WO2024154441A1 - Magnetoresistance effect element - Google Patents

Abstract

This magnetoresistance effect element comprises a first ferromagnetic layer, a second ferromagnetic layer, a nonmagnetic layer, and a first intermediate layer. The nonmagnetic layer is sandwiched between the first ferromagnetic layer and the second ferromagnetic layer and contains a cubic or tetragonal nitride. The first intermediate layer is sandwiched between the first ferromagnetic layer and the nonmagnetic layer and contains a cubic or tetragonal nonmagnetic material.

Description

Magnetoresistance effect element

The present invention relates to a magnetoresistance effect element. This application has priority to the contents described in Japanese Patent Application No. 2023-005203, filed in Japan on January 17, 2023.

A magnetoresistance effect element has two ferromagnetic layers and a nonmagnetic layer sandwiched between the two ferromagnetic layers. The resistance value of the magnetoresistance effect element changes as the relative angle of magnetization of the two ferromagnetic layers sandwiching the nonmagnetic layer changes. Magnetoresistance effect elements exhibit what is known as the magnetoresistance effect. Taking advantage of this resistance change characteristic, magnetoresistance effect elements are used in a variety of applications such as magnetic sensors, high-frequency components, magnetic heads, and magnetic memories.

As magnetic memories become larger in capacity, there is a demand for miniaturizing magnetoresistance effect elements. Miniaturized magnetoresistance effect elements have a small volume of ferromagnetic layer and low thermal stability of magnetization. For example, perpendicular magnetization magnetoresistance effect elements are known that use a CoFe-based ferromagnetic material for the ferromagnetic layer and MgO or Mg-Al-O nonmagnetic material for the nonmagnetic layer. Magnetoresistance effect elements that use a CoFe-based ferromagnetic material may not have sufficient thermal stability when miniaturized. For this reason, research is underway into the materials of the ferromagnetic layer to be used in magnetoresistance effect elements.

For example, Patent Documents 1 and 2 disclose magnetoresistance effect elements that use Mn-based magnetic layers. The Mn-based magnetic layer is not a ferromagnetic material such as an Fe-based material, and when the magnetoresistance effect element is miniaturized, the magnetization has less of an effect on the surrounding area. Mn-based magnetic materials are attracting attention as a material that may be applied to magnetoresistance effect elements in the future, with the aim of increasing the capacity of magnetic memories.

Patent Documents 1 and 2 propose a magnetoresistance element that has perpendicular magnetic anisotropy and is capable of exhibiting a larger magnetoresistance effect. However, Patent Documents 1 and 2 do not disclose specific data on the magnetoresistance effect of this magnetoresistance element. To date, there have been no reports of a magnetoresistance effect element using a Mn-based magnetic layer (particularly an element having a Mn-based magnetic layer on a barrier layer (non-magnetic layer)) exhibiting a large magnetoresistance effect.

JP 2010-232499 A JP 2015-176930 A

The performance of a magnetoresistance effect element is not determined only by the ferromagnetic layer. The performance of a magnetoresistance effect element is also affected by the nonmagnetic layer. A magnetoresistance effect element is often connected to a semiconductor element that controls the operation of the magnetoresistance effect element. It is preferable that the operating resistance of this semiconductor element and the resistance in the stacking direction of the magnetoresistance effect element do not deviate significantly. For example, MgO and Mg-Al-O have high resistance. When using these materials for the nonmagnetic layer, the nonmagnetic layer may be thinned to bring the resistance of the magnetoresistance effect element closer to the operating resistance of the semiconductor element. However, a thin nonmagnetic layer may undergo dielectric breakdown when a voltage is applied. Therefore, a nonmagnetic layer containing a material with lower resistance is required.

For example, AlN, GaN, and InN are candidates for low-resistance nonmagnetic materials. However, the crystal structures of AlN, GaN, and InN are generally wurtzite structures. Magnetoresistance effect elements are fabricated by stacking thin films, so it is difficult to stack a ferromagnetic layer with a different crystal structure on a nonmagnetic layer with a wurtzite structure.

The inventors of the present application believed that the material and crystal structure of the nonmagnetic layer would be important factors in improving the performance of the magnetoresistance effect element by using a low-resistance nonmagnetic material and in realizing a magnetoresistance effect element with perpendicular magnetic anisotropy made of an Mn-based magnetic layer, and so they investigated the nonmagnetic layer. As mentioned above, there have been no reports of magnetoresistance effect elements that include an Mn-based magnetic layer.

This disclosure has been made in consideration of the above problems, and aims to provide a magnetoresistance effect element with a low-resistance tunnel barrier layer.

To solve the above problems, this disclosure provides the following means.

The magnetoresistance effect element according to the first aspect includes a first ferromagnetic layer, a second ferromagnetic layer, a nonmagnetic layer, and a first intermediate layer. The nonmagnetic layer is sandwiched between the first ferromagnetic layer and the second ferromagnetic layer, and includes a cubic or tetragonal nitride. The first intermediate layer is sandwiched between the first ferromagnetic layer and the nonmagnetic layer, and includes a cubic or tetragonal nonmagnetic material.

The magnetoresistance effect element according to the above embodiment has a low-resistance tunnel barrier layer and has low resistance in the stacking direction.

1 is a cross-sectional view of a magnetoresistive effect element according to a first embodiment. FIG. 11 is a cross-sectional view of a magnetoresistive effect element according to a second embodiment. 4 shows the simulation results of Experimental Example 1. 1 is a cross-sectional TEM image of a magnetoresistive effect element produced in Experimental Example 2. 13 is a cross-sectional TEM image of the magnetoresistive effect element produced in Experimental Example 3. 13 shows the results of X-ray diffraction of the magnetoresistive effect element produced in Experimental Example 3. 13 shows the results of X-ray diffraction of the magnetoresistance effect element of the modified example of Experimental Example 3. 13 shows the results of X-ray diffraction of the magnetoresistance effect element of the modified example of Experimental Example 3. 13 shows the results of Experimental Example 4, illustrating the MR ratio of the magnetoresistive element when the thickness ratio between the first intermediate layer and the nonmagnetic layer is changed. 13 shows the MR ratio of the magnetoresistive element of Experimental Example 5. RA of the magnetoresistive effect element of Experimental Example 5 is shown. 4 shows the magnetic properties of the sample of Experimental Example 6. 1 shows the results of X-ray diffraction of the sample of Experimental Example 6. 13 shows magnetic properties of the sample of Experimental Example 7. 13 shows the magnetic properties of the sample of Experimental Example 8. 13 shows the results of X-ray diffraction of the magnetoresistive effect element of Experimental Example 9. 13 shows the magnetic characteristics of the magnetoresistance effect element of Experimental Example 9. 13 shows the results of X-ray diffraction of the magnetoresistive effect element of Experimental Example 10. This shows the results of calculating the band structure of a GaN nonmagnetic layer. This shows the relationship between the thickness of the nonmagnetic layer and the element resistance of the magnetoresistive effect element of Experimental Example 10 (the nonmagnetic layer is GaN), and the relationship between the thickness of the nonmagnetic layer and the element resistance of a magnetoresistive effect element having the same configuration as Experimental Example 10 except that the nonmagnetic layer is MgO. 13 shows the MR ratio of the magnetoresistive element of Experimental Example 10.

The present embodiment will now be described in detail with reference to the drawings as appropriate. The drawings used in the following description may show enlarged characteristic parts for the sake of convenience in order to make the features of the present invention easier to understand, and the dimensional ratios of each component may differ from the actual ones. The materials, dimensions, etc. exemplified in the following description are merely examples, and the present invention is not limited to them, and may be modified as appropriate within the scope of the present invention.

"First embodiment"
1 is a cross-sectional view of a magnetoresistance effect element 10 according to the first embodiment. The magnetoresistance effect element 10 has a first ferromagnetic layer 1, a second ferromagnetic layer 2, a nonmagnetic layer 3, a first intermediate layer 4, and an underlayer 9.

The magnetoresistance effect element 10 is a columnar body. The shape of the magnetoresistance effect element 10 when viewed from the z direction in a plane is, for example, circular, elliptical, or rectangular.

The resistance value of the magnetoresistance effect element 10 changes when the relative angle between the magnetization of the first ferromagnetic layer 1 and the magnetization of the second ferromagnetic layer 2, which sandwich the nonmagnetic layer 3, changes. When the magnetization of the first ferromagnetic layer 1 and the magnetization of the second ferromagnetic layer 2 are antiparallel, the resistance value of the magnetoresistance effect element 10 increases, and when the magnetization of the first ferromagnetic layer 1 and the magnetization of the second ferromagnetic layer 2 are parallel, the resistance value of the magnetoresistance effect element 10 decreases. The magnetoresistance ratio (MR ratio) of the magnetoresistance effect element is expressed as (R _A -R _P )/R _P ×100. The larger the MR ratio, the greater the magnetoresistance effect of the magnetoresistance effect element. R _A is the resistance value in the stacking direction of the magnetoresistive effect element 10 when the magnetization of the first ferromagnetic layer 1 and the magnetization of the second ferromagnetic layer 2 are antiparallel, and R _P is the resistance value in the stacking direction of the magnetoresistive effect element 10 when the magnetization of the first ferromagnetic layer 1 and the magnetization of the second ferromagnetic layer 2 are parallel.

The underlayer 9 is, for example, on a substrate. The substrate is, for example, silicon with a thermal oxide film, glass, ceramic (for example, alumina, magnesium oxide), quartz, etc. The substrate may be amorphous or crystalline. If the substrate is crystalline, it is preferable that the layered surface of the substrate is (001) oriented.

The underlayer 9 has, for example, a cubic or tetragonal crystal structure. The underlayer 9 is, for example, Cr, V, TiN, MnN, NiAl, CoAl, CoGa, or NiGa. The underlayer 9 has a cubic or tetragonal crystal structure with an appropriate lattice constant and is oriented in the (001) crystal plane, so that the layer deposited on the underlayer 9 can be cubic or tetragonal.

The first ferromagnetic layer 1 is, for example, on the underlayer 9. The first ferromagnetic layer 1 is closer to the underlayer 9 than the second ferromagnetic layer 2. The crystal structure of the first ferromagnetic layer 1 is, for example, cubic or tetragonal.

The first ferromagnetic layer 1 includes a ferromagnetic material. The ferromagnetic material may include ferrimagnetic material, and the first ferromagnetic layer 1 may be a ferrimagnetic material. The first ferromagnetic layer 1 may be a ferromagnetic material including magnetic elements such as Co, Fe, and Ni. For example, Fe, Co, CoFe, CoFeB, CoMnFe, _Co2MnSi , _Co2FeAl , CoPt, NiFe, and FePt are examples of ferromagnetic materials including magnetic elements such as Co, Fe, and Ni. The first ferromagnetic layer 1 may also be a Mn-based ferrimagnetic material. The Mn-based ferrimagnetic material has high magnetic anisotropy and small saturation magnetization. For example, an MnGa alloy is an example of a Mn-based ferrimagnetic material. The MnGa alloy may contain Mn and Ga elements, and examples of the MnGa alloy include Mn ₃ Ga, Mn ₂ FeGa, and Fe ₂ MnGa.

The first ferromagnetic layer 1 may be, for example, a magnetization fixed layer or a magnetization free layer. The magnetization of the magnetization fixed layer is less likely to move than that of the magnetization free layer when a predetermined magnetic field is applied. The magnetization stability of the first ferromagnetic layer 1 is higher when the first ferromagnetic layer 1 is closer to the substrate than the second ferromagnetic layer 2. The magnetoresistance effect element 10 in which the first ferromagnetic layer 1 is a magnetization fixed layer has high reliability (stability against thermal oscillations).

The first intermediate layer 4 is, for example, on the first ferromagnetic layer 1. The first intermediate layer 4 is, for example, between the first ferromagnetic layer 1 and the nonmagnetic layer 3. The first intermediate layer 4 is a nonmagnetic layer made of a nonmagnetic material, and may partially fulfill the role of the nonmagnetic layer 3 described below.

The first intermediate layer 4 includes a cubic or tetragonal nonmagnetic material. The first intermediate layer 4 may be made of a cubic or tetragonal nonmagnetic material. The crystal structure of the nonmagnetic material included in the first intermediate layer 4 is, for example, a rock-salt or distorted rock-salt tetragonal crystal.

The non-magnetic material constituting the first intermediate layer 4 is, for example, any one selected from the group consisting of Cr, Ti, Mn, CrN, TiN, MgO, and MnN. When producing the first intermediate layer 4 consisting of CrN or MnN, the first intermediate layer 4 is formed using Cr or Mn, and then the nitride non-magnetic layer 3 is deposited. The first intermediate layer 4 consisting of CrN or MnN is obtained by reacting some of the nitrogen in the non-magnetic layer 3 with Cr or Mn, or by reacting Cr or Mn with nitrogen generated in the process of depositing the nitride non-magnetic layer 3. The first intermediate layer 4 may contain, for example, CrN or MnN produced by such a process.

The first intermediate layer 4 may function as a seed layer (a layer that acts as a seed to promote crystal growth). The thickness of the first intermediate layer 4 may be extremely thin (one or two layers at the atomic level, specifically, about 0.1 nm to 0.2 nm). In this case, the extremely thin CrN or MnN combines with the non-magnetic layer 3 to increase the MR ratio of the magnetoresistance effect element 10 and produce an undisclosed effect of lowering the resistance value of the magnetoresistance effect element 10.

When the first intermediate layer 4 is made of MgO, which has a high resistance value, the thickness may be a certain degree. The certain degree of thickness is, for example, about 0.8 nm. On the other hand, when the second ferromagnetic layer 2 is made of a ferrimagnetic material, it is preferable that the thickness of the first intermediate layer 4 is thin. The thickness of the first intermediate layer 4 is, for example, less than 1.1 nm, and preferably 0.8 nm or less.

The thickness of the first intermediate layer 4 can be determined, for example, by measuring the cross section of the magnetoresistance effect element 10 with a transmission electron microscope (TEM), measuring the film thickness at five different points within the plane in which the first intermediate layer 4 extends, and averaging the measured values.

The first intermediate layer 4 functions as a seed layer between the first ferromagnetic layer 1 and the nonmagnetic layer 3. If a material with a high resistance value such as MgO is used for the first intermediate layer 4, the first intermediate layer 4 may also function as a tunnel barrier layer like the nonmagnetic layer 3. The presence of the first intermediate layer 4 between the first ferromagnetic layer 1 and the nonmagnetic layer 3 makes the nonmagnetic layer 3 a cubic or tetragonal crystal.

The nonmagnetic layer 3 is, for example, on the first intermediate layer 4. The nonmagnetic layer 3 is between the first ferromagnetic layer 1 and the second ferromagnetic layer 2. The nonmagnetic layer 3 is, for example, between the first intermediate layer 4 and the second ferromagnetic layer 2. The nonmagnetic layer 3 is a tunnel barrier layer.

The nonmagnetic layer 3 of the present disclosure includes a cubic or tetragonal nitride. The nonmagnetic layer 3 may be made of a cubic or tetragonal nitride. The nonmagnetic layer 3 includes, for example, a nitride of at least one element selected from the group consisting of Al, Ga, and In. The nonmagnetic layer 3 is, for example, AlN, GaN, or InN. These nitrides do not usually become cubic or tetragonal because the wurtzite structure is stable. By forming the nonmagnetic layer 3 on the cubic or tetragonal first intermediate layer 4, the nonmagnetic layer 3 becomes cubic or tetragonal.

When the first intermediate layer 4 is an insulator such as MgO, the nonmagnetic layer 3 preferably has a lower resistance than the first intermediate layer 4. A nonmagnetic layer 3 containing a nitride has a lower resistance than a nonmagnetic layer made of MgO. When the resistance of the nonmagnetic layer 3 is low, the resistance value of the magnetoresistance effect element 10 and the resistance value of the other semiconductor element connected to the magnetoresistance effect element 10 are closer in order of magnitude. Therefore, even when the magnetoresistance effect element 10 is miniaturized, the thickness of the nonmagnetic layer 3 can be set thicker than when the nonmagnetic layer 3 is made of MgO alone, and insulation breakdown of the nonmagnetic layer 3 is less likely to occur.

When the resistance value of the first intermediate layer 4 is small (for example, when the first intermediate layer 4 is made of CrN or MnN, etc.), the nonmagnetic layer 3 mainly plays the role of a tunnel barrier layer. The nonmagnetic layer 3 functions as a significant tunnel barrier layer, allowing the magnetoresistance effect element to function properly.

On the other hand, if the first intermediate layer 4 is made of a material with high resistance, such as MgO, the nonmagnetic layer 3 and the first intermediate layer 4 will act as a tunnel barrier layer.

The resistance of the nonmagnetic layer 3 of the present disclosure is lower than the resistance value of the first intermediate layer 4 when the first intermediate layer 4 is MgO. In order to prevent a short circuit through the nonmagnetic layer 3 and the first intermediate layer 4 while keeping the resistance value of the first intermediate layer 4 of the magnetoresistance effect element 10 small, it is preferable that the thickness of the nonmagnetic layer 3 is thicker than the thickness of the first intermediate layer 4. On the other hand, when the first intermediate layer 4 is MgO, if the thickness of the first intermediate layer 4 is too thin compared to the thickness of the nonmagnetic layer 3, the MR ratio of the magnetoresistance effect element 10 tends to decrease, so the thickness of the first intermediate layer 4 is preferably 0.2 nm or more.

In addition, in this disclosure, the nonmagnetic layer 3 is a cubic or tetragonal crystal because the second ferromagnetic layer 2 formed on the nonmagnetic layer 3 is a cubic or tetragonal crystal. Taking all of this into consideration, the materials and thicknesses of the first intermediate layer 4 and the nonmagnetic layer 3, and the ratio of their respective thicknesses are examined.

The second ferromagnetic layer 2 is, for example, on the nonmagnetic layer 3. The second ferromagnetic layer 2 is located farther from the underlayer 9 than the first ferromagnetic layer 1. The crystal structure of the second ferromagnetic layer 2 is, for example, cubic or tetragonal.

The second ferromagnetic layer 2 includes a ferromagnetic material. The second ferromagnetic layer 2 may be a ferromagnetic material including a magnetic element such as Co, Fe, Ni, etc. For example, Fe, Co, CoFe, _CoFeB , CoMnFe, Co2MnSi, _Co2FeAl , CoPt, FePt, etc. are examples of ferromagnetic materials including a magnetic element such as Co, Fe, Ni, etc.

The second ferromagnetic layer 2 may also be a Mn-based ferrimagnetic material. The Mn-based ferrimagnetic material has high magnetic anisotropy and low saturation magnetization. For example, a MnGa alloy is an example of a Mn-based ferrimagnetic material. The MnGa alloy may contain Mn and Ga elements, and may be, for example, Mn ₃ Ga, Mn ₂ FeGa, or Fe ₂ MnGa.

For example, if the non-magnetic layer 3 is MgO, a cubic or tetragonal Mn-based ferrimagnetic material cannot be formed on the non-magnetic layer 3. In contrast, a Mn-based ferrimagnetic material can be formed on a non-magnetic layer 3 that contains a cubic or tetragonal nitride.

The second ferromagnetic layer 2 may be, for example, a magnetization fixed layer or a magnetization free layer. It is preferable that the second ferromagnetic layer 2 is a magnetization free layer.

Each layer of the magnetoresistance effect element 10 can be fabricated by, for example, sputtering, vapor deposition, laser ablation, molecular beam epitaxial, etc. Furthermore, photolithography or electron beam lithography can be used to process the magnetoresistance effect element 10 into a desired shape.

The magnetoresistance effect element 10 according to the first embodiment has a nonmagnetic layer 3 containing a cubic or tetragonal nitride, and has a lower resistance than a magnetoresistance effect element having a nonmagnetic layer made only of MgO. In addition, since the nonmagnetic layer 3 of the magnetoresistance effect element 10 according to the first embodiment contains a cubic or tetragonal nitride, the second ferromagnetic layer 2 can be made of an Mn-based ferrimagnetic material. It is not possible to apply an Mn-based ferrimagnetic material to the upper ferromagnetic layer away from the substrate when the nonmagnetic layer is MgO, but this can be achieved only by making the nonmagnetic layer 3 a cubic or tetragonal nitride. Mn-based ferrimagnetic materials have high magnetic anisotropy and small saturation magnetization, and are therefore less susceptible to thermal fluctuations and the like. Therefore, the magnetoresistance effect element 10 in which an Mn-based ferrimagnetic material is applied to the second ferromagnetic layer 2 has excellent reliability.

Second Embodiment
2 is a cross-sectional view of the magnetoresistance effect element 11 according to the second embodiment. The magnetoresistance effect element 11 differs from the magnetoresistance effect element 10 according to the first embodiment in that it has a second intermediate layer 5. In the description of the magnetoresistance effect element 11 of the second embodiment, the same components as those of the magnetoresistance effect element 10 according to the first embodiment are denoted by the same reference numerals and description thereof will be omitted.

The magnetoresistance effect element 11 has a first ferromagnetic layer 1, a second ferromagnetic layer 2, a nonmagnetic layer 3, a first intermediate layer 4, a second intermediate layer 5, and an underlayer 9. The magnetoresistance effect element 11 is a columnar body similar to the magnetoresistance effect element 10. The resistance value of the magnetoresistance effect element 11 changes when the relative angle between the magnetization of the first ferromagnetic layer 1 and the magnetization of the second ferromagnetic layer 2, which sandwich the nonmagnetic layer 3, changes.

The underlayer 9, the first ferromagnetic layer 1, the first intermediate layer 4, and the nonmagnetic layer 3 are similar to those in the magnetoresistance effect element 10 according to the first embodiment. The second ferromagnetic layer 2 is similar to the magnetoresistance effect element 10 according to the first embodiment, except that it is located on the second intermediate layer 5.

The second intermediate layer 5 is, for example, on the nonmagnetic layer 3. The second intermediate layer 5 is, for example, between the nonmagnetic layer 3 and the second ferromagnetic layer 2.

The second intermediate layer 5 includes a cubic or tetragonal nonmagnetic material. The second intermediate layer 5 may be made of a cubic or tetragonal nonmagnetic material. The crystal structure of the nonmagnetic material included in the second intermediate layer 5 is, for example, a rock-salt or distorted rock-salt tetragonal crystal.

The non-magnetic material constituting the second intermediate layer 5 is, for example, one selected from the group consisting of Cr, Ti, Mn, CrN, TiN, and MnN.

The thickness of the second intermediate layer 5 is, for example, less than 1.1 nm, and preferably 0.8 nm or less. The thickness of the second intermediate layer 5 is determined, for example, by measuring the cross section of the magnetoresistance effect element 11 with a transmission electron microscope (TEM), measuring the film thickness at five different points in the plane in which the second intermediate layer 5 extends, and averaging the measured values.

The second intermediate layer 5 functions as a buffer layer between the second ferromagnetic layer 2 and the nonmagnetic layer 3. The presence of the second intermediate layer 5 between the second ferromagnetic layer 2 and the nonmagnetic layer 3 makes it easier for the nonmagnetic layer 3 to stabilize in a cubic or tetragonal crystal structure. Furthermore, when the second intermediate layer 5 is made of CrN or MnN, the effect of reducing the resistance of the magnetoresistance effect element and increasing the MR ratio is obtained.

The magnetoresistance effect element 11 can be fabricated in a similar manner to the magnetoresistance effect element 10 according to the first embodiment.

The magnetoresistance effect element 11 according to the second embodiment has the same effect as the magnetoresistance effect element 10 according to the first embodiment.

The present invention has been described in detail above using several embodiments as examples, but the present invention is not limited to the above embodiments and modifications, and various modifications and variations are possible within the scope of the gist of the present invention described in the claims. For example, the magnetoresistance effect element may have layers other than the first ferromagnetic layer, the second ferromagnetic layer, the nonmagnetic layer, the first intermediate layer, the second intermediate layer, and the underlayer.

(Experimental Example 1: Simulation Results of the Resistance Value of a Magnetoresistance Effect Element Having a Nonmagnetic Layer Made of a Cubic or Tetragonal Crystal)
Theoretical calculations were performed using simulations to determine the resistance value and MR ratio of the magnetoresistance effect element when the material of each layer of the magnetoresistance effect element was changed. The simulations were performed in the following procedure. First, the structure of the element was optimized using the first-principles calculation code VASP (Vienna Ab initio Simulation Package) based on the density functional theory. However, the in-plane lattice constant of the element was adjusted to the lattice constant of Fe. Next, the effective potential of the element required for the conduction calculation was obtained using the pwscf code in the first-principles calculation package Quantum ESPRESSO. Finally, ballistic conduction calculations were performed using the pwcond code included in Quantum ESPRESSO to obtain the magnetoresistance value.

The simulation was carried out on the following eight samples:
Sample s1: Fe/MnN/AlN/MnN/Fe
Sample s2: Co/CrN/AlN/CrN/Co
Sample s3: Co/MnN/AlN/MnN/Co
Sample s4: Fe/CrN/AlN/CrN/Fe
Sample s5: Fe/AlN/Fe
Sample s6: Fe/MgO/Fe
Sample s7: Co/MgO/Co
Sample s8: Co/AlN/Co

In this simulation, AlN and MgO were calculated as cubic crystals. Samples s1 to s4 have compositions in the order of first ferromagnetic layer/first intermediate layer/non-magnetic layer/second intermediate layer/second ferromagnetic layer. Samples s5 to s8 have compositions in the order of first ferromagnetic layer/non-magnetic layer/second ferromagnetic layer. Each of samples s1 to s8 has the same number of atomic layers in the first ferromagnetic layer and the second ferromagnetic layer. Each of samples s1 to s4 has the same number of atomic layers in the first intermediate layer, non-magnetic layer, and second intermediate layer, and each of samples s5 to s8 has the same number of atomic layers in the non-magnetic layer. The number of atomic layers in the non-magnetic layer of each of samples s1 to s4 is also equal to the number of atomic layers in the non-magnetic layer of each of samples s5 to s8. The number of atomic layers in each of the first intermediate layer and the second intermediate layer of samples s1 to s4 is one layer. The number of atomic layers is the number of layers in the stacking direction of the magnetoresistance effect element.

Figure 3 shows the simulation results of Experimental Example 1. The vertical axis of Figure 3 is the MR ratio of the magnetoresistance effect element, and the horizontal axis is the resistance value of each sample normalized by the resistance value of sample s6. Samples s1 to s5 and s8, which have an AlN nonmagnetic layer, had magnetoresistance effect element resistances equal to or lower than samples s6 and s7, which have an MgO nonmagnetic layer. Cubic AlN has a low resistance of the nonmagnetic layer, and by applying this nonmagnetic layer to the magnetoresistance effect element, it is expected that the performance of the magnetoresistance effect element will be improved. Although sample s8 has a high resistance of the magnetoresistance effect element, it is difficult to fabricate it as an actual element because it does not have a first intermediate layer.

(Experimental Example 2: Actual magnetoresistance effect element in which the ferromagnetic layer is a CoFe-based ferromagnetic material and the non-magnetic layer is a cubic or tetragonal crystal)
In Experimental Example 2, an actual magnetoresistance effect element was fabricated. On an underlayer made of Cr, CoFe with a thickness of 10 nm, CrN with a thickness of 0.1 nm, cubic AlN with a thickness of 2 nm, CoFe with a thickness of 8 nm, IrMn with a thickness of 10 nm, and Ru with a thickness of 10 nm were successively formed.

The specific configuration of the manufactured magnetoresistance effect element is as follows.
Underlayer: Cr
First ferromagnetic layer: 10 nm thick CoFe (oriented in the (001) crystal plane)
First intermediate layer: 0.1 nm thick CrN
Non-magnetic layer: cubic AlN with a thickness of 2 nm
Second ferromagnetic layer: 8 nm thick CoFe (oriented in the (001) crystal plane)
Antiferromagnetic layer: 10 nm thick IrMn
Cap layer: Ru with a thickness of 10 nm

Figure 4 is a cross-sectional TEM (Transmission Electron Microscope) (JEM-ARM200F manufactured by JEOL) image of the magnetoresistance effect element produced in Experimental Example 2. As shown in Figure 4, the atoms are arranged without disorder near the nonmagnetic layer. Also as shown in Figure 4, spots caused by cubic crystals were confirmed even when nanobeam electron diffraction method was used. In other words, it can be confirmed that the nonmagnetic layer is cubic AlN. Cubic AlN is a metastable phase, and is not usually formed unless a specific underlayer is used. Furthermore, atoms are arranged without disorder in the layers above the nonmagnetic layer, confirming that the layers above the nonmagnetic layer also have a cubic crystal structure.

(Experimental Example 3: Actual magnetoresistance effect element in which the ferromagnetic layer is a CoFe-based ferromagnetic material and the non-magnetic layer is a cubic or tetragonal crystal)
In Experimental Example 3, an actual magnetoresistance effect element was fabricated by changing the film configuration from Experimental Example 2. On a substrate made of MgO, an underlayer made of Cr, a 10 nm thick _Co2FeAl layer, a 0.14 nm thick TiN layer, a 2.5 nm thick cubic AlN layer, an 8 nm thick _Co2FeAl layer, a 10 nm thick IrMn layer, and a 10 nm thick Ru layer were successively formed.

The specific configuration of the manufactured magnetoresistance effect element is as follows.
Underlayer: Cr
First ferromagnetic layer: 10 nm thick _Co2FeAl (oriented in the (001) crystal plane)
First intermediate layer: TiN with a thickness of 0.14 nm
Non-magnetic layer: cubic AlN with a thickness of 2.5 nm
Second ferromagnetic layer: 8 nm thick _Co2FeAl (oriented in the (001) crystal plane)
Antiferromagnetic layer: 10 nm thick IrMn
Cap layer: Ru with a thickness of 10 nm

Figure 5 is a cross-sectional TEM image of the magnetoresistance effect element produced in Experimental Example 3. As shown in Figure 5, it can be confirmed that in Experimental Example 3 as well, the atoms are arranged without disorder near the nonmagnetic layer, and cubic AlN is formed. It can also be confirmed that the atoms are arranged without disorder in the layers above the nonmagnetic layer, and the layers above the nonmagnetic layer also have a cubic crystal structure.

Figure 6 shows the results of X-ray diffraction (Rigaku, SmartLab) of the magnetoresistance effect element produced in Experimental Example 3. As shown in Figure 6, a peak derived from the IrMn (002) crystal plane was confirmed in the X-ray diffraction spectrum of Experimental Example 3. IrMn is located in a layer above the nonmagnetic layer, and the fact that this peak can be confirmed confirms that the nonmagnetic layer and the layer above the nonmagnetic layer also have a cubic crystal structure.

As described above, it has been shown that by adjusting the thickness and composition of the first intermediate layer, a cubic or tetragonal nonmagnetic layer (AlN) can be formed on the first ferromagnetic layer, which would not normally be formed.

(Modifications 1 to 3: When MgO is used in the first intermediate layer)
7 shows the results of X-ray diffraction of the magnetoresistance effect element of the modified example of Experimental Example 3. The top graph in Fig. 7 shows the results of X-ray diffraction of the first modified example, the center graph shows the results of X-ray diffraction of the second modified example, and the bottom graph shows the results of X-ray diffraction of the third modified example.

The first modified example differs from the element configuration of Experimental Example 3 in that the first and second ferromagnetic layers are CoMnFe, the first intermediate layer is 1.5 nm of MgO, and the nonmagnetic layer is 1.0 nm of cubic AlN.

The second modified example differs from the element configuration of Experimental Example 3 in that the first and second ferromagnetic layers are CoFe, the first intermediate layer is 1.0 nm of MgO, and the nonmagnetic layer is 1.5 nm of cubic AlN.

The third modified example differs from the element configuration of experimental example 3 in that the first and second ferromagnetic layers are CoMnFe, there is no first intermediate layer, and the nonmagnetic layer is 2.0 nm of MgO.

In Figure 7, a peak originating from the IrMn (002) crystal plane was confirmed in all of the first to third modifications. In other words, it can be said that the crystallinity of the nonmagnetic layer and the layer stacked on top of the nonmagnetic layer is comparable to that of the nonmagnetic layer made of MgO (third modification) even when the nonmagnetic layer is a nitride film (first and second modifications).

(Modification 4: When MnN is used for the first intermediate layer)
The fourth modified example differs from the element configuration of Experimental Example 3 in that the first ferromagnetic layer and the second ferromagnetic layer are CoMnFe, the first intermediate layer is MnN having a thickness of 0.2 nm to 0.8 nm, and the nonmagnetic layer is cubic AlN having a thickness of 2.0 nm. Note that Mn was deposited when the first intermediate layer was formed. It is believed that nitrogen diffused from the AlN adjacent to the first intermediate layer into the first intermediate layer, causing the first intermediate layer to become MnN.

Figure 8 shows the X-ray diffraction results for the magnetoresistance effect element of Variation 4 of Experimental Example 3. The results for four samples with different film thicknesses for the first intermediate layer are shown in Figure 8. As shown in Figure 8, a peak derived from the IrMn (002) crystal plane was also confirmed in the fourth variation. In other words, it can be confirmed that the nonmagnetic layer and the layer stacked on top of the nonmagnetic layer have a cubic crystal structure.

As described above, it has been shown that by adjusting the thickness and composition of the first intermediate layer, a cubic or tetragonal nonmagnetic layer (AlN) can be formed on the first ferromagnetic layer, which would not normally be formed.

(Experimental Example 4: MR ratio when MgO non-magnetic layer in CoFe-based ferromagnetic material is replaced with cubic AlN)
In Experimental Example 4, the possibility of replacing a part of the non-magnetic layer, which is conventionally made only of MgO, with cubic AlN was examined. From Modifications 1 and 2, the MR ratio was examined when MgO was used as the first intermediate layer and cubic AlN was used as the non-magnetic layer to replace the non-magnetic layer, and a part of MgO was replaced with cubic AlN. The film structure of Experimental Example 4 was the same as that of the above-mentioned first modification. That is, the layer structure of each layer was as follows. The total thickness of the first intermediate layer and the non-magnetic layer was fixed at 2.5 nm.
Underlayer: Cr
First ferromagnetic layer: 10 nm thick CoMnFe (Co 66 at%, Mn 17 at%, Fe 17 at%)
First intermediate layer: MgO ((1) 0.5 nm, (2) 0.7 nm, (3) 1.0 nm, (4) 1.5 nm, (5) 2.0 nm)
Nonmagnetic layer: cubic AlN ((1) 2.0 nm, (2) 1.8 nm, (3) 1.5 nm, (4) 1.0 nm, (5) 0.5 nm)
Second ferromagnetic layer: 8 nm thick CoMnFe
Antiferromagnetic layer: 10 nm thick IrMn
Cap layer: Ru with a thickness of 10 nm

Figure 9 shows the results of Experimental Example 4, and shows the MR ratio of the magnetoresistance effect element when the ratio of the first intermediate layer and the non-magnetic layer is changed. The thicknesses at the five points from left to right in Figure 9 indicate (1) to (5) above. As shown in Figure 9, when the non-magnetic layer 3, which is usually composed of only MgO, is replaced with MgO as the first intermediate layer and cubic AlN as the non-magnetic layer, it was revealed that even if about 50% of the MgO is replaced with cubic AlN, the MR ratio remains almost 100%, and cubic AlN functions as a barrier layer of the magnetic element. It was also shown that even if about 80% of the MgO is replaced with cubic AlN, the MR ratio decreases, but the magnetoresistance effect element still functions. If more MgO can be replaced with cubic AlN, the resistance value can be reduced, so there is a possibility that there will be no problem with the performance of the magnetoresistance effect element 10.

(Experimental Example 5: MR ratio and resistance value)
In the fifth experimental example, including the fourth experimental example, the film configuration of the first intermediate layer, the non-magnetic layer, and the second intermediate layer was changed to obtain the MR ratio and RA (area resistance product) of the magnetoresistance effect element. The RA can be obtained by the product of the resistance R _P when the magnetization directions of the first ferromagnetic layer 1 and the second ferromagnetic layer 2 are parallel and the area in the in-plane direction of the magnetoresistance effect element 10. FIG. 10 shows the MR ratio of the magnetoresistance effect element of the fifth experimental example, and FIG. 11 shows the RA of the magnetoresistance effect element of the fifth experimental example. The horizontal axis of FIG. 10 and FIG. 11 is the film thickness of the first intermediate layer and the second intermediate layer. Also, FIG. 10 and FIG. 11 show the film configuration of the first intermediate layer, the non-magnetic layer, and the second intermediate layer, and in the case of a three-layer structure, the order is the first intermediate layer/non-magnetic layer/second intermediate layer, and in the case of a two-layer structure, the order is the first intermediate layer/non-magnetic layer. The numerical value written after the material of the non-magnetic layer is the thickness of the non-magnetic layer in nm. Note that, for MgO/AlN, an example of Experimental Example 4 is shown in Figures 10 and 11. As shown in Figures 10 and 11, even if the film configuration of the non-magnetic material constituting the first intermediate layer and non-magnetic layer of the magnetoresistance effect element in the CoFe system is changed to something other than MgO, the magnetoresistance change of the magnetoresistance effect element is confirmed, and the magnetoresistance effect element of each film configuration functions properly.

(Experimental Example 6: Characteristics of MnGa-based (ferrimagnetic) magnetic film formed on a nonmagnetic layer)
A sample was produced in which the second ferromagnetic layer 2 on the nonmagnetic layer 3 was a MnGa-based magnetic film, and the magnetic properties were measured. There have been no reports on the magnetoresistance effect of a magnetoresistance effect element in which the second ferromagnetic layer 2 on the nonmagnetic layer 3 is a MnGa-based magnetic film. The nonmagnetic layer 3 is made to have a cubic or tetragonal crystal structure, thereby making the second ferromagnetic layer 2 on the nonmagnetic layer 3 a MnGa-based magnetic film.

In Experimental Example 6, three samples having the following layer configuration were prepared.
Sample MnGa1: Cr/2 nm thick MgO/3.0 nm thick MnGa
Sample MnGa2: Cr/0.8 nm thick MgO/2.5 nm thick cubic AlN/3.0 nm thick MnGa
Sample MnGa3: Cr/2.5 nm thick cubic AlN/3.0 nm thick MnGa

Figure 12 shows the magnetic properties of the samples of Experimental Example 6. The magnetic properties shown in Figure 12 were determined using the surface magnetic Kerr effect, with the vertical axis of Figure 12 representing the rotation of the polarization plane (Kerr rotation angle) and the horizontal axis representing the magnetic field strength applied to the sample. The external magnetic field was applied perpendicular to the surface. Figure 12 (a) shows the results for sample MnGa1, (b) shows the results for sample MnGa2, and (c) shows the results for sample MnGa3.

As shown in FIG. 12(a), no hysteresis loop was observed in the MnGa film formed on the MgO film. That is, the MnGa of sample MnGa1 does not exhibit ferromagnetic properties. In other words, it is presumed that a Mn-based ferrimagnetic material could not be formed on MgO. In contrast, as shown in FIG. 12(b) and (c), the MnGa of sample MnGa2 and sample MnGa3 formed on cubic AlN exhibited a hysteresis loop and ferromagnetic properties. That is, in this disclosure, it has been discovered that a magnetoresistance effect can be obtained by forming a ferrimagnetic material film such as MnGa on a cubic or tetragonal nonmagnetic layer 3.

Figure 13 shows the X-ray diffraction results for the samples of Experimental Example 6. Figure 13 (a) shows the results for sample MnGa1, (b) shows the results for sample MnGa2, and (c) shows the results for sample MnGa3. Peaks derived from the (002) plane of MnGa were confirmed in samples MnGa2 and MnGa3. In contrast, no peaks derived from the (002) plane of MnGa were confirmed in sample MnGa1. This is thought to be because tetragonal MnGa could not be formed on MgO. This result was consistent with the results shown in Figure 12.

(Experimental Example 7: Dependence of Magnetic Properties of MnGa-Based (Ferrimagnetic) Magnetic Film on Film Thickness)
In Experimental Example 7, samples were prepared with a layer structure in which Cr/cubic AlN with a thickness of 2.5 nm/MnGa were stacked in this order from the bottom, and the thickness of the MnGa layer was changed to measure the magnetic properties. Samples of MnGa layers with thicknesses of 3 nm, 5 nm, 10 nm, 20 nm, and 50 nm were prepared. Figure 14 shows the magnetic properties of Experimental Example 7. The magnetic properties shown in Figure 14 were obtained using the surface magnetic Kerr effect, and the vertical axis of Figure 14 represents the rotation of the polarization plane (Kerr rotation angle), and the horizontal axis represents the magnetic field strength applied to the sample.

As shown in Figure 14, a hysteresis loop was observed in all cases. In other words, even when the MnGa layer was thick, the crystal structure of the ferromagnetic layer was maintained, and the ferromagnetic layer exhibited magnetic anisotropy.

(Experimental Example 8: Dependence of the Magnetic Properties of MnGa-Based (Ferrimagnetic) Magnetic Film on the Thickness of the Nonmagnetic Layer)
In Experimental Example 8, samples were prepared with a layer structure in which Cr/MgO/cubic AlN with a thickness of 2.5 nm/MnGa with a thickness of 40 nm were stacked from the bottom, and the thickness of the MgO layer was changed to measure the magnetic properties. Samples with MgO layers having thicknesses of 0 nm, 0.5 nm, 0.8 nm, 1.1 nm, and 2.0 nm were prepared. Figure 15 shows the magnetic properties of Experimental Example 8. The magnetic properties shown in Figure 15 were obtained by utilizing the surface magnetic Kerr effect, and the vertical axis of Figure 15 is the rotation of the polarization plane (Kerr rotation angle), and the horizontal axis is the magnetic field strength applied to the sample.

As shown in Figure 15, samples with MgO layer thicknesses of 1.1 nm and 2.0 nm showed a near-zero value, and no hysteresis loop was observed. On the other hand, a hysteresis loop was observed in samples with MgO layer thicknesses of less than 1.1 nm. In other words, when using a MnGa-based (ferrimagnetic) magnetic film, it was found that the magnetic properties of the ferromagnetic layer decrease as the MgO layer thickness increases, and this is consistent with the results of Experimental Example 4 shown in Figure 8.

(Experimental Example 9: Magnetoresistance effect element using MnGa-based (ferrimagnetic) magnetic film as the second ferromagnetic layer)
In Experimental Example 9, a magnetoresistance effect element having a second ferromagnetic layer of MnGa was actually fabricated and its characteristics were evaluated. On an underlayer made of Cr, a 10 nm thick FeCo layer, a 0.9 nm thick Mg layer, a 0.6 nm thick MgO layer, a 2.0 nm thick cubic AlN layer, a 3 nm thick MnGa layer, and a protective layer were successively formed.

The specific configuration of the manufactured magnetoresistance effect element is as follows.
Underlayer: Cr
First ferromagnetic layer: 10 nm thick FeCo (oriented in the (001) crystal plane)
First intermediate layer: 0.9 nm thick Mg and 0.6 nm thick MgO
Nonmagnetic layer: cubic AlN
Second ferromagnetic layer: 3 nm thick MnGa (oriented in the (001) crystal plane)
Protective layer: Ta/Ru

Fig. 16 shows the results of X-ray diffraction of the magnetoresistance effect element of Experimental Example 9. As shown in Fig. 16, a peak derived from the MnGa ferrimagnetic material having the _L10 structure (tetragonal crystal) was confirmed. That is, Fig. 16 shows that a magnetoresistance effect element having a MnGa ferrimagnetic material on a nonmagnetic layer was successfully fabricated.

Figure 17 shows the magnetic characteristics of the magnetoresistance effect element of Experimental Example 9. The magnetic characteristics shown in Figure 17 were obtained using the surface magnetic Kerr effect, with the vertical axis of Figure 17 representing the rotation of the polarization plane (Kerr rotation angle) and the horizontal axis representing the magnetic field strength applied to the sample. As shown in Figure 17, magnetic characteristics with a hysteresis loop were confirmed. This indicates that the MnGa ferrimagnetic material functions as a second ferromagnetic layer.

(Experimental Example 10: Magnetoresistance effect element with GaN nonmagnetic layer)
In Experimental Example 10, a magnetoresistance effect element was fabricated in which the nonmagnetic layer was made of GaN. On an underlayer made of Cr, a 10 nm thick CoMnFe film, a 0.6 nm thick MgO film, cubic GaN, a 0.6 nm thick MgO film, a 8 nm thick CoMnFe film, a 10 nm thick IrMn film, and a 10 nm thick Ru film were successively formed.

The specific configuration of the manufactured magnetoresistance effect element is as follows.
Underlayer: Cr
First ferromagnetic layer: 10 nm thick CoMnFe (oriented in the (001) crystal plane)
First intermediate layer: MgO having a thickness of 0.6 nm
Nonmagnetic layer: cubic GaN
Second intermediate layer: MgO with a thickness of 0.6 nm
Second ferromagnetic layer: 8 nm thick CoMnFe (oriented in the (001) crystal plane)
Antiferromagnetic layer: 10 nm thick IrMn
Cap layer: Ru with a thickness of 10 nm

In Experimental Example 10, four samples were produced with GaN thicknesses of 0 nm, 0.2 nm, 0.6 nm, and 0.9 nm. When the GaN thickness was 0 nm, the 1.2 nm MgO layer formed by combining the first and second intermediate layers became the nonmagnetic layer.

Figure 18 shows the X-ray diffraction results for the magnetoresistance effect element of Experimental Example 10. As shown in Figure 18, a peak derived from the IrMn (002) crystal plane was also confirmed in the magnetoresistance effect element of Experimental Example 10. In other words, it can be confirmed that the nonmagnetic layer and the layer stacked on top of the nonmagnetic layer have a cubic crystal structure.

FIG. 19 shows the result of calculating the band structure of the GaN nonmagnetic layer. The band structure was obtained by simulation using the first-principles calculation package quantum espresso. The zinc-blende structure was assumed as the crystal structure of the cubic GaN. The calculation conditions for simulating the band structure were the generalized gradient approximation for the correlation exchange potential, 600 eV for the cutoff energy of the plane wave basis, and 0.458 nm obtained from the first-principles calculation for the lattice constant. In FIG. 19, κ _Δ5 indicates the attenuation constant when Δ ₅ band electrons tunnel, κ _Δ1 indicates the attenuation constant when Δ ₁ band electrons tunnel, Δ ₅ indicates an energy band having Δ ₅ symmetry, and Δ ₁ indicates an energy band having Δ ₁ symmetry. The band gap of the MgO nonmagnetic layer is about 8 eV, the band gap of the AlN nonmagnetic layer is about 5 eV, and the band gap of the GaN nonmagnetic layer is about 3 eV. A magnetoresistance effect element using GaN for the nonmagnetic layer has a lower resistance than a magnetoresistance effect element using MgO or AlN for the nonmagnetic layer.

Figure 20 shows the relationship between the thickness of the nonmagnetic layer and element resistance of the magnetoresistance effect element of Experimental Example 10 (with a GaN nonmagnetic layer), and the relationship between the thickness of the nonmagnetic layer and element resistance of a magnetoresistance effect element having the same configuration as Experimental Example 10 except that the nonmagnetic layer is MgO. As with the results obtained from the simulation results above, the magnetoresistance effect element of Experimental Example 10 with a GaN nonmagnetic layer has a lower element resistance than a magnetoresistance effect element with an MgO nonmagnetic layer.

Figure 21 shows the MR ratio of the magnetoresistance effect element of Experimental Example 10. The magnetoresistance effect element shown in Figure 21 is a sample with a GaN thickness of 0.9 nm. The horizontal axis of Figure 21 shows the heat treatment temperature of the magnetoresistance effect element. This heat treatment is annealing performed after the magnetoresistance effect element is laminated. Annealing the magnetoresistance effect element causes atomic rearrangement, improving performance. The vertical axis of Figure 21 is the MR ratio of the magnetoresistance effect element. As shown in Figure 21, the magnetoresistance effect element of Experimental Example 10 shows a magnetoresistance change rate of more than 100%.

From the above, the non-magnetic layer containing cubic or tetragonal nitrogen disclosed herein has low resistance and functions as a tunnel barrier layer. It has been shown that a magnetoresistance effect occurs in an FeCo-based magnetic film, and that a magnetoresistance effect can occur in a ferrimagnetic magnetic film.

Furthermore, by forming a ferrimagnetic material on the non-magnetic layer containing cubic or tetragonal nitrogen of the present disclosure, a magnetoresistance effect can be obtained. This magnetoresistance effect element has the potential to realize high performance and high density magnetoresistance effect elements.

1...first ferromagnetic layer, 2...second ferromagnetic layer, 3...non-magnetic layer, 4...first intermediate layer, 5...second intermediate layer, 9...underlayer, 10, 11...magnetoresistance effect element

Claims (12)

A first ferromagnetic layer;
A second ferromagnetic layer;
a nonmagnetic layer sandwiched between the first ferromagnetic layer and the second ferromagnetic layer and containing a cubic or tetragonal nitride;
a first intermediate layer sandwiched between the first ferromagnetic layer and the nonmagnetic layer and including a cubic or tetragonal nonmagnetic material; The magnetoresistance effect element of claim 1, further comprising a second intermediate layer sandwiched between the second ferromagnetic layer and the nonmagnetic layer and including a cubic or tetragonal nonmagnetic material. Further comprising a base layer;
the first ferromagnetic layer is closer to the underlayer than the second ferromagnetic layer;
2. The magnetoresistance effect element according to claim 1, wherein the second ferromagnetic layer includes a Mn-based ferrimagnetic material. The magnetoresistance effect element of claim 3, wherein the second ferromagnetic layer includes an MnGa alloy. The magnetoresistance effect element according to claim 3, wherein the first ferromagnetic layer includes a Mn-based ferrimagnetic material. The magnetoresistance effect element of claim 1, wherein the nonmagnetic layer contains a nitride of at least one element selected from the group consisting of Al, Ga, and In. The magnetoresistance effect element according to claim 1, wherein the crystal structure of the non-magnetic material of the first intermediate layer is a rock-salt type or a distorted rock-salt type tetragonal crystal. The magnetoresistance effect element according to claim 1, wherein the non-magnetic material of the first intermediate layer is any one selected from the group consisting of Cr, Ti, Mn, CrN, TiN, MgO, and MnN. The magnetoresistance effect element of claim 3, wherein the thickness of the first intermediate layer is less than 1.1 nm. The magnetoresistance effect element according to claim 2, wherein the crystal structure of the non-magnetic material of the second intermediate layer is a rock-salt type or a distorted rock-salt type tetragonal crystal. The magnetoresistance effect element according to claim 2, wherein the non-magnetic material of the second intermediate layer is any one selected from the group consisting of Cr, Ti, Mn, CrN, TiN, MgO, and MnN. The magnetoresistance effect element of claim 2, wherein the thickness of the second intermediate layer is less than 1.1 nm.

Images