Patent Grant
9034662
The invention relates to the general field of magneto-resistance (MR) devices with particular reference to improving their performance.
A typical example of a spintronic device is a magneto-resistance sensor. As shown schematically in
As recording densities increase, device sizes become correspondingly smaller. Consequently the sensor (or other spintronic device) has to continue having a high enough magneto-resistance (MR) ratio, as well as a low enough resistance x area product (RA), for it to still have sufficient output amplitude and also for it to be suitable for future extendibility. It is, however, becoming increasingly more difficult for MgO-based TMR devices to maintain their current high MR ratio in the lower RA region. Current Perpendicular to the Plane (CPP) GMR or Confined Current Path (CCP) GMR devices have not yet been able to produce high enough MR ratios for them to take over as the future read head technologies. There is thus an urgent need for sensors that have higher output amplitude and/or a higher MR ratio.
The present invention, as will be detailed below, discloses some novel methods for further enhancing the MR ratio.
A routine search of the prior art was performed with the following references of interest being found:
U.S. Pat. No. 7,596,015 (Kitigawa et al) shows a free layer including an intermediate layer of Cu and/or Zn or Sn. (See col. 5, lines 58-63 free layer is the same as recording layer 12, and see col. 11, lines 1-15 for materials added to the recording layer—at least one of the materials including Cu, Zn, and Sn).
U.S. Pat. No. 7,223,485 (Yuasa et al) teaches an insertion layer in the free layer. (See col. 2-3 where the insert layer contains at least one element from Cu, Zn, O). K. Zhang et al, HT06-055 “A novel CPP device with enhanced MR ratio” and the reference therein. Y. Chen et al, “Spin-dependent CPP transport properties of ZnO/ferromagnet Heterostructures” Physics Letters A 303 (2002) 91-96.
It has been an object of at least one embodiment of the present invention to provide a method for enhancing the performance, particularly the MR and R.A ratios, of magneto-resistance devices.
Another object of at least one embodiment of the present invention has been to provide a process for manufacturing the invention.
A further object of at least one embodiment of the present invention has been that said process be fully compatible with existing processes for manufacturing spintronic devices.
Still another object of at least one embodiment of the present invention has been to provide a detailed description thereof.
These objects have been achieved by inserting one or more Magneto-Resistance Enhancing Layers (MRELs) into approximately the center of one or more of the active layers (such as the magnetic AP1, spin induction layer or SIL, and free layers). An MREL is a layer of a low band gap, high electron mobility semiconductor such as ZnO or a semimetal such as Bi. Optionally, the interface(s) between the MREL and the magnetic layer into which it has been inserted may be bridged by a thin layer of a highly conductive metal such as copper so as to ensure an ohmic contact across the full interface between the MREL and the magnetic layer.
We have found that when a trilayer of a low band gap semiconductor such as ZnO, sandwiched between layers of a conductive metal such as Cu (to form Cu3/ZnO15/Cu3 for example), is inserted into a spintronic device the MR ratio of that device is increased significantly. An early example of this was discovered in the course of experiments with a spin torque oscillator (STO), patterned to around 45×45 nm, whose structure was the following: Ta10/Ru20/Cu20/[Co2/Ni6]x15/Cu20/FeCo150/Ru10/Ta40/Ru30. This device had almost no MR ratio for the in-plane RH measurement. However, when the applied field was close to the perpendicular direction (specifically, H was tilted 7 degrees from the perpendicular direction), the observed MR ratio was typically about 1.5% with an RA of about 0.06. A typical RH curve is shown in
After insertion of the above-specified trilayer (MREL) into the ferromagnetic (FM) FeCo layer, the resulting full structure became Ta10/Ru20/Cu20/[Co2/Ni6]x15/Cu20/FeCo75/Cu3/ZnO15/Cu3/FeCo75/Ru10/Ta40/Ru30). The MR ratio of the MREL modified structure was found to have increased significantly (from about 1.3% to as high as 17%) with little change in the RA product. A typical curve is shown in
In
Second, an in-plane measurement showed almost no MR ratio, which is very similar to what was observed for the reference case. These facts indicate that the high MR ratio cannot be coming from the magnetic switching of the [FeCo75/Cu3/ZnO15/Cu3/FeCo75] stack of layers, but rather from the [Co2/Ni6]x15/Cu20/[FeCo75/Cu3/ZnO15/Cu3/FeCo75] switching across the Cu spacer. Thus, the [FeCo75/Cu3/ZnO15/Cu3/FeCo75] stack behaves as though it is a single layer (like the original FeCo150 layer that it replaced).
These findings confirm that the Cu3/ZnO15/Cu3 trilayer acts as an MR enhancing layer. It is thought that the spin scattering coefficient of the [FeCo75/Cu3/ZnO/Cu3/FeCo75] stack is significantly enhanced by the insertion of the Cu3/ZnO15/Cu3 trilayer. This is in accordance with the Schmitt theory, which shows that when spin is injected from a FM metal (FeCo) into a semiconductor (e.g. ZnO), the spin coefficient is proportional to the ratio of the conductivity of the semiconductor to that of the FM metal.
It is also possible that the Cu3/ZnO/Cu3 MREL acts as a spin filtering layer, possibly even one operating under a ballistic scattering regime. Regardless of the precise mechanism at work here, the MR enhancement that results from inserting the Cu/ZnO/Cu (or similar layer) can be realized in a wide variety of applications that we will detail below.
Before describing several applications for these MRELs we note the following. In the general case, the MREL's structure is M1/low band gap, high mobility, N-type semi-conductor/M2, where
(a) M1 and M2 are high conductivity metals such as (but not limited to) Cu, Ag, Au, C (including Graphene, nano-tubes etc), Zn, Ti, Sn, Cr, Al, Mg, and Ru. M1 and M2 will generally be the same material and will have the same thickness (ranging from 0 to 50 Å), with about 3 Å Cu for both M1 and M2 being preferred. However, M1 and M2 may be different materials and they may have different thicknesses without significantly affecting the effectiveness of the invention.
(b) Additionally, the basic M1/semiconductor/M2 trilayer structure that makes up the MREL may be repeated in situ one or more times, each such additional trilayer not necessarily comprising the same materials and/or individual thicknesses as other trilayers in the same MREL.
(c) The band gap of the semiconductor portion of the MREL should be in a range of from 0.3 to 8 eV, with from 1 to 6 eV being preferred.
(d) The electron mobility of the semiconductor portion of the MREL should be in a range of from 10 to 2,000,000_cm2·sec−1·V−1 with from 50 to 50,000 cm2·sec−1·V−1 being preferred.
Examples of suitable semi-conductor materials include (but are not limited to) ZnO, ZnS, ZnxMg(1-x)O (x ranging from 0 to 0.99), ZnCuO, ZnCdO, ZnAlO, ZnSe, ZnTe, Si, Ge, TiO2, AlN, GaN, InN, AlP, AlAs, AlSb, GaP, GaAs, GaSb, InP, InAs, ZnS, CdS, CdTe, HgTe, PbS, PbSe, PbTe, SnO, SnTe, Cu2O, FeSi2, CrMnSi, Mg2Si, RuSi3, and Ir3Si5.
The semiconductor portion may be undoped or its conductivity may be adjusted by the inclusion of a dopant selected from the group consisting of Si, B, Mg, Mn, Al, Cu, Cd, Cr, Zn, Ti, Sn, Zr, Hf, Ru, Mo, Nb, Co, Fe, and Ni. Moreover, the semiconductor portion may be replaced by a semi-metal such as Sb, Bi, CoSi, CoxFe(1-x)Si, CoxNi(1-x)Si, CoxMn(1-x)Si, FeSi, or CoxCr(1-x)Si. The semiconductor/semimetal layer should have a thickness in a range of from 1 to 50 Å, with from 3 to 20 Å being preferred.
We now describe several examples of how these MR enhancing layers may be used to improve the performance of existing MR devices. These include (but are not limited to) GMR devices (including CIP, CPP, and CPP/CCP versions), TMR devices, spin injection devices, magnetic random access memories (MRAMs) and microwave assisted magnetic recording (MAMR) devices (see item 4 below).
It will be understood that those skilled in the art may be able to identify additional spintronic devices not listed above (such as, for example, sensors of all types (including biosensors), thermally assisted magneto-resistance (TAMR) devices, or dual spin valve devices to which the principles disclosed by the present invention would obviously be applicable.
1. GMR Devices:
A prior art structure is seed/AFM/AP2/Ru/AP1/conductive spacer/FL/cap. Possible structures based on the invention (see
The spacer layer could be a uniform metallic layer (CPP devices) or it could be in the form of conductive islands embedded in an insulating matrix such as, for example, Cu in Al2O3 (CPP/CCP devices).
2. TMR Devices:
A prior art structure is seed/AFM/AP2/Ru/AP1/barrier layer/FL/cap. Possible structures based on the invention (see
3. Spin Injection Layer (SIL) Devices that Operate without an AFM/AP2/Ru/AP1 Pinning Stack:
A prior art structure is seed/FM/spacer/FL/cap (including an H field normal to the layers). In other words, a field (H) is applied in a direction normal (perpendicular) to the layers after the SIL stack of layers is completed in order to magnetize the FM layer in a direction perpendicular to the top and bottom surfaces of the FM layer. Possible structures based on the invention (see
A prior art structure is seed/SIL/spacer/FGL/cap (including an H field normal to the layers). Possible structures based on the invention (see
For the examples listed in the above four categories, the seed layer could be Ta/Ru, Ta/Cu, Ta/NiFe, Ta/NiCr, Ta/NiFeCr, Ta/Ti, Ta/Ti/Cu, Ta/Ti/Ru/Cu, etc, the AFM layer (when present) could be IrMn, PtMn, PtPdMn etc, and AP2 (where present) could be CoFe, CoFe/FeTa/CoFe, or CoFe/CoFeB/CoFe, for example.
This is a Divisional application of U.S. patent application Ser. No. 12/799,468, filed on Apr. 26, 2010, which is herein incorporated by reference in its entirety, and assigned to a common assignee.
| Number | Name | Date | Kind |
|---|---|---|---|
| 5627704 | Lederman et al. | May 1997 | A |
| 5668688 | Dykes et al. | Sep 1997 | A |
| 5715121 | Sakakima et al. | Feb 1998 | A |
| 6876523 | Takahashi et al. | Apr 2005 | B2 |
| 6917088 | Takahashi et al. | Jul 2005 | B2 |
| 7116529 | Yoshikawa et al. | Oct 2006 | B2 |
| 7223485 | Yuasa et al. | May 2007 | B2 |
| 7381480 | Nakamura et al. | Jun 2008 | B2 |
| 7479394 | Horng et al. | Jan 2009 | B2 |
| 7525772 | Koui et al. | Apr 2009 | B2 |
| 7596015 | Kitagawa et al. | Sep 2009 | B2 |
| 8609262 | Horng et al. | Dec 2013 | B2 |
| 20020048128 | Kamiguchi et al. | Apr 2002 | A1 |
| 20070014054 | Zhang et al. | Jan 2007 | A1 |
| 20070070556 | Zhang et al. | Mar 2007 | A1 |
| 20070297103 | Zhang et al. | Dec 2007 | A1 |
| 20080180991 | Wang | Jul 2008 | A1 |
| 20080278864 | Zhang et al. | Nov 2008 | A1 |
| 20090108383 | Horng et al. | Apr 2009 | A1 |
| 20110014500 | Horng et al. | Jan 2011 | A1 |
| 20140099735 | Horng et al. | Apr 2014 | A1 |
| Entry |
|---|
| “Spin-dependent CPP transport properties of ZnO/ferromagnetic heterostructures,” by Yanxue Chen et al., Physics Letters A 303, (Oct. 7, 2002) 91-96, NH Elsevier. |
| “MgO-Based Tunnel Junction Material for High-Speed Toggle Magnetic Random Access Memory,” by Renu W. Dave et al., IEEE Transactions on Magnetics, vol. 42, No. 8, Aug. 2006, pp. 1935-1939. |
| “The scalability of CPP-GMR heads toward over 100Gbpsi, compared with TMR heads,” by M. Takagishi et al., Toshiba TMRC 2001, Aug. 20, 2001, 10 pgs. |
| Japanese Office Action 2011-098800 Mailed: Aug. 19, 2014, Headway Technologies Inc. |
| Number | Date | Country | |
|---|---|---|---|
| 20140220708 A1 | Aug 2014 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 12799468 | Apr 2010 | US |
| Child | 14244923 | US |
