Abstract:
The series resistance of a CPP GMR stack can be reduced by shaping it into a small upper, on a somewhat larger, lower part. Because of the sub-micron dimensions involved, good alignment between these is normally difficult to achieve. The present invention discloses a self-alignment process based on first laying down a mask that will determine the shape of the top part. Ion beam etching is then initiated, the ion beam being initially applied from one side only at an angle to the surface normal. During etching, all material on the near side of the mask gets etched but, on the far side, only material that is outside the mask's shadow gets removed so, depending on the beam's angle, the size of the lower part is controlled and the upper part is precisely centrally aligned above it.
Abstract:
A current-perpendicular-to-plane (CPP) giant magnetoresistive (GMR) sensor of the synthetic spin valve type is provided, the sensor comprising a GMR stack having a substantially square lateral cross-section, a Cu spacer layer of smaller square cross-section formed centrally on the GMR stack and a capped ferromagnetic free layer of substantially square, but even smaller cross-sectional area, formed centrally on the spacer layer. The stepped, reduced area geometry of the sensor provides a significant improvement in its GMR ratio (DR/R), a reduced resistance, R, and elimination of Joule heating hot-spots in regions of high resistance such as the antiferromagnetic pinning layer and its seed layer.
Abstract:
Using a beam of xenon ions together with a suitable mask, a MTJ stack is ion milled until a part of it, no more than about 0.1 microns thick, has been removed so that a pedestal, having sidewalls comprising a vertical section that includes all of the free layer, has been formed. This is followed by formation of the longitudinal bias and conductive lead layers in the usual way. Using xenon as the sputtering gas enables the point at which milling is terminated to be more precisely controlled.
Abstract:
Single write poles tend to large shape anisotropy which results in a very large remnant field when not actually writing. This has now been eliminated by giving the write pole the form of a three layer laminate in which two ferromagnetic layers are separated by a non-magnetic or antiferromagnetic coupling layer. Strong magnetostatic coupling between the outer layers causes their magnetization directions to automatically be antiparallel to one another, unless overcome by the more powerful write field, leaving the structure with a low net magnetic moment. The thickness of the middle layer must be carefully controlled.
Abstract:
The series resistance of a CPP GMR stack can be reduced by shaping it into a small upper, on a somewhat larger, lower part. Because of the sub-micron dimensions involved, good alignment between these is normally difficult to achieve. The present invention discloses a self-alignment process based on first laying down a mask that will determine the shape of the top part. Ion beam etching is then initiated, the ion beam being initially applied from one side only at an angle to the surface normal. During etching, all material on the near side of the mask gets etched but, on the far side, only material that is outside the mask's shadow gets removed so, depending on the beam's angle, the size of the lower part is controlled and the upper part is precisely centrally aligned above it.
Abstract:
As the recording density of magnetic disk drives approaches 100 Gbits/in2, write track lengths of about 0.10 microns will be required. This cannot be accomplished using conventional photolithography. The present invention solves this problem by first forming on the bottom pole of the write head a cavity in a layer of photoresist, using conventional means. A seed layer of non-magnetic material is electrolessly laid down, following which a second layer of photoresist is deposited and patterned to form a second cavity that symmetrically surrounds the first one, thereby forming a mold around it. Ferromagnetic metal is then electro-deposited in this mold to form the top magnetic pole. Following the removal of all photoresist and a brief selective etch of the bottom pole, an extremely narrow write head is obtained.
Abstract:
Using a beam of xenon ions together with a suitable mask, a stack is ion milled until a part of it, no more than about 0.1 microns thick, has been removed so that a pedestal having sidewalls, including a vertical section and a shortened taper portion, has been formed. This is followed by formation of conductive lead layers as needed. Using xenon as the sputtering gas enables the point at which milling is terminated to be more precisely controlled.
Abstract:
Using a beam of xenon ions together with a suitable mask, a stack is ion milled until a part of it, no more than about 0.1 microns thick, has been removed so that a pedestal having sidewalls, including a vertical section and a shortened taper portion, has been formed. This is followed by formation of conductive lead layers as needed. Using xenon as the sputtering gas enables the point at which milling is terminated to be more precisely controlled.
Abstract:
Using a beam of xenon ions together with a suitable mask, a MTJ stack is ion milled until a part of it, no more than about 0.1 microns thick, has been removed so that a pedestal, having sidewalls comprising a vertical section that includes all of the free layer, has been formed. This is followed by formation of the longitudinal bias and conductive lead layers in the usual way. Using xenon as the sputtering gas enables the point at which milling is terminated to be more precisely controlled.
Abstract:
A GMR read pillar formed by ion milling has both vertical and sloping sidewall sections with the free layer normally being located within the latter. Using xenon as the sputtering gas enables the vertical section of the pedestal to be made longer, relative to the sloping portion, without requiring an increase in the sputtering rate, so the free layer can have vertical sidewalls. This allows the point at which milling is terminated to be controlled more precisely and, by more precisely defining the width of the free layer, also improves design tolerances.