly correlate with the signal output. The PET base with ca 3% reduced size of coated and inner filler was adopted in order to compensate for it. This construction lead to the reduction of durability, and therefore, DLC layer onto the magnetic layer [7], and newly developed amine salt modified lubricant [8] was adopted. The similar technique with DVC ME magnetic layer and base film was applied to the AIT system [9]. This system could achieve low space digital data storage with 25GB per one small 8mm cassette. This tape is 168Mb/in 2 (72.6 kbpi by 2.31ktpi) in areal recording density, and the volume density is 75 GB/in 3, owe to the high read/write characteristics in short wave length, and 4 % thin base film. The AIT-2 system, as the next application, is capable of recording 5 GB per one 8 mm cassette, which means the volume recording density of 146 GB/ in 3. In order to meet this requirement, 3 % reduction of the base film was needed again. This time, it was difficult to earn enough signal outputs only by the surface designing. Therefore, as the high young's modulus tape, Aramid was adopted for this application to acquire the stiffness. Among the high young's modulus polymers, excluding polyester film, only Aramid is succeeded in providing the smooth base film suitable for ME tape. On the Aramid film, however, the surface control technique has not been established unlike the case of PET and PEN. Therefore, to establish it, the protrusion of large SiO 2 filler in the film (spike filler) and novel fine organic filler near the film surface was examined before [1]. In that report, it was suggested to be suitable to use inner fillers and narrower shape of fine fillers to achieve the comparable tape characteristics versus DVC PET. As concluded in this section, precise surface control had been successfully accomplished on the base film, to perform both durability and electro-magnetic characteristics. 3 SURFACE CONTROL OF THE ARAMID BASE FILM 3.1 Nano filler coating For the achievement of further high volume density, such as 1TB per one 8mm cassette, it is indispensable to use Aramid base film and to control its surface morphology in high degree. In this section, we report our study to understand the effect of the protrusions, intending to form ideal base surface. We used Aramid base film whose thickness is 4.4 µm, same as AIT-2, and formed fine protrusions onto it with controlled various densities. Table 1 base films for 8mm ME tape applications

3 The experimental details are as follows. An ultrasmooth Aramid base film (Ra=.9nm) with and without spike protrusions was used for this investigation. The areal density with this protrusion was 2 x 1 4 /mm 2. The spherical SiO 2 minute particles (12 / 18 nm in diameter) were coated onto it with a binder solution. Then the Co deposition were carried out under the induction of O 2 to form 5nm of Co-CoO magnetic layer, whose Hc were adjusted to be about 11 ka/m (or 1.4 koe). The DLC protective layer, back layer, and lubricant layer formation were followed. The fine protrusion densities, formed with nano filler coating, are easily controlled with the filler concentration. Fig. 1 shows the relation between the filler concentration and the filler density coated on the film. The densities are linearly incleased along with the filler content. The difference in the slope for 12nm and 18nm particles, can be simply understood as the volume difference, i.e., the 18nm particle is about 3 times as large as 12nm in the volume. Those fillers are firmly immobilised with the cobalt layer. The sample tapes are listed in Table 2. A wide range of fine filler density was examined. Conventional PET film for DVC was also applied for the comparison. SEM images of PET film and Aramid films with and without spike protrusions are shown in Fig. 2. The surface roughness is enhanced by shadowing method, that is, to deposit thin silver vapor at low incident angle. The Aramid film is proved to have quite smooth surface. The sprinkled protrusions in Fig. 2 (c) are the spike protrusion, and the surface asperity observed in the conventional PET is the binder coating layer to add the runnability for the ME tape as said above. The filler coated surface is shown in Fig 3 as an AFM image. Fig. 3 is the base film surface before the coating. The smoothness of the Aramid film can be confirmed again, and it is also understood that the coated filler forms uniform and well dispersed fine protrusions. Typical surface images of each sample tape (Sample A, B, D and E in Table 2) are displayed in Fig. 4. In comparing those figures, the size difference in fine fillers (D vs. E), and fine and spike protrusion (in B) can be observed. 3.2 Electro-magnetic characteristics The Read/Write characteristics of those tapes were measured with using drum tester. The tester was equipped with MIG (gap =.22 µm) and MR (g =.18 µm) head, and the head-media relative speed was 6.8m/s. The carrier signal was recorded at 24MHz (18kfci) and the noise level was calculated from the total background signal intensity, read at 1 to 26 MHz. The result of this measurement is displayed as the relative db value, by comparison with Sample A (the conventional PET based tape). The output intensity is likely to depend upon the stiffness of the base films. In Fig. 5, it is clear that the output diminishes along with the decreasing stiffness. Figure 6 shows the signal output, noise output and S/N ratio for each sample tape, using db scale versus conventional PET based tape (Sample A). filler density / 1 7 mm filler concentration / wt% Fig. 1 Relationship between filler concentration and protrusion density for a) 12nm and b) 18nm SiO 2 fillers Sample Base film Coated filler (1 7 /mm 2 ) A PET for DVC (6.3µm) B Aramid with 2x1 4 /mm 2 spike 12nm C protrusions (4.4 µm) 18nm.5.9 D Ultra-Smooth Aramid 12nm E (4.4 µm) 18nm Table 2 sample ME tapes Comparing with the reference, the signal intensities were low in every case in spite of the smoother surface. This should be caused by the lower stiffness of the thinner film. The 4.4 µm Aramid shows 13 % lower stiffness than 6.3 µm PET film though the Young's modulus is 2.6 times higher, whichh is estimated by the former equation. Based on the finding of Fig. 5 about 1dB of disadvantages can be estimated. However, higher S/n ratio has been achieved for the tape without spike protrusions than the PET tape. From the results in Fig. 6, it is also proven that the signal level diminishes with the fine filler density while the noise level seems to increase suddenly at fine protrusion over a certain density. Form the result in Fig. 6, it may between x1 7 /mm 2 for 12 nm fine fillers. The spike protrusions do not affect to the signal but to the noise level that reduces the S/N ratio. Noticeable is that the signal intensity differs about 2dB between 12nm and 18nm filler coating, which well corresponds to the calculated spacing loss. That is to say, the spacing loss is said to be calculated in kd/λ, where k is the spacing loss coefficient (~1 for ME tape [11]), λ is the wave length 3 nm and d is the spacing, and from which, 2 db corresponds to 6 nm difference in spacing. It is probably that the filler size directly affect the spacing and further improvement can be expected by applying finer fillers. As a summary of the electro-magnetic characteristics, the two protrusions were known to change the tape S/N ratio. The existence of the spike protrusions rises the noise level to reduce the S/N ratio. The fine protrusions change the signal and noise intensity, according to their densities and sizes, that is, the height of the fine protrusions seems directly affect the spacing losses, while the higher filler densities lowers the signal level and rises the noise level. Therefore, the finer protrusions are and smoother surface is needed for the media with high S/N ratio.

4 (c) 6 µm Fig. 2 SEM images of applied films, that is, conventional PET for DVC, ultra-smooth Aramid and (c) spike filler added Aramid. 15 nm Fig. 3 AFM images of the Aramid tapes before and after fine filler coating. A B D E 1 µm Fig. 4 SEM images of sample ME tapes. The note A-E corresponds to Table 1, and the fine filler densities for B,D,E are.3, 2.5, and.6 x17/mm2, respectively 3.3 Durability In order to evaluate the durability of these tapes, friction test and drive short shuttle test have been carried out. The friction tests were made against stainless guide pin (2mmφ.1S). The sample tapes were cut into 8 mm in width and were wrapped around a quadrant and 1 g weight was set to it. Then it was scraped back and forth repeatedly. The friction coefficients were calculated from the pulling tension and were recorded in every reciprocating motion. The drive short shuttle test was carried out using AIT drive, recording random signal and reading the defined region of the tape repeatedly. The error rate was recorded in every pass. Figure 7 shows the kinetic frictional coefficient (µk) change of the filler coated tapes. Because each tape shows almost same change in the µk, only the results for the 12 nm filler coated tapes without spike protrusions (Sample D) are presented in the figure, as the typical example. The µk shows gradual increase till reaches to µk~.4, and rises immediately. Therefore, in Fig. 8, the pass numbers at µk of exceeding.4 are summarised as friction durability against stainless pin. From the result, it was suggested that Aramid tapes with spike protrusions show high durability and some of them show better result than conventional PET (Sample A). On the other hand, the result for Sample D and E (only with fine protrusions, without spike prtrusions) shows almost same low durability. It is certainly that the spike protrusions contribute to the durability against the stainless pin rubbing. In general, however, future tape storage systems would adopt milder condition than DVC system, because MR head, which has low margin to the head wear, will be equipped. Consequently, the tapes without spike protrusions showing 5 to 9% durability versus PET, is

5 possible to be enough. The proper combination should be selected according to the applied systems. Figure 9 shows the typical error rate change in short shuttle test. The error rate rises gradually at first, and suddenly increases. Fig. 1 summarises this point of inflection, as a short shuttle durability. I / db I / db stiffness / cm g 2 s -2 Fig. 5 The signal output level dependence on tape stiffness B C D E S/N media Noise Signal filler density / 1 7 mm -2 Fig. 6 Signal and Noise intensity, and the S/N ratio of the filler coated tapes vs. PET. The note B-E corresponds to Table 2 The durability shows apparent relationship to fine protrusion densities. Spike protrusions are also effective to the durability. But in this case, differs from friction test, the spike protrusions seems not always required for the high durability. The Aramid tape, modifying the base surface only by certain densities of fine fillers, obtained a comparable result with conventional PET tape. When the fine protrusion densities are high enough, the tape without spike protrusions also shows good short shuttle durability. To investigate the effect of the protrusions, the tape surface before the test and after the error rate went up were observed by AFM. Fig. 11 shows the AFM images and the bearing curves for Sample D (1.7x1 7 /mm 2 ). The abrasion around the fine filler was observed at some part of the tape surface after the test, and the shown results are for this part. This observed salient was parallel to the head tracing direction. It was also observed in the tested tape which error rate did not start to go up. Hence, this salient itself, is not the origin of error rate. From the results, it can be said that the fine fillers protect the magnetic layer from head scraping, and they are gradually grounded down as the head tracing. When the wear reaches to a certain stage, the error rate may starts to rise. The wear rate of a protrusion should be inversely proportional to the filler density, and that may originates the relationship between the durability and the density. µ k filler density (x1 7 /mm 2 ) shuttle passes Fig. 7 Typical change in µk against friction test. The results for Sample D are shown. The numbers in the figure are denote of the filler density. guide pin test durability / pass A B C D E filler density / 1 7 /mm 2 Fig. 8 Correlation between filler density and durability against guide pin test. The note A-E corresponds to the Sample A-E in Table 2, and the fine filler densities are displayed at the bottom. Error Rate Number of Cycles Fig. 9 Typical change in error rate of short shuttle test. The results for Sample D with 2.5 x1 7 /mm 2 of fine protrusions is shown.

6 short shuttle durability / pass A B C D filler density / 1 7 / mm 2 Fig. 1 Correlation between filler density and durability against drive short shuttle test. The note A-D corresponds to the Sample A-D in Table 2, and the fine filler densities are displayed at the bottom. 2 1 height / nm 3 2µm bearing area / % Fig. 11 The surface AFM images (z: 5nm full scale) and bearing curves for Sample D (1.7x1 7 /mm 2 ), before and after the short shuttle test. height / nm µm bearing area / % Fig. 12 The surface AFM images (z: 5nm full scale) and bearing curves for Sample B (1.4x1 7 /mm 2 ) before and after the drive short shuttle test. In the case with spike protrusions, the tapes show relatively long durability. Fig. 12 is the AFM images and bearing curves of Sample B (12nm 1.4x1 7 mm 2 ), before and after the drive short shuttle test. The large salient around the spike protrusion was observed. But the wear rate seems smaller than the Sample D. The bearing curve suggests that the fine protrusions were worn out, that is to say, the spike protrusion does not worn out beforehand of the fine protrusions. The spike protrusions may, therefore, work to reduce the wear of fine protrusions, to elongate the durability. The mechanism of this error rate change has not been understood in details. The error rate may goes up immediately when the wear reaches to a certain stage, that is, cobalt layer. The protrusion is also said to have the effect of cleaning to prevent drive heads from clogging with abrasions of lubricant or cobalt. Therefore, the error rate may rise when the protrusions lose their cleaning effect by the wear. In conclusion of the durability tests, it was understood that the spike protrusions improve durability of the tapes and is especially mighty to undergoing severe rubbing conditions, such as against stainless guide pin. The fine protrusions, on the other hand, have less durability against the severe rubbing, but very effective to the drive short shuttle durability. Those Aramid tapes show comparable property with conventional PET tapes. 4 FOR THE FURTHER HIGH CHARACTERISTICS In this section, we made further study on the tape morphology, to provide higher characteristics for the future high recording density. As the ME layer is formed onto the roughened surface, the Co-O column around the filler tends to grow irregularly (schematically represented in Fig. 13 ). This configuration may result in increasing the medium noise level. More over, one reason of the signal degradation in still or shuttle test is said to be caused by the severe rubbing between the metallic head and outcropped cobalt from the protective layer or lubricant layer. Consequently, the ideal ME tape should be formed with the magnetic layer deposited on the smooth base film, followed by the layer to add the roughness (Fig. 13 ). It may lessen the noise level, provide wide margin for wear, and improve the ME oxidation property. For the achievement of this construction, we applied two method on the smooth metal evaporated layer, i.e., one is to coat the 12 nm SiO 2 filler and one is to form sputter deposited texture, both of which are followed by carbon protective layer deposition. The conventional AIT-2 base was also applied as a reference. The samples are listed in Table 3. The sputtering texture can be shaped by hard high surface energy materials onto the magnetic layer. We have examined over some materials and conditions, and successively formed the texture onto the ME layer. AFM images of the tape with sputter induced texture is shown in Fig.14.

7 a) b) DLC layer ME layer base film Fig. 13 A schematic drawing of the cross-sectional view of the ME tape. is the conventional method to add the roughness by the base morphlogy. In, the roughness is added onto the ME layer. The bearing curves for Sample I III in table 3 and D, E in Table 2 are shown in Fig.15. It is clear that the newly developed tapes are modified with protrusions, with narrow bearing curves. At first, S/N ratio of Sample I was measured. The result is represented in Fig. 16 using conventional AIT-2 film based tape (Sample IV) as a reference. The filler under coat tape (Sample D) is also shown. As the stiffness is identical, the differences in surface morphology is purely compared. From the result, it was known that no fewer than 4.3 db can potentially be acquired using this base film even after forming 6nm thickness of carbon protective layer. The filler under coated tape could earn 1.8 db of S/N ratio, but seems leaving much room for improvement. The S/N ratio of Sample II and III are also graphed out in Fig. 16. The S/N ratio for the filler top coated tape (II) reaches +3.1dB versus conventional AIT-2 based tape, that assures 2.2 Gb/in 2, i.e., no less than Sample Base film After treatment I 4.4µm ultra- None II smooth Aramid 12nm filler coating III film Sputter texture IV conventional AIT-2 ase film (4.4µm Aramid) Table 3 Sample tapes 2µm Fig. 14 AFM image of the sputter induced texture (Sample III). (z: 3nm full scale) height / nm D E 1 II III 5 I bearing area / % Fig. 15 The bearing curves for Sample D(2.5x1 7 /mm 2 ), E(.6x1 7 /mm 2 ) and I - III. 1.4 TB/in 3 of recording density. The value measured at sputter textured tape (III) shows +1.3dB of S/N ratio, reagardless of the similar bearing curve with II. To the end to evaluate the runnability of those tapes, friction tests were made. The results are shown in Fig. 17 with the result of smooth tape (I) and filler under coated tape (D). Friction of the filler top coated base was not improved from that of the smooth tape. But the sputter-induced texture shows alomost same behaviour and coefficient with that of the filler under coated tape. Though the output level seem to be insufficient, this new configuration tape with sputter induced texture is able to control its surface morphology easily by the sputtering conditions, and is thought to have expected high potential to combine tape durability and low noise intensity. We have already succeed in developing an over 5 Gb/in 2 recording, which will lead to 1TB recording per one AIT cassette scale. This high density recording is accomplished mainly owe to size reducing of the magnetic domain and to GMR head. In such a system, the thickness of the magnetic layer will be less than 3 nm. The 1 nm of under coated filler would lead the crucial increase in the noise intensity. I / db D I II III S N S/N -3 Fig. 16 Signal, noise and S/N of the ME tapes sample tape D (2.5x1 7 /mm 2 )and I - III (in table 3)

8 µ k III shuttle / pass Fig. 17 The change in µ k of the Sample I - III and D (2.5x1 7 /mm 2 ) Hence, we are now combining these sputtering methods aiming at the data storage system of next generation. 5 REFERENCES [1] K. Chiba, K. Sato, Y. Ebine, Metal evaporated tape for high band 8mm video system, IEEE Transactions on consumer electronics, Vol.35, No.3, pp , 1989 [2] H. Osaki, K. Fukushi, K. Ozawa, Wear mechanisms of metal-evaporated magnetic tapes in helical scan videotape recorders, IEEE Transactions on magnetics, Vol.26, No.6, pp , 199 [3] H. Osaki, Role of surface aspirates on durability of metal evaporated magnetic tapes, IEEE transactions on magnetics, vol.29, No.1, pp. 11-2, 1993 I II D [4] H. Kondo, M. Aoki, J. Seto, Lubrication mechanism of phosphite and thiophosphite on magnetic thin film media, Wear, vol. 16, pp , 1993 [5] T. Kawana, S. Onodera, T. Samoto, Advanced metal evaporated tape, IEEE transaction on magnetics, Vol. 31, No.6, , 1995 [6] K. Kanota, H. Inoue, A. Uetake, M. Kawaguchi, K. Chiba, Y. Kubota, A high density recording technology for digital VCRs, IEEE Transaction on consumer electronics, Vol. 36, No.3, pp , 199 [7] Y. Kaneda Tribology of Metal-evaporated tape for high density magnetic recording, IEEE transactions on magnetics, vol.33, No.2, pp , 1997 [8] H. Kondo, A. Seki, A. Kita, Comparison of an amide and amine salt as friction modifiers for a magnetic thin-film medium, STLS Transaction on Tribology, Vol.37, pp99-14, 1994 [9] T. Ozue, Y. Watanabe, T. Kohno, T. Hirose, K. Koseki, Development of high density recording for advanced tape streamer, IEEE Transaction on Magnetics, Vol. 32, No.5, pp , 1996 [1] S. Sato, Y. Arisaka, S. Matsumura, Surface design of Aramid film for future ME tapes, IEEE Transaction on Magnetics, Vol.35, No.5, pp , 1999 [11] T. Samoto, S. Onodera, K. Sato, Analysis of spacing loss factor of obliquely evaporated tape, Technical report of IEICS, No.6, pp.7-12, 1996 (in Japanese)