WO2000072315A1

WO2000072315A1 - Carbon nitride overcoat layer and method for forming

Info

Publication number: WO2000072315A1
Application number: PCT/US2000/014641
Authority: WO
Inventors: Xi Chu; Bing Zhang
Original assignee: Hyundai Electronics America, Inc.
Priority date: 1999-05-26
Filing date: 2000-05-25
Publication date: 2000-11-30
Also published as: AU5168100A

Abstract

A carbon nitride overcoat layer of a magnetic recording medium is deposited onto a magnetic layer within a plasma chamber. The magnetic layer is supported by a substrate. A negative bias voltage, typically in the range of -50 volts to -400 volts, and more preferably in the -75 volts to -250 volts range, is applied to the magnetic layer during the depositing step. A 5nm thick carbon nitride overcoat layer applied in this manner has a durability at least as good as a conventional carbon nitride overcoat layer 10nm thick. Therefore, by providing equal durability at reduced thickness, the signal to noise ratio is improved significantly.

Description

CARBON NITRIDE OVERCOAT LAYER AND METHOD FOR FORMING

BACKGROUND OF THE INVENTION A magnetic recording medium, such is used with magnetic recording disks, typically includes the following layers in the following order: a substrate, an underlayer such as chromium, a magnetic layer such as CoCrTa, an overcoat layer and a lubricant layer such as perfluoropolyether. A magnetic recording head is typically mounted on a slider that is supported above the rotating disk by a suspension. The moving air created by the rotating disk can provide an air bearing which supports the slider and magnetic head at an appropriate height above the rotating disk, such as about 10 to 20 nm.

An overcoat layer is necessary to protect the magnetic layer from damage when the slider happens to contact the magnetic recording medium or when other objects come into contact with the recording medium. When the disk starts rotating, the slider takes off, being suspended by a thin layer of air above the disk, and then lands on the disk when the disk stops rotating. This is referred to as contact-start-stop(CSS). To have acceptable durability, the overcoat layer, usually covered with a lubricant, must withstand several tens of thousands of CSS cycles without failure.

It is relatively simple to provide an effective overcoat layer if thickness is not a concern. However, the distance between the magnetic head on the slider and a magnetic layer should be minimized to reduce magnetic signal strength loss:

Overcoat layers are typically made from hydrogenated carbon (CH_X) or carbon nitride (CN_X) with CN_X typically preferred. Conventional magnetic recording disks require a CN_X overcoat thicknesses of about 12 nm or greater to be exceptionally durable. However, it would be useful to be able to reduce the thickness of the overcoat layer to improve the signal to noise ratio (SMNR). As suggested in Fig. 1, an SMNR enhancement of about -2 dB can be achieved by reducing the thickness of the overcoat layer from about 12nm to about 3nm and about -1.5dB when the thickness is reduced from about 12nm to about 5nm.

SUMMARY OF THE INVENTION The present invention provides for a toughened carbon nitride overcoat layer which provides substantially the same durability as conventional carbon nitride overcoat layers of a much greater thickness. Acceptable durability can be achieved when a carbon nitride overcoat layer made according to the invention is less than about 8nm thick, typically about 5-7nm thick. The carbon nitride overcoat layer of a magnetic recording medium is formed by first depositing a carbon nitride overcoat layer onto a magnetic layer within a plasma chamber. The magnetic layer is supported by a substrate. A negative bias voltage, typically in the range of -50 volts to -400 volts, and more preferably in the -75 volts to -250 volts range, is applied to the magnetic layer during the depositing step. A 5nm thick carbon nitride overcoat layer applied in this manner has a durability at least as good as a conventional carbon nitride overcoat layer lOnm thick. Therefore, by providing equal durability at reduced thickness, the signal to noise ratio is improved significantly. In addition to maintaining the carbon nitride overcoat layer at a negative bias during the deposition step, the durability of the overcoat layer can be increased over conventional carbon nitride layers by employing low levels of sputtering wattage, typically in the range of about 2 W/cπT to about 14 W/cm and more preferably in the range of about 3 W/cm² to about 8 W/cm², and a high ion-to-neutral ratio within the deposition chamber, such as about 0.4 to 2 and more preferably about 0.8 to 1.5. These values can be contrasted with a typical conventional ion-to-neutral ratio of about 0 and a sputtering range of about 12-24 W/cm ^'

Other features and advantages of the invention will appear from the following description in which the preferred embodiment has been set forth in detail in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 is a plot of signal-to-noise ratio SMNR vs. carbon overcoat thickness obtained from magnetic disks using giant magneto resistance (GMR) heads at a recording density of 220 kFCI; Fig. 1A is a simplified cross-sectional schematic representation of a magnetic recording medium of the type including a substrate, an underlayer, a magnetic layer, an overcoat layer and a lubricant layer;

Fig. 2 is a schematic diagram of a DC magnetron deposition system with a magnetic recording disk element coupled to a bias power supply;

Fig. 3 plots wear depth vs. substrate bias as the result of an atomic force microscopy (AFM) nano-scratch test of magnetic disks with a 7.5 nm thick carbon nitride overcoat layer at various substrate bias voltages;

Fig. 4 is a plot similar to Fig. 3 but, showing the AFM nano-scratch test results for different concentrations of nitrogen in the carbon nitride overcoat layer;

Fig. 5 shows test results similar to those of Figs. 3 and 4 but where target power densities are varied;

Fig. 6 illustrates the results of a contact-start-stop (CSS) test of a 5nm thick carbon nitride layer deposited at a power density of 8 Watts/cm without substiate bias showing failure at about 27,000 CSS cycles;

Fig. 7 shows CSS test results of a 5nm thick carbon nitride film deposited at a power density of 8 Watts/cm² at a bias voltage of -220V showing no failure at 80,000 cycles;

Fig. 8 illustrate the results of a CSS test of a 5nm thick hydrated carbon (CH_X) overcoat layer sputtered in a mixture of argon and H₂ gas at a power density of 8 Watts/cm² showing failure at only about 1000 cycles;

Fig. 9 illustrates the results of a CSS test of a 4nm thick carbon nitride film deposited at a power density of 7 Watts/cm² and a bias voltage of -220V showing no failure through at least about 35,000 CSS cycles; Fig. 10 shows G peak position derived from deconvoluted Raman spectra of CN_X films sputtered with increasing substrate bias voltage; and

Fig. 11 shows Raman G peak width from the same series of CN_X films shown in Fig. 10.

DESCRIPTION OF THE SPECIFIC EMBODIMENTS

Fig. 1A illustrates, quite schematically, a section of a magnetic recording medium 2, typically a magnetic disk 2, including a substrate 4 made of Al-Mg alloy on which an underlayer 6 of chromium is optionally deposited. A magnetic layer 8, typical CoCrTa, is applied to underlayer 6. Carbon nitride overcoat layer 10 is, in one preferred embodiment, produced in a magnetron DC plasma chamber 12 as shown in Fig. 2. A lubricant layer 11, typically of pefluoropolyether, is applied on top of overcoat layer 10. Plasma chamber 12 includes a housing 14 having an inlet 16 through which argon and nitrogen is directed and an outlet 18 coupled to a conventional vacuum pump 20. Two graphite (carbon) targets 22 are supported within the interior 24 of housing 14 and are coupled to a target power supply 26 which supplies the targets with negative DC voltage. The values described below were for a plasma chamber 12 made by Intervac of San Jose, California as model number MDP 250B. In one test run, chamber 12 was back filled with nitrogen and argon gas at a pressure of 5 mTorr. Target power supply 26 provided lkW sputter power to each target 22, each target, in this embodiment, having an area of about 85 cm². This created a plasma within chamber 12 which caused argon ions to bombard targets 22 resulting in carbon atoms being sputtered from the targets. The carbon atoms were transported through the plasma and reacted with nitrogen atoms to form carbon nitride on the surface of disk element 28. Disk element 28 included substrate 4, underlayer 6, and a magnetic layer 8; overcoat layer 10 was created within chamber 12. Since it is typically desired that magnetic disk 2 be a two-sided disk, substrate 4 is typically sandwiched between an underlayer 6 and a magnetic layer 8 on each side so that overcoat layers 10 can be formed on both sides of disk element 28.

The above-described test run procedures and structure is generally conventional. What is not conventional is the use of a bias power supply 30 to negatively bias disk element 28 during deposition. Bias power supply 30 is connected to disk element 28 through a disk element holder 32. The use of bias power supply 30 significantly increases the durability of overcoat layer 10. It is believed that when negative voltage is applied to disk element 28, some of the argon and carbon ions in the plasma are accelerated towards disk element 28 and impinge on the disk element surface. The ion bombardment is believed to knock out loosely bonded surface atoms improving the overall film strength. (See, R. Messier et al., "Revised Structure Zone Model for Thin Film Physical Structure" J. Vac.Sci.TechnoL, A2 (1984) 500.)

The voltage applied to disk element 28 is typically between about -50VDC and about -400VDC, and more preferably between about -75VDC and about -250VDC. The duration of the processes using the above described equipment typically takes about between about 1 to 5 seconds.

In addition to the application of a negative bias to disk element 28, is been found that durability of carbon nitride overcoat layer 10 can be further enhanced by a high ion-to-carbon-neutral ratio. Ion-to-carbon-neutral ratio is the ratio of bombarding ions (typically argon and carbon) to carbon atoms being deposited.

The cuπent collected through bias power supply 30 measures the amount of ion bombardment on the surfaces of disk element 28 during deposition. For deposition rates of about 5 to 15 A (0.5 to 1.5nm) per second, the ion-to-carbon neutral ratio is calculated to be between about 1.5 to 0.5. The durability of carbon nitride overcoat layer 10 improves with increasing ion-to-carbon neutral ratio; that is, an ion-to-carbon neutral ratio of about 1.0 provides an improved durability nitride overcoat layer 10 as opposed to an ion-to-carbon neutral ratio of about 0.4; in one example the durability increased from 18K (for a 0.4 ratio) to 48K (for a 1.0 ratio). During testing, sputtering power densities of about 2 to 24 Watts/cm² were used. At a power density of about 10 Watts/cm², the durability of carbon nitride overcoat layer 10 improves and continues to improve with lower sputtering power densities. Fig. 5 illustrates results of an AFM nano-scratch test of a magnetic disk with a 7.5nm thick carbon nitride overcoat layer 10 at different target power densities. It can be seen that wear depths, and thus surface abrasion resistance, is minimized with the target power density in the 2 to 12 Watts per centimeter range.

To vary the amount of nitrogen in overcoat layer 10, tests were conducted in which the sputter gas N₂ to Ar+N₂ volume ratio was in the range of 0.05 to 0.9 resulting in a range of nitrogen in overcoat layer 10 of about 5% nitrogen to about 28% nitrogen as is measured by x-ray photoelectron spectroscopy (XPS). It was found that the durability of carbon nitride overcoat layer 10 degraded to an undesirable extent if the percent of nitrogen in layer 10 was higher than about 25% or lower than about 5% nitrogen by atomic percent, and more preferably between about 20% and 8%. Results of these tests are shown in the graph of Fig. 4. All the wear depth measurements were determined using an atomic force microscopy (AFM) equipped with a diamond tip having a three-sided pyramid tip with an apex angle of 60°. This diamond tip was used to scratch overcoat layer 10 using a load of about 3x10 ^"5N normal to the surface of layer 10. The wear mark was imaged by AFM after the test. The wear depth measured was used with Figs. 3, 4 and 5 to characterized the film mechanical properties; that is, deeper wear depths were considered to indicate poorer durability of the overcoat layer 10. Also, the"results of contact-start-stop (CSS) tests are plotted at Fig. 6-9 to compare wear qualities of the disks and show failures of the disks in Figs. 6 and 8.

Fig. 3 illustrates results of an AFM nano-scratch test of magnetic disks 2 having a 7.5nm thick CN_X overcoat layer 10 deposited at various substrate bias voltages. The carbon nitride overcoat layers were DC sputtered as discussed above with a substrate bias potential from 0 to -500VDC. As can be seen from this test, the wear depth of the carbon nitride overcoat layer is reduced substantially when disk element 28 is negatively biased during the process of depositing carbon nitride overcoat layer 10. While even a small amount of negative bias, such as -100VDC, improves the durability of overcoat layer 10, even more substantial improvements are achieved when the negative bias voliages, for this series of tests, range from about -150 volts to about -300 volts, and more preferably about -200 to -300 volts.

Fig. 5 illustrates the results of an AFM nano-scratch test of magnetic disks with a 7.5 nm thick carbon nitride overcoat layers 10 formed at different power densities of targets 22. This test illustrates that for this particular chamber 12, power density in the range of about 2 to 12 Watts/cm² results in the overcoat layers 10 having the best durability.

Fig. 6 illustrates the result of a CSS test of a magnetic medium having a 5nm thick carbon nitride overcoat layer deposited at a power density of 8 Watts/cm but without biasing of the substrate. The disk lasted about 27,000 cycles and then failed. In contrast, Fig. 7 illustrates the results of a CSS test of a disk substantially identical to that tested in Fig. 6 but where a bias voltage of -220 volts was applied to disk element 28 during the deposition of overcoat layer 10. As can be seen from the figure,- the disk lasted to 80,0000 cycles without failure; at that time the test was stopped. This is a graphic illustration of how biasing disk element 28 while depositing overlayer 10 increases the durability of the carbon nitride overcoat layer 10. Fig. 8 illustrates the results of a test similar to that shown in Fig. 6 but in which overcoat layer 10 is a hydrogenated carbon film. This disk failed at about 1,000 cycles showing that a 5nm thick carbon nitride overcoat layer has significantly greater durability than a similar thickness of a hydrogenated carbon overcoat layer. Fig. 9 illustrates the results of a CSS test of a 4nm thick carbon nitride overcoat layer deposited at a power density of 7 Watts/cm² and a bias voltage of -220 volts. The disk lasted about 35,000 cycles without failure; at that time the test as halted. This illustrates the fact that even a 4nm thick carbon nitride overlayer has very good durability when deposited with a negative bias voltage.

Raman spectroscopy is a useful technique for obtaining information regarding the bonding characteristics of carbon atoms within overcoat layer 10. (See M.A. Tamor and W. C. Vassell, "Raman Fingerprinting of Amorphous Carbon Films ", J.Appl.Phys., 76(6)( 1944)3823.) Fig. 10 plots G peak position vs. bias voltage. Raman spectra is obtained by shining a laser light normal to film surface and plotting the intensity of inelastic scattered light as a function of the frequency difference with respect to the incident laser light. Each intensity peak corresponds to a vibrational mode of scattering molecule. Raman spectra of CN_X films consists of two broad and overlapped peaks, i.e. G peak (-1570 cm^"1) and D peak (-1400 cm^"'). The position and peak width correlate to bonding state and properties of DLC films. (See Tamor et al.).

The measurements were made with Renishaw (UK) Raman Imaging Microscope. G peak position shifts to a lower wave number indicating an increase of compressive stress in overcoat layer 10. In general, low G band bond structure correlates with better CSS performance. Fig. 11 shows the Raman G peak width from the same series of disks shown in Fig. 10. G peak width is maximized as bias voltage moves towards about -200V. An increase of G peak width has been correlated to higher film hardness in hydrogen- free carbon films. Accordingly, Figs. 10 and 11 illustrate that Raman spectra show a clear indication of changes in bonding of the carbon atoms within carbon nitride overcoat 10 when the substrate is biased during the deposition. The changes in the carbon bonding state are believed to be responsible for the enhanced durability observed.

U. S. Patent No. 4,664,976 and 5,767,512 discuss sputtering of carbon nitride overcoat layers. See also the following references: M.Y. Chen et al, "Synthesis and Tribological Properties of Carbon Nitride As A Novel Superhard Coating and Solid Lubricant", Trib.Trans., 36 (1993) 491; and R.L. White et al "RF-Sputtered Amorphous CN_X for Contact Recording Applications", Triboiogy, 6 (1996) 33.

Any and all patents, patent applications and the printed publications referred to above are incorporated by reference. Modification and variation can be made to the disclosed embodiment and method without departing from the subject invention as defined in the following claims.

Claims

WHAT IS CLAIMED IS:

1. A magnetic recording medium comprising: a substrate; a magnetic layer on top of the substrate; a carbon nitride overcoat layer on top of the magnetic layer; the overcoat layer having a thickness of less than 8nm; the overcoat layer having a wear depth of no more than about 2nm using a surface-scratching tip, having the following characteristics: a three-sided pyramid tip having a 60° apex angle to scratch the overcoat layer using a normal force of 3xlO^"5N.

2. A method for forming a carbon nitride overcoat layer of a magnetic recording medium comprising: depositing a carbon nitride overcoat onto a magnetic layer within a deposition chamber, the magnetic layer supported by a substrate; and applying a negative bias voltage to the magnetic layer during the depositing step.

3. The method according to claim 2 wherein the depositing step is carried out in a plasma chamber.

4. The method according to claim 2 wherein the depositing step is carried out in a DC plasma deposition chamber containing Ar and N₂ gas and a carbon target.

5. The method according to claim 2 wherein the depositing step is carried out so that the deposition chamber has a carbon ion-to-carbon neutral ratio of between about 1.5 to 0.5.

6. The method according to claim 2 wherein the applying step is carried out with the bias voltage being a negative bias voltage between about -50VDC and -400VDC.

10 7. The method according to claim 2 wherein the applying step is carried out with the bias voltage being a negative bias voltage between about -75VDC and -200VDC.

8. The method according to claim 2 wherein the depositing step is carried out in a plasma deposition chamber containing a carbon target by sputtering at a sputtering power density of between about 2 and 12 Watts/cm².

9. The method according to claim 2 wherein the overcoat layer has between about 8 and 20 % N by atomic percent.

10. The method according to claim 2 wherein the overcoat layer has between about 5 and 28% N by atomic percent.

11. The method according to claim 2 wherein the applying step is carried out with the total gas pressure within the chamber between about 2 to lOmTorr.

12. The method according to claim 2 further comprising the step of applying a lubricant to the overcoat layer.

13. The method according to claim 4 further comprising the step of applying a lubricant to the overcoat layer.

14. A magnetic recording medium made according to the method of claims 12.

15. A magnetic recording medium made according to the method of claim 13.