JP2002329902A - Spin valve transistor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a spin valve transistor which can obtain a large current change rate. SOLUTION: The spin valve transistor is constituted in such a way that its emitter is constituted of a tunnel or semiconductor-metal Schottky barrier, that its base is constituted of a GMR film, and that its collector is constituted of a Schottky barrier. A current source is provided between the emitter and the base, and a voltage source is provided between the base and the collector.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a spin valve transistor.

[0002]

2. Description of the Related Art A CIP-GMR is used for a current GMR head. CIP-GMR does not seem to achieve an MR ratio of 20% or more. Therefore, CPP-GMR is being studied in order to obtain a larger MR ratio. As one of them, there has been reported a spin valve transistor (SVT) using hot electrons which can obtain a change rate of nearly 100%. The spin valve transistor has a three-terminal structure including a tunnel or a semiconductor-metal Schottky barrier as an emitter, a GMR film as a base, and a Schottky barrier as a collector, and hot electrons are injected from the emitter.

[0003] However, there is a demand for a spin-valve transistor capable of obtaining a higher current change rate than at present.

[0004]

SUMMARY OF THE INVENTION An object of the present invention is to provide a spin valve transistor capable of obtaining a large current change rate.

[0005]

A spin valve transistor according to the present invention comprises a tunnel or semiconductor-metal Schottky barrier for an emitter, a GMR film for a base, and a Schottky barrier for a collector. It has a current source and a voltage source between the base and the collector.

[0006]

EXAMPLES Example 1 A sample was prepared by n-Si (100 nm) / Si by magnetron sputtering using an Al ₂ O ₃ target for AlO _x barrier deposition.
O / NiFe (5.0 nm) / Cu (3.5 nm) / C
o (2.0 nm) / AlO _x (1.0 nm) / Cu (1
(00 nm). No attempt was made to remove the native oxide layer on the Si wafer prior to deposition. A metal mask was used to determine the pattern of the AlO _x junction. The diameter of the joint is about 1m
m. When the L _max and _{L min} the maximum and minimum collector currents respectively, the magnetic field dependence collector current change rate _{(| L max -L mi n |} ) / | L min | to define.

FIG. 1 shows a magnetic hysteresis curve of this sample. Since a symmetric curve is obtained, it can be seen that the sample exhibits a magnetization reversal process indicating an antiparallel arrangement of the magnetic layers within a certain magnetic field range. It is observed that the switching fields of the NiFe and Co layers are about 10 Oe and 60 Oe, respectively. MR measurements were also performed on this sample using in-plane geometry current. The MR ratio was about 3.5% at room temperature.

[0008] A schematic representation of the measurement settings is shown in FIG. Two power supplies were used for emitter bias and collector bias. A constant current power supply was used for emitter bias.
The emitter current was 1A. The magnetic field dependence of the collector current and the base current was measured at room temperature for various collector bias voltages. In this experiment, the collector bias voltage was changed from 0.1 V to -0.1 V with respect to the base electrode. When applying a negative bias to the collector barrier, two-way electrons moving from emitter to collector and from collector to emitter must be considered. The former electron is called a reverse electron (forward current), and the latter is called a forward electron (reverse current).

Before measuring the GMR effect, the current-voltage (IV) characteristics of the emitter and the collector barrier were examined. AlO
_The IV characteristics of the _x- layer showed linear dependence between -0.5V and + 0.5V. The resistance was several hundred ohms. The IV characteristics of the collector barrier exhibited a non-linear and asymmetrical dependence, which seems to be a superposition of the characteristics of the tunnel barrier on that of the Schottky barrier. This property may be due to the thick native oxide layer on the Si substrate.

FIG. 3 shows the magnetic field dependence of the collector current up to 200 Oe. In this figure, there are three graphs with different collector bias voltages. The top, middle and bottom panels are
Collector current vs. magnetic field (I) with collector bias voltages of 0.05V, -0.05V and -0.10V respectively.
_C- H) shows a curve.

When the collector bias voltage is 0.01 V, the collector current is extremely large. It is conceivable that this large collector current arises from the characteristics of the collector and the emitter barrier. Nevertheless, a change in collector current due to the magnetic field was obtained. This change is considered to correspond to the magnetization process of FIG. When the magnetizations of the two ferromagnetic regions are in an antiparallel arrangement, the collector current decreases. In this case, G
The MR effect is attributable to the spin-dependent scattering of inverse electrons traveling from the emitter to the collector. The collector current change and the collector current change rate are about 7 μA and about 0.5%, respectively.
It is.

The collector bias voltage shown in FIG.
For 0.10V, the collector current has a negative sign. A negative collector current indicates that the predominant electrons that primarily affect the collector current are forward electrons traveling from the collector to the emitter. The GMR effect can be expected to be obtained for forward electrons. In fact, it can be seen that the absolute value of the collector current decreases when the magnetization is in an antiparallel arrangement. This result coincides with that shown in FIG. 3A except for the dominant electrons that contribute to the GMR effect. From FIG. 3C, the changes in the collector current and the collector current change rate are estimated to be about 16 μA and about 4%, respectively. The change in the collector current is as large as that estimated from FIG.

[0013] Interesting characteristics were observed when the collector bias was -0.05V. Figure 3 shows the experimental results.
(B). In this figure, it is found that the sweep of the magnetic field changes the collector voltage from negative to positive. The collector current was about −8 μA in the parallel arrangement of magnetization, while it was about +8 μA in the anti-parallel arrangement. The collector current is smaller than the results shown in FIGS. 3 (a) and 3 (c).
Relatively small. From this result, it is estimated that the collector current change rate is about 200% and the collector current change is about 16 μA.

In this experiment, the behavior of a two-way electron, that is, an electron in the reverse order was observed. When the collector bias voltage is positive, the behavior of the reverse electron can be mainly observed. Conversely, when the bias voltage is negative, the behavior of forward electrons can be observed. Therefore, it can be considered that the I _C -H curves shown in FIGS. 3A and 3C reflect the spin-dependent scattering of forward and reverse electrons, respectively.
It is also considered that the dominant electrons that affect the GMR effect can be controlled using the collector bias voltage. The small collector current in FIG. 3 (b) indicates that the currents due to the forward and reverse electrons almost cancel each other. In this case, the sign of the collector current changes from negative to positive due to the decrease of the forward electrons in the antiparallel arrangement of the magnetization. GM shown in FIG. 3 (b)
The dominant electrons affecting the R effect are forward electrons. It can be taken into account that when the two-way electrons are balanced, a large MR ratio is obtained. In fact, very large, ie 10
An MR ratio exceeding 0% can be obtained by controlling the collector bias voltage.

In conclusion, an SVT with a NiFe / Cu / Co metal base was investigated. It can be observed that both the forward and reverse electrons exhibited spin-dependent scattering. The predominant electrons affecting the GMR effect can be controlled by the collector bias voltage. When the collector bias voltage was -0.05 V, the sign of the collector current changed from negative to positive due to the sweep of the magnetic field. In this state, an extremely large collector current change rate of about 200% could be obtained at room temperature.

In the end, it was observed that both forward and reverse electrons exhibited spin-dependent scattering. The predominant electrons affecting the GMR effect changed the sign of the collector current from negative to positive due to the sweep of the magnetic field when the collector bias voltage was -0.05V. In this state, an extremely large change rate of the collector current of about 200% could be obtained at room temperature.

Example 2 In this example, an AlO _x layer was used for an emitter barrier. To investigate the behavior of bidirectional electrons in the SVT, electrons were injected from the emitter and collector electrodes.

The samples were prepared by magnetron sputtering using an Al ₂ O ₃ target for AlO _x deposition using n-Si (100 nm) / SiO / NiFe
(5.0 nm) / Cu (3.5 nm) / Co (2.0 n
m) / AlO _x (1.0 nm) / Cu (100 nm) structure. No attempt was made to remove the native oxide layer on the Si wafer prior to deposition. A metal mask was used to determine the pattern of the AlO _x junction. The diameter of the joint was about 1 mm.
When L _max and L _min are the maximum and minimum collector currents, respectively, the magnetic field-dependent collector current change rate is (| L
_max− L _min |) / | L _min |.

The experimental setup for the collector current measurement is the same as in the first embodiment. Power supplies were used for emitter and collector bias. A constant current power supply was used for the emitter bias.
The emitter current was 1 mA. The collector voltage was changed from -0.1 V to +0.1 V with respect to the base electrode.
The magnetic field dependence of the collector and base currents was measured at room temperature. One can take into account the presence of bidirectional electrons in the SVT.

FIG. 1 shows a typical magnetic hysteresis curve of the SVT. The symmetry curve shows that an antiparallel arrangement of the magnetic layers is obtained within a certain magnetic field range. MR measurements using in-plane geometry currents were also performed on this sample. The MR ratio was about 3% at room temperature.

Before measuring the magnetic field-dependent collector current, the current-voltage (IV) characteristics of the emitter barrier were examined. AlO
_The IV characteristics of the _x- layer showed linear dependence between -0.5V and + 0.5V. AlO _x appears to contain many defects. The resistance was several hundred ohms.

The IV characteristics of the collector barrier were also investigated. Many IV characteristics were obtained. The samples were classified into two, namely, type A (diode characteristics) and type B (resistance characteristics). Typical IV characteristics are shown in FIG. FIG. 4 (a)
The IV characteristics of type A shown in FIG.
However, large leakage currents can be seen. On the contrary, it can be seen from FIG. 4B that the IV characteristic of the type B sample shows linear dependence. These characteristics are considered to affect the magnetic field dependence of the collector current.

FIG. 5 shows the dependence of the collector current on the magnetic field up to 2000 Oe (I _c -H). In this case, the collector bias voltage is positive. FIG. 5B shows the type B sample. This result indicates that the dominant electrons affecting GMR are reverse electrons. Type A samples show different _I c -H curve. The results are shown in FIG. The collector current increases when the magnetization is in an anti-parallel arrangement. It can be taken into account that an increase in collector current corresponds to a decrease in forward electrons at the collector electrode.

[0024] Figure 6 shows an I _c -H curve when adding a negative collector bias voltage to the sample. _I c -H curves of type A and type B samples show the same characteristics. Since the sign of the collector current is negative, the absolute value of the collector current decreases in the antiparallel arrangement of magnetization. In this case, GMR
The effect is mainly attributable to the spin-dependent scattering of forward electrons.

In the experiment shown above, two-way electrons,
That is, the GMR effect of forward and reverse electrons could be observed. The predominant electrons affecting the GMR effect were affected by the collector bias voltage and also by the collector barrier properties. When the collector currents based on bidirectional electrons almost cancel each other, a large change rate of the collector current was obtained. The I _c -H curve is shown in FIG. In both types, the sweep of the magnetic field changes the sign of the collector current. When the magnetization of the base GMR film is in a parallel configuration, the sign of the collector current is negative. The collector current has a positive sign when an antiparallel arrangement of the magnetization layers is obtained within a constant magnetic field range. This property results from the reduction of forward electrons due to spin-dependent scattering. About 1200% for each sample
Was obtained. These samples are
It can be considered that the rate of change of the collector current has the possibility of an infinite value.

After all, it was observed that both forward and reverse electrons exhibited spin-dependent scattering. The dominant electrons affecting the GMR effect could be controlled by the collector bias voltage. When the two-way current based on the two-way electrons is made substantially parallel using an appropriate collector bias voltage, the sign of the collector current changes due to the sweep of the magnetic field, and the collector current change rate is about 1200% regardless of the collector barrier characteristics. Was.

Example 3 A sample was prepared using a DC / RF magnetron sputtering apparatus. The film structure is Si
/ NiFe / Cu / Co / (AlO) coercive force difference type;
Si / Ta / NiFe / Cu / Co / FeMn / Ta /
(AlO) spin valve type.

FIG. 2 shows an example of the formed film structure. The Si substrate was prepared in a case where the natural oxidation m section on the surface was not treated and a case where the surface was removed using hydrofluoric acid. In this example, an untreated Si substrate was used.

The AlO layer was divided into _two types: one directly sputtered by an Al ₂ O ₃ target, and one spontaneously oxidized in a chamber after sputtering Al. It has a three-terminal structure in which the upper part of the film surface is an emitter, Si is a collector, and a Co or Ta part is a base. Current source between emitter and base, voltage source between base and collector (bias voltage)
And the collector current was measured by changing the voltage between the base and the collector.

Si / NiFe (5 nm) / Cu (3.5
nm) / Co (2 nm) / (AlO) Coercive force difference type film has the same magnetization curve as that shown in FIG. It was found that there was a step in a state corresponding to the antiparallel state of the magnetization of the two ferromagnetic layers. The CIPGMR of this film was 3.5%. The IV characteristics between the base and the collector of this film were the same as those shown in FIG. One exhibiting rectifying properties (type A) and one exhibiting ohmic properties in the measurement range (type B) were obtained. A between emitter and base
10 was almost ohmic because sufficient tunnel characteristics could not be obtained.

The change in collector current when a magnetic field was applied to the type A sample was similar to that shown in FIG. When the bias voltage is +0.01 V, a downward convex change is obtained, and it can be seen that electrons flowing from the emitter to the collector are dominant and exhibit GMR. At -0.1 V, on the contrary, electrons flowing from the collector to the base are dominant, indicating GMR. At -00535V, the currents in the two directions are substantially balanced, but a convex change is obtained. These current changes substantially correspond to the magnetic field where a stepwise change is obtained in the magnetization curve shown in FIG. The change of the current of this sample is 10 μA or more, and the value is slightly increased when the bias voltage is negative. Change in collector current ΔIc
FIG. 8 shows the result of defining the MR ratio as / Ic (parallel).
Shown in Large MR where collector current is almost zero
The ratio has been obtained.

FIG. 9 shows a change in the collector current when a magnetic field is applied to the type B sample. It can be seen from the comparison with FIG. 3 that the value and the change of the collector current are small.
Although the collector current flows from the emitter to the collector when the + voltage is applied, the change due to GMR shows an upward convexity, which is different from the type A. This 2
It is also considered that the change appears in the two samples and that the change in the current has asymmetry with respect to the positive or negative of the bias voltage.

FIG. 10 shows the results of the spin valve method. Also in this case, the same result as described above was obtained, and it was found that the same could be applied to the spin valve structure.

[0034]

According to the present invention, a large current change rate can be obtained.

[Brief description of the drawings]

FIG. 1 is a magnetic hysteresis curve of a sample in an example.

FIG. 2 is a diagram illustrating a measuring apparatus according to an embodiment.

FIG. 3 is a graph showing the magnetic field dependence of the collector current in Example 1.

FIG. 4 is an IV characteristic diagram between a base and a collector in the second embodiment.

FIG. 5 is a graph showing magnetic field dependence of a collector current in Example 2.

FIG. 6 is a graph showing the magnetic field dependence of a collector current in Example 2.

FIG. 7 is a graph showing the magnetic field dependence of the collector current in Example 2.

FIG. 8 is a graph showing a change rate of a collector current of a type A sample.

FIG. 9 is a graph showing a change in a collector current of a type B sample with respect to a magnetic field.

FIG. 10 is a graph showing a change in a magnetization curve and a collector current of a spin valve type sample.

 ──────────────────────────────────────────────────続 き Continued from the front page F term (reference) 2G017 AA01 AD58 AD61 AD62 5E049 AA01 AA04 AA07 AC05 BA30

Claims (8)

[Claims] 1. A semiconductor device comprising a tunnel or semiconductor-metal Schottky barrier as an emitter, a GMR film as a base, a Schottky barrier as a collector, a current source between the emitter and the base, and a voltage between the base and the collector. A spin-valve transistor having a source. 2. A method of changing a voltage between a base and a collector,
2. The spin valve transistor according to claim 1, wherein a two-way current from a metal to a semiconductor and from a semiconductor to a metal can be controlled. 3. The spin valve transistor according to claim 2, wherein the two-way electrons are substantially balanced. 4. The spin-valve transistor according to claim 1, wherein the current-voltage characteristics between the base and the collector exhibit rectification. 5. The current-voltage characteristic between a base and a collector shows ohmic characteristics.
The spin valve transistor according to any one of the preceding claims. 6. The spin valve transistor according to claim 1, wherein the base has a sandwich structure of NiFe / Cu / Co. 7. The spin valve transistor according to claim 1, wherein the emitter is made of AlO _x . 8. The spin valve transistor according to claim 1, wherein the collector is made of n-Si.