Method to fabricate a magnetic head including ion milling of read gap using dual layer hard mask

Magnetic hard disk drives retrieve and/or store data in computers and other consumer electronics devices. A magnetic hard disk drive includes one or more heads that can read and write information on a corresponding magnetic surface of a spinning disk. For convenience, all heads that can read are referred to as “read heads” or “heads” herein, regardless of other devices and functions the read head may also perform (e.g. writing, micro-actuation, flying height control, touch down detection, lapping control, localized disk media heating, etc). Each read head is a sub-component of a head gimbal assembly (HGA). The HGA also includes a suspension assembly for holding the head and providing a plurality of electrical connections thereto. The suspension assembly typically includes a fragile laminated flexure to carry the electrical signals to and from the head.

The head typically comprises a slider that includes an air bearing surface (ABS) that faces the magnetic disk surface and a trailing face. A magnetic transducer and a plurality of head bond pads are typically disposed on the trailing face of the slider. The mounting face of the slider is typically permanently bonded to a tongue portion of the fragile laminated flexure by an adhesive, in a position such that the plurality of head bond pads are aligned with corresponding bond pads on the laminated flexure.

Conventionally, the magnetic transducer on the trailing face of the slider can write tiny magnetic transitions on the magnetic disk surface by applying sufficient magnetic field to a desired microscopic disk surface location, to overcome the coercivity of the disk surface material there and thereby change the remnant field there. The magnetic transducer also includes a read sensor that must be capable of sensing magnetic transitions written along very narrow concentric tracks around the disk surface. For example, the width of the read sensor may be required to be less than 50 nm in modern magnetic recording applications. As the areal density of data written on magnetic disk media further increases, ever-narrower read sensors are required.

Conventional methods to fabricate read sensors of narrow width may suffer from inadequate cleaning of photo mask residue and/or undesirable metal re-deposition or corrosion during certain manufacturing steps, leading to unreliability in ultimately producing the desired read sensor widths. Accordingly, there is a need in the art for improved methods to reliably fabricate very narrow read transducers (e.g. with read widths of 30 nm or less) for magnetic heads, practical for high volume manufacturing with acceptable yield and cost.

FIG. 1 is an exploded perspective view of a disk drive capable of including read heads fabricated according to an embodiment of the present invention.

FIG. 2 is a top perspective view of a head stack assembly (HSA), capable of including read heads fabricated according to an embodiment of the present invention.

FIG. 3 depicts a read head capable of including a read sensor fabricated according to an embodiment of the present invention, as viewed from a perspective angle towards the air bearing surface of the head.

FIG. 4 depicts method steps according to an embodiment of the present invention.

FIG. 5 depicts method steps according to an embodiment of the present invention.

FIG. 1 is an exploded perspective view of a disk drive 1 including read heads 40 that may be fabricated according to an embodiment of the present invention. The disk drive 1 includes a head disk assembly (HDA) 10 and a printed circuit board (PCB) 14 . The HDA 10 includes a disk drive base 16 and a disk drive cover 18 , that together enclose other components of the HDA 10 . At least one disk 20 is rotatably mounted to the disk drive base 16 by a spindle motor 26 . The HDA 10 optionally includes one or more additional disks 20 , also rotatably mounted to the disk drive base 16 by spindle motor 26 , with the disks 20 then separated by one or more spacer rings 12 .

The disks 20 may comprise an aluminum, glass, or ceramic substrate, with the substrate optionally being coated with a NiP under-layer, at least one thin-film magnetic layer, a diamond-like amorphous carbon protective layer, and a very thin lubricant layer, for example. The disks 20 are annular in shape, having an outer periphery 24 and an inner periphery 22 , and may be clamped to a rotating hub of the spindle motor 26 , by a clamp 11 . The rotating hub of the spindle motor 26 rotates the disks 20 about a disk axis of rotation 28 .

The HDA 10 also includes a head stack assembly (HSA) 30 that includes an actuator 32 that is pivotably mounted to the disk drive base 16 by a pivot bearing 44 that is inserted as a cartridge into a bore in the actuator 32 . The pivot bearing 44 allows the actuator 32 to pivot about an actuator pivot axis 46 . The actuator 32 may be fabricated from any suitable material having sufficient stiffness to weight, for example aluminum, magnesium, beryllium, or stainless steel. The pivot bearing 44 may be retained in the bore by a tolerance ring or may be otherwise retained (e.g. by an adhesive). The angular range of motion of the actuator 32 may be limited by a latch and crash stop mechanism 52 . The actuator 32 includes at least one actuator arm 36 that extends away from the pivot bearing 44 , and an actuator coil 50 that extends away from the pivot bearing 44 in a direction generally opposite the actuator arm 36 .

The actuator coil 50 may fit in a yoke structure that optionally comprises a top plate 56 and a bottom plate 58 , to form a voice coil motor (VCM). One or both of the top plate 56 and the bottom plate 58 may support a permanent magnet(s) of the VCM (e.g. permanent magnet 54 ). The top plate 56 and/or the bottom plate 58 preferably comprise a ferromagnetic metal so as to provide a return path for magnetic flux from the permanent magnet(s) through the yoke structure. The ferromagnetic metal yoke structure including the top plate 56 and the bottom plate 58 is preferably affixed to the disk drive base 16 , for example by an adhesive, one or more fasteners, and/or magnetic attraction.

A plurality of head gimbal assemblies 42 may be attached to arms 36 of the actuator 32 , for example by swaging. A flexible printed circuit (FPC) 62 (also known as a “flex cable”) may also be attached to the actuator 32 . Each of the head gimbal assemblies 42 may include a read head 40 that is bonded to a laminated flexure having a flexure tail that is electrically connected to the FPC 62 . The FPC 62 can communicate electrical signals, via these electrical connections and via a flex bracket 64 , between the head gimbal assemblies 42 and an external electronic system that is disposed on the PCB 14 .

The actuator 32 , head gimbal assemblies 42 , and FPC 62 all may be considered as components of the HSA 30 . The HDA 10 may have additional components that are not shown in FIG. 1 . For example, the HDA 10 may also include a conventional head loading ramp (not shown), that may be positioned adjacent the disks 20 to facilitate merging of the head gimbal assemblies 42 onto surfaces of the disks 20 .

FIG. 2 is a top perspective view of a head stack assembly (HSA) 200 , including read heads 292 that may be being fabricated according to an embodiment of the present invention. The HSA 200 includes an actuator 232 pivotably attachable to a disk drive base by a pivot bearing 244 that is inserted as a cartridge into a bore in the actuator 232 . The actuator 232 may be fabricated from a suitable metal such as aluminum, magnesium, beryllium, or stainless steel, or a relatively low mass density non-metallic or composite material having sufficient stiffness. The pivot bearing 244 may be retained in the bore by a tolerance ring or may be otherwise retained (e.g. by an adhesive). The actuator 232 includes a plurality of actuator arms 236 that extends away from the pivot bearing 244 , and an actuator coil 250 that extends away from the pivot bearing 244 in a direction generally opposite the actuator arms 236 .

In the embodiment of FIG. 2 , a plurality of head gimbal assemblies (HGAs) 242 are shown to be attached to arms 236 of the actuator 232 . Each HGA includes a read head 292 and a suspension assembly to which the read head 292 is attached. Specifically, each read head 292 is attached and electrically connected to a laminated flexure 282 of the suspension assembly of the HGA 242 . The suspension assembly of each HGA 242 is shown to be attached to one of the actuator arms 236 , for example by swaging. A flexible printed circuit (FPC) 262 is also shown to be attached to the actuator 232 . Each laminated flexure 282 includes a plurality of conductive traces, and each laminated flexure 282 has a flexure tail that extends to the FPC 262 , where the plurality of conductive traces is connected to the FPC 262 near a pre-amplifier chip 266 .

FIG. 3 depicts a read head 304 , as viewed from a perspective angle towards its air bearing surface 324 . The read head 304 comprises a slider 326 that includes the air bearing surface 324 , and a trailing face 328 that is normal to the air bearing surface 324 . The air bearing surface 324 may be of any conventional design, and is not restricted to the design shown in FIG. 3 . The head 304 also comprises a plurality of head bond pads 352 that are disposed on the trailing face 328 and that are electrically conductive. The head bond pads 352 may comprise gold and/or copper, for example.

In the embodiment of FIG. 3 , a magnetic transducer 350 is disposed on the trailing face of the slider 326 . The magnetic transducer 350 may include a magneto-resistive read sensor (e.g. a tunneling magneto-resistive read sensor) fabricated according to an embodiment of the present invention, and a magneto-inductive write transducer, for example. The magnetic transducer 350 may also include other functions and structures, such as a micro-actuator, flying height control heater, a touch down sensor, a lapping guide, and/or a laser guide or mirror for heating a localized region of the disk media. The slider 326 may comprise a ceramic material such as AlTiC, for example.

FIG. 4 and FIG. 5 depict steps of a method to manufacture a read sensor for a magnetic head, according to an embodiment of the present invention. Now referring to FIG. 4 , a first layered structure 400 is provided, for example by depositing a plurality of layers on to a conventional substrate. In the embodiment of FIG. 4 , the first layered structure 400 includes a magnetic layer 402 , that itself may be a tunneling magnetoresistive laminate comprising a plurality of conventional sublayers.

In the embodiment of FIG. 4 , a first hard mask layer 404 is shown to be disposed immediately adjacent and contacting the magnetic layer 402 . The first layered structure 400 also includes a second hard mask layer 406 immediately adjacent and contacting the first hard mask layer 404 . The first layered structure 400 may also include a first stepping layer 408 immediately adjacent and contacting the second hard mask layer 406 , and a second stepping layer 410 immediately adjacent and contacting the first stepping layer 408 .

In the embodiment of FIG. 4 , the first stepping layer 408 may optionally comprise amorphous carbon. The first stepping layer 408 is shown in FIG. 4 to have a thickness 428 that, in certain embodiments, may preferably be in the range of 20 nm to 200 nm. In the embodiment of FIG. 4 , the second stepping layer 410 is shown to have a thickness 430 that, in certain embodiments may preferably be in the range of 1 nm to 10 nm. In certain embodiments, the second stepping layer 410 may preferably comprise a material that is not easily removed by an oxygen plasma etching process, such as tantalum, silicon oxide, chromium, silicon nitride, or ruthenium.

In the embodiment of FIG. 4 (prior to step A→B), a patterned photoresist layer 412 is shown to be immediately adjacent and contacting the second stepping layer 410 , although in embodiments having only the first stepping layer 408 and not the optional second stepping layer 410 , the patterned photoresist layer 412 may be immediately adjacent and contacting the first stepping layer 408 .

In step A→B of FIG. 4 , the first and second stepping layers 408 and 410 are shown to have been etched through to create a sidewall 418 of the first and second stepping layers 408 , 410 . For example, the second stepping layer 410 may be ion milled or etched away where exposed (i.e. where not protected from the ion milling by the patterned photoresist layer 412 ). Also for example, the first stepping layer 408 may be etched to create the sidewall 418 by reactive ion etching, for example using oxygen plasma. The patterned photoresist layer 412 is removed after etching through the second stepping layer 410 during the process of creating the sidewall 418 , and therefore the patterned photoresist layer 412 is absent after step A→B of FIG. 4 . In step A→B of FIG. 4 , the total height of the sidewall 418 is shown to equal the sum of the thickness 428 of the first stepping layer 408 plus the thickness 430 of the second stepping layer 410 .

In step B→C of FIG. 4 , a mask-width definition layer 414 is deposited on and adjacent to the sidewall 418 , until a mask-width definition layer thickness 434 is achieved adjacent to the sidewall 418 . In certain embodiments, the mask-width definition layer thickness 434 is preferably less than 30 nm, for example in the range of 3 nm to 30 nm. In certain embodiments, the mask-width definition layer 414 optionally may comprise alumina, silicon dioxide, or silicon nitride. As will be described herein, the thickness 434 of the mask-width definition layer can be determinative of the reader width according to certain embodiments of the inventive method.

In step C→D of FIG. 4 , the mask-width definition layer 414 is removed except on the sidewall 418 , for example by ion etching (e.g. ion milling or reactive ion etching). Optionally, and as shown in step C→D of FIG. 4 , the mask-width definition layer 414 removal step may also remove the second stepping layer 410 . If not, the second stepping layer 410 may be removed as a separate step either before or during the removal of the first stepping layer 408 , which may leave a narrow portion of the mask-width definition layer 414 as shown before step A→B of FIG. 5 . In certain embodiments, oxygen plasma is used to remove the first stepping layer 408 , for example if the first stepping layer 408 comprises amorphous carbon.

As shown in FIG. 4 , the first hard mask layer 404 has a thickness 424 that, in certain embodiments, may preferably be in the range of 10 nm to 50 nm. In certain embodiments, the first hard mask layer 404 may preferably be a material removable by etching with an oxygen plasma, such as amorphous carbon (e.g. diamond-like carbon), or a cured photoresist material (e.g. baked above the cross-link temperature to lose photo-sensitivity). The second hard mask layer 406 has a thickness 426 that, in certain embodiments, may preferably be in the range of 5 nm to 50 nm. In certain embodiments, the second hard mask layer 406 may preferably comprise silicon carbide. Silicon dioxide or silicon nitride may also be used for the second hard mask layer 406 , so long as the material is different from that of the mask width definition layer 414 .

In the step A→B of FIG. 5 , the second hard mask layer 406 is etched away (for example using reactive ion etching), except for a remainder of the second hard mask layer 406 that was beneath the narrow portion of the mask-width definition layer 414 . The narrow portion of the mask-width definition layer 414 is also removed, for example by wet etching.

In the step B→C of FIG. 5 , a patterned partial over-layer of photoresist material 512 optionally may be deposited on the first hard mask layer 404 , so that only exposed regions 503 and 505 of the first hard mask layer 404 are exposed, the exposed regions 503 and 505 being separated from each other by the narrow remainder of the second hard mask layer 406 .

In the step C→D of FIG. 5 , the exposed regions 503 and 505 of the first hard mask layer 404 are etched away around the remainder of the second hard mask layer 406 , for example using oxygen plasma. Such etching defines a remainder 504 of the first hard mask layer 404 beneath the remainder of the second hard mask layer 406 . A dual layer hard mask is thereby formed, comprising the remainder of the second hard mask layer 406 and the remainder 504 of the first hard mask layer 404 .

In the step D→E of FIG. 5 , the magnetic layer 402 is ion milled around the dual hard mask 406 , 504 to define a reader stripe 502 . In certain embodiments, ion milling the magnetic layer 402 round the dual hard mask 406 , 504 also erodes away the remainder of the second hard mask layer 406 , with less undesirable metal re-deposition than prior methods. Hard bias or soft bias material 542 and 544 may then be optionally back-filled into the voids left by the ion milling of the magnetic layer 402 .

In the step E→F of FIG. 5 , the patterned partial over-layer of photoresist material 512 may be removed by a conventional photoresist material removal process, and the first hard mask layer 404 (including the remainder 504 ) may also be removed, for example using oxygen plasma. In certain embodiments, such removal of the first hard mask layer 404 by oxygen plasma may obviate the need for a cleaning step (to otherwise remove residue from the patterned photoresist material over-layer 512 ). In certain embodiments, the ability to remove the first hard mask layer 404 by oxygen plasma may reduces corrosion of the magnetic layer 402 and the reader stripe 502 (relative to the corrosion that might be induced by the use of a fluorine plasma). The reader stripe 502 is shown in FIG. 5 to have a reader width w, which one of ordinary skill will recognize depends substantially upon the mask-width definition layer thickness 434 shown in FIG. 4 .

In the foregoing specification, the invention is described with reference to specific exemplary embodiments, but those skilled in the art will recognize that the invention is not limited to those. It is contemplated that various features and aspects of the invention may be used individually or jointly and possibly in a different environment or application. The specification and drawings are, accordingly, to be regarded as illustrative and exemplary rather than restrictive. For example, the word “preferably,” and the phrase “preferably but not necessarily,” are used synonymously herein to consistently include the meaning of “not necessarily” or optionally. “Comprising,” “including,” and “having,” are intended to be open-ended terms.