Thermoelectric material

Abstract

Provided is a thermoelectric material which can increase its anomalous Nernst angle. The thermoelectric material of a magnetic material for a thermoelectric power generation device employs the anomalous Nernst effect, including iron doped with iridium.

Claims

1. A thermoelectric material comprising iron doped with iridium, wherein a doping amount of the iridium in the thermoelectric material is 15 at % or more and 22 at % or less.

Description

BRIEF DESCRIPTION OF DRAWINGS

(1) FIG. 1 is a view to explain a thermoelectric power generation device 10 employing a thermoelectric material 1 of the present invention;

(2) FIG. 2 is a view to explain an embodiment of a thermoelectric power generation device of Example;

(3) FIG. 3A is a graph to explain the relationship between the Seebeck coefficient and the doping amount of Ir in the thermoelectric power generation device of Example;

(4) FIG. 3B is a view to explain the relationship between the Nernst coefficient and the doping amount of Ir in the thermoelectric power generation device of Example; and

(5) FIG. 3C is a view to explain the relationship between the anomalous Nernst angle and the doping amount of Ir in the thermoelectric power generation device of Example.

DESCRIPTION OF EMBODIMENTS

(6) Hereinafter the present invention will be explained referring to the drawings. The embodiments shown below are examples of the present invention, and the present invention is not limited to the embodiments.

(7) FIG. 1 is a view to explain a thermoelectric power generation device 10 employing a thermoelectric material of the present invention. The thermoelectric device 10 shown in FIG. 1 includes a substrate 3, a thermoelectric material 1 and a connector 2 that are arranged on the substrate 3.

(8) The thermoelectric material 1 is formed of iron (Fe) doped with iridium (Ir). The connector 2 is formed of: a nonmagnetic material not showing anomalous Nernst effect (e.g. copper (Cu) and chromium (Cr)); or a ferromagnetic material having a magnetization opposite to that of the thermoelectric material 1; or a ferromagnetic material having anomalous Nernst coefficient opposite to that of the thermoelectric material 1 (e.g. MnGa). The substrate 3 is formed of silicon, magnesium, and the like.

(9) The thermoelectric material 1 is formed of a line-thinned thin film of Fe doped with Ir formed on the substrate 3. The thermoelectric material 1 is magnetized in the direction shown in FIG. 1. The thermoelectric material 1 is configured so as to generate power in the direction of the electric field shown in FIG. 1, against the temperature difference in the direction (direction of heat flow shown in FIG. 1) perpendicular to the direction of magnetization, by the anomalous Nernst effect.

(10) The connector 2 is arranged on the surface of the substrate 3, parallel to each of thermoelectric bodies 1, 1, . . . . One connector 2 is arranged between one pair of the thermoelectric bodies 1, 1 adjacent to each other. The connector 2 electrically connects one end side of one of the pair of the thermoelectric bodies 1, 1, and the other end side of the other one of the pair of the thermoelectric bodies 1, 1. This makes the thermoelectric bodies 1, 1, . . . electrically connected in series by the connector 2.

(11) As described above, the thermoelectric power generation device 10 includes the thermoelectric material 1 formed of Fe doped with Ir. The thermoelectric material 1 formed of Fe doped with Ir can increase its anomalous Nernst angle more than a conventional magnetic material. Therefore, according to the present invention, it is possible to provide the thermoelectric material 1 which can increase the anomalous Nernst angle. By using such a thermoelectric material 1, it is possible to provide the thermoelectric power generation device 10 having a configuration easy to be practically used.

(12) In the present invention, the doping amount of Ir in the thermoelectric material 1 can be adequately determined, depending on the size of a necessary anomalous Nernst angle. As described later, by doping Fe with Ir, it is possible to enlarge the size of anomalous Nernst angle compared to the case where Fe is not doped with Ir. Therefore, the doping amount of Ir in the present invention is more than 0 at %, On the other hand, the inventors of the present invention have found the followings: (1) it is possible to increase the anomalous Nernst angle by increasing the doping amount of Ir. For example, it is possible to increase the anomalous Nernst angle more than the material described in Pu et al., by making the doping amount of Ir 7.9 at % or more; and (2) if the doping amount of Ir is excessively increased, the anomalous Nernst angle tends to decrease. According to researches of the inventors of the present invention, it is considered that the doping amount of Ir with which the anomalous Nernst angle starts to decrease is between 18 at % and 22 at %. Therefore, the upper value of the doping amount of Ir is preferably between 18 at % and 22 at %.

EXAMPLES

(13) It has been suggested that anisotropic scatterings by impurities having large spin orbit interactions to the spin direction of electrons originate effectively the anomalous Nernst effect, though details are unknown. Therefore, whether the anomalous Nernst effect was increased or not was confirmed by doping Fe which is a magnetic material, with heavy elements such as Ir, Ta and Bi, to generate scatterings of electron spins.

1. Sample Production

Example

(14) A thin film formed of Fe doped with Ir was produced on a substrate of magnesium oxide and silicon single crystal, via a process of simultaneously discharging Fe target and Ir target by means of a magnetron sputtering apparatus (BC6155, manufactured by ULVAC, INC.). Thereafter, a thermoelectric material attached with a thin wire of Fe doped with Ir and an Au electrode for measurement was formed by means of photolithography. The doping amount of Ir was adjusted by a simultaneous film deposition method for Fe target and Ir target. In producing a thermoelectric material whose doping amount of Ir was zero, only Fe target was discharged in producing a thin film. The configuration of a thermoelectric power generation device of Example produced as described is shown in FIG. 2.

Comparative Example 1

(15) A thermoelectric power generation device of Comparative Example 1 was produced in the same manner as in producing the thermoelectric power generation device of Example, except that Ta target was used instead of Ir target in producing a thin film.

Comparative Example 2

(16) A thermoelectric power generation device of Comparative Example 2 was produced in the same manner as in producing the thermoelectric power generation device of Example, except that Bi target was used instead of Ir target in producing a thin film.

2. Measurement of Seebeck Coefficient

(17) Each of the produced thermoelectric devices of Example, Comparative Example 1, and Comparative Example 2 was installed in a two-terminal prober apparatus. A heat gradient T was applied in the in-plain direction of the thin film by a heater arranged on an electrode 21 side shown in FIG. 2. Terminal probes were put on an electrode 22 and an electrode 24, and Seebeck electromotive force V.sub.SE was measured by means of a nanovoltmeter. The Seebeck coefficient was measured by measuring the heat gradient dependency of T dependency of V.sub.SE, changing the power of the heater. Table 1 shows the results of the Seebeck coefficient of the thermoelectric device of Example. Table 2 shows the results of the Seebeck coefficient of the thermoelectric device of Comparative Example 1. Table 3 shows the results of the Seebeck coefficient of the thermoelectric device of Comparative Example 2. FIG. 3A shows the relationship between the Seebeck coefficient and the doping amount of Ir in the thermoelectric device of Example.

3. Measurement of Nernst Coefficient

(18) Each of the produced thermoelectric devices of Example, Comparative Example 1, and Comparative Example 2 was installed in a two-terminal prober apparatus. A heat gradient T was applied in the in-plain direction of the thin film by a heater arranged on the electrode 21 side shown in FIG. 2. Terminal probes were put on an electrode 23 and an electrode 27. A magnetic field was applied in the direction perpendicular to the surface of the thin film by an electromagnet, and the anomalous Nernst voltage V.sub.ANE generated when Fe was magnetized in the direction perpendicular to the surface was measured by means of a nanovoltmeter. The anomalous Nernst coefficient (Nernst coefficient) was measured by measuring the heat gradient dependency of T dependency of V.sub.ANE, changing the power of the heater. Table 1 shows the results of the Nernst effect of the thermoelectric device of Example. Table 2 shows the results of the Nernst effect of the thermoelectric device of Comparative Example 1. Table 3 shows the results of the Nernst effect of the thermoelectric device of Comparative Example 2. FIG. 3B shows the relationship between the Nernst coefficient and the doping amount of Ir in the thermoelectric device of Example. In Table 3, - means that the value of the Nernst coefficient could not be identified since the Seebeck coefficient was small, or that the value of the anomalous Nernst angle could not be identified since the value of the Nernst coefficient could not be identified.

4. Identification of Anomalous Nernst Angle

(19) Seebeck coefficients measured in the above 2, and Nernst coefficients measured in the above 3. were substituted in anomalous Nernst angle (%)=100*Nernst coefficient/Seebeck coefficient, whereby the anomalous Nernst angle (%) was derived. Table 1 shows the results of the anomalous Nernst angle of the thermoelectric device of Example. Table 2 shows the results of the anomalous Nernst angle of the thermoelectric device of Comparative Example 1. Table 3 shows the results of the anomalous Nernst angle of the thermoelectric device of Comparative Example 2. FIG. 3C shows the relationship between the anomalous Nernst angle and the doping amount of Ir in the thermoelectric device of Example.

(20) TABLE-US-00001 TABLE 1 Ir doping Seebeck Nernst anomalous amount coefficient coefficient Nernst corner [at %] [V/K] [V/K] [%] 0 3.1 0.055 1.8 1.0 2.7 0.12 4.3 2.3 3.2 0.14 4.5 3.9 3.0 0.21 6.1 7.9 3.3 0.28 8.5 15 1.9 0.29 15 18 0.84 0.32 38

(21) TABLE-US-00002 TABLE 2 Ta doping Seebeck Nernst anomalous amount coefficient coefficient Nernst corner [at %] [V/K] [V/K] [%] 0 3.1 0.055 1.8 0.74 3.3 0.070 2.1 2.5 3.6 0.052 1.4 3.6 3.6 0.051 1.4 6.3 6.3 0.077 1.2 11 6.4 0.070 1.1

(22) TABLE-US-00003 TABLE 3 Bi doping Seebeck Nernst anomalous amount coefficient coefficient Nernst corner [at %] [V/K] [V/K] [%] 0 3.1 0.055 1.8 1.7 1.6 2.6 0.96 4.3 0.18

5. Results

(23) As shown in Table 1 and FIG. 3A, the Seebeck coefficient decreased as the doping amount of Ir increased. In contrast, as shown in Table 1 and FIG. 3B, the Nernst coefficient increased as the doping amount of Ir increased. As a result, as shown in Table 1 and FIG. 3C, a tendency of the anomalous Nernst angle to increase as the doping amount of Ir increases appeared. From the above results, it was possible to considerably increase the anomalous Nernst angle by doping Fe with Ir. With a Fe sample including 18 at % of Ir, it was possible to obtain up to 38% of the anomalous Nernst angle.

(24) As shown in Table 1 and FIG. 3C, the anomalous Nernst angle in a case where the doping amount of Ir was zero (pure Fe) was slightly less than 2%. Therefore, the conversion efficiency from the Seebeck current to the anomalous Nernst current was able to improve by approximately one digit by doping Fe with Ir. The finding that it is possible to convert 30% or more of the Seebeck current to the anomalous Nernst current by doping Fe with Ir is useful and novel for the practical use of a thermoelectric power generation employing the anomalous Nernst effect.

(25) In contrast, as shown in Table 2, the Seebeck coefficient increased as the doping amount of Ta increased, whereas the anomalous Nernst angle was 2.1% at most. As shown in Table 3, the Seebeck coefficient decreased to be rarely observed as the doping amount of Bi increased. As a result, the observation of the Nernst coefficient got difficult and the anomalous Nernst angle could not be calculated.

REFERENCES SIGN LIST

(26) 1 thermoelectric material 2 connector 3 substrate 10 thermoelectric power generation device 21, 22, 23, 24, 25, 26, 27, 28 electrode