Electricity: measuring and testing – Magnetic – Stress in material measurement
Reexamination Certificate
2002-07-15
2003-12-16
Strecker, Gerard R. (Department: 2862)
Electricity: measuring and testing
Magnetic
Stress in material measurement
C324S228000, C324S232000
Reexamination Certificate
active
06664783
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to systems for measuring magnetostriction in thin magnetic films and more particularly to such a system employing multiple harmonics of an external rotating magnetic field to improve measurement accuracy.
2. Description of the Related Art
Areal data storage densities are increasing at an unprecedented rate owing to advances in the development of magnetic medium and read head materials. Exploiting increasing storage density requires increased signal output from the read sensor. For this reason, thin film inductive read heads were eventually replaced by read heads using the anisotropic magnetoresistance (AMR) effect. Subsequently, a class of metallic multilayer films exhibiting the giant magnetoresistive (GMR) effect were introduced as read heads for hard-disk (HD) and tape magnetic storage devices. GMR effect read sensors exhibit an MR ratio, &Dgr;R/R, that is typically 8% or higher compared with perhaps 2% for earlier AMR sensors. The increased sensitivity allows acceptable signal levels from smaller read element track widths, thereby increasing track density and areal storage density.
The exchange biased spin-valve (SV) is a multilayer GMR device that is most useful for read head applications. A SV consists of several layers: a free and pinned layer both made of a soft ferromagnetic (FM) material such as Permalloy (NiFe) or a cobalt (Co) alloy. These two FM layers are separated by a non-magnetic conductive spacer layer such as copper (Cu). An exchange layer of antiferromagnetic (AFM) material, commonly made of a manganese (Mn) alloy such as NiMn, is deposited next to the FM pinned layer. The FM free layer is thin enough to allow conduction electrons to frequently move back and forth between the free and pinned layers via the conducting spacer layer. The magnetic moment of the FM pinned layer is fixed and held in place by the AFM layer, while the FM free layer magnetic moment changes in response to the external magnetic field, such as that from a bit stored on a hard disk.
The quantum spin property of electrons, i.e., either spin up or spin down, is exploited in SV sensors. Conduction electrons with spin parallel to the FM material's magnetization (spin “up”) move freely, while the motion of those electrons with anti-parallel orientation (spin “down”) is impeded via collisions with atoms in the material. When the FM free and pinned layer magnetic moments are parallel, spin up electrons move freely in both FM layers, corresponding to a relatively low effective resistance. Conversely, when the free and pinned layer magnetic moments are anti-parallel, movement of spin up electrons is hampered by one layer, while the movement of spin down electrons is hampered by the other, so that neither move freely through both FM layers, leading to a relatively high effective resistance. In the GMR sensor, the external field from a recorded bit rotates the FM free layer magnetic moment relative to that of the FM pinned layer, effectively switching the SV device between the high and low resistance states.
Much of the cost in manufacturing read head sensors is incurred in processing the individual sensors after the actual material deposition process. So a means for qualifying the post deposition product is critical to reduced cost (improved yield). In fabricating GMR SV devices, a variety of in-process metrology is required to provide requisite material deposition process control. In-process measurement of magnetic and magnetoresistive properties is one method by which the film deposition process is qualified and controlled. One exemplary concern is the uniformity and thickness of the deposited layers (the copper spacer layer is typically less than 15 atoms thick). Another is surface roughness, which affects the coupling between layers, the coercivity of the FM free layer, and the effectiveness of the AFM layer in pinning the FM pinned layers. Other concerns include the thin-film properties that significantly affect device performance, such as magnetoresistance, resistivity and magnetostriction. Many of these parameters interact with one another, so fabricating acceptable read sensors requires strict quality control tolerances and processes.
The saturation magnetostriction in the magnetically soft GMR films may induce undesirable magnetic anisotropy changes during head fabrication and therefore must be tightly controlled. Anisotropy is controlled during wafer production using a complex combination of process parameters during magnetic film deposition. Eventually, the wafers are diced and the critical aerodynamic surfaces are polished. Both of these mechanical processes create unpredictable changes in the stress level between the substrate and the magnetic films. The magnetostriction of the film material translates these unpredictable stress changes into changes in the anisotropy of the magnetic layer. Also, an inverse relationship between applied magnetostrictive stress and the MR sensitivity ratio, &Dgr;R/R, has been noted. The preferred way to control anisotropy after deposition is to keep magnetostriction below 10
−7
. Accordingly, the accurate measurement and analysis of thin-film magnetostriction is critical to effective process control during manufacture of modern GMR SV sensors.
Determination of the magnetostriction coefficient of ferromagnetic materials by inference from the measurement of the deflection in an external magnetic field H
ext
of a substrate element on which a layer of the material under test has been applied has been known in the art for several decades. As understood in the 1960's and 1970's, this procedure suffered from numerous well-known disadvantages, including the difficulties of accurate characterization of substrate material parameters, accurate detection of microscopic deflections, heating effects from the requisite external field intensity, unsuitability of the requisite procedures to automation, and the like.
A major improvement in this technique was disclosed in the commonly-assigned U.S. Pat. No. 4,310,798 issued to Brunsch et al. and entirely incorporated herein by this reference. The Brunsch et al. invention optically measures the dynamic displacement of the free end of a cantilever substrate element on which is applied a layer as thin as 5 nm of the FM material under test, in the presence of an external rotating magnetic field. The substrate element deflection arising from magnetostriction of the test material in the external rotating magnetic field occurs at twice the rotation frequency f so this rotation frequency f is tuned to half of the mechanical resonant frequency f
0
of the cantilever element to permit accurate detection of the deflection amplitude, which is otherwise too small. Because Brunsch et al. assume the test material to have a linear magnetization characteristic, leading to a quadratic dependence of magnetostriction &lgr; on external field intensity H
ext
, their procedure looks for a maximum resonant deflection RMS amplitude A
0
and presumes it to be proportional to the saturated magnetostriction &lgr;
S
. The exact proportionality is a function of substrate material characteristics. The Brunsch et al. procedure resolved several problems known in the art and, for the first time, permitted automated magnetostriction testing, but did not eliminate the need for accurate characterization of the substrate material parameters.
Later, Cheng et al. (Cheng et al., “Device to Measure Magnetostriction of Thin Film on a Substrate”,
IBM Technical Disclosure Bulletin
, July 1988, pp. 59-60) propose an improvement of the Brunsch et al. procedure that adds a lock-in amplifier to the deflection amplitude sensor that permits operation well below mechanical resonant frequency f
0
, thereby distinguishing the magnetostrictive deflection A
2
at 2f from many other sources of resonant deflection at f
0
. The lock-in amplifier provided enhanced sensitivity needed to accurately detect the microscopic deflection amplitudes below mechanical resonance.
The
Baril Lydia
Mackay Kenneth Donald
Johnston Ervin F.
Strecker Gerard R.
LandOfFree
System for measuring magnetostriction employing a plurality... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System for measuring magnetostriction employing a plurality..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System for measuring magnetostriction employing a plurality... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3170069