The present invention relates to thin film magnetic heads, and more particularly, this invention relates to a head with a write head positioned under a read head.
In a typical head, an inductive write head includes a coil layer embedded in first, second and third insulation layers (insulation stack), the insulation stack being located between first and second pole piece layers. A gap is formed between the first and second pole piece layers by a gap layer at an air bearing surface (ABS) of the write head. The pole piece layers are connected at a back gap. Currents are conducted through the coil layer, which produce magnetic fields in the pole pieces. The magnetic fields fringe across the gap at the ABS for the purpose of writing bits of magnetic field information in tracks on moving media, such as in circular tracks on a rotating magnetic disk or longitudinal tracks on a moving magnetic tape.
The second pole piece layer has a pole tip portion which extends from the ABS to a flare point and a yoke portion which extends from the flare point to the back gap. The flare point is where the second pole piece begins to widen (flare) to form the yoke. The placement of the flare point directly affects the magnitude of the magnetic field produced to write information on the recording medium. Since magnetic flux decays as it travels down the length of the narrow second pole tip, shortening the second pole tip will increase the flux reaching the recording media. Therefore, performance can be optimized by aggressively placing the flare point close to the ABS.
Another parameter important in the design of a write head is the location of the zero throat height (ZTH). The zero throat height, for the prior art designs, is the location where the first and second pole pieces first separate from one another after the ABS. ZTH separation is imposed by an insulation layer, typically the first insulation layer in the insulation stack. Flux leakage between the first and second pole pieces is minimized by locating the ZTH as close as possible to the ABS.
Unfortunately, the aforementioned design parameters require a tradeoff in the fabrication of the second pole tip. The second pole tip should be well-defined in order to produce well-defined written tracks on the rotating disk. Poor definition of the second pole tip may result in overwriting of adjacent tracks. A well-defined second pole tip should have parallel planar side walls which are perpendicular to the ABS and a predictable size. This definition is difficult to achieve because the second pole tip is typically formed along with the yoke after the formation of the first insulation layer, the coil layer and the second and third insulation layers. Each insulation layer includes a hard-baked photoresist having a sloping front surface.
After construction, the first, second and third insulation layers present front sloping surfaces which face the ABS. The ZTH defining layer rises from a plane normal to the ABS at an angle (apex angle) to the plane. After hard baking of the insulation layers and deposition of a metallic seedlayer the sloping surfaces of the insulation layers exhibit a high optical reflectivity. When the second pole tip and yoke are constructed, a thick layer of photoresist is spun on top of the insulation layers and photo patterned to shape the second pole tip, using the conventional photo-lithography technique. In the photo-lithography light imaging step, ultraviolet light is directed vertically through slits in an opaque mask, exposing areas of the photoresist which are to be removed by a subsequent development step. One of the areas to be removed is the area where the second pole piece (pole tip and yoke) is to be formed by plating. Unfortunately, when ultraviolet light strikes the sloping surfaces of the insulation layers in a flaring region of the second pole piece, the ultraviolet light is reflected forward, toward the ABS, into photoresist areas at the sides of the second pole tip region. After development, the side walls of the photoresist extend outwardly from the intended ultraviolet pattern, causing the pole tip plated therein to be poorly formed. This is called “reflective notching”. As stated hereinabove this causes overwriting of adjacent tracks on a rotating disk. It should be evident that, if the flare point is recessed far enough into the head, the effect of reflective notching would be reduced or eliminated since it would occur behind the sloping surfaces. However, this solution produces a long second pole tip which quickly reduces the amount of flux reaching the recording medium.
The high profile of the insulation stack causes another problem after the photoresist is spun on a wafer. When the photoresist is spun on a wafer it is substantially planarized across the wafer. The thickness of the resist in the second pole tip region is higher than other regions of the head since the second pole tip is substantially lower on the wafer than the yoke portion of the second pole piece. During the light exposure step the light progressively scatters in the deep photoresist like light in a body of water causing poor resolution during the light exposure step.
Another prior art scheme for minimizing the reflective notching and poor resolution problems is to construct the high resolution pole tip atop a planarized surface. The bottom second pole tip is constructed first, before the insulation layers to eliminate the reflective notching problem. After forming the first high resolution first pole piece layer and the write gap layer, processing can proceed to make the coil layer and the second and third insulation layers. The top second, wider, pole tip is then stitched (connected) to the bottom second pole tip and extends from the ABS to the back gap. Since the bottom second pole tip is well-formed and its size predictable, this fabrication method is an advance on the longitudinal head design. However, with this head, the ZTH is dependent upon the location of the insulation layer with respect to the bottom high resolution pole tip. Since the bottom pole tip has to be long enough to provide a sufficient stitching area, this length may result in undesirable flux leakage between the first and second pole pieces. Since the top second pole tip is typically wider than the bottom high resolution pole tip, the second pole piece has a T-shape at the ABS. The upright portion of the T is the front edge of the high resolution pole tip, and the cross of the T is the front edge of the top second pole tip. A problem with this configuration is that during operation, flux fringes from the outer corners of the top second pole tip to a much wider first pole piece at the ABS, causing adjacent tracks to be overwritten.
In another prior art embodiment, the high resolution pole tip is placed atop the stack. Once the high resolution second pole tip is formed, it is desirable to notch the underlying first pole tip of the first pole piece opposite the first and second corners at the base of the second pole tip so that flux transfer between the pole tips does not stray beyond the track width defined by the bottom second pole tip. Notching provides the first pole piece with a track width that substantially matches the track width of the high resolution second pole tip. A prior art process for notching the first pole piece entails ion beam milling the gap layer and the first pole piece, employing the bottom second pole tip as a mask. The gap layer is typically alumina and the first and second pole pieces and pole tips are typically Permalloy (NiFe). The alumina mills more slowly than the Permalloy; thus the top of the bottom second pole tip and a top surface of the first pole piece are milled more quickly than the gap layer. Further, during ion milling, there is much redeposition (redep) of alumina on surfaces of the workpiece. In order to minimize redep, the milling ion beam is typically directed at an angle to a normal through the layers, which performs milling and cleanup simultaneously. The gap layer in the field remote from the first and second corners of the bottom second pole tip is the first to be milled because of a shadowing effect at the first and second corners caused by the bottom second pole tip when the ion beam is angled. In this case, the ion stream will overmill the first pole piece before the gap layer is removed adjacent the first and second corners of the bottom second pole tip in the region where the notching is to take place. After the gap layer is removed above the sites where the notching is to take place, ion milling continues in order to notch the first pole piece. Overmilling of the first pole piece continues to take place in the field beyond the notches, thereby forming surfaces of the first pole piece that slope downwardly from the notches. As is known, such overmilling of the first pole piece can expose leads to the MR sensor or the wider head's coils, thereby rendering the head inoperative.
Even if overmilling of the first pole piece can be controlled, there is potentially a more troublesome problem, namely overmilling the top of the high resolution second pole tip when the unwanted portions of the gap layer are milled and notches are formed. In order to compensate for this overmilling, the aspect ratio (ratio of thickness of photoresist to track width of the high resolution second pole tip) is increased so that a top portion of the top of high resolution bottom second pole tip can be sacrificed during the milling steps. When the aspect ratio is increased, definition of the high resolution second pole tip is degraded because of the thickness of the photoresist, discussed hereinabove, resulting in track overwriting.
Another problem with the prior art merged MR head is that the profile of the MR sensor between the first and second gap layers is replicated through the second shield/first pole piece layer to the write gap layer causing the write gap layer to be slightly curved concave toward the MR sensor. When the write head portion of the merged MR head writes data the written data is slightly curved on the written track. When the straight across MR sensor reads this curved data there is progressive signal loss from the center of the data track toward the outer extremities of the data track.
All merged magnetic heads have a separation between the read and write gaps. This separation causes misregistration between the read and write gaps when the magnetic head is located at outer tracks on the magnetic disk. In the magnetic disk drive, an actuator swings the magnetic head across the rotating disk to various circular tracks on the disk. At the innermost track the read and write gaps are substantially aligned with one another and there is substantially no misregistration. At the innermost track the read gap follows within the track written by the write gap. However, when the actuator swings the magnetic head to the outermost track the read and write gaps are misaligned with respect to the track. If the write gap is within the track being written the read gap may be partially in the track and partially in an adjacent track. The misregistration increases with an increase in the separation between the read and write gaps. In magnetic heads where the write head is constructed before the read head the profile of the insulation stack of the write head raises the height of the first shield layer of the read head. It would be desirable if this profile could be reduced so that the read and write gaps are closer together.
Still another problem with prior art magnetic heads is that heating of high magnetic moment pole tips risks damage to the read sensor of the underlying read head. A high magnetic material is Ni45Fe55 as compared to Ni80Fe20. Pole tips constructed of high magnetic material are desirable because they will conduct higher flux density without saturating. Because the read head is placed under the write head while sharing the second shield/first pole tip elements (merged design), the read head is exposed to intense flux variations during the operation of the pole tips. This can result in damage and ultimate failure of the read sensor.
Heat also causes problems during use of the head. All of the coils are found in the write head. Modern disks are made with higher coercivity. To write to disks with high coercivity, more moment must come from the head, so more current must pass through the coils. The increased current generates more heat, which causes the copper in the coils to expand, causing the materials in front of them to protrude. Further, the coils are typically sandwiched between photoresist, which has less than ½ the expansion coefficient of Cu, resulting in an uneven air bearing surface (ABS). The ferromagnetic element also protrudes when heated. To compensate for protrusion, more flying gap is needed between the disk and head, resulting in lower resolution. This in turn lowers the maximum disk density, as the head needs to be very close to the disk for high areal density. Thus it is desirable to decrease heat to reduce expansion of the head layers. To provide the required read current with lower heat, adding additional turns to the coils has been attempted, but this increases the size of the head. While more turns can carry more current, this increases reluctance, so more magnetic moment is needed because there are more losses. Making the copper coils bigger with less separation between the coils decreases parasitic resistance and thus produces less heat, but at some point the coil becomes difficult to fabricate because after electroplating the copper, an ion milling process or sputtering process must be used to remove the Cu seed layer that was used to nucleate the copper during the electroplating operation. If the coils are too close together, the ions can't penetrate the gaps to clean up the seed layer, which ultimately has the effect of shorting the coil.
Furthermore, for efficiency, inductive heads are now being designed with very short distance between the ABS and the closure back gap. Patents on this concept include U.S. Pat. No. 6,259,583 which uses a pedestal-like structure to bring the coils up to 4 μm from the ABS. However, by having the coils so close to the ABS and the pitch of the coils very small, the heat dissipation over such small area creates thermal protrusion causing unwanted larger magnetic separation between the disk and head.
What is needed is a way to reduce the heat in the write head (i.e., improve heat dissipation) in order to keep the protrusion under control while avoiding the aforementioned problems.
Yet another problem is that during fabrication, the write head is formed last. As touched on above, the write head contains the most difficult layers to fabricate, thereby increasing the chance of error. For instance, the write head has many layers of thick dissimilar material, some hard and some soft. Each of these requires special processing (heating, milling, etc.) are prone to errors and/or can damage existing layers. Other difficulties arise when constructing the read portion which include bad imaging, bad composition when plating, smudges, lint, residue from prior processes, scratching, etc. If these difficulties extend to the write head, an error could occur during fabrication of the write head, so the entire wafer may have to be discarded, causing loss of precious time as well as substantial sums of money.
What is needed is a way to fabricate the most difficult layers first, thereby minimizing loss in case of error. This means fabricating the write head first.
Prior art U.S. Pat. No. 6,130,809 discloses a longitudinal write head placed under a read head. Because the head is designed for longitudinal recording, inherent with that concept is the need to produce very thick pole tips with large aspect ratio demands (>20:1). It also precludes the use of vacuum deposited films for the poles because it would require patterning using ion milling incapable of producing such aspect ratios.
A further problem encountered in the prior art, and particularly in longitudinal writing, is fringing fields. To increase bit density, tracks must be positioned very closely together. Similarly, the bit to bit density must be high. The close proximity of tracks and bits increases the risk of fringe writing. As areal density increases, the magnetic moment of the disk increases. To abate the paramagnetic limit of the media, more magnetic flux must be emitted from the pole tip because it also needs much thinner write gaps to increase the on-track bit density. This further increases the size of the fringing fields.
What is therefore needed is a perpendicular recording head, and process for manufacturing the same, that overcomes the problems inherent in the prior art.
The present invention solves the problems described above by providing standard and thin film magnetic head structures for perpendicular recording and reading. One head structure includes a write head portion for writing data to magnetic media via lines of flux oriented substantially perpendicular to a surface of the media facing the write head portion. The write head portion includes a first pole piece having a first pole tip, a probe pole piece with a probe pole tip for emitting magnetic flux, an insulation stack positioned between the pole pieces, one or more write coils embedded in the insulation stack, and a shaping layer positioned between the second pole piece and the insulation stack for focusing flux to the probe pole tip. A read head portion that has been fabricated after the write head portion is coupled to the write head portion. In one embodiment, the top pole piece functions as a shield layer for the read portion. In another embodiment, the read head portion includes first shield layer separated from the top pole layer, and a sensor positioned therebetween.
The write head portion is preferably positioned between the read head portion and a slider. This aids in conducting heat from the write head portion by allowing the slider to function as a heat sink. Also preferably, the probe pole piece is constructed at least in part of a deposited (vacuum deposited, plated, etc.) film that has been ion-milled. A preferred aspect ratio of the probe pole tip is less than 5:1.
A method for constructing a magnetic head for perpendicular recording and reading includes fabricating a probe pole piece having a probe pole tip for writing data to magnetic media via lines of flux oriented substantially perpendicular to a surface of magnetic media facing the write head portion. An insulation stack and write coil are fabricated above the probe pole piece, the write coil being embedded in the insulation stack. A second pole piece is fabricated above the insulation stack. Finally, a read head portion is fabricated above the second pole piece. Preferably, a shaping layer for focusing flux to the probe pole tip is formed between the probe pole tip piece and the insulation stack.
A magnetic storage system according to one embodiment includes magnetic media, a head for reading from and writing to the magnetic media via lines of flux oriented substantially perpendicular to a surface of the media facing the head, a slider for supporting the head, and a control unit coupled to the head for controlling operation of the head. The magnetic media preferably has a magnetically hard first layer positioned towards the head, and a magnetically soft second layer positioned behind the first layer.
For a fuller understanding of the nature and advantages of the present invention, as well as the preferred mode of use, reference should be made to the following detailed description read in conjunction with the accompanying drawings.
The following description is the best embodiment presently contemplated for carrying out the present invention. This description is made for the purpose of illustrating the general principles of the present invention and is not meant to limit the inventive concepts claimed herein.
Magnetic Disk Drive
Referring now to the drawings wherein like reference numerals designate like or similar parts throughout the several views there is illustrated in
Prior Art Merged MR Head
The prior art write head portion of the merged MR head includes a coil layer 84 located between first and second insulation layers 86 and 88. A third insulation layer 90 may be employed for planarizing the head to eliminate ripples in the second insulation layer caused by the coil layer 84. The first, second and third insulation layers are referred to in the art as an “insulation stack”. The coil layer 84 and the first, second and third insulation layers 86, 88 and 90 are located between first and second pole piece layers 92 and 94. The first and second pole piece layers 92 and 94 are magnetically coupled at a back gap 96 and have first and second pole tips 98 and 100 which are separated by a write gap layer 102 at the ABS. As shown in
As shown in
The location of the flare point 130, shown in
Another important design parameter in making the write head is the location of a zero throat height (ZTH), which is where the first and second pole piece layers 92 and 94 first separate from one another behind the ABS. It is important to locate the ZTH as close as possible to the ABS (typically within about 1 μm) in order to reduce flux loss between the pole pieces before the fields reach the gap layer 102 at the ABS. In the prior art, locating the ZTH close to the ABS contributed to the aforementioned problem of fabricating a well-defined second pole tip 100.
The photoresist pattern for the second pole piece is shown in
When the second pole piece 94 is plated and the photoresist layer 141 is removed the head is complete, as shown in
It can be seen from
Another problem with the prior art head in
As mentioned above, prior art fabrication of DASD thin film heads have positioned the write heads above the read heads, except for U.S. Pat. No. 6,130,809 where the longitudinal write head is placed under the read head. Because the head in the patent is for longitudinal recording, inherent with that concept is the need to produce very thick pole tips with large aspect ratio demands (>20:1). It also precludes the use of vacuum deposited films for the poles because it would require patterning using Ion Milling incapable of producing such aspect ratios.
In
The Invention and Method of Construction
Two embodiments of storage systems with perpendicular heads 190 are illustrated in
By this structure the magnetic lines of flux extending between the poles of the recording head loop into and out of the outer surface of the recording medium coating with the high permeability under layer of the recording medium causing the lines of flux to pass through the coating in a direction generally perpendicular to the surface of the medium to record information in the magnetically hard coating of the medium in the form of magnetic impulses having their axes of magnetization substantially perpendicular to the surface of the medium. The flux is channeled by the soft underlying coating 192 back to the return layer (P1) of the head 190.
As shown in
As depicted in
Once the shaping layer 212 is made, gaps can be filled by a conventional process and the wafer is again planarized, preferably by using alumina and CMP.
The disclosed perpendicular design also brings together very efficient head performance, better heat dissipation, and very thin writer pole needs. Fewer coils are needed, as the shaping layer focuses flux into the probe pole. Thus, because the embodiments of the present invention use perpendicular recording resulting in a basically fringeless process, tighter tracks are possible.
As mentioned above, for efficiency, inductive heads are now being designed with a very short distance between the ABS and the closure back gap. By having the coils so close to the ABS and the pitch of the coils very small, the heat dissipation over such small area creates thermal protrusion causing unwanted larger magnetic separation between the disk and head. By placing the write heads first (and consequently near the heat-sinking slider), heat dissipation is improved, thereby helping to reduce the protrusion problem. Further, while the head in U.S. Pat. No. 6,259,583 uses a pedestal like structure to bring the coils up to 4 μm from the ABS, no pedestal is necessary in preferred embodiments of the present invention. In conventional heads, the P1 and P2 layers are wider than the corresponding layers in the disclosed embodiments, so side writing occurs. Fringing fields from conventional heads are up to 40% wider than those of perpendicular recording heads. As described above, the recording head does not require a pedestal according to preferred embodiments. In perpendicular recording, the flux jumps into the disk which acts as a pedestal to return flux back to the larger P1 layer. By removing the pedestal the flux is focused, and by having the head use the disk as a return layer, high resolution is guaranteed because the problems of conventional heads (fringing fields, heat, protrusion, etc.) are almost completely avoided.
Though the read head is now more separated from the heat sinking slider, the combined metallic components underneath act as an excellent heat sinking element. Again by fabricating the write head first, none of its fabrication demands for high temperature process are felt by the read head. The fact that the write head's more demanding layer (the probe poles) are much thicker than the K5 or K3 layers of the read heads it means that alignment targets will be of better resolution.
While various embodiments have been described above, it should be understood that they have been presented by way of example only, and not limitation. Thus, the breadth and scope of a preferred embodiment should not be limited by any of the above-described exemplary embodiments, but should be defined only in accordance with the following claims and their equivalents.
Number | Name | Date | Kind |
---|---|---|---|
4277809 | Fisher et al. | Jul 1981 | A |
5073836 | Gill et al. | Dec 1991 | A |
5455730 | Dovek et al. | Oct 1995 | A |
5486967 | Tanaka et al. | Jan 1996 | A |
5668689 | Schultz et al. | Sep 1997 | A |
5691867 | Onuma et al. | Nov 1997 | A |
5710683 | Sundaram | Jan 1998 | A |
5822153 | Lairson et al. | Oct 1998 | A |
5901432 | Armstrong et al. | May 1999 | A |
6130809 | Santini | Oct 2000 | A |
6154346 | Sasaki | Nov 2000 | A |
6172848 | Santini | Jan 2001 | B1 |
6212047 | Payne et al. | Apr 2001 | B1 |
6259583 | Fontana, Jr. et al. | Jul 2001 | B1 |
6339523 | Santini | Jan 2002 | B1 |
6697221 | Sato et al. | Feb 2004 | B2 |
6754947 | Stageberg et al. | Jun 2004 | B2 |
6903900 | Sato et al. | Jun 2005 | B2 |
6943988 | Ueyanagi | Sep 2005 | B2 |
6954340 | Shukh et al. | Oct 2005 | B2 |
20020176214 | Shukh et al. | Nov 2002 | A1 |
20030137779 | Santini et al. | Jul 2003 | A1 |
20040196589 | Lille | Oct 2004 | A1 |
Number | Date | Country |
---|---|---|
60-177420 | Sep 1985 | JP |
1155510 | Jun 1989 | JP |
04-216311 | Aug 1992 | JP |
4286702 | Oct 1992 | JP |
06-295418 | Oct 1994 | JP |
2001-101612 | Sep 1999 | JP |
2003-16608 | Jun 2001 | JP |
Number | Date | Country | |
---|---|---|---|
20040223257 A1 | Nov 2004 | US |