WO1999046824A1 - Method of producing thermoelectric semiconductor material - Google Patents

Abstract

A powder sintered material of a thermoelectric semiconductor having rhombohedral structure (hexagonal structure) is subjected to hot forging and is plastically deformed so that crystal grains in powder sinter structure are oriented in the direction of a crystal orientation having excellent figure of merit, to produce a thermoelectric semiconductor material which has a satisfactory strength and performance and also can be used for producing a product in a high yield.

Description

 Description Manufacturing method for thermoelectric semiconductor materials

Technical field

 The present invention relates to a method for producing a thermoelectric semiconductor material. Background art

 Electro-cooling elements using the Peltier effect or the Etching-Shausen effect, or thermoelectric generation elements using the Seebeck effect have attracted attention in a wide range of applications because they have a simple structure, are easy to handle, and can maintain stable characteristics. ing. Especially for thermoelectric coolers, since local cooling and precise temperature control near room temperature are possible, research is being widely conducted toward constant temperature control of optoelectronics and semiconductor lasers.

In the thermoelectric module used for this electronic cooling and thermoelectric power generation, as shown in FIG. 12, a p-type semiconductor 5 and an n-type semiconductor 6 are joined via a metal electrode 7 to form a pn element pair. Are arranged in series, and one end is heated while the other end is cooled depending on the direction of the current flowing through the junction. The material of this thermoelectric element has a large figure of merit Z (two a ² / p) expressed by the Seebeck coefficient, the specific resistance p, and the thermal conductivity K, which are constants specific to the substance, in the operating temperature range. Is used. Many thermoelectric semiconductor materials have anisotropy in thermoelectric performance due to their crystal structure. In other words, the figure of merit Ζ differs depending on the crystal orientation. For this reason, single-crystal materials are used by energizing in the crystal orientation with high thermoelectric performance. In general, since anisotropic crystals have cleavage properties and weak material strength, single crystals are not used as practical materials, but are polycrystals that are unidirectionally solidified by the Pridgeman method and oriented to crystal orientations with high thermoelectric performance. Is used. However, although polycrystalline materials are not as strong as single crystals, their material strength is weak, and there is a problem that cracking or cracking of the element is likely to occur during element processing. In contrast to these crystalline materials, powdered sintering materials have no cleavage and material strength is dramatically increased. Although the crystal orientation is improved, there is a problem that the thermoelectric performance is inferior to that of the crystalline material because the crystal orientation is random or has a crystalline orientation but has a gradual distribution. There has been no thermoelectric semiconductor material having sufficient strength and performance. That is, in general the crystalline material used as the electronic cooling element, tellurium mold bismuth _{(B i 2 T e 3)} , antimony telluride (S b ₂ T e ₃₎ and selenide bis mass (B i ₂ S e ₃₎ However, these crystals have remarkable cleavage properties, and after slicing and dicing processes to obtain thermoelectric elements from ingots, the yield is extremely low due to cracking and chipping. There was a problem.

 Therefore, attempts have been made to form a sintered powder element to improve the mechanical strength. Thus, when used as a powdered sintered body instead of as a crystal, the problem of cleavage is eliminated, but as described above, its performance is inferior due to low orientation. That is, there is a problem that the figure of merit Z is small.

 The present invention has been made in view of such circumstances, and has as its object to provide a thermoelectric semiconductor material having sufficient strength and performance and a high production yield. Disclosure of the invention

 Therefore, in the present invention, a heating step of mixing and heating and melting material powders to have a desired composition, and a solidification step of forming a solid solution ingot of a thermoelectric semiconductor material having a rhombohedral structure (hexagonal structure) are provided. A pulverizing step of pulverizing the solid solution ingot to form a solid solution powder; a sizing step of uniformizing the particle diameter of the solid solution powder; and a sintering of pressure-sintering the solid solution powder having a uniform particle diameter. And a hot upsetting forging process in which the powder sintered body is plastically deformed by heat and expanded to orient the crystal grains of the powder sintered structure in a crystal orientation having an excellent figure of merit. Is assumed. In the first aspect of the present invention, the hot upsetting forging step is characterized in that the powder sintered body is hot plastically deformed and spread while being pressed perpendicularly to a pressing direction in the sintering step. I do.

In the second aspect of the present invention, in the hot upsetting forging step, the powder sintered body is plastically deformed hot, and is expanded while being pressed in a direction coinciding with a pressing direction in the sintering step. Thereafter, it is characterized by extending in a direction perpendicular to the pressing direction in the sintering step. According to a third aspect of the present invention, in the hot upsetting forging step, the powder sintered body is hot plastically deformed, and is spread while being pressed in a direction perpendicular to a pressing direction in the sintering step. It is characterized in that it extends in a direction coinciding with the pressing direction in the sintering step. In a fourth aspect of the present invention, the hot upsetting forging step is characterized in that the powder sintered body is plastically deformed and spread hot by an initial load pressure of 100 Okg / cm ² or less. .

In a fifth aspect of the present invention, the hot upsetting forging step is characterized in that the powder sintered body is plastically deformed and spread hot by a load pressure of 100 Okg / cm ² or less. The method of the present invention focuses on the anisotropy of the thermoelectric performance inherent in a single crystal of a thermoelectric semiconductor material, and hot-pours a sintered powder ingot that is strong but has poor crystal grain orientation. The deformation improves the crystal orientation. Thereby, a thermoelectric semiconductor material having good performance while maintaining strength can be obtained. Due to the plastic deformation of hot forging, the ingot shrinks in the compression direction, while it expands in the direction of the compression surface. Due to this deformation, the crystal grains in the ingot undergo flat plastic deformation and are oriented so that the cleavage plane is perpendicular to the compression direction. As a result, a thermoelectric semiconductor material having sufficient strength and performance can be obtained. In the B i ₂ Te ₃ based thermoelectric semiconductor material, c-axis is oriented in the compression Direction.

 Therefore, it is possible to obtain a highly reliable thermoelectric module having high mechanical strength and excellent orientation.

 In addition, in the present invention, since the powder is formed by hot upsetting and forging a powdered sintered body instead of a single crystal, the composition ratio can be selected relatively freely, and a high figure of merit Z can be obtained. .

 In addition, compared to the case where a single crystal or polycrystalline ingot is used as it is, a decrease in manufacturing yield due to cracks or the like is greatly reduced.

Here, the terms B i ₂ T e ₃ based thermoelectric semiconductor _{materials, B i 2 -xSb x Te 3} - y - z Se y S z (0≤x≤2, 0≤y + z≤3) t reveal of And those containing impurities in the crystal. Similarly, Bi-Sb-based semiconductor is Bit- _x Sb 0 <x <1), and includes those containing impurities as dopants in the crystal. BRIEF DESCRIPTION OF THE FIGURES

 FIG. 1 is a conceptual diagram showing hot upsetting forging in the method for producing a thermoelectric semiconductor of the present invention.

 FIG. 2 is a micrograph of the thermoelectric semiconductor material after hot upsetting forging according to the present invention.

 Figure 3 is a micrograph of the thermoelectric semiconductor material before hot upsetting forging according to the present invention.

 FIG. 4 is a view showing a flowchart of the method for manufacturing a thermoelectric semiconductor according to the present invention. FIG. 5 is a view showing an upsetting device used in the method of the first embodiment of the present invention.

 FIG. 6 is a view showing the result of measuring the relationship between the forging time and the ingot thickness in the hot upsetting forging of the present invention.

 Figure 7 shows the distribution of the density ratio of this ingot to a single crystal in a quarter part.

 FIG. 8 is a diagram showing the relationship between the pressing force and the thickness reduction of the ingot in the hot upholstery curtain of the present invention.

 FIG. 9 is a diagram showing an upsetting device used in the method of the second embodiment of the present invention. FIG. 10 is a diagram showing the relationship between pressure, time, and ingot thickness in the method of the present embodiment. .

 Fig. 11 shows the distribution of the density ratio of this ingot to a single crystal in a quarter part.

 FIG. 12 is a diagram showing a thermoelectric module.

FIG. 13 is a diagram showing the relationship between the heat radiation surface temperature and the maximum temperature difference of a thermoelectric module formed using the thermoelectric element material formed by the method of the present invention. FIG. 14 is a diagram showing the relationship between the heat radiating surface temperature and the maximum temperature difference of the thermoelectric module formed by using the thermoelectric element material formed by the method of the present invention.

 FIG. 15 is a diagram showing time variations of pressure and ingot height in the method of the embodiment of the present invention.

 FIG. 16 is a diagram showing how the forging process of the embodiment of the present invention is repeatedly performed. FIG. 17 is a diagram for explaining how to measure the strength of the thermoelectric module created in the embodiment of the present invention.

 FIG. 18 is a diagram showing time variations of pressure and ingot height in the method of the embodiment of the present invention.

 FIG. 19 is a diagram showing time variations of pressure and ingot height in the method of the embodiment of the present invention.

 FIG. 20 is a view showing a measuring piece for measuring physical property values of a forged ingot. FIG. 21 is a diagram showing the time variation of the pressure and the ingot height in the method according to the embodiment of the present invention.

 FIG. 22 is a diagram showing time changes of pressure and ingot height in the method of the embodiment of the present invention.

 FIG. 23 is a diagram showing the time variation of the pressure and the ingot height in the method according to the embodiment of the present invention.

 FIG. 24 is a diagram showing the time variation of the pressure and the height of the ingot in the method according to the embodiment of the present invention.

Figure 2 5 shows a ^v definitive pressure and Ingotto time changes in height in the method of Example of the present invention.

 FIG. 26 is a diagram showing time changes of pressure and ingot height in the method of the embodiment of the present invention.

 FIG. 27 is a diagram showing the time variation of the pressure and the height of the ingot in the method according to the embodiment of the present invention.

FIG. 28 is a diagram showing a time change of the pressure and the ingot height in the method according to the embodiment of the present invention. FIG. 29 is a diagram showing time changes of pressure and ingot height in the method of the embodiment of the present invention.

 FIG. 30 is a view for explaining the upsetting apparatus according to the embodiment of the present invention.

 FIG. 31 is a view for explaining the upsetting apparatus according to the embodiment of the present invention. BEST MODE FOR CARRYING OUT THE INVENTION

 Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

 According to the method of the present invention, a sintered ingot obtained by powder sintering is subjected to hot upsetting forging as shown in the conceptual views of FIGS. 1 (a) and 1 (b), and crystal grains are arranged in layers as shown in FIG. It is characterized by having been processed so as to face. Fig. 2 is a micrograph showing the state after free forging, and Fig. 3 is a micrograph before free forging.

The photographs shown in Fig. 2 and Fig. 3 show that the material after hot upsetting and the material before hot upsetting are embedded in epoxy resin ^S and polished until the surface of each material becomes mirror-finished. This is a structure photograph etched later with an acidic solution. The vertical direction of the photograph is the pressing direction during forging and sintering.

 In the state before forging in Fig. 3, the structure is observed in a wavy shape. The wavy grains are crystal grains resulting from the grains of the powder when forming the powder sintered material.

 In the state after forging in Fig. 2, it can be seen that the wavy crystal grains shown in Fig. 3 are elongated and elongated by the forging, and the crystal grains are oriented in layers perpendicular to the pressing direction.

 In addition, the circular hole in the photograph was generated at the time of etching, and does not indicate the structure itself.

FIG. 4 is a flowchart showing a manufacturing process of an n-type Bi ₂ Te ₃ thermoelectric semiconductor material.

¾

That is, as shown in FIG. 4, bismuth B i, tellurium T e, the single element selenium S e, the stoichiometric ratio _{B i 2 T e 2. 7} S e. ₃ . The mixture was weighed so as to obtain an appropriate amount of a compound for adjusting the carrier concentration, and then dissolved, mixed and coagulated to prepare a soluble material. After pulverizing this soluble material with a stamp mill, ball mill, etc., 15 The mixture is passed through a 0-mesh and a 400-mesh sieve, and what is left on the 400-mesh sieve is selected to prepare powder having a particle size of about 34 to 106 1m. After sizing, a predetermined volume of powder is supplied into a predetermined volume of glass ampoule under vacuum evacuation, hydrogen is injected and sealed to 0.9 atm, and then the mixture is heated in a 350 ° C heating furnace. Hydrogen reduction was performed by performing heat treatment for 0 hours. The powder was sintered by a hot press apparatus in an argon atmosphere at a sintering temperature of 500 ° C. and a pressure of 7500 kg / cm ² . The size of the sintered ingot was 32 mm x 32 mm in cross section and 20 mm in thickness. The ingot has a negative Seebeck coefficient and the material has n-type. This is forged by hot upsetting forging.

In the forging process, as shown in Fig. 5, this ingot was installed in a cemented carbide upsetting device, and was subjected to 450 in an argon atmosphere. This is performed by pressing at 150 kg / cm ² at the top in the same direction as the pressing direction during pressure sintering. As a result, the sintered ingot is compressed. Here, FIG. 5 (a) is a top view, FIG. 5 (b) is a cross-sectional view a of _c The upsetting apparatus, the base 1 and the base 1 a cylindrical sleeve 2 orthogonally was allowed erected this sleeve 2 And a punch 3 formed so as to be inserted into the base 1. The ingot 4 is placed on the base 1 and pressed by the punch 3. This device is made of carbide and is configured to be heated to about 450 ° C. by a heating device (not shown). According to this device, the ingot 4 is pressurized only from the upper and lower directions, and is free in the other directions, and is forged so that the c-axis directions are aligned. Figure 6 shows the results of measuring the relationship between forging time and ingot thickness. Pressing for 12 hours increased the thickness of the ingot to about 1/2, about 8 mm, and the bottom area was about twice as large as 48.5 mm square. Figure 7 shows the distribution of the density ratio of the ingot to the single crystal in the quarter part. Since the thermoelectric performance decreases as the density decreases and the strength also decreases, the usable part is the part with a density ratio of 97% or more.

Table 1 shows the thermal performance of this hot forged ingot and the central part of the ingot before forging. In the case of this n-type material, carriers are reduced due to crystal defects due to hot forging, and the specific resistance and Seebeck constant are increased. Both were sealed in a glass ampule and heat-treated at 400 ° C for 48 hours to eliminate lattice defects for forging, and the carrier concentration became the same as the starting ingot. This can be understood from the fact that the Seebeck constant is the same. table 1

In addition, hot forging was performed by changing the compression ratio, and the performances of the heat-treated materials were compared (the results are shown in Table 2.

As is evident from Table 2, as the compression ratio increases, the amount of plastic deformation increases, the orientation improves, and the thermoelectric performance improves. Increasing the pressure increases the compression speed (thickness reduction amount) as shown in Fig. 8 and shortens the processing time, but cracks occur during processing and the high-density part decreases. The number of usable parts decreases. Therefore, it is desirable to set the applied pressure to such an extent that cracking is not possible. This pressing force also depends on the frictional force at the portion where the ingot contacts the base and the punch. Carbon powder, BN powder, etc. were applied to the bottom of the punch and the top of the base before forging in order to make the spreading smooth. This facilitates deformation without cracking. Thus, according to the method of the present invention, there is no decrease in density due to cracks and the like during forging. It is possible to obtain a thermoelectric semiconductor material having a high yield and an extremely good orientation. The hot forging step was performed in an argon atmosphere, but is not limited to this, and may be performed in a vacuum or in another inert gas atmosphere.

Next, as a second embodiment of the present invention, a description will be given of an upsetting forging method for freely extending only in one axial direction. In this method, as shown in Fig. 9, the forging process is performed by placing this ingot on a cemented carbide upsetting device and subjecting it to 100-500 kg / cm ² at 450 ° 〇. For 5 hours by pressing in the same direction as the pressing direction during pressure sintering. As a result, the sintered ingot is compressed. Here, FIG. 9 (a) shows the state before upsetting forging, and FIG. 9 (b) shows the state after upsetting forging. This upsetting device is erected perpendicularly to the base 11 and the base 11 so as to be inserted into the cylindrical sleeve 12 having a rectangular parallelepiped cavity H therein and the cavity H of the sleeve 12. With the formed punch 13, the ingot 14 is placed on the base 11 and pressed by the punch 13. This device is made of carbide and is configured to be heated to about 450 ° C. by a heating device (not shown). Here, the ingot had a thickness of 30 mm, a width of 40 mm, and a length in the extending direction of 18 mm. According to this device, the ingot 14 is regulated from above and below and from the width direction of the cavity in the sleeve, and in the other two directions, these directions are free until they contact the wall of the sleeve, Forged so that the cleavage planes are aligned. Figure 10 shows the results of measuring the thickness of the ingot against the applied pressure and the elapsed time in this step. As can be seen from the figure, the thickness of the ingot was reduced to about 7 mm by pressing for 5 hours, and the length was pressed to the wall of the sleeve. It was the same 80 mm. About 90% of these ingots had a density ratio of 98% or more. FIG. 11 shows the distribution of the density ratio with respect to a half of the ingot single crystal. The thermoelectric performance decreases as the density ratio decreases, and the material strength also decreases. In this example, the usable part is the part with a density ratio of 97% or more, so almost all of the usable part can be used.

Table 3 shows the thermoelectric properties of the heat-treated ingot after hot forging and the center of the ingot before forging. In the case of this n-type material, crystal defects caused by hot forging As a result, the lattice resistance for forging is eliminated by heat treatment in an argon gas atmosphere as in the first embodiment, and the carrier concentration is reduced. Will be the same as the departure ingot. This can be understood from the same Seebeck constant. Table 3

In the above-described embodiment, during the forging process, the pressure is applied in the same direction as the pressing direction during the pressure sintering, but during the forging process, the direction perpendicular to the pressing direction during the pressure sintering is used. Even if pressure is applied in the opposite direction, the thermoelectric performance can be similarly improved. Hereinafter, this embodiment will be described.

In the forging process of the present embodiment, a sintered ingot having a height of 30 mm, a width of 40 mm, and a length of 20 mm in a direction in which it is extended was cemented into a cemented carbide upsetting device shown in FIG. And a pressure of ²⁰⁰ kg / cm2 at 450 ° C in an argon atmosphere for 6 hours from the top in the direction perpendicular to the pressing direction during pressure sintering. Press and freely expand only in one axis direction.

 As a result, the sintered ingot was compressed, and the dimensions of the forged ingot after forging were 7.5 mm in height, 40 mm in width, and 80 mm in the spreading direction.

Table 37 below shows a comparison of the physical properties of the sintered ingot (hot pressed product) before forging and the forged ingot (hot forged product) heat treated after hot forging at the center and at the end. Table ^

As shown in Table 37, it can be seen that the performance index indicating the thermoelectric performance of the forged ingot after forging is improved as compared with the sintered ingot before forging. When the figure of merit is 2.45 or more, the thermoelectric performance is evaluated as "good." The ingot after forging shows a figure of merit of 2.45 or more, indicating that the thermoelectric performance is “good”.

 In the above-described embodiment, the forging process is performed by pressing in the same direction as the pressing direction during pressure sintering, or by pressing in the direction perpendicular to the [] pressure direction during pressure sintering. The forging process is performed by first pressing in the same direction as the pressing direction during pressure sintering, and then in the direction perpendicular to the pressing direction during pressure sintering. It is possible to obtain the final forged ingot by applying pressure. Also in this case, the thermoelectric performance can be improved as in the above-described embodiment. Hereinafter, this embodiment will be described.

In the forging process of this example, a sintered ingot having a height of 30 mm, a width of 15 mm, and a length of 20 mm in the direction of extension was machined into a cemented carbide upsetting device shown in FIG. to Installation, in an argon atmosphere 4 5 0 ° with 2 5 0 k loading pressure of gZ cm ² at C, 6 hours, pressed in the direction that matches the upper and pressurizing direction at the time of sintering under pressure, Free extension only in one axis direction. Thereafter, the ingot was inverted 90 ° and pressed from above in a direction perpendicular to the pressing direction during pressure sintering.

 As a result, the sintered ingot was compressed, and the dimensions of the forged ingot after forging were 7.5 mm in height, 40 mm in width, and 80 mm in length in the spreading direction.

The following Table 38 compares the physical properties at the center and end of the sintered ingot (hot-pressed product) before forging and the forged ingot (hot forged product) heat-treated after hot forging. Shown in comparison. Table 38

As shown in Table 38, it can be seen that the figure of merit indicating the thermoelectric performance of the forged ingot after forging has been improved as compared to the sintered ingot before forging.

In the above-described embodiment, the forging process is performed by first pressing in the same direction as the pressing direction during the pressure sintering, and thereafter, the forging process is performed in a direction perpendicular to the pressing direction during the pressure sintering. By pressing, the final forged ingot is obtained, but the forging process is performed by first pressing in the direction perpendicular to the pressing direction during calopress sintering, and then pressing sintering The final forging ingot may be _obtained by pressing in the same direction as the pressing direction. Also in this case, the thermoelectric performance can be similarly improved. Hereinafter, this embodiment will be described.

In the forging process of the present embodiment, a sintered ingot having a height of 30 mm, a width of 15 mm, and a length of 2 O mm in the extending direction was machined into a cemented carbide upsetting apparatus shown in FIG. to Installation, in an argon atmosphere 4 5 0 ° with 2 5 0 loading pressure of kg / cm ² at C, 6 hours, pressurized in a direction perpendicular from the top to the pressing direction during sintering under pressure, Free extension only in one axis direction. Thereafter, the ingot was turned 90 ° and pressed from above in a direction corresponding to the pressing direction during pressure sintering.

 As a result, the sintered ingot was compressed, and the dimensions of the forged ingot after forging were 7.5 mm in height, 40 mm in width, and 80 mm in length in the spreading direction.

In this case, as in Table 38 above, it was confirmed that the performance index indicating the thermoelectric performance of the forged ingot after forging was improved as compared to the sintered ingot before forging. Next, as a third embodiment of the present invention, a method for forming a p-type element will be described. Bismuth B i, tellurium Te, the single element of antimony S b, those stoichiometric ratio B io. ₄ Sb and weighed so as to be ₆ T e _3, was further added to the appropriate amount of Te to adjust the carrier concentration , Dissolved, mixed and solidified to create a soluble material. This soluble material is pulverized with a stamp mill, a ball mill, or the like, passed through a 150-mesh or 400-mesh sieve, and the material remaining on the 400-mesh sieve is selected to have a powder having a particle size of about 34 to 10 6 / m. In the case of p-type material, the hydrogen reduction step was not performed because the effect of fine particles and powder oxidation was small. The powder was sintered by a hot press at a sintering temperature of 500 ° C and a pressure of 75 Okg / cm ² . The size of the sintered ingot was cut into a height of 30 mm, a width of 40 mm, and a length of 18 mm, and it was installed in the same upsetting apparatus as used in Example 2, and this was forged by hot upsetting forging. To

In the forging process, this ingot was installed in a cemented carbide upsetting device, as shown in Fig. 9, and pressed and sintered at 500 ° 〇 at 100 to 500 kg / cm ² for 5 hours. By pressing in the same direction as the pressing direction, the sintered ingot is compressed. Table 4 shows the thermoelectric performances of the hot forged ingot and the center of the ingot before forging. In the case of this p-type material, heat treatment was not performed because the carrier was not reduced by hot forging as compared with the n-type material. When heat treatment is performed, the carrier concentration becomes lower than that of the starting ingot.

 Table 4

Furthermore B 1 ₂ chome 6 ₂ in the same manner ₈₅ 36 _{0 15} and 8;.... 1 _{0 5} 313 ₁ for ₅ Te ₃ also measures the thermoelectric performance of Ingotto heart before forging and hot forging Ingo' DOO The results are shown in Tables 5 and 6, respectively. Table 5

 Table 6

Then, in this way B i ₂ T e ₂ is an n-type formed by the method of the second embodiment are. ₇ of Ingodzuto the S Eo.3 density less than 97%, using 80% of Ingodzuto Then, the wafer is sliced perpendicularly to the spreading direction to form a 1.33 mm thick wafer. Electrode metal layers were formed on the front and back surfaces of the wafer. Then, dicing was performed to form a chip of 0.64 mm square. An n-type element was randomly selected from these (see Table 3). Further, the p-type Bi formed by the method of the third embodiment. The AS b L ₆ Te ₃ ingot was machined into a chip of the same size and used as a P-type element (see Table 4). Then, 18 pairs of the pn element pair including the n © element and the p-type element were mounted to form a thermoelectric module as shown in FIG. Then, 16 thermoelectric modules were formed and the maximum temperature difference was measured. The average value of this maximum temperature difference was calculated, and the relationship between this and the heat radiation surface temperature is shown in FIG. For comparison, the same material was used, hot pressing was performed without hot upsetting and dicing was performed, and the thermoelectric module was formed in exactly the same manner as in the other steps. . The maximum temperature difference of the thermoelectric module formed by hot upsetting forging in the region where the heat radiation surface temperature is 0 ° (~ 80 ° C) is significantly higher than the maximum temperature difference of the module formed by hot pressing. , Tables 3 and 4 Enhance the improvement of thermoelectric performance by the structure. Here, the current value that gives the maximum temperature difference was 1.5 to 1.6 A for both modules. The standard deviation of the maximum temperature difference was 0.4 to 0.5 ° C. Furthermore, for example, when the heat radiation surface temperature is 27 ° C, the maximum temperature difference of the thermoelectric module formed by hot upholstery is over 75 ° C, which is an extremely excellent result.

 Here, the strength of the thermoelectric module formed as described above is examined.

 When a thermoelectric module breaks, shear stress is applied to the thermoelectric module, and the P-type and n-type elements are often broken.

 Therefore, as shown in Fig. 17, as a test material, a p-type element 5 and an n-type element 6 soldered to one side of a ceramic plate 15 obtained in the process of manufacturing a thermoelectric module were used, and these p-type elements were used. The shear strength of the element and the n-type element was measured.

 That is, as shown in the side view of FIG. 17 (a) and the partial perspective view of FIG. 17 (b), a push-pull gauge 16 is used to connect the p-type element 5 and the n-type element 6 to the base. It measures the shear strength when a 0.15 mm wire 17 is pulled up at a speed of 10 mm / min.

 Table 10 below shows the results of measuring the shear strength of P-type and n-type elements generated based on the forged ingots subjected to hot upsetting forging.

 Table 11 below shows the results of measuring the shear strength of p-type and n-type elements produced based on sintered ingots that were not hot-pressed and hot-pressed and forged with the same material.

Table 12 below shows that the same material is used instead of hot pressing to produce a unidirectionally solidified soluble material by the Stockberger method, and to measure the shear strength of P-type and n-type elements generated based on this soluble material. The results are shown. Table 10 Shear strength of elements using hot forging material

Table 11 Shear strength of devices using hot-press materials

Table 12 Shear strength of elements using ingot materials

Comparing these tables, the shear strength of the hot-forged material element and the hot-pressed material element of the n-type element 6 is larger than that of the ingot material element (compared to 117 6). 185, 191 4), and the strength of the hot forged material element is greater than that of the hot press material (2 185 for 191 4). . It can also be seen that the standard deviation of the shear strength decreases in the order of the element of the soluble material, the element of the hot breath material, and the element of the hot forging material (34). 224 for 7 and 158 for 224).

 On the other hand, as for the p-type element 5, the magnitude of the shear strength increases in the order of the element of the soluble material, the element of the hot press material, and the element of the hot forging material, though slightly different (as compared to 1413). 1430, 1430 vs. 1472). The standard deviation of the shear strength also decreases in the order of the element of the soluble material, the element of the hot press material, and the element of the hot forging material (132 for 429, 112 for 132).

 Here, the larger the standard deviation of the shear strength, the higher the possibility of fracture even below the average value of the shear strength, and the higher the probability of failure below the average value of the shear strength.

 Therefore, it can be concluded that the elements made of hot forged material not only have higher shear strength than elements of other materials, but also have a lower probability of fracture below the average value of shear strength and higher reliability. .

Therefore, by using a hot forging material for the thermoelectric module, damage during assembly of the module can be reduced, durability can be increased, and reliability can be improved. Also _{similarly, B i 2 Te 2. 85} Se. . ₁₅ For even and _{B io.sSbi. 5 T e 3,} ( see Table 5 and Table 6) in the same manner to form it it n-type device and p-type devices, have created a thermoelectric module. The average value of the maximum temperature difference of this thermoelectric module was calculated, and the relationship between this and the heat radiation surface temperature is shown by the curve a in FIG. For comparison, the curve b shows the result of forming a thermoelectric module by hot-pressing without hot upsetting and dicing after hot pressing without forging. In the region where the heat radiation surface temperature is between 0 ° C and 80 ° C, the maximum temperature difference of the thermoelectric module formed by hot upsetting forging exceeds the maximum temperature difference of the module formed by hot pressing. The current value giving the maximum temperature difference was 1.3 to 1.4 A for both modules.

 From a comparison between FIG. 13 and FIG. 14, although there are some changes depending on the material, it is understood that the present invention is effective in any case.

By the way, the present inventors have further conducted experiments, and as a result, the following parameters have been found to have an effect on performance. The outline is as follows. (1) Density ratio

 If the density ratio of the sintered ingot is low, the thermal conductivity will decrease, but the thermoelectric performance will decrease due to the increase in electrical resistance. Also, the strength of the material is reduced.

 After all, there is a range of density ratios that can improve thermoelectric performance and does not impair the strength of the material, and the sintered ingot finally has a density ratio of 97% or more by hot forging. It became clear that we should do it.

 In this case, the sintered ingot having a density ratio of 97% or more may be finally hot-forged so as to have a density higher than the density ratio, or a sintered ingot having a density ratio of less than 97%. Even so, the final hot-forging density ratio should be 97% or more.

 Here, the density ratio is the ratio of the green density to the true density (ideal density) of a substance (single crystal before being ground) having the same composition as the green compact.

 (2) Plastic working temperature

 In hot forging and warm forging, the recovery of crystal proceeds simultaneously with the crystal distortion accompanying the progress of plastic working. It is thought that in the hot plastic working of thermoelectric semiconductors, this crystal distortion and recovery, and also the flow of the powder grain boundaries and the crystal grain boundaries of the sintered material occur, but the details are still unknown.

 However, the higher the hot forge ratio, the more uniform the structure as viewed with a polarizing microscope, and it is clear that crystal distortion and recovery due to plastic working greatly contribute to the improvement of orientation. Here, when the temperature increases, crystal distortion and recovery are promoted, but at temperatures higher than the recrystallization temperature, crystal grains grow regardless of orientation, the degree of orientation decreases, and the material strength decreases. I will. The higher the temperature, the faster the ingot deforms. In other words, if the temperature is high, the tissue becomes fluid, plastically deforms before the orientation is aligned, and the orientation does not progress further.

 Conversely, if the temperature is too low, the plastic deformation itself becomes slow, which is not suitable for practical processing.

After all, there is a temperature range that is optimal for plastic working of sintered ingots, and it is necessary that the temperature be below the grain growth temperature at which crystallites increase and orientation is lost. It became clear.

 Specifically, it has been clarified that the temperature should be not higher than 550 ° C. and should be in the range of 350 ° C. or higher at which plastic deformation can be performed at a practical speed.

 It is desirable that the hot press temperature range and the plastic working temperature range be the same.

 (3) Plastic working load

 The greater the load applied to the sintered ingot, the faster the deformation rate. However, as the load increases, the frictional resistance increases and the ingot buckles. Also, when the load was weakened to reduce the deformation speed, no significant improvement in orientation was observed, probably because only the flow at the grain boundaries occurred. The rate of deformation is determined by the initial loading pressure applied to the sintered ingot first. The initial load pressure must be equal to or higher than the yield stress of the sintered ingot.

In the end, it was found that there was an appropriate range of initial loading pressure or loading pressure, and the range was less than 100 kg / cm ² .

 (4) Forging process

 Although the present invention basically presupposes the upsetting forging process, by appropriately adding a die-forging process to the upsetting forging process, the density (1: 1) of the above (1) can be reduced. It became clear that it could be improved.

 In other words, after the sintered ingot is spread in the free direction by upsetting forging, it is further pressed in a state where the free direction is regulated by a jig or the like, thereby recovering the once reduced density ratio, It has been found that the ratio can be improved.

 In addition, it is also clear that once the sintered ingot is hot upset forged and then subjected to the hot die forging process, it is also possible to recover the once lowered density ratio or to increase the density ratio. became.

 (5) Number of times of forging

 It was also found that by increasing the number of forging steps and sequentially increasing the compression ratio, the orientation was improved and the thermoelectric performance could be improved.

The above (1) to (5) will be described below with reference to specific examples. , Fourth embodiment

 First, a specific example of performance improvement when the density ratio in (1) is 97% or more will be described.

In this embodiment, B i ₂ T e ² of the same composition as the second embodiment described above. ₇ S e. ³ . The n-type thermoelectric semiconductor was pulverized to have an average particle diameter of 20 m, and this was hot-pressed at a sintering temperature of 500 ° C and a pressing force of 750 kg / cm ² to perform powder sintering. The average particle size (20 m) used was smaller than that of the second example.

Thereafter, as in the second embodiment, hot forging was performed by applying a load pressure shown in FIG. 15 at CD at 400 ° C. by an upsetting device that freely extends only in one axial direction shown in FIG. 9. Was. That is, the load pressure was set to 0, the initial load was set to 100 kg / cm ^2, and finally increased to 450 kg / cm ² .

 Thereafter, the forged ingot after the forging was annealed at 400 ° C. for 48 hours in an argon reduced glass sealed tube.

Table 7 below shows a comparison result between the second embodiment and the fourth embodiment. In Table 7, hot press 1 refers to the case where only the hot press is performed in the second embodiment, and the hot forging (hot forging) process is omitted, and hot forge 1 refers to the second embodiment. Shows the case where the hot forge process was performed. The hot press 2 is a case where only the hot press is performed in the fourth embodiment, and the hot forging (hot forging) process is omitted. In this example, the case of performing the hot forging process is shown. Table 7 Hot press 1 Hot forge 1 Hot press 2 Hot horn 2 Specific resistance (ηηΩ / cm) 0.950 0.950 1.283 1.056 Seebeck constant OiV / deg) 1 200 1 200 -192 1 192 Heat Conductivity 13.8 15.0 13.8 14.4 Resistance anisotropy ratio 1.7 2.25 1.01 2.42 Density ratio * 99.3% 96.8% 96.8% Figure of merit ( X10 one ^{3 / deg) 2. 40 2. 81} 2. 08 2. 45 Here, the anisotropy ratio of the resistance is a value indicating the directionality of the resistance, and the larger the value is, the more remarkable the effect of improving the crystal orientation is.

 In addition, as a criterion for evaluating the figure of merit Z, 2.45 was set as the threshold. When the figure of merit Z is 2.45 or more, it is determined that the thermoelectric performance has been improved.

 As is clear from the table, in the fourth embodiment, since a relatively fine powder (average particle size 2 O ^ m) was used, the density ratio of the sintered ingot subjected to hot forging was low (96 8%), the anisotropy ratio of the resistance indicating the degree of crystal orientation is low (1.01).

 Also, when this sintered ingot is forged in the uniaxial direction, the processing speed is lower than that of the second embodiment, although the forging temperature is lower (450 ° C. compared to 450 ° C. of the second embodiment). It turns out that it is fast (see Figures 10 and 15).

 The density ratio of the forged ingot after hot forging in the fourth embodiment is the same as that of the sintered ingot before forging (96.8%), but the figure of merit is improved (2.45). You can see that it is. This is thought to be because the crystal orientation was improved by forging. This is why the anisotropy ratio of resistance has increased (2.42).

 However, as compared with the forged ingot after hot forging in the second embodiment, the figure of merit is lower due to the lower density ratio (2.85 compared to 2.81 in the second embodiment). You can see that.

 As described above, in the fourth embodiment, although the anisotropy ratio of resistance is improved, the figure of merit is lower than that of the second embodiment as follows. It is explained in.

 That is, the figure of merit Z is determined by both the resistance anisotropy ratio and the density ratio. The contribution ratio of the density ratio is larger than the contribution ratio of the resistance anisotropy ratio.

 Therefore, in the case of the fourth embodiment in which the density ratio is smaller than 97%, despite the improvement in the resistance anisotropy ratio, should the performance index Z be the evaluation standard of 2.45? Will not rise to a value that exceeds.

Based on the above, the density ratio of sintered ingot before hot forging was 97% or more. It was concluded that setting a value is desirable for improving thermoelectric performance. It was concluded that the density ratio of the forged ingot finally obtained by hot forging should be 97% or more. Then, it was concluded that it is desirable to finally increase the density ratio of the sintered ingot before forging by hot forging a sintered ingot having a density ratio of 13: a 97% or more.

Even if the sintered ingot has a density ratio of less than 97%, it is possible to improve the thermoelectric performance without deteriorating the strength by forming a forged ingot having a density ratio of 97% or more. it can. Next, this embodiment will be described. In this embodiment, B i ₂ T e ₂ of the same composition as the second embodiment described above. ₇ S e. ₃ . The n-type thermoelectric semiconductor was pulverized with an average particle size of 40 zm, and was subjected to powder sintering by hot pressing at a sintering temperature of 300 ° (: 1000 kg / cm ² ).

After that, the powder sintered ingot was cut into dimensions of (length in the direction of extension) X (width) X (thickness) of 15 x 40 x 33 mm, and the same as in the second embodiment. Then, hot forging was performed at 450 ° C by a swaging device that can freely spread only in the uniaxial direction as shown in Fig. 9 and spread in the direction of 40 mm in length. In this case, the range of the applied pressure was 250 kg / cm ² to 1000 kg / cm ² .

 Thereafter, the forged ingot after the forging was annealed at 400 ° C. for 48 hours in an argon reduced glass sealed tube.

 Tables 35 and 36 below show the results of comparison of the physical properties of the sintered ingot before forging and the forged ingot after forging.

table:

 t 'Fuku coefficient density ratio

(μ Y / deg) (% ) anisotropy ratio X10- ³ (w / craK ²⁾

1.709 -196.7-89.4 2.26 2.26 Table 3

Here, the power factor is a value obtained by dividing the square of the Seebeck coefficient by the resistivity, and the larger this value is, the better the thermoelectric performance is. As the evaluation criteria, those with a noise factor of 3.2 or more were judged to have “good thermoelectric performance”.

 The sintered ingot before forging shown in Table 35 has a low density ratio (89.4%), the power factor value indicating thermoelectric performance is less than 3.2, and the thermoelectric performance is not good. On the other hand, the forged ingot after forging shown in Table 36 has a density ratio of 97% or more in all parts of the ingot by hot upsetting forging. As a result, the value of the power factor indicating the thermoelectric performance was 3.2 or more, indicating that the thermoelectric performance was improved.

 • Fifth embodiment

 Next, a specific example of the performance improvement by increasing the number of times of the forging step (5) will be described.

In this embodiment, B i ₂ T e ₂ of the same composition as the second embodiment described above. ₇ S eQ. _3. This n-type thermoelectric semiconductor material was sintered by the same manufacturing method as in the second example. As shown in Fig. 16, the sintered ingot 14 obtained at this time had a height (thickness) of 60 mm, a width of 40 mm, and a length in the extending direction of 40 mm.

By hot-forging the sintered ingot 14 once, twice, and three times, a forging material having a large compression ratio is sequentially generated. That is, the compression ratio becomes 1/2 by the first forging, and the compression ratio becomes 1/8 by the second forging. By the third forging, the compression ratio becomes 1/16.

 Thereafter, the forged ingot after the final forging was subjected to a heat treatment at 400 ° C. for 24 hours in an atmosphere substituted with argon.

 Table 8 below shows the average value of the thermoelectric performance and the resistance anisotropy in the part where the density ratio is 97% or more in the forged ingot.

 Table 8

The data for one forging cycle (compression ratio 1/5) shown in Table 8 is based on the average value of the parts with a density ratio of 97% or more in the curtain ingot obtained in the second example. I have.

 As shown in the table, as the compression ratio increases, the amount of plastic deformation increases, the orientation is improved, and the thermoelectric performance improves (from a performance index of 2.19 to 2.52, from 2.52 to 2.55). To) You can see that

 Also, a slight change in carrier concentration occurs due to an increase in the compression ratio. This can be understood by seeing that the Zebeck constant changes according to the compression ratio. Despite the anisotropy ratio of resistance being large for forged ingots that were forged multiple times (2.52, 2.55 for 2.19 for the first forged ingot) The figure of merit does not increase significantly (2.61, 2.62 vs. 2.56 for the first forged ingot) because the composition of the sintered ingot as the starting material maximizes the thermoelectric performance. This is probably because the forging ingot that was forged multiple times had a composition that deviated from this optimum carrier concentration, while the carrier concentration was formed with the optimum carrier concentration that could be drawn out.

However, this is due to the fact that the carrier of the sintered ingot of the starting material has a concentration suitable for the compression ratio. It can be solved by changing to

 , Sixth embodiment

 Next, a specific example of performance improvement by the number of times of forging is performed for a p-type material as in the fifth embodiment.

In this embodiment, the third embodiment of the same composition as _{B i 0. 4 S b 6 T} e 3 types thermoelectric semiconductor material described above was sintered in the third embodiment and the same process. As shown in FIG. 16, the sintered ingot 14 obtained at this time had a height (thickness) of 60 mm, a width of 40 mm, and a length in the extending direction of 40 mm.

 By hot-forging the sintered ingot 14 once, twice and three times, a forging material having a large compression ratio is sequentially produced. In other words, the compression ratio becomes 1/2 by the first forging, the compression ratio becomes 1/8 by the second forging, and the compression ratio becomes 1/16 by the third forging .

 Thereafter, the forged ingot after the final forging was subjected to a heat treatment at 400 ° C. for 24 hours in an atmosphere substituted with argon.

 Table 9 below shows the average value of the thermoelectric performance and the resistance anisotropy of the part where the density ratio is 97% or more in the forged ingot. Table 9

The data for one forging cycle (compression ratio 1/5) shown in Table 9 are based on the average value of the part with a density ratio of 97% or more in the forged ingot obtained in the third example. I have.

As shown in the table, as the compression ratio increases, the plastic deformation increases and the orientation is more improved. It can be seen that the thermoelectric performance has been improved (performance index from 3.2 to 3.22 and from 3.22 to 3.35).

 • Seventh embodiment

Next, the effect of the hot press sintering temperature on the performance will be examined. In this embodiment, B i ₂ T e ₂ of the same composition as the second embodiment described above. ₇ S e. The so-type n-type thermoelectric semiconductor was pulverized to an average particle size of 40 μm, and was hot-pressed at a pressing force of 75 Okg / cm ² to perform powder sintering. Hot press 400. C, 450 ° C., 500 ° C., and 550 ° C. were performed at four sintering temperatures.

After that, similarly to the second embodiment, pressurization was performed at 450 ° C. with a load pressure of 100 kg / cm ^{2 to} 450 kg / cm ² by an upsetting device that freely extends only in one axial direction shown in 9. .

 Thereafter, the forged ingot after the forging was annealed at 400 ° C. for 24 hours in an argon reduced glass sealed tube.

 The physical properties of the central part of the forged ingot (“forged product” in the table) and the sintered ingot after hot pressing (“pressed product” in the table) were obtained by sintering at 400 ° C and 450 ° C. C, 500 ° C, 550. For each C, it is shown in the following Tables 13, 14, 15, and 16.

Table 13 Hot breath temperature 400 ° C

 Breath product Hoge product

 Specific resistance (τηΩ / cm) 1.93 1.62

 Seebeck constant OiV / deg) -236 -235

 Thermal conductivity (mW / cmdeg) 13. 8 14. 1

Figure of merit ³ / d _e g) 2.09 2.58

 Anisotropy ratio of resistance 1.87 2.38

Density ratio (%) 98. 4 99. 6 Table 14

 Hot press temperature 450

Table 15

 Hot press temperature 500

Table 16

Hot

As is clear from these tables, forged ingots that have been hot upset forged are more than 99% regardless of the production conditions (sintering temperature conditions) of the sintered ingot, which is the starting material for the forging process. It can be seen that the material strength and thermoelectric performance are improved. Furthermore, the sintering ingot of the starting material, which has a density ratio of about 98% after hot pressing (when the sintering temperature is 400 ° C or 450 ° C), is subjected to hot upsetting forging. (When the sintering temperature is 400 ° C, it is improved from 98.4% to 99.6%. When the sintering temperature is 450 ° C, it is 98.5% to 99.9%.) To improve).

 Regarding the crystal orientation (anisotropy ratio of resistance), hot upsetting forging is performed regardless of the conditions (sintering temperature conditions) of the sintered ingot, which is the starting material of the forging process. , Improvement, and improvement.

 If the hot pressing temperature is high, the structure changes due to recrystallization during hot pressing, and the anisotropy ratio of the resistance decreases (the difference in the resistance of the sintered ingot at a sintering temperature of 550 ° C). The anisotropy ratio is 1.34). For this reason, when hot forging is performed using a sintered ingot having such a reduced resistance anisotropy ratio, the orientation is improved (from 1.34 to 2.25). Since the orientation itself at the start of forging is low (the anisotropy ratio of the resistance of the sintered ingot is low), the figure of merit is low (the evaluation criterion is 2.45 or less in 2.43). For example, the performance index of a forged ingot at a sintering temperature of 550 ° C is 2.43 (the resistance anisotropy ratio of the sintered ingot is 1.34), and the performance index of a forged ingot at a sintering temperature of 500 ° C is The index is lower than 2.65 (the resistance anisotropy ratio of the sintered ingot is 1.83).

 '' Eighth embodiment

 Next, as described in (4) above, a specific example that can improve the density ratio and improve the thermoelectric performance by adding a die-forging-like process to the upsetting forging process will be described. .

In this embodiment, B i ₂ Te ₂ having the same composition as the first actual TU previously described. ₇ S e. ₃ . The n-type thermoelectric semiconductor material was sintered by the same manufacturing method (sintering temperature: 500 ° C., pressure: 750 kg / cm ² ) as in the first and second embodiments. From the powder sintered body thus obtained, as in the second embodiment, a sintered ingot having a height (thickness) of 30 mm, a width of 40 mm, and a length in the extending direction of 18 mm was obtained. By cutting out the individual pieces and making the forging process different for them, as in the second embodiment, only one axial direction shown in FIG. 9 is free. Hot forging was performed at 450 ° C. using a spreading upsetting device.

Fig. 18 shows that one of the two sintered ingots was placed in the middle of spreading, and a jig added to the upsetting device (the spreading direction was regulated by a wall, and the hot Changes in ingot height (solid line) and load pressure change (dashed line) when pressure is continued, that is, when the ingot's extended end is constrained to the jig wall in the second half of upsetting forging. (Approximately 3 and 15 minutes after the start of forging), the load pressure was further increased to 450 kg / cm2, and the pressurization was continued for about 5 hours. The ^second half of the forging process was performed by die forging. It is something to do.

 As shown in the figure, it can be seen that in the latter half of the forging process, there is no change in the height of the ingot because no spreading takes place.

 On the other hand, Fig. 19 shows that the other sintered ingot of the two sintered ingots was spread in the same manner, and the spread was stopped before the spread end came into contact with the jig. It shows the change in ingot height and the change in load pressure when the stop was performed.

Incidentally, FIG. 1 8, 1 9 together, the initial heavy pressure was ² 5 0 kg / cm 2. Observation of the forged ingot obtained in the forging process shown in Fig. 18 showed that the end was formed into a square shape and the surface was smooth because it was pressed for a long time using a forging die. On the other hand, the forged ingot obtained by the process in which forging was stopped halfway as shown in Fig. 19 had an arc-shaped end and many fine cracks on the surface.

 From each of the two forging ingots 14 obtained in these different forging processes, measurement beads shown in FIG. 20 were cut out, and for each measurement bead, the resistivity, the density ratio, the anisotropy ratio of the resistivity, Physical property values such as one Beck constant and power factor were measured. The results are shown in Tables 17 and 18 below. The power factor is the value obtained by dividing the square of the Seebeck constant by the resistivity, and the larger the value, the better the thermoelectric performance. As the evaluation criteria, a hot press product (before forging) of a comparative example having an n-type power factor of 3.2 or more was determined to have “good thermoelectric performance”.

There was almost no variation in the physical property values in the height direction and the width direction of the measurement bead (forged ingot 14), but there was variation in the distribution of the physical properties in the spreading direction D. The average was taken in the width direction, and the physical properties were shown at each position in the spreading direction D. Tables 17 and 18 show these.

 Table 17 shows the physical property values of the forged ingot obtained by the forging process shown in FIG. 18 at each distance in the spreading direction D from the center C of the forged ingot 14 shown in FIG.

 Table 18 also shows the physical properties of the forged ingot obtained in the forging process shown in Fig. 19 for each distance in the spreading direction D from the center C of the curtain ingot 14. 1

Table 18

Resistivity from the center portion Seebeck constant density ratio, the distance of the power factor of the resistivity (mm) (mQcm) (μν / deg) (%) Anisotropic ^{ffit XI 0 - 5 (w /} cmK 2)

 2 0.993 -195 97.7 2.34 3.82

 6 0.992 -193 97.6 2.31 3.77

 10 1.005 -193 97.4 2.23 3.70

1 1.026 -192 97.0 2.12 3.59

18 1.023 -190 97.2 2.06 3.54

22 1.122 -188 96.5 1.85 3.16 As shown in these tables, in the forged ingot of Table 17 in which the latter half was made by die forging, the density ratio of each part of the ingot was 98% or more, and if only this was seen, the material strength and thermoelectric performance improved. You can see that

 However, the anisotropy ratio of the resistance decreases from the center of the ingot to the end, and it can be seen that the orientation of the crystallites gradually becomes less uniform. Also, as with the anisotropy ratio of resistance, the Seepeck constant also increases in absolute value from the center of the ingot to the end.

,

It can be seen that the resistivity decreases and the resistivity increases toward the end. The reason why the Seebeck constant decreases toward the end is thought to be that lattice defects due to forging increase toward the end. Also, the reason why the resistivity increases toward the end is considered to be that the orientation of crystal grains deteriorates toward the end.

 In addition, the power factor decreases from the center of the ingot to the end, and this indicates that the crystal orientation deteriorates toward the end and the thermoelectric performance decreases. However, up to a distance of 26 mm from the center of the ingot, the power factor maintained a value exceeding 3.6, which was the evaluation criterion, of 3.2 or more, and the range of use as a thermoelectric material was wide. You can see that there is.

On the other hand, in the forged ingot of Table 18 obtained by stopping the forging before the die forging, it can be seen that the density ratio is lower than that of Table 17. Since the density ratio at a distance exceeding 22 mm from the center of the ingot is too low (density ratio is less than 97%), Table 18 does not show the data at a distance exceeding 22 mm. I have. At the center of the ingot (for example, at a distance of 2 mm from the center of the ingot), although the anisotropy ratio of resistance and the Seebeck constant are almost the same as the values in Table 17, the resistivity increases. (0.93 vs. 0.948) because the density ratio is lower than that in Table 17 (97.7 vs. 98.8) it is conceivable that. The reason why the power factor is lower than that in Table 17 (3.892 compared to 3.99) is also considered to be due to the effect of the decrease in the density ratio. Only at a distance of 10 mm from the center of the ingot, a power factor value of 3.6 or more can be maintained, indicating that the range of use as thermoelectric material is narrower than Table 17 O

 From these comparison results, it is understood that the density ratio in addition to the orientation greatly affects the thermoelectric performance.

 Also, considering the density ratio shown in Table 18 as the density ratio up to the middle of the spreading, and the density ratio shown in Table 17 as the density ratio in the case of further die forging from the middle of the spreading, It can be seen that even if the density ratio is reduced by the above, the density ratio can be improved and recovered by further die forging.

 This recovery of the density ratio was similarly confirmed for p-type materials with different compositions. Further, in the above description, it is assumed that the ingot is freely extended in one axial direction and then restrained in this uniaxial direction. However, as in the first embodiment, the ingot is freely extended in two axial directions. After the spreading, the spreading in these two directions may be restricted at the same time. FIG. 30 and FIG. 31 show the configuration of the upsetting device for restraining the spreading in both directions. Fig. 30 is a view showing the sintered ingot 14 in a free-spreading state in which the sintered ingot 14 is not restrained. It is a figure which shows the state in which was restrained. FIGS. 30 and 31 (a) are top views of the upsetting device, and FIGS. 30 and 31 (b) are side views of the upsetting device.

 That is, as shown in FIGS. 30 and 31, the upsetting device includes a die 20 on which the sintered ingot 14 is placed, and a punch for compressing the sintered ingot 14 from above. 18, and four side walls 19, which are driven as shown by the arrows, come into contact with the respective side surfaces of the sintered ingot 14, and prevent free spreading.

 In the first half of the forging process, hot forging is performed in the free spreading state shown in Fig. 30. In the second half of the forging process, four side walls 19 are driven as shown in Fig. 31 and these four sides are driven. Hot forging is continued in a state where the biaxial free extension of the sintered ingot 14 is prevented by the wall 19 (the state where the punch 18 is lowered).

Thus, even when the free spreading in the biaxial direction is prevented in the latter half of the forging process, the density of the forged ingot is reduced in the same manner as in the above-described embodiment in which the free spreading in the uniaxial direction is prevented in the latter half of the forging process. Ratio of more than 97% to improve thermoelectric performance Can be.

 By the way, in the above-described embodiment, the latter half of the forging process is performed by die forging. However, it is also possible to perform the die forging process after performing the upsetting forging process. The case where this is performed using the upsetting apparatus shown in FIGS. 30 and 31 will be described below.

 In other words, in the first upsetting forging process, hot upsetting forging is performed in a free spreading state as shown in FIG. Then, the punch 18 is once raised to bring the sintered ingot 14 into a non-compressed state. In the next die forging process, as shown in Fig. 31, the four side walls 19 are driven, and the four side walls 19 prevent the sintered ingot 14 from free spreading in the biaxial direction. The punch 18 is lowered again. Thus, the sintered ingot 14 is vertically compressed, and hot die forging is performed.

 The hot upsetting forging step and the hot die forging step may be repeated a plurality of times.

 Thus, even when the die forging process is performed following the upsetting forging process, the density ratio of the forged ingot is increased to 97% or more, as in the above-described embodiment in which free spreading is prevented in the latter half of the forging process. The thermoelectric performance can be improved. • Ninth embodiment

 Next, specific examples of the range of the plastic working temperature at which the thermoelectric performance described in (2) can be improved and the range of the plastic working load at which the thermoelectric performance described in (3) can be improved will be described. explain.

Table 19 below shows the ingot deformation rate when hot forging was performed under the conditions 1 to 7 with the temperature, initial load pressure, and ¾ changed. The hot pressing process is the same as in the second embodiment. In the hot forging process, the height (thickness) is 30 mm and the width is 40 mm, as in the second embodiment. A sintered ingot having a length of 18 mm in the direction was cut out and hot forged under the conditions shown in Table 19 below using an upsetting device shown in FIG. Table 19

Here, condition 1 is data when the hot forging step is performed under the same conditions as in the second embodiment. The changes in load pressure and ingot height during the forging process are as shown in FIG.

 Table 20 below shows the physical property values of the sintered ingot as a starting material for forging under each of the conditions 1 to 7. Table 20

Figure 21 shows the change in load pressure during the forging process (dashed line) under condition 3 and the ingot Fig. 22 shows the change in load pressure (dashed line) and the change in ingot height (solid line) during the forging process under condition 4, and Fig. 23 shows the change in condition. The change in load pressure during the forging process (break, change in ingot height (solid line)) for case 5 is shown in Fig. 24. The change in load pressure during the forging process (dashed line) for condition 6 with ingot is shown in Fig. 24. Figure 25 shows the change in load pressure (dashed line) and the change in ingot height (solid line) during the forging process under condition 7.

 As is clear from Table 19 above, the higher the temperature, the faster the ingot deformation speed.If the initial load pressure is the same, the deformation speed of the ingot becomes almost the same, but the firing of the forging starting material It can be seen from FIGS. 21 to 25 that buckling occurs depending on the shape of the ingot.

 In particular, when the deformation speed of the ingot was high, large buckling was observed. For example, when the ingot deformation speed is high in conditions 3 and 7, as shown in Figs. 21 and 25, it is clear that large buckling occurs.

 Buckling occurs because the upper and lower surfaces of the ingot become a rigid region due to the frictional force of the upper and lower surfaces, and the vicinity of the unconstrained surface becomes a deformation region. Is formed into barrels, which eventually result from the ingot being sheared at this boundary. Cracking etc. due to shearing is recovered by subsequent pressing or forming pressure (die forging) after free forging described in the eighth embodiment, but shearing must be prevented in advance to prevent buckling. There is.

 Buckling can be avoided in advance by using a sintered ingot that is a starting material for forging and having a low ingot height with respect to the pressed area. However, if it is desired to increase the housing ratio, using a sintered ingot of such a shape has a disadvantage.

Tables 21 to 27 below show the physical properties of the forged ingots after forging under the conditions 1 to 7 described above. Table 2 1

Distance from Table 22 the central portion a power factor of resistivity Ze "click constant density ratio resistivity (mm) (mQcm) OiV / deg) (%) anisotropy ratio x10- ⁵ (w / cmK ²⁾

 2 1.046 -193 98.8 2.38 3.63

6 1.044 -199 99.2 2.46. 3.78

10 1.058 -198 98.8 2.41 3.71

14 1.073 -198 99.2 2.34 3.64

18 1.068 -197 98.8 2.32 3.62

22 1.079 -19 98.8 2.21 3.49

26 1.120 -192 98.9 2.10 3.29 Table 23

Table 24

Resistivity Zeck constant Density ratio Resistivity power factor distance from center (mm) (mQcm) xV / deg) (%) Anisotropic ffit xlO- ⁵ (w / cmK ² )

 2 1.124 -197.2 99.5 1.98 3.46

6 1.081 -197.0 99.5 2.01 3.59

10 1.1 11 1 195.5- 99.4 2.00 3.45

14 1.095 -195.6 99.5 2.01 3.49

18 1.419 -191.5 97.4 1.39 2.59

22 1.425 -190.7 97.6 1.08 2.55

26 1.593 1 190.6 94.9 1.09 2.29

30 2.027 -189.0 98.1 0.66 1.76 Table 25

Table 26

Distance from the center of the power factor of the Seebeck constant density ratio resistivity (mm) xV / deg) ( %) Anisotropic ¾ !: spoon x10- ⁵ (w / cmK ²⁾

 2 1.051 -204 99.5 2.45 3.95

6 1.077 -203 98.9 2.42 3.83

10 1.086 -203 99.2 2.29 3.83

14 1.115 -202 99.2 2.26 3.67

18.1.101 -203 99.5 2.20 3.75

22 .1.118 -203 99.1 2.16 3.68

26 1.132 -202 98.6 2.10 3.59

30 1.173 -199 98.0 1.93 3.38 Table 27

 That is, after forging under the respective conditions 1 to 7, heat treatment was performed at 400 ° C. for 24 hours, and the measurement pieces shown in FIG. 20 were cut out from the forged ingots 14 obtained, and each measurement piece was cut out. For, physical properties such as resistivity, density ratio, resistivity anisotropy ratio, Seebeck constant, and power factor were measured. The results are shown in Tables 21 to 27 above.

 There was almost no variation in the physical property values in the height direction and the width direction of the measurement bead (forged ingot 14), but there was variation in the distribution of the physical properties in the spreading direction D. In Tables 21 to 27, the physical properties are shown for each distance in the spreading direction D by averaging in the width direction.

 Table 20 compares the anisotropy ratio of the resistance of the sintered ingot as the starting material for forging and the values of the anisotropy ratio of the resistance of the forged ingot after forging shown in Tables 21 to 27. As can be seen, the anisotropy ratio of the resistance at the center of the forged ingot was larger than that before the forging in each of the conditions 1 to 7, indicating that the orientation of the crystal grains was improved by the forging. . For example, in the case of condition 2, the resistance anisotropy ratio is increased from 1.0 to 2.38 at a position 2 mm from the center of the ingot.

In addition, it is presumed that the power factor was increased by forging in each of conditions l to 7, and that the thermoelectric performance was improved by improving the orientation of crystal grains. For example, in the case of condition 2, the distance from 3.22 to 3.63 at a position 2 mm from the center of the ingot The power factor is increasing.

 In this respect, in the case of conditions 3 and 7, the orientation was improved only at a part of the center of the forged ingot.Comparative results between Table 20 and Table 23, and Table 20 This is clear from the comparison results in Table 27. For example, the power factor of Condition 3 after forging is larger than the value before forging (3.53) only in the portion 6 mm from the forged ingot center (3.79).

 This means that, as described above, large buckling occurs during the port forging process (see FIGS. 21 and 25), the ingot breaks and separates, and at the time of this separation, the separation part rotates and the part is rotated. This is probably because the orientation of the crystal changed.

 However, even under such forging conditions where buckling is likely to occur, buckling can be avoided and orientation can be improved by using a sintered ingot with a shape that does not easily cause buckling as described above. It is. This is because the orientation of the center of the ingot, which is less susceptible to buckling, is originally good.

 Regarding the forging temperature, the higher the temperature, the easier the plastic deformation proceeds, but it must be below the grain growth temperature at which the crystal grains grow and lose their orientation. Specifically, from the results of the ninth embodiment and the first to eighth embodiments, it is desirable that the temperature is not higher than 550 ° C.

 Conversely, if the forging temperature is low, plastic deformation will be slow and impractical, but forging is possible at temperatures that allow hot press sintering. Specifically, from the results of the ninth embodiment and the first to eighth embodiments, the temperature is desirably 350 ° C. or more.

Regarding the load pressure during forging, it is necessary that the initial weight of the load be equal to or higher than the yield stress of the sintered ingot. In addition, the load pressure must be such that the ingot deformation speed is lower than the ingot deformation speed at which buckling occurs. Specifically, from the results of the ninth embodiment and the first to eighth embodiments, the range is preferably from 70 kg / cm ^{2 to} 350 kg / cm ² .

In addition, as shown in FIGS. 21 to 25, in this embodiment, buckling is minimized by changing the load pressure according to the shape (height) change of the ingot during forging. Yo I'm trying. The results of this ninth embodiment and the eighth embodiment, in order to avoid buckling pressurizes the sintered Ingo' preparative loading pressure exceeding the forging in 500 kg / dm ² is I knew it wouldn't.

However, initial load pressure is in the range of 70 kg / cm ² or more 350 kg / cm ² or less, because the load pressure not exceeding forged in 500 kg / cm ² is desired, sintering Ingo' bets are buckling This is the case for dimensions that are likely to occur. For example, in a sintered ingot having a height of 30 mm, a width of 40 mm, and a length of 18 mm in the extending direction assumed in the above-described embodiment, such a range of the initial load pressure and the load pressure is desirable.

However, as described below, in the case of sintering Ingo' preparative hardly occurs buckling dimensions, the initial load pressure, be in the range of 100 Okg / cm ² or less to the load pressure both during forging, A forged ingot having high thermoelectric performance can be obtained without buckling. Hereinafter, this embodiment will be described.

In the forging process of the present embodiment, a sintered ingot having a size of 2 Omm in height, 15 mm in width, and 4 Omm in length in the extending direction was installed in a cemented carbide upsetting apparatus shown in FIG. And 450 in an argon atmosphere. C, with a load pressure of 1000 kg / cm ² , press for 1.5 hours from the top in the direction that matches the pressing direction during pressure sintering, and freely expand only in one axial direction.

 As a result, the sintered ingot was compressed, and the dimensions of the forged ingot after forging were 1 Omm in height, 15 mm in width, and 8 Omm in length in the spreading direction.

 Table 39 below shows the physical properties of the sintered ingot (hot pressed product) before forging and the forged ingot (hot forged product) heat-treated after hot forging at the center and at the end.

 Table 3, Hot Breath Hot Forge, Part, Hot Forge Cham constant (μν / Κ) —192 —192 —192 Specific resistance (m Q cm) 1.150 0.990 1.06

 Conductivity (mW / cmIO 14 14.5 14.3

Figure of merit (10 ^A -3 / ¾ 2.29 2.57 2.43 As shown in Table 39, it can be seen that the figure of merit indicating the thermoelectric performance of the forged ingot after forging has been improved as compared to the sintered ingot before forging.

 Another embodiment will be described below.

In the forging process of this embodiment, a sintered ingot having a height of 80 mm, a width of 80 mm, and a length of 4 O mm in the direction in which it was extended was machined into a cemented carbide upsetting device shown in FIG. and Installation, in an argon atmosphere with a load pressure of 5◦ kg / cm ² at 4 5 0 ° C, 6 hours, pressed in the direction that matches the upper and pressurizing direction at the time of sintering under pressure, 1 Free extension only in the axial direction.

 As a result, the sintered ingot was compressed, and the dimensions of the forged ingot after forging were 20 mm in height, 80 mm in width, and 160 mm in length in the spreading direction.

 Table 40 below shows the physical properties at the center of the sintered ingot before hot forging (hot pressed product) and the hot forged ingot after hot forging (hot forged product). table ·.

As shown in Table 40, it can be seen that the performance index indicating the thermoelectric performance of the forged ingot after forging is improved as compared with the sintered ingot before forging.

 Another embodiment will be described below.

In the forging process of the present example, a sintered ingot having a height of 20 mm, a width of 15 mm, and a length of 40 mm in the direction of extension was inserted into a cemented carbide upsetting apparatus shown in FIG. to Installation, in an argon atmosphere 4 5 0 ° with a load pressure of 5 0 0 kg / cm ² at C, 1. 5 hours, in a direction that matches the upper and pressurizing direction during pressure sintering additive Press and freely expand only in one axis direction. As a result, the sintered ingot was compressed, and the dimensions of the forged ingot after forging became 10 mm in height, 15 mm in width, and 80 mm in length in the spreading direction.

Table 41 below compares the physical properties of the sintered ingot (hot pressed product) before forging and the forged ingot (hot forged product) heat treated after hot forging at the center and at the end. table

As shown in Table 41, it can be seen that the figure of merit indicating the thermoelectric performance of the forged ingot after forging has been improved as compared to the sintered ingot before forging.

 Next, a description will be given of an example in which forging conditions were similarly changed for a ρ-type material.

That is, Bi having the same composition as that of the third embodiment. . The ₄ S b ₆ T e ₃ of ρ-type thermoelectric semiconductor material was sintered in the third embodiment and the same process. Then, as in the third embodiment, a sintered ingot having a height (thickness) of 30 mm, a width of 40 mm, and a length of 18 mm in the extending direction was cut out. Hot forging was performed under the conditions 8 to 12 shown in Table 28 below using the upsetting device that can freely expand only in the uniaxial direction shown in Table 2 below.

Table 28 below shows the ingot deformation rate when hot forging was performed under the conditions 8 to 12 with the temperature and initial load pressure changed. Table 28

Here, the condition 8 is a time when the hot forging process is performed under the same conditions as the third embodiment.

 Table 29 below shows the physical property values of the sintered ingot as the starting material for forging for each of the conditions 8 to 12.

 Table 2 9

Figure 26 shows the change in load pressure during the forging process (dashed line) and the change in ingot height (solid line) under condition 9, and Figure 27 shows the load pressure during the forging process under condition 10. Figure 28 shows the change in load pressure (dashed line) and the change in ingot height (solid line) during the forging process under condition 11 (1). Fig. 29 shows the change in load pressure (broken line) and the change in ingot height (solid line) during the forging process under condition 12.

In particular, when the deformation speed of the ingot was high, large buckling was observed. For example, in the case of condition 10 where the ingot deformation speed is high, it is clear from Fig. 27. As can be seen, large buckling has occurred.

 Tables 30 to 34 below show the physical property values of the forged ingot after forging under the above conditions 8 to 12. Table 30

Table 3 1

From the center resistivity resistivity Seebeck constant density ratio resistivity power factor distance (mm) (mQcm) OiV / deg) (%) anisotropy ratio x10 " ⁵ ^ / cmK ² )

 2 0.804 191 99.9 1.52 4.51

 6.0.823 191 100.2 1.43 4.46

 10 0.863 191 98.7 1.28 4.21

 14 0.871 192 100.7 1.35 4.25

 18 0.91 1 195 98.0 1.33 4.16

 22 0.963 195 96.1 1.26 3.95

26 1.072 196 95.2 1.16 3.60 Table 3 2

Table 3 3

From the center Resistivity Specific constant Density ratio Distance of resistivity power factor (mm) (mQcm) OiV / deg) (%) Anisotropy ratio x10 ~ ⁵ (w / cmK ² )

 2 0.846 197 99.9 1.57 4.60

6 0.850 197 98.4 1.57 4.56

10 0.879 196 98.6 1.49 4.39

14 0.927 197 99.1 1.38 4.18

18 0.914 197 98.7 1.41 4.23

22 1.950 196 97.8 1.52 4.03

26 1.191 197 93.0 1.14 3.25

Table 3 4

That is, after forging under the respective conditions 8 12, the measurement beads shown in FIG. 20 were cut out from each of the obtained forged ingots 14, and the difference in resistivity, density ratio, and resistivity was measured for each measurement piece. Physical properties such as anisotropy ratio, Seebeck number, and power factor were measured. The results are shown in Tables 30 to 34 above.

 Although there was almost no variation in the physical property values of the measuring piece (forged ingot 14) in the height direction and the width direction, the distribution of the physical property values in the spreading direction D varied. In Table 304, the average is taken in the width direction, and the physical properties are shown for each distance in the spreading direction D.

 It can be seen by comparing the anisotropy ratio of the resistance of the sintered ingot which is the starting material for forging in Table 29 with the anisotropy ratio of the resistance of the forged ingot after forging shown in Table 334. Thus, the anisotropy ratio of the resistance at the center of the forged ingot was larger than that before the forging under each of the conditions 8 1 and 2, indicating that the orientation of the crystallite was improved by the forging.

 In addition, in the case of p-type material, the range of improvement in orientation is larger at the end of the forged ingot in the case of p-type material than in n-type material. It can be seen that is improved.

In the fourth to ninth embodiments, the Bi ₂ T-based thermoelectric semiconductor material is mainly used. However, similar results can be obtained when the present invention is applied to a BiSb-based thermoelectric semiconductor material.

In the first to ninth embodiments described above, B i ₂ T e ₃ based thermoelectric semiconductor and B i have been described S b based thermoelectric semiconductor, it can be applied to the thermoelectric semiconductor material having other rhombohedral is there.

 In the present embodiment, the description has been made mainly on the assumption that the powder sintered body obtained by hot pressing (pressure sintering) is hot upset forged. However, the present invention is not limited to this. It is not done.

 The pressurized body obtained by pressing the solid solution powder may be hot upset forged, or the sintered body obtained by pressing the solid solution powder and then sintered may be hot upset forged. Further, as shown in the comparative example, a material obtained by melting and solidifying a material (melted material) may be cut out as a desired solid solution block, and this may be directly subjected to hot upsetting forging.

 Further, in the present embodiment, description has been made on the assumption that the thermoelectric semiconductor material is obtained by hot forging, but the hot forging method described in the present embodiment can be applied to any material. The materials to which the hot forging of the present invention is applied include magnetic materials, dielectric materials, and superconductor materials having a hexagonal structure, a layered structure, or a tungsten bronze structure, such as a bismuth layered structure ferroelectric and bismuth. High-temperature superconductors having a layered structure are exemplified.

 As described above, according to the method of the present invention, it is possible to obtain a thermoelectric semiconductor material having a high orientation and a high production yield.

Claims

The scope of the claims

1. A heating step of mixing and heating and melting the material powders to have a desired composition; a solidification step of forming a solid solution ingot of a thermoelectric semiconductor material having a rhombohedral structure (hexagonal structure);

 A crushing step of crushing the solid solution ingot to form a solid solution powder,

 A sizing step for uniformizing the particle diameter of the solid solution powder,

 A sintering step of pressure-sintering the solid solution powder having a uniform particle size,

 The powder sintered body is plastically deformed by heat and spread while being pressed perpendicularly to the pressing direction in the sintering step, so that the crystal grains of the powder sintered structure are brought into a crystal orientation having an excellent figure of merit. A method for producing a thermoelectric semiconductor material, comprising: a hot upsetting forging step for orienting.

 2. A heating step of mixing and heating and melting the material powders to have a desired composition, a solidification step of forming a solid solution ingot of a thermoelectric semiconductor material having a rhombohedral structure (hexagonal structure),

 A crushing step of crushing the solid solution ingot to form a solid solution powder,

 A sizing step for uniformizing the particle diameter of the solid solution powder,

 A sintering step of pressure-sintering the solid solution powder having a uniform particle size,

 The powder sintered body is plastically deformed by heating, and is spread while being pressed in a direction coinciding with the pressing direction in the sintering step, and is then spread in a direction perpendicular to the pressing direction in the sintering step. A hot upsetting forging process for orienting the crystal grains of the powder sintered structure in a crystal orientation having an excellent figure of merit.

3. A heating step of mixing and heating and melting the material powders to have a desired composition; a solidification step of forming a solid solution ingot of a thermoelectric semiconductor material having a rhombohedral structure (hexagonal structure);

 A crushing step of crushing the solid solution ingot to form a solid solution powder,

A sizing step for uniformizing the particle diameter of the solid solution powder, A sintering step of pressure-sintering the solid solution powder having a uniform particle size,

 This powder sintered body is plastically deformed by heat, spread while pressing in a direction perpendicular to the pressing direction in the sintering step, and then spread in a direction coinciding with the pressing direction in the sintering step. A hot upsetting forging process for orienting the crystal grains of the powdered sintered structure in a crystal orientation having an excellent figure of merit.

 4. A heating step of mixing and heating and melting the material powders to have a desired composition, and a solidification step of forming a solid solution ingot of a thermoelectric semiconductor material having a rhombohedral structure (hexagonal structure).

 A crushing step of crushing the solid solution ingot to form a solid solution powder,

 A sizing step for uniformizing the particle diameter of the solid solution powder,

 A sintering step of pressure-sintering the solid solution powder having a uniform particle size,

This powder sintered body is plastically deformed and spread by hot under an initial load pressure of 1000 kg / cm ² or less, so that the crystal grains of the powder sintered structure have excellent performance index. A hot upsetting forging process for orienting in a direction.

 5. A heating step of mixing and heating and melting the material powders so as to have a desired composition; a solidification step of forming a solid solution ingot of a thermoelectric semiconductor material having a rhombohedral structure (hexagonal structure);

 A crushing step of crushing the solid solution ingot to form a solid solution powder,

 A sizing step for uniformizing the particle diameter of the solid solution powder,

 A sintering step of pressure-sintering the solid solution powder having a uniform particle size,

The powder sintered body is plastically deformed and spread under heat at a load pressure of 100 kg / cm ² or less, so that the crystal grains of the powder sintered structure have a crystal orientation with an excellent figure of merit. A hot upsetting forging process for orienting the thermoelectric semiconductor material.