JP2018178242A - MANUFACTURING METHOD OF Ti-Al-BASED ALLOY - Google Patents

Abstract

PROBLEM TO BE SOLVED: To effectively manufacture a high quality Ti-Al-based alloy from a low quality titanium material at good yield.SOLUTION: A manufacturing method of a Ti-Al-based alloy 1 has a first process that a dissolution raw material 2 by blending a CaO-CaF-based flux 3 by blending 35 to 95 mass% of calcium fluid with calcium oxide with a Ti-Al-based alloy consisting of a titanium material and an aluminum material and containing oxygen of total 0.1 mass% or more and Al of 40 mass% or more to have 3 to 20 mass% based on the Ti-Al-based alloy, is divided into n so that weight of the dissolution raw material 2 after division becomes 4/5 or less of whole based on final target ingot weight after casting end as maximum, a second process that operations for charging the first divided body 41 into a water-cooled copper crucible 5, dissolving the same under inert gas atmosphere of 1.33 Pa or more and pulling a bottom of water-cooled copper crucible 5 at a rate of 15 mm or less per minute are repeated by the n-th divided body to form an ingot, and a third process that a flux layer 8 adhered to a surface of the ingot formed in the second process is mechanically removed.SELECTED DRAWING: Figure 1

Description

The present invention is high quality, ie, low oxygen, by dissolving a low grade titanium raw material such as low grade sponge titanium, scrap titanium, titanium oxide (TiO ₂ ) such as rutile ore and adding an aluminum raw material and a flux. The present invention relates to a method for producing a Ti—Al based alloy for producing a Ti—Al based alloy.

In recent years, the demand for Ti—Al based alloys as materials for aircraft and automobiles has increased. Conventionally, since Ti-Al alloys are very active against oxygen, vacuum arc melting (VAR), electron beam melting (EB), plasma arc melting (PAM), which can reduce the influence of oxygen Melting and casting are performed using techniques such as vacuum induction melting (VIM) and water-cooled copper induction melting (CCIM).
Among the melting and casting methods described above, in the melting process based on VAR, EB, VIM, etc. performed under a vacuum atmosphere, not only the alloy element Al but also Ti is lost by volatilization, so in the industrial process It is difficult to control the composition after dissolution, and it is feared that the cost of production will increase.

In addition, high-grade raw materials with low oxygen content are generally used as titanium raw materials for melting Ti-Al alloys, but the price of high-grade titanium raw materials has risen sharply in recent years There is an increasing need to obtain a high grade, low oxygen Ti-Al alloy even when using a titanium raw material such as low grade ore or scrap having a high oxygen content.
Therefore, instead of using dissolution methods such as VAR, EB, VIM, etc., where loss of Ti is likely to occur due to volatilization, low-quality raw materials (of 0.1% by mass or more of oxygen content) are used using dissolution methods such as PAM or CCIM A technology has been proposed for producing a Ti—Al based alloy by deoxidation from titanium raw materials while suppressing the volatilization loss of Al and Ti.

For example, in Patent Document 1, when 40 mass% or more of Al is added to high oxygen content Ti in an atmosphere of 1.33 Pa or more using PAM or CCIM and dissolved and held, oxygen in the Ti—Al based alloy is Al In the form of Al ₂ O ₃ , it is discharged from Ti-Al in the form of Al ₂ O ₃ to proceed with deoxidation, and when the CaO-CaF ₂ system flux is added, the activity of Al ₂ O ₃ decreases, It is further stated that deacidification proceeds.
In the manufacturing method of Patent Document 1 mentioned above, the oxygen in the Ti—Al based alloy is certainly discharged in the form of Al ₂ O ₃ and deoxidized. However, simply dissolving and retaining using PAM or CCIM, Al ₂ O ₃ , which is a by-product of deoxidation, or a CaO-CaF ₂ system flux added for deoxidation promotion is a Ti-Al alloy In the Ti-Al alloy, there are sites where deoxidation proceeds and sites where deoxidation does not proceed (sites where Al ₂ O ₃ remains). Will be mixed.

Further, according to Patent Document 1, low-oxygen Ti, in other words, high-quality Ti such as pure Ti, is added to a Ti—Al-based alloy in which Al ₂ O ₃ or CaO—CaF ₂ based flux partially remains. It is also described that if Al is diluted, a Ti-Al based alloy having an Al content of less than 40% by mass and a low oxygen content can be produced.
However, there are sites where Al ₂ O ₃ and flux remain in the Ti-Al alloy, and pure Ti is added to this to melt the Ti-Al alloy with an Al content of less than 40% by mass. If it is attempted, Al ₂ O _{3 and the} like inside the flux will be decomposed / redissolved, and the oxygen concentration etc. will rise. Therefore, in the manufacturing method of Patent Document 1, it is not easy to obtain a high grade, ie, low oxygen, Ti—Al based alloy having an Al content of less than 40% by mass.

In addition, after mechanically removing the site where Al ₂ O _{3 and the} like remain by cutting etc., adding high grade, low oxygen Ti and dissolving it, low oxygen with an Al content of less than 40% by mass Although a Ti-Al alloy can be obtained, the portion where the nonreusable Al ₂ O _{3 and the} like remain remains to be discarded, and in addition, when mechanical removal of Al ₂ O _{3 and the} like, part of metal Ti is Also, since the yield of the Ti-Al based alloy becomes very bad because it is removed together, it leads to cost increase.

That is, in the deoxidizing method of Patent Document 1, for a Ti—Al alloy containing 40% by mass or more of Al, a flux of Al ₂ O ₃ or CaO—CaF ₂ is added to the Ti—Al alloy. Whether it remains or not, in other words, when Al ₂ O ₃ or CaO-CaF ₂ -based flux remains, it is important how to separate / remove the remaining flux and other substances from the Ti—Al based alloy.
In this respect, Patent Document 2 and Patent Document 3 disclose techniques useful for separating / removing a substance such as residual flux from a Ti—Al based alloy.

For example, Patent Document 2 uses a cold crucible floatation / dissolution apparatus (CCIM) and adds rare earth metal (cerium or misch metal in the example) as a deoxidizing agent to titanium and calcium fluoride (CaF ₂ ) as a flux. When it melts, a molten flux can be present between the molten metal and the water-cooled copper crucible, and by moving solid or liquid nonmetallic inclusions in this molten flux layer, melting improves the cleanliness of the molten metal Methods of purification of metals are described.

Generally, when melting metals containing non-metallic inclusions using a high-frequency induction furnace such as CCIM, non-metallic inclusions have lower electrical conductivity compared to metals such as Ti, and so molten metal It is known that it tends to gather on the outside (water-cooled copper crucible side) of Therefore, if the method of Patent Document 2 is used, there is a possibility that nonmetallic inclusions such as Al ₂ O ₃ can be localized and physically removed from the Ti—Al based alloy.

Further, Patent Document 3 discloses an induction melting furnace provided with a water-cooled copper crucible in which a crucible main body is relatively movable in a vertical direction with respect to a container contained in the furnace. This patent document 3 also describes that melting of a metal raw material and pulling down of a water-cooled copper crucible are repeated to produce a long active metal ingot. This Patent Document 3 is also similar to Patent Document 2 in that a high frequency induction furnace provided with a water-cooled copper crucible is used, and in principle there is a tendency for nonmetallic inclusions to gather on the outside of the metal ingot. Because of this, as in Patent Document 2, there is a possibility that nonmetallic inclusions such as Al ₂ O ₃ can be physically removed from the Ti—Al alloy.

JP 2016-135907 JP-A-11-246919 Unexamined-Japanese-Patent No. 2006-122920

In the method of purifying molten metal of Patent Document 2, a metal containing non-metallic inclusions is simply dissolved using a cold crucible floatation / dissolution apparatus (CCIM), and non-metals such as Al ₂ O _{3 and the} like are used. The conditions are not suitable for the inclusions to move into the flux. Therefore, even if migration of non-metallic inclusions to the flux occurs, the possibility of quantitatively sufficient migration is low and the possibility of non-metallic inclusions remaining in the molten metal is high (described later) In Comparative Examples 1 and 2 according to the present invention, the result in which non-metallic inclusions remain is actually obtained).

In addition, even if migration of non-metallic inclusions such as Al ₂ O ₃ is sufficiently performed, there is no description in Patent Document 2 on how to treat the flux remaining in the molten metal. . That is, if the flux remaining in the molten metal can not be removed from the molten metal by any method, a Ti-Al-based alloy having a high degree of cleanliness can not be obtained using the technique of Patent Document 2.
In addition, the method of producing an active metal ingot of Patent Document 3 is also capable of collecting non-metallic inclusions on the outside of a metal ingot, in principle similar to Patent Document 2, but non-metallic inclusions In fact, non-metallic inclusions such as Al ₂ O ₃ in the Ti—Al alloy may not be able to be removed because the condition is not suitable for transferring into the flux (in Comparative Example 3 described later) The result of remaining nonmetallic inclusions is actually obtained).

  Furthermore, in patent document 3, the melt | dissolution which added the flux is not assumed, and there is also no description regarding deoxidation or inclusion removal. Therefore, when the reduction is performed at the reduction rate specified in Patent Document 3, it is also unclear whether deoxidation or inclusion removal is performed without any problem. For example, in patent document 3, although the speed at the time of pull-down is 30 mm / min or less, in the case of such a fast pull-down speed, the flux layer breaks and the molten metal leaks, etc. It may not happen.

  The present invention has been made in view of the above problems, and a high quality and low oxygen Ti-Al alloy is efficiently produced from a low grade titanium material containing high concentration of oxygen at a high yield. An object of the present invention is to provide a method of producing a Ti—Al based alloy that can be produced.

In order to solve the above problems, the method for producing a Ti—Al based alloy of the present invention adopts the following technical means.
That is, according to the method for producing a Ti—Al based alloy of the present invention, an alloy material of a Ti—Al based alloy containing 0.1 mass% or more of total oxygen and 40 mass% or more of Al consisting of a titanium material and an aluminum material, The dissolved raw material is a molten raw material in which a CaO-CaF ₂ system flux containing 35 to 95% by mass of calcium fluoride mixed with calcium oxide is blended to be 3 to 20% by mass with respect to a Ti-Al based alloy The first step of n-splitting so that the weight of the melted raw material after splitting is at most 4/5 or less of the final target ingot weight at the end of casting, and the melting split in the first step The first divided body of the raw material is charged into a water-cooled copper crucible and melted under an inert gas atmosphere of 1.33 Pa or more, and the bottom of the water-cooled copper crucible is pulled downward at a speed of 15 mm or less per minute; Thereafter, the second divided body of the melting raw material divided in the first step is treated as the water-cooled copper. Load in a crucible, melt under an inert gas atmosphere of 1.33 Pa or more, and pull down the water-cooled copper crucible at a speed of 15 mm / min or less, and repeat the above operation until the nth divided body thereafter And a third step of mechanically removing a flux layer attached to the surface of the ingot formed in the second step.

  Preferably, a titanium material is added to the ingot from which the flux layer has been removed in the third step, and the ingot is dissolved by a melting method using a water-cooled copper container in an atmosphere of 1.33 Pa or more. Preferably, a Ti—Al based alloy having an Al content of less than 40% by mass is obtained.

  According to the method of manufacturing a Ti-Al alloy of the present invention, a high quality and low oxygen Ti-Al alloy is efficiently manufactured with a high yield from a low grade titanium material containing high concentration of oxygen. can do.

It is the figure which divided the manufacturing method of the Ti-Al system alloy concerning the present invention into each process, and was shown typically.

Hereinafter, an embodiment of a method of manufacturing a Ti—Al based alloy according to the present invention will be described in detail based on the drawings.
As shown in FIG. 1, the method of manufacturing the Ti—Al-based alloy 1 according to the present embodiment passes through three steps of the first to third steps, and preferably further after the first to third steps. The high-quality Ti-Al alloy 1 having an oxygen content of less than 0.1 mass% is manufactured from the alloy material of the Ti-Al alloy 1 containing 0.1 mass% or more of oxygen by performing four steps. ing.

Specifically, the alloy material used in the method of manufacturing the Ti—Al alloy 1 is a mixture of a titanium material and an aluminum material. Thus, when the molten raw material 2 in which aluminum is blended is melted in the second step, the aluminum reacts with oxygen in the alloy material to perform deoxidation. Further, in the manufacturing method of the present invention, the flux 3 of the CaO-CaF ₂ system is added in the first step in order to accelerate deoxidation in the alloy material. When such a flux 3 is added in the first step, deoxidation in the second step is further promoted, and from the alloy material containing 0.1 mass% or more of oxygen, finally, high grade with less than 0.1 mass% of oxygen Can be obtained.

Further, in the first step, the dissolved raw material 2 to which the flux 3 is added is divided into a plurality of pieces to form divided bodies. Then, in the second step, a pulling operation is performed on all the divided bodies following the dissolving operation of the divided bodies of the dissolved raw material 2, and nonfluxes of nonmetallic inclusions such as Al ₂ O ₃ are transferred to the flux 3; The flux 3 to which nonmetallic inclusions have been transferred can be localized (localized) on the outer peripheral surface of the ingot. Thus, the Ti—Al alloy 1 of the present invention can be obtained by mechanically removing the localized nonmetallic inclusions and the flux 3 in the third step, or preferably by further adjusting the components in the fourth step. Manufactured.

Hereinafter, each process of the 1st process-the 4th process provided in the manufacturing method of this invention is each demonstrated.
In the first step, the flux 3 is added to the alloy material of the Ti—Al alloy 1 to produce the molten raw material 2, and the produced molten raw material 2 is divided into n pieces. The n-th divided body is formed. This dividing operation of the melting material 2 is a feature of the production method of the present invention.

Specifically, the alloy material of the Ti—Al alloy 1 used for the molten raw material 2 is made of a titanium material and an aluminum material, and contains 0.1 mass% or more of oxygen in total and 40 mass% or more of Al in total. doing. Further, the flux 3 blended in the alloy material is a CaO-CaF ₂ system flux 3 in which 35 to 95 mass% of calcium fluoride is blended with calcium oxide. And what mix | blended the flux 3 so that it may become 3-20 mass% with the alloy material mentioned above becomes the melt | dissolution raw material 2 of this embodiment.

The titanium material constituting the above-mentioned alloy material includes sponge titanium having a low grade and a large amount of oxygen, a scrap raw material, titanium oxide (TiO ₂ ) such as rutile ore, and the like. The reason for using a low grade titanium material as the alloy material in this way is that these titanium materials are inexpensive and easy to procure.
Moreover, the alloy material mentioned above makes the total content of oxygen 0.1 mass% or more. For example, when the total content of oxygen in the alloy material is less than 0.1% by mass, the content of oxygen is small and deoxidation itself is not necessary. In the present invention, although the upper limit of the content of oxygen is not defined, the upper limit of the total content of oxygen actually contained in the alloy material is considered to be about 25 mass%.

The reason for using the Ti—Al-based alloy 1 containing 40% by mass or more of Al as the alloy material used for the molten raw material 2 in the first step is as follows.
For example, according to the known ternary Ti—Al—O ternary phase diagram (see FIG. 5 etc. of WO 2016/035824), the maximum amount of oxygen dissolved in the Ti—Al alloy 1 is Ti— As the Al content in the Al-based alloy 1 is increased, the solid solution oxygen concentration tends to decrease. That is, even in the case of the dissolved raw material 2 containing the Ti—Al alloy 1 manufactured using a low grade titanium material, if the content of Al is increased to 40 mass% or more, deoxidation was performed in the second step The present inventors came to complete the present invention, thinking that oxygen in the alloy material could be lowered at that time.

The flux 3 described above has a function of reducing the activity of Al ₂ O ₃ in the dissolved raw material 2 by being added to the dissolved raw material 2 and promoting the deoxidation reaction in the second step described later. That is, by dissolving Al ₂ O ₃ which is a deoxidation product of the Ti—Al-based alloy 1, the flux 3 reduces the activity of Al ₂ O ₃ which is a generation species in the deoxidation reaction, thereby removing the flux. It has the effect of promoting the acid reaction.

The dissolution of Al ₂ O ₃ in the flux 3 occurs only when the flux 3 is melted. Therefore, if the melting point of the flux 3 becomes too high, the flux 3 will not melt and dissolution of Al ₂ O ₃ will not occur. That is, in the case of the flux 3 of the CaO-CaF ₂ system, it is necessary to lower the melting point of the flux 3 itself by, for example, increasing the content of CaF ₂ having a low melting point. Specifically, in the deoxidation of the present embodiment, the content of CaF ₂ in the flux 3 is set to 35% by mass or more so that the melting point of the flux 3 is 1800 K or less. Moreover, so as not to Ti-Al-based alloy 1 obtained as a product after the third step or the fourth step is contaminated with fluorine in CaF _2, the content of CaF ₂ in the flux 3 is less than 95 wt% It is assumed.

In addition, the addition amount of the CaO-CaF ₂ -based flux 3 to the molten raw material 2 described above is 3% by mass to 20% by mass with respect to the Ti-Al-based alloy 1 as a product. If the amount added to the Ti—Al alloy 1 is less than 3% by mass, the Al ₂ O ₃ activity does not decrease appreciably, and the deoxidation promoting effect is hardly obtained. If the amount added to the Ti—Al alloy 1 is more than 20% by mass, the risk of the added flux 3 remaining in the manufactured Ti—Al alloy 1 increases.

Thus, in the first step, the flux 3 is mixed with the alloy material to prepare the molten material 2 first.
By the way, the molten raw material 2 in which the flux 3 is blended with the alloy material of the Ti-Al based alloy 1 in the first step described above is charged into the water-cooled copper crucible 5 and melted (performing the solidification method in the crucible), Nonmetallic inclusions such as Al ₂ O ₃ can not be removed sufficiently. In order to remove nonmetallic inclusions sufficiently, it is necessary to carry out the operation of the second step as shown below.

Specifically, when the dissolution raw material 2 described above is prepared, the prepared dissolution raw material 2 is divided into n pieces to form the first divided body 41 to the n-th divided body. The division number "n" of the melting material 2 is an integer of 2 or more, and in the first step, the melting material 2 is divided into a plurality of divided bodies.
Also, when the molten material 2 is divided in the first step, the weight of each divided body (the first divided body 41 to the n-th divided body) is at most the entire target ingot weight at the end of casting. The dissolved raw material 2 after adjustment is divided into n so as to be 4/5 or less. For example, as shown in FIG. 1, when the final target ingot weight is 100 tons, in the case of dividing the molten material 2 into three, all of the first divided body 41 to the third divided body 43 become 80 tons or less Need to be split. Industrially, although it is preferable to divide the split body equally (in this case, 33.3 ton), the present invention is not limited to dividing the split body equally.

If such division is performed, none of the first divided body 41 to the n-th divided body obtained in the first step will exceed 4/5 of the final target ingot weight at the end of casting, and water cooling at the time of casting The heat removal from the bottom 7 of the copper crucible 5 can be suppressed, and the deoxidation reaction can be sufficiently promoted.
More specifically, as described above, the raw material weight of each divided body after dividing the molten raw material 2 is divided so as to be at most 4/5 or less of the total with respect to the final ingot weight. In other words, at least two dissolution operations and two pull-down operations (division / insertion → reduction → division / insertion → dissolution) are required. This is because the portion of the bottom 7 close to the water-cooled copper crucible 5 is strongly cooled by only one melting operation, and solidification proceeds before the input flux 3 is sufficiently melted, so Al ₂ O ₃ And other nonmetallic inclusions and the flux 3 hardly react. In this respect, since the heat removal from the bottom portion 7 is suppressed in the second and subsequent melting operations, the flux 3 and the nonmetallic inclusion sufficiently react with each other in the second and subsequent melting to promote deoxidation. Do.

  If the weight of each divided body is more than 4/5 of the final target weight, the separation efficiency between metal and flux 3 may be reduced even if the second and subsequent melting operations are performed. There is sex. For example, when the weight of the first divided body 41 exceeds 4/5 of the final target weight, the non-metallic inclusions and the flux may be dissolved even if less than the remaining 1/5 is dissolved in the second and subsequent dissolutions. The reaction with 3 is less likely to occur. Therefore, it is desirable that the weight of the divided body is at most 4/5 or less of the whole, preferably 2/3 or less of the whole, more preferably 1/2 or less of the whole.

In the present embodiment shown in FIG. 1, the present invention is described using an example in which the dissolved raw material 2 is divided into three and deoxidized in the first step, but the number of divisions is five. There is no problem at all even with 11 divisions.
In the second step, the first divided body 41 to the n-th divided body of the molten raw material 2 divided in the first step are charged into the water-cooled copper crucible 5 (bottom-lined water-cooled copper crucible 5) to deactivate 1.33 Pa or more An operation of melting under a gas atmosphere and pulling down the water cooled copper crucible 5 at a speed of 15 mm or less per minute is performed, and then the second divided body 42 of the melted raw material 2 divided in the first step is Perform the operation of charging to 5 and melting under an inert gas atmosphere of 1.33 Pa or more and pulling down the water-cooled copper crucible 5 at a speed of 15 mm or less per minute, thereafter repeating the same operation up to the n-th divided body It is what forms an ingot.

  In other words, the second step is a melting operation in which the divided body of the molten raw material 2 is charged into the water-cooled copper crucible 5 and melted under an inert gas atmosphere of 1.33 Pa or more, and the water-cooled copper crucible 5 following this melting operation. The operation of performing the pull-down operation of pulling down downward at a speed of 15 mm or less per minute, as one basic operation, is one operation. Then, this basic operation is performed once for each of the first divided body 41, the second divided body 42,..., And the n-th divided body.

In the case of the present embodiment, the dissolving material 2 is divided into three, so the dissolving operation of the first divided body 41 → the pulling-down operation of the first divided body 41 → the dissolving operation of the second divided body → the second The second process proceeds by treating the molten material 2 in the order of the pulling-down operation of the divided body 42, the dissolving operation of the third divided body 43, and the pulling-down operation of the third divided body 43.
Next, the dissolution operation and the pull-down operation performed on each divided body of the dissolution raw material 2 in the second step will be described in detail.

In the second step, each divided body of the molten raw material 2 is charged into a water-cooled copper crucible 5 (water-cooled copper mold) in which the inside of the container is prepared in an inert gas atmosphere of 1.33 Pa or more. The ingot is cast by melting each divided body of the melting material 2 using plasma arc or induction heating as a heat source. In addition, when melt | dissolving the division body of the melt | dissolution raw material 2 using the water-cooled copper crucible 5, it is preferable to melt | dissolve preferably using induction heating as a heat source. Generally, when a metal containing nonmetallic inclusions such as oxides is dissolved by induction heating, the nonmetallic inclusions gather on the outside of the molten metal due to the difference in the electrical conductivity between the nonmetallic inclusion and the metal. It is known. That is, the flux 3 in which Al ₂ O ₃ in the molten metal supplied to the inside of the water-cooled copper crucible 5 is melted is concentrated in the vicinity of the inner peripheral surface of the mold by induction heating, and solidification is performed in a concentrated state. As a result, in the ingot melted and produced in the second step, the surface-adhered flux layer 8 in which the flux 3 is biased and exists on the outer peripheral side of the ingot to be drawn downward is formed. Thus, in the case of an ingot in which the surface adhering flux layer 8 exists on the outer peripheral side, the surface adhering flux layer 8 can be scraped off by mechanical means such as shot blasting or grinding in the third step. _This is because nonmetallic inclusions such as ₂ O ₃ can be removed.

  In the second step, the pulling-down operation performed subsequent to the melting operation of the divided body of the melting material 2 is a surface-adhering flux layer 8 formed on the surface of the ingot by dissolving the flux 3 by the above-described melting operation. , Uneven distribution between the mold and the ingot. Specifically, the water-cooled copper crucible 5 described above has a combined structure of a cylindrical crucible main body 6 opened toward both the upper side and the lower side, and a bottom portion 7 disposed on the bottom side of the crucible main body 6 It has become. The bottom 7 of the water-cooled copper crucible 5 is mounted movably in the vertical direction with respect to the crucible main body 6, and if the bottom 7 is lowered with respect to the crucible main body 6, the casting placed on the upper side of the bottom 7 It also becomes possible to lower the mass. In the second step, the surface adhering flux layer 8 is unevenly distributed on the outer peripheral side of the ingot, in other words, between the mold and the ingot by performing such a pulling-down operation.

  Further, in the pulling-down operation of the present embodiment, the pulling-down speed of the ingot, in other words, the falling speed of the bottom portion 7 of the water-cooled copper crucible 5 is preferably 15 mm / min or less, more preferably 10 mm / min or less. When the draw down speed of the ingot exceeds 10 mm / min, preferably 15 mm / min, the surface adhering flux layer 8 formed on the outer peripheral surface of the ingot in the second step described above is broken, and the surface adhering flux layer 8 It is difficult to make the local distribution between the mold and the ingot sufficiently.

Performing the above-described dissolution operation and the pull-down operation one by one successively is the “basic operation” for a certain divided body. This "basic operation" is performed once for each of the first divided body 41, the second divided body 42, ..., and the n-th divided body.
For example, in the case of the second step of the illustrated example, the dissolution raw material 2 is divided into three, and since the first divided body 41, the second divided body 42, and the third divided body 43 exist, the first divided body Dissolution operation of 41 → pull-down operation of the first divided body 41 → dissolution operation of the second divided body → pull-down operation of the second divided body → dissolution operation of the third divided body 43 → pull-down operation of the third divided body 43 Finally, a cylindrical ingot long in the vertical direction is melted (casted) in the second step.

On the outer peripheral surface of the ingot cast in the second step described above, there is formed a surface-adhered flux layer 8 solidified in a state where the flux 3 is biased. This surface-adhered flux layer 8 is made of Al ₂ O ₃ or the like. Nonmetallic inclusions are also contained in high concentration. Therefore, if the surface adhering flux layer 8 formed on the outer peripheral surface of the ingot is scraped off by mechanical means such as shot blasting or grinding in the third step, the flux 3 and nonmetallic inclusions such as Al ₂ O ₃ etc. It becomes possible to remove, and the oxygen concentration contained in the ingot can be lowered as a whole.

  In the Ti—Al alloy 1 obtained in the first to third steps described above, the surface-adhering flux layer 8 formed on the outer peripheral surface of the ingot in the second step is shot blasted or ground in the third step. Etc., so that the oxygen content contained in the Ti-Al alloy 1 is greatly reduced, and the oxygen originally contained in the alloy material is surely deoxidized and reduced. It has become. That is, according to the method of manufacturing the Ti-Al alloy 1 of the present embodiment, the high quality, that is, the low oxygen Ti-Al alloy 1 can be efficiently obtained with a high yield from low grade titanium containing high concentration of oxygen. Can be manufactured.

  In addition, the manufacturing method of Ti-Al type alloy 1 of this embodiment makes the total of the oxygen content contained in Ti-Al type alloy 1 by making Al content in an alloy material 40 mass% or more. Since the content is less than 0.1% by mass, the Ti—Al-based alloy 1 to be produced necessarily has an Al content of 40% by mass or more. However, when using the obtained Ti—Al based alloy 1, there is also a demand to reduce the Al content to less than 40% by mass.

In such a case, in addition to the first to third steps described above, the fourth step shown below may be performed.
That is, in the fourth step, an Al content is less than 40 mass% by adding a titanium material to the ingot and dissolving it by a melting method using a water-cooled copper mold (water-cooled copper container 9) in an atmosphere of 1.33 Pa or more. The Ti-Al alloy 1 is obtained. Although the dissolution method illustrated in FIG. 1 uses a water-cooled copper container, the dissolution method used in the fourth step is a dissolution method other than water-cooled copper induction melting (CCIM), for example, vacuum arc melting method ( (VAR), vacuum induction melting (VIM), or the like.

  Specifically, the titanium material to be added to the ingot in the fourth step has an Al content of 40 if the Ti content is less than 40% by mass after the fourth step. Preferably, the titanium material is less than mass%. For example, if a titanium material having an Al content of less than 40% by mass, such as pure Ti containing no aluminum as an impurity, is added, the Al content in the ingot is reduced by dilution, so the Al content is 40 mass. It is possible to obtain a Ti—Al-based alloy 1 of less than 10%.

The titanium material to be added in the fourth step changes depending on the required quality of the Ti-Al alloy 1 to be manufactured, so components other than aluminum in the titanium material (such as Sn, V and Mn other than aluminum) The concentration of metal) can not be defined and can be arbitrarily changed.
Further, in the manufacturing method of the present embodiment, the ratio (H / D) of the final ingot height H (final target ingot height) to the ingot diameter D (H / D) is not particularly limited. It is preferable to set it as one or more from a viewpoint.

  By performing the fourth step in addition to the first to third steps described above, it is possible to obtain the Ti—Al-based alloy 1 that meets the required qualities with respect to compositions other than oxygen and aluminum. It is possible to further enhance the convenience.

Next, the effects of the method for producing the Ti—Al-based alloy 1 of the present invention will be described in detail using Comparative Examples and Examples.
In Examples and Comparative Examples, a CaO-CaF ₂ flux 3 is added to an alloy material prepared by mixing an aluminum material with a titanium material to deoxidize O (oxygen) contained in the alloy material. It is

  In Example 1, the processing of the first to fourth steps is performed on an alloy material containing 40% by mass of Al and 0.8% by mass of O. Moreover, Example 2 processes the 1st process-the 4th process about the alloy material which contains Al 60 mass% and O 0.8 mass%. In Example 3, the processing of the first to fourth steps is performed on an alloy material containing 45% by mass of Al and 0.8% by mass of O. In Example 4, the processes of the first to fourth steps are performed on an alloy material containing 52% by mass of Al and 0.8% by mass of O.

In addition, the division | segmentation number of the melt | dissolution raw material 2 of Examples 1-4 is "11" with respect to each of the 1st division body 41-the 11th division body 51, one dissolution operation and one pull-down operation. "Basic operation" which is composed of
Moreover, although the comparative example 1 uses the alloy material of the same composition as Example 1, in the 1st process, division of the melting raw material 2 is not performed, only the melting operation is performed in the 2nd process, and casting is performed. It has become a lump. In Comparative Example 1, a fourth step is further performed on the obtained ingot to obtain a Ti-30Al ingot.

Furthermore, Comparative Example 2 is obtained by performing the same treatment as Comparative Example 1 to obtain an ingot, and the fourth step is performed using only a specific portion of the obtained ingot.
Further, Comparative Example 3 is the same as Example 1, in which the alloy material containing 40% by mass of Al and 0.8% by mass of O is subjected to the processes of the first to fourth steps. Comparative Example 3 differs from Example 1 in that flux 3 is not blended in the dissolved raw material 2.

Specifically, the first to fourth steps described above were performed under the following conditions.
First, in the first step, scrap Ti, titanium oxide (TiO ₂ ), and pure Al are used as alloy materials, and in Comparative Example 1, Comparative Example 2 and Example 1, a CaO—CaF ₂ flux 3 is used as a Ti—Al based alloy. It melt | blended 10% with respect to the weight of the alloy 1, and about the comparative example 3, the melt | dissolution raw material 2 was produced, without mix | blending the flux 3. FIG. Flux 3 used in these comparative examples and examples is flux 3 (CaO-CaF ₂ system flux 3) containing CaO and CaF ₂ so that the weight ratio of CaO: CaF ₂ = 2: 8.

In the second step, the molten material 2 prepared in the first step is placed in a water-free copper crucible 5 (inner diameter 80 mm) having no bottom (no bottom), and an Ar atmosphere using argon as an inert gas Under the conditions, the pressure was set to 6.6 × 10 ⁴ Pa, and dissolution was performed using an induction dissolution apparatus. In the second step of Example 1 and Comparative Example 3, first to fourth divided bodies 41 to 51 are formed in advance by dividing the melting material 2 into 11 in the first step, and the dissolution operation and the pull-down operation Was carried out once for each divided body, and the dissolved raw material 2 was dissolved to proceed with deacidification. Moreover, about the comparative example 1 and the comparative example 2, as above-mentioned, 2nd process itself is not implemented.

With respect to the ingot thus cast in the second step, the inside of the ingot after melting was observed by SEM.
In the third step, the ingot cast in the second step is shot blasted as a mechanical means on the surface of the ingot to remove the surface-adhering flux layer 8 attached to the surface of the ingot. About the ingot from which the surface adhesion flux layer 8 was removed at the 3rd process, the oxygen concentration contained in the ingot was analyzed by the inert gas melting method.

Furthermore, in the fourth step, with respect to the ingot from which the surface adhering flux layer 8 has been removed in the third step, pure Ti having an oxygen concentration of 0.05% by mass is added using a plasma arc melting furnace, and Ti-30% by mass Al alloy was melted. The oxygen concentration of the ingot of the Ti-30 mass% Al alloy melted in the fourth step was also analyzed by the inert gas melting method in the same manner as the ingot after the third step.
The analysis results of Examples and Comparative Examples are shown in Table 1.

"Comparative example 1"
In Comparative Example 1, as described above, division of the dissolved raw material 2 is not performed in the first step, and in the second step, only the melting operation not accompanied by the pulling-down operation is performed on the undissolved divided raw material 2 The
When the ingot obtained in Comparative Example 1 was taken out after the second step and visually observed, a surface-adhering flux layer 8 was formed on the surface of the ingot. Moreover, in the comparative example 1, the flux 3 wound in the inside of alloy structure was also visually confirmed.

Moreover, about the ingot of "comparative example 1", SEM observation with respect to the inside of an ingot is also performed. As a result of observation by SEM, it was confirmed that a portion where Al ₂ O ₃ and flux 3 are not present and a portion where flux 3 is present are mixed in the ingot.
Furthermore, as a result of analyzing the oxygen concentration of the obtained ingot by the inert gas melting method, it is found that 0.50 mass% of oxygen is present in the absence of inclusions and oxygen is present in the presence of inclusions. 1.16% by mass was detected.

Furthermore, the ingot described above is melted in a fourth step using a plasma arc melting furnace, and pure Ti (oxygen concentration 0.05 mass%) is added to the above ingot to melt a Ti-30 mass% Al alloy. The oxygen concentration of the obtained ingot of Ti-30% by mass Al alloy was analyzed by an inert gas melting method. As a result of analysis, 0.79% by mass of oxygen was detected. Furthermore, when the inside of the ingot of Ti-30 mass% Al alloy was observed by SEM, the presence of oxide inclusions such as Al ₂ O ₃ was not confirmed.

From the results of the visual inspection and SEM observation described above, in Comparative Example 1, the flux 3 and the oxide inclusion (nonmetallic inclusion) remain in the obtained ingot, and the flux 3 and the oxide inclusion are It is judged that it remains inside the ingot, and it can be seen that oxygen in the Ti—Al alloy 1 can not be sufficiently reduced.
In addition, when the oxygen concentration in the Ti-Al alloy 1 of the comparative example is measured, oxygen significantly exceeding 0.1 mass% is detected regardless of the presence or absence of inclusions. It was found that oxygen could not be reduced sufficiently.

Furthermore, when the ingot once melted was redissolved in the fourth step, the oxygen concentration increased from 0.50 mass% to 0.79 mass% in the portion where the inclusions were not present. This is because when the pure Ti 10 is added to the ingot in the fourth step to adjust the composition, the oxide inclusions left in the ingot are decomposed by the pure Ti and redissolved in Ti-Al. It is believed that the oxygen concentration has risen.
"Comparative example 2"
The comparative example 2 melts the ingot by the melt | dissolution raw material 2 similar to the comparative example 1, and the melt | dissolution method. Therefore, the oxygen concentration of the site | part in which the inclusion of the melted ingot does not exist is 0.51 mass%, and the site | part in which the inclusion exists is 1.12 mass%. Among the ingots of this Comparative Example 2, a site where no inclusions are present is selected, melted using a plasma arc melting furnace, and pure Ti 10 (oxygen concentration 0.05 mass%) is added to Ti-30 mass%. An ingot of Al alloy was melted. As a result of analyzing the oxygen concentration of the melted ingot, the oxygen concentration was 0.42 mass%, which is a result of being lower than that of Comparative Example 1.

From this, unlike in Comparative Example 1, when the fourth step is performed using a site where inclusions such as Al ₂ O ₃ are hardly confirmed, oxidation of Al ₂ O ₃ etc. contained in the ingot is performed. It can be seen that the increase in oxygen concentration due to the decomposition of substance inclusions is suppressed. However, since the site where the inclusions are present can not be used as a raw material for re-dissolution, the evaluation of "yield" in Table 1 is an evaluation result of x while Comparative Example 1 is ○. From this, in Comparative Example 2, although the concentration of oxygen contained in the ingot can be reduced, the result is that the yield is bad, and the overall evaluation is x.
"Comparative example 3"
In the same manner as Comparative Example 1 and Comparative Example 2, a molten raw material 2 was prepared so as to be Ti-40 mass% Al-0.8 mass% O, and an ingot (2800 g) was melted. In addition, unlike Comparative Example 1 and Comparative Example 2, the molten raw material 2 of Comparative Example 3 is a molten raw material 2 prepared without blending the flux 3 in the alloy material in the first step. It is divided into 11 pieces. Specifically, the first divided body 41 initially charged in the water-cooled copper crucible 5 has a raw material weight of 800 g (excluding the flux 3), and the second divided body 42 to the eleventh divided additionally charged thereafter The body weighs 200g for all ingredients. The raw material (800 g) of the first divided body 41 is placed on a pure Ti starting block (the bottom 7 of the water-cooled copper crucible 5), and under Ar atmosphere, the pressure is 6.6 × 10 ⁴ Pa in the crucible Dissolved in After the first divided body 41 was melted, the crucible bottom 7 was pulled down at a speed of 2 mm / min for 5 minutes (10 mm) to carry out a pulling operation. Thereafter, the additional raw material (200 g) of the second and subsequent divided bodies previously charged in the additional charge raw material feeder is charged into the water-cooled copper crucible 5 and melted, and after melting, the crucible bottom 7 is pulled down. The ingot was melted by performing on all of the second to eleventh divided bodies 42 to 51.

As a result of observing the inside of the ingot of Comparative Example 3 after melting by SEM, as in Comparative Example 1 and Comparative Example 2, the non-metallic inclusion Al ₂ O ₃ hardly exists and It was found that there was an internal tissue in the In addition, as a result of analyzing the oxygen concentration of the ingot by the inert gas melting method, the oxygen concentration of the portion where the nonmetal inclusion hardly exists is 0.75 mass%, and the portion where the nonmetal inclusion is present The oxygen concentration was 0.94% by mass. From this fact, when dissolution and deoxidation are carried out without using flux 3, Al ₂ O ₃ which is a nonmetallic inclusion in Ti-Al hardly transfers to the ingot surface, and is contained in the ingot. Since it has stayed, it can be seen that the oxygen concentration can hardly be reduced even if the dissolved raw material 2 is divided in the first step.

Moreover, pure Ti10 (oxygen concentration 0.05 mass%) is added to the ingot of the comparative example 3 mentioned above using a plasma arc melting furnace, Ti-30 mass% Al alloy is melted, and Ti-30 mass is produced. A cast ingot of% Al alloy was cast, and the oxygen concentration of the ingot of Ti-30 mass% Al alloy was analyzed, and the result of 0.60 mass% was obtained. From this, as in Comparative Example 1, even if oxide Ti inclusions are added to pure ingots to the ingot remaining to the inside of the ingot to adjust the components, they are left in the ingots. Since the oxide inclusions are decomposed by pure Ti and the oxygen concentration in Ti—Al increases, it is judged that the oxygen in the Ti—Al alloy 1 can not be sufficiently reduced.
"Example 1"
Example 1 is the same as Comparative Example 3, in which the melting material 2 is divided into 11 pieces and the ingot is melted. Example 1 differs from Comparative Example 3 in that flux 3 is blended at 10% by mass with respect to 1% by weight of the Ti—Al alloy.

When the surface of the ingot taken out after completion of the melting operation was visually confirmed, a layer (surface-adhering flux layer 8) which was considered to be formed by dissolution / solidification of the added flux 3 was adhering to the surface of the ingot. When the inside of a similar ingot was observed by SEM, it was found that almost no non-metallic inclusions (oxide inclusions) such as Al ₂ O ₃ were present in the inside of the ingot. It is presumed that this is because non-metallic inclusions such as Al ₂ O ₃ have migrated to the flux layer formed on the ingot surface.

Moreover, the flux layer (surface adhesion flux layer 8) on the surface of the ingot could be easily removed by shot blasting. As a result of analyzing the oxygen concentration of the ingot from which the surface adhesion flux layer 8 has been removed by shot blasting by the inert gas melting method, the oxygen concentration in the portion where Al ₂ O ₃ is not present is 0.30 mass%, slightly The oxygen concentration was as very low as 0.40% by mass even at sites where oxide inclusions such as Al ₂ O ₃ were observed. From this, the flux 3 is blended in the first step and the melting material 2 is divided, and in Example 1 in which the melting operation and the pulling-down operation are repeatedly performed for each divided body in the second step, the oxygen concentration in the ingot Can be greatly reduced.

Further, when the ingot of Example 1 described above is melted using a plasma arc melting furnace and pure Ti 10 (oxygen concentration 0.05 mass%) is added to melt a Ti-30 mass% Al alloy ingot, The oxygen concentration of the Ti-30 mass% Al alloy ingot decreased to 0.25 mass%.
From this, in the fourth step of Example 1, unlike in Comparative Examples 1 to 3, almost no oxide inclusions were present in the ingot, so the oxygen concentration increased even if it was remelted. It turned out that there was no decline in yield either. Therefore, using the production method of the present invention, it is possible to efficiently produce low-oxygen Ti-Al alloy 1 having an Al content of less than 40% by mass using low-grade raw materials with high yield. It is determined that
"Example 2"
Example 2 is the same as Comparative Example 3, in which the melting material 2 is divided into 11 pieces and the ingot is melted. Example 2 differs from Comparative Example 3 in that flux 3 is blended at 10% by mass with respect to 1% by weight of the Ti—Al alloy.

When the surface of the ingot taken out after completion of the melting operation was visually confirmed, a layer (surface-adhering flux layer 8) which was considered to be formed by dissolution / solidification of the added flux 3 was adhering to the surface of the ingot. When the inside of a similar ingot was observed by SEM, it was found that almost no non-metallic inclusions (oxide inclusions) such as Al ₂ O ₃ were present in the inside of the ingot. It is presumed that this is because non-metallic inclusions such as Al ₂ O ₃ have migrated to the flux layer formed on the ingot surface.

Moreover, the flux layer (surface adhesion flux layer 8) on the surface of the ingot could be easily removed by shot blasting. As a result of analyzing the oxygen concentration of the ingot from which the surface adhesion flux layer 8 has been removed by shot blasting by the inert gas melting method, the oxygen concentration in the area where Al ₂ O ₃ is not present is 0.045 mass%, which is slight The oxygen concentration was as very low as 0.065% by mass even at sites where oxide inclusions such as Al ₂ O ₃ were observed. From this, the flux 3 is blended in the first step and the melting material 2 is divided, and in Example 2 in which the melting operation and the pulling-down operation are repeatedly performed for each divided body in the second step, the oxygen concentration in the ingot Can be greatly reduced.

Further, when the ingot of Example 2 described above is melted using a plasma arc melting furnace and pure Ti 10 (oxygen concentration 0.05 mass%) is added to melt a Ti-30 mass% Al alloy ingot, The oxygen concentration of the Ti-30 mass% Al alloy ingot decreased to 0.057 mass%.
From this fact, in the fourth step of Example 2, unlike in Comparative Examples 1 to 3, almost no oxide inclusions were present in the ingot, so the oxygen concentration increased even if it was remelted. It turned out that there was no decline in yield either. Therefore, using the production method of the present invention, it is possible to efficiently produce low-oxygen Ti-Al-based alloy 1 having an Al content of less than 60% by mass using low-grade raw materials with high yield. It is determined that
"Example 3"
Example 3 is the same as Comparative Example 3, in which the melting material 2 is divided into 11 pieces and the ingot is melted. Example 3 differs from Comparative Example 3 in that flux 3 is blended at 10% by mass with respect to 1% by weight of the Ti—Al alloy.

When the surface of the ingot taken out after completion of the melting operation was visually confirmed, a layer (surface-adhering flux layer 8) which was considered to be formed by dissolution / solidification of the added flux 3 was adhering to the surface of the ingot. When the inside of a similar ingot was observed by SEM, it was found that almost no non-metallic inclusions (oxide inclusions) such as Al ₂ O ₃ were present in the inside of the ingot. It is presumed that this is because non-metallic inclusions such as Al ₂ O ₃ have migrated to the flux layer formed on the ingot surface.

Moreover, the flux layer (surface adhesion flux layer 8) on the surface of the ingot could be easily removed by shot blasting. As a result of analyzing the oxygen concentration of the ingot from which the surface adhesion flux layer 8 has been removed by shot blasting by the inert gas melting method, the oxygen concentration in the area where Al ₂ O ₃ is not present is 0.16 mass%, which is slight The oxygen concentration was as very low as 0.20 mass% even at the site where oxide inclusions such as Al ₂ O ₃ were identified. From this, the flux 3 is blended in the first step and the melting material 2 is divided, and in Example 3 in which the melting operation and the pulling-down operation are repeatedly performed for each divided body in the second step, the oxygen concentration in the ingot Can be greatly reduced.

Further, when the ingot of Example 3 described above is melted using a plasma arc melting furnace and pure Ti 10 (oxygen concentration 0.05 mass%) is added to melt a Ti-30 mass% Al alloy ingot, The oxygen concentration of the Ti-30 mass% Al alloy ingot decreased to 0.15 mass%.
From this, in the fourth step of Example 3, unlike in Comparative Examples 1 to 3, almost no oxide inclusions were present in the ingot, so the oxygen concentration increased even if it was remelted. It turned out that there was no decline in yield either. Therefore, using the production method of the present invention, it is possible to efficiently produce low-oxygen Ti-Al alloy 1 having an Al content of less than 45% by mass using low-grade raw materials with high yield. It is determined that
"Example 4"
Example 4 is the same as Comparative Example 3, in which the melting material 2 is divided into 11 pieces and the ingot is melted. Example 4 is different from Comparative Example 3 in that the flux 3 is blended at 10% by mass with respect to 1% by weight of the Ti—Al alloy.

When the surface of the ingot taken out after completion of the melting operation was visually confirmed, a layer (surface-adhering flux layer 8) which was considered to be formed by dissolution / solidification of the added flux 3 was adhering to the surface of the ingot. When the inside of a similar ingot was observed by SEM, it was found that almost no non-metallic inclusions (oxide inclusions) such as Al ₂ O ₃ were present in the inside of the ingot. It is presumed that this is because non-metallic inclusions such as Al ₂ O ₃ have migrated to the flux layer formed on the ingot surface.

Moreover, the flux layer (surface adhesion flux layer 8) on the surface of the ingot could be easily removed by shot blasting. As a result of analyzing the oxygen concentration of the ingot from which the surface adhesion flux layer 8 has been removed by shot blasting by the inert gas melting method, the oxygen concentration in the area where Al ₂ O ₃ is not present is 0.042 mass%, which is slight The oxygen concentration was as very low as 0.060% by mass even at the site where oxide inclusions such as Al ₂ O ₃ were observed. From this, the flux 3 is blended in the first step and the melting material 2 is divided, and in Example 4 in which the melting operation and the pulling-down operation are repeatedly performed for each divided body in the second step, the oxygen concentration in the ingot Can be greatly reduced.

Further, when the ingot of Example 4 described above is melted using a plasma arc melting furnace and pure Ti 10 (oxygen concentration 0.05 mass%) is added to melt a Ti-30 mass% Al alloy ingot, The oxygen concentration of the Ti-30 mass% Al alloy ingot decreased to 0.055 mass%.
From this, in the fourth step of Example 4, unlike in Comparative Examples 1 to 3, almost no oxide inclusions were present in the ingot, so the oxygen concentration increased even if it was remelted. It turned out that there was no decline in yield either. Therefore, using the production method of the present invention, it is possible to efficiently produce low-oxygen Ti-Al-based alloy 1 having an Al content of less than 52% by mass using low-grade raw materials with high yield. It is determined that

  It should be understood that the embodiments disclosed herein are illustrative and non-restrictive in every respect. In particular, in the embodiment disclosed this time, matters not explicitly disclosed, such as operating conditions and conditions, various parameters, dimensions of components, weights, volumes, etc., deviate from the range normally practiced by those skilled in the art. It is not necessary for the person skilled in the art to use values that can easily be assumed.

DESCRIPTION OF SYMBOLS 1 Ti-Al type alloy 2 Melt raw material 3 Flux 41 1st division body 42 2nd division body 43 3rd division body 5 Water-cooled copper crucible 6 Crucible main body 7 Bottom part 8 Surface adhesion flux layer 9 Water-cooled copper container 10 Pure Ti

Claims (2)

CaO-CaF ₂ in which 35 to 95 mass% of calcium fluoride is mixed with calcium oxide in a Ti—Al alloy containing 0.1 mass% or more of total oxygen and 40 mass% or more of Al consisting of a titanium material and an aluminum material With respect to the molten raw material compounded so as to be 3 to 20% by mass with respect to the Ti-Al based alloy, the molten raw material after the division with respect to the final target ingot weight at the end of casting The first step of dividing into n parts so that the weight of the part is at most 4/5 or less of the whole,
The first divided body of the molten raw material divided in the first step is placed in a water-cooled copper crucible and melted under an inert gas atmosphere of 1.33 Pa or more, and the bottom of the water-cooled copper crucible is 15 mm or less per minute And the second divided body of the molten raw material divided in the first step is charged into the water-cooled copper crucible and melted under an inert gas atmosphere of 1.33 Pa or more and A second step of forming the ingot by performing the operation of pulling down the water-cooled copper crucible at a speed of 15 mm or less per minute and thereafter repeating the operation up to the n-th divided body;
A third step of mechanically removing a flux layer attached to the surface of the ingot formed in the second step. A method of manufacturing a Ti-Al based alloy, comprising: A titanium material is added to the ingot from which the flux layer is removed in the third step, and the fourth step of melting the ingot by a melting method using a water-cooled copper container in an atmosphere of 1.33 Pa or more is performed. The method for producing a Ti-Al based alloy according to claim 1, wherein a Ti-Al based alloy having an Al content of less than 40% by mass is obtained.