Copyright 2008 ASM International. All rights reserved.

Stainless Steels for Design Engineers (#05231G) www.asminternational.org

CHAPTER 6

Austenitic Stainless Steels

Summary because their greater thermal expansion co-

efficient tends to cause the protective oxide
AUSTENITIC STAINLESS STEELS are the coating to spall.
most common and familiar types of stainless 2. They can experience stress corrosion crack-
steel. They are most easily recognized as non- ing (SCC) if used in an environment to which
magnetic. They are extremely formable and they have insufficient corrosion resistance.
weldable, and they can be successfully used 3. The fatigue endurance limit is only about
from cryogenic temperatures to the red-hot tem- 30% of the tensile strength (vs. ~50 to 60%
peratures of furnaces and jet engines. They con- for ferritic stainless steels). This, combined
tain between about 16 and 25% chromium, and with their high thermal expansion coeffi-
they can also contain nitrogen in solution, both cients, makes them especially susceptible to
of which contribute to their high corrosion re- thermal fatigue.
sistance. Were it not for the cost of the nickel However, the risks of these limitations can be
that helps stabilize their austenitic structure, avoidable by taking proper precautions.
these alloys would be used even more widely.

Introduction Alloy Families in Perspective

Austenitic stainless steels have many advan- The fundamental criterion in the selection of
tages from a metallurgical point of view. They a stainless steel is generally that it can survive
can be made soft enough (i.e., with a yield with virtually no corrosion in the environment
strength about 200 MPa) to be easily formed by in which it is to be used. Good engineering
the same tools that work with carbon steel, but practice sometimes requires that materials be
they can also be made incredibly strong by cold selected for sufficient, but finite, service life.
work, up to yield strengths of over 2000 MPa This is especially true for high-temperature
(290 ksi). Their austenitic (fcc, face-centered service, for which creep and oxidation lead to
cubic) structure is very tough and ductile down limited life for all materials. The choice among
to absolute zero. They also do not lose their the stainless steels that can be used in that envi-
strength at elevated temperatures as rapidly as ronment is then based on the alloy from which
ferritic (bcc, body-centered cubic) iron base al- the component can be produced at the lowest
loys. The least corrosion-resistant versions can cost, including maintenance, over the intended
withstand the normal corrosive attack of the service life. The ferritic stainless steels are less
everyday environment that people experience, expensive for the same corrosion resistance but
while the most corrosion-resistant grades can sometimes are found lacking because of:
even withstand boiling seawater.
Lack of toughness, as is the case at subambi-
If these alloys were to have any relative
weaknesses, they would be: ent temperatures or in thicknesses greater
than about 1.5 mm
1. Austenitic stainless steels are less resistant Lack of great ductility, specifically if more
to cyclic oxidation than are ferritic grades than about 30% elongation is needed
70 / Stainless Steels for Design Engineers

Susceptibility to high-temperature embrit- The austenitic alloys can have compositions

tling phases when moderately alloyed anywhere in the portion of the Delong diagram
labeled austenite shown in Fig. 1 (Ref 1). This
The less-expensive martensitic grades are diagram was designed to show which phases are
used instead of austenitic when high strength present in alloys in the as-solidified condition,
and hardness are better achieved by heat treat- such as found in welds. Thus it also applies to
ing rather than by cold work, and mechanical castings and continuously cast products. As a
properties are more important than corrosion re- practical matter of castability, the composition
sistance. This is also the case for the more ex- of most commercial alloys falls along the zone
pensive PH grades, which can achieve corrosion of several percent ferrite as cast. The salient fea-
resistance only matching the least corrosion re- ture of austenitic alloys is that as chromium and
sistant of the austenitic alloys. molybdenum are increased to increase specific
Duplex grades match austenitic grades in cor- properties, usually corrosion resistance, nickel
rosion resistance and have higher strength in the or other austenite stabilizers must be added if
annealed condition but present the designer the austenitic structure is to be preserved.
with challenges with regard to embrittling The traditional way of displaying the
phases that can form with prolonged exposure austenitic stainless steels is to present 302 as a
to elevated temperatures and only moderate base. Figure 2 shows one such diagram. Dia-
ductility like the ferritic alloys. grams such as these treat alloys as an evolution-
So, the austenitic grades are the most com- ary family tree and subtly mislead. Many alloys
monly used grades of stainless mainly because, were pushed toward obsolescence because of
in many instances, they provide very predictable advances in processing. For instance, 321 was
levels of corrosion resistance with excellent me- developed as an alloy in which the detrimental
chanical properties. Using them wisely can save effects of carbon were negated by addition of ti-
the design engineer significant costs in his or her tanium. The widespread adoption of the argon
product. They are a user-friendly metal alloy oxygen decarburization (AOD) in the 1970s
with life-cycle cost of fully manufactured prod- made this alloy unnecessary, except for special
ucts lower than many other materials. circumstances, since carbon could be cheaply

Fig. 1 Schaeffler-Delong stainless steels constitution diagram. Adapted from Ref 1, 2

 Chapter 6: Austenitic Stainless Steels / 71

Fig. 2 The austenitic stainless family. Source: Ref 3

removed routinely. Likewise, 302 gave way to nickel equivalents (manganese, nitrogen, carbon,
the lower-carbon 304, for which the even lower- etc.) must also be added in matching amounts,
carbon 304L is commonly substituted and du- austenite stability is also increased. If molybde-
ally certified to qualify as either grade. While num, a chromium equivalent, is added, corro-
low carbon prevents sensitization, stabilized sion resistance but not oxidation resistance is
grades may still be preferred for special applica- enhanced. And, if nitrogen is the austenite stabi-
tions such as type 321 in aerospace and type lizer added to balance increases chromium or
347 in refinery service. Similar inertia keeps the molybdenum, then corrosion resistance is also
higher-nickel 300 series as the de facto standard increased. With small exceptions, that is the
when the more cost-efficient high-manganese rationale of austenitic grade design. Silicon
200 series is the logical basic grade. The rele- is used as an alloy to promote oxidation resist-
vant types of austenitic alloys can nonetheless ance and resistance to corrosion by oxidizing
be rationalized with this diagram. acids. Copper is used to promote resistance to
As chromium is added, oxidation resistance sulfuric acid. Rare earths make a more stably
and corrosion resistance increase. Because oxidation-resisting scale. Niobium increases
72 / Stainless Steels for Design Engineers

creep resistance. Sulfur and selenium increase Lean Alloys

machinability.
In this chapter, austenitic alloys are classified Lean austenitic alloys constitute the largest
into three groups: portion of all stainless steel produced. These are
Lean alloys, such as 201 and 301, are gener- principally 201, 301, and 304. Alloys with less
ally used when high strength or high forma- than 20% chromium and 14% nickel fall into
bility is the main objective since the lower, this unofficial category. Since they are stainless,
yet tailorable, austenite stability of these al- it is generally taken for granted that these alloys
loys gives a great range of work-hardening will not corrode, and these alloys have suffi-
rates and great ductility. Richer alloys, such cient corrosion resistance to be used in any in-
as 305, with minimal work hardening are the door or outdoor environment, excluding coastal.
high-alloy, lowest work-hardening rate These grades are easily weldable and formable
grade for this purpose. The general-purpose and can be given many attractive and useful sur-
alloy 304 is within this group. face finishes, so they are very much general-
Chromium nickel alloys when the objective purpose alloys. Table 1 lists some typical com-
is high temperature oxidation resistance. positions of the most commonly used lean
This can be enhanced by silicon and rare austenitic alloys. These typical compositions
earths. If the application requires high-tem- vary with end use, raw material cost factors, and
perature strength, carbon, nitrogen, niobium, the preference of a given manufacturer. The
and molybdenum can be added. 302B, 309, compositions of standard alloys are often fine-
310, 347, and various proprietary alloys are tuned to the intended end use. In this table, the
found in this group. word drawing indicates higher nickel for lower
Chromium, molybdenum, nickel, and nitro- work hardening, while tubing indicates alloys
gen alloys when corrosion resistance is the with higher sulfur to facilitate gas tungsten arc
main objective. Alloys such as silicon and welding (GTAW) penetration. Tensile indicates
copper are added for resistance to specific lower alloy levels to increase the work-harden-
environments. This group includes 316L, ing rate for material that is intended to be used
317L, 904L, and many proprietary grades. in the cold-worked, high-strength condition.
316L is included in its most common tubing end
Wrought alloys generally have cast counter- use chemistry even though it is a corrosion-re-
parts that differ primarily in silicon content. sisting alloy because it is so pervasively used as
Versions that require enhanced machinability a service center sheet item.
have a high content of controlled inclusions, The main difference among the lean
sulfides, or oxysulfides, which improve machin- austenitic alloys lies in their work-hardening
ability at the expense of corrosion resistance. rate: the leaner the alloy, the lower the austenite
Carbon is kept below 0.03% and designated an stability. As unstable alloys are deformed, they
L grade when prolonged heating due to multi- transform from austenite to the much harder
pass welding of heavy section (greater than martensite. This increases the work-hardening
about 2 mm) or when welds requiring a post- rate and enhances ductility since it delays the
weld stress relief are anticipated. onset of necking since greater localized

Table 1 Typical compositions of the most commonly used lean austenitic alloys
Alloy Designation C N Cr Ni Mo Mn Si Other Other Other
201 S20100 0.08 0.07 16.3 4.5 0.2 7.1 0.45 0.001 S 0.03 P 0.2 Cu
201 drawing S220100 0.08 0.07 16.9 5.4 0.02 7.1 0.5 0.001 S 0.30 P 0.6 Cu
201LN S20153 0.02 0.13 16.3 4.5 0.2 7.1 0.45 0.001 S 0.03 P 0.5 Cu
301 tensile S30100 0.08 0.4 16.6 6.8 0.2 1.0 0.45 0.001 S 0.03 P 0.3 Cu
301 drawing S30100 0.08 0.04 17.4 7.4 0.02 1.7 0.45 0.007 S 0.03 P 0.6 Cu
303 S30300 ... ... ... ... ... ... ... ... ... ...
304 S30400 0.05 0.05 18.3 8.1 0.3 1.8 0.45 0.001 S 0.03 P 0.3 Cu
304 drawing S30400 0.05 0.04 18.4 8.6 0.3 1.8 0.45 0.001 S 0.03 P 0.3 Cu
304 extra drawing S30400 0.06 0.04 18.3 9.1 0.3 1.8 0.45 0.001 S 0.030 P 0.4 Cu
304L tubing S30403 0.02 0.09 18.3 8.1 0.3 1.8 0.45 0.013 S 0.030 P 0.4 Ci
305 S30500 0.05 0.02 18.8 12.1 0.2 0.8 0.60 0.001 S 0.02 P 0.2 Cu
321 S32100 0.05 0.01 17.7 9.1 0.03 1.0 0.45 0.001 S 0.03 P 0.4 Ti
316L S31603 0.02 0.0 16.4 10.5 2.1 1.8 0.50 0.010 S 0.03 P 0.4 Cu
 Chapter 6: Austenitic Stainless Steels / 73

deformation is more than offset by greater lo- such as occurs when they are sensitized or when
calized strain hardening. solute segregation occurs, as from welding, then
These grades are best viewed as a continuum the equation applies on a microscopic scale.
with a lower boundary at 16%Cr-6%Ni and an Sensitized zones (i.e., the regions near grain
upper boundary at 19%Cr-12%Ni. This repre- boundaries where chromium carbides have pre-
sents the range from minimum to maximum cipitated) will have a much higher tendency to
austenite stability. Since that is the main distinc- transform to martensite. Figures 3(a) and (b)
tion within this grade family, let us examine its show the changes in phase structure as a func-
basis. tion of composition over ranges that encompass
Martensite and Austenite. Stability. The these alloys.
formation of martensite at room temperature Martensite can be present in two different
may be thermodynamically possible, but the forms. The -form is the bcc magnetic form,
driving force for its formation may be insuffi- while is a nonmagnetic, hcp (hexagonal close-
cient for it to form spontaneously. However, packed) version. The formation of versus is
since martensite forms from unstable austenite related to the stacking fault energy of the alloy,
by a diffusionless shear mechanism, it can occur which is given by (Ref 6):
if that shear is provided mechanically by exter-
nal forces. This happens during deformation, Y300SF (mJ m-2) = Y0SF + 1.59Ni 1.34Mn
and the degree to which it occurs varies with + 0.06Mn2 1.75Cr + 0.01Cr2
composition according to (Ref 4): + 15.21Mo 5.59Si
Md30 (C) = 551 462(%C + %N) 60.69(C + 1.2N)1/2
9.2(%Si) 8.1(%Mn) 13.7(%Cr) + 26.27(C + 1.2N)
29(%Ni + Cu) 18.5(%Mo) (Cr + Mn + Mo)1/2
68(%Nb) 1.42 (GS 8) (Eq 1) + 0.61[Ni(Cr + Mn)]1/2 (Eq 2)

This is the temperature at which 50% of the Epsilon martensite formation is favored in
austenite transforms to martensite with 30% alloys of lower stacking fault energy. The fcc
true strain (Ref 5). It should be noted that even structures deform by slip between (111)
elements that are chromium equivalents in pro- planes. Viewed from these planes, the structure
moting ferrite are austenite stabilizers in that is a series of ABCABC atom arrangements.
they impede martensite formation. This temper- Slip between planes can result in an
ature is the common index of austenite stability. ABCA/CAB structure. This so-called stacking
This regression analysis was generated for ho- fault generates an hcp structure. With lower
mogeneous alloys. If alloys are inhomogeneous, stacking fault energies, these are more readily

Fig. 3 (a) Iron-chromium phase diagram at 8% nickel; (b) iron-nickel phase diagram at 18% chromium
74 / Stainless Steels for Design Engineers

Fig. 4 Variation of martensite formation with temperature and true strain for 304. Source: Ref 7

formed, and predominates. The stacking fault and tensile strength, respectively, are reported
can also be viewed as two partial dislocations (Ref 10) to follow the equations:
with the material between them faulted. These
partial dislocations, when generated in abun- YS( MPa) = 15.4[4.4 + 23(%C) + 32(%N)
dance, cannot readily slip past one another + 0.24(%Cr ) + 0.94(%Mo)
and thus pile up, increasing work-hardening + 1.3(%Si ) + 1.2(%V)
rates. + 0.29(%W ) + 2.6(%Nb)
As in carbon and alloy steels, the martensite
+ 1.7(%Ti ) + 0.82(%Al )
transformation can take place simply by cool-
ing, but in the lean austenitic alloys the temper- + 0.16(%Ferrite )
atures are well below ambient. The more stable + 0.46(d 1/1/ 2 )
(Eq 3)
alloys do not transform even with cryogenic
treatment. Figure 4 shows the variation of TS ( MPa) = 15.4[29 + 35(%C) + 55(%N)
martensite formation with temperature and true
+ 2.4(%Si ) + 0.11((%Ni ) + 1.2(%Mo)
strain for 304. Martensite formed in these alloys
is quite stable and does not revert until heated + 5.0(%Nb) + 3.0(%Ti ) + 1.2(%Al )
well above the temperatures (Fig. 5) at which it + 0.14(%Ferrite ) + 0.82(d 1/ 2 ) (Eq 4)
was formed. The carbon levels of austenitic
stainless steels are always relatively low, so In each case, d is the grain diameter in mil-
strain-induced martensite is self-tempering and limeters.
not brittle. Another researcher (Ref 11) gave the rela-
Martensite has been found to form in unstable tionships as:
austenite due to the electrochemically induced
supersaturation by hydrogen (Ref 9). Under YS ( MPa ) = 120 + 210 N + 0.02 + 2 Mn + 2Cr
conditions of cathodic charging, superficial lay- + 14 Mo + 10 Cu + (6.15 0.054)
ers were found to transform to under condi- + (7 + 35(N + 0.2))d 1/ 2 (Eq 5)
tions of intense hydrostatic compression. Dur-
ing subsequent outgassing, was found to
form due to reversals in the stress state. Marten- TS = 470 + 600( N + 0.2)
site thus formed is, of course, susceptible to + 14 Mo + 1.5 + 8d 1/ 2 (Eq 6)
hydrogen embrittlement.
Mechanical Properties. The tensile proper- Again, d is grain diameter in millimeters, and
ties in the annealed state not surprisingly relate is percent ferrite. The claimed accuracy for
well to composition. The 0.2% yield strength the latter set of equations is 20 MPa and is said
 Chapter 6: Austenitic Stainless Steels / 75

Fig. 6 Variation of impact strength with temperature for (a)

austenitic, (b) duplex, and (c) ferritic stainless steels

form more stable dislocation arrays that break

loose at a higher and distinct yield point.
The tensile properties of austenitic stainless
steels with unstable austenite, that is, those with
Md30 temperatures (Eq 1) near room tempera-
ture, are very strain rate dependent. This is sim-
Fig. 5 Reversion of martensite formed by cold work. Source:
ply due to the influence of adiabatic heating
Ref 8
during testing increasing the stability of the
to apply to both austenitics and duplex stainless austenite. Tests run under constant temperature
steels, but clearly the tensile strength relation- conditions, either by slow strain rates or use of
ship must break down for leaner alloys, such as heat sinks, produce lower tensile strengths.
301, in which tensile strength increases with de- Thus, reported tensile strengths should not be
creasing alloy content because of the effect of taken as an absolute value but a result that can
increasing alloying causing less transformation be significantly changed by changes in testing
to martensite, which inarguably produces higher procedure, even with accepted norms and
tensile strengths in austenitic stainless steels. standards.
Equation 3 must also be favored over Eq 5 in Highly cold-worked austenitic stainless
that it accounts for carbon explicitly. steels are often used for their robust mechanical
One other hardening mechanism is possible properties. Few metallic materials can match
in austenitic stainless steels, and that is precipi- the very high strengths they can achieve. Very
tation hardening. Most precipitation-hardening lean 301 can be cold worked to yield strengths
stainless steels are unstable austenite, which is on the order of 2000 MPa because of its unsta-
transformed to martensite before the precipita- ble austenite transforming to martensite. When
tion hardening takes place. One commercial cold worked to lower degrees, it can provide
alloy, A-286, is entirely austenitic and employs very high strength while keeping impressive
the precipitation within the austenite matrix of ductility.
Ni3 (titanium, aluminum) for strengthening. Austenitic stainless steels have exceptional
This is dealt with in a separate section. toughness. The ambient temperature impact
Austenitic stainless steels do not have a clear strength of austenitic stainless steels is quite
yield point but can begin to deform at as little as high. This is not surprising in view of their high
40% of the yield strength. As a rule of thumb, tensile strengths and high elongations. What is
behavior at less than half the yield strength is most remarkable is the absence of a transition
considered fully elastic and stresses below two- temperature, which characterizes ferritic and
thirds of the yield strength produce negligible martensitic materials. Figure 6 shows impact
plastic deformation. This quasi-elastic behavior strength of the various stainless steel types ver-
is a consequence of the many active slip sys- sus temperature. This again is due to the multi-
tems in the fcc structure. Even highly cold- plicity of slip systems in the fcc structure and
worked material exhibits this phenomenon, al- the fact that they do not require thermal activa-
though stress-relieving cold-worked material tion. This makes the austenitic stainless steels,
will cause dislocations to lock in place and especially the 200 series, the optimal cryogenic
76 / Stainless Steels for Design Engineers

material, surpassing the 9% nickel martensitic tion in austenite, and diffusion rates are suffi-
steels in cost, toughness, and, of course, corro- cient for carbon and chromium to segregate into
sion resistance. precipitates. The solubility of carbon in austen-
Precipitation of Carbides and Nitrides. ite is over 0.4% at solidification but decreases
Carbon is normally considered as an undesir- greatly with decreasing temperature. The solu-
able impurity in austenitic stainless steel. While bility is given by (Ref 12):
it stabilizes the austenite structure, it has a great
thermodynamic affinity for chromium. Because 6272
log (C ppm ) = 7771 (Eq 7)
of this affinity, chromium carbides, M23C6, form T (K )
whenever carbon reaches levels of supersatura-
The equilibrium diagram for carbon in a basic
18%Cr10%Ni alloy is shown in Fig. 7. At room
temperature, very little carbon is soluble in
austenite; even the 0.03% of L grades is mostly
in a supersaturated solution. The absence of car-
bides in austenitic stainless is due to the slow
diffusion of carbon and the even slower diffu-
sion of chromium in austenite. At a carbon level
of 0.06%, which is found in most 304, supersat-
uration is reached below about 850 C. Below
this temperature, supersaturation increases ex-
ponentially, while diffusion decreases exponen-
tially. This results in precipitation rates that vary
with temperature and carbon level as shown in
Fig. 8. At these temperatures, grain boundary
diffusion is much more rapid than bulk diffu-
sion, and grain boundaries provide excellent nu-
cleation sites, so precipitation occurs along
grain boundaries. And, because carbon diffuses
several orders of magnitude more rapidly than
chromium, carbon diffuses to and combines
Fig. 7 Carbon solubility in 1810 austenitic stainless. Source: with chromium essentially in situ, depleting the
Ref 13 grain boundaries of chromium in solution.

Fig. 8 The precipitation rates for Cr23C6 as a function of carbon content

 Chapter 6: Austenitic Stainless Steels / 77

Fig. 10 Variation of carbide precipitation locus with time.

Source: Ref 16

Much longer term heat treatment is required to

eliminate these depleted zones by rehomoge-
nization of slowly diffusing chromium than the
short time required to form them. This is very
evident for carbides, but also true for oxides.
Underneath chromium-rich oxide scales is a
layer depleted in chromium and lower in corro-
sion resistance. This is why not only scale from
welding must be removed, but also the underly-
ing chromium-depleted zone.
Other precipitation processes that give rise to
chromium depletion are and and the solid-
state precipitation of oxides, nitrides, and sul-
fides. Chromium precipitates that form in the
liquid alloy do not cause depletion of chromium
Fig. 9 Depletion of chromium from the austenite near grain
locally because no chromium gradients are set
boundaries due to chromium carbide precipitation.
Source: Ref 14 up around them during precipitation as diffu-
sion in the liquid is very rapid. Thus, primary
Figure 9 shows that the local chromium de- carbides, oxides, and sulfides are not per se
pletion is such that the chromium level can be- harmful to corrosion resistance. But, if the same
come low enough that it has not even enough to compounds form and grow in the solid state,
be stainless and certainly much lower corrosion chromium depletion occurs (Ref 15).
resistance than the surrounding area. This zone, Alloying elements can have a major influence
because it is lower in chromium, also has very on carbide precipitation by their influence on the
unstable austenite and is quite prone to marten- solubility of carbon in austenite. Molybdenum
site formation. Figure 10 shows how the locus and nickel accelerate the precipitation by dimin-
of precipitation changes with time and tempera- ishing the solubility of carbon. Chromium and
ture. Carbon relatively far from grain bound- nitrogen increase the solubility of carbon and
aries in the interior of grains remains in super- thus retard and diminish precipitation. Nitrogen
saturation until much longer times and much is especially useful in this regard (Fig. 11).
greater supersaturation since bulk diffusion is Increasing austenite grain size accelerates
required for the nucleation and growth of these precipitation, as does cold work, especially in
precipitates. the interior of grains, where diffusion is en-
The key observation is that any solid-state hanced by increased defect density.
precipitation of a chromium-rich precipitate Nitrogen is much more soluble than carbon
will necessarily cause local chromium depletion and does not give rise to sensitization phenom-
and a resulting loss of corrosion resistance. ena as does carbon even though Cr2N can be a
78 / Stainless Steels for Design Engineers

Fig. 11 Delay in carbide precipitation induced by nitrogen

level. Source: Ref 17

stable phase when the solubility limit is ex-

ceeded. The solubility is over 0.15% in austen-
ite, so its precipitation seldom has the possibil-
ity of occurring, but it does become an issue in Fig. 12 Variation of hardness with depth and therefore car-
ferritic stainless steels in this regard, for which bon content in colossal supersaturation
solubility is much lower. Manganese and
chromium increase the solubility of nitrogen in to ignore the titanium-consuming capacity of
austenite. oxygen and sulfur unless they have been mini-
Stabilization. Before carbon was easily low- mized by refining, which can be done quite
ered to harmless levels, it was found that adding readily.
more powerful carbide formers than chromium Even if sufficient titanium or niobium is pres-
could preclude the precipitation of chromium ent to combine with all carbon, kinetic consid-
carbides. Titanium and niobium are the most erations may result in that not occurring. High
useful elements in this regard. They form car- temperatures, such as encountered in welding,
bides with solubility that follows the following dissociate carbides. If quenched from this state,
equation type: carbon can be free to form Cr23C6 if it is re-
heated to temperatures above 500 C.
log [ M] [ X] = + A H (Eq 8) Carbon has always been considered totally
RT undesirable from a corrosion point of view be-
cause of its tendency to form chromium car-
For titanium carbide and niobium carbide, the bides. Recently, however, new processes have
respective solubilities are: been developed to supersaturate carbon in
6780 austenite below the temperatures at which it has
log [Ti ] [C] = 2.97 (Eq 9) sufficient mobility to form carbides. This so-
T
called colossally supersaturated austenite re-
sults in very high hardness (Fig. 12) and corro-
9350 sion resistance over limited depths. From this,
log [ Nb] [C] = 4.55 (Eq 10)
T however, we can see that carbon, like nitrogen,
is actually beneficial to corrosion resistance in
Oxides and sulfides are more energetically fa- solid solution, although this is not observed at
vorable than are carbides and nitrides of these normal concentrations. It is possible to see that
metals. Thus, any additions made to form car- if it could be kept in solution it would be appro-
bides must be sufficient to account for the prior priate to give it a factor of around 10 in the pit-
formation of these compounds. Nitrogen also ting resistance equivalent number (PREN)
competes with carbon for available titanium or equation:
niobium. Thus, for successful gettering of all
carbon, there must be sufficient titanium or nio- PREN = %Cr + 3.3(%Mo)
bium to combine stoichiometrically with all
these species present. + 30(%N) + 10(%C) (Eq 11)
This requires in rough terms that titanium ex-
ceed four times the carbon plus nitrogen, or that This is consistent with the similar thermody-
niobium exceed eight times, after accounting namic interaction coefficients that carbon and
for the oxygen and sulfur. It would be a mistake nitrogen share with regard to chromium.