US2531155A - Stainless steel - Google Patents

Description

Nav. 2l, 1950 Filed Nov. 6. 1945 2 Sheets-Sheet 1 FEE..

om, oc o oo o x/ o x nc o0 o on Q o o o0 x o o o o o o o o v o o o u o o o YT T o 4` l\l\ o oo o o o0 o o MMMMMMMMMMWMM Z. ZZ Z7. 0. a@

F. J. PHILLIPS STAINLESS STEEL ov. 2L 1950 2 Sheets-Sheet 2 Filed Nov. 6. 1945 y W w ww Patented Nov. 21, 1950 anni STAINLESS STEEL Freeman J. Phillips, Mcleesport, Pa., assigner to Carnegie-Iliinois Steel Corporation, a corporation of New Jersey Application November 6, 1945, Serial No. 627,077

(Cl. 'i5-428) 3 Claims.

The present invention relates to austenitic stainless steels immune to intergranular corrosion and, particularly, to nickel-chromium steels of the class described capable of resisting intergranular corrosion attack under any given set of conditions.

Austenitic stainless steels--namely, steels comprising a, sufficiency of chromium for assuring corrosion resistance, an adequate nickel or manganese content for converting the system into substantially austenitic state at room temperature and increasing the mechanical strength of the alloy at elevated temperatures, and a percentage of carbon inherent to the manufacturing practice-enjoy an enviable reputation among structural materials of the present day and are employed for a multitude of uses calling for corrosion resistance under ambiant atmospheric conditions and for heat and corrosion resistance associated with a suflicient strength at elevated temperatures encountered in many industrial processes.

Industrial experience has indicated a particular suitability of steels of the class described to the second of the-above groups of applicationsy but has demonstrated, at the same time, the tendency of said steels to disintegrate, or to lose the major portion of their strength, after being subjected tothe action of temperature in the range of substantially 100G-1400" F. and then f to a corroding action of various reagents.

Comprehensive investigations, abundantly reported in the literature, have been undertaken and successfully concluded with the purpose of explaining and clarifying the phenomena underlying the mechanism of disintegration taking place in the above steels under aforementioned conditions; and it appears that the said disintegration accompanies preoipitation of chromium carbides at the grain boundaries. This conception is based on the observation that austenitic stainless steel, usually carrying 7 to 9 per cent nickel and 16 to 19 per cent chromium, reaches ultimate use in4 an unstable state. Since the maximum corrosion resistance is commonly connected with single-phase alloys, industrial practice stipulates bringing all constituents of stainless steels into a solid solution by heating in the neighborhood of 1900-2l00 F. and retaining said solid solution by a rapid cooling, thus preventing equilibrium conditions.

On heating in the 100G-1400o F. range, unstable austenitic alloys tend to reach a more stable state through precipitation, as separate phases, of elements held in super-saturation.

Carbon occupies a prominent place among such elements, and is the first to precipitate. Since elemental carbon is unstable, and quite active chemically, precipitation of molecular carbon is seldom observed. Carbon tends to combine with chromium prior to precipitation and separates from the system as chromium carbides, the exact composition thereof being still unknown. Physical metallurgy teaches that both carbon and compounds thereof tend to precipitate at the grain boundaries. Apparently carbon migrates to the boundaries as such, the high diffusibilty thereof in solid steel being well known, and combines with other elements, to form corresponding carbides in situ.

Chromium migration rate in solid steel is greatly less than the rate of migration of carbon atoms in ferrous alloys. During the time required to achieve sensitizing at G-1409" F., substantially the whole amount of carbon ultimately reaching grain boundaries is segregated therein. Migration of chromium in the same direction is, however, incapable of transferring atoms thereof to the grain boundaries at a rate at all commensurate to that of carbon, and the concentration of chromium remains substantially unchanged in comparison with the original content thereof in the metal. Since precipitation of carbon takes place through the intermediate phase of chromium-carbide formation, and as the said phase has a greater concentration of chromium than the average for the metal, the formation of chromium carbides at the grain boundaries depletes surrounding areas of chromium. f

Modern conception of corrosion-protection mechanism attributes a major place to protective oxide lms. forefront of substances suitable for forming the said protective films. An increased concentration thereof in the base metal is, furthermore, associated with a greater resistance to corrosion attack under given oxidizing conditions, and vice versa. An alloy capable of withstanding a given corrosive attack when carrying a speciiied percentage of chromium, frequently fails under identical corrosion conditions when chromium contained therein is reduced below appropriate minimum values. Impoverishment of grain boundaries in respect to chromium caused by the aforementioned chromium-carbide precipitation and the solution thereof under conditions incapable of affecting the original metal may be held as a classical example of the above observations.

All early realization of the picture dening Chromium is generally held in the intergranular corrosion of the type described led to devising alleged remedies for avoiding undesirable features associated therewith. Since carbon migration towards the grain boundaries cannot be avoided, nor the mobility of chromium atoms increased, the desired effect was sought to be obtained through stabilization of carbon within metallic grains through the formation of difcultly soluble stable carbides intended to be present as a second phase substantially insoluble in the main austenitic system. Titanium and columbium have been found to be particularly meritorious in connection with the aforementioned carbon stabilization, and addition thereof, in amounts empirically found to be optimum, was generally adopted by the steel industry.

Alloying with carbon-stabilizing elements recited hereinbefore did not, however, solve the problem of intergranular-corrosion elimination. The necessity of a number of widely used, carefully devised methods for detecting intergranular corrosion, exemplified by the well known Strauss acid-copper-sulphate test, to be applied to carefully compounded and properly made steels of austenitic stainless steel type, conclusively demonstrated insufficiency of the said additions as a universal cure for intercrystalline corrosion. Fully austenitic steels containing titanium and columbium in respective ratios of 6 and 10 times to the carbon content of the steel frequently failed, while steels having the same composition and, apparently, identical thermal treatment, were completely unaffected by the reagents.

No indisputable authority can be readily cited on the exact nature of the mechanism leading to intercrystalline corrosion of the above stabilized steels, It is now thought, however, that irregular stability in corroding media may be tentatively associated with the thermal treatment to which the steels have been subjected in the process of manufacture. Titanium and columbium carbides, reputedly stable in ferrous alloys, remain insoluble therein until a certain temperature minimum is reached. Once heated above said temperature minimum, the said carbides enter a solid solution and lose their identity as separate phases. Slow cooling from above solution temperature might regenerate the carbides concerned. Cooling at a faster rate than required for precipitation retains carbon and stabilizing elements in a solid solution, thus being free to move. Migration velocity thereof differs greatly. During sensitizing treatment, carbon atoms become segregated at the grain boundaries, while those of the stabilizing elements remain in situ, being impotent to exert any action on carbon.

Steelmaking operations are always characterized by a certain latitude of thermal ranges involved. These variations usually do not lead to any ill-effect in the application thereof to common grades of steel. In case of stabilized austenitic stainless steels, the above digressions heavily bear on the quality of the metal produced, particularly when met in the latter stages of the processing. Austenitizing heat treatment is particularly prone to result in a portion, or even totality, of stabilized carbides entering a solid solution leading to the aforementioned uncertain results from the standpoint of stability towards intercrystalline corrosion.

The present invention provides means for overcoming the foregoing disadvantages and has for an important object the provision of an improved austenitic stainless steel immune to intercrystalline corrosion.

A further object of the invention is to provide austenitic stainless steels of improved composition that are immune to intergranular corrosion.

Other objects of this invention will become apparent from the following description and the appended claims.

The invention will be understood more readily by reference to the accompanying drawings, wherein:

Figure 1 is a chart showing a series of graphs wherein free carbon content is plotted as a function of chromium content and of grain size of the steels.

Figure 2 is a chart which shows the distribution of compositions of satisfactory and unsatisfactory corrosion resistances when the behavior of steel after corrosion attack is plotted as a function of the carbon concentration and the chromium content.

Generally speaking, the present invention provides an austenitic stainless steel containing from substantially seven to substantially twelve per cent nickel, 16 to 25 per cent chromium, and titanium from zero per cent up to amounts equal to approximately six times the carbon content of the steel, or columbium in amounts equal to substantially ten times the carbon content, the said steels being insensitive to intergranular corrosion.

In accordance with the present invention, it has been discovered that a definite relation exists between the percentage of carbon precipitabe during a heat treating step inducive to carbide precipitation and intergranular corrosion in the steels of the class described and the chromium content thereof. A further discovery demonstrated a pronounced difference, viewed from the corrosion-resistance standpoint, between the total percentage of carbon precipitated and the percentage of the precipitate thereof at the grain boundaries at a given chromium content of steel. Finally, it has been found that the concentration of chromium carbides precipitate at the grain boundaries at a given chromium content of steel is the controlling factor in determining the stability of a steel under conditions leading to intergranular corrosion.

An austenitic stainless steel is substantially free from intercrystalline corrosion when the carbon concentration available for combination with chromium at the grain boundaries is equal to or less than a specic maximum thereof, which is a function of the chromium content and grain size of steel involved. In general, said maximum free carbon concentration can be expressed for a given grain size of steel, as a function of chromium content of the steel multiplied by the tangent of an experimentally determinable angle. Assuming E being the permissible maximum of free carbon concentration which is plotted on the axis of ordinates, A being the percentage of chromium content at the origin of coordinates, B the chromium content of the steel under scrutiny plotted on the axis of abscissae, the aforementioned maximum free carbon concentration can be expressed as E=(B-A) tan a where a is an angle determined as recited in my copending application Serial No. 627,076, led November 6, 1945.

Otherwise stated, the above relation can be rephrased to read that the permissible maximum concentration of free carbon at the grain boundaries, assuring a substantial freedom of austenitic stanless steels from intercrystalline corrosion, is equal to the difference between the actual chromium content of a steel and the percentage of chromium chosen for the origin of coordinates used for presentation of said relation, said difference being multipled by the tangent of an experimentally determined angle. When, for example, the chromium content at the origin of coordinates is chosen as 16.75% and angle a, is found to be 18, the expression for the permissible maximum of free carbon concentration becomes E: (B-16.75) tan a=O.375 (B-16.75)

This tangential relation between free carbon concentration and chromium content has been found to prevail throughout the whole eld of austenitic chromium-bearing stainless steels investigated in the course of the study leading to the present invention. While the angle involved varies, the determination thereof does not present any substantial diiiiculties.

It may be noted here that intercrystalline coru rosion, as the name implies, deals with phenomena occurring at the grain boundaries. The amount of chromium carbide precipitation at the grain boundaries must be considered as precipi tation taking place within the crystals has no bearing on this type of corrosion other than to lower the amount of chromium carbide precipitated intergranularly. The pronounced mobility of carbon atoms at elevated temperatures permits a tentative assumption of precipitation taking place entirely at the grain boundaries at temperatures in the sensitizing range. A correct representation of the process of chromium carbide precipitation and subsequent corrosion is obtained, therefore, through reference of the total precipitable amount of carbon found in a given volume of the metal to the area of grain boundaries present in this Volume.

Corrosion resistance of stainless steels of the type described is generally held to be associated with the chromium content, higher chromiuml content causing a greater resistance. Impoverishment of an area of steel by chromium carbide precipitation might reduce the chromium concentration to a dangerous minimum in case of low original chromium content, while precipitation induced by the same amount of carbon may leave a sufficient chromium concentration in a steel alloyed originally with a higher chromium content. An accurate picture of potential corrosion characteristics can be obtained by relating the amount of carbon precipitated at the grain boundaries to the chromium content of the steel.

A predetermining of corrosion characteristics of stainless austenitic steel requires, therefore, ascertaining the percentage of chromium contained in the metal. determining free carbon content thereof, finding potential free carbon concentration at the grain boundaries, determiningv carbon concentration limits inducive to noncorrodibility. and comparing for carbon concentration found in a specimen under investigation with the above limits.

In accordance with the present invention, the Y chromium content of the steel is determined by following generally available methods of quantitative chemical analyses. Also, there is ascertained the percentage of free carbon by Aconventional methods of analysis, but preferably the total percentage of carbon is determined by a combustion method. Then there is determined by known methods the amount of carbon which is tied up by carbon stabilizing elements, principally as titanium and columbium carbides, and subtracting the latter value from the former.

The carbon concentration at the boundaries is determined in accordance `with the present in-V vention, by dividing the free carbon content as determined above, by the area of the grain boundaries. It may be considered that the grain of any specimen of steel is sufficiently uniform to hold a planar section through it as representative of the three-dimensional relations actually extent. In this light, the area of a specimen can be substituted, for the purpose concerned, for its volume, and the length of the grain boundaries for their area. Equally valid results are obtained by substitution of any desired unit area for the total surface examined, and measuring the length of boundaries of the grains contained in this unit area.

Instead of actual measurements of grain boundaries length, it is preferred to utilize generalizations laid as the foundation for the A. S. T. M. grain size chart, to determine by observation under a microscope the grain size number, and to calculate the corresponding length of the grain boundaries by using a suitable formula expressing the length of grain boundaries as a function of the unit of area used and grain size observed.

Such a formula may be derived as follows:

Let it be assumed that the grains have uniformly hexagonal shape and the same size. The number N of grains per square inch, at 10Q diameter magnification, will be:

N=2nl where n is the A. S. T. M. grain size. Then a C, l aX tan 60 =l square meh where a is the side of the hexagon.

Hence the perimeter P of the grain will be:

=l square inch rIhe total perimeter A of grains totaling in their area l square inch is i'ra The actual method of deriving the numerical value ofthe formulasobtained are of minor'importance. The said formulas'are intended to present the periphery of grainscontained in a unit of area expressed in the same-units, for example, inches per square inch- The free carbon content, determined as previously indicated, divided by the length of grain boundaries, in a unit of area determined as indicated above, gives the concentration of free carbonin per unit length, for example, per inch per square inch.

Selecting a representative number of specimens depicting adequate variations of free carbon and of chromium content, determining boundaries-carbon concentration thereof, sensitizing the-metalby proper heat treatment, and subjecting the sensitized specimens to actual intercrystallinev corrosion testing, provide a set of values for predetermining corrosion behavior of similar steels.

Figure 1 of the accompanying drawings presents the distribution of the aforementioned values when the behavior of steel after corrosion attack is plotted as a function of the carbon concentration and the chromium content.

In this view, circles represent specimens which failed from the corrosion resistancev standpoint, and crosses indicate steels which withstood the attack of the corrosive media. A sharp separation between the area comprising satisfactory specimens and the area comprising defective specimens is shown clearly, the separation taking place substantially along the line C D. When free carbon concentration at the grain boundaries is equal to or less than a critical value E, dependent upon the chromium content of the steel, no intercrystalline corrosion will occur in s the corresponding austenitic steel of the chromium-nickel type.

This critical value E can be determined from Figure 2. Let B indicate the chromium percentage of the steel. Then the critical value E can be expressed for any point on the line C-D as E: (Ee- 16.75) tan a which is the formula given above.

As has been pointed out above, actual measurements show the value of the angle a is substantially 18. This value is Valid in connection with the above formula for grain periphery determination. Further simplified, the above formula becomes Other formulas for grain periphery determination give different values for the angle a without departing from the straight-line law of the present invention. A lower or higher chromium concentration than the 16.75% used in the above example for arriving at a formula for E requires merely replotting of experimental results for determining angle a corresponding to the new chromium content range.

The steps of ascertaining average grain size of a steel according to A. S. T. M. values, determining the chromium content thereof, computing carbon concentration at the grain boundaries through a suitable formula as set forth above, and employing the formula indicated above for the critical value E of carbon concentration, permit definite predetermination of corrosion characteristics of any steel embraced by the corresponding limits of composition, prior to actual corrosion testing.

The application of the general law disclosed above to practical'uses canbe illustrated in reference to steels containing 9 to 11% nickel and 16.75 to 18.40% chromium and varying in grain size from No. 1 to No. 8 A. S. T. M. standard grain size chart, by the accompanying diagram of Figure 2 which represents graphs wherein free carbon content is plotted as a function of chromium content and of grain size of the steels. These graphs are interpreted by regarding the carbon content for any grain size and a given chromium content as potentially inducive to intercrystalline corrosion when said carbon content is above the point of intersection of the grain-size line with chromium content of the steel, and indicates a steel immune to intercrystalline corrosion When said carbon content is found therebelow. Thus, for instance, a steel having No. 6 grain and containing, for example, 8% nickel and 17.90% chromium, is liable to fail through intergranular corrosion when the free carbon content thereof is 0.04%, and be immune thereto when free carbon content thereof is 0.03%, since the aforementioned point of intersection A is below the upper value and above the lower one.

The present invention provides a series of alloys assuring the freedom from intercrystalline attack, through adjusting free carbon content thereof to chromium concentration and the grain size of' steel. Three aforementioned factors entering compositional limitations of the present invention render impossible the outlining of the preferred range of the composition in numerical values valid for all possible combinations thereof and require the use of charts as shown in Figure 2.

These charts show, among other things, a pronounced decrease of sensitivity towards intergranular corrosion of steels containing from '7 to 11% nickel and 16.75 to 18.40% chromium with the grain size thereof increasing above sub-v stantially the No. 5 grain size, offering thereby further means for avoiding the deleterious effect of intergranular corrosion by keeping the size of the grains as small as possible by a suitable control of deoxidizing practice. It is to be understood that the criterion of No. 5 A. S. T. M. grain size is used exemplarily in connection withv the specific range of compositions chosen as the basis of calculations leading to Figure 2 of the drawings, and may be conceivably shifted either upwards or downwards when another compositional range is used without contradicting the abovestatement relative to desirability of smaller grain s ze.

Chromium-bearing austenitic stainless steels of the present invention embrace alloys comprising, in addition to chromium, nickel and carbon, certain other elements introduced therein for specific purposes without destroying the substantially austenitic character thereof.

While new compositional requirements for austenitic stainless steels immune to intercrystalline corrosion have been recited and illustrated by factual examples, those skilled in the art will readily enlarge on the scope thereof while still remaining within the teachings and spirit of the present invention as dened in the appended claims.

I claim:

1- As a new metallurgical product, austenitic stainless steel alloyed with a metal of the group consisting of titanium and columbium and immune to intercrystalline corrosion, containing substantially 8 per cent nickel, more than 17.50

per cent chromium, 0.02 per cent carbon uncombined with the metal of said group, and having a grain size number greater than No. 6 according to the A. S. T. M. grain size standards.

2. Austenitic stainless nickel-chromium titanium bearing steels immune to-intercrystalline corrosion containing more than 16.75% chromium, 7 to 12 per cent nickel and a percentage of carbon uncombined with titanium below a minimum defined as a tangential function of the chromium content thereof, said tangential function being Within the range of an angle of about 4 degrees for steels having a No. 1 grain size according to the A. S.,T. M. grain size standards to an angle of substantially 25 degrees for steels having a grain size corresponding to No. 8 of the A. S. T. M. grain size standard in accordance with Figure 2 of the drawing.

3. Austenitic stainless nickel-chromium columbium bearing steel immune to intercrystalline corrosion containing more than 16.75% chromium, 7 to 12 per cent nickel, and a percentage of carbon uncombined with columbium below a minimum defined as a tangential function of the REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name Date 2,073,901 Newell Mar. 16, 1937 2,159,497 Becket et al. May 23, 1939 2,186,710 Schafmeister et a1. Jan. 9, 1940 2,374,396 Urban Apr. 24, 1945 OTHER REFERENCES American Society for Steel Treating, vol. 19. pages 673 and 696 to 710, June 1932.

Claims (1)

1. 2. AUSTENITIC STAINLESS NICKEL-CHROMIUM TITANIUM BEARING STEELS IMMUNE TO INTERCRYSTALLINE CORROSION CONTAINING MORE THAN 16.75% CHROMIUM, 7 TO 12 PER CENT NICKEL AND A PERCENTAGE OF CARBON UNCOMBINED WITH TITANIUM BELOW A MINIMUM DEFINED AS A TANGENTIAL FUNCTION OF THE CHROMIUM CONTENT THEREOF, SAID TANGENTIAL FUNCTION BEING WITHIN THE RANGE OF AN ANGLE OF ABOUT

Images