Tantalum sputtering target, method for manufacturing same, and barrier film for semiconductor wiring formed by using target

Abstract

Provided is a tantalum sputtering target, which is characterized that an average crystal grain size of the target is 50 m or more and 200 m or less, and variation of a crystal grain size in the target plane is 40% or higher and 60% or less. This invention aims to provide a tantalum sputtering target capable of improving the uniformity of the film thickness and reducing the variation of the resistance value (sheet resistance).

Claims

1. A tantalum sputtering target, wherein an average crystal grain size of the target is 50 m or more and 200 m or less, and variation of a crystal grain size in the target plane is 40% or higher and 50% or less on a surface of the target, wherein the variation is defined as a ratio of standard deviation to an average of five average crystal grain sizes measured at five different locations selected on the surface of the sputtering target.

2. A method of producing a tantalum sputtering target wherein an ingot or a billet obtained by subjecting a tantalum raw material to electron beam melting and casting is heat treated at a temperature of 900 C. or higher and 1400 C. lower, subsequently subjected to cold forging processing consisting of upset forging and extend forging by a total amount of true strain of 3.0 or more and 5.0 or less, thereafter heat-treated at 850 C. or higher and 1100 C. lower subsequently rolled at a rolling reduction of 80% or higher and 90% or less, thereafter heat-treated at 750 C. or higher and 1000 C. lower, and additionally subject to finish processing to obtain a target, wherein an average crystal grain size of the target is 50 micrometers or more and 200 micrometers or less, and a variation of crystal grain size in a target plane is 40% or higher and 50% or less on a surface of the target, wherein the variation is defined as a ratio of standard deviation to an average of five average crystal grain sizes measured at five different locations selected on the surface of the sputtering target.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a schematic diagram showing the locations where the crystal grain size of the target was measured in the Examples and Comparative Examples of the present invention.

(2) FIG. 2 is a schematic diagram showing the locations where the sheet resistance value and the film thickness of the thin film were measured in the Examples and Comparative Examples of the present invention.

DETAILED DESCRIPTION

(3) The tantalum sputtering target of the present invention can be produced based on the following processes.

(4) Foremost, high purity tantalum having a purity of 99.99% or higher is prepared as a tantalum raw material. While there is no particular limitation regarding the purity, high purity tantalum is preferably used in order to acquire favorable device properties.

(5) Next, the tantalum raw material is melted via electron beam melting or other methods, and then cast to prepare an ingot or a billet. Subsequently, the ingot or the billet is subject to a series of processes such as heat treatment, forging, heat treatment, rolling, heat treatment, and finish processing.

(6) For example, the sputtering target of the present invention can be obtained by subjecting the ingot to heat treatment at a temperature of 900 C. to 1400 C., cold forging (upset forging and extend forging), heat treatment at a temperature of 850 C. to 1100 C., cold (hot) rolling, heat treatment at a temperature of 750 C. to 1100 C., and finish processing (machine processing, polish processing, etc.).

(7) In the foregoing manufacturing processes, cold forging and the subsequent heat treatment may be performed as one cycle, but may also be repeated as needed. Moreover, cold (hot) rolling and the subsequent heat treatment may be performed as one cycle, but may also be repeated as needed. It is thereby possible to effectively reduce the structural defects of the target.

(8) The cast structure can be destroyed and pores and segregations can be diffused or eliminated by forging or rolling the ingot, and, by additionally heat-treating the forged or rolled ingot, recrystallization is achieved, and by repeating the foregoing forging or rolling and heat treatment processes, it is possible to densify and refine the target structure and increase the target strength.

(9) In the present invention, in order to control the variation in the average crystal grain size of the target, it is desirable to perform forging so that the true strain becomes 3.0 or more and 5.0 or less. Here, true strain is calculated from the ratio of the height of the ingot or billet before and after forging. For example, when the thickness of the billet before forging is a, and the thickness of the billet after forging is b, the true strain can be calculated as ln(a/b) (ln is a natural logarithmic). When the foregoing cold forging and subsequent heat treatment are repeatedly performed, the true strain is evaluated based on the integrated value ln(a/b) of the overall forging process.

(10) Moreover, in the present invention, the heat-treated forged product is desirably subject to cold (hot) rolling at a rolling reduction of 80% or higher and 90% or less. It is thereby possible to eliminate the non-recrystallized structure, and adjust the variation in the crystal grain size to be within the intended range.

(11) Moreover, in the present invention, it is necessary to destroy the cast structure based on casting and rolling as well as sufficiently perform recrystallization in the series of manufacturing processes. In the present invention, it is desirable to refine and uniformize the structure by subjecting the melted and cast tantalum ingot or billet to post-casting heat treatment at a temperature of 900 C. to 1400 C., and post-rolling heat treatment at a temperature of 750 C. to 1000 C.

(12) With the structure of the tantalum sputtering target obtained as described above, the average crystal grain size will be 50 m or more and 200 m or less, and the variation of the crystal grain size in the target plane will be 40% or higher and 60% or less. Accordingly, the present invention can obtain a tantalum sputtering target in which the variation in the crystal grain size of the target is controlled.

(13) In the present invention, the variation in the crystal grain size is 40% or higher and 60% or less, and more preferably 40% or higher 50% or less. When the variation in the crystal grain size is too great, the variation in the film thickness uniformity and specific resistance of the thin film obtained via sputter deposition will deteriorate. Meanwhile, by leaving a certain level of variation in the crystal grain size, the target structure will become a duplex grain structure, and stable sputtering characteristics can be obtained.

(14) Variation in the crystal grain size was obtained by measuring the crystal grain size at five locations in the target plane, calculating the average value and standard deviation thereof, and attaining variation (%)=standard deviation/average value100.

(15) Note that the crystal grain size was measured by using analySIS FIVE (Soft Imaging System) to analyze the crystal structure photograph taken with an optical microscope in a visual field of 1500 m1200 m. The crystal grain size was calculated in accordance with the average crystal grain size area method of ASTM, and the area of the crystal grains was converted into a circle, and the diameter of that circle was used as the crystal grain size. Moreover, the crystal grain size was measured at a total of five locations as shown in FIG. 1; namely, one location at the center of the target, two locations at the R (diameter) point, and two locations at the R point (outer peripheral part) were sampled.

(16) With the thin film formed by sputtering the target of the present invention, the variation in the sheet resistance Rs relative to the target life is 5.0% or less, preferably 4.0% or less, and more preferably 3.0% or less. Moreover, the smaller the variation of the sheet resistance Rs, the better, and the present invention can achieve up to around 2.2%.

(17) Here, the term target life refers to the life of a target from the start of use of that target to the time that the target can no longer be used as a result of the decrease in the target thickness caused by the advancement or erosion due to the sputtering phenomenon.

(18) The target life can be represented based on the integration of the power during sputtering and the total sputtering time and, for example, when the power is 15 kW, the target life of a target that can be used for 100 hours will be 1500 kWh.

(19) In the present invention, the variation in the sheet resistance Rs relative to the target life was obtained by sputtering one wafer for each 300 kWh and obtaining the average value of the sheet resistance Rs at 49 locations in the respective wafer planes, using the obtained average value to calculate the average value and the standard deviation among the wafers, and attaining variation (%) relative to target life=standard deviation among wafers/average value among wafers100.

(20) Note that the sheet resistance was measured at a total of 49 locations; namely, one location at the center of the wafer, 8 locations at the R (diameter) point (45 intervals), 16 locations at the R (diameter) point (22.5 intervals), and 24 locations at the R point (outer peripheral part) (15 intervals) were sampled.

(21) Moreover, with the thin film formed by sputtering the target of the present invention, the variation in the film thickness uniformity relative to the target life is 5.0% or less, preferably 4.0% or less, and more preferably 3.0% or less. Moreover, the smaller the variation of the film thickness uniformity, the better, and the present invention can achieve up to around 1.2%.

(22) In the present invention, the variation in the film thickness uniformity relative to the target life was obtained by sputtering one wafer for each 300 kWh and obtaining the average value of the film thickness at 49 locations in the respective wafer planes, using the obtained average value to calculate the average value and the standard deviation among the wafers, and attaining variation (%) in film thickness uniformity relative to target life=standard deviation among wafers/average value among wafers100.

(23) Note that the film thickness was measured at a total of 49 locations; namely, one location at the center of the wafer, 8 locations at the R (diameter) point (45 intervals), 16 locations at the R (diameter) point (22.5 intervals), and 24 locations at the R point (outer peripheral part) (15 intervals) were sampled.

EXAMPLES

(24) The present invention is now explained based on the following Examples and Comparative Examples. Note that the following Examples are merely illustrative, and the present invention is not in any way limited by these Examples. In other words, the present invention is limited only by its scope of patent claims, and covers various modifications other than the Examples included in the present invention.

Example 1

(25) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1200 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 900 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 900 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 900 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 4.1. Subsequently, the billet was subject to cold rolling at a rolling reduction of 88% and thereafter heat-treated at 800 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(26) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 68.1 m, and the variation thereof was 47.9%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, the variation in the sheet resistance Rs was 2.3% and the variation in the film thickness uniformity was 2.0% and small in both cases, and an extremely favorable thin film with superior uniformity was obtained.

Example 2

(27) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1100 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 900 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 900 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 900 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 4.1. Subsequently, the billet was subject to cold rolling at a rolling reduction of 88% and thereafter heat-treated at 800 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(28) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 69.7 m, and the variation thereof was 47.3%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, the variation in the sheet resistance Rs was 4.5% and the variation in the film thickness uniformity was 1.2% and small in both cases, and an extremely favorable thin film with superior uniformity was obtained.

Example 3

(29) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1200 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 900 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 900 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 900 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 4.0. Subsequently, the billet was subject to cold rolling at a rolling reduction of 87% and thereafter heat-treated at 850 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(30) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 71.0 m, and the variation thereof was 50.1%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, the variation in the sheet resistance Rs was 3.7% and the variation in the film thickness uniformity was 1.2% and small in both cases, and an extremely favorable thin film with superior uniformity was obtained.

Example 4

(31) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1000 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 900 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 900 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 900 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 4.2. Subsequently, the billet was subject to cold rolling at a rolling reduction of 88% and thereafter heat-treated at 800 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(32) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 67.9 m, and the variation thereof was 48.2%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, the variation in the sheet resistance Rs was 2.3% and the variation in the film thickness uniformity was 2.1% and small in both cases, and an extremely favorable thin film with superior uniformity was obtained.

Example 5

(33) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1300 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 900 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 900 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 900 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 4.2. Subsequently, the billet was subject to cold rolling at a rolling reduction of 88% and thereafter heat-treated at 800 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(34) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 70.3 m, and the variation thereof was 48.2%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, the variation in the sheet resistance Rs was 2.2% and the variation in the film thickness uniformity was 3.9% and small in both cases, and an extremely favorable thin film with superior uniformity was obtained.

Example 6

(35) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1350 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 850 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 850 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 850 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 4.4. Subsequently, the billet was subject to cold rolling at a rolling reduction of 89% and thereafter heat-treated at 750 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(36) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 34.4 m, and the variation thereof was 53.6%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, the variation in the sheet resistance Rs was 3.5% and the variation in the film thickness uniformity was 4.7% and small in both cases, and an extremely favorable thin film with superior uniformity was obtained.

Example 7

(37) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1200 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 1100 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 1100 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 1100 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 3.8. Subsequently, the billet was subject to cold rolling at a rolling reduction of 85% and thereafter heat-treated at 1000 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm. As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 154.3 m, and the variation thereof was 43.4%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, the variation in the sheet resistance Rs was 3.0% and the variation in the film thickness uniformity was 2.6% and small in both cases, and an extremely favorable thin film with superior uniformity was obtained.

Example 8

(38) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1200 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 1000 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 1000 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 1000 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 3.6. Subsequently, the billet was subject to cold rolling at a rolling reduction of 82% and thereafter heat-treated at 900 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(39) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 102.1 m, and the variation thereof was 46.3%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, the variation in the sheet resistance Rs was 3.1% and the variation in the film thickness uniformity was 3.8% and small in both cases, and an extremely favorable thin film with superior uniformity was obtained.

Comparative Example 1

(40) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1200 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 1100 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 1100 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 1100 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 2.2. Subsequently, the billet was subject to cold rolling at a rolling reduction of 82% and thereafter heat-treated at 1000 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(41) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 137.9 m, and the variation thereof was 36.7%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, while the variation in the sheet resistance Rs was small at 3.2%, the variation in the film thickness uniformity was great at 6.9%, and a thin film with inferior uniformity was obtained.

Comparative Example 2

(42) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1100 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 1100 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 1100 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 1100 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 2.4. Subsequently, the billet was subject to cold rolling at a rolling reduction of 82% and thereafter heat-treated at 1000 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(43) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 137.0 m, and the variation thereof was 36.1%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, the variation in the sheet resistance Rs was 6.3% and the variation in the film thickness uniformity was 5.3% and great in both cases, and a thin film with extremely inferior uniformity was obtained.

Comparative Example 3

(44) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 am. The obtained billet was thereafter heat-treated at 1350 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 1100 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 1100 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 1100 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 2.2. Subsequently, the billet was subject to cold rolling at a rolling reduction of 82% and thereafter heat-treated at 1000 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(45) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 135.9 m, and the variation thereof was 37.3%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, while the variation in the sheet resistance Rs was small at 4.6%, the variation in the film thickness uniformity was great at 6.3%, and a thin film with inferior uniformity was obtained.

Comparative Example 4

(46) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1300 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 900 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 900 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 900 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 5.2. Subsequently, the billet was subject to cold rolling at a rolling reduction of 91% and thereafter heat-treated at 800 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(47) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 50.0 m, and the variation thereof was 23.6%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, while the variation in the film thickness uniformity was small at 3.5%, the variation in the sheet resistance Rs was great at 6.5%, and a thin film with inferior uniformity was obtained.

Comparative Example 5

(48) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1100 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 900 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 900 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 900 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 5.3. Subsequently, the billet was subject to cold rolling at a rolling reduction of 91% and thereafter heat-treated at 800 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(49) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 67.1 m, and the variation thereof was 22.2%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, while the variation in the sheet resistance Rs was small at 3.7%, the variation in the film thickness uniformity was great at 6.3%, and a thin film with inferior uniformity was obtained.

Comparative Example 6

(50) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1300 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 1000 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 1000 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 1000 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 5.5. Subsequently, the billet was subject to cold rolling at a rolling reduction of 95% and thereafter heat-treated at 900 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(51) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 99.8 m, and the variation thereof was 35.8%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, while the variation in the sheet resistance Rs was small at 2.9%, the variation in the film thickness uniformity was great at 5.6%, and a thin film with inferior uniformity was obtained.

Comparative Example 7

(52) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1100 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 850 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 850 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 850 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 5.4. Subsequently, the billet was subject to cold rolling at a rolling reduction of 95% and thereafter heat-treated at 775 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm.

(53) As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 39.0 m, and the variation thereof was 66.2%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, while the variation in the sheet resistance Rs was small at 2.8%, the variation in the film thickness uniformity was great at 17.3%, and a thin film with inferior uniformity was obtained.

Comparative Example 8

(54) A tantalum raw material having a purity of 99.997% was subject to electron beam melting, and then cast to obtain a billet having a thickness of 54 mm and a diameter of 195 mm. The obtained billet was thereafter heat-treated at 1200 C. Subsequently, the billet was subject to extend forging at room temperature, and thereafter heat-treated at a temperature of 1200 C. Subsequently, the billet was subject to extend forging and upset forging at room temperature, and once again heat-treated at 1200 C. Subsequently, the billet was once again subject to extend forging and upset forging at room temperature, and heat-treated at a temperature of 1200 C. As a result of repeating the forging and heat treatment processes as described above, the integrated value of the true strain of the overall forging process was 2.0. Subsequently, the billet was subject to cold rolling at a rolling reduction of 80% and thereafter heat-treated at 1100 C., and then subject to finish processing to obtain a target having a thickness of 10 mm and a diameter of 450 mm. As a result of using an electron microscope and analyzing the target obtained from the foregoing processes, the average crystal grain size was 214.3 m, and the variation thereof was 67.2%. Moreover, this target was sputtered to form a tantalum thin film on a silicon wafer (12-inch), and the sheet resistance was measured. As a result, as shown in Table 1, the variation in the sheet resistance Rs was 6.6% and the variation in the film thickness uniformity was 16.2% and great in both cases, and a thin film with inferior uniformity was obtained.

(55) TABLE-US-00001 TABLE 1 Heat treatment Heat treatment Crystal grain temp. after Integrated value temp. after Rolling Heat treatment size of casting of true strain forging reduction temp. after rolling target ( C.) of forging process ( C.) (%) ( C.) Avg. (m) Min. (m) Example 1 1200 4.1 900 88 800 68.1 30.1 Example 2 1100 4.1 900 88 800 69.7 30.1 Example 3 1200 4.0 900 87 850 71.0 30.2 Example 4 1000 4.2 900 88 800 67.9 30.1 Example 5 1300 4.2 900 88 800 70.3 30.1 Example 6 1350 4.4 850 89 750 34.4 2.1 Example 7 1200 3.8 1100 85 1000 154.3 50.2 Example 8 1200 3.6 1000 82 900 102.1 40.6 Comparative 1200 2.2 1100 82 1000 137.9 68.2 Example 1 Comparative 1100 2.4 1100 82 1000 137.0 59.8 Example 2 Comparative 1350 2.2 1100 82 1000 135.9 62.3 Example 3 Comparative 1300 5.2 900 91 800 50.0 32.4 Example 4 Comparative 1100 5.3 900 91 800 67.1 40.1 Example 5 Comparative 1300 5.5 1000 95 900 99.8 30.3 Example 6 Comparative 1100 5.4 850 95 775 39.0 1.3 Example 7 Comparative 1200 2.0 1200 80 1100 214.3 72.6 Example 8 Crystal grain size of Film target Sheet resistance thickness Max. (m) Variation Rs variation variation Determination* Example 1 194.0 47.9% 2.3% 2.0% Example 2 204.1 47.3% 4.5% 1.2% Example 3 234.5 50.1% 3.7% 1.2% Example 4 199.7 48.2% 2.3% 2.1% Example 5 199.2 48.2% 2.2% 3.9% Example 6 69.0 53.6% 3.5% 4.7% Example 7 282.9 43.4% 3.0% 2.6% Example 8 185.4 46.3% 3.1% 3.8% Comparative 242.9 36.7% 3.2% 6.9% X Example 1 Comparative 233.7 36.1% 6.3% 5.3% X Example 2 Comparative 239.8 37.3% 4.6% 6.3% X Example 3 Comparative 69.1 23.6% 6.5% 3.5% X Example 4 Comparative 90.8 22.2% 3.7% 6.3% X Example 5 Comparative 159.5 35.8% 2.9% 5.6% X Example 6 Comparative 73.1 66.2% 2.8% 17.3% X Example 7 Comparative 393.2 67.2% 6.6% 16.2% X Example 8 Determination: shows that the sheet resistance Rs variation is 3% or less, and the film thickness variation is within 3%. shows that the sheet resistance Rs variation is greater than 3% and 5% or less, or the film thickness variation is greater than 3% and within 5%. X shows that the sheet resistance Rs variation and/or the film thickness variation exceeds 5%.

(56) As described above, in all cases of Examples 1 to 8, it was confirmed that the variation in the average crystal grain size of the tantalum target is 40% or higher and 60% or less. Thus, it has been confirmed that this kind of structure plays an extremely important role for uniformizing the film thickness and the specific resistance.

(57) Moreover, in Examples 1 to 8, while the series of processes of forging and heat treatment was repeated three to four times, favorable results were obtained when the true strain was 3.0 or more and 5.0 or less even when the foregoing series of processes was performed only once or repeated in a number other than three or four times.

(58) As a result of controlling the variation in the average crystal grain size of a tantalum sputtering target, the present invention can improve the uniformity of the film thickness and specific resistance of the thin film formed using the foregoing target. Accordingly, fine wiring can be stably produced by using the target of the present invention. The present invention is useful as a tantalum sputtering target that is used for the deposition of a barrier film for a wire layer in a semiconductor device.