JP7400560B2 - Tunnel barrier layer, magnetoresistive element, method for manufacturing tunnel barrier layer, and insulating layer - Google Patents

Description

The present invention relates to a tunnel barrier layer, a magnetoresistive element, a method for manufacturing a tunnel barrier layer, and an insulating layer.

Giant magnetoresistive (GMR) elements consisting of a multilayer film of a ferromagnetic layer and a nonmagnetic layer, and tunnel magnetoresistive (TMR) elements using an insulating layer (tunnel barrier layer, barrier layer) as the nonmagnetic layer are known. There is. Generally, a TMR element has a higher element resistance than a GMR element, but a higher magnetoresistance (MR) ratio. TMR elements are attracting attention as elements for magnetic sensors, high-frequency components, magnetic heads, and nonvolatile random access memories (MRAM).

TMR elements can be classified into two types depending on the mechanism of electron tunnel conduction. One is a TMR element that utilizes only the seepage effect (tunnel effect) of a wave function between ferromagnetic layers. The other is a TMR element in which coherent tunneling (only electrons having a specific wave function symmetry tunnel) that maintains the symmetry of the wave function is dominant. A TMR element in which coherent tunneling is dominant can obtain a larger MR ratio than a TMR element that uses only tunneling effects.

MgO is an example of a tunnel barrier layer in which coherent tunneling occurs. Furthermore, as a material to replace MgO, for example, a ternary oxide consisting of Mg, Al, and O (Mg-Al-O) is being considered. Mg-Al-O has improved lattice matching with ferromagnetism compared to MgO, and its MR ratio is less likely to decrease than conventional MgO even when a high voltage is applied.

For example, Patent Document 1 describes an example in which MgAl2O4 having a spinel type crystal structure is used for a tunnel barrier layer.

Further, Patent Document 2 describes a ternary oxide (Mg-Al-O) having a cubic crystal (irregular spinel structure) having a lattice constant half that of a spinel structure. Since the irregular spinel structure is a metastable structure, the tunnel barrier layer can be constructed without being limited to the stoichiometric composition of the spinel structure. The lattice constant of the irregular spinel structure can be continuously changed by adjusting the Mg--Al composition ratio. Further, Patent Document 2 describes a magnetoresistive element that combines a tunnel barrier layer with an irregular spinel structure and a BCC type Co--Fe based ferromagnetic layer. By combining a tunnel barrier layer with an irregular spinel structure and a BCC type Co--Fe ferromagnetic layer, the band folding effect is suppressed, and the magnetoresistive element stably exhibits a large MR ratio.

Patent No. 5586028 Patent No. 5988019

The tunnel barrier layer having an ordered spinel structure described in Patent Document 1 cannot obtain a sufficiently large MR ratio. Further, the tunnel barrier layer having an irregular spinel structure described in Patent Document 2 tends to have a reduced MR ratio when a high voltage is applied. In order to improve the output voltage of a magnetoresistive element, a large MR ratio and high voltage resistance are required.

In view of these circumstances, it is an object of the present invention to provide a magnetoresistive element that can improve the output voltage.

The present inventors introduced both an ordered spinel structure and an irregular spinel structure into the tunnel barrier layer. The present invention provides the following means to solve the above problems.

(1) The tunnel barrier layer according to the first aspect includes a nonmagnetic oxide, and has a crystal structure including both an ordered spinel structure and a disordered spinel structure.

(2) In the tunnel barrier layer according to the above aspect, the lattice constant of the irregular spinel structure may be approximately half of the lattice constant of the regular spinel structure.

(3) The tunnel barrier layer according to the above aspect has a first part showing the first electron beam pattern, a second part showing the second electron beam pattern, and a pseudo pattern in nanoelectron diffraction using a transmission electron microscope. and a third portion showing one electron beam pattern.

(4) The tunnel barrier layer according to the above aspect may consist only of a third portion that shows a pseudo first electron beam pattern in nanoelectron diffraction using a transmission electron microscope.

(5) In the tunnel barrier layer according to the above aspect, the proportion of the ordered spinel structure may be 10% or more and 90% or less.

(6) In the tunnel barrier layer according to the above aspect, the proportion of the ordered spinel structure may be 20% or more and 80% or less.

(7) In the tunnel barrier layer according to the above aspect, the nonmagnetic oxide may contain Mg and at least one of Al and Ga.

(8) The tunnel barrier layer according to the above aspect includes an oxide containing Mg and Ga, and an oxide containing Mg and Al, and the oxide containing Mg and Ga has the ordered spinel structure. The oxide containing Mg and Al may have the irregular spinel structure.

(9) In the tunnel barrier layer according to the above aspect, the crystal orientation direction may be (001) orientation.

(10) A magnetoresistive element according to a second aspect includes the tunnel barrier layer according to the above aspect, and a first ferromagnetic layer and a second ferromagnetic layer sandwiching the tunnel barrier layer in the thickness direction.

(11) In the magnetoresistive element according to the second aspect, at least one of the first ferromagnetic layer and the second ferromagnetic layer may contain Fe element.

(12) The method for manufacturing a magnetoresistive element according to the third aspect includes a first condition in which sufficient oxygen is supplied to form an ordered spinel structure, and a second condition in which the amount of oxygen supplied is smaller than the first condition. including.

(13) The insulating layer according to the fourth aspect includes a nonmagnetic oxide, and has a crystal structure including both an ordered spinel structure and a disordered spinel structure.

The magnetoresistive element according to one embodiment of the present invention has improved output voltage.

1 is a cross-sectional view of a magnetoresistive element according to an embodiment. This is the result of performing nanoelectron diffraction (NBD) on an ordered spinel structure using a transmission electron microscope (TEM). These are the results of nanoelectron diffraction (NBD) of an irregular spinel structure using a transmission electron microscope (TEM). FIG. 2 is a diagram showing a crystal structure of an ordered spinel structure. FIG. 2 is a diagram showing a crystal structure of an irregular spinel structure. FIG. 3 is a diagram showing an example of a crystal structure of a tunnel barrier layer. FIG. 7 is a diagram showing another example of the crystal structure of a tunnel barrier layer. FIG. 7 is a diagram showing another example of the crystal structure of a tunnel barrier layer. FIG. 2 is a cross-sectional view of a magnetic recording element according to Application Example 1 of a magnetoresistive element. FIG. 3 is a cross-sectional view of a magnetic recording element according to Application Example 2 of a magnetoresistive element.

Hereinafter, the present invention will be described in detail with reference to the drawings as appropriate. In the drawings used in the following explanation, characteristic parts of the present invention may be shown enlarged for convenience in order to make it easier to understand, and the dimensional ratio of each component may differ from the actual one. be. The materials, dimensions, etc. exemplified in the following description are merely examples, and the present invention is not limited thereto, and can be implemented with appropriate changes within the scope of the invention.

"Magnetoresistance effect element"
FIG. 1 is a cross-sectional view of the magnetoresistive element according to this embodiment. FIG. 1 is a cross-sectional view of the magnetoresistive element 10 taken along the lamination direction of each layer of the magnetoresistive element. The magnetoresistive element 10 includes a first ferromagnetic layer 1 , a second ferromagnetic layer 2 , and a tunnel barrier layer 3 . The magnetoresistive element 10 may have a cap layer, a base layer, etc. in addition to these layers. Tunnel barrier layer 3 is an example of an insulating layer.

(First ferromagnetic layer, second ferromagnetic layer)
The first ferromagnetic layer 1 and the second ferromagnetic layer 2 are magnetic materials. The first ferromagnetic layer 1 and the second ferromagnetic layer 2 each have magnetization. The magnetoresistive element 10 outputs a change in the relative angle between the magnetization of the first ferromagnetic layer 1 and the magnetization of the second ferromagnetic layer 2 as a change in resistance value.

The magnetization of the second ferromagnetic layer 2 is, for example, more difficult to move than the magnetization of the first ferromagnetic layer 1. When a predetermined external force is applied, the direction of magnetization of the second ferromagnetic layer 2 does not change (fixed), and the direction of magnetization of the first ferromagnetic layer 1 changes. As the direction of magnetization of the first ferromagnetic layer 1 changes with respect to the direction of magnetization of the second ferromagnetic layer 2, the resistance value of the magnetoresistive element 10 changes. In this case, the second ferromagnetic layer 2 is sometimes called a magnetization fixed layer, and the first ferromagnetic layer 1 is sometimes called a magnetization free layer. Hereinafter, a case where the first ferromagnetic layer 1 is a magnetization free layer and the second ferromagnetic layer 2 is a magnetization fixed layer will be described as an example.

The first ferromagnetic layer 1 and the second ferromagnetic layer 2 contain ferromagnetic material. The ferromagnetic material constituting the first ferromagnetic layer 1 and the second ferromagnetic layer 2 is, for example, a metal selected from the group consisting of Cr, Mn, Co, Fe, and Ni, or an alloy containing one or more of these metals. , an alloy containing these metals and at least one or more elements of B, C, and N. The first ferromagnetic layer 1 and the second ferromagnetic layer 2 contain, for example, Fe element. The first ferromagnetic layer 1 and the second ferromagnetic layer 2 containing the Fe element have high spin polarization, and the MR ratio of the magnetoresistive element 10 increases. The first ferromagnetic layer 1 and the second ferromagnetic layer 2 are, for example, Fe, Co--Fe, Co--Fe-B, or Ni--Fe.

The first ferromagnetic layer 1 and the second ferromagnetic layer 2 may be Heusler alloys. Heusler alloys are half metals and have high spin polarizability. Heusler alloy is an intermetallic compound with a chemical composition of _XYZ or It is an elemental species of Cr, a transition metal of the Ti group, or X, and Z is a typical element of groups III to V. Examples of the Heusler alloy include Co ₂ FeSi, Co ₂ FeGe, Co ₂ FeGa, Co ₂ MnSi, Co ₂ Mn _1-a Fe _a Al _b Si _1-b , Co ₂ FeGe _1-c Ga _c , and the like.

The thickness of the first ferromagnetic layer 1 and the second ferromagnetic layer 2 is, for example, 3 nm or less. When the first ferromagnetic layer 1 and the second ferromagnetic layer 2 are thin, interfacial magnetic anisotropy occurs at the interface between the first ferromagnetic layer 1 and the second ferromagnetic layer 2 and the tunnel barrier layer 3. The magnetization directions of the first ferromagnetic layer 1 and the second ferromagnetic layer 2 are easily oriented perpendicular to the lamination plane.

An antiferromagnetic layer may be provided on the surface of the second ferromagnetic layer 2 opposite to the tunnel barrier layer 3 with a spacer layer interposed therebetween. The second ferromagnetic layer 2, the spacer layer, and the antiferromagnetic layer have a synthetic antiferromagnetic structure (SAF structure). A synthetic antiferromagnetic structure consists of two magnetic layers sandwiching a nonmagnetic layer. Due to the antiferromagnetic coupling between the second ferromagnetic layer 2 and the antiferromagnetic layer, the coercive force of the second ferromagnetic layer 2 becomes larger than that in the case without the antiferromagnetic layer. The antiferromagnetic layer is, for example, IrMn, PtMn, or the like. The spacer layer includes, for example, at least one selected from the group consisting of Ru, Ir, and Rh.

(Tunnel barrier layer)
Tunnel barrier layer 3 includes nonmagnetic oxide. The nonmagnetic oxide is, for example, an oxide containing Mg and at least one of Al and Ga. The nonmagnetic oxide is expressed as, for example, Mg-(Al, Ga)-O. Since the ratio of Mg to Al or Ga is not determined, the composition of nonmagnetic oxides is often written as described above without using subscripts.

The tunnel barrier layer 3 has a regular spinel structure and an irregular spinel structure as crystal structures. The compositions of the oxides constituting the ordered spinel structure and the oxides constituting the disordered spinel structure may be the same or different. For example, an oxide constituting an ordered spinel structure is an oxide containing Mg and Ga (Mg-Ga-O), and an oxide constituting an irregular spinel structure is an oxide containing Mg and Al (Mg-Ga-O). Al--O) may also be used.

FIG. 2 shows the results of nanoelectron diffraction (NBD) performed on a crystal with an ordered spinel structure using a transmission electron microscope (TEM). Further, FIG. 3 shows the results of nanoelectron diffraction (NBD) performed on a crystal with an irregular spinel structure using a transmission electron microscope (TEM). Nanoelectron diffraction measures electronic patterns. The electron beam pattern is an electron image obtained by irradiating a thin sample obtained by cutting the magnetoresistive element 10 into thin pieces with an electron beam focused to a diameter of about 1 nm, and transmitting and diffracting the electron beam. 2 and 3 show the results of an electron beam incident on a thin sample in the [100] direction. Hereinafter, the electron beam pattern shown in FIG. 2 will be referred to as a first electron beam pattern, and the electron beam pattern shown in FIG. 3 will be referred to as a second electron beam pattern.

The white parts shown in FIGS. 2 and 3 are diffraction spots where diffracted light was detected, and are regularly arranged. Differences in the size of the diffraction spots are due to the distance from the central axis of the incident light, the ease of diffraction of atoms serving as diffraction points, and other factors. The first electron beam pattern (FIG. 2) and the second electron beam pattern (FIG. 3) are different. Compared to the second electron beam pattern (FIG. 3), the first electron beam pattern (FIG. 2) has spots S1, S2, S3, and S4 (hereinafter collectively referred to as first order (sometimes referred to as spots) are increasing. In the first electron beam pattern, the spots S1 to S4 have equal brightness. Depending on the electron beam pattern, the brightness of the spots S1 to S4 may differ or the spots may not be visible among the spots S1 to S4. An electron beam pattern in which the spots S1 to S4 have different luminances and an electron beam pattern in which there are invisible spots are referred to as pseudo first electron beam patterns.

The electron beam pattern can be regarded as a Fourier transform of a crystal lattice, and changes in effective lattice constants and crystal symmetry are observed. The difference in the electron beam pattern between the ordered spinel structure and the disordered spinel structure is due to the difference in the crystal structure between the ordered spinel structure and the disordered spinel structure.

FIG. 4 is a diagram showing a crystal structure of an ordered spinel structure. In the ordered spinel structure, the site where element A is ionized and enters and the site where element B is ionized and enter are fixed, and the arrangement of these elements is regular. The A element is, for example, one or more elements selected from the group consisting of Mg and Zn, and the B element is one or more elements selected from the group consisting of Al, In, and Ga. The A element is, for example, Mg, and the B element is, for example, Al or Ga. An oxide of Mg and Ga (Mg-Ga-O) tends to have an ordered spinel structure.

FIG. 5 is a diagram showing a crystal structure of an irregular spinel structure. In the case of a disordered spinel structure, the ionized A element or B element can be present at either a site that coordinates tetrahedrally or a site that coordinates octahedrally with respect to oxygen. Which site the A element or B element enters is random. Element A and element B, which have different atomic radii, randomly enter these sites, making the crystal structure disordered. The irregular spinel structure has, for example, a space group symmetry of Fm-3m or a space group symmetry of F-43m. An oxide of Mg and Al (Mg-Al-O) tends to have an irregular spinel structure.

The effective lattice constant (a/2) of the irregular spinel structure is approximately half the effective lattice constant (a) of the ordered spinel structure. For example, the substantial lattice constant of a crystal with an ordered spinel structure is 0.808 nm, and the substantial lattice constant of a crystal with a disordered spinel structure is 0.404 nm. Here, the term "substantive lattice constant" refers to a lattice constant that is allowed when crystallized as spinel or disordered spinel, although the lattice constant may vary somewhat due to oxygen vacancies and the like. Here, "substantial" refers to the amount of lattice misalignment that does not cause loss of crystallinity, and includes a misalignment of about 3% based on the value of the above-mentioned lattice constant. Moreover, "approximately half" includes a deviation of 4% around the half value of the lattice constant (a) of the regularized spinel structure.

The difference in effective lattice constant between an ordered spinel structure and an irregular spinel structure causes a difference in electron beam pattern between an ordered spinel structure and an irregular spinel structure. The second electron beam pattern (FIG. 3) observes fundamental reflections from a face- centered cubic (FCC) lattice. In the first electron beam pattern (FIG. 2), in addition to the fundamental reflection from the face-centered cubic (FCC) lattice, reflection from the {220} plane is observed. The reflection from the {220} plane corresponds to the first order spot. When an ordered spinel structure and an irregular spinel structure coexist, the electron beam diffraction intensity becomes weaker, and the brightness of an ordered spot becomes weaker only in part.

The thickness of the tunnel barrier layer 3 is, for example, 3 nm or less. When the thickness of the tunnel barrier layer 3 is set to 3 nm or less, the wave functions of the first ferromagnetic layer 1 and the second ferromagnetic layer 2 tend to overlap beyond the tunnel barrier layer 3, resulting in a tunneling effect and a wave function of the wave function between the ferromagnetic layers. It becomes easier to obtain a coherent tunnel effect.

6 and 7 are diagrams showing an example of the crystal structure of the tunnel barrier layer 3. As described above, the tunnel barrier layer 3 has a thickness of, for example, 3 nm or less, and in the case of a regular spinel structure with a lattice constant of 0.808 nm, for example, the tunnel barrier layer 3 consists of several layers of unit lattices. FIG. 6 shows a case where the tunnel barrier layer 3 has one layer of unit cells. FIG. 7 shows a case where the unit cell forming the tunnel barrier layer 3 is a plurality of layers.

The tunnel barrier layer 3 shown in FIG. 6 has a regular spinel structure C1 and an irregular spinel structure C2. In the tunnel barrier layer 3 shown in FIG. 6, a region consisting only of the regular spinel structure C1 is referred to as a first region R1, and a region consisting only of the irregular spinel structure C2 is referred to as a second region R2. The first region R1 is an example of the first portion, and the second region R2 is an example of the second portion. The first region R1 shows a first electron beam pattern in nanoelectron diffraction (NBD) using a transmission electron microscope (TEM). The second region R2 shows a second electron beam pattern in nanoelectron diffraction (NBD) using a transmission electron microscope (TEM).

Further, the tunnel barrier layer 3 shown in FIG. 7 also has a regular spinel structure C1 and an irregular spinel structure C2. In the tunnel barrier layer 3 shown in FIG. 7, a region consisting only of the ordered spinel structure C1 in the stacking direction is called a first region R1, a region consisting only of the irregular spinel structure C2 in the stacking direction is called a second region R2, and a region consisting only of the ordered spinel structure C1 in the stacking direction is called a second region R2. The region where the spinel structure C1 and the irregular spinel structure C2 are stacked in this order is called a third region R3, and the region where the irregular spinel structure C2 and the regular spinel structure C1 are stacked in that order is called a fourth region R4. Further, a region in which the ordered spinel structure C1, irregular spinel structure C2, and ordered spinel structure C1 are stacked in this order is referred to as a fifth region R5, and the region in which the ordered spinel structure C2, ordered spinel structure C1, and irregular spinel structure C2 are stacked in this order. The layered region is referred to as a sixth region R6. The third region R3, the fourth region R4, the fifth region R5, and the sixth region R6 are examples of the third portion. Since crystals grow in the stacking direction, the probability that the third region R3, fourth region R4, fifth region R5, and sixth region R6 in which different crystal structures are stacked is the same as that of the first region R1 and the second region R2. It is low compared to the probability of occurrence. The third region R3, the fourth region R4, the fifth region R5, and the sixth region R6 are mixed crystals containing an ordered spinel structure and an irregular spinel structure. Note that the tunnel barrier layer 3 may consist of a first region R1 and at least one region selected from a third region R3, a fourth region R4, a fifth region R5, and a sixth region R6. It may consist of the second region R2 and at least one region selected from the third region R3, the fourth region R4, the fifth region R5, and the sixth region R6. Further, as shown in another example in FIG. 8, the tunnel barrier layer 3 may include at least one region selected from a third region R3, a fourth region R4, a fifth region R5, and a sixth region R6. good.

The first region R1 shows a first electron beam pattern in nanoelectron diffraction (NBD) using a transmission electron microscope (TEM). The second region R2 shows a second electron beam pattern in nanoelectron diffraction (NBD) using a transmission electron microscope (TEM). The third region R3, the fourth region R4, the fifth region R5, and the sixth region R6 show a pseudo first electron beam pattern in nanoelectron diffraction (NBD) using a transmission electron microscope (TEM). The first region R1 is an example of the first portion. The second region R2 is an example of the second portion. The third region R3, the fourth region R4, the fifth region R5, and the sixth region R6 are examples of the third portion.

The first order spots in the first electron beam pattern (spots surrounded by dotted lines in FIG. 2) are generated due to the ordered spinel structure whose effective lattice constant is approximately twice that of the irregular spinel structure. In the first portion, a first order spot is generated in the electron beam pattern, resulting in a first electron beam pattern. The third region R3, the fourth region R4, the fifth region R5, and the sixth region R6 include the ordered spinel structure C1, and thus represent ordered spots. However, the third region R3, the fourth region R4, the fifth region R5, and the sixth region R6 have a pseudo first electron beam pattern because the peak intensity differs depending on the position.

The first region R1 consisting only of a regular spinel structure exhibits a first order spot with high brightness and uniformity, compared to the third region R3 and the fourth region R4, which partially have a regular spinel structure. As mentioned above, the spots obtained by irradiating the third region R3, the fourth region R4, the fifth region R5, and the sixth region R6 with the electron beam are the same as the spots obtained by irradiating the first region R1 with the electron beam. It may be thinner than the order spot (the electron beam diffraction intensity is weak), or a part of the first order spot may not be seen.

The proportion of the ordered spinel structure in the tunnel barrier layer 3 is preferably 10% or more and 90% or less, more preferably 20% or more and 80% or less. The proportion of the irregular spinel structure from which the second electron beam pattern can be obtained in the tunnel barrier layer 3 is preferably greater than 0% and less than 10%, or greater than 90% and less than 100%, and more preferably greater than 0% and less than 20%. or less than or equal to or greater than 80% and less than 100%.

The proportion of the regular spinel structure in the tunnel barrier layer 3 is determined as follows. First, a thin sample of the magnetoresistive element 10 is prepared by cutting the magnetoresistive element 10 along a plane along the stacking direction using a focused ion beam. Next, ten different locations on the tunnel barrier layer 3 of the thin sample are irradiated with an electron beam using a transmission electron microscope (TEM), and nanoelectron diffraction (NBD) is performed at each location. The 10 locations to which the electron beam is irradiated are, for example, the positions of the tunnel barrier layer 3 divided into 11 sections equally spaced in one direction within the plane, and between adjacent sections. Then, the electron beam patterns obtained from each of the ten locations are divided into a first electron beam pattern, a pseudo first electron beam pattern, and a second electron beam pattern based on the presence or absence of spots. Next, the brightness of spots S1 to S4 is measured for the electron beam pattern obtained from each of the 10 locations. For each of the 10 samples, the average brightness (brightness average) of the spots S1 to S4 is determined, assuming that the brightness of the spot with the highest brightness among the spots S1 to S4 is 100. In each of the ten samples, the average brightness of the four spots S1 to S4 is the proportion of the ordered spinel structure of the sample. Further, the average ratio of the ordered spinel structure in the ten electron beam patterns is the ratio of the ordered spinel structure in the tunnel barrier layer 3.

Further, the tunnel barrier layer 3 is preferably (001) oriented. The tunnel barrier layer 3 may be a (001) oriented single crystal or a polycrystalline material mainly containing (001) oriented crystals. When the tunnel barrier layer 3 is (001) oriented, lattice matching with the first ferromagnetic layer 1 and the second ferromagnetic layer 2 increases, making it easier to obtain a coherent tunnel effect. In particular, lattice matching is enhanced when the first ferromagnetic layer 1 or the second ferromagnetic layer 2 contains Fe element, such as Fe, Co--Fe, Co-based Heusler alloy.

The magnetoresistive element 10 is, for example, columnar. The magnetoresistive element 10 has, for example, a circular shape, an elliptical shape, a quadrangular shape, a triangular shape, or a polygonal shape when viewed in plan from the stacking direction. The shape of the magnetoresistive element 10 in plan view from the stacking direction is preferably circular or elliptical from the viewpoint of symmetry.

The length of the long side of the magnetoresistive element 10 in plan view from the stacking direction is preferably 80 nm or less, more preferably 60 nm or less, and even more preferably 30 nm or less. When the length of the long side is 80 nm or less, it becomes difficult to form a domain structure in the ferromagnetism, and there is no need to consider a component different from the spin polarization in the ferromagnetic metal layer. Furthermore, when the length of the long side is 30 nm or less, a single domain structure is formed in the ferromagnetic layer, and the magnetization reversal speed and probability are improved. In addition, there is a strong demand for reduced resistance of magnetoresistive elements that have been miniaturized.

Next, a method for manufacturing the magnetoresistive element 10 according to this embodiment will be explained. The magnetoresistive element 10 is obtained by laminating a first ferromagnetic layer 1, a tunnel barrier layer 3, and a second ferromagnetic layer 2 in this order. The method for forming each layer is, for example, a sputtering method, a vapor deposition method, a laser ablation method, or a molecular beam epitaxial (MBE) method.

The tunnel barrier layer 3 includes film formation conditions of a first condition and a second condition. For example, the tunnel barrier layer 3 is formed under the first condition and then under the second condition. Further, the tunnel barrier layer 3 may be formed under the first condition after being formed under the second condition, for example.

The first condition is a film forming condition in which enough oxygen is supplied to the layer that will become the tunnel barrier layer 3 to form an ordered spinel structure. For example, the tunnel barrier layer 3 can be obtained by laminating an alloy and sufficiently oxidizing the alloy. The oxidation is, for example, plasma oxidation or oxidation by introducing oxygen. The tunnel barrier layer 3 becomes an ordered spinel structure when oxygen is sufficiently supplied.

Whether the ionized elements A and B enter a site that coordinates tetrahedrally or octahedrally with respect to oxygen is largely influenced by the energy potential. When sufficient oxygen is supplied to the tunnel barrier layer 3, from the viewpoint of energy potential, sites where each of the ionized elements A and B are stabilized are fixed, and the ordered spinel structure becomes stable.

The amount of oxygen sufficient to form an ordered spinel structure can be determined from the amount of oxygen theoretically taken in from the compositional formula. For example, if oxygen is supplied into the oxidation treatment chamber in excess of the amount of oxygen that can be theoretically taken in based on the compositional formula, enough oxygen to form an ordered spinel structure will be taken into the layer that will become the tunnel barrier layer 3. The amount of oxygen taken into the layer that will become the tunnel barrier layer 3 can be freely controlled by, for example, adjusting the flow rate of oxygen introduced into the oxidation chamber, the pressure of the oxidation chamber, the oxidation time, and the amount of oxygen relative to the composition of the target alloy. can. For example, under the first condition, the pressure in the oxidation treatment chamber is set to 100 Pa.

The second condition is a condition in which the amount of oxygen supplied to the layer that will become the tunnel barrier layer 3 is smaller than the first condition. The tunnel barrier layer 3 becomes a disordered spinel structure when the amount of oxygen supplied is insufficient.

When the tunnel barrier layer 3 is deficient in oxygen, oxygen is one of the elements responsible for the crystal lattice, and the crystal structure is disturbed. When the crystal structure is disturbed, the energy states at sites that coordinate tetrahedrally and octahedrally with respect to oxygen are also disturbed. When the energy state is disturbed, element A, which should be stabilized at a site that coordinates tetrahedrally to oxygen, becomes stabilized at a site that coordinates octahedrally to oxygen, or vice versa. Therefore, which site the A element and the B element enter is entirely random, and as a result, it becomes easier to stabilize the spinel structure with a more disordered structure.

In the second condition, for example, the flow rate of oxygen introduced into the oxidation treatment chamber is reduced compared to the first condition. As another example, the second condition makes the pressure in the oxidation treatment chamber lower than the first condition. As another example, the second condition makes the oxidation time shorter than the first condition. Another example of the second condition is to reduce the amount of oxygen relative to the composition of the target alloy. For example, in the second condition, the pressure in the oxidation treatment chamber is set to 1 Pa.

Since the tunnel barrier layer 3 is thin, the regular spinel structure and the irregular spinel structure coexist within the tunnel barrier layer 3 even if the first condition and the second condition are performed in order. The tunnel barrier layer 3 may be heated after being formed.

Next, an application example of the magnetoresistive element 10 according to this embodiment will be explained. The magnetoresistive element 10 can be used, for example, as a magnetic sensor, a memory such as an MRAM, and the like.

FIG. 9 is a cross-sectional view of the magnetic recording element 100 according to Application Example 1. FIG. 9 is a cross-sectional view of the magnetoresistive element 10 taken along the lamination direction of each layer of the magnetoresistive element. A magnetic recording element 100 shown in FIG. 9 is an example of an application of the magnetoresistive element 10.

The magnetic recording element 100 includes a magnetoresistive element 10 , a first electrode 11 , a second electrode 12 , a power source 13 , and a measuring section 14 . The first electrode 11 is connected to the first surface of the magnetoresistive element 10 in the stacking direction. The second electrode 12 is connected to the second surface of the magnetoresistive element 10 in the stacking direction. The first electrode 11 and the second electrode 12 are conductors, and are made of, for example, Cu. The power source 13 and the measuring section 14 are connected to the first electrode 11 and the second electrode 12, respectively. The power supply 13 applies a potential difference in the stacking direction of the magnetoresistive element 10. The measurement unit 14 measures the resistance value of the magnetoresistive element 10 in the stacking direction.

When a potential difference is generated between the first electrode 11 and the second electrode 12 by the power source 13, a current flows in the stacking direction of the magnetoresistive element 10. The current becomes spin-polarized when passing through the second ferromagnetic layer 2, and becomes a spin-polarized current. The spin-polarized current reaches the first ferromagnetic layer 1 via the tunnel barrier layer 3. The magnetization of the first ferromagnetic layer 1 is reversed in response to spin transfer torque (STT) caused by a spin-polarized current. By changing the magnetization direction of the first ferromagnetic layer 1 and the magnetization direction of the second ferromagnetic layer 2, the resistance value of the magnetoresistive element 10 in the stacking direction changes. The resistance value of the magnetoresistive element 10 in the stacking direction is read out by the measurement unit 14. That is, the magnetic recording element 100 shown in FIG. 9 is a spin transfer torque (STT) type magnetic recording element.

FIG. 10 is a cross-sectional view of a magnetic recording element 101 according to Application Example 2. FIG. 10 is a cross-sectional view of the magnetoresistive element 10 taken along the stacking direction of each layer of the magnetoresistive element. A magnetic recording element 101 shown in FIG. 10 is an example of an application of the magnetoresistive element 10.

The magnetic recording element 101 includes a magnetoresistive element 10 , a first electrode 21 , a first wiring 22 , a power source 23 , and a measuring section 24 . The first electrode 21 is connected to the first surface of the magnetoresistive element 10 in the stacking direction. The first wiring 22 is connected to the second surface of the magnetoresistive element 10 in the stacking direction. The first electrode 21 is a conductor, for example, Cu. The first wiring 22 includes any one of a metal, an alloy, an intermetallic compound, a metal boride, a metal carbide, a metal silicide, and a metal phosphide, which have the function of generating a spin current by the spin Hall effect when current flows. . The first wiring 22 is, for example, a nonmagnetic metal having an atomic number of 39 or more and having d electrons or f electrons in its outermost shell. The power supply 23 is connected to the first end and the second end of the first wiring 22 . The measuring section 24 is connected to the first electrode 21 and the first wiring 22 and measures the resistance value of the magnetoresistive element 10 in the stacking direction.

When a potential difference is generated between the first end and the second end of the first wiring 22 by the power supply 23, a current flows along the first wiring 22. When a current flows along the first wiring 22, a spin Hall effect occurs due to spin-orbit interaction. The spin Hall effect is a phenomenon in which moving spins are bent in a direction perpendicular to the direction of current flow. The spin Hall effect produces uneven distribution of spins within the first interconnect 22 and induces a spin current in the thickness direction of the first interconnect 22. Spin is injected from the first wiring 22 into the first ferromagnetic layer 1 by a spin current.

The spins injected into the first ferromagnetic layer 1 give spin-orbit torque (SOT) to the magnetization of the first ferromagnetic layer 1 . The first ferromagnetic layer 1 receives spin-orbit torque (SOT) and undergoes magnetization reversal. By changing the magnetization direction of the first ferromagnetic layer 1 and the magnetization direction of the second ferromagnetic layer 2, the resistance value of the magnetoresistive element 10 in the stacking direction changes. The resistance value of the magnetoresistive element 10 in the stacking direction is read out by the measurement unit 14. That is, the magnetic recording element 101 shown in FIG. 10 is a spin-orbit torque (SOT) type magnetic recording element.

The magnetoresistive element 10 according to this embodiment can output a larger output voltage than, for example, when the tunnel barrier layer 3 is made of only an ordered spinel structure. In order to improve the output voltage of a magnetoresistive element, a large MR ratio and high voltage resistance are required. Since the crystal structure of the ordered spinel structure is more stable than that of the irregular spinel structure, the MR ratio is less likely to decrease even when a large voltage is applied. On the other hand, a disordered spinel structure can increase the MR ratio compared to an ordered spinel structure. This is because the irregular spinel structure has an effective lattice constant halved compared to the regular spinel structure, and can easily improve lattice matching with the first ferromagnetic layer 1 and the second ferromagnetic layer 2. Since the magnetoresistive element 10 according to the present embodiment has both an ordered spinel structure and an irregular spinel structure, it is possible to achieve both the characteristics of each structure. Therefore, the magnetoresistive element 10 according to this embodiment has a large MR ratio and high voltage tolerance, and the output voltage is improved.

(Example 1)
A magnetoresistive element 10 shown in FIG. 1 was fabricated on an MgO (001) single crystal substrate. First, 40 nm of Cr was laminated as an underlayer on a substrate, and heat treatment was performed at 800° C. for 1 hour. As the first ferromagnetic layer 1, Fe was laminated to a thickness of 30 nm and heat treated at 300° C. for 15 minutes.

Next, a 0.5 nm film of an alloy expressed as Mg _0.33 Al _0.67 was formed on the first ferromagnetic layer 1, and natural oxidation was performed. Natural oxidation was performed by exposing the sample to air at a pressure of 100 Pa for 600 seconds (first condition). Thereafter, a film of 0.5 nm of an alloy expressed as Mg _0.33 Al _0.67 was formed, and natural oxidation was performed by exposing it to air at a pressure of 1 Pa for 600 seconds (second condition). Thereafter, the laminated film was heat-treated at 400° C. for 15 minutes in vacuum to obtain tunnel barrier layer 3 (Mg--Al--O layer).

Next, 6 nm of Fe was laminated as the second ferromagnetic layer 2 on the tunnel barrier layer 3 and heat treated at 350° C. for 15 minutes to obtain a ferromagnetic tunnel junction. Next, a 12 nm thick IrMn film was formed as an antiferromagnetic layer, and a 20 nm thick Ru film was formed as a cap layer, to obtain the magnetoresistive element 10. Finally, heat treatment was performed at a temperature of 175° C. for 30 minutes while applying a magnetic field of 5 kOe to impart uniaxial magnetic anisotropy to the second ferromagnetic layer 2. The magnetoresistive element 10 had a cylindrical shape with a diameter of 80 nm.

The magnetoresistive element 10 produced using a focused ion beam was cut along a plane along the stacking direction to produce a thin sample of the tunnel barrier layer 3. Then, nanoelectron diffraction (NBD) of the thin sample was performed using a transmission electron microscope (TEM). Specifically, a thin sample was irradiated with an electron beam focused to a diameter of about 1 nm, and the electron beam pattern resulting from transmission diffraction was measured. Next, the brightness of the spot was measured using image analysis software for the electron beam pattern in which the spot was observed. Ten locations on the tunnel barrier layer 3 were irradiated with the electron beam. The 10 locations irradiated with the electron beam correspond to the positions of the tunnel barrier layer 3 divided into 11 sections equally spaced in one direction within the plane, and between adjacent sections. The electron beam was incident in the Mg-Al-O [100] direction.

Five of the electron beam patterns measured at each of the ten locations were the first electron beam pattern, and the remaining five were the second electron beam patterns. The proportions of the ordered spinel structure and the irregular spinel structure in the tunnel barrier layer 3 were 50%, respectively.

(Examples 2 to 5)
Examples 2 to 5 differ from Example 1 in that the manufacturing conditions for the tunnel barrier layer 3 were changed from Example 1. Electron beam patterns were measured at 10 locations under the same conditions.

In Example 2, an alloy expressed as Mg _0.33 Al _0.67 was naturally oxidized under the first condition to a thickness of 0.1 nm, and an alloy expressed as Mg _0.33 Al _0.67 was naturally oxidized under the second condition. This example differs from Example 1 in that the film thickness was 0.9 nm. One of the electron beam patterns measured at each of the 10 locations was the first electron beam pattern, and the remaining nine were the second electron beam patterns. In the tunnel barrier layer 3, the proportion occupied by the ordered spinel structure was 10%, and the proportion occupied by the irregular spinel structure was 90%.

In Example 3, an alloy expressed as Mg _0.33 Al _0.67 was naturally oxidized under the first condition to a thickness of 0.2 nm, and an alloy expressed as Mg _0.33 Al _0.67 was naturally oxidized under the second condition. This example differs from Example 1 in that the film thickness was 0.8 nm. Two of the electron beam patterns measured at each of the ten locations were the first electron beam pattern, and the remaining eight were the second electron beam patterns. In the tunnel barrier layer 3, the proportion occupied by the ordered spinel structure was 20%, and the proportion occupied by the irregular spinel structure was 80%.

In Example 4, an alloy expressed as Mg _0.33 Al _0.67 which was naturally oxidized under the first condition was 0.8 nm, and an alloy expressed as Mg _0.33 Al _0.67 was naturally oxidized under the second condition. This example differs from Example 1 in that the film thickness was 0.2 nm. Eight of the electron beam patterns measured at each of the ten locations were the first electron beam pattern, and the remaining two were the second electron beam patterns. In the tunnel barrier layer 3, the proportion occupied by the ordered spinel structure was 80%, and the proportion occupied by the irregular spinel structure was 20%.

In Example 5, an alloy expressed as Mg _0.33 Al _0.67 which was naturally oxidized under the first condition was 0.9 nm, and an alloy expressed as Mg _0.33 Al _0.67 was naturally oxidized under the second condition. This example differs from Example 1 in that the film thickness was 0.1 nm. Nine of the electron beam patterns measured at each of the ten locations were the first electron beam pattern, and the remaining one was the second electron beam pattern. In the tunnel barrier layer 3, the proportion occupied by the ordered spinel structure was 90%, and the proportion occupied by the irregular spinel structure was 10%.

(Comparative example 1)
In Comparative Example 1, when stacking the tunnel barrier layer 3, a 1.0 nm thick film of an alloy expressed as Mg _0.33 Al _0.67 was formed and natural oxidation was performed. Natural oxidation was performed by exposing the sample to air at a pressure of 1 Pa for 600 seconds. Comparative Example 1 differs from Example 1 in that alloy film formation and natural oxidation were performed at one time. All of the electron beam patterns measured at each of the 10 locations were the second electron beam pattern. In the tunnel barrier layer 3, the proportion occupied by the ordered spinel structure was 0%, and the proportion occupied by the irregular spinel structure was 100%.

(Comparative example 2)
In Comparative Example 2, when stacking the tunnel barrier layer 3, a 1.0 nm thick film of an alloy expressed as Mg _0.33 Al _0.67 was formed and natural oxidation was performed. Natural oxidation was performed by exposing the sample to air at a pressure of 100 Pa for 600 seconds. Comparative Example 1 differs from Example 1 in that alloy film formation and natural oxidation were performed at one time. All of the electron beam patterns measured at each of the 10 locations were the first electron beam pattern. In the tunnel barrier layer 3, the proportion occupied by the ordered spinel structure was 100%, and the proportion occupied by the irregular spinel structure was 0%.

The MR ratios of the magnetoresistive elements of Examples 1 to 5 and Comparative Examples 1 and 2 were measured. The measurement was performed at 300K (room temperature). The output voltages of the magnetoresistive elements of Examples 1 to 5 and Comparative Examples 1 and 2 were also determined. The MR ratio and output voltage were determined based on the following equations. The output voltage was determined based on the MR ratio and the applied voltage when a voltage (V _half ) that halved the MR ratio was applied.
“MR ratio (%)” = (R _AP - R _P )/R _P ×100
"Output voltage" = 1/2V _bias (TMR(V)/1+TMR(V))
R _P is the resistance when the magnetization directions of the first ferromagnetic layer 1 and the second ferromagnetic layer 2 are parallel, and R _AP is the resistance when the magnetization directions of the first ferromagnetic layer 1 and the second ferromagnetic layer 2 are parallel. This is the resistance in the antiparallel case. TMR is (R _AP - R _P )/R _P.
V _bias is an applied voltage, and was set to a voltage at which the MR ratio was halved (V _half ). TMR (V) is the MR ratio when voltage is applied. The voltage characteristics were determined by first measuring the MR ratio when a low voltage of 1 mV was applied, and increasing the applied voltage. The results are summarized in Table 1 below.

As can be seen from Table 1, the magnetoresistive elements shown in Examples 1 to 5 had better output voltages than the magnetoresistive elements shown in Comparative Examples 1 and 2.

1 First ferromagnetic layer 2 Second ferromagnetic layer 3 Tunnel barrier layer 10 Magnetoresistive elements 11, 21 First electrode 12 Second electrode 22 First wiring 13, 23 Power supply 14, 24 Measuring section

Claims (15)

Contains non-magnetic oxides,
The crystal structure includes both an ordered spinel structure and an irregular spinel structure,
In nanoelectron diffraction using a transmission electron microscope,
A tunnel barrier layer including at least two of a first portion showing a first electron beam pattern, a second portion showing a second electron beam pattern, and a third portion showing a pseudo first electron beam pattern. Contains non-magnetic oxides,
The crystal structure includes both an ordered spinel structure and an irregular spinel structure,
A tunnel barrier layer consisting of only a third portion showing a pseudo first electron beam pattern in nanoelectron diffraction using a transmission electron microscope . Contains non-magnetic oxides,
The crystal structure includes both an ordered spinel structure and an irregular spinel structure,
It has an oxide containing Mg and Ga and an oxide containing Mg and Al,
The oxide containing Mg and Ga has the ordered spinel structure,
The oxide containing Mg and Al is a tunnel barrier layer having the irregular spinel structure . 4. The tunnel barrier layer according to claim 1 , wherein the lattice constant of the irregular spinel structure is approximately half the lattice constant of the ordered spinel structure. The tunnel barrier layer according to any one of claims 1 to 4, wherein the proportion of the ordered spinel structure is 10% or more and 90% or less. The tunnel barrier layer according to any one of claims 1 to 4, wherein the proportion of the ordered spinel structure is 20% or more and 80% or less. The tunnel barrier layer according to any one of claims 1 to 6, wherein the nonmagnetic oxide contains Mg and at least one of Al and Ga. It has an oxide containing Mg and Ga and an oxide containing Mg and Al,
The oxide containing Mg and Ga has the ordered spinel structure,
The tunnel barrier layer according to claim 1 or 2 , wherein the oxide containing Mg and Al has the irregular spinel structure. The tunnel barrier layer according to any one of claims 1 to 8, wherein the crystal orientation direction is (001) orientation. A tunnel barrier layer according to any one of claims 1 to 9,
a first ferromagnetic layer and a second ferromagnetic layer sandwiching the tunnel barrier layer in the thickness direction;
A magnetoresistance effect element including. The magnetoresistive element according to claim 10, wherein at least one of the first ferromagnetic layer and the second ferromagnetic layer contains Fe element. The tunnel according to any one of claims 1 to 9 , comprising a first condition in which sufficient oxygen is supplied to form an ordered spinel structure, and a second condition in which the amount of oxygen supplied is smaller than the first condition. A tunnel barrier layer manufacturing method for manufacturing a barrier layer. Contains non-magnetic oxides,
The crystal structure includes both an ordered spinel structure and an irregular spinel structure,
In nanoelectron diffraction using a transmission electron microscope,
An insulating layer including at least two of a first portion showing a first electron beam pattern, a second portion showing a second electron beam pattern, and a third portion showing a pseudo first electron beam pattern. Contains non-magnetic oxides,
The crystal structure includes both an ordered spinel structure and an irregular spinel structure,
An insulating layer consisting of only a third portion exhibiting a pseudo first electron beam pattern in nanoelectron diffraction using a transmission electron microscope . Contains non-magnetic oxides,
The crystal structure includes both an ordered spinel structure and an irregular spinel structure,
It has an oxide containing Mg and Ga and an oxide containing Mg and Al,
The oxide containing Mg and Ga has the ordered spinel structure,
The oxide containing Mg and Al has the irregular spinel structure .

Images