GB2106138A - Single crystal nickel alloy casting - Google Patents

Abstract

The alloy comprises by weight percent: 8-10% Chromium 0-15% Cobalt 1.5-3% Titanium+1 DIVIDED 2 Niobium 5-6.5% Aluminium 3-10.5% Tungsten 0-3.5% Molybdenum 0-3.5% Tantalum 0-0.5% Hafnium 0-1.5% Vanadium 0.015-0.05% Carbon 0-0.01% Boron 0-0.05% Zirconium Balance Nickel plus incidental impurities The invention also contemplates a heat treatment for the alloy comprising initially heating to 1300-1320 DEG C followed by maintaining at a temperature of about 870 DEG C for 16 hours, with an intermediate step at 1080-1120 DEG C for one hour if desired.

Description

SPECIFICATION An alloy suitable for making single-crystal castings and a casting made thereof This invention relates to an alloy suitable for making single-crystal castings and to a casting made thereof.

In the manufacture of castings of nickel-based alloys which are intended.for use at high temperatures under difficult conditions of stress and corrosion attack, it has been appreciated that castings made as a single crystal offer potential advantages in their combination of life and resistance to elevated temperatures. The main area in which these properties are of use lies in the hotter parts, such as nozzle guide vanes and turbine rotor blades, of gas turbine engines. However, the commonly-used nickel-based superalloys, although highly developed as materials for equi-axed conventional castings, have ranges of constituents which may not be beneficial to the properties of a single-crystal casting of the material.

Taking a modern nickel-based superalloy as a basis, we have invented a range of alloy compositions giving very good properties when used in the form of a single crystal casting.

According to the present invention an alloy suitable for use in the form of a single-crystal casting comprises, by weight percent, 8-10% Chromium 0-1 5% Cobalt 1.5-3% Titanium + 2 Niobium 56.5% Aluminium 3-10.5% Tungsten 0-3.5% Molybdenum 0-3.5% Tantalum 04.5% Hafrium 01.5% Vanadium 0.01 5.05% Carbon 0-0.01% Boron 0-0.05% Zirconium Balance Nickel plus incidental impurities.

Preferably the alloy comprises, by weight percent, 8-10% Chromium 2-11% Cobalt 1.7-2.6% Titanium + 21 Niobium 5.25-5.75% Aluminium 8.5-10.5% Tungsten 2.5-3.2% Tantalum 0.015-0.05% Carbon 0-0.01% Boron 0.0.01% Zirconium Balance Nickel plus incidental impurities.

The alloy may be solution heated to a temperature of about 13000 C; thus in specific cases the heat treatment was in the range 1260-1 3200C for 4 hours followed by a gas fan quench and treatment at 1090%C for 1 hour and followed by 8700C for 1 6 hours.

The invention also includes a cast object in single crystal form made of alloy, and particularly a cast gas turbine rotor blade in single crystal form made of the alloy.

Tests were carried out to confirm the properties of alloys in accordance with the invention, and these are described below with reference to the accompanying drawings in which: Figure 1 is a bar-chart of the life of various alloy test pieces under stress at a first, lower temperature, and Figure 2 is a chart similar to Figure 1 but of results obtained at a second, higher temperature.

In order to test alloys in accordance with the invention, test pieces of the various alloys were made in cast single crystal form. There are two basic methods by which single-crystal castings may be made; these are now well-known in the art, and generally comprise either the use of a seed crystal or the use of directional solidification followed by the use of a labyrinthine passage which serves to select a single crystal of the alloy which grows to form the test piece.

Although either method could have been used, we found it convenient to use the latter technique and for each of the various test alloys a test piece in the form of a single crystal was made. In order to provide a control, a similar test piece was cast in the form of a directionally solidified material. As will be appreciated, the directionally solidified material consists of a plurality of individual grains all of which are parallel, and this type of material usually has better properties than would a conventional orequi-axed casting of the same alloy.

In the accompanying drawings the results are compared for ten single-crystal test pieces and one directionally solidified control which were made and were tested for their stress-rupture lives at a lower temperature (7600C) and at a higher temperature (1 0400 C).

The compositions each were as follows, although it should be noted that in the case of alloy 9 the actual analysis may have differed slightly from the 'target' figures:- TABLE 1 Chromium Cobalt Titanium Aluminium Tungsten Tantalum Hafnium Carbon Boron Zirconium Control 8.7 9.8 1.6 5.4 9.9 2,5 1.23 0.166 0.014 0.038 1 8.7 9.8 1.6 5.4 9.9 2.5 1.23 0.166 0.014 0.038 2 8.6 10 1.6 5.3 10 2.6 0.51 0.015 0* 0* 3 8.5 10 1.6 5.3 9.9 2.55 0.94 0.015 0* 0* 4 8.7 10 1.6 5.5 9.7 2.4 1.16 0.015 0* 0* 5 8.6 9.9 1.6 5.4 10 2.55 1.15 0.015 0* 0.035 6 8.5 10.2 1.6 5.2 9.8 2.6 1.18 0.15 0* 0* 7 8.8 9.9 1.6 5.5 9.7 2.6 1.15 0.015 0.015 0* 8 8.6 5.1 1.74 5.45 9.9 2.35 0* 0.015 0* 0* 9 9 0* 1.5 5.5 10 2.5 0* 0.015 0* 0* 10 8.75 10 1.71 5.23 9.84 2.73 0* 0.015 0* 0* lt should noted thet it is not possible to remove all trace of these slements; therefore although not intentionally present traces of these elements will be left in the alloy.

It will be seen that alloy 1 is identical to the control alloy; any difference in properties is therefore due to the structural difference between a single crystal casting (alloy 1) and a directionally solidified casting (control). The remaining alloys each demonstrate the effect of a change in the constitution of the material. It will be seen that alloys 2,8,9 and 10 fall within the range of the present invention, while 1 and 3-7 and outside the invention.

In all cases illustrated in Figures 1 and 2 in unbroken lines the alloys were solution heat treated at a temperature of 8700 C for 16 hours.

It should be noted that the alloys in accordance with the invention are all capable of being heat treated to a higher temperature than are the other alloys in the diagram. Thus they can be solution treated at 1 3200C, above the solvus and yet below the incipient melting point. In comparing the results for the various alloys it must be bonie in mind that the unbroken lines represent the best results possible for the alloys outside the invention, but not for alloys 2, 8, 9 and 10 which are in accordance with the invention. The optimum results for alloys 8, 9 and 10 are indicated in broken lines, but those for alloy 2 are not available. Clearly however, alloy 2 is capable of better results than those displayed.

Referring to the diagrams, Figure 1 illustrates the life of samples of the various alloys when maintained under a stress of 730 MPa at a temperature of 7600C. This is a standard test procedure to determine the stress-rupture properties of the alloy and will not be further elaborated upon in this specification. It will be seen that the control alloy has a life of less than 80 hours, while the same alloy in single crystal form (alloy 1) has almost double life at 145 hours. The first alloy in accordance with the invention (alloy 2) has a life of almost 260 hours, a dramatic improvement. Alloys 3,4 and 5 which are outside the invention, do not reach this peak although they demonstrate an improvement with lives of over 200, over 200 and over 230 hours respectively.

Alloys 6 and 7 fall outside the invention by virtue of high carbon level for alloy 6 and in the case of alloy 7, high Boron level. Neither of these alloys is as good as Alloy 1, and in the case of alloy 7 the life is almost as low as in the case of the control. Alloys 8 and 10 again return to the levels shown in the case of alloys 3 and 4 (over 200 hours) while alloy 9, on the borderline of the invention because of its very low Cobalt level, is no better than the control. It is interesting to note that this alloy is somewhat unstable after soaking for 500 hours at elevated temperatures in the range 850-1 0500 C. The better figures achieved after high temperature treatment (iliustrated in broken lines) are discussed below.

Turning now to Figure 2, this shows the results obtained by a similar test procedure to that referred to above but at a higher temperature (1 0400 C) and corresponding lower stress (128 MPa). It will be seen that the lower temperature behaviour is largely repeated. Thus the single crystal alloy 1 is similar to the control, the alloys 2-5 show a considerable improvement over these two. Alloy 6 is much less effective, and alloy 9 is rather better while alloy 8 shows considerable promise at this condition and in fact comprises a good all-round balance of properties especially when heat treated (see below). Alloy 7 is similar to Alloy 5 in this hot condition, but of course its very poor lower temperature result tells against it.

Although the stress-rupture life is very important to the utility of an alloy, it is not the sole parameter to be considered, however, the changes made in accordance with the invention are unlikely to deleteriously affect any significant alloy parameters and are found to improve certain of them.

Thus the incipient melting point is increased by the lower levels of Boron and Zirconium and Hafnium which are melting point depressants. This is advantageous both because it gives a straightforward opportunity to raise the operating temperature of the alloy and because it allows a higher temperature solution heat treatment to be used. By raising the temperature of the heat treatment the degree to which the 1/ strengthening phase is taken into solution is increased, and the stress-rupture life is increased even more. Thus the alloys of the invention generally possess a 'window' of temperatures above the solvus but below the incipient melting point and in which they may be heat treated. A convenient general technique for this heat treatment is to heat at 1260-1 3200C for 4 hours followed by 16 hours at 870"C.

The following table demonstrates the increase in incipient melting point and the stress rupture results obtained using this higher temperature heat treatment available with alloys of the invention. It will be seen that the stress-rupture life is again quoted at the conditions used to obtain the results shown in Figures 1 and 2, the results are therefore comparable. The modified heat treatment includes cooling from the solution treatment temperature to room temperature and a subsequent ageing heat treatment at a lower temperature.

TABLE II

Life at stress Life at stress Incipient 730 MPa 128 MPa Melting temp. temp.

Material Point Heat Treatment 7600C 1 0400C Alloy 1 11800C 16hrsat87O0C 145 hrs 104 hrs Alloy 9 1320 C 16 hrs at 8700C 75 hrs 109 hrs 1 hr at 1320 C +16 hrs at 870 C 187 hrs 284 hrs Alloys 13200C 15 hrs at 870 c 216 hrs 203 hrs 1 hr at 1320 C +16hrsat87O0C 435 hrs 325 hrs Alloy 10 13000C 16hrsat87O0C 210 hrs 177 hrs 1 hr at 1300 C +16 hrs at 870 C These results for alloys 8, 9 and 10 are shown in broken lines in the diagrams Figs. 1 and 2, and it will be noted that all three alloys demonstrate properties better than those outside the invention. Alloy 2, which may also be heat treated in the vanes, would be expected to show a similar improvement although in this case the borderline level of Hafnium leads to the incipient melting point becoming perilously close to the heat treatment temperature. The temperature control must therefore be very accurate.

It should be noted that between the initial high temperature solution treatment and the later lower temperature ageing treatment and the piece was cooled to room temperature. We find that this cooling step and the rate at which it is carried out can provide significant differences in the final properties. Thus a cooling rate of 70-2000C per minutes for this step can lead to a marked improvement in properties.

It is clear from the above table that a considerable increase in stress-rupture life can be achieved by the use of a high temperature solution heat treatment. This is especially true of Alloy 8 which approximates to a preferred alloy.

It will also be noted that in the case of alloys 8 and 10 the level of Titanium is relatively high at 1.7% or more. We find this level is very important, and the following table Ill shows the result of tests carried out to demonstrate this.

We also find that the properties after heat treatment may be further improved by an intermediate step between the solution and ageing heat treatments. This will comprise keeping the alloy at some 1080-1120 C for about one hour between these treatments. In tests using an alloy nominally the same as alloy 8 we found that solution heat treating at 1 3000 C for 4 hours followed by a gas fan quench to room temperature, 1 hour at 1080 C and 16 hours at 870 C gave a stress-rupture life of 180 hours at 750 c and a load of 850 MPa. The same alloy with the same heat treatment sequence except that the intermediate step was at 1 0900C for one hour gave a life of 200 hours under the same conditions, while a third test with the intermediate step at 11 000C for one hour gave a life of 170 hours.It should be noted that these results are not directly comparable with those in Table II since the test conditions differed, but they do demonstrate an improvement in properties.

TABLE III

Heat Stress Stress Alloy Composition (Wt.%) Treatment Rupture Rupture Window Solution Conditions Life Alloy Cr Al Tt Co W Ta C ( C) Treatment (MPa/RC) (hrs) 11 8.7 5.32 1.77 5.14 9.65 2.36 .01 1260-1320 1 hr 1320 C 730/760 435 Air Cool + 128/1050 325 16 hrs 870 C 12 8.57 5.44 1.48 5.02 8.86 2.12 .0.15 1260-1320 1 hr 1320 C 730/760 110 Air Cool + 128/1040 103 16 hrs 870 C 13 8.40 5.41 1.45 5.04 9.3 3.01 .01 1280-1310 2 hrs 1300 C, 730/760 268 Air Cool + 128/1040 241 16 hrs 870 C 14 8.38 5.43 2.13 4.91 9.24 2.98 0.015 1290-1300 4 hrs 1300 C 730/760 598 Air Cool+ 128/1040 497 16 hrs 870 C It will be seen that a series of alloys were tested. Alloy 11 approximates to the preferred alloy 8 of Table I. The significant differences between alloys 11 and 12 are the reduced Titanium from 1.77 to 1.48 and the reduced Tantalum from 2.36 to 2.12. It will be seen that the stress rupture properties of alloy 12 are considerably reduced compared with those of alloy 11.

Alloy 13 has the Tantalum level restored to 3.01%, and although this does have some effect on the properties, it does not restore them to the level of Alloy 11. Alloy 1 4, however, which has a high Titanium level of 2.13%, has properties which not only equal but considerably exceed those of Alloy 11.

These and other tests lead us to the conclusion that there is a sharp change in properties at a Titanium level of from 1.5 to 1.6%. At 1.5% or thereabouts the properties are adequate, but at 1.6% or above the properties become exceedingly good. We feel that this is susceptible to explanation in terms of a change in deformation mechanism which occurs as a result of the change in constituents.

It will be noted that the alloys in accordance with the invention are defined as having a content of Titanium plus 2 Niobium greater than 1.5% rather than Titanium alone. This is because we believe that Niobium can be used to substitute for Titanium in equivalent atomic amounts, and this again implies that approximately two parts by weight of Niobium should be substituted for one part of Titanium. Hence the requirement that the Titanium content plus 2 that of Niobium should be kept within the specified limits.

This thinking is reflected in the composition of the first three alloys, referred to as alloys 1 5-17 inclusive, of the following table, IV in which properties of constituents are set out in weight percent.

TABLE N Alloy Cr Cn Ti Al W Ta Mo Nb V C 15 8.5 5 2.2 5.5 9.5 2.8 - 0.015 16 8.5 5 1.1 5.5 9.5 2.8 - 2.1 - 0.015 17 8.5 5 - 5.5 9.5 2.8 - 4.3 - 0.015 18 8.5 5 2.6 5.5 9.5 1.4 - - - 0.015 19 8.5 5 2.9 5.5 9.5 - - - - 0.015 20 8.5 5 2.2 5.5 7.75 2.8 - - 0.5 0.015 21 8.5 5 2.2 5.5 6.5 2.8 - - 1.0 0.015 22 8.5 5 2.2 5.5 6.6 2.8 1.5 - - 0.015 23 8.5 5 2.2 5.5 3.8 2.8 3.0 - 0.015 24 8.5 5 2.2 6.25 3.2 2.8 - - - 0.015 25 8.5 2.2 5.5 9.5 2.8 - - - 0.015 26 8.5 10 2.2 5.5 9.5 2.8 - - - 0.015 27 8.5 15 2.2 6.5 9.5 2.8 - - - 0.015 Alloy 13 represents an optimum composition, while Alloys 14 and 13 are versions in which the Titanium content is replaced partially and fuliy by Niobium. This substitution gives a potential hardening of the y' phase of the material.

A similar effect is achieved in Alloys 16 and 17 by substituting Titanium for Tantalum. The y phase is hardened in Alloys 18 and 19 by the substitution of Vanadium for Tungsten, which provides a stronger alloy but of somewhat reduced oxidation resistance. The same effect is achieved in Alloys 20 and 21 by substituting Molybdenum forTungsten. In Alloy 22 the balance of the alloy is altered by the substitution of Aluminium plus Titanium for Tungsten, softening they phase but providing an improvement in other properties of the alloy.

Finally, alloys 22, 23 and 24 alter the alloy balance by variation in the Cobalt level.

It will be noted that some of the alloys have any deliberate addition of Boron or zirconium.

Although the above results concentrate on stress-rupture properties, we find that other parameters are also improved.

Thus we find that by using Tantalum to replace some of the Tungsten and Molybdenum solid solution strengtheners, an improved resistance to corrosion and oxidation may be achieved. We have demonstrated the importance of this Tantalum addition by comparative tests of alloys with and without Tantalum. Thus for alloy 10 its stress rupture life at a stress of 730 MPa and temperature of 7600C was 210 hours, while its life at 128 MPa and 1 0400C was 177 hours (see table). Test pieces were prepared of an alloy nominally identical but without Tantalum, and the corresponding lives were 67 and 51 hours. The Tantalum is clearly necessary at the amounts tested, and in fact we believe that additions up to and perhaps just above 3% may be beneficial.

Further testing of the oxidation resistance of the alloy confirmed the importance of the Tantalum addition. Thus three samples, one of the control alloy, one of Alloy 8 but without Tantalum and one of Alloy 8 with Tantalum replaced by Tungsten were subject to 90 hours at 1 0500C in an atmosphere of air plus 4 parts per million of salt. After this time the control alloy showed 140,u attack, while both the other samples had a more severe attack of 200,u.

Also the low levels of Carbon give rise to relatively small carbide particles, and thus avoid the large, scriptic morphology carbides which can reduce the oxidation performance of aluminide coated (also true of D.S.) nickel-based superalloys. As an example, we carried out corrosion tests on an aluminide coated isotropic material otherwise similar to the control material referred to above, and found that with fine carbides the life was,more than doubled compared with the normal state.

We also find that the impact strength of the materials in accordance with the invention is considerably improved when compared with that of the unmodified directionally solidified material. We believe that this effect is due to the absence of the scriptic carbides.

The stability of the alloy, which is determined by the Aluminium, Titanium, Chromium, Tungsten and Cobalt levels is not affected to any serious degree, by the modifications specified, however, it will be appreciated that by taking alloys whose constituents are at the extremes of the claimed ranges it may be possible to produce alloys of reduced stability, which may of course still be useful alloys.

It will therefore be seen that the alloys in accordance with the invention provide properties suitable for articles operating under highly stressed conditions at high temperatures, such for instance as turbine rotor blades of gas turbine engines. However, the properties are such that the alloy could be used in a wide range of other uses, for instance in other parts of gas turbines.

Claims (10)

1. An alloy suitable for use in the form of a single-crystal casting comprising, by weight percent, 8-10% Chromium 0-1 5% Cobalt 1.5-3% Titanium + 2 Niobium 56.5% Aluminium 310.5% Tungsten 0-3.5% Molybdenum 0-3.5 Tantalum 0-0.5 Hafnium 0-1.5% Vanadium 0.0150.05% Carbon 0.01% Boron 0#).05% Zirconium Balance Nickel plus incidental impurities.

2. An alloy as claimed in claim 1 and comprising by weight percent 8-10% Chromium 2-11% Cobalt 1.7-2.6% Titanium + 2 Niobium

5.25-5.75% Aluminium 8-5-10.5% Tungsten

2.5-3.2% Tantalum 0.015-0.05% Carbon 0-0.01% Boron 0-0.05% Zirconium Balance Nickel plus incidental impurities.

3. An alloy as claimed in claim 2 and comprising by weight percent: 8-10% Chromium 3-7% Cobalt

1.7-2.6% Titanium + 2 Niobium

5.25-5.75% Aluminium

8.5-10.5% Tungsten

2.5-3.2% Tantalum 0.015-0.05% Carbon 0-0.01% Boron 0-0.05% Zirconium Balance Nickel plus incidental impurities.

4. An alloy as claimed in claim 3 and comprising by weight percent:

8.5% Chromium 5% Cobalt

2.2% Titanium

5.5% Aluminium

9.5% Tungsten

2.8% Tantalum 0.015% Carbon Balance Nickel plus incidental impurities.

5. A solution heat treatment for the alloy of any one of the claims 1-4 and comprising heating the alloy to a temperature in the range 1260-1 3200C.

6. A solution heat treatment as claimed in claim 5 and comprising cooling the alloy from said temperature to room temperature at a rate of 70 to 200 C per minute.

7. A heat treatment as claimed in claim 5 or claim 6 in which the alloy is subsequently heated to a temperature of about 870 C.

8. A solution heat treatment as claimed in claim 7 and comprising heating the alloy to a temperature of about 1 3000C for 4 hours, cooling the alloy to room temperature of about 8700C and maintaining it at this temperature for 16 hours.

9. A heat treatment as claimed in claim 7 or claim 8 and comprising an intermediate heat treatment step in which the alloy is heated to a temperature of between 1080 and 11 200C for about one hour.

10. A single crystal casting formed from an alloy as claimed in any one of claims 1 4 above.