US3523788A - Austenitic stainless steel of improved stress corrosion resistance - Google Patents

Description

United States-Patent O 3,523,788 AUSTENITIC STAINLESS STEEL F INLPROVED STRESS CORROSION RESTSTANCE John F. Bates, Franklin Township, Westmoreland County,

and Alexei W. Loginow, Monroeville Borough, Pa., as-

shown, however, that austenitic stainless steels in such cold worked, and therefore highly strained state, are less resistant to stress-corrosion cracking than the same steels in the annealed condition. It is similarly desirable, therefore, to provide an austenitic stainless steel that is resissigllol's United States Steel Corporation, a 'P 5 tant to stress-corrosion cracking in both the annealed and Nd la r fvi g l inuation iii-part of application Ser No Cold-Worked conditions 619,552, Mar. 1, 1%7. This application June 2, 19.67, Alihmgh {t Posslble P l the E seLNo. 643,068 stress corrosion by the appl1cat1on of stress rehevmg Int CL u 39 14 10 treatments and proper des1gn and process control, these US, Cl, 75-123 2 Cl i preventive measures are usually expensive and are not always reliable or practical. There is provided in accordance with the invention an austenitic stainless steel hav- ABSTRACT OF THE DISCLOSURE ing a composition and structure which renders it resistant 1 to stress-corrosion cracking without special preventive in 2 2322 22 2::z igggj ggi g zi ss treatments, and possesses as good or greater resistance to res istance to aid and pitting corrosion as well as liighpitting an? acid cOrro-sion than Type siainless' te e1 t mperature oxidation The steel consists essentially of and reslstance to lgh'temperamre loxldatlon comm-1rto 0 15% carbon; 1 0 to 2 5% manganese o 02% able to Type 310 sta nless steel. Steels 1n accordance wrth i p gp 0 1 sillfur 165 to 2 the lnventlon contain, 1n percent by welght, 0.05 to con 8 0 to 19 6 nickel to chromium 0.15% carbon, 1.0 to 2.5% manganese, 0.00% max. 005% .max min-(gen and 0,05% max my Wit}; phosphorus, 0.03% max. sulfur, 1.65 to 3.0% slhcon, 8.0 the aforementioned elements present according to a relato 190% mckel to 190% Chrommm 005% nitrogen and 0.05% max. molybdenum, with the aforetlonshlp expresscd by regressxon equation mentioned elements present according to a relationship expressed by a regression equation. A presently preferred composition range of the new This application is a continuation-in-part of applicasteel is 0.06 to 0.09% carbon, 1.0 to 2.0% manganese, tion Ser. No. 619,552 filed Mar. 1, 1967 now abandoned. 0.01% max. phosphorus, 0.01% max. sulfur, 1.8 .to 2.2% This invention relates to austenitic stainless steel. More silicon, 15.0 to 19.0% nickel, 17.0 to 19.0% chromium, particularly, the invention relates to an austenitic stainless 0.05% max. nitrogen and 0.02% max. molybdenum. steel composition which is resistant to stress-corrosion Several previous attempts have been made to provide cracking, and possesses good resistance to pitting corrothe industry with an austenitic stainless steel that is resion and acid corrosion as well as high-temperature oxisistant to stress-corrosion cracking by virtue of its comdation. position. One such steel is disclosed in US. Pats. Nos. Austenitic stainless steels have had widespread use as 3,159,479 and 3,159,480. Stress-corrosion resistance of constructional materials in the nuclear and chemical inthis and the other presently available commercial steels dustries. Unfortunately, in certain environments it has is described in Table I which reports the composition been shown that stress-corrosion cracking of these steels and the results of stress-corrosion testing in both the cold can occur. Adverse conditions such as exposure to chloo k d nd a n al d di ion, A can b seen, some ride environments may precipitate stress-corrosion crackalloys have a higher resistance to stress-corrosion failure ing. Obviously, because of the hazards associated with the in the annealed condition but considerably less resistance failure of critical components in nuclear reactors and bein the cold-Worked form. For comparison purposes, the cause of the expense involved in replacing components composition of steels in accordance with the invention which fail in both the nuclear and chemical applications, are also provided in Table I; however, the stress-corrosion it is desirable to provide an austenitic stainless steel havresistance properties are omitted here and will be discussed ing mechanical properties comparable with those of preshereinafter. As will be seen, the stress-corrosion reently available austenitic stainless steels but which is resistance of the balanced austenitic steel in accordance with sistant to stress-corrosion cracking to a greater degree. the invention possesses outstanding resistance in the cold- However, it is also important that such a steel be also reworked conditionv as well as in the annealed state.

TABLE I Time to failure, hours Cold Alloy 0 Mn P -S Si Cu Ni Cr Mo N Al Annealed worked AISI 304L 0. 029 1. 2s 0.019 0. 015 0. 35 ND 9.52 18.6 0.19 0.040 0. 02 5 2 A181 310 0. 061 1. 09 0.027 0. 029 0. 35 ND 12.06 18.09 2.30 0. 041 ND 3 4 A181 310 0.063 1.15 0. 024 0.013 0.32 ND 20.0 24.5 0.15 0. 038 0.01 15 11 IncoloySOO 0.038 0.83 0.017 0.005 0. 33 0.31 32.1 20.2 0. 074 0. 013 0. 24 1,189 167 Patent 3,159,479 (Alloy 84) 0.082 0. 50 0. 014 0.013 1.28 0. 05 20.9 17.7 0.03 0.029 0.005 NF 2,000 363 SCR m Q Max. Max. J 1L0 .Max. Max} a .212 at as as is? 12::- sCR'(Prefmd) 0:00 2. 0 0.01 0.01 "2. 2i "1?? TX) 0.02 0.05}

Norn.-ND =Not determined; NF=No iailureet indicated times; SCR=Stress corrosion resistant steel in accordance with the invention.

sistant to pitting corrosion, acid corrosion and high-temperature oxidation if it is to enjoy the generality of applications to which its outstanding properties render it Well suited.

In many applications, austenitic stainless steels are used in cold-worked conditions. Typical of such uses are cold rolled and formed sheets, drawn cups and trays, as well as cold-formed tubing and piping. Experience has To obtain resistance to stress-corrosion cracking with ,good hot workability, the steel in accordance with our invention must be balanced within the specified coniposition limits, with respect to the carbon, nitrogen, manganese, phosphorus, molybdenum, sulfur and silicon contents. It is necessary to maintain the carbon, manganese and silicon contents at a relatively high level and to restrict the nitrogen, phosphorus, molybdenum and sulfur contents to low levels. It is additionally important to maintain the manganese relatively high, e.g. 1% or more, for sulfur control to achieve good hot workability of the steel. It is also very important that the effects of crackpromoting elements (molybdenum, phosphorus, nitrogen, chromium and sulfur) be greatly overcome by the crackretarding elements (nickel, carbon and silicon) for best performance in the annealed as well as in the cold worked condition. In addition, the effect of the total austenitizing elements, such as carbon, manganese, nickel and nitrogen, must be greater than that of the ferritizing elements, such as chromium, molybdenum and silicon, to render the steel predominantly austenitie at both ambient and hotworking temperatures. The austenitic structure is desirable for best cold forming and hot-working performance.

Balancing of the elements is accomplished by use of a regression equation which has been developed to describe the relationship of these elements necessary for improved stress-corrosion resistance in accordance with the invention. The regression equation for stress-corrosion behavior of austenitic stainless steels in the cold worked condition in accordance with the invention is as follows:

Natural logarithm of cracking time:

10.2 ---41 P 75 S +1.8 Si +0.33 Ni 0.60 Cr -13 Mo +23 (P0.016) (Si1.1) 6.1 (P-0.0l6) (Ni-16) -0.0058 (Ni16) Error in prediction at the 95% probability limit=i0.92 Coefiicient of determination (R =95.9%.

The aforementioned regression equation which was developed after extensive investigation relates the compositional elfects to the failure times of cold-worked specimens. A logarithm transformation was made on the mean cracking times prior to analysis and the equation is expressed in terms of natural log cracking time. A positive sign associated with single element effects and interaction effects indicates a beneficial effect, and a negative sign indicates a detrimental effect. The interaction effects included in the regression equations indicate that the effect on the cracking time of one chemical element differs at different levels of another element. Those knowledgeable in the art will appreciate that there is usually a wide spread in stress-corrosion data. To indicate the high level of reliability, also included above, are a measure of the error in predicting the natural log cracking time at the 95% probability limit and a measure (R of the percent of total variation in the natural logarithm of the cracking time explained by the composition eflects.

We have found that the desirable combination of stresscorrosion resistance and hot workability can be achieved if the composition is critically controlled within the relationship described by the aforementioned regression equation with each element present within the ranges previously given. Moreover, we have discovered that manganese need not be, as has been previously believed, detrimental to stress corrosion resistance, and it is possible to include the relatively high manganese level necessary for improved hot workability. Thus, with silicon within the range of 1.65 to 3.0%, manganese contents within the range of 1.0 to 2.5% may be maintained without sacrificing resistance to stress-corrosion cracking and will result in outstanding hot workability for these types of steels.

We have also discovered that it is necessary to restrict molybdenum to relatively low levels on the order of 0.05% max. to provide the desired resistance to stresscorrosion cracking. In addition, for good resistance to corrosion and high-temperature oxidation, the steel should contain sufficient chromium (up to about 19%) although lower contents of chromium (about 15%) would impart improved stress-corrosion performance. Nickel is added to the steel to ensure an austenitic structure, high ductility, good cold-working properties and resistance to stress-corrosion cracking. Although greater amounts of nickel, i.e. 20% or more, are beneficial to obtain in-' creased resistance to stress-corrosion cracking, we have found that the detrimental effects of phosphorus are more pronounced at increasing nickel levels and, of course, hot workability sufiers. The nickel content of the balanced steel composition in accordance with the invention is maintained at a level such that the steel still has good hot workability. The carbon content is controlled closely at a relatively high level to provide austenite stabilization and resistance to stress-corrosion cracking. The nitrogen content of the steel is kept at a low level. For best resistance to stress-corrosion cracking, the nitrogen content is preferably lower than 0.05%. The phosphorus content is kept below 0.01% for resistance to stress-corrosion cracking of the cold-worked steel and the silicon content is kept as high as is consistent with good hot-working characteristics. Sulfur is kept low for hot workability and for resistance to stress-corrosion cracking. Other elements, such as copper and aluminum, when contained in the new steel, are present in the small amounts normally present in 18% chromium, 8% nickel (AISI Type 304) austenitic stainless steels.

The following examples will aid in understanding the invention.

For the examples, heats were melted in an induction furnace and hot rolled to /2-inch thick plate. Specimen blanks (V: x /2 x 3 /2-inches) were cut with the long dimension parallel to the rolling direction. The blanks were solution annealed and water quenched. Tension specimens, 0.252-inch in diameter, were machined from the blanks. Some of the specimens were tested in the annealed condition by loading to 75% of the yield strength. Other specimens were elongated 30% in a tensile machine and tested in a cold-worked condition by loading to 75% of the new yield strength. A standard hot salt test was used in which the specimens were totally immersed in a boiling 42% magnesium chloride solution.

Usually three specimens were tested in the annealed and three in the cold-worked condition for each heat. Cold-worked specimens were removed after a minimum of 2000 hours if they had not failed; annealed specimens :were removed after times varying from 2000 to 4500 hours if there had been no failure. The mean times to failure were used to evaluate the materials. Where some specimens from a given heat failed and others did not fail, there is no simple mean that will satisfactorily describe such behavior. Consequently, the value used was arbitrarily chosen'to be the mean of the times to failure and the no-failure times; such values are indicated as erratic (E) in the tables to recognize the inadequacy of the numerical value. Unlike other alloys, austenitic steels in accordance with the invention may provide a time-to-failure in the cold worked condition of as much as 450 hours or more when tested as described above.

The compositions and the stress-corrosion behavior of a series of known steels, as discussed previously, are

given in Table I above. One composition is a melted heat within the range of the aforementioned US. Pat. No. 3,159,479. It is noted that although steel of typical compositions within the aforementioned patent have good resistance to stress-corrosion cracking in the annealed condition, resistance in the cold-worked condition is limited. Moreover, as will be hereinafter pointed out, steel in accordance with the invention possesses good resistance to acid and pitting corrosion as well as to high-temperature oxidation. Thus, a superior characteristic of steels 5 in accordance with the invention is the unique combination of outstanding properties with respect to (1) stresscorrosion resistance, (2) pitting-corrosion resistance, (3) acid-corrosion resistance, and (4) high-temperature oxi- TABLE II.EFFECT OF NICKEL Composition, percent Time to failure, hours Alloy Mn P S Si Cu Ni Cr Mo N Al Annealed worlad= NOTE.NF=N0 failure at indicated times.

dation resistance. Carbon is a beneficial element in stress-corrosion re- The cr1ticality of the comp0s1t1on may be seen by exsistance as indicated by the data shown in Table III. As can be seen by comparing the annealed condition properties, maintaining carbon over about 0.05% materially improves stress-corrosion resistance. It is also noted; that the high carbon content is desirable for the beneficial effect. of low nitrogen to become operative (carbon-nitrogen in teraction). In comparison Groups A, B, C and. D, increased carbon improved the stress-corrosion resistance in the annealed condition. In threeof the five comparison groups, Groups B, D and E, carbon is shown to he distinctly beneficial for cold worked materiaL,

TABLE TIL-EFFECT OF CARBON Composition, percent Time to failure, hours Cold Group Alloy 0 Mn P S Si Cu Ni Or Mo N AI Annealed worked A 9 0. 034 1. 53 0. 025 0. 017 0. 53 0. 01 17. 8 17. 8 0. 01 0. 006 0. 004 220 38 10 0. 073 1. 0. 029 0. 015 0. 50 0. 01 17. 9 17. 8 0. 01 0. 005 0. 010 NE 4,500 48 B 1 0. 010 2. 39 0. 005 0. 011 0. 19 0. 01 18. 1 16. 4 0. 01 0. 007 0. 007 627 138 2 0. 07 2. 53 0. 007 0. 011 0. 24 0. 01 17. 9 17. 3 0. 01 0. 010 0. 012 NF 3,300 520 C, 3 0. 044 1. 51 0. 025 0. 017 0. 49 0. 01 7. 9 17.6 0. 01 0. 005 0. 011 921 4 4 0.075 1. 50 0. 028 0. 012 0. 50 0. 01 8. 0 17. 7 0. 01 0.006 0.008 NF 4,500 2 I) 90 0. 021 1. 53' 0. 008 0. 014 1. 96 0. 11 18. 2 17. 8 0.028 0. 033- 0. 011 974 223 75 0. 074 1. 52 0. 007 0. 013 1. 99 0. 10 18. 1 18. 5 0. 030 0. 035 0. 010 NF 2,000 E 1,649

E 21 0. 053 1. 44 0. 010 0. 016 0. 28 0.01 17. 2 17. 9 0.015 0.009 0.021 NF 4,600 50 22 4 0.075 1. 44 0. 008 0.019 0. 34 0. 01 18. 3 17. 6 0. 015 0. 011 NF 4,600 90 NOTE.NF=N0 failure at indicated times; E =Erratic behavior.

creases resistance of austenitic stainless steels to stresscorrosion cracking in hot chloride solutions. However, very high nickel contents, i.e. 20% or greater, are not desirable because of a detrimental nickel-phosphorus interaction which is hereinafter discussed with the effects of phosphorus and because of the detrimental efiect on hot workability.

The efiect of nickel is shown in Table II which is divided into two groups of data, and each group is a comparison series with increasing nickel content. It will be noted that for both cold worked and annealed material, increasing nickel content is beneficial. The properties of these steels are all poorer than the optimum composition because the optimum contents of silicon, phos- The effect of silicon is shown infTable IVwhich-indi cates that silicon is beneficial in the cold-worked material but has a less significant effect in the annealed material. However, with a crack-promoting element (molybdenum) at a high level, the beneficial eifect of silicon in the annealed condition is clearly evident. An interaction of the beneficial silicon effect with the detrimental phosphorus effect is discussed with elfects of phosphorus. It will be seen from Table-.IV that the firsttwo-groups at'the:9% nickel level show little effect of increasing. silicon from the 0.4 to the 0. 9% level. However, increaseiof silicon to 2.0% was invariablybeneficial.at'the. 18% nickel'level; The last two comparison groups show the great advantage of having silicon higher than 1.65%

TABLE IV.-Efleet of Silicon Time to failure, hours Composition, percent Cold Mn P S Si Cu Ni Cr Mo N Al Annealed Worked 1. 53 0. 003 0. 015 0. 45 0. 02 9. 05 18. 9 0. 01 0. 012 0. 012 NF 2, 000 2 1. 56 0. 003 O. 015 0. 89 0. 02 9. 05 18. 7 0. 01 0. 014 0. 015 NF 2, 000 2 1. 55 0. 031 o. 016 0. 41 0. 05 s. 90 17. 0. 01 0. 010 0.017 E 1, 371 2 1. 52 0. 030 0. 016 0. 86 0. 05 8. 88 16. 9 0. 01 0. 011 0. 017 NF 2, 000 G 0. 50 0. 014 0. 013 0. 05 20. 9 17. 7 O. 03 0. 029 0. 005 NF 2, 000 363 1. 45 0. 014 0. 013 2. 00 0. 17. 8 17. 8 0. 034 0. 036 0. 009 NF 2, 000 674 1. 51 0. 025 0. 017 0. 40 0. 05 17. 7 18. 0 0. 01 0. 010 0. 016 NF 2, 000 81 1. 52 0. 026 O. 017 1. 98 0. 05 18. l 18. 0 0. 01 0. 009 0. 010 NF 2, 000 786 1. 47 0. 010 0. 016 0.90 0. 10 17. 6 17. 6 0. 030 0. 042 0. 003 NF 2, 000 100 1. 47 0. 010 0. 015 1. 44 0. 10 17. 5 17. 8 0. 030 0. 038 0. 003 209 1. 51 0. 009 0. 014 2.05 0. 10 17. 8 18. 0 0. 026 0. 034 0. 005 NF 2, 000 1, 012

N 0TE.-N F=N0 failure at indicated times; E=Erratic behavior.

low a level as is presently technologically possible, even 40 below 0.015%, will result in further increases in stresscorrosion resistance.

such that the detrimental effect of phosphorus becomes less pronounced as the silicon content increases. A phosphorus-nickel interaction also occurs with the result that the detrimental efiect of phosphorus becomes more pronounced as the nickel content increases. This interaction is of very great importance in itself and in context with other discoveries described herein. Moreover, these interactions very closely point up the necessity of critically controlling the composition. Thus, the content of each element must be counterbalanced in accordance with the regression equation by other elements for an optimum combination of properties.

The single element comparisions of variouss phosphorus contents shown in Table VI are divided into seven com- TAB LE V.-Efiect of Molybdenum Time to failure, hours Composition, percent Cold Mn P S S1 Cu N 1 Or Mo N Al Annealed worked 1. 52 0. 010 0. 016 0. 42 0. 05 17. 9 18. 4 0. 01 0. 009 0. 012 NF 2, 100 E 976 1. 44 0. 008 0. 019 0. 34 0. 01 18. 4 17. 6 0. 015 0. 011 0. 008 NF 4, 600 96 1. 89 0. 008 0. 026 1. 94 0. 10 17. 8 17. 9 0. 028 0. 043 0. 008 NF 2, 000 408 1. 60 0. 006 0. 016 1. 89 0. 09 17. 7 17. 5 0.22 0. 043 0. 007 NF 2, 200 67 1. 0. 008 0. 013 2. 00 0. 1O 17. 8 17. 7 0. 033 0. 031 0. 009 NF 2, 000 Y 814 1. 45 0. 010 0. 012 1. 98 0. 10 17. 9 17. 7 0. 072 0. 032 0. 007 NF 2, 000 368 1. 0. 010 0. 012 1. 99 0. 11 18. 0 17. 6 0. 21 0. 027 0. 011 NF 2, 000 106 NoTE.NF=No failure at indicated times; E=Erratic behavior.

The effects of phosphorus are shown in Table VI. These examples demonstrate that phosphorus is quite detrimental in the cold-worked material, but is somewhat detrimental in the annealed state when the molybdenum content is parison groups. Especially noteworthy is the first comparison group of four heats which Were made as a split heat, varying only phosphorus. Without exception, each group shows phosphorus to be detrimental in the coldrelatively high. A phosphorus-silicon interaction occurs Worked state.

TABLE VL-Effeet of Phosphorus Time to failure, hours Composltiompercent Cold Mn P S Si Cu Ni Cr Mo N Al Annealed worked 1. 48 lfii 0. 011 2. 02 0. 11 18. 17. 9 0. 21 0. 026 0. 009 NF 2, 000 119 1. 50 0.010 0.012 1. 99 0.11 18.0 17.6 0.21 0. 027 0.011 NF 2, 000 106 l 1. 48 0. 020 0. 012 2. 00 0. 11 18. 0 17. 9 0. 22 0. 028 0.012 1, 235 81 1. 49 0. 022 0. 011 1. 98 0. 11 17. 9 17. 9 0. 21 0. 029 0. 009 916 70 1. 46 0. 004 0. 014 1. 97 0. 10 17. 8 17. 7 0. 034 0. 030 0. 010 NF 2, 000 E 1, 208 1. 45 O. 008 0. 013 2. 00 0. 10 17. S 17. 7 0. 033 0. 031 0. 009 NF 2, 000 814 1. 45 0. 014 0. 013 2. 00 0. 10 17. 8 17. 8 0.034 0. 036 0. 009 NF 2, 000 674 1. 45 0. 004 0. 014 1. 85 0. 10 17. 9 17. 7 0.074 0. 030 0. 005 NF 2, 000 428 1. 45 0. 011 0. 012 1. 81 0. 10 17. 9 17. 6 0. 072 0.090 0. 008 NF 2, 000 353 1. 45 0. 010 0. 012 1. 98 0. 10 17. 9 17. 7 0. 072 0. 032 0. 007 N F 2, 000 308 1. 52 0. 010 0. 016 0. 42 0. 17. 9 18. 4 0. 01 0. 009 0. 012 NF 2, 000 E 976 1. 51 0. 025 0. 017 0.40 0. 05 17. 7 18. 0 0. 01 0. 010 0. 016 NF 2, 000 81 23 0. 083 1. 63 0. 009 0. 008 1. 94 0. 030 18. 1 18. 0 0. 01 0. 009 0. 026 NF 2, 100 NF 2, 100 1. 52 0. 026 0. 017 1. 98 0. 05 18. 1 18. 0 0. 01 0. 009 0. 010 NF 2, 000 786 1. 50 0. 007 0. 014 1. 92 0. 01 17. 5 17. 3 0.'01 '0. 021 0. 005 NF 2, 000 NF 2, 000 1. 45 0. 030 0. 013 1. 96 0. 01 17. 4 17. 5 0. 01 0. 022 0. 011 841 2 0. 07 2. 53 0. 007 0. 011 0. 24 0. 01 17. 9 17. 3 0. 01 0. 010 0. 012 NF 3, 300 520 10 0. 073 1. 55 0. 029 0. 015 0. 50 0. .01 17. 9 17.8 0. 01 0. 005 0. 010 NF 4, 500 48 N OTE.NF=N0 failure at indicated time; E=Erratie behavior.

It has been found that nitrogen is detrimental in the annealed material, but that the effect for cold-worked material is too small to be significant. A carbon-nitrogen in teraction for annealed steels occurs such that the beneficial efiect of low nitrogen content is less pronounced at low carbon levels and the beneficial effect of carbon is less pronounced at high nitrogen levels. Technological considerations limit the lowest practical level for nitrogen and metallurgical factors, such as intergranular carbide precipitation, control the upper limit for carbon.

In Table VII the first and second comparison groups are split heats varying only in nitrogen. The stress-corrosion behavior of the steels in the first group are rather poor because all of these heats are high in molybdenum (0.22%). The second comparison group represents additions of nitrogen above the levels normally encountered in commercial austenitic stainless steels .(0.03 to 0.05% N). Both of these groups, however, indicate a detrimental effect for nitrogen in'the cold-worked condition.

The fourth comparison group, with up to 0.021% nitrogen, comprises steels that did not fail in the cold-worked condition. These steels, however, contain a very low level of molybdenum, less than 0.01%. The last two comparison groups show the marked detrimental etfect of nitrogen on annealed steels having low nickel and high phosphorus contents.

Composition, percent Time to failure, hours Cold 0 Mn 1? S Si Cu Ni Cr Mo N Al Annealed worked 0. 069 1. 50 0. 011 0. 012 2. 01 0. 11 18. 0 17. 7 0. 22 0. 005 0. 008 NF 2, 000 129 0.071 1. 50 0.011 0.013 2. 04 0. 10 18.0 17.8 0.22 0. 010 0.010 E 1, 300 115 0. 072 1. 50 0. 013 0. 014 2. 08 0. 11 18. 0 l7. 7 0. 22 0. 015 0. 009 E 1, 450 110 0. 068 1. 50 0. 007 0. 011 2. 03 0. 11 18. 0 17. 6 O. 22 0. 025 0. 010 NF 2, 000 9G 0. 051 1. 50 0. 008 0. 010 1 98 0. 11 18. 0 17. 8 0. 22 0. 036 0. 010 E 1, 500 79 0. 081 1. 48 0.011 0. 016 2. 04 0. 10 19. 6 l7. 7 0. 028 0. 052 0. 006 NF 2, 000 526 0. 084 1. 48 0. 010 0. 016 2. 02 0. 10 19. 6 17. 6 0. 026 0. 072 0. 006 NF 2, 000 253 0.080 1. 48 0.006 0. 017 2. 06 0. 10 19. 6 17. 7 0.028 0. 105 0.005 NF 2, 000 141 0. 075 1. 44 0. 008 0. 019 0. 34 0. 01 18. 4 17. 6 0. 015 0. 011 0. 008 NF 4, 600 96 0. 070 1. 49 0. 013 0. 024 0. 37 0. 02 18. 2 17. 7 0. 017 0. 030 0. 010 NF 2, 000 69 0.083 1. 63 0. 009 0. 008 1. 94 0. 030 18. 1 18. 1 0. 01 0. 009 0. 026 NF 2, 100 NF 2, 100 0. 080 1. 46 0. 010 0. 013 1. 86 0. 01 17. 4 17. 3 0. 01 0. 015 0. 007 NF 2, 000 NF 2, 000 0. 067 1. 50 0. 007 0. 014 1. 92 0. 01 17. 5 17.3 0. 01 0. 021 0. 005 NF 2, 000 NF 2, 000

0. 072 1. 62 0. 003 0. 016 1. 95 0. 10 17.4 17. 5 0. 033 0. 005 NF 2, 000 E 1, 728 0. 065 1. 42 0. 003 0. 015 1. 90 .0. 10 17. 8 17. 8 0. 035 0. 023 0. 003 NF 2, 000 823 0. 078 1. 46 0. 004 0. 014 1. 97 0. 10 17. 8 17. 7 0. 034 0. 0 0. 010 NF 2, 000 E 1, 208

0. 072 1. 54 0. 009 0.015 1. 99 0.10 18. 0 18.1 0. 030 0. 010 NF 2, 000 752 0. 075 1. 0. 008 0. 013 2. 00 0. 10 17. 8 l7. 7 0. 033 0. 031 0. 009 NF 2, 000 814 0. 074 1. 89 0. 008 0. 026 1. 94 0. 10 17. 8 17. 9 0.028 0. 043 0. 008 NF 2, 000 408 0.075 1. 0. 028 0.012 0 50 0. 01 8.0 17.7 0. 01 0 006 0.008 NF 4, 500 29 0.057 1. 51 0.024 0.017 0 50 0. 01 7. 9 17.6 0. 01 0 043 0. 100 4 Z2 NOTE.NF=NO failure at indicated times; E=Erratio behavior.

We have also observed that increasing chromium above A test for pitting corrosion is conducted in an acidified about 19% is detrimental in cold-worked material. Thus, ferric chloride solution (108 g. FeCl -6H O, 4.5 ml. conc. it is desirable to have as low chromium as possible con- HCl per liter of solution) for 4 hours at 95 F. A series sistent with the needed resistance to general corrosion. of specimens of the composition of steels shown in Sulfur in the range 0.01 to 0.02% is also detrimental for 5 Table IX were treated in the test solution and the weight cold-worked materials and, therefore, the sulfur content loss of the specimens was calculated as milligrams per should be kept as low as possible. Furthermore, a low square decimeter per day (m.d.d.), a unit commonly used sulfur content improves the hot-working properties of the for non-uniform (pitting) corrosion testing. The results steel. Manganese in the range 1.0 to 2.5%, aluminum in of the test are reported in Table X and as can be seen, the range 0.005 to 0.03% and copper in the range 0.01 to Alloy 78, a composition in accordance with the invention,

0.12% do not have a significant effect on stress-corrosion had a somewhat lower corrosion rate than Type 304 stainless steel and an appreciably lower corrosion rate than behavior.

A number of steels within the scope of the invention are Alloy TABLE IX.-STAINLESS STEELS TESTED Alloy 0 Mn P S Si Cu Ni Cr Mo -Al N 1.45 0. 00s 0. 013 2.00 0.10 17.8 17.7 0. 03 0. 009 0. 03s 0. 50 0. 014 0. 013 1. 2s 0. 05 20.9 17.7 0. 03 0. 005 0.021 1.70 0. 029 0.011 0.25 0.14 0.50 18.4 0.10 ND 0. 039 1. 4.7 0. 010 0.016 0.67 ND 20.1 24.4 0. 086 ND 0. 02s

NoTE.ND=Not determined.

TABLE X summarized in Table VIII. These examples show that varying degrees of resistance to stress-corrosion cracking may be obtained within the limits imposed by our invention. However, they also show that the steels of our inven Resistance to pitting corrosion in a ferric chloride solution tion have a considerably better resistance to stress-corro- AHOY: Corroslon Tate 1 sion cracking in both annealed and cold-worked conditions 78 310 than currently available steels also presented in the table, 5 550 and in particular, in the cold-worked condition a time-toiype 304 390 failure Of greater than 450 hours. 1 M.d.d.Milligrams per square decimeter per day.

TABLE VIII Time to failure, hours Composition, percent Cold Alloy 0 Mn P S Si Cu Ni Cr Mo N Al Annealed worked Stress Corrosion Resistant Steels 1. 63 0. 009 0. 008 1. 94 0. 03 18. 1 18. 0 0 01 0. 009 0. 026 NF 2, 100 NF 2, 100 1. 46 0.010 0.013 1. 86 0. 01 17. 4 17.3 0 01 0.015 0.007 NF 2, 000 NF 2, 000 1. 50 0.007 0. 014 1.92 0. 01 17. 5 17.3 0 01 0. 021 0.005 NF 2 000 NF 2, 000 1. 62 0.003 0. 016 1. 95 0. 10 17.4 17. 5 0 033 0.019 0. 005 NF 2, 000 E 1, 728 1. 51 0.007 0. 013 1. 92 0.10 18.4 18.3 0 031 0.032 0.010 NF 2, 000 E 1, 486 1. 52 0.007 0.013 1. 09 0. 10 18.1 18. 6 0 030 0. 035 0.010 NF 2, 000 E 1, 64

1. 46 0. 004 0.014 1.97 0. 10 17.8 17. 7 0 034 0.030 0.010 NF 2,000 E 1,208 1. 51 0. 009 0. 014 2. 05 0. 10 17. 8 18. 0 0 026 0. 034 0. 005 NF 2, 000 1, 01d 1. 42 0. 003 0.015 1. 90 0. 10 17.8 17.8 0 035 0.023 0.003 NF 2, 000 S2 1. 45 O. 008 0. 013 2. 00 0. 10 17. 8 17. 7 0 035 0. 031 0. 009 NF 2, 000 814 1. 54 0. 009 0. 015 1. 99 0. 10 18. 0 18. 1 0 030 0. 017 0. 010 NF 2, 000 752 1. 45 0. 014 0 013 2. 00 0. 10 17. 8 17. 8 0 034 0. 036 0. 009 NF 2, 000 674 Commercial Steels 1. 28 0. 019 0. 015 0. 35 ND 9. 52 18. 6 0 19 0. 040 0. O2 5 2 1. 69 0.027 0. 029 0.35 ND 12. 06 18. 09 2 31 0.041 ND 8 4 1. 15 0.024 0. 013 0. 32 ND 20. 0 24. 5 0. 15 0.038 0. 01 15 11 0. 83 0. 017 0. 005 0. 33 0. 31 32. l 20. 2 0. 074 0. 013 0. 24 1, 189 167 N 0TE.N F=N 0 failure at indicated times; E=E1Tatic behavior; N D=Not determined.

As is seen from the above, the invention involves an Resistance to acid corrosion is evaluated by testing in austenitic stainless steel with good corrosion properties, a deaerated 10% sulfuric acid t 86 F. for 96 hours. resistance to stress-corrosion cracking in hot chloride en- The corrosion rate l ulated a mils per year (m.p y VirOIlmentS, and g hot workability- These Properties a unit commonly used within the chemical industry to deare achieved by a carefully balanced composition which 55 scribed general or uniform corrosion, are an indication is especially critical because of the interacting effects of of the teel a idorro ion re i tance, The results of the C rbon a d ge ph ph and nickel, and P acid-corrosion tests are described in Table XI which shows phorus and silicon. A range of optimum compositions is that the corrosion rates of Alloy 78 and Alloy 84 were achieved Within the scope of our invention by the effect of about an order of magnitude lower than Type 304 t i crack-retarding elements (nickel, carbon and silicon) over l nd therefore, superior to it in this property. It is coming the effect of crack-promoting elements (molybfurther noted, however, that the corrosion rate of Alloy denum, phosphorus, nitrogen, chromium and sulfur). Fur- 78 was a re iably lower than that of Alloy 84, thermore, the effects of austenitizing elements (carbon, TABLE XI manganese, nickel and nitrogen) as provided herein are greater than that of the ferritizing elements (chromium, Resistance to corrosion in 10% sulfuric acid molybdenum and silicon) so as to produce an austenitic Alloy: Corrosion rate m py 1 structure. However, as discussed above, there are additional and other surprising advantages of steels in accordance with the invention. Unlike some generally similar but actually diiierent compositions, our steel possesses a high degree of resistance to high-temperature oxidation as Well as to pitting and acid corrosion. The following examples demtion with respect to high-temperature oxidation resistance onstrate these outstanding properties in addition to the also materially contributed to broader application. In a stress-corrosion resistance discussed above. series of tests in which Alloy 78, Alloy 84 and Type 310 M.p.y.-Mils (0.00 1 lllph) per year. The properties of steel in accordance with the invenstainless steel were tested for high-temperature oxidation resistance, the steel in accordance with the invention demonstrated a degree of resistance comparable to Type 310 stainless which is a material frequently recommended for such applications. The test was performed at temperatures of 1800 and 2000 F. for a period of 8 days. Because oxidation rate varies with time, the test results are expressed as weight loss in milligrams per square decimeter in the 8 days of tests. The results are summarized in Table XII and show that at both temperatures Alloy 78 has had an oxidation resistance essentiatlly the same as Type 310 stainless and both Alloy 78 and Type 310 stainless performed appreciably better than Alloy 84 in this test.

TABLE XII.-RESISTANGE TO HIGH-TEMPERATURE It is apparent from the above that the balanced steel composition in accordance with the invention provides superior properties with respect to a number of important corrosion environments. It is unusual for a single steel to possess a combination of properties as described above in connection with steels in accordance with the invention. While some steels have been provided with stresscorrosion resistance, the steel composition in accordance with the invention is the only one known to possess the variety of outstanding properties as described herein. This outstanding combination of properties is achieved by carefully balancing the composition. However, other elements may be incorporated within the steel to perform known functions. Thus, for example, where it is desirable to avoid carbide precipitation as a continuous grain boundary network, elements such as titanium, columbium, zirconium and tungsten may be added to minimize or avoid this problem. These elements can be added in amounts which will tie up carbon and/or nitrogen with the net result of minimizing or precluding a decrease in stress-corrosion resistance while alleviating the continuous grain boundary network problem.

We claim:

1. A balanced austenitic stainless steel having good hot workability, stress-corrosion resistance, pitting and acidcorrosion resistance, as well as high-temperature oxidation resistance, consisting essentially of 0.05 to 0.15% carbon, 1.0 to 2.5% manganese, 0.02% max. phosphorus, 0.0 3% max. sulfur, 1.65 to 3.0% silicon, 15.0 to 19.0% nickel, 15.0 to 19.0% chromium, 0.05 max. nitrogen and 0.05% max. molybdenum.

2. A balanced austenitic stainless steel in accordance with claim 1 having 0.06 to 0.09% carbon, 1.0 to 2.0% manganese, 0.01% max. phosphorus, 0.01% max. sulfur, 1.8 to 2.2% silicon, 15.0 to 19.0% nickel, 17.0 to 19.0% chromium, 0.05% max. nitrogen and 0.02% max. molybdenum.

References Cited UNITED STATES PATENTS 1,745,360 2/ 1930 Antoinette.

2,5 90 ,83 5 4/ 1952 Kirkby.

2,820,708 1/ 1958 Waxweiler.

3,152,934 10/ 1964 Lula -428 3,276,864- 10/ 1966- Loginow 75128 HYLAND BIZOT, Primary Examiner UNITED STATES PATENT OFFICE CERTIFICATE OF CORRECTION Patent No 3, S23 788 August 11 1970 John F Bates et a1 It is certified that error appears in the above identified patent and that said Letters Patent are hereby corrected as shown below:

Column 8, line 36, "closely" should read clearly line 41, "variouss" should read various Columns 9 and 10, Table VII, line 20, under the heading "Cold Worked" "29" should read 2 line 21, under the headin "Cold Worked" "22" should read 2 Columns 11 and 12 Table III, line 6 under the heading "Cold Worked", "E 1,64" should read E 1,649 line 8, under the heading "Cold Worked", "1,013" should read 1,012 line 9, under the heading "Cold Worked", "82" should read 823 Table IX, line 1, under "N", "0.038" should read 0.031 line 2, under "N", "0.021" should read H 0.029 line 3, under "N", "0.039" should read 0.036 line 4, under "N", "0.026" should read 0.028 Column 12, lines 54 and 55, "described" should read describe Column 13, line 11, "essentiatlly" should read essentially Signed and sealed this 23rd day of March 1971 (SEAL) Attest:

EDWARD M. FLETCHER,JR. WILLIAM E. SCHUYLER, JR.

Attesting Officer Commissioner of Patents