Dynamic magnetic information storage or retrieval – Head – Magnetoresistive reproducing head
Reexamination Certificate
1999-04-15
2001-05-01
Klimowicz, William (Department: 2754)
Dynamic magnetic information storage or retrieval
Head
Magnetoresistive reproducing head
Reexamination Certificate
active
06226160
ABSTRACT:
TECHNICAL FIELD
This invention relates to magnetic tunnel junction (MTJ) magnetoresistive (MR) devices for use as read heads for reading magnetically-recorded data and as memory cells in nonvolatile magnetic random access memory (MRAM) arrays.
BACKGROUND OF THE INVENTION
A magnetic tunnel junction (MTJ) device is comprised of at least two ferromagnetic layers separated by a thin insulating tunnel barrier layer and is based on the phenomenon of spin-polarized electron tunneling. The insulating tunnel barrier layer is thin enough that quantum mechanical tunneling occurs between the ferromagnetic layers. The tunneling phenomenon is electron-spin dependent, making the magnetic response of the MTJ a function of the relative orientations and spin polarizations of the two ferromagnetic layers. MTJ devices have been proposed as memory cells for nonvolatile solid state memory and as external magnetic field sensors, such as MR read sensors or heads for magnetic recording systems. For a memory cell application one of the ferromagnetic layers in the MTJ has its magnetic moment fixed or pinned so as to be parallel or antiparallel to the magnetic moment of the other free or sensing ferromagnetic layer in the absence of an applied magnetic field within the cell. For a MR field sensor or read head application one of the ferromagnetic layers has its magnetic moment fixed or pinned so as to be generally perpendicular to the magnetic moment of the free or sensing ferromagnetic layer in the absence of an external magnetic field. The use of an MTJ device as a memory cell in an MRAM array is described in IBM's U.S. Pat. No. 5,640,343. The use of an MTJ device as a MR read head has been described in U.S. Pat. No. 5,390,061 and in IBM's U.S. Pat. Nos. 5,650,958; 5,729,410 and 5,764,567.
What is important for MTJ device applications is the signal-to-noise ratio (SNR). The magnitude of the signal is dependent upon the magnetoresistance or MR (&Dgr;R/R) exhibited by the device. The signal is given by i
B
&Dgr;R, which is the bias current (i
B
) passing through the MTJ device (assuming a constant current is used to detect the signal) times the resistance change (&Dgr;R) of the device. However, the noise exhibited by the MTJ device is determined, in large part, by the resistance R of the device. It is well known that any conductor exhibits thermal noise commonly referred to as Johnson-Nyquist noise. The magnitude of the Johnson-Nyquist noise can be expressed as the root mean square voltage across the resistor, V
J
=(4k
B
TR&Dgr;f)
1/2
. Thus if the noise introduced by the Johnson-Nyquist noise is the dominant noise source (i.e., the measurement electronics do not introduce any significant noise compared to the Johnson-Nyquist noise) then the SNR will be given by i
B
2
&Dgr;R
2
/(4k
B
TR&Dgr;f), where k
B
is Boltzmann's constant, T is the absolute temperature, R is the resistance of the device, and &Dgr;f is the bandwidth of the measurement electronics. Thus to obtain the maximum SNR for constant power used to sense the device the resistance (R) of the device must be small and the change in resistance (&Dgr;R) of the device large.
The resistance of a MTJ device is largely determined by the resistance of the insulating tunnel barrier layer for a device of given dimensions since the resistance of the metal layers in the MTJ device, for example, the electrical leads and the ferromagnetic layers, contribute little to the resistance. Moreover, because the sense current passes perpendicularly through the ferromagnetic layers and the tunnel barrier layer, the resistance of a MTJ device increases inversely with the area of the device. This is in contrast to conventional MR devices, such as those based on the anisotropic magnetoresistance (AMR) effect and the giant magnetoresistance (GMR) effect, where the sense current passes parallel to the layers. Prior art MTJ devices have resistance values (~10
4
-10
9
&OHgr;) that are several orders of magnitude higher than the resistance values of conventional AMR or GMR devices (~10-40 &OHgr;) of the same size, which means that the noise they exhibit is also much higher than conventional AMR or GMR devices.
The requirement for low resistance MTJ devices, coupled with the inverse relationship of resistance with area, is especially troublesome because an additional requirement for MTJ device applications is small area. For MRAM applications the density of the array depends on small area MTJ cells. For read head applications, high storage density on the media requires that the trackwidth (TW) be small (the area of the MTJ device is given by h×TW, where h is the height of the MTJ).
Since the resistance R of a MTJ device scales inversely with the area A, it is convenient to characterize the resistance of the MTJ device by the specific resistance, i.e., the product of the resistance R times the area A. Thus the specific resistance, R
S
, is independent of the area A of the MTJ device.
The specific resistance R
S
of prior art MTJ devices has ranged from more than 10
9
&OHgr;(&mgr;m)
2
down to ~10
3
&OHgr;(&mgr;m)
2
. For example, in 1992 J. Nowak et al. (
J. Magn. Mat.
109, 79-90 (1992)) reported specific resistance values of ~10
8
-10
9
&OHgr;(&mgr;m)
2
in Fe/GdO
x
/Fe junctions, although these junctions exhibited very small magnetoresistance (&Dgr;R/R) values of ~0.7%. In 1995 T. Miyazaki et al. reported magnetoresistance values of up to 18% at 300 K in Fe/Al
2
O
3
/Fe planar tunnel junctions for which the specific resistance was ~6×10
3
&OHgr;(&mgr;m)
2
. However, these junctions were formed with very large areas, A~1×1 mm
2
, which suggests that the measured magnetoresistance values were not intrinsic values but were enhanced by non-uniform current flow through the large-area junctions, as originally discovered in work on Josephson junctions (R. J. Pedersen et al.,
Applied Physics Letters,
10, 29 (1967)). In 1996 J. S. Moodera, et al (
Phys. Rev. Lett.
74, 3273 (1996)) observed magnetoresistance values of ~10% at room temperature in planar tunnel junctions of area A=0.3×0.3 mm
2
but the specific resistance of these junctions was very high at ~10
9
&OHgr;(&mgr;m)
2
. In 1996 S. S. P. Parkin et al. (
J. Appl. Phys.
81, 5521 (1997) reported magnetoresistance values of up to 25% at room temperature in planar exchange biased magnetic tunnel junctions with specific resistance values of ~10
6
&OHgr;(&mgr;m)
2
for junctions with areas of ~0.1×0.1 mm
2
. Also in 1996 W. J. Gallagher et al. (
J. Appl. Phys.
81, 3741 (1997)) reported magnetoresistance values of up to 20% at room temperature in junctions with specific resistance values of ~5×10
4
&OHgr;(&mgr;m)
2
for junctions ranging in size from approximately 100 &mgr;m×100 &mgr;m (10,000 &mgr;m
2
) down to junctions with sub-micron dimensions.
The prior art that describes magnetic tunnel junctions with relatively low specific resistance values also suggests that these devices have other undesirable properties, such as unacceptably low magnetoresistance, and are made with complicated processes. For example, in 1997 H. Tsuge et al. (
Appl. Phys. Lett.
71, 3296 (1997)) reported Fe/Al
2
O
3
/CoFe/ junctions in which the tunnel barrier layer was formed by first depositing Al layers 2 nm thick using electron beam evaporation and then exposing these layers to oxygen. These junctions exhibited R
S
values of ~1.5×10
3
&OHgr;(&mgr;m)
2
but very low magnetoresistance values of ~5% at room temperature. Only one Al thickness was reported and no studies of the dependence of the junction resistance on deposited Al thickness were reported and thus no method for controlling or varying the barrier thickness was indicated. The authors suggested that the use of an in-situ deposited barrier aluminum layer and oxidation of this layer in the vacuum chamber without breaking vacuum are the critical steps to obtaining a low resistance tunnel barrier irrespective of the thickness of the deposited Al layer. P. K. Wong et al. (
J. Appl. Phys.
83, 6697 (1998)) re
Gallagher William Joseph
Papworth Parkin Stuart Stephen
Berthold Thomas R.
International Business Machines - Corporation
Klimowicz William
LandOfFree
Small area magnetic tunnel junction devices with low... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Small area magnetic tunnel junction devices with low..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Small area magnetic tunnel junction devices with low... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2458000