CN105242094A - FM/NM thin-film structure inverse spin hall voltage value measurement method - Google Patents

Abstract

The invention discloses a method for measuring an inverse spin Hall voltage value in an FM/NM film structure, and belongs to the field of spintronics research and related spin device manufacturing. The present invention is based on the method of testing the spin pumping-inverse spin Hall effect voltage of the FM/NM film structure sample by using a suspended microstrip transmission line fixture with a short-circuit terminal, and adopts the method of vertically flipping the sample in the microstrip line fixture, successively The voltage across the sample is measured when the membrane is facing up and the membrane is facing down. According to the difference in the opposite direction of spin injection in the sample configuration, the spin rectification component and the inverse of the measured voltage can be accurately obtained by simple calculation. The voltage component of the spin Hall effect provides a reference for accurate calculation of the spin Hall angle.

Description

A Measuring Method of Inverse Spin Hall Voltage in FM/NM Thin Film Structure

technical field

The invention belongs to the field of spintronics research and related spin device manufacturing, and specifically relates to a spin pumping-inverse spin Hall effect voltage based on testing FM/NM film structure samples with a suspended microstrip transmission line fixture using terminal short circuit The method uses the method of turning the sample vertically in the microstrip line fixture, and measures the voltage at both ends of the sample when the film surface is vertically upward and the film surface is vertically downward. The difference in the direction of the spin injection is opposite, and the spin rectification component and the inverse spin Hall effect voltage component in the measurement voltage can be accurately obtained through calculation.

Background technique

In spintronics, spin is used as the carrier of information storage and transmission. Therefore, the generation, manipulation and detection of spin currents are always the most basic and critical scientific issues for the realization of spintronic devices. With people's extensive research, a variety of technologies for generating spin currents have been developed, such as nonlocal electric injection spin current technology, spin current injection technology based on ferromagnetic resonance spin pumping effect (SpinPumping) , a spin-current technology based on the spin Seebeck effect, using circularly polarized light injection into the spin-current technology. Among them, the ferromagnetic resonance spin pumping method is to deposit ferromagnetic thin film material (Ferromagnetic, FM) and nonmagnetic thin film material (Nonmagnetic, NM) together. The spin current is injected into the magnetic metal. Due to the inverse spin Hall effect (Inverse Spin Hall Effect, ISHE), that is, based on the spin-orbit coupling effect, the electrons with opposite spins are deflected to the direction perpendicular to the spin current, and the spin electrons move directional A charge flow is formed, so that the spin flow is converted into a charge flow in the non-magnetic metal material, and the magnitude of the spin flow can be characterized by detecting the voltage across the sample, so as to further study the spin injection efficiency and spin diffusion of the material structure. The length provides the basis and guidance. This method of generating spin current and detecting spin current by combining the spin pumping effect and the inverse spin Hall effect has become the research frontier and hotspot of current spintronics.

In addition, in the selection of magnetic materials with FM/NM thin film structure, NiFe is the most researched by scholars at home and abroad. This is because NiFe has high magnetic permeability and low coercive force under weak and medium magnetic fields. Exhibits good soft magnetic properties, easy to magnetize and demagnetize. In terms of non-magnetic metal materials, Ta, Pt, Pd, etc. have strong spin-orbit coupling characteristics, so when there is spin injection, a large inverse spin Hall voltage will be generated. Although a lot of research has been carried out so far, there are still some difficulties in the measurement of the inverse spin Hall voltage. The reason is that in FM/NM, when the FM layer is ferromagnetically resonant, the magnetic moment will precess to The NM layer generates spin injection, and the inverse spin Hall voltage V _ISHE is generated in the NM layer by the inverse spin Hall effect. At the same time, there is also a spin rectification (Spin Rectification Effect, SRE) voltage V _SRE in the FM film, so the measured The resulting voltage is the superposition of V _ISHE and V _SRE . Theoretical studies show that V _ISHE presents a symmetric Lorentz lineshape with the external magnetic field, while the V _SRE lineshape includes symmetric and antisymmetric Lorentz lineshape. Some researchers did not consider the contribution of spin rectification during the test, or directly attributed the symmetric Lorentz component in the voltage obtained by the test to the contribution of V _ISHE , and attributed the anti-symmetric component to the contribution of V _SRE , resulting in resulting in an imprecise V _ISHE . Although the external magnetic field is rotated, that is, the angle φ _H between the external magnetic field _H and the microwave magnetic field h provided by the fixture is changed, and then the test signal V _ISHE and V The contribution of _SRE to the voltage can get a more accurate V _ISHE , but because the change trend of the test curve is more complicated, in order to make the test data better fit the fitting formula, it is generally necessary to rotate 360 degrees with a step size of 10° °Measuring 36 times, so the test steps are complicated, and the reliability of the test results depends on the number of fitting data.

Contents of the invention

In order to overcome the defects of inaccurate measurement methods and complicated steps in the background technology, the present invention provides a measurement method for accurately obtaining the inverse spin Hall voltage value in the FM/NM thin film structure, based on two effects on the direction of spin injection Dependence, the method of testing the spin pumping-inverse spin Hall effect voltage of FM/NM film structure samples by using the suspended microstrip transmission line fixture with short-circuited terminals, using the method of vertically flipping the sample in the microstrip line fixture, successively The voltage across the sample is measured when the membrane surface is vertically upward and the membrane surface is vertically downward. According to the difference in the opposite direction of spin injection in the two sample configurations, the measured voltage can be accurately obtained by simple calculation. The middle spin rectification component and the inverse spin Hall effect voltage component provide a reference for accurate calculation of the spin Hall angle.

In the present invention, the film material used is NiFe/Pt; in terms of the test fixture, the test fixture used in the present invention is a suspended microstrip transmission line with terminal short circuit. As shown in Figure 1, in the microstrip test fixture, it is only necessary to place the substrate deposited with the thin film in the microstrip line fixture, and load a test DC at both ends of the sample. Using the test fixture, on the one hand, compared with the traditional microwave transmission line method that needs to make an insulating layer and perform multiple photolithography processes, the test fixture used in the present invention is simple to prepare samples, and the film can be directly deposited on the substrate. Photolithography; on the other hand, compared to the fixed resonant frequency compared to the operating frequency of the microwave resonator fixture, the test fixture of the present invention can be tested in a wide frequency range, considering the application background of this experiment and the frequency range of use of instruments and transmission lines , so the test frequency used is 1-10GHz. The size of the film along the direction of the transmission line can be selected from 1 mm to 10 mm. For the convenience of the implementation of the experiment and the consideration of the size of the test fixture, the present invention uses 5 mm, which is only 1/60 to 1/6 of the microwave wavelength in the test frequency range, which can be approximated It is considered that the microwave magnetic field received by the film is uniform.

Based on the theory of magnetic field angle, the present invention proposes a vertical flipping sample, and by changing the relative positions of FM and NM layers in the fixture, the V _ISHE value can be obtained conveniently and accurately only by two measurements. The underlying theory is as follows:

For a single-layer NiFe sample prepared on a _SiO2 substrate, the microwave magnetic field with an applied static magnetic field The included angle is φ _H , since NiFe is a soft magnet, the coercive force is very low, so that the magnetic moment can be with keep parallel. The voltage generated by the spin rectification effect can be obtained:

V _SRE ＝A _L ·L+A _D ·D(1)

In the formula, _D and _L are the coefficients of the antisymmetric and symmetrical linear components respectively, and AL and AD are the magnitudes of the symmetrical and antisymmetrical components in the voltage respectively, which can be expressed as a function of φ _H :

A _L ＝f ₁ (φ _H )(2)

A _D =f ₂ (φ _H )(3)

For the double-layer Pt/NiFe sample prepared on the SiO2 substrate of the same size, not only the spin rectification effect of the NiFe film must be considered, but also the ISHE voltage generated by injecting spin waves from NiFe to Pt. :

V _ISHE = ISHE · sin ³ φ _H · L (4)

This voltage value is in the shape of a Lorentz symmetric line, therefore, in this sample, the measured voltage is the sum of the contributions of formula (1) and ISHE, namely:

${\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; c</span>}}^{\mathrm{<span\; class="notranslate">\; N</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; f</span>}\mathrm{<span\; class="notranslate">\; e</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; P</span>}\mathrm{<span\; class="notranslate">\; t</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; S</span>}\mathrm{<span\; class="notranslate">\; R</span>}\mathrm{<span\; class="notranslate">\; E.</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; I</span>}\mathrm{<span\; class="notranslate">\; S</span>}\mathrm{<span\; class="notranslate">\; h</span>}\mathrm{<span\; class="notranslate">\; E.</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; A</span>}}_{\mathrm{<span\; class="notranslate">\; D.</span>}}<span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; D.</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

In the above formula, the symmetrical component of the corrected bilayer film sample is

A′ _L ＝A _L +ISHE′sin ³ (φ _H )(6)

It is also worth noting that, by the formula

in γ is the gyromagnetic ratio, e is the electronic charge, is the reduced Planck constant, is the injected spin current vector, is the unit vector of the spin current injection direction, and J _S is the spin current magnitude. It can be seen that the direction of spin injection determines the sign of the ISHE voltage, indicating that the direction of the spin current is perpendicular to the film surface. Therefore, the spin injection direction of the sample Pt/NiFe is in the -y direction, while that of the sample NiFe/Pt is in the +y direction. Therefore, formula (6) can be set as the spin flow is injected in the +y direction, that is, vertically upward along the membrane surface normal, and at the same time, when the spin current is injected in the -y direction, that is, vertically downward along the membrane surface normal direction, formula (6) can be revised as

A′ _L ＝A _L -ISHE′sin ³ (φ _H )(8)

At this time, in the two equations (6) and (8), the inverse spin Hall voltage component ISHE is both positive.

The concrete steps of the present invention about NiFe/Pt film test method are as follows:

(1) NiFe/Pt/SiO _The sample is placed in the microstrip line test fixture, start the test platform, and when φ _H is 90 °, the curve of the voltage at both ends of the test sample changes with the applied static magnetic field;

(2) Turn the NiFe/Pt/SiO ₂ sample vertically 180° and then place it in the microstrip line test fixture. At this time, the surface of the film is downward, the substrate is above the film layer, the interface between the substrate and the film and the NiFe and Pt At the same time, insert a piece of SiO ₂ substrate of the same size under the sample to ensure that the film is at the same height as the position in the fixture before flipping, start the test platform, and the voltage at both ends of the test sample changes with the applied static magnetic field the curve;

(3) After the sample is turned over, the spin injection in the film is reversed, causing the curves in steps (1) and (2) to show opposite trends. Adding the two can eliminate the contribution of ISHE to the voltage, Get twice the SRE voltage value;

(4) Subtract the voltage curve after the sample is vertically flipped before the flip, that is, the curve obtained in step (2) minus step (1), then the contribution of SRE to the voltage can be eliminated, and twice the ISHE voltage can be obtained value, divide the subtracted curve by 2, the voltage at the peak is twice the maximum contribution value of ISHE to the voltage, and divide this voltage by 2 to obtain the inverse spin Hall voltage V _ISHE .

In particular, the length and width range of NiFe/Pt film is (5×1～10×10mm), the thickness of NiFe is (10～50nm), the thickness of Pt is (5～20nm), and the length and width range of _SiO2 is ( 5×5～10×10mm), the thickness of SiO ₂ is (0.2～0.5mm).

The invention is based on the spin pumping-inverse spin Hall effect, and measures the voltage at both ends of the FM/NM film structure sample when ferromagnetic resonance occurs under different applied static magnetic field angles, and then separates the ISHE voltage test method. This test method fully considers the spin rectification effect of the FM layer and the ISHE generated by spin injection, and overcomes the cumbersome steps of the traditional test method and the deviation caused by formula fitting. Afterwards, test twice to obtain the voltage curves when the spin current injection directions are vertically downward and vertically upward, respectively, and the ISHE voltage can be obtained by simple calculation. This method is simple and quick, and fully considers the contribution of various effects to the voltage. The obtained ISHE is relatively accurate, which provides a reference for further obtaining important parameters such as accurate spin Hall angles and developing related devices for spintronics.

Description of drawings

Figure 1 is a schematic diagram of the microstrip line fixture, H is the external static magnetic field, h is the microwave magnetic field, and the microwave signal is fed through the SMA port;

Figure 2 is a schematic diagram of the microwave magnetic field and the static magnetic field in the film plane, M is the magnetization of the NiFe film, j _x and j _z are the microwave current densities in the x direction and z direction generated by the sample under microwave excitation, respectively;

Figure 3 is a schematic diagram of the test platform, which is composed of a microwave source, a lock-in amplifier, a sample and a fixture, an electromagnet, and a current source;

Fig. 4 is the voltage of measuring sample NiFe when φ _H =270 ° and the separated symmetrical and antisymmetrical components;

Fig.5 Curves of (c) symmetric component A′ _L and (d) antisymmetric component A _D of NiFe/Pt sample as a function of φ _H ; (e) symmetric component A′ _L and (f) antisymmetric component of Pt/NiFe sample The curve of A _D changing with φ _H ;

Figure 6 (a) Schematic diagram of the position of the double-layer thin film NiFe/Pt/ _SiO2 sample when measuring the voltage; (b) schematic diagram of the position of the double-layer thin film Pt/NiFe/SiO2 sample when measuring the voltage; (c) double-layer thin film NiFe Schematic diagram of the position of the /Pt/SiO2 sample when measuring the voltage after vertical flipping; (d) Schematic diagram of the configuration of the double-layer thin film Pt/NiFe/SiO2 sample when measuring the voltage after vertical flipping;

The dark gray dotted line in Fig. 7 is the voltage curve V ₁ of the sample after inversion, and the light gray solid line is the voltage curve V ₂ of the sample before inversion. For the double-layer film NiFe/Pt/SiO ₂ sample, when φ _H =90°, (a) the black solid line is the curve of the voltage across the sample measured after the inversion with the applied static magnetic field minus the curve before inversion; (b) The black solid line is the curve after flip plus the curve before flip. For the double-layer film Pt/NiFe/SiO2 sample, when φ _H =270 °, (c) black solid line is the curve of the voltage across the sample measured after the reversal with the applied static magnetic field minus the curve before the reversal; ( d) The black solid line is the curve after flip plus the curve before flip.

detailed description

The present invention will be further described below by steps in the summary of the invention and accompanying drawing:

Place a sample with a length × width of 10 × 5mm in a microstrip line fixture (Figure 1), and use SMB100A (Rohde&Schwarz) as a microwave source. The frequency of the microwave signal is 4.4GHz, the power is 16dBm, and the frequency is 10KHz. Square wave modulation, at the same time, a static magnetic field with an equal step size within a certain range and a microwave magnetic field at an angle of φ _H is applied to the sample surface, and the SR830 lock-in amplifier (Standard Research System) is used to receive the 10KHz square wave emitted by the microwave source. When the microwave is used as a reference signal, the voltage generated by the excitation of the microwave magnetic field at both ends of the sample in the microstrip line fixture is detected. Since the magnitude of the magnetic field applied by the electromagnet is linearly proportional to the magnitude of the input current, the current source can be programmed and controlled by LabVIEW software to output a current that changes in equal steps within a certain range to the electromagnet, so that all The applied static magnetic field changes in equal steps, and at the same time, the voltage across the sample is recorded through the lock-in amplifier, and the curve of the voltage changing with the applied static magnetic field can be drawn. The test platform is shown in Figure 3. By changing the size of the φ _H angle, the curves of the voltage changing with the applied static magnetic field under different magnetic field directions are measured, and the symmetrical component _AL and the antisymmetrical component _AD are separated by fitting. For example, the measured voltage of sample NiFe (20nm) at φ _H of 270° and the separated symmetric and antisymmetric components are shown in Fig. 4, which can be obtained by fitting equation (1), AL = 1.098 μV, A _{D =} -4.262μV.

The traditional measurement method is shown in Figure 5. Rotate the external electromagnet of the fixture so that φ _H takes 10 degrees as the step size, and rotates from 10 degrees to 360 degrees. The voltage measured at each φ _H is separated into symmetrical components and opposing Weigh the components, and then draw the curves of the two changes with φ _H , (a) and (b) are the curves of the symmetric and antisymmetric components of the NiFe(20nm)/Pt(10nm)/SiO ₂ sample with the angle φ _H , respectively , (c) and (d) are the variation curves of the symmetric and antisymmetric components of the Pt(10nm)/NiFe(20nm)/SiO ₂ sample with the angle φ _H , respectively. Figure 6 is a schematic diagram of the above two samples during measurement, it can be seen that (a) the spin injection direction of the double-layer thin film NiFe(20nm)/Pt(10nm)/ _SiO2 sample is the -y direction; (b) Pt The spin injection direction of (10nm)/NiFe(20nm)/ _SiO2 sample is +y direction. Through fitting, it can be concluded that the ISHE=0.852μV in the NiFe(20nm)/Pt(10nm) sample, and the ISHE=0.423μV in the Pt(10nm)/NiFe(20nm) sample, the difference between the two sizes is due to the two samples The voltage detection points are in contact with NiFe and Pt respectively, and the conductivity of Pt is greater than that of NiFe, which leads to a small voltage value measured by the Pt(10nm)/NiFe(20nm) sample. This method of testing the ISHE voltage by rotating the angle and fitting the formula is too cumbersome, and during the rotation of the electromagnet, the magnetic field received by the sample at each angle may not always be uniform, and it will also appear when fitting the formula Certain deviations.

Therefore, this patent adopts a method of flipping the sample test. Without changing the angle of the magnetic field, the ISHE voltage can be obtained by simply subtracting the voltage curves obtained from the two tests before and after the vertical flipping of the sample. value. Specific steps are as follows:

For the double-layer film NiFe(20nm)/Pt(10nm)/SiO ₂ sample, the measurement position is shown in Figure 6(a), and its spin current is injected in the -y direction, and the measured voltage versus magnetic field curve is obtained by separation The resulting symmetric components can be described by (8). At this time, the sample is turned over vertically, so that the film faces downward. At the same time, in order to make the microwave magnetic field and the applied static magnetic field of the sample in the microstrip line fixture the same as before the flip test, insert a piece of the same size under the film surface. _SiO2 substrate, so that the relative position of the film has the same height as before flipping. The measurement position is shown in Figure 6(c), and the spin current is injected in the +y direction, and the symmetrical components obtained by separating the measured voltage versus magnetic field curve can be described by equation (6). From the formulas (6) and (8), it can be known that the ISHE voltage value can reach the maximum at 90 degrees, and the curve after the reversal is subtracted from the curve before the reversal, and the obtained curve is V′ _ISHE =2×V _ISHE (H ),but When in the ferromagnetic resonance field, the maximum value can be obtained, that is, the sought V _ISHE ; if the curve after the flip is added to the curve before the flip, that is, (6) plus (8), the obtained curve is V ′ _SRE =2×V _SRE (H), that is, twice the spin rectification voltage, as shown in Figure 7(b). Therefore, based on this model, it is only necessary to test the curves V ₁ and V ₂ of the voltage changing with the applied static magnetic field before and after turning over at 90 degrees. The difference between the two voltage curves is divided by 2 to obtain the maximum value of the curve It is the maximum value of the voltage generated by ISHE, as shown in Figure 7(a). After calculation, when φ _H =90°, the difference between the two voltage curves is V' _ISHE = 1.74μV, then V _ISHE = 0.87μV.

For the double-layer thin film Pt(10nm)/NiFe(20nm)/SiO ₂ sample, the measurement position is shown in Figure 6(b), the spin current is injected in the +y direction, and the measured voltage versus magnetic field curve is obtained by separation The resulting symmetric components can be described by (6). At this time, the sample was turned over vertically, so that the film surface was vertically downward. At the same time, in order to make the microwave magnetic field and the applied static magnetic field of the sample in the microstrip line fixture the same as before the flip test, a piece of the same film was inserted under the film surface. Standardize the size of the _SiO2 substrate so that the relative position of the film has the same height as before flipping. The measurement position is shown in Figure 6(d), and the spin current is injected in the -y direction, and the symmetrical components obtained by separating the measured voltage versus magnetic field curve can be described by equation (8). It can be seen from formulas (6) and (8) that the ISHE voltage can reach the maximum value at 270 degrees, and the curve after subtracting the one before the inversion is used to obtain the curve V′ _ISHE =2×V _ISHE (H ),but When in the ferromagnetic resonance field, the maximum value can be obtained, that is, the sought V _ISHE ; if the curve after the flip is added to the curve before the flip, that is, (6) plus (8), the obtained curve is V ′ _SRE =2×V _SRE (H), that is, twice the spin rectification voltage, as shown in Figure 7(d). Therefore, based on this model, it is only necessary to test the curves V ₁ and V ₂ of the voltage before and after the reversal with the applied static magnetic field at 270 degrees. The difference between the two voltage curves is divided by 2, and the ISHE The maximum value of the generated voltage is shown in Fig. 7(c). After calculation, when φ _H = 270°, the difference between the two voltage curves is V' _ISHE = 0.862μV, then V _ISHE = 0.431μV.

The test method of this patent can also be extended to the ISHE voltage measurement of general FM/NM/Substrate or NM/FM/Substrate thin film structure samples, only need to flip the sample vertically, so that the spin injection direction is reversed after the sample is flipped. Turn, test twice before and after flipping, and make a difference between the two symmetrical component change curves to eliminate the voltage component generated by the spin rectification effect, get twice the inverse spin Hall voltage value, and then divide by 2 , the ISHE voltage can be obtained, and this test method has been verified by experiments. The results obtained by this method are very similar to the results obtained by the method of rotating the angle and fitting the formula. It can be explained that this is a convenient and accurate way to obtain the ISHE voltage Value test method.

Claims (2)

1. a method for measuring reverse spin Hall voltage value in a FM/NM film structure, comprising the following steps: (1) NiFe/Pt/SiO _The sample is placed in the microstrip line test fixture, start the test platform, and when φ _H is 90 °, the curve of the voltage at both ends of the test sample changes with the applied static magnetic field; (2) Turn the NiFe/Pt/SiO ₂ sample vertically 180° and then place it in the microstrip line test fixture. At this time, the surface of the film is downward, the substrate is above the film layer, the interface between the substrate and the film and the NiFe and Pt The interface remains unchanged. Insert a _SiO2 substrate of the same size under the sample to ensure that the film is at the same height as the position in the fixture before flipping. Start the test platform and test the curve of the voltage at both ends of the sample with the external static magnetic field. ; (3) After the sample is turned over, the spin injection in the film is reversed, causing the curves in steps (1) and (2) to show opposite trends. Adding the two can eliminate the contribution of ISHE to the voltage, Get twice the SRE voltage value; (4) Subtract the voltage curve after the sample is vertically flipped before the flip, that is, the curve obtained in step (2) minus step (1), then the contribution of SRE to the voltage can be eliminated, and twice the ISHE voltage can be obtained value, divide the subtracted curve by 2, the voltage at the peak is twice the maximum contribution value of ISHE to the voltage, and divide this voltage by 2 to obtain the inverse spin Hall voltage V _ISHE . 2. The measuring method of reverse spin Hall voltage value in a kind of FM/NM film structure as claimed in claim 1, it is characterized in that: the length and width range of NiFe/Pt film is 5 * 1～10 * 10mm, NiFe The thickness of Pt is 10-50nm, the thickness of Pt is 5-20nm, the length and width range of SiO ₂ is 5×5-10×10mm, and the thickness of SiO ₂ is 0.2-0.5mm.