JP2016008339A - Cu-Ga alloy sputtering target - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a Cu-Ga alloy sputtering target having a high Ga ratio and high strength.SOLUTION: The Cu-Ga alloy sputtering target formed of Cu-Ga alloy including Ga of more than 22 at% and less than 30 at% on an average and the remainder consisting of Cu and inevitable impurities has a columnar structure consisting of mixed phases of γ and ζ phases having Ga dissolved to Cu. Out of the mixed phases observed by the COMPO image of a backscattered electron image, the median of an aspect ratio of a major axis length to a minor axis length in a phase (an inclusion phase) intervened between the structures is 5-60.

Description

  The present invention relates to a Cu-Ga alloy sputtering target. In particular, the present invention relates to a Cu-Ga alloy sputtering target used when forming a Cu-In-Ga-Se (hereinafter referred to as CIGS) quaternary alloy thin film which is a light absorption layer of a thin film solar cell layer. .

  In recent years, mass production of CIGS solar cells having high photoelectric conversion efficiency as thin film solar cells has progressed. A CIGS thin film solar cell generally has a structure in which a back electrode, a light absorption layer, a buffer layer, a transparent electrode, and the like are sequentially laminated on a substrate. As a method for producing the light absorption layer, an evaporation method and a selenization method are known. Solar cells manufactured by vapor deposition have advantages of high conversion efficiency, but have disadvantages of low film formation speed, high cost, and low productivity, and selenization is more suitable for industrial mass production.

  The outline process of the selenization method is as follows. First, a molybdenum electrode layer is formed on a soda lime glass substrate, a Cu—Ga layer and an In layer are formed thereon by sputtering, and then a CIGS layer is formed by high-temperature treatment in selenium hydride gas. A Cu—Ga alloy sputtering target is used during the sputter deposition of the Cu—Ga layer during the CIGS layer formation process by this selenization method.

  As the shape of the sputtering target, there are a flat plate type and a cylindrical type. Since the entire surface of the cylindrical target is eroded by rotating around the cylindrical axis, the material utilization efficiency is higher than that of the flat target, and the cooling efficiency is improved by continuously changing the plasma irradiation surface. Therefore, output can be kept high and mass productivity is high. However, the cylindrical target is more difficult to manufacture because the shape is more complicated than the flat target, and the risk of cracking and chipping during the manufacturing increases. If cracks or defects occur during sputtering, the fragments or cracks generated thereby can cause generation of particles or abnormal discharge. In addition, the flat target is required to have a high strength that is not easily damaged during transportation or sputtering.

  Here, as a method for producing a Cu—Ga alloy target, a melt casting method and a powder sintering method are known. Inevitable pores remain in the powder sintering method. The vacancies not only cause abnormal discharge, but also make it difficult to increase the density and cause cracking and chipping during cutting and sputtering. JP 2013-138232 A (Patent Document 1) discloses a method of forming a two-phase structure by mixing and sintering a high-concentration Ga powder and a low-concentration Ga powder in order to prevent segregation that causes cracking. Although disclosed, the process is complex and expensive.

  On the other hand, regarding the melt casting method, JP-A-2000-073163 (Patent Document 2) discloses a Cu—Ga alloy cast by the melt method with a Ga composition of 15 wt% to 70 wt%, As a method for producing a Cu—Ga alloy, a method is described in which casting is performed by a melting method by controlling the temperature to a cooling rate that does not cause brittle cracks and segregation by a mold having heating means and cooling means. Since the Cu—Ga alloy obtained by this method is not brittle and segregated, it is easy to form and can be processed into an arbitrary shape.

  Japanese Patent Laying-Open No. 2013-76129 (Patent Document 3) describes a sputtering target of a Cu—Ga alloy having a Ga concentration of 27 wt% or more and 30 wt% or less formed in a cylindrical shape by melt casting. It is described that the structure of the sputtering target is characterized by being equiaxial in a cut surface cut in parallel to the solidified surface of the sputtering target. It is also described that the sputtering target can be mass-produced with high quality.

  Japanese Patent Laying-Open No. 2013-204081 (Patent Document 4) discloses a molten and cast plate-like Cu—Ga alloy sputtering target in which Ga is 15 at% or more and 22 at% or less, and the balance is Cu and inevitable impurities. Yes. The Cu-Ga alloy has a structure composed of an α phase in which Ga is dissolved in Cu or a mixed phase of an α phase and a ζ phase, and a dendritic structure in a structure formed of the α phase or a mixed phase of the α phase and the ζ phase. The dendrite structure is composed of a primary arm and a secondary arm that grows laterally from the primary arm, the average length of the secondary arm is 30 to 60 μm, and the average width of the secondary arm is A Cu—Ga alloy sputtering target has been proposed which is 10 to 30 μm, and the average distance between the secondary arms is 20 to 80 μm.

JP 2013-138232 A JP 2000-073163 A JP2013-76129A JP2013-204081A

  In producing a cylindrical target, it is considered that the melt casting method is more suitable than the powder sintering method, but in any of the above-mentioned documents, the consideration on the strength of the target is insufficient.

Patent Document 2 describes that the temperature is controlled to a cooling rate that does not cause brittle cracking and segregation, but the control of only the cooling rate cannot suppress the formation of shrinkage cavities that cause abnormal discharge during sputtering. Because it is difficult to keep the solidification rate constant during solidification in the casting method in which hot water is poured, even if solidified in one direction from the mold bottom, the solidification rate is reduced by the solidification latent heat released at the top of the mold, This is because shrinkage nests occur frequently. Furthermore, Patent Document 2 describes that the cooling rate is controlled in the range of 1.0 × 10 ⁻¹ ° C./sec to 1.5 × 10 ⁻² ° C./sec, but the cooling rate is slow. The crystal structure obtained at the cooling rate is equiaxed. High strength cannot be obtained with equiaxed crystals. Patent Document 2 does not describe a cylindrical target.

  Although a cylindrical target is specifically described in Patent Document 3, a target having sufficient strength cannot be obtained because the crystal structure is an equiaxed crystal as in Patent Document 2.

  Although it is described that the sputtering target described in Patent Document 4 has a dendrite structure, since the Ga concentration is low, manufacturing and processing are relatively easy, and necessary strength is easily obtained. Recently, there is an increasing need for Cu—Ga alloys having a high Ga ratio. A target having a high Ga ratio tends to be easily broken, but Patent Document 4 does not make any study for increasing the strength of a Cu-Ga alloy having a high Ga ratio. Moreover, there is no description about the cylindrical target.

  The present invention was created in view of the above circumstances, and an object of the present invention is to provide a Cu-Ga alloy having a high Ga ratio and a high-strength sputtering target, particularly a cylindrical sputtering target.

  The present inventor has intensively studied to solve the above problems. As a result, when the Cu—Ga alloy is a columnar crystal and has an intervening phase having a predetermined aspect ratio, the Cu—Ga alloy has a high Ga ratio. However, the present inventors have found that strength is easily developed and completed the present invention.

  Therefore, in one aspect, the present invention is a Cu-Ga alloy sputtering target made of a Cu-Ga alloy containing Ga on average exceeding 22 at% to less than 30 at%, with the balance being Cu and inevitable impurities. A short axis length in a phase intervening in the structure (intervening phase) among the mixed phases observed in the COMPO image of the reflected electron image, having a columnar structure composed of a mixed phase of γ phase and ζ phase in which Ga is dissolved in Cu This is a sputtering target having a median of the aspect ratio of the major axis length to the thickness of 5 to 60.

  In one embodiment of the sputtering target according to the present invention, the value obtained by subtracting the standard deviation from the average value of the three-point bending strength is 300 MPa or more.

  In another embodiment, the sputtering target according to the present invention has a relative density of 99 to 100%.

  In yet another embodiment, the sputtering target according to the present invention has a plate shape or a cylindrical shape.

  In still another embodiment, the sputtering target according to the present invention has a cylindrical shape.

  According to the present invention, a sputtering target having a high bending strength and a Cu-Ga alloy having a high Ga ratio can be obtained. The effect of the sputtering target according to the present invention appears more remarkably when it is made cylindrical. The sputtering target according to the present invention is not easily damaged during transportation or sputtering, and is excellent in practicality.

It is an example of the optical microscope photograph of the Cu-Ga alloy sputtering target cross section which concerns on this invention. It is an example of the reflected electron image (COMPO image) by EPMA of the cross section of the Cu-Ga alloy sputtering target which concerns on this invention (magnification 200 times). It is a schematic diagram of a dendritic tissue. It is an example of the reflection electron image (COMPO image) by EPMA of the cross section of the conventional Cu-Ga alloy sputtering target (magnification 200 times). It is a schematic diagram of an amoeba-like structure. It is a schematic diagram which shows the structure of the vertical type continuous casting apparatus used in the Example. It is a schematic diagram which shows the structure of the gravity casting apparatus used by the comparative example. It is a schematic diagram which shows the relationship between the average value (micro | micron | mu) in a normal distribution, and a standard deviation ((sigma)).

(composition)
In one embodiment, the Cu—Ga alloy sputtering target according to the present invention contains an average of more than 22 at% to less than 30 at% Ga, typically 25 at% to 29 at%, with the balance being Cu and inevitable. It has a composition consisting of impurities. The Ga content is derived from a request for forming a Cu—Ga alloy sputtered film that is required when manufacturing a CIGS solar cell. In the present invention, however, the Ga content is relatively high. One of the features is the setting point. As can be understood from the Cu—Ga binary phase diagram, as the Ga content increases, the proportion of γ phase from ζ phase increases. However, γ phase is more easily cracked than ζ phase, ensuring strength. Is difficult. In the present invention, since the crystal structure and the aspect ratio of the phase intervening in the structure of these two phases are appropriately controlled, it has succeeded in obtaining high strength even with such a high Ga content. . From the Cu-Ga binary phase diagram, the γ phase becomes dominant when the Ga content is 26.7 at% or more, and thus the strength improvement effect according to the present invention is particularly significant. Is 26.7 at% or more.

(Crystal structure)
In one embodiment, the Cu—Ga alloy sputtering target according to the present invention can have columnar crystals composed of a mixed phase of γ phase or ζ phase in which Ga is dissolved in Cu. By using columnar crystals, it is possible to have higher strength than equiaxed crystals. The columnar crystal can be confirmed by observing a linear grain boundary by macro observation as in the optical micrograph illustrated in FIG. Moreover, it can be understood from the Cu—Ga binary phase diagram that the Cu—Ga alloy according to the present invention is a mixed phase of γ phase and ζ phase. The γ phase is hard and brittle as a single substance, but a flexible structure can be obtained by using a mixed phase with a relatively soft ζ phase. As described below, the aspect ratio is increased by elongating the ζ phase or γ phase forming the phase intervening in the tissue (hereinafter referred to as “intervening phase”), which is characteristic of the flexible tissue. Thus, a high bending strength can be obtained. This intervening phase refers to a portion inside a closed region surrounded by a curve observed when a cross section is observed.

(Organization)
When the microstructure of the cross section of the Cu—Ga alloy sputtering target according to the present invention is observed by a reflected electron image (COMPO image) of EPMA, a two-phase structure as shown in FIG. 2 can be confirmed (the Ga concentration of the Cu—Ga alloy is 28 at%). In FIG. 2, it is confirmed by EBSP analysis that the black part which is the intervening phase is the ζ phase and the white part is the γ phase. The structure representing the black ζ phase has a dendritic shape, and a structure having a primary arm and a secondary arm that grows laterally from the primary arm is confirmed. A schematic diagram of the dendritic tissue is shown in FIG. According to the results of the study by the present inventors, the aspect ratio of the major axis length to the minor axis length in the intervening phase has a significant effect on the target strength. Specifically, it is good when the median aspect ratio is 5 or more, preferably 10 or more, more preferably 20 or more, even more preferably 30 or more, even more preferably 40 or more, and even more preferably 50 or more. Strength can be obtained. However, if the aspect ratio is excessively increased, shrinkage cavities are likely to be generated, and conversely, the strength may be lowered. The intervening phase may be a γ phase in a concentration range where the Ga concentration is more than 22 at% to less than 30 at%.

  FIG. 4 shows an example of a reflected electron image (COMPO image) by EPMA of a cross section of a conventional Cu—Ga alloy sputtering target (the Ga concentration of the Cu—Ga alloy is 28 at%). As can be seen from FIG. 4, an amoeba-like intervening phase having an unspecified shape is dominant in the alloy structure. Also in the intervening phase, it is confirmed by EBSP analysis that the black portion is the ζ phase and the white portion is the γ phase. FIG. 5 shows a schematic diagram of the amoeba-like intervening phase.

In the present specification, the aspect ratio of the major axis length to the minor axis length in the intervening phase is defined as follows, regardless of the shape, the major axis length (y) and the minor axis length (x). To measure.
The major axis length (y) is defined as the diameter of the smallest circle that can surround one continuous intervening phase, and the minor axis length (x) is the maximum that can be surrounded by one continuous intervening phase. It is defined as the diameter of the circle.

  In this way, the major axis length (y) and minor axis length (x) of each intervening phase is obtained, the aspect ratio in each intervening phase is calculated, and the median aspect ratio in 100 or more intervening phases is measured. Value. In addition, the intervening phase for which the aspect ratio is calculated has a major axis (y) of 50 μm or more and a minor axis (x) of less than 50 μm. This is because the intervening phase deviating from this is considered not to contribute much to the improvement of the bending strength.

(Relative density)
In general, the target for sintered products is a relative density of 95% or more. When the relative density is low, when the internal vacancies are exposed during sputtering, the generation of particles and surface irregularities on the film due to splash and abnormal discharge starting from the periphery of the vacancies progress early, and surface protrusions (nodules) This is because an abnormal discharge or the like starting from () is likely to occur. The cast product can achieve a relative density of almost 100%, and as a result, it has an effect that the generation of particles due to the difference in sputtering can be suppressed. This is one of the major advantages of castings. Since the Cu—Ga alloy sputtering target according to the present invention can be manufactured by casting, it can have a high relative density. For example, in one embodiment, the Cu—Ga alloy sputtering target according to the present invention can be 99% or more, preferably 99.5% or more, and more preferably 100%. For example, it may be 99 to 100%.

(Folding strength)
In one embodiment, the Cu—Ga alloy sputtering target according to the present invention has an average value of three-point bending strength measured in accordance with JIS R1601: 2008 of 340 MPa or more. In a preferred embodiment of the Cu—Ga alloy sputtering target according to the present invention, an average value of three-point bending strength measured in accordance with JIS R1601: 2008 is 350 MPa or more. In a more preferred embodiment of the Cu—Ga alloy sputtering target according to the present invention, the average value of the three-point bending strength measured according to JIS R1601: 2008 is 360 MPa or more. In a still more preferred embodiment, the Cu—Ga alloy sputtering target according to the present invention has an average value of three-point bending strength measured in accordance with JIS R1601: 2008 of 370 MPa or more. In a still more preferred embodiment, the Cu—Ga alloy sputtering target according to the present invention has an average value of three-point bending strength measured in accordance with JIS R1601: 2008 of 380 MPa or more. In a typical embodiment of the Cu—Ga alloy sputtering target according to the present invention, the average value of the three-point bending strength measured in accordance with JIS R1601: 2008 is 340 to 410 MPa.

  Further, the smaller the standard deviation of the three-point bending strength, the better the quality stability and the better. Therefore, in the Cu-Ga alloy sputtering target according to the present invention, in one embodiment, the standard deviation of the three-point bending strength is 40 MPa or less. can do. In a preferred embodiment of the Cu—Ga alloy sputtering target according to the present invention, the standard deviation of the three-point bending strength can be within a range of 30 MPa or less, for example, 10 to 40 MPa, typically 20 It can be set to -30 MPa.

  Therefore, in one embodiment of the Cu—Ga alloy sputtering target according to the present invention, a value obtained by subtracting the standard deviation from the average value of the three-point bending strength can be set to 300 MPa or more. In a preferred embodiment of the Cu—Ga alloy sputtering target according to the present invention, the value obtained by subtracting the standard deviation from the average value of the three-point bending strength can be 310 MPa or more. In a more preferred embodiment of the Cu—Ga alloy sputtering target according to the present invention, a value obtained by subtracting the standard deviation from the average value of the three-point bending strength can be set to 320 MPa or more. In a more preferred embodiment of the Cu—Ga alloy sputtering target according to the present invention, the value obtained by subtracting the standard deviation from the average value of the three-point bending strength can be set to 330 MPa or more. In a more preferred embodiment of the Cu—Ga alloy sputtering target according to the present invention, a value obtained by subtracting the standard deviation from the average value of the three-point bending strength can be set to 340 MPa or more. In a typical embodiment of the Cu—Ga alloy sputtering target according to the present invention, a value obtained by subtracting the standard deviation from the average value of the three-point bending strength can be set to 300 to 380 MPa.

  The significance of defining a value obtained by subtracting the standard deviation from the average value will be supplemented. For example, the value obtained by subtracting the standard deviation from the average value of the three-point bending strength is 300 MPa or more when the three-point bending strength in the same target follows a normal distribution (see FIG. 8). “2/3 of all data is 300 MPa or more”. Even if the average value is the same, if the variation in the measured value is large, the quality assurance value must be reduced. However, according to the Cu-Ga alloy sputtering target according to the present invention, the variation in the measured value is Even in consideration of the above, a practical three-point bending strength can be secured. By comparing values obtained by subtracting the standard deviation from the average value of the three-point bending strength, a difference in strength from the gravity cast product in which the measured value tends to vary significantly appears.

  The Cu—Ga alloy sputtering target according to the present invention can be provided as, for example, a plate shape or a cylindrical shape. Moreover, since it has high strength, it can be easily processed into a desired shape.

(Casting method)
The example of the suitable manufacturing method of the Cu-Ga alloy sputtering target which concerns on this invention is demonstrated. The Cu—Ga alloy sputtering target according to the present invention can be manufactured using, for example, a vertical continuous casting apparatus 30 having a structure shown in FIG. 6 provided with a high frequency induction heater, a graphite crucible, and a water cooling probe. The target raw material is melted in the graphite crucible 31, and this molten metal 38 is continuously poured into the mold 20 and cooled in accordance with the drawing of the pulling member 34 installed at the bottom of the crucible, so that Cu—Ga An alloy casting (billet) 39 is continuously produced. Depending on the shape of the pull-down member 34, the shape of the casting 39 can be changed. For example, if the pulling member 34 is cylindrical, a cylindrical casting 39 is obtained, and if the pulling member 34 is flat, a flat casting 39 is obtained. The obtained casting 39 can be further machined and polished to obtain a sputtering target of a desired shape Cu—Ga alloy.

  A water-cooled copper jacket 33, which is a cooling unit for cooling the casting space from the outer peripheral side, is provided on the outer peripheral side of the crucible 31. At this time, since the refrigerant does not directly contact the molten metal 38, there is no risk of a steam explosion even if a hot water leak occurs. The crucible 31 is provided with an inert gas introduction part 42 for introducing an inert gas, and the oxygen partial pressure in the molten metal 38 is reduced.

  A heater 45 is provided on the outer periphery of the crucible 31. A crucible temperature control thermocouple 44 is provided on the wall of the crucible 31. A molten metal temperature measuring thermocouple 43 for measuring the molten metal temperature at the molten metal supply portion that supplies the molten metal 38 from the crucible 31 to the casting space passes through the upper surface of the columnar core 32 while being accommodated in a predetermined protective tube. It passes through the thermocouple protection tube insertion port formed as described above and reaches the molten metal supply site. A plurality of refrigerant probes 46 such as water for cooling the casting space from the inner peripheral side are inserted into the core 32 from the refrigerant probe insertion port 36 in a concentric manner. The vertical continuous casting apparatus 30 forms a cast product 39 by cooling and solidifying the molten metal 38 supplied between the mold 20 and the core 32 disposed inside the mold 20 directly from the metal melting furnace. The casting member is continuously extracted from the mold 20 and the core 32 by the extraction mechanism 47, thereby continuously producing a cast product.

  Here, in order to control the crystal structure and the intervening phase, and further to prevent the shrinkage and ensure the strength, it is important to control the drawing rate of the cast product and the cooling rate [° C./sec] at the solidification interface. is there. By increasing the drawing speed, it is possible to progress unidirectional solidification and grow columnar crystals. In addition, the intervening phase is also affected by the cooling rate. If the cooling rate in the unidirectional solidification is high, an advantage is obtained that the elongated and fine intervening phase grows rapidly, and thus the crystal is difficult to break.

  Specifically, the drawing speed is preferably 30 to 120 mm / min, more preferably 60 to 120 mm / min, and even more preferably 90 to 120 mm / min. The average cooling rate is preferably 1.7 to 14.5 ° C / sec, more preferably 3.3 to 14.5 ° C / sec, and more preferably 5.0 to 14.5 ° C / sec. Even more preferably.

  The pulling operation can be performed while repeatedly driving and stopping the pulling mechanism. In the present invention, the drawing speed refers to a value calculated from the length of the cast product drawn with respect to the entire time of driving and stopping. The drawing speed can be changed by controlling the rotation speed of the pinch roller 48 in the drawing mechanism. Even if the drawing speed is the same, if the balance between driving and stopping is not good, a desired tissue may not be obtained. The driving time and the stopping time are, for example, driving time / stop time = 0.1 to 0.3. Typically 0.15 to 0.25. In addition, the cooling rate can be controlled by changing the drawing speed. Cooling speed of solidification interface = [temperature gradient (° C./mm)]×[drawing speed (mm / min)]. This equation means that the cooling rate increases in proportion to the drawing rate when the temperature gradient is constant. The temperature gradient is obtained from the temperature measurement distance of the thermocouple inserted into the mold and the core and the temperature difference. Specifically, a graph (horizontal axis: thermocouple position vs. vertical axis: temperature) was created by connecting the measurement points with a straight line, and a temperature gradient in the range of melting point ± 30 ° C. was obtained.

  Examples for better understanding of the present invention and its advantages will be shown below, but the present invention is not limited to these examples.

(1. Vertical continuous casting: Examples 1-9, Comparative Examples 1-2)
Cylindrical Cu-Ga alloy having an outer shape of 159 mm, a thickness of 14 mm, and a height of 620 mm using a vertical continuous casting apparatus having the structure shown in FIG. 6 equipped with a resistance heater (graphite element), a graphite crucible and a water-cooled probe. A sputtering target was manufactured.

  35 kg of Cu—Ga alloy raw materials with respective Ga concentrations shown in Table 1 (Cu purity is 4N, Ga purity is 4N) were introduced into the crucible, and the inside of the crucible was placed in an argon gas atmosphere and heated to 1250 ° C. This high-temperature heating is for welding the cylindrical pulling member installed at the bottom of the crucible and the molten Cu-Ga alloy.

  After the raw material was melted, the molten metal temperature was lowered to 1080 ° C., and when the molten metal temperature and the crucible temperature were stabilized, drawing of the pull-down member was started. By pulling out the pulling member, the solidified cylindrical casting was continuously drawn. In the drawing pattern, the drawing mechanism was operated by repeating driving for 0.5 seconds and stopping for 2.5 seconds, and the frequency was changed to change the drawing speed, thereby changing the cooling rate. At the time of drawing, the drawing speed was limited to 120 mm / min or less so as not to increase the cooling speed in order to prevent the formation of shrinkage cavities near the solidification interface.

<Crystal structure>
A cross section parallel to the solidification direction and the central axis direction of the cylinder was polished, etched with nitric acid and hydrochloric acid, and observed visually and with an optical microscope. As illustrated in FIG. 1, a case where a grain boundary solidified and grown from a heat removal portion of a cylindrical ingot collides near the center of the plate thickness is regarded as a columnar crystal, and a case where the grain boundary is distributed in a spot shape is an equiaxed crystal. (Here, the heat removal portion means a mold, a core, and a cooling space in contact with the ingot).

<Aspect ratio of intervening phase>
The microstructure of the cross section (cross section perpendicular to the solidification direction) of the obtained sputtering target was observed with a reflected electron image (COMPO image) of EPMA (manufactured by JEOL Ltd., apparatus name: XJA-8500F). The black part is the ζ phase, and the white part is the γ phase. The median aspect ratio of the intervening phase was measured according to the definition described above.

<Relative density>
The density of the obtained sputtering target was measured by the Archimedes method, and the ratio (%) with respect to the theoretical density determined by the composition was determined to obtain the relative density.

<Folding strength>
The three-point bending strength of the obtained sputtering target was measured based on JIS R1601: 2008. The test jig was set at 3p-30. Ten test pieces were cut out from each target and measured, and an average value and a standard deviation were obtained. The measurement was performed by applying pressure in a direction perpendicular to the length direction using a square piece cut in the length direction of the target as a test piece. The length direction refers to the target installation direction, that is, the direction of the backing plate or the backing tube.

(2. Gravity casting: Comparative Examples 3 to 5)
A cylindrical Cu—Ga alloy sputtering target having an outer diameter of 162 mm, a thickness of 18 mm, and a height of 630 mm was manufactured using the gravity casting apparatus 50 shown in FIG. 7 equipped with a graphite crucible 51, a tundish 52, and a mold 53. 44 kg of Cu—Ga alloy raw material (Cu purity is 4N, Ga purity is 4N) was introduced into the crucible 51, and the casting apparatus 50 was heated to 1300 ° C. in a vacuum atmosphere of about 10 Pa. Next, the molten metal in the crucible 51 was poured into the mold via the tundish 52.

  Since the molten metal poured into the mold from the tundish 52 jumps at the bottom of the mold, holes may remain in the lower part of the ingot. Moreover, since the solidification latent heat released as heat is extracted from the bottom of the mold and solidification progresses upward, shrinkage cavities tend to occur frequently at the top of the ingot. Therefore, in quality evaluation, sampling was performed from a position 350 mm from the bottom of the ingot.

  Here, the cooling rate was obtained by monitoring the temperature change of the thermocouple (installed at 300 mm and 600 mm from the bottom) inserted into the mold and drawing a graph of temperature vs time. As the temperature of the poured hot water decreases, the solidification latent heat is released and the temperature gradient on the graph becomes gentler. The temperature gradient becomes steep again as the latent heat is removed. The slope of the tangent at the inflection point of the graph curve showing the above behavior was defined as the cooling rate [° C./sec] at the position of the thermocouple. Therefore, the cooling rate is a measured value at each thermocouple position. The cooling rate described in the table is the average of the measured values obtained.

  With respect to the obtained cylindrical sputtering target, the crystal structure, aspect ratio, relative density, and bending strength were evaluated in the same manner as described above.

(3. Discussion)
The results are shown in Table 1. The sputtering targets of Examples 1 to 9 having a large aspect ratio compared to the comparative example had high bending strength. It is considered that an improvement in strength that could not be obtained with a single phase was obtained by extending and spreading the phase interposing the structure in the phase existing outside. In other words, the soft structure intervenes so as to connect the brittle structure, or vice versa, and the effect of the softness of the ζ phase to compensate for the brittleness of the γ phase structure makes it possible to reduce the brittleness and hardness in the single phase. It is thought that is not reflected in the bending strength. In Comparative Examples 4 and 5, although the cooling rate itself was high, a high aspect ratio could not be obtained because unidirectional solidification did not occur due to gravity casting.

20 Mold 30 Vertical Continuous Casting Device 31 Crucible 32 Core 33 Water-cooled Copper Jacket 34 Pull-down Member 36 Refrigerant Probe Insertion Port 38 Molten Metal 39 Cast Product (Billette)
42 Inert gas introduction part 43 Thermocouple for measuring molten metal temperature 44 Thermocouple for controlling crucible temperature 45 Heater 46 Refrigerant probe 47 Pulling mechanism 48 Pinch roller 50 Gravity casting device 51 Crucible 52 Tundish 53 Mold

Claims (5)

  A Cu—Ga alloy sputtering target containing an average of more than 22 at% to less than 30 at% Ga, the balance being Cu and an inevitable impurity Cu—Ga alloy sputtering target, in which Ga is a solid solution in Cu The aspect ratio of the major axis length to the minor axis length in the phase intervening in the structure (intervening phase) among the mixed phases observed in the COMPO image of the reflected electron image, which has a columnar structure composed of a mixed phase of the phase and the ζ phase Sputtering target whose median is 5-60.   The sputtering target according to claim 1, wherein a value obtained by subtracting a standard deviation from an average value of the three-point bending strength is 300 MPa or more.   The sputtering target according to claim 1 or 2, wherein the relative density is 99 to 100%.   It is plate shape or cylindrical shape, The sputtering target as described in any one of Claims 1-3.   The sputtering target according to claim 4, which has a cylindrical shape.