JP2018032459A - Magnetoresistance effect element and magnetic sensor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a magnetoresistance effect element with high SN ratio, and a magnetic sensor.SOLUTION: A magnetoresistance effect element 1 comprises a channel layer 7, a first ferromagnetic layer 12A, a second ferromagnetic layer 12B, and a reference electrode 20. The first ferromagnetic layer 12A, the second ferromagnetic layer 12B, and the reference electrode 20 are separate from each other and electrically connected to each other via the channel layer 7. The channel layer 7 includes: a first area A1 overlapping with the first ferromagnetic layer 12A when viewed from a depthwise direction; a second area A2 overlapping with the second ferromagnetic layer 12B when viewed from the depthwise direction; and a third area A3 between the first area A1 and the second area A2. At least part of the reference electrode 20 is in contact with an area including the first area, the second area A2 and the third area A3.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a magnetoresistive effect element and a magnetic sensor.

  A magnetoresistive element is known as a reproducing element used in a thin film magnetic recording / reproducing head or the like. Since a general magnetoresistive effect element passes a current between the magnetization fixed layer, the magnetization free layer, and these layers, a high output can be obtained.

  On the other hand, a spin accumulation type magnetoresistive effect element is known in which a magnetization free layer and a magnetization fixed layer are formed on the same horizontal plane (on a channel layer for accumulating spins). The spin accumulation type magnetoresistive effect element is expected to obtain a high spatial resolution when used in a magnetic sensor such as a thin film magnetic recording / reproducing head. In addition, an advantage that the degree of freedom in device design can be improved is also expected. Patent Document 1 discloses a four-terminal spin accumulation type magnetoresistive element having two ferromagnetic electrodes and two nonmagnetic electrodes. Patent Document 2 discloses a three-terminal spin accumulation type magnetoresistive effect element having two ferromagnetic electrodes and one nonmagnetic electrode.

  Patent Document 2 describes a magnetoresistive effect element 1x as shown in FIG. The magnetoresistive effect element 1x has a channel layer 7x on a support substrate 30x through a base insulating layer 80x, and has a first ferromagnetic layer 12Ax on the channel layer 7x through a barrier layer 14Ax. The second ferromagnetic layer 12Bx is provided via 14Bx. The channel layer 7x includes a first region A1x that overlaps the first ferromagnetic layer 12Ax when viewed from the thickness direction of the channel layer 7x, and a second region A2x that overlaps when viewed from the thickness direction of the second ferromagnetic layer 12Bx and the channel layer 7x. have. Further, the channel layer 7x has an extending portion AEx as shown in FIG. In the magnetoresistive effect element 1x, the extending portion AEx includes a reference electrode 20x (nonmagnetic electrode) provided apart from the second region A2x and a region overlapping with the channel layer 7x as viewed in the thickness direction, This is a region composed of a region between two regions A2x.

JP 2010-287666 A International Publication No. 2015/076187

  However, in order to make a spin accumulation type magnetoresistive element practical, the signal-to-noise ratio (S / N ratio) is still small, and further improvement of the S / N ratio is necessary. An object of this invention is to provide the magnetoresistive effect element and magnetic sensor which have a high S / N ratio.

  In order to solve the above problem, a magnetoresistive element of the present invention includes a channel layer, a first ferromagnetic layer, a second ferromagnetic layer, and a reference electrode, and the first ferromagnetic layer, The second ferromagnetic layer and the reference electrode are separated from each other and electrically connected to each other through the channel layer, and the first ferromagnetic layer and the second ferromagnetic layer are The channel layers are separated from each other without overlapping with each other when viewed from the thickness direction of the channel layer, and the channel layer includes a first region overlapping with the first ferromagnetic layer when viewed from the thickness direction, the second ferromagnetic layer, and the thickness. A second region that overlaps when viewed from the direction, and a third region between the first region and the second region, wherein at least part of the reference electrode includes the first region, the second region, and the second region It is characterized in that it is in contact with a region composed of the third region.

  For the magnetoresistive element described above, by causing spin-polarized carriers to flow from the first ferromagnetic layer to the second ferromagnetic layer through the channel layer between the first ferromagnetic layer and the second ferromagnetic layer, Spin is accumulated in the second region of the channel layer, and a voltage (output signal) is detected between the second ferromagnetic layer and the reference electrode via the channel layer, so that the magnetization direction of the first ferromagnetic layer and the second A resistance change corresponding to a change in the relative angle of the magnetization direction of the ferromagnetic layer is detected. According to the magnetoresistive effect element described above, the channel layer extending portion (channel layer) can be formed as compared with the case where the region composed of the first region, the second region, and the third region and the reference electrode are arranged apart from each other. The volume of the entire portion excluding the portion where spin-polarized carriers flow can be reduced. As a result, diffusion of spin accumulated in the second region of the channel layer from the second region to the extending portion can be suppressed, and spins can be accumulated in the second region with high density. Therefore, the output signal of the magnetoresistive effect element of the present invention can be increased, and a high S / N ratio can be realized.

  Furthermore, the magnetoresistive effect element of the present invention is characterized in that at least a part of the reference electrode is in contact with the second region.

  According to this, as compared with the case where at least a part of the reference electrode is in contact with the region including the first region, the second region, and the third region, the second region and the reference electrode are spaced apart from each other. When the contact area of the layer and the reference electrode is the same, the volume of the region where the part where the spin-polarized carriers flow (spin transport path) and the part where the voltage is detected (spin detection path) overlaps in the channel layer Can be reduced. As a result, the output noise of the magnetoresistive element of the present invention can be reduced.

  Furthermore, the magnetoresistive element of the present invention is characterized in that the first region is not in contact with the reference electrode.

  According to this, compared with the case where the first region and the reference electrode are in contact with each other, there are a portion where the spin-polarized carriers flow in the channel layer (spin transport path) and a portion where the voltage is detected (spin detection path). The volume of the overlapping region can be reduced. As a result, the output noise of the magnetoresistive element of the present invention can be reduced.

  Furthermore, the magnetoresistive element of the present invention is characterized in that the third region is not in contact with the reference electrode.

  According to this, as compared with the case where the third region and the reference electrode are in contact with each other, the portion where the spin-polarized carriers flow in the channel layer (spin transport path) and the portion where the voltage is detected (spin detection path) are formed. The volume of the overlapping region can be reduced. As a result, the output noise of the magnetoresistive element of the present invention can be reduced.

  In the magnetoresistive element of the invention, the thermal conductivity of the reference electrode is smaller than the thermal conductivity of the channel layer.

  According to this, the heat radiation effect by the reference electrode of Joule heat generated in the spin transport path of the channel layer can be suppressed, and the temperature gradient inside the channel layer can be reduced. Since the output noise decreases as the temperature gradient inside the channel layer decreases, the SN ratio of the magnetoresistive element of the present invention can be further increased.

  Furthermore, the magnetic sensor of the present invention is characterized by having the magnetoresistive element described above.

  According to this, a magnetic sensor having a high S / N ratio can be realized.

  ADVANTAGE OF THE INVENTION According to this invention, the magnetoresistive effect element and magnetic sensor which have high SN ratio can be provided.

FIG. 1 is a perspective view of the magnetoresistive effect element according to the first embodiment. FIG. 2 is a perspective view of a modification of the magnetoresistive effect element according to the first embodiment. FIG. 3 is a perspective view of the magnetoresistive effect element according to the second embodiment. FIG. 4 is a perspective view of a modification of the magnetoresistive effect element according to the second embodiment. FIG. 5 is a perspective view of the magnetoresistive effect element according to the third embodiment. FIG. 6 is a perspective view of a modification of the magnetoresistive effect element according to the third embodiment. FIG. 7 is a cross-sectional view of a main part of a magnetic sensor according to the fourth embodiment. FIG. 8 is a side view seen from the magnetic medium facing surface (X direction) of FIG. FIG. 9 is a perspective view of a magnetoresistive effect element according to a conventional example.

Preferred embodiments for carrying out the present invention will be described in detail with reference to the drawings. The present invention is not limited by the contents described in the following embodiments. The constituent elements described below include those that can be easily assumed by those skilled in the art, those that are substantially the same, and those that are equivalent. Furthermore, the constituent elements described below can be appropriately combined. In addition, various omissions, substitutions, or changes of components can be made without departing from the scope of the present invention.

(First embodiment)
Hereinafter, the magnetoresistive effect element 1 according to the first embodiment will be described with reference to the drawings.

(Basic structure)
FIG. 1 is a perspective view of a magnetoresistive element 1 according to the first embodiment. As shown in FIG. 1, the magnetoresistive effect element 1 includes a channel layer 7, a first ferromagnetic layer 12 </ b> A, a second ferromagnetic layer 12 </ b> B, and a reference electrode 20. The channel layer 7 and the reference electrode 20 are provided on the support substrate 30 through the base insulating layer 80. The first ferromagnetic layer 12A, the second ferromagnetic layer 12B, and the reference electrode 20 are separated from each other and are electrically connected to each other through the channel layer 7. The first ferromagnetic layer 12 </ b> A and the second ferromagnetic layer 12 </ b> B are separated from each other without overlapping each other when viewed from the thickness direction of the channel layer 7. In the channel layer 7, the direction in which the first ferromagnetic layer 12 </ b> A and the second ferromagnetic layer 12 </ b> B are aligned (the direction in the film surface) is the long side direction, and the direction perpendicular to the long side direction (the direction in the film surface) ) Has a rectangular plan view shape that is a short side direction, and has a rectangular parallelepiped shape.

  The channel layer 7 has a first surface and a second surface that face each other in the thickness direction of the channel layer 7. In the magnetoresistive effect element 1, the first ferromagnetic layer 12A and the second ferromagnetic layer 12B are provided on the first surface side of the channel layer 7 (in the example shown in FIG. 1, the upper surface side of the channel layer 7). It has been.

  Between the first ferromagnetic layer 12A and the second ferromagnetic layer 12B, spin-polarized carriers (electrons or holes) flow through the channel layer 7 from the first ferromagnetic layer 12A to the second ferromagnetic layer 12B. The second ferromagnetic layer 12 </ b> B and the reference electrode 20 are portions where a voltage is detected via the channel layer 7.

  The channel layer 7 includes a first region A1 that overlaps the first ferromagnetic layer 12A when viewed from the thickness direction of the channel layer 7, a second region A2 that overlaps when viewed from the thickness direction of the second ferromagnetic layer 12B and the channel layer 7, and A third area A3 is provided between the first area A1 and the second area A2.

  The reference electrode 20 is provided so that at least a part thereof is in contact with a region composed of the first region A1, the second region A2, and the third region A3. In the example shown in FIG. 1, at least a part of the reference electrode 20 is provided in contact with the second region A2. In the magnetoresistive effect element 1, the reference electrode 20 is provided on the second surface side of the channel layer 7 (in the example shown in FIG. 1, on the lower surface side of the channel layer 7), and is in contact with the lower surface of the second region A2. Is arranged. That is, the reference electrode 20 is arranged at a position overlapping the second ferromagnetic layer 12B and the channel layer 7 when viewed from the thickness direction. In the magnetoresistive effect element 1, the reference electrode 20 is in contact with the channel layer 7 only on the entire lower surface of the second region A2, and is not in contact with the first region A1 and the third region A3.

  In the magnetoresistive element 1, the first ferromagnetic layer 12A, the first region A1, the third region A3, the second region A2, and the second ferromagnetic layer 12B are regions corresponding to the spin transport path (spin-polarized carriers flow). And the second ferromagnetic layer 12B, the second region A2, and the reference electrode 20 are regions corresponding to the spin detection path (portions where voltage is detected).

A barrier layer 14 A is provided between the first ferromagnetic layer 12 A and the channel layer 7, and a barrier layer 14 B is provided between the second ferromagnetic layer 12 B and the channel layer 7.

(Support substrate and base insulating layer)
Examples of the material of the support substrate 30 include AlTiC or Si. Examples of the material of the base insulating layer 80 provided on the support substrate 30 include SiO _x (silicon oxide), HfO _x (hafnium oxide), SiN _x (silicon nitride), and the like. The thermal conductivity of the base insulating layer 80 is smaller than the thermal conductivity of the channel layer 7.

(Ferromagnetic layer)
As a material of the first ferromagnetic layer 12A and the second ferromagnetic layer 12B, for example, a metal selected from the group consisting of Co, Fe and Ni, a metal of the group consisting of Cr, Mn, Co, Fe and Ni are 1 An alloy containing at least seeds, or one or more metals selected from the group consisting of Cr, Mn, Co, Fe and Ni, and a group consisting of B, C, N, Al, Si, Ga and Ge Examples include alloys containing one or more elements, and specific examples include CoFeB or NiFe.

(Barrier layer)
Since the barrier layer 14A is provided between the first ferromagnetic layer 12A and the channel layer 7, many spin-polarized carriers (electrons or holes) are injected from the first ferromagnetic layer 12A to the channel layer 7. Thus, the output signal of the magnetoresistive effect element 1 can be increased. Since the barrier layer 14B is provided between the second ferromagnetic layer 12B and the channel layer 7, the spin polarization rate of the spin-polarized carriers between the second ferromagnetic layer 12B and the channel layer 7 can be reduced. The decrease is suppressed, and efficient spin transport from the first ferromagnetic layer 12A to the second ferromagnetic layer 12B becomes possible.

The barrier layers 14A and 14B are preferably tunnel barrier layers. As a material of the tunnel barrier layer, for example, magnesium oxide, aluminum oxide, titanium oxide, spinel oxide film, zinc oxide, or the like can be used. From the viewpoint of suppressing an increase in resistance and functioning as a tunnel barrier layer, the thickness of the tunnel barrier layer is preferably 3 nm or less. The thickness of the tunnel barrier layer is preferably 0.4 nm or more in consideration of the thickness of one atomic layer.

(Channel layer)
The channel layer 7 is a layer in which spin is transported and accumulated. The material of the channel layer 7 is preferably a nonmagnetic material and a material having a long spin diffusion length. In the magnetoresistive effect element 1, the material of the channel layer 7 is a nonmagnetic semiconductor. In this case, the material (base material) of the channel layer 7 can be a semiconductor including any one of Si, Ge, GaAs, and C. The material of the channel layer 7 can also be a nonmagnetic metal such as Cu, Ag, or Al. In the magnetoresistive effect element 1, the material of the channel layer 7 is the same in the first region A1, the second region A2, and the third region A3.

When a nonmagnetic semiconductor is used as the material of the channel layer 7, it is preferable to add a small amount of impurities (impurity doping) for imparting conductivity to the semiconductor material as a base material. Examples of the impurity doping method include an ion implantation method and a thermal diffusion method. For example, when a group 14 (group IV) element such as Si, Ge, or C is used as a base semiconductor material, a group 15 (group V) element such as N, P, As, or Sb is doped as an impurity. In this case, since electrons become majority carriers, the electron concentration can be adjusted as the carrier concentration. In this case, the channel layer 7 is an n-type semiconductor. In addition, if a Group 13 (Group III) element such as B, Al, or Ga is doped as an impurity, holes become majority carriers, so that the hole concentration can be adjusted as the carrier concentration. In this case, the channel layer 7 is a p-type semiconductor. The impurity concentration of the channel layer 7 is preferably 1.0 × 10 ^{16 to} 1.0 × 10 ²¹ cm ⁻³ .

  When the material of the channel layer 7 is an n-type semiconductor, electrons become spin-polarized carriers. In this case, a current flowing from the second ferromagnetic layer 12B to the first ferromagnetic layer 12A via the channel layer 7 is applied from the first ferromagnetic layer 12A and the second ferromagnetic layer 12B. In this case, electrons whose spin polarization is caused by the first ferromagnetic layer 12A are injected from the first ferromagnetic layer 12A, and the inside of the channel layer 7 is directed from the first ferromagnetic layer 12A to the second ferromagnetic layer 12B. Be transported to. As a result, a spin non-equilibrium state (spin accumulation) is generated inside the channel layer 7 (second region A2).

  When the material of the channel layer 7 is a p-type semiconductor, holes become spin-polarized carriers. In this case, a current flowing from the first ferromagnetic layer 12A to the second ferromagnetic layer 12B via the channel layer 7 is applied from the first ferromagnetic layer 12A and the second ferromagnetic layer 12B. In this case, holes whose spin polarization is caused by the first ferromagnetic layer 12A are injected from the first ferromagnetic layer 12A, and the inside of the channel layer 7 is directed from the first ferromagnetic layer 12A to the second ferromagnetic layer 12B. Be transported to. As a result, a spin non-equilibrium state (spin accumulation) is generated inside the channel layer 7 (second region A2).

(Reference electrode)
As shown in FIG. 1, in the magnetoresistive effect element 1, the reference electrode 20 is provided in the concave portion of the base insulating layer 80 and is in contact with the lower surface of the second region A <b> 2 of the channel layer 7.

The material of the reference electrode 20 is preferably a nonmagnetic material that is a conductor, particularly a nonmagnetic metal. Further, the thermal conductivity of the reference electrode 20 is preferably smaller than the thermal conductivity of the channel layer 7. Table 1 shows examples of main nonmagnetic materials that can be used for the reference electrode 20 or the channel layer 7 and their thermal conductivity. For example, when Si is used as the material of the channel layer 7, preferable materials used for the reference electrode 20 include Ti, Zr, V, Ta, Cr, Nb, and the like. When Cu is used as the material of the channel layer 7, preferable materials used for the reference electrode 20 include Al, W, Mo, Ti, Zr, V, Ta, Cr, or Nb. Note that, as a material of the reference electrode 20, an alloy or nitride containing these elements as main components may be used.

In the magnetoresistive effect element 1, a current flows between the first ferromagnetic layer 12A and the second ferromagnetic layer 12B in order to allow spin-polarized carriers to flow in a region corresponding to the spin transport path. While a current flows between the first ferromagnetic layer 12A and the second ferromagnetic layer 12B, an amount of heat (Joule heat) in the spin transport path is proportional to the square of the current value and the resistance value of the material in the spin transport path. However, when the material of the channel layer 7 is the same in the first region A1, the second region A2, and the third region A3 as in the magnetoresistive element 1, the Joule heat is uniform in each region of the channel layer 7. Occurs.

Here, when the thermal conductivity of the reference electrode 20 is larger than the thermal conductivity of the channel layer 7, the reference electrode from the channel layer 7 (the side with the lower thermal conductivity) is formed on the surface where the reference electrode 20 and the channel layer 7 are in contact with each other. Thermal diffusion (heat dissipation) occurs in the direction toward 20 (the side with higher thermal conductivity), and a temperature gradient is generated inside the channel layer 7. More specifically, the temperature distribution of the channel layer 7 is high on the first region A1 side (side away from the reference electrode 20) and low on the second region A2 side (side near the reference electrode 20). Become a thing.

On the other hand, when the thermal conductivity of the reference electrode 20 is smaller than the thermal conductivity of the channel layer 7, the diffusion (heat dissipation) of Joule heat generated in the channel layer 7 in the direction of the reference electrode 20 is suppressed, and the channel layer The temperature gradient generated inside 7 is suppressed to be small. The smaller the temperature gradient inside the channel layer 7, the smaller the output noise. In this case, the SN ratio of the magnetoresistive effect element 1 can be further increased. Therefore, the thermal conductivity of the reference electrode 20 is preferably smaller than the thermal conductivity of the channel layer 7.

(Production method)
For example, the magnetoresistive effect element 1 is preferably manufactured by the following procedure.

A substrate having an insulating layer on a support substrate 30 and comprising an insulating layer corresponding to the base insulating layer 80 and a nonmagnetic conductor layer corresponding to the reference electrode 20 by photolithography, ion milling, sputtering, or the like. Is made.

Next, the surface of the structure is flattened by CMP (Chemical Mechanical Polishing) or the like, and the base insulating layer 80 and the reference electrode 20 are formed. In addition, deposits (particles, organic substances, natural oxide film, etc.) on the surface of the structure including the base insulating layer 80 and the reference electrode 20 are removed and cleaned by chemical cleaning or plasma treatment.

Next, a nonmagnetic semiconductor layer, a barrier layer, and a ferromagnetic layer are formed on the structure by MBE (molecular beam epitaxy) or the like to form a laminated film.

Next, the laminated film is patterned into a rectangular shape by photolithography, ion milling, or the like, so that a laminated structure including a channel layer 7, a barrier layer, and a ferromagnetic layer is formed on the structure. The laminated structure is patterned so that only the lower surface of the region corresponding to the second region of the channel layer 7 is in contact with the upper surface of the reference electrode 20.

Finally, the barrier layer and the ferromagnetic layer of the stacked structure are patterned by photolithography and ion milling to form the barrier layer 14A, the first ferromagnetic layer 12A, the barrier layer 14B, and the second ferromagnetic layer 12B. To do. As a result, the barrier layer 14A and the first ferromagnetic layer 12A are stacked above the first region A1 of the channel layer 7, and the barrier layer 14B and the second strong layer are stacked above the second region A2 of the channel layer 7. A structure in which the magnetic layer 12B is laminated is produced.

In the magnetoresistive effect element 1 manufactured in this way, the side surface on the long side of the channel layer 7 and the side surfaces of the first ferromagnetic layer 12A and the second ferromagnetic layer 12B (the first ferromagnetic layer 12A and the second strong layer). The side surfaces along the arrangement direction of the magnetic layer 12B) are present on the same surface. Also, one side surface of the short side of the channel layer 7 and the side surface of the first ferromagnetic layer 12A (the side surface intersecting the arrangement direction of the first ferromagnetic layer 12A and the second ferromagnetic layer 12B, The side surface opposite to the ferromagnetic layer 12B) exists on the same surface. In addition, the other side surface of the short side of the channel layer 7 and the side surface of the second ferromagnetic layer 12B (the side surface intersecting the arrangement direction of the first ferromagnetic layer 12A and the second ferromagnetic layer 12B, The side surface opposite to the ferromagnetic layer 12A) exists on the same surface.

Note that the distance from the first ferromagnetic layer 12 </ b> A to the second ferromagnetic layer 12 </ b> B in the channel layer 7 is preferably equal to or shorter than the spin diffusion length of the material used for the channel layer 7.

(Explanation of effect)
When an n-type semiconductor is used as the material of the channel layer 7 and a voltage with the second ferromagnetic layer 12B being positive and the first ferromagnetic layer 12A being negative is applied in the spin transport path, the second ferromagnetic layer 12B A current flows through the channel layer 7 to the first ferromagnetic layer 12A, and an electric field is generated inside the channel layer 7 with the second region A2 side being positive and the first region A1 side being negative. In this case, electrons that have undergone spin polarization by the first ferromagnetic layer 12A are injected from the first ferromagnetic layer 12A and transported to the second ferromagnetic layer 12B via the channel layer 7, but the channel layer 7 Inside, the force (drift electric field) for accelerating and transporting the spin-polarized electron in the same direction as the direction of flow of the spin-polarized electron is applied to the spin-polarized electron. As a result, spin-polarized electrons are accumulated at high density in the second region A2 (particularly, near the interface between the second ferromagnetic layer 12B and the second region A2).

  When a p-type semiconductor is used as the material of the channel layer 7 and a voltage with the first ferromagnetic layer 12A being positive and the second ferromagnetic layer 12B being negative is applied in the spin transport path, the first ferromagnetic layer 12A A current flows through the channel layer 7 to the second ferromagnetic layer 12B, and an electric field is generated inside the channel layer 7 with the first region A1 side being positive and the second region A2 side being negative. In this case, holes whose spin polarization is caused by the first ferromagnetic layer 12A are injected from the first ferromagnetic layer 12A and transported to the second ferromagnetic layer 12B via the channel layer 7, but the channel layer 7 Inside the, a force (drift electric field) that accelerates and transports the spin-polarized hole in the same direction as the spin-polarized hole flows is applied to the spin-polarized hole. Thereby, spin-polarized holes accumulate at high density in the second region A2 (particularly, near the interface between the second ferromagnetic layer 12B and the second region A2).

  Thus, spin-polarized carriers are transported in the direction from the first ferromagnetic layer 12A to the second ferromagnetic layer 12B, and a spin non-equilibrium state (spin accumulation) is generated in the second region A2. In the magnetoresistive effect element 1, a magnetoresistive effect caused by the magnetization of the first ferromagnetic layer 12A and the magnetization of the second ferromagnetic layer 12B occurs, and the magnetization direction of the first ferromagnetic layer 12A and the second ferromagnetic layer A resistance change corresponding to a change in the relative angle of the magnetization direction of 12B occurs. Among these, the voltage change (output value) corresponding to the resistance change due to the magnetoresistive effect between the second ferromagnetic layer 12B and the second region A2 of the channel layer 7 mainly passes through the channel layer 7 to the second. It is detected between the ferromagnetic layer 12B and the reference electrode 20.

  In the magnetoresistive effect element 1, at least a part of the reference electrode 20 is in contact with a region composed of the first region A1, the second region A2, and the third region A3. Therefore, compared with the case where the region composed of the first region A1, the second region A2, and the third region A3 and the reference electrode 20 are arranged apart from each other (conventional example), the extending portion of the channel layer 7 (channel The volume of the entire layer excluding the region corresponding to the spin transport path can be reduced.

  A perspective view of a magnetoresistive effect element 1x according to a conventional example is shown in FIG. As shown in FIG. 9, the channel layer 7x of the magnetoresistive effect element 1x includes a reference electrode 20x provided apart from the second region A2x and a region overlapping with the channel layer 7x as viewed in the thickness direction, It has the extension part AEx which is an area | region which consists of an area | region between 2 area | regions A2x. In the magnetoresistive effect element 1x, the channel layer 7x includes a first region A1x, a second region A2x, a third region A3x that is a region between the first region A1x and the second region A2x, and an extending portion AEx. Yes. The reference electrode 20x is not in contact with the second region A2x, but is in contact with the extending part AEx.

  The spin accumulated in the second region A2 (A2x) has a property of three-dimensionally diffusing inside the channel layer 7 (7x) around the second region A2 (A2x). Therefore, as in the magnetoresistive effect element 1, by reducing the volume of the extending portion of the channel layer 7, the spin accumulated in the second region A2 is diffused in the direction from the second region A2 to the extending portion. Can be suppressed. As a result, spins can be accumulated in the second region A2 with high density, and the output signal of the magnetoresistive effect element 1 can be increased. Therefore, the magnetoresistive effect element 1 can have a high SN ratio.

  In the magnetoresistive effect element 1, there is a region in the channel layer 7 where a portion where a spin-polarized carrier flows (spin transport path) and a portion where a voltage is detected (spin detection path) overlap. As the volume of the region where the spin transport path and the spin detection path overlap in the channel layer 7 is reduced, the influence of the voltage drop generated in the spin transport path can be reduced and the output noise can be reduced. Further, as the contact area between the channel layer 7 and the reference electrode 20 is larger, the resistance value in the spin detection path can be reduced, and the output noise can be reduced. In the magnetoresistive effect element 1, since at least a part of the reference electrode 20 is in contact with the second region A2, at least a part of the reference electrode 20 is a region composed of the first region A1, the second region A2, and the third region A3. As compared with the case where the contact area between the channel layer 7 and the reference electrode 20 is the same as compared with the case where the second region A2 and the reference electrode 20 are spaced apart from each other, the spin in the channel layer 7 is compared. It is possible to reduce the volume of the region where the part where the polarized carriers flow (spin transport path) and the part where the voltage is detected (spin detection path) overlap. As a result, the output noise of the magnetoresistive effect element 1 can be reduced.

  Further, in the magnetoresistive effect element 1, since the reference electrode 20 is not in contact with the first region A1, a portion where the spin-polarized carriers flow in the channel layer 7 (spin transport path) and a portion where the voltage is detected (spin detection path) ) Does not exist in the first region A1, and the volume of the region where the spin transport path and the spin detection path in the channel layer 7 overlap is small. As a result, the output noise of the magnetoresistive effect element 1 can be reduced.

  Further, in the magnetoresistive effect element 1, since the reference electrode 20 is not in contact with the third region A3, a portion where the spin-polarized carriers flow in the channel layer 7 (spin transport path) and a portion where the voltage is detected (spin detection path) ) Does not exist even in the third region A3, and the volume of the region where the spin transport path and the spin detection path in the channel layer 7 overlap is small. As a result, the output noise of the magnetoresistive effect element 1 can be further reduced.

In the magnetoresistive effect element 1 of the first embodiment, the first ferromagnetic layer 12A and the second ferromagnetic layer 12B are provided on the upper surface side of the channel layer 7, and the reference electrode 20 is provided on the lower surface side of the channel layer 7. However, the first ferromagnetic layer 12 </ b> A and the second ferromagnetic layer 12 </ b> B may be provided on the lower surface side of the channel layer 7, and the reference electrode 20 may be provided on the upper surface side of the channel layer 7. Further, the first ferromagnetic layer 12 </ b> A and the reference electrode 20 may be provided on the upper surface side of the channel layer 7, and the second ferromagnetic layer 12 </ b> B may be provided on the lower surface side of the channel layer 7. In addition, the first ferromagnetic layer 12 </ b> A and the reference electrode 20 may be provided on the lower surface side of the channel layer 7, and the second ferromagnetic layer 12 </ b> B may be provided on the upper surface side of the channel layer 7.

Moreover, in the magnetoresistive effect element 1 of 1st Embodiment, although the reference electrode 20 demonstrated the example arrange | positioned so that the lower surface of 2nd area | region A2 might be contacted, it seems that the reference electrode 20 contacts the side surface of 2nd area | region A2. May be arranged. In FIG. 2, the magnetoresistive effect element 1 a in which the reference electrode 20 is disposed on the end surface (side surface on the short side of the channel layer 7) of the second region A <b> 2 facing the surface in which the second region A <b> 2 contacts the third region A <b> 3. FIG. Also in the magnetoresistive effect element 1a, since the reference electrode 20 is provided so as to be in contact with the second region A2, a high SN ratio can be obtained.

(Second Embodiment)
Hereinafter, the magnetoresistive effect element 2 according to the second embodiment will be described mainly with respect to differences from the magnetoresistive effect element 1 of the first embodiment, and description of common matters will be omitted as appropriate.

The magnetoresistive effect element 2 is different from the magnetoresistive effect element 1 of the first embodiment in the configuration of the channel layer 7. As shown in FIG. 3, in the magnetoresistive effect element 2, the second region A2 extends from the end surface of the second region A2 facing the surface in contact with the third region A3 toward the direction away from the third region A3. Part AE. In the magnetoresistive effect element 2 shown in FIG. 3, the extending portion AE is a region obtained by removing the second region A2 from the region where the channel layer 7 overlaps the reference electrode 20 and the channel layer 7 when viewed in the thickness direction. . The reference electrode 20 is in contact with the second region A2 and the lower surface of the extending part AE. The other points of the magnetoresistive effect element 2 are the same as those of the magnetoresistive effect element 1 of the first embodiment.

In the magnetoresistive effect element 2, as in the conventional example (the magnetoresistive effect element 1x), the extending portion AE is provided in the channel layer 7. However, at least a part of the reference electrode 20 includes the first region A1 and the second region. By being provided so as to be in contact with the region composed of A2 and the third region A3, the volume of the extending portion AE is smaller than that of the conventional example (the magnetoresistive effect element 1x). Therefore, in the magnetoresistive effect element 2, the diffusion of the spin accumulated in the second region A2 in the direction from the second region A2 to the extending portion AE can be suppressed. Thereby, although the output signal increasing effect is reduced as compared with the magnetoresistive effect element 1, the output signal can be increased as compared with the conventional example.

Further, in the magnetoresistive effect element 2, since the reference electrode 20 is provided on the lower surface of the second region A2 and the extending portion AE, the contact area between the channel layer 7 and the reference electrode 20 rather than the magnetoresistive effect element 1. Is getting bigger. Therefore, the magnetoresistive effect element 2 can reduce the output noise as compared with the magnetoresistive effect element 1. As a result, a high SN ratio can be obtained also in the magnetoresistive effect element 2.

In the magnetoresistive effect element 2 according to the second embodiment, the channel layer 7 has a rectangular parallelepiped shape. However, the channel layer 7 may have a tapered shape. FIG. 4 shows a side surface of the extending portion AE of the channel layer 7 (a side surface intersecting the arrangement direction of the first ferromagnetic layer 12A and the second ferromagnetic layer 12B and opposite to the second region A2). The perspective view of the magnetoresistive effect element 2a whose shape of a part is a taper shape is shown. In the magnetoresistive effect element 2a, the channel layer 7 has a cross section perpendicular to the thickness direction of the channel layer 7 from the upper surface side (opposite surface side to the second ferromagnetic layer 12B) to the lower surface side (opposite the reference electrode 20). The shape becomes wider toward the surface side. In the magnetoresistive effect element 2a, the side surface of the channel layer 7 on the first ferromagnetic layer 12A side (the side surface intersecting the arrangement direction of the first ferromagnetic layer 12A and the second ferromagnetic layer 12B) and the first ferromagnetic layer The side surface of 12A (the side surface intersecting the arrangement direction of the first ferromagnetic layer 12A and the second ferromagnetic layer 12B and opposite to the second ferromagnetic layer 12B) exists on the same surface. doing.

The magnetoresistive effect element 2a, for example, by increasing the inclination angle of ion milling when forming a side surface having a taper in a step of patterning a laminated film composed of a nonmagnetic semiconductor layer, a barrier layer, and a ferromagnetic layer, Can be produced.

  In the magnetoresistive effect element 2a, compared with the case where the contact area between the channel layer 7 and the reference electrode 20 is the same as in the case where the shape of the channel layer 7 is a rectangular parallelepiped (the magnetoresistive effect element 2), The volume of the extending part AE can be reduced. Thereby, in the magnetoresistive effect element 2a, while providing the output noise reduction effect obtained by enlarging the contact area of the channel layer 7 and the reference electrode 20, the upper part of the second region A2 (with the second ferromagnetic layer 12B) The diffusion of the spin accumulated in the vicinity of the interface between the second region A2 and the second region A2 in the direction from the second region A2 to the extending portion AE can be suppressed, and the output signal can be increased. Accordingly, the magnetoresistive effect element 2a can obtain a higher SN ratio than the magnetoresistive effect element 2.

(Third embodiment)
Hereinafter, the magnetoresistive effect element 3 according to the third embodiment will be described mainly with respect to differences from the magnetoresistive effect element 1 of the first embodiment, and description of common matters will be omitted as appropriate.

The magnetoresistive effect element 3 is different from the magnetoresistive effect element 1 of the first embodiment in the installation position of the reference electrode 20. As shown in FIG. 5, in the magnetoresistive effect element 3, the reference electrode 20 is provided in contact with a part of the lower surface of the second region A <b> 2 of the channel layer 7.

The magnetoresistive effect element 3 uses the pattern layout of the step of forming the reference electrode 20, the step of forming the channel layer 7, and the step of forming the second ferromagnetic layer 12 </ b> B as a manufacturing method of the magnetoresistive effect element 1. It can produce by changing with respect to it.

The magnetoresistive effect element 3 has a smaller contact area between the channel layer 7 and the reference electrode 20 than when the reference electrode 20 is in contact with the entire lower surface of the second region A2 (the magnetoresistive effect element 1). . As a result, the volume of the region where the portion where the spin-polarized carriers flow and the portion where the voltage is detected overlaps in the channel layer 7 can be reduced. As a result, the output noise of the magnetoresistive effect element 3 can be reduced.

The magnetoresistive effect element 3 has a resistance value at the interface between the channel layer 7 and the reference electrode 20 as compared with the case where the reference electrode 20 is in contact with the entire lower surface of the second region A2 (the magnetoresistive effect element 1). It is getting bigger. In this case, the output noise due to the resistance value of the spin detection path is larger than that of the magnetoresistive effect element 1, but the volume of the region in the channel layer 7 where the portion where the spin-polarized carriers flow and the portion where the voltage is detected overlaps. An output noise reduction effect can be obtained by reducing. Therefore, a high SN ratio can be obtained also in the magnetoresistive effect element 3.

In the first to third embodiments described so far, the example in which the reference electrode 20 is disposed so as to be in contact with the channel layer 7 only in the second region A2 is described. However, the reference electrode 20 is formed in the second region A2. You may arrange | position so that it may contact | connect also not only on the lower surface but on the lower surface of 3rd area | region A3 or 1st area | region A1. FIG. 6 shows a perspective view of the magnetoresistive effect element 3a in which the reference electrode 20 is disposed so as to be in contact with the lower surfaces of the second region A2 and the third region A3. Also in the magnetoresistive effect element 3a, the reference electrode 20 is provided apart from the first ferromagnetic layer 12A and the second ferromagnetic layer 12B.

In the magnetoresistive effect element 3a, compared with the case where the reference electrode 20 is disposed so as to be in contact with the channel layer 7 only in the second region A2 (the magnetoresistive effect element 1), at the interface between the channel layer 7 and the reference electrode 20 The resistance value is small. Thereby, the output noise of the magnetoresistive effect element 3a can be reduced.

On the other hand, in the magnetoresistive effect element 3a, the spin-polarized carriers in the channel layer 7 are smaller than in the case where the reference electrode 20 is disposed so as to be in contact with the channel layer 7 only in the second region A2 (magnetoresistance effect element 1). The volume of the region where the flowing part and the part for detecting the voltage overlap is large. In this case, the output noise due to the volume of the region where the portion where the spin-polarized carriers flow in the channel layer 7 and the portion where the voltage is detected overlaps is larger than that in the magnetoresistive element 1, but the resistance value of the spin detection path An output noise reduction effect can be obtained by reducing. Therefore, a high SN ratio can be obtained also in the magnetoresistive effect element 3a.

In the first to third embodiments, the example in which at least a part of the reference electrode 20 is in contact with the second region A2 has been described. However, the reference electrode 20 may not be in contact with the second region A2. For example, at least a part of the channel layer 7 and the reference electrode 20 may be provided so as to be in contact with both or only one of the first region A1 and the third region A3. Even in the magnetoresistive effect element in such a case, at least a part of the reference electrode 20 is in contact with the region composed of the first region A1, the second region A2, and the third region A3, so that the first region Compared with the case where the reference electrode 20 and the region composed of A1, the second region A2, and the third region A3 are separated from each other, the volume of the extending portion of the channel layer 7 can be reduced, and the high SN Ratio can be realized.

(Fourth embodiment)
Hereinafter, the magnetic sensor 100 according to the fourth embodiment will be described. 7 is a cross-sectional view of the main part of the magnetic sensor 100, and FIG. 8 is a side view (side view seen from the X direction) seen from the magnetic medium M in FIG. As shown in FIGS. 7 and 8, the magnetic sensor 100 includes the magnetoresistive effect element 1 and the magnetoresistive effect element 1 in a film surface orthogonal direction (Z direction) via a base insulating layer 80 on a support substrate 30. A lower magnetic shield 40 and an upper magnetic shield 50 are provided so as to be sandwiched therebetween. As shown in FIG. 8, the magnetic sensor 100 includes bias magnetic field application layers 16 provided on both sides in the Y direction (the direction perpendicular to the paper in FIG. 7) that is the width direction of the magnetoresistive element 1. Yes. An insulating film 4 is provided between the channel layer 7 and the lower magnetic shield 40 and between the magnetoresistive effect element 1 and the bias magnetic field applying layer 16. The insulating film 4 is formed to insulate and separate the bias magnetic field application layer 16 and the magnetoresistive element 1 and to suppress spin absorption by the lower magnetic shield 40 and the bias magnetic field application layer 16. The bias magnetic field application layer 16 applies a Y-direction bias magnetic field to the first ferromagnetic layer 12A in order to make the first ferromagnetic layer 12A of the magnetoresistive effect element 1 into a single magnetic domain. The material of the insulating film 4 is, for example, Al ₂ O ₃ or SiO ₂ , the material of the bias magnetic field applying layer 16 is, for example, CoPt or CoCrPt, and the material of the lower magnetic shield 40 and the upper magnetic shield 50 is preferably, for example, NiFe .

As shown in FIG. 7, the lower magnetic shield 40 is provided below the channel layer 7 and the reference electrode 20. The reference electrode 20 is provided in contact with the concave portion of the lower magnetic shield 40. The upper magnetic shield 50 is provided on the first ferromagnetic layer 12A. A wiring 5 is provided on the second ferromagnetic layer 12B. The upper magnetic shield 50 also serves as a wiring for applying a current to the magnetoresistive effect element 1. The lower magnetic shield 40 also serves as a wiring for detecting the voltage of the magnetoresistive effect element 1. The material of the wiring 5 may be the same as the material of the lower magnetic shield 40 and the upper magnetic shield 50. A current source 60 is connected between the upper magnetic shield 50 and the wiring 5, and a current is applied between the first ferromagnetic layer 12 </ b> A and the second ferromagnetic layer 12 </ b> B via the channel layer 7. ing. A voltmeter 70 is connected between the wiring 5 and the lower magnetic shield 40 so that the voltage between the second ferromagnetic layer 12B and the reference electrode 20 is detected.

The magnetic sensor 100 will be described using an example in which a magnetic field generated from the magnetic medium M is detected as an external magnetic field to be detected. As shown in FIG. 7, in the magnetic sensor 100, the tip portion (one end of the first ferromagnetic layer 12 </ b> A) of the magnetoresistive effect element 1 is disposed on the surface facing the magnetic medium M of the magnetic sensor 100. In the magnetic sensor 100, the first ferromagnetic layer 12A functions as a layer (magnetization free layer) whose magnetization direction changes according to the magnetic field generated from the magnetic medium M. As the first ferromagnetic layer 12A, in particular, a soft magnetic material is applied. A protective film such as DLC (diamond-like carbon) is preferably formed on the surface of the magnetic sensor 100 facing the magnetic medium M, but the protective film is omitted in FIGS. 7 and 8. Yes.

As shown in FIG. 7, in the magnetic sensor 100, the second ferromagnetic layer 12 </ b> B is disposed at a position away from the magnetic medium M. For this reason, since the second ferromagnetic layer 12B is not easily affected by the magnetic field generated from the magnetic medium M, the second ferromagnetic layer 12B functions as a layer whose magnetization direction is fixed in one direction (magnetization fixed layer). In order to make the magnetization of the second ferromagnetic layer 12B fixed more firmly (increase the coercive force of the second ferromagnetic layer 12B), shape anisotropy may be imparted to the second ferromagnetic layer 12B. Alternatively, an exchange coupling acting between the second ferromagnetic layer 12B and the antiferromagnetic layer may be used by stacking an antiferromagnetic layer on the second ferromagnetic layer 12B. Alternatively, exchange coupling acting inside the second ferromagnetic layer 12B may be used by forming the second ferromagnetic layer 12B in a synthetic pinned structure.

A current is applied between the first ferromagnetic layer 12A and the second ferromagnetic layer 12B via the channel layer 7, and the voltage between the second ferromagnetic layer 12B and the reference electrode 20 is measured, whereby the magnetic medium It becomes possible to detect the magnetic field generated from M. In accordance with the change in the magnetic field generated from the magnetic medium M, the relative angle between the magnetization direction of the first ferromagnetic layer 12A and the magnetization direction of the second ferromagnetic layer 12B changes, and the second ferromagnetic layer 12B and the reference electrode 20 The voltage between changes.

Since the magnetic sensor 100 has the magnetoresistive effect element 1 that can obtain a high S / N ratio, it can detect an external magnetic field with high accuracy.

The magnetic sensor 100 of the fourth embodiment is an example having the magnetoresistive effect element 1, but instead of the magnetoresistive effect element 1, the magnetoresistive effect element 1a, the magnetoresistive effect element 2, the magnetoresistive effect element 2a, and the magnetism The resistance effect element 3 or the magnetoresistance effect element 3a may be used.

DESCRIPTION OF SYMBOLS 1, 1a, 2, 2a, 3, 3a ... Magnetoresistance effect element 7 ... Channel layer 12A ... 1st ferromagnetic layer 12B ... 2nd ferromagnetic layer 14A, 14B ... Barrier layer 20 ... Reference electrode A1 ... 1st area | region A2 ... 2nd area A3 ... 3rd area

Claims (6)

A channel layer, a first ferromagnetic layer, a second ferromagnetic layer, and a reference electrode;
The first ferromagnetic layer, the second ferromagnetic layer, and the reference electrode are separated from each other and electrically connected to each other through the channel layer,
The first ferromagnetic layer and the second ferromagnetic layer are separated from each other without overlapping each other when viewed from the thickness direction of the channel layer,
The channel layer includes a first region overlapping with the first ferromagnetic layer as viewed from the thickness direction, a second region overlapping with the second ferromagnetic layer as viewed from the thickness direction, and the first region and the second region. A third region between the regions,
At least a part of the reference electrode is in contact with a region composed of the first region, the second region, and the third region.   The magnetoresistive effect element according to claim 1, wherein at least part of the reference electrode is in contact with the second region.   The magnetoresistive effect element according to claim 2, wherein the first region is not in contact with the reference electrode.   The magnetoresistive effect element according to claim 2, wherein the third region is not in contact with the reference electrode.   The magnetoresistive effect element according to any one of claims 1 to 4, wherein the thermal conductivity of the reference electrode is smaller than the thermal conductivity of the channel layer.   The magnetic sensor which has a magnetoresistive effect element as described in any one of Claims 1 thru | or 5.

Images