Volume 13 - Corrosion Part 12 potx

16 Effect of composition, cooling rate, and sintering temperature on corrosion resistance of type 304L and tin-modified type 304L P/M stainless steels sintered density: 6.5 g/cm 3 ; sint

absorption follows Sievert's law; that is, absorption is proportional to the square root of the partial pressure of nitrogen in the sintering atmosphere This nitrogen absorption provides significant strengthening (Fig 3) Upon completion of sintering, when the part enters the cooling zone of the furnace, the solubility of nitrogen decreases sharply with temperature (Fig 12) As a result, Cr2N begins to precipitate at the temperature at which the nitrogen content crosses the solubility limits More important, below about 1150 °C (2100 °F), additional nitrogen is absorbed from the sintering atmosphere, leading to more Cr2N precipitation and chromium depletion along the grain boundaries The net result is inferior corrosion resistance due to grain-boundary corrosion

Fig 12 Solubility of nitrogen in austenitic stainless steel in equilibrium with gaseous nitrogen or Cr2 N Source: Ref 9

The rate of this detrimental nitrogen absorption increases with decreasing part density and with decreasing dew point A high dew point, however, leads to the problem of excessive oxidation The basic relationship of this phenomenon is shown in Fig 13 The data in Fig 13, which were developed for the bright annealing of stainless steel in dissociated NH3atmospheres, show the extent of nitrogen and oxygen absorption as a function of dew point At high dew points (higher than about -37 °C, or -35 °F, depending on part size), the rate of oxidation is severe enough to produce a dull surface At dew points of about -45 °C (-50 °F) or lower, nitrogen absorption increases so much that the corrosion resistance deteriorates because of excessive Cr2N formation Thus, optimum bright annealing of austenitic stainless steels must be done within a narrow dew-point range Although the authors (Ref 14) caution against applying these findings to sintered stainless steels based on the unexplained higher nitrogen contents found for their parts sintered in dissociated NH3, it should be noted that such higher nitrogen contents are expected on the basis of known solubility data for nitrogen in type 316L (Fig 12) considering the differing methods of nitrogen analysis used

Fig 13 Safe operating parameters with respect to dew point can be developed for a specific set of operating

conditions and quality requirements The safe zone here is for sintering in an atmosphere of 30% H2-70% N2 at

1035 °C (1900 °F) Source: Ref 14

Chromium nitride sensitization may in some cases be limited to a very shallow surface depth of the part With very slow cooling, however, absorption and precipitation proceed toward the interior of the porous part Figure 14 shows Cr2N precipitates in the grain boundaries of parts that were sintered under conditions that produced nitrogen contents from 55

to 6650 ppm Increasing nitrogen content correlates with increasing amounts of precipitation and increasing localized corrosion (Fig 14) Figure 15 shows the microstructure of a type 316L part that was sintered in dissociated NH3 and cooled very slowly Slow cooling produced a lamellar structure of Cr2N and low-chromium austenite of very poor corrosion resistance

Fig 14 Scanning electron micrographs of type 316L stainless steel (a) Sintered 45 min in 100% H2 at 1350 °C (2460 °F); 66 ppm N (b) Sintered 45 min in 75% H 2 at 1350 °C (2460 °F); 3100 ppm N (c) Sintered 45 min

in 25% H 2 at 1350 °C (2460 °F); 4300 ppm N (d) Sintered 45 min in 25% H 2 at 1150 °C (2100 °F); 6650 ppm

N The amount of intergranular precipitate increases with nitrogen content Source: Ref 13

Fig 15 Micrograph showing the lamellar structure of Cr2N and low-chromium austenite in sintered type 316L that was slowly cooled in dissociated NH 3 Etched with Marble's reagent 700× Source: Ref 9

Corrosion resistance data for sintered types 304L and 316L in NaCl solutions and in 10% HNO3, reflecting the effect of

Cr2N precipitation, are shown in Fig 8, 16, and 17 Figures 8 and 16 show that a higher sintering temperature, fast cooling rates (75 °C/min, or 135 °F/min, versus 8 °C/min, or 14 °F/min), and the use of type 316L rather than type 304L provide better corrosion resistance That these measures are beneficial follows directly from the austenite-nitrogen phase diagram (Fig 12)

Fig 16 Effect of composition, cooling rate, and sintering temperature on corrosion resistance of type 304L and

tin-modified type 304L P/M stainless steels (sintered density: 6.5 g/cm 3 ; sintering atmosphere: dissociated

NH3) in 5% aqueous NaCl B rating indicates <1% of specimen surface stained Parenthetical values designate sintering temperature and cooling rate Source: Ref 19

Fig 17 Weight loss of austenitic stainless steel in 10% aqueous HNO3 as a function of absorbed nitrogen content Curve A: sintered in dissociated NH3 at 1150 °C (2100 °F) with a dew point of -43 °C (-45 °F)

Density: 5.10 to 5.20 g/cm 3 Curve B: sintered in various atmospheres with different dew points Density: 5.2

to 5.8 g/cm 3 Source: Ref 9

Figure 18 shows potentiodynamic corrosion curves for sintered type 316L in 10% HNO3 The corrosion current density in the passive range increases and the corrosion potential decreases under conditions that promote Cr2N precipitation, that is, lower sintering temperature, slower cooling rate, and high nitrogen concentration of the sintering atmosphere Figure 19 is similar to Fig 18 except that internal rather than external cross sections were used The significantly lower corrosion currents of the internal surface confirm that Cr2N precipitation is most severe on the surface of a sintered part

Fig 18 Forward scan potentiodynamic corrosion curves for external surfaces of three sintered type 316L

stainless steel samples in 10% HNO3 at 25 °C (75 °F) Note the increasing corrosion currents in the 0- to 1-V range and the decreasing corrosion potential with nitrogen additions to the atmosphere, slow cooling, and lower sintering temperatures SCE, saturated calomel electrode See also Fig 19 Source: Ref 20

Fig 19 Forward scan potentiodynamic corrosion curves of the internal microstructure (metallographic cross

section) for type 316L stainless steel samples sintered in 25% H 2 Corrosion susceptibility in 10% HNO 3 at 25

°C (75 °F) increases with a lower sintering temperature and slow cooling Cr 2 N precipitation is most severe on the surface of a sintered part See also Fig 18 Source: Ref 20

Recently developed tin-containing grades of type 304L (Table 2) and 316L stainless steels have shown less sensitivity to nitride precipitation and correspondingly improved corrosion resistance (Fig 8 and 16) The beneficial effect of tin has been confirmed in several studies (Ref 9, 10, 16, 19, 20, 21, 22) and has been attributed to an enrichment of the surfaces

of both the water-atomized powder and the sintered part with tin, presumably as a result of the low solubility of tin in solid stainless steel (Ref 10) Tin may also form stable acid-resistant passive films in a crevice and may cause cathodic surface poisoning, but its major beneficial effect is believed to lie in its formation of an effective barrier to nitrogen (and possibly also to oxygen) diffusion This reduces the rate at which nitrogen is absorbed on the surface of the sintered part

as it enters the cooling zone of the furnace Auger composition depth profiles of regular and (1.5%) tin-containing type 316L parts sintered in dissociated NH3 (Fig 20) show that the presence of tin on the surface effectively suppresses nitrogen absorption In addition, on the basis of potentiodynamic polarization tests in 10% HNO3 and 5 N H2SO4, improvement in corrosion resistance has also been reported due to the presence of tin (Ref 20, 22)

Fig 20 Auger composition depth profiles of P/M type 316L stainless steel parts sintered in dissociated NH3 at

1175 °C (2150 °F) (a) Type 316 L (b) Tin-modified type 316L Source: Ref 9

The effect of oxygen on the corrosion resistance of sintered stainless steels is probably the most complex and least understood variable for several reasons First, commercial water-atomized compactible stainless steel powders have typical oxygen contents of about 2000 ppm or more Although much of this oxygen resides on the surfaces of individual powder particles as oxidized silicon (Fig 21a), the exact nature and distribution of the oxides depends on atomizing conditions Second, with typical industrial sintering practice, the reduction of these oxides remains incomplete and depends on many process parameters Lastly, as a sintered part enters the cooling zone of the furnace, certain elements will oxidize upon reaching the temperature for the oxide-metal equilibrium of the high oxygen affinity elements (Fig 22) Thus, a sintered part still reflects the history of its powder-making process, compaction, and sintering Figure 21(b) shows the Auger composition depth profile of a type 316L part after sintering in hydrogen at 1260 °C (2300 °F) It is apparent that much of the oxidized silicon present in the green part has become reduced and that severely depleted chromium has been replenished

Fig 21 Auger composition depth profiles of P/M type 316L stainless steel (a) Green part (b) Sintered part

Fig 22 Redox curves for chromium and silicon alone and in solution Source: Ref 9

An empirical correlation between the saltwater corrosion resistance of sintered type 316L and the oxygen content of the sintered parts suggests that sintering conditions resulting in lower oxygen contents provide better corrosion resistance (Fig 23) With excessive dew points (>-34 °C, or -30 °F), the oxygen content of a sintered part may increase considerably The microstructure (Fig 24) of such a part shows a lack of particle bonding (compare with Fig 10a for low oxygen content), and its mechanical strength and corrosion resistance are both inferior

Fig 23 Effect of oxygen content on corrosion resistance of sintered type 316L and tin-modified type 316L

(sintered density: 6.65 g/cm 3 ; cooling rate: 75 °C/min, or 135 °F/min) Parenthetical values are sintering temperature (°C), dewpoint (°C), and nitrogen content (ppm), respectively Time indicates when 50% of specimens showed first sign of corrosion in 5% aqueous NaCl Source: Ref 10

Fig 24 Microstructure of type 316L stainless steel sintered in a high dew point atmosphere Oxygen content:

5100 ppm; sintered density: 7.5 g/cm 3 Etched with Marble's reagent 200× Source: Ref 9

For optimum corrosion resistance, it appears that the following precautions are beneficial:

• Use of a powder with low oxygen content

• Sintering conditions that ensure a high degree of oxide removal

• Fast cooling through the high-temperature range after sintering

Cooling in a hydrogen atmosphere should be done with a water vapor content of less than 50 ppm (Ref 12) Cooling in a nitrogen-containing atmosphere should be done with a dew point between about -37 and -45 °C (-35 and -50 °F) (Ref 14)

Effect of Sintered Density. Applications of sintered stainless steels cover a wide density spectrum Low densities of about 5 g/cm3 may be typical of filters, but densities of 6.5 g/cm3 or greater are typical of structural parts It is therefore of interest to know the effect of density on corrosion resistance Corrosion studies of sintered austenitic stainless steels have shown that the corrosion resistance improves significantly with increasing density in acidic environments, such as dilute

H2SO4, HCl, and HNO3 Figure 25 illustrates this behavior for three austenitic stainless steels (18Cr-11Ni to 18Cr-14Ni) that were vacuum sintered 1 h at 1150 and 1250 °C (2100 and 2280 °F) and tested in boiling 40% HNO3

Fig 25 Relationship between sintered density and weight decrease of three austenitic stainless steels in 40%

HNO3 solution Source: Ref 23

For saline solutions, some investigations have found the effect of increasing density to be beneficial (Ref 20) while others have found it to be detrimental (Ref 10, 16, 23) This lack of agreement is perhaps not surprising considering that concentration changes in several of the critical variables, such as oxygen, carbon, and nitrogen, also depend on the density

of a part It should be noted, however, that the positive relationship between density and corrosion resistance was derived from short-term potentiodynamic polarization measurements (Ref 20), whereas the negative relationships were all derived from longer-term salt immersion tests

Table 6 summarizes recent results on the effect of density on the salt corrosion resistance immersion in 5% aqueous NaCl) of vacuum-sintered type 316L parts Unlike sintering in a reducing atmosphere, vacuum sintering does not lower the oxygen content with decreasing density Thus, an improvement in corrosion resistance with decreasing density, as shown in Table 6, should not be attributed to a lower oxygen content, but is perhaps better explained in terms of reduced crevice corrosion as a result of the improved circulation of the corrodent through large pores (Ref 9) The average pore diameters of the parts pressed at the lower compacting pressures (Table 6), as measured by mercury porosimetry, were 60

to 80% larger than the size of pores of the high-density parts The standard deviations of the pore size distributions were similar and were around 2 Therefore, sintered stainless steel parts with densities from about 60 to 90% of theoretical have average pore sizes from about 10 to 2 or 3 m that are likely to affect the circulation of the corrodent and thus its resistance to crevice corrosion

Table 6 Effect of density on corrosion performance of vacuum-sintered type 316L stainless steel

min

Sintered density, g/cm3

Median pore size

of pore volume(a)

(b) A, sample free from any corrosion; B, 1% of surface covered by stain; C, 1 to 25% of surface covered by stain with slight corrosion

product; D, >25% of surface covered by stain with heavy corrosion product

Effect of Copper Additions to Type 304L. One study found that the corrosion resistance of copper-containing type 304L vacuum-sintered parts (1 h at 1200 °C, or 2190 °F; 88% dense) improved with increasing copper content (Ref 17) Figure 26 shows the weight loss of the parts kept for 6 h in boiling 5% H2SO4 Higher nickel content is also beneficial Salt spray testing for 24 h with 5% NaCl solution resulted in almost no pitting The effect of copper in P/M stainless steels

is said to be identical to that observed in cast stainless steels

Fig 26 Effect of nickel and copper additions on the corrosion rate of sintered austenitic stainless steel

compacts exposed to boiling H 2 SO 4 for 6 h Relative sintered density is 88% Source: Ref 17

Oxidation Resistance. Sintered stainless steels are not widely used for elevated-temperature service Thus, information on elevated-temperature oxidation resistance is scarce

Figure 27 shows the weight gain in air at 700 °C (1290 °F) for type 310L stainless steel parts that were vacuum sintered 1

h at 1250 °C (2280 °F) as a function of sintered density (circular plates), mesh size of powder used, and sintering temperature This initial weight gain did not always show a parabolic course of oxidation Within the density range studied, oxidation increased almost exponentially with decreasing density Silicon-modified (4.06% Si) type 310L stainless steel showed weight gains that were less than 50% of those of regular type 310L The increased oxidation of the parts made from the finer powder fraction is due to their large internal surface area Higher sintering temperature and higher compacting pressure (higher densities) reduce surface porosity and specific pore surface area, thus lessening

internal oxidation through early pore closure The maximum recommended operating temperature for sintered austenitic stainless steels is 700 °C (1290 °F)

Fig 27 Weight gain versus sintered density curves for materials prepared from powders of various particle

sizes Parts were sintered 1 h at 1250 °C (2280 °F) Source: Ref 24

Higher-Alloyed Stainless Steels. Although the common stainless steel grades used in industry have maximum chromium and nickel contents of 20 and 14%, respectively (Table 2), higher-alloyed stainless steels have been used in the past to obtain improved corrosion resistance Such steels are available from powder procedures In one investigation, a high-nickel/chromium/molybdenum austenitic stainless steel P/M material (SS-100) performed comparably to wrought type 216, 316, or 317 in 16-h salt solution immersion tests (Ref 25)

Other Approaches to Improving the Corrosion Resistance of Sintered Stainless Steels. If the corrosion resistance of sintered stainless steel parts remains inadequate after composition and process optimization, passivation and coating treatments are sometimes used Chemical and thermal passivation treatments for sintered type 316L, effective in dilute H2SO4, are described in Ref 26 Chemical passivation with HNO3 solutions similar to those applied to wrought stainless steels is not suitable for every material On the basis of rest potential measurements of sintered type 316L, thermal passivation by heating the sintered parts for 20 to 30 min in air at temperatures of 400 to 500 °C (750 to 930 °F)

is recommended

In another study, the corrosion resistance of vacuum-sintered type 304L (6.9 g/cm3) in 5% H2SO4 was improved by activating the parts in a mixture of 13 to 15% HNO3, 2% hydrofluoric acid (HF), and 0.3% hydrochloric acid (HCl), followed by passivation for 30 min in 30% HNO3 at 70 °C (160 °F) (Ref 27) After testing for 2 h in 5% H2SO4 (Fig 28), the passivated specimens showed no weight loss, whereas the as-sintered specimens rapidly lost weight and turned the solution green In addition, Ref 28 describes a phosphate-base passivating treatment for sintered stainless steels that is effective in acetic acid

Fig 28 Relationship between weight loss and corrosion time of vacuum-sintered type 304L stainless steel in

5% H2SO4

Improvement of the corrosion resistance of sintered stainless steel through the chemical vapor deposition (CVD) or chromium is discussed in Ref 29 The chemical vapor deposition of chromium onto sintered stainless steel parts was applied by pack cementation Considerable infilling of the pores with chromium takes place; 50- m thick pores with diameters of up to 50 m may become sealed Immersion of coated and uncoated specimens in 5 and 10% H2SO4solutions for 168 h at room temperature showed significant attack of the uncoated specimens and no noticeable attack of the coated specimens Electrochemical testing in 5% H2SO4 gave similar results, and a 3% salt spray test at room temperature showed many local sites of corrosion for the uncoated specimen and no corrosion after 250 h for the coated specimen Sealing or coating of the pores of a sintered stainless steel part with an organic resin (Ref 25) or with water glass is sometimes recommended, but performance data proving the effectiveness of this treatment are lacking

Fully Dense P/M Stainless Steels

In the fully dense category of P/M stainless steels, parts made from water-atomized powders must be distinguished from those made from inert gas atomized powders

Water-atomized powders, because of their irregular particle shape, are cold compactible and permit the pressing of complex parts, which, at temperatures approaching the melting point of the material, can be sintered to nearly full density However, water-atomized stainless steel powders typically have oxygen contents of 2000 ppm or more, and sintering to full density usually does not reduce the oxygen content to the low level of the corresponding ingot material The commercial production of such parts is still in its infancy, and corrosion data are not yet available

Inert gas or centrifugally atomized powders are spherical and noncompactible They have low oxygen contents (about 50 to 200 ppm) and are consolidated to full density by such process as hot isostatic pressing (HIP), hot forging, and extrusion

One company has manufactured seamless stainless steel tubes from gas-atomized powder since 1980 The P/M method is said to offer a competitive alternative to conventional production methods due to:

• Efficient use of raw materials

• Low energy consumption

• Short total production time

• High flexibility (less material in process; short delivery times)

• The ability to make difficult compositions (Ref 30)

The process consist of cold isostatic compaction of the encapsulated nitrogen-atomized powder, followed by heating to the extrusion temperature and hot extrusion The capsule material is removed by decladding Standard grades include most of the common austenitic stainless steels as well as some special austenitic, ferritic-austenitic, and ferritic stainless steels, together with nickel-base alloys

In comparison to conventional material, the extruded P/M products possess a more homogeneous structure with reduced microsegregation due to the rapid cooling of the powder particles Also, the grain size is somewhat finer, slag inclusions (particularly sulfides) are smaller, and the nitrogen content is somewhat higher (900 versus 500 ppm for wrought type 316)

Mechanical Properties and Corrosion Resistance. Attributed to the above differences are slightly higher yield and tensile strength (Table 7) without a loss in elongation Mechanical properties at elevated temperatures are practically identical to conventionally produced materials The impact toughness of the P/M material, although good, is lower than that of conventional material when tested in the longitudinal direction Creep strength is similar to that of conventional material

Table 7 Typical mechanical properties of cold-worked and annealed stainless steel tubes extruded from powders

P/M 112 382 55 681 99 43

Source: Ref 30

(a) C, conventional production; P/M, powder metallurgy

No difference between P/M and conventional material has been found regarding the resistance to intergranular corrosion according to practice C and practice E of ASTM A 262 (Ref 31) As shown in Fig 29, the resistance to pitting attack, as measured by the pitting corrosion breakthrough potential, is superior for several P/M grades compared to the corresponding conventional grades Table 8 gives the general and selective corrosion information from tests according to ASTM A 262, practice C (Ref 31) for two austenitic P/M grades The improved corrosion resistance of the P/M grades is attributed to their lower segregation rate, their finer and more uniform distribution of inclusions, and their finer grain size

Table 8 Huey test (ASTM A 262, practice C) corrosion data for two P/M extruded stainless steels

Corrosion rate, m/48 h Selective attack, m Grade Number of

samples

Average Specific Average Specific

Type 725LN 14 0.57-0.69 1.5 max <50 100 max

Type 724L 14 1.48-1.79 3.3 max <30 200 max

Fig 29 Comparison of pitting resistance of P/M and conventional stainless steels Source: Ref 30

The enhancement of the corrosion resistance of stainless steel parts made from rapidly solidified powders has been confirmed by several investigators For example, the significantly superior oxidation resistance of type 303 stainless steel, made by extrusion of rapidly solidified powder, was attributed to the elevated-temperature grain growth inhibiting effect

of uniformly dispersed manganese sulfide (MnS) particles (Ref 32) Figure 30 shows that this material maintains its good

corrosion performance in aqueous environments, and potentiodynamic polarization curves in 1 M H2SO4 indicate that the P/M material exhibits the lowest corrosion rate at the corrosion potential Finally, although wrought type 303 was highly susceptible to pitting, the P/M alloy showed no obvious pits on the surface and only a low pit density within the material The pits were related to the presence of sulfide stringers in the wrought material, from which it was concluded that P/M steels with lower sulfur contents and with spherical sulfide morphology, such as type 304 and 316, might exhibit improved pitting resistance

Fig 30 Potentiodynamic polarization curves for conventional type 303 and 304 stainless steels and for rapidly

solidified type 303 in deaerated 1 M H2SO4 at 30 °C (85 °F) Source: Ref 33

Injection molding technology (see the article "Powder Injection Molding" in Powder Metal Technologies and Applications, Volume 7 of the ASM Handbook) is currently used to manufacture small and nearly fully dense stainless

steel parts from the powders However, information on the corrosion performance of such parts is unavailable

P/M Superalloys

Development of P/M superalloys began in the 1960s with the search by the aerospace industry (and later the electric power industry) for stronger high-temperature alloys in order to operate engines at higher temperatures and thus improve fuel efficiency Figure 31 illustrates the great advances achieved since the 1940s by the introduction of new processes and alloys, such as vacuum melting, directional solidification of eutectics, development of alloys with high volume fractions

of ' phase, and P/M processing with and without oxide dispersions

Fig 31 Trends in alloy processing and development Source: Ref 34

Initially, lower production costs were a major objective in exploring the P/M approach Figure 32 illustrates the material savings possible with two different P/M methods due to their near net shape capabilities Later, specific advantages linked

on the P/M approach, such as the use of more complex and greater volume fractions of dispersoids, reduced segregation, and improved workability, led to the development of stronger alloys and to the use of these alloys not only in turbine disks but also in the higher-temperature turbine blades

Fig 32 Processing sequence in the production of jet engine compressor disks

Much effort is currently being directed toward reducing the cost of consolidating superalloy powders, particularly of oxide dispersion strengthened (ODS) superalloys, through the development of suitable forging techniques (Ref 34) Efforts are also underway to exploit the advantage of microcrystallinity and extended solid solutions of rapid solidification technology

Uses and State of Commercialization. P/M superalloys were first used in military engines in the mid-1970s Table

9 summarizes the uses of P/M superalloys in terms of components, engine use, and reasons for using P/M technology Other uses of superalloys include nuclear reactors, heat exchangers, furnaces, sour gas well equipment, and other high-temperature applications

Table 9 Aerospace applications of P/M superalloys

Reasons for using P/M

technology

manufacturer

Cost reduction

Improved properties

IN-100 Turbine disks, seals, spacers F-100 Pratt &

Whitney

René 95 Turbine disks, cooling plate T-700 Helicopter/G.E

René 95 Turbine disks, compressor shaft F-404 F-18 Fighter X

René 95 High-pressure turbine blade retainer, disks, forward

Inconel MA-754 Turbine nozzle vane F-404 F-18 Fighter X

Inconel MA-754 High-and low-pressure turbine vanes Selected

as atomization by the rotating electrode process, are known to be suitable for producing powders with the required low oxygen content and low degree of contamination (details on these processes are available in the article "Atomization" in

Powder Metal Technologies and Applications, Volume 7 of the ASM Handbook) The so-called prior particle boundary

(PPB) problem, that is, the presence of carbides segregated at PPBs, was solved through the development of low-carbon alloys Special equipment is used for removing ceramic particles and particles containing entrapped argon Some of these problems are minimized or avoided in ODS alloys made by mechanical alloying In mechanical alloying, elemental and master alloy powders as well as refractory compounds are mechanically alloyed by high-energy milling (Ref 36, 37)

Two established powder consolidation techniques for P/M superalloys are hot isostatic pressing (HIP) and isothermal forging Figure 33 illustrates schematically the steps of the P/M processes in comparison to conventional processing Both P/M methods permit the manufacture of so-called near net shape parts with attendant improved material use and reduced machining costs Powder metallurgy forging exploits the improved forgeability deriving from the higher incipient melting temperature and reduced grain size of P/M material Hot compaction by extrusion leads to very fine grain size, improved hot ductility, and superplasticity

Fig 33 Comparison of conventional processing and P/M processing for the fabrication of superalloy disks

Compositions and Properties. Table 10 shows the compositions of the best known P/M superalloys Many have the same compositions as cast alloys but are manufactured similarly to wrought alloys The important P/M superalloys IN-

100, René 95, and Astroloy were adapted to the P/M process by reducing their carbon content and by adding stable carbide formers to eliminate the problem of PPB carbides To facilitate HIP, alloy compositions were modified to increase the temperature gap between the ' solvus (above which HIP has to be carried out for increasing grain size) and the solidus temperature

Table 10 Nominal compositions of several P/M superalloys

Composition, % Alloy

Fig 34 Advancing steps is the protection of superalloys against oxidation at high temperatures showing life (in

hours) to 0.25-mm (10-mils) penetration at 980 °C (1800 °F) Source: Ref 39

Cyclic oxidation causes protective scales of Al2O3 and Cr2O3 to crack and spall Regeneration of the scales will eventually result in the complete depletion of chromium and aluminum The length of time for which superalloys are Al2O3 or Cr2O3formers under given conditions is very important because of the subsequent appearance of less protective oxides The importance of chromium for imparting oxidation resistance is demonstrated in Table 11, which lists fatigue crack growth rates for different alloys

Table 11 Relative increase in fatigue crack growth rates after 15 min at 650 °C (1200 °F)

Alloy Relative increase in

crack growth rates

Astroloy after HIP + forge 3.5 50-100 14.7

Source: Ref 40

Figures 35 and 36 show comparisons of the oxidation resistances of superalloys with and without oxide dispersions May studies have confirmed the beneficial effect of dispersed oxides on oxidation resistance The lower oxidation rates of ODS alloys have been attributed to the reduced time required to form a continuous Cr2O3 scale due to the presence of dispersed oxides, which act as nuclei for oxidation (Ref 41) Based on marker studies with platinum, one investigation attributed the beneficial effect of oxide dispersions to the predominant, inward diffusion of oxygen ion (O2-) and slowdown of the chromic ion (Cr3+), diffusion (Ref 42) The latter may be caused by the blocking of the dispersions in the

Cr2O3 scale With the dispersed oxides becoming dissolved in the scale, it also appears possible that trivalent ions, such as yttrium (Y3+) and lanthanum (La3+), will reduce the number of vacant cation sites, thus lowering the diffusivity of Cr3+

Fig 35 Comparison of the oxidation resistance of ODS alloys MA 956, MA 754, and MA 6000 with that of other

superalloys Testing conditions: 504 h at 1100 °C (2010 °F) in air containing 5% H2O Temperature was cycled between test temperature and room temperature every 24 h Source: Ref 40

Fig 36 Cyclic oxidation of ODS alloys MA 956, MA 8077, MA 953, and TD-NiCr compared to that of coated alloy

MM 200 Testing conditions: held at 1100 °C (2010 °F) for 1 h and cooled by a 3-min air blast Source: Ref 40

Dispersed oxides may also improve scale adhesion because of the thinner scale or because of increased porosity or smaller grain size in the oxide scale (Ref 40) It was reported that at 1300 °C (2370 °F) the outer regions of the Al2O3 film

of MA 956 (Fe-20Cr-4.5Al-0.5Ti-0.5Y2O3) became enriched with titanium, giving rise to a continuous layer of rich oxide (Ref 43) Pegging of the oxide titanium carbide particles and the irregular metal/oxide interface is said to contribute to the good spalling resistance of the alloy

titanium-In oxidation tests in air and in an inert atmosphere at 1260 °C (2300 °F) for MA 754 (Ni-20Cr-05Ti-0.5Y2O30.05C), Ni-20Cr (cast/wrought), and an ODS nickel-chromium alloy, subsurface porosity was attributed to the oxidation

-0.3Al-of chromium and aluminum (Kirkendall porosity), and thermally induced porosity was excluded as a cause (Ref 44) This type of porosity decreases with improving oxidation resistance of the alloy

Results of cyclic oxidation tests at 1100 °C (2010 °F) for MA 956, TD-NiCr, and Hastelloy X are given in Table 12 The superior resistance of MA 956 attributed to a very stable Al2O3 film and parabolic oxidation for over 500 h Tables 13 and

14 list sulfidation and carburization resistance data for MA 956 As in the case of oxidation, the alloy shows marked superiority to the other alloys tested Tables 15 and 16 provide similar data for alloy MA 6000E (Ni-15Cr-4.5Al-4W-2Mo-2Ta-2.5Ti-1.1Y2O3) (Ref 45) The functions of the various alloying elements are as follows:

• Aluminum, titanium, and tantalum for ' hardening

• Y2O3 for high-temperature strength and stability

• Aluminum and chromium for oxidation resistance

• Titanium, tantalum, chromium, and tungsten for sulfidation resistance

• Tungsten and molybdenum for solid-solution strengthening

Table 12 Cyclic oxidation resistance of superalloys at 1100 °C (2010 °F)

504-h test in atmosphere of air containing 5% H2O; temperature cycled between 1100 °C (2010 °F) and room temperature every 24 h

Table 13 Sulfidation resistance of superalloys at 925 °C (1700 °F)

312-h test in burner rig with air/fuel ratio of 30:1; fuel contained 0.3% S and 5 ppm seawater Temperature cycle: 58 min at temperature, followed by 2 min cool to room temperature

Table 14 Carburization resistance of superalloys at 1095 °C (2000 °F)

100-h test in atmosphere of hydrogen containing 2% methane

Source: Ref 45

Table 15 Sulfidation resistance of superalloys at 925 °C (1700 °F)

Tested in burner rig with air/fuel ratio of 30:1; fuel contained 0.3% S and 5 ppm seawater Temperature cycle: 58 min at temperature, followed by 2 min cool to room temperature

Alloy Exposure

time, h

Descaled weight loss, mg/cm2

Table 16 Cyclic oxidation resistance of superalloys at 1100 °C (2010 °F)

504-h test in air containing 5% H2O; temperature cycled from 1100 °C (2010 °F) to room temperature every 24 h

Alloy Descaled weight change, mg/cm2

sulfur-of rapid degradation The difference, compared to oxidation, is that the conditions causing hot corrosion simply shorten the time in which superalloys form protective Al2O3 or Cr2O3 scales by selective oxidation (Ref 38) Factors affecting the length of the initiation stage (at the end of the initiation stage, the superalloy must be removed from service because of the start of excessive corrosion) include alloy composition, alloy fabrication conditions, gas composition and velocity, deposit composition and its physical state, amount of deposit, temperature, temperature cycles, erosion, and specimen geometry

When a protective scale dissolves into a liquid deposit, so-called fluxing reactions can occur with the appearance of other basic or acidic nonprotective reaction products Propagation may also be caused by components from the deposit that can accumulate in the deposit or the alloy and thus cause a nonprotective scale to form Chlorine and sulfur produce such effects, and hot corrosion caused by the latter is known as sulfidation (see the section "High-Temperature

Oxidation/Sulfidation" in the article "General Corrosion" in this Volume) Figure 37 shows the temperature ranges over which the various hot corrosion propagation modes are important Additional information on the effects of individual elements on corrosion resistance is available in Ref 38 and in the articles "Corrosion of Nickel-Base Alloys" and

"Corrosion of Cobalt-Base Alloys" in this Volume

Fig 37 Schematic showing the temperature regimes over which different propagation modes are most

prevalent

Some superalloys corrode in several modes For example, hot corrosion of IN-738 proceeds by alloy-induced acidic fluxing, but is preceded by other propagation modes, including a basic fluxing mode The higher chromium content alloys IN-738 and IN-939 were developed to improve the hot corrosion resistance of land-based gas turbines Carbide stabilization through tungsten and tantalum and delay of M23C6 formation in service were expected to allow the large chromium content to impart improved hot corrosion resistance Increasing the chromium and decreasing the Al2O3, however, lowered ' solution temperatures and strength, which necessitated the use of coatings The use of coatings led

to the current use of enhanced aluminum, that is, carefully balanced coating alloys (based on nickel), iron, or cobalt with chromium, aluminum, and other active elements) Generally, all superalloy load-bearing parts used at very high temperatures under dynamic conditions are coated (Ref 39) Nevertheless, coatings generally last longer on more corrosion-resistant base materials

In a model study, IN-738 was used to demonstrate the effect of grain size and Y2O3 dispersions on hot corrosion behavior (Ref 46) Under gas turbine simulated hot gas corrosion test conditions at 850 and 950 °C (1560 and 1740 °F) (Fig 38), the presence of a Y2O3 dispersion lowered the corrosion rate At 950 °C (1740 °F) a finer grain size further reduced the corrosion rate, which was thought to be mainly due to a higher diffusion rate of chromium and aluminum The effect of

the dispersion was predominant at 850 °C (1560 °F) Reduced sulfate formation at 850 °C (1560 °F) was attributed to the likely formation of yttrium oxysulfide

Fig 38 Corrosion resistance of alloy IN-738LC in hot (850 °C, or 1560 °F) gases A, IN-738LC; B, IN-738LC

with Y2O3 dispersion, annealed at 1270 °C (2320 °F); C, IN-738LC with Y2O3 dispersion, annealed at 1100 °C (2010 °F) Source: Ref 34

In a study of the oxidation and hot corrosion resistance of P/M LC Astroloy and IN-100, isostatically pressed samples were found to be moderately attacked in a sulfate-chloride environment and heavily corroded by pure sodium sulfate (Na2SO4) (Ref 47) Heat treatment and the use of coarse powder (62 to 150 m for Astroloy and 88 to 200 m for IN-100) lowered the susceptibility to catastrophic corrosion Additions of yttrium to IN-100 improved the corrosion resistance in pure sulfate, but were detrimental when NaCl was present Therefore, yttrium additions to IN-100 cannot be recommended for marine turbines It was concluded that in many cases impregnation coatings must be considered for components made of IN-100 alloys

As a part of an evaluation of improved alloys for use in oil and gas drilling a depths of 6100 m (20,000 ft) HIP base alloy Inconel 625 was studied in a simulated deep, hot, sour cell environment (Ref 48) The alloy demonstrated resistance to pitting and crevice corrosion, sulfide stress cracking, chloride stress-corrosion cracking (SCC), and elevated-temperature anodic stress cracking Hot isostatic pressed Inconel 625 exhibit essentially the same corrosion resistance as wrought Inconel 625

nickel-Fatigue and Creep Crack Growth. Fatigue crack growth rates of nickel-base superalloys measured at frequencies above 0.1 Hz, at intermediate temperatures, and at an intermediate stress intensity range, K, were found to be several

times higher than those measured in inert atmospheres (Ref 49) The buildup of corrosion products with decreasing K,

however, was thought to enhance crack closure, thus reducing the effective stress intensity range and leading to fatigue thresholds higher than those in inert environments

Table 17 shows creep crack growth rates of IN-750 with various grain-boundary carbide microstructures (Ref 40, 49) Aggressive environments (helium + 3% sulfur dioxide, SO2, and air) produce order of magnitude increases over the rates

in inert gas In general, the reaction of both oxide dispersoid-free P/M superalloys and cast and wrought superalloys to aggressive environments is similar This suggests that crack growth is governed mainly by microstructure and alloy chemistry

Table 17 Dependence on carbide microstructure of creep crack growth rates of alloy IN-750 in four environments

Crack growth rate (da/dt), mm/min(a)

Fig 39 Comparison of the corrosion resistance of MA 6000 and MA 956 with that of other superalloys Tested

in a burner rig for 312 h using a 30:1 air to fuel ratio Fuel contained 0.3% S and 5 ppm seawater, and specimens were held at temperature for 58 min of each hour, then cooled 2 min with an air blast Source: Ref

40

Fig 40 Corrosion rate (a) and temperature capability (b) of MA 6000, MA 754, and non-ODS superalloys as a

function of chromium content A stress of 200 MPa (29 ksi) was applied to the specimens during the 10,000-h test Source: Ref 40

Fig 41 Temperature capability as a function of corrosion rate for various superalloys Same data as in Fig 40

Source: Ref 40

Fig 42 Hot corrosion of alloys MA 953, HDA 8077, and MA 956 compared to that of some non-ODS alloys Test

conditions: 900 °C (1650 °F), 1 h, followed by a 3-min air blast, 5 ppm sea salt Source: Ref 40

Coatings for ODS alloys. As mentioned above, for extended high-temperature service, superalloys require additional protection through coatings The use of aluminide coatings appears to be unsatisfactory due to the development of subsurface Kirkendall porosity and early spalling of the protective scale Kirkendall porosity decreases with increasing aluminum content of the substrate alloy as well as with decreasing grain size (Ref 50) Only limited information exists on the properties of chromium-aluminum-yttrium coatings (Ref 40) and on diffusion barrier coatings (Ref 51, 52, 53, 54)

P/M Aluminum Alloys

Although low- to medium-strength sintered (porous) P/M aluminum alloy parts were reported over 40 years ago and then commercialized in recent years, interest in the corrosion properties of P/M aluminum alloys paralleled the development of fully dense P/M aluminum products, which dates back some 25 years Early work showed that heat-treated extrusions of alloy powders higher in zinc, magnesium, and copper than the conventional, 7000-series ingot metallurgy (I/M) wrought alloys provided higher tensile strength (Ref 55) In addition, it was found that high strength and resistance to SCC superior to that of I/M alloys could be obtained (Ref 56) Subsequently, rapid solidification processing and mechanical alloying were further exploited with additions of iron, nickel, cobalt, and oxides for grain refinement and stabilization of the structure without the deleterious segregation effects that occur when I/M alloys are overalloyed (Ref 57)

The large, active constituents often found in conventional alloys are absent in these P/M alloys Also absent are impurities and grain-boundary depletion of alloying elements, both of which can cause localized attack Elements that are insoluble

in the solid state are often soluble in the liquid state and may be uniformly dispersed in the powder particles by rapid quenching; metastable phases may also be formed (Ref 58) Figure 43 illustrates the differences in appearance between P/M and I/M microstructures Additional information on the development of high-strength aluminum P/M alloys is available in Ref 59, 60, 61, 62, 63, 64

Fig 43 Comparison of grain structure in P/M and I/M aluminum alloys (a) P/M alloy X7090 (b) I/M alloy 7178

Both 500×

Because of higher solute contents, the P/M aluminum alloys can be aged to considerably higher yield strength than conventional alloys, with accompanying improvements in fatigue and toughness and without loss in SCC resistance Rapid soldification processing and mechanical alloying also permit increased alloying contents For example, increased amounts of lithium are possible (supersaturation), which results in lower density (7 to 20% weight savings) and higher specific elastic modulus

Uses, Market, and State of Commercialization

Depending on alloy type, the attractiveness of these alloys in comparison to conventional alloys lies in their superior room-temperature strength combined with excellent corrosion and stress-corrosion resistance, their improved elevated-temperature properties, and their lower density and higher elastic modulus High-performance P/M aluminum alloys are expected to find increasing use in aerospace, military, and marine applications and to replace heavier and more costly titanium alloys for example, in major airframe primary load-carrying structural members, such as upper wing skins and landing gear components; helicopter rotors; low-temperature fan and compressor cases, vanes, and blades for gas-turbine engines; and fins, winglets, and rocket motor cases in missiles

The addition of silicon carbide (SiC) to various aluminum alloy matrices by P/M or I/M processing results in lightweight, high-modulus composites Potential uses include antennae yolks, torpedo hulls, mobile bridges, gyroscope supports, tractor tread shoes, and helicopter landing skids (Ref 60) The limited ductility and high price of these composite alloys, however, are impeding commercialization for use in sports equipment

According to a recent market assessment, the targets of the high-performance P/M aluminum alloys are the high-strength

2xxx and 7xxx alloys (Ref 65) About 50,000 short tons (45,000 Mg), out of a total of 100,000 to 200,000 short tons

(90,000 to 180,000 Mg) per year, are used by the United States aerospace industry If the functional price can be made comparable to that of conventional alloys, a large potential also exists in the automotive market

Of the many P/M alloys under development, only the aluminum-zinc-magnesium-cobalt alloys X7090 and X7091, which were introduced in 1981, are being produced commercially Alloy X7090 is expected to be used for the landing gear support beam and the landing gear door actuator for the Boeing 757 aircraft According to Ref 65, the following barriers must be overcome to achieve market introduction:

• Lack of exposure and experience

• Inadequate industry standards and the lack of reliable, repeatable, nondestructive tests for P/M parts

• The rate of technology change, which can inhibit major capital investment

The manufacture of high-performance fully dense P/M aluminum alloys is based on the use of either rapidly solidified materials, powders and particulates or mechanically attrited powders Detailed descriptions of how these powders are

produced are available in Ref 66 and the articles "Atomization" and "Milling of Brittle and Ductile Materials" in Powder Metal Technologies and Applications, Volume 7 of the ASM Handbook Both methods can be used to produce

compositions that are not practical with I/M and that have very fine and uniform dispersions of intermetallic particles

The powders are usually degassed and then consolidated through the application of heat and pressure, that is, by forging, extrusion, hot pressing, or hot rolling, into a billet or near net shape part Additional information on the critical steps of degassing and consolidation of these alloys is available in Ref 59

Classes of High-Performance Aluminum P/M Alloys

As noted above, from an applications point of view, it is convenient to distinguish among three classes of performance P/M aluminum alloys (Ref 59, 60, 62, 63): high room-temperature strength, SCC/corrosion-resistant alloys; low-density high-stiffness alloys; and elevated-temperature alloys

high-High Room-Temperature Strength SCC/Corrosion-Resistant Alloys. Table 18 shows the composition of six P/M aluminum alloys characterized by high room-temperature strength and high SCC resistance Their room-temperature

properties are given in Tables 19 and 20 Alloy CW67 (Table 20), a recent second-generation P/M product in the 7xxx

series, is a zirconium- and nickel-containing aluminum-zinc-magnesium-copper alloy It is still in the laboratory evaluation stage Strength increases of 20 to 30% over wrought aluminum alloy 7075 translate into weight savings of 10%

or more in aerospace structures In the damage-tolerant T7 temper, CW67 demonstrates superior combinations of strength, fracture toughness, and resistance to fatigue and SCC compared to other I/M and P/M alloys (Ref 67, 68) Of the alloys shown in Table 18, alloys X7090 and X7091 are farthest along in the development of P/M high-strength aluminum alloys In these alloys, cobalt is present as an intermetallic dispersoid of the composition Co2Al9 (in iron- and nickel-containing alloys, the analogous dispersoid is (Fe,Ni)2Al9) and is beneficial to both strength and SCC resistance Figure

44 shows a transmission electron micrograph of a P/M MA 67 (the experimental precursor of alloy X7090) heat-treated die forging showing the uniformly distributed and small (0.05- m) Co2Al9 particles as dark spheroids present on grain

boundaries and within grains In the T7x tempers, there were generally no precipitate-free zones surrounding the Co2Al9particles The magnesium- and aluminum-base oxide particles, another distinctive P/M feature, appear as clusters of dark particles, lighter and much finer than Co2Al9 (0.02 to 0.004 m) and predominantly at grain boundaries

Table 18 Nominal compositions of high strength corrosion-resistant P/M aluminum alloys

(a) For selected applications only

Table 20 Typical room-temperature properties of I/M and P/M high strength aluminum alloys

SCC

resistance

Ultimate tensile

strength

Yield strength(b)

Fracture toughness (c) Alloy

MPa ksi

Exfoliation rating(a)

MPa ksi MPa ksi

Fig 44 Structure of P/M MA 67 die forging that was solution heat treated 2 h at 495 °C (920 °F), water

quenched, and aged 7 days at room temperature plus 24 h at 120 °C (250 °F), plus 12 h at 165 °C (325 °F) Transverse section showing Co2Al9 (arrow C), oxide cluster (arrow O), and a precipitate-free zone (arrow PFZ- GB) at a grain boundary 18,000× Source: Ref 58

In 1975, it was proposed that the superior SCC resistance of P/M aluminum-zinc-magnesium-copper alloys containing iron and nickel or cobalt may be related to the following factors: grain morphology (SCC fracture is intergranular and the crack path is tortuous because of the presence of fine dispersoids); grain-boundary movement restriction from the dispersoids (including oxides); hydrogen recombination at the cathodic Co2Al9 dispersoids; blunting of the leading edge

of cracks by dispersoid particles; and shifting of the chemistry within cracks to more alkaline conditions at grain boundaries containing corroding oxide particles (Ref 58) The effect of different cobalt concentrations on the SCC behavior of several aluminum powder alloys was recently investigated in more detail (Ref 69, 70, 71) It was concluded that the increased SCC resistance was not due to the smaller grain size but to a direct effect of cobalt attributable to the cathodic behavior of the Co2Al9 particles that serve as sites for hydrogen recombination, thus reducing both the absorption

of atomic hydrogen into the grain boundaries and hydrogen embrittlement

Based on precracked double-cantilever beams in chromate inhibited brine, alloys X7090 and X7091 exhibited the greatest susceptibility to SCC in their undertempers (Ref 72) Susceptibility decreased in the peak-aged tempers and was absent in the overaged tempers after 500 h of immersion In comparison, conventional alloy 7075, which uses chromium and manganese as dispersoid-forming elements, has a lower strength in its peak-aged temper and exhibits similar SCC propagation behavior to X7091 under the same test conditions

Figure 45 shows time to failure curves for unnotched (reflecting both initiation and propagation behavior) specimens of alloys 7075 and X7091 in acetic acid brine For a given strength, the P/M alloys exhibit superior SCC resistance An investigation of the effect of varying amounts of dispersoid contents of Co2Al9 and (Fe,Ni)2Al9 found that the SCC resistances of Co2Al9 and (Fe,Ni)2Al9 dispersoid-containing alloys are similar in their T7 tempers (Ref 73) The influence

of dispersoids on SCC was evident in the T6 but not the T7 tempers Electrochemical tests showed a tendency for high