US2172427A - Corrosion-resistant ferrous alloys - Google Patents

Description

Patented Sept. 12, 1939 UNITED STATES CORROSION-RESISTANT FERROUS ALLOYS John O. Wulfl, Cambridge, Mass., assignmto The Chemical Foundation,

Incorporated, ew

York, N. Y., a corporation of Delaware No Drawing.

pplication May 11, 1937,

Serial No. 141,921

4 Claims. (Cl. 148-215) This invention relates to the production of corrosion resistant ferrous alloys, more particularly to ferrous alloys which are characterized by a singular resistance to pit corrosion.

As is known, an outstanding advance in the metallurgical field has been the. development of the corrosion resistant steels of the stainless type. This development has been the result of long and patient work and in the broad sense, the product of many individual contributions.

The general method of solving the basic problem of protecting ferrous alloys against general corrosion, in its broadest aspect, has always been recognized. This method is, in substance, the ennobling. of baser ferruginous material by the addition of variant amounts of nobler metals. While workers in the field are not in agreement as to the precise mechanism of this ennoblement of iron, they are in agreement on the ultimate efiect, namely, that certain alloying components 'tend to passivize the ferruginous base metal.

As a result of the research work in this field, there have been developed a large number of ferruginous alloys, which exhibit a markedly increased resistance to general corrosion. These alloys can generally be defined as iron base alloys, which contain effective amounts of passivizing alloy components. Examples of such alloys are the iron chromium series, such as the rustless irons and steels, the ferro molybdenum alloys, the ferro manganese alloys, and similar alloys, 1. e., those which contain suflicient amounts of passivizing constituents to render the iron itself substantially passive. The classic example of this type of product is the standard 18 chromium, 8 nickel stainless.

Passive ferrous alloys of this type, while of increased resistance to general, or surface oxidation, corrosion, are not resistant to all forms of corrosion. As the art advanced and as the use of such alloys was extended to fields where drastic and special corrosive environments were encountered, it was soon ascertained that even the most improved forms of these alloys, such as the stainless steels, were susceptible to special or specific forms of corrosion. Of these, the so called intergranular corrosion and pit corrosion were particularly dangerous because of the 10- calized and insidious nature of the attack.

Each of these forms of corrosion is recognized to be a distinct problem. That form known as intergranular corrosion, in its effect, is somewhat analogous to caustic embrittlement of steels in alkaline solutions. The action appears to take place in the area contiguous the grain boundaries. It was generally accepted that the action wasdue to the tendency of chromium in the grains to combine with carbon to form carbides and thus commensurately to impoverish the grain of the passiv'izing chromium addition, leaving the denuded areas susceptible to attack.

Recent advances in the art, however, have substantially solved this problem of intergranular corrosion. Such steels are now rendered substantially resistant to intergranular corrosion by proper control of the carbon content and heat treatment or by utilizing metals such as columbium or titanium which appear to have a preferential afiinity for the carbon, thus in effect, leaving the chromium in the grains to exert its beneficial passivizing action.

In the present state of the art, therefore, many of the major problems involved in the corrosivity of passivized ferruginous alloys have efiectively been solved. However, a major and peculiarly diflicult problem still confronts the art. This is the problem of effectively inhibiting pit corrosion in these passive iron alloys.

As has been recognized, the phenomenon of pit corrosion is distinguishable as well from general surface corrosion, on the one hand, as intergranular corrosion on the other, by the nature and the locale of the attack. Indeed, up to the present improvement, the rationale of the attack itself was not definitely established. The capricious nature of this form of corrosive attack renders it peculiarly difficult of definition in any case and predetermination in a particular case.

A great amount of research work has been done on this problem. In general, the conclusions which have been reached are of a negative character. Thus, it is generally agreed that this form of corrosion is not due to uniform or g en-' eral chemical reaction with sea water, nor to corrosion by direct oxidation, nor to'electrolyte corrosion due to differential potentials set up by dissimilar metals.

The investigators appear to be in agreement only on the fact that it is due to some other cause. One group attributes the trouble to what is termed contact corrosion under non-metallic substances (such as scale or inclusions) or under consider that electrical cells are established by reason of a differential chemical composition in the individual micro-crystals of the metal itself.

Pit corrosion is distinguished from general surface corrosion in that it is initiated in localized or segregated areas of relatively small cross section and proceeds or continues perpendicularly into the body of the metal rather than longitudinally across the surface of the metal. The development of the depth of pit, after corrosion has been initiated, is generally concluded to be due to the establishment of an oxygenreduction cell. In these circumstances, assuming the localized'pit attack to have been started, and a pore of small cross sectional area to have been established, the electrolyte at the bottom of the pit then tends to become poorerr'in oxygen than that at the top or surface of the pit. In such circumstances, the base of the pit, or oxygen denuded area, becomes anodic to the surface area, thus establishing an electrical cell. The products of electrolytic corrosion, i. e., the ions of the metal of the alloy, then tend to be carried to the cathodic regions where they may be oxidized. If the oxidized metal is deposited at the entrance of the pit, it nevertheless does not serve to protect the metal at the base of the pit so that the action is progressive and, in a sense, autocatalytic.

This phenomenon of pit corrosion is generally encountered in saline media and particularly ,saline media in which, by reason of stagnation of the medium itself, or some surface impurity, a local concentration of chlorine ions is engendered. It appears that in these circumstances electrolysis sets'in with the formation of soluble products which are leached away, establishing a locus for a potential pit, after which, by the progressive action described, the

corrosion continues of the metal.

There are many factors which may possibly play a prominent role in the initiation of such pits. These factors may broadly be classified as metallurgical factors and mechanical factors. Specifically, such factors may include: the influence of the alloy concentration gradients, or the differential constituency of the grains of the metal; the influence of dissolved or occluded gases; the influence of the sonims, or nonmetallic inclusions; the influence of carbides; the influence of work strains; the influence of the products produced as a result of working, such as ferrite; and the influence of working, as such. It is apparent that certain of these several factors may be interrelated in their ultimate effect on the susceptibility of the steel to pit progressively into the body corrosion..

As a result of extensive experimentation in this fleld, it has been determined that of these factors the character of the cold work to which the article is subjected during the fabrication processes largely determines its susceptibility to pit corrosion. It was further determined that this susceptibilityis due mainly to the mechanical eflects of cold work, i. e., to the institution gbimperfections in the surface, such as cracks, pores, fissures, and the like, and that a presumptively important metallurgical incident of cold working, namely, the production of alpha ferrite, does not, per se, have any particular "effect on the susceptibility to pit corrosion.

Similarly, it was found that such factors as the presence of carbides and the diflerential strains 1: in the surface of the material do not exercise samples were cut.

a major influence on the institution of pit corrosion.

These conclusions, and the method of immum'zing the potentially passive steels, were developed as a result of a series of tests in which, insofar as possible, each presumptively important factor was segregated and its individualeffect determined. 1

The tests were carried out upon specimens which were of established uniformity. For this purpose a forging was produced from a stainless steel ingot whose actual composition was 18.13 Cr, 8.94 Ni, 0.08 C. The forging was hot rolled to produce a sheet and from this sheet of the specimen with an aqueous solution containing 10% of FeCl: and 20 cc. of 2.45 N.HC1,

per litre. The solution was caused to circulate through a. thermal pump system under such circumstances that a drop of the solution was maintained in contact with the specimen to be tested fora period of four hours. With this method, it was found that, as a general rule, there was little general surface corrosion but that pits were rapidly developed in a steel which is susceptible to pit corrosion.

"The samples which were subjected to this type of test were then examined to determine the loss of weight 'per unit area, the total number of pits per unit area, the number of large pits per unit area, and the depth of the largest pits. After considerable experience with the method, it was determined that in normal circumstances the depth of the deepest pit is fairly constant for all steels which are susceptible to the attack. For quickand reasonably accurate determinations therefore, it is found that it was suflicient "this particular factcr, magnetic measurements on cold worked samples were made. The basis of this type of test and evaluation is the known fact that there is a difference in magnetic permeability between strained and unstrained sections of the alloy, the unstrained sections being characterized by higher permeability than the strained portions.

In carrying out these tests, a number of uniform specimens were subjected to difler'ent de- The several samples were grees of cold working. These specimens were then individually tested as to magnetic properties by suspension in a large magnetic field of known inhomogeneity. The testing apparatus was so constructed that the specimens were surrounded by a non-inductively covered electric furnace so that. measurements could be taken under different thermal conditions up to 800 C.

As a result of numerous tests of this character,

on specimens which had been cold worked to different degrees, it was-found that there was a marked increase in magnetism for the severely worked specimens when such specimens were annealed for a long period of time below 450 C. It was further found that if these specimens were heated much above this temperature two major changes occurred; a phase change from ferrite to austenite, and some carbide precipitation.

A large number of these differentially strained specimens were submitted to the accelerated corrosion tests, of the character described, to determine the difference in pit corrosion susceptibility due to the surface strains. Similarly, equivalent specimens which were strained by cold working and in which the strains were relieved by annealing were tested in the same manner. It was found as a result of these tests and direct comparison of the pit corrosion of thetwo groups that there was no perceptible difference. In other words, there was no material difference in pitting suscept'bility between the specimens which had been strained and the same type of material in which such strains were relieved to the optimum extent by a prolonged low temperature anneal.

Similarly, a series of experiments was conducted to determine, insofar as possible, the effect of carbides on the susceptibility of the stainless steel to pit corrosion. As will be appreciated by those skilled in the art, this type of determination is exceedingly difficult to make. It is known that the precipitation of carbides in the temperatur range of from 500 C. to 800 C. affects the corrodibility of stainless steel, and that the precipitation of'such carbides is largely responsible for the intergranular corrosion "of such steels. It is also known that if such stainless steels are heated within this temperature range, the carbides may precipitate within the grain as well as in the grain boundaries.

The identification of the carbide particles within the grain of the alloy is obviously quite diiiicult, even under the best conditions of microscopy. However, in those instances where the particles could be microscopically identified, an examination of the specimen, while in contact with the corrodent, showed quite conclusively that the corrosion initiated around the whole grain boundary more often than directly at the carbide'inclusion. In some cases, the corrosion did initiate within the grain, but even in these instances, the attack proceeded immediately around the whole periphery of the grain Such microscopic examination, while not of itself conelusive, would appear to indicate that inclusions are not a determining factor in effecting the locus of the corrosive attack.

This assumption is considerably strengthened by correlation with observations made during the magnetic measurements. In the course of such measurements it 'was observed that when severely cold worked specimens were heated in a furnace and their magnetic characteristics observed and recorded, in every case there was a decided break or kink in the curve, at a temperature slightly below 200 C. It is reasonableto attribute this break in the curve to the presence of iron carbide which has a Curie point at 200 C.

- Furthermore, it was found that the magnitude of this depression or break was substantially directly proportional to the amount of ferrite produced by the cold rolling operation and the amount of carbon present in the alloy. It is, of course, diflicult to check the presence of the carreasonable to assume, however, that such carbide may be formed since f errite holds considerably less carbon in solution than austenite. Likewise, since carbon has little mobility at the temperature of the cold working operations, it is expectable that when segregated from the gamma solid solution it is quite likely to combine with iron to form the carbide.

From this series of experiments, therefore, it is quite evident that carbide inclusions, as such, are not an important factor in determining the initiation of pit corrosion.

Certain investigators have suggested that pit susceptibility of these passive ferrous alloys may be due to electrochemical potential differences set up between the alpha and gamma phases. As is known, any cold working on austenitic 18-8 steels results in the product of alpha ferrite. The amount of ferrite produced is not strictly directly proportional to the degree of cold working, for even under the most severe working conditions it is impossible to convert a sample into pure ferrite.

In order to determine the role played by the ferrite in pit corrosion, a special series of experiments was run. In these experiments the relative quantity of the ferrite produced, atthe surface of the specimen, was quickly evaluated method, quite accurate determinations could immediately be made and a tween the quantity of ferrite produced and its effect on pit corrosion could be established. It is particularly to be observed, at this point, that in respect to studies of the causes of pit corrosion a clear distinction must be maintained between the total quantity of ferrite in the host austenite and that exposed or potentially available at the surface.

The analysis of the quantity and distribution of the ferrite at the surface of the test samples was made, as indicated, by a special method. For this purpose, a given area vof the surface was covered with a colloidal suspension of gamma phase F8203. Under the microscope, there was no change in the surface of the specimen, except, of course, a noticeable change in color. When, however, the specimen was placed on a pole of an energized electro-magnet, the colloidal particles are attracted to and precipitated on the magnetic (ferretic) areas ofthe surface.

A series of samples was subjected to different degrees of cold work varying from about 2% to 50% reduction. Upon examination of these, utilizing the method described, it was found that the total amount of ferrite formed was approximately proportional to the degree of cold work. Upon a check comparison by X-ray intensity measurements, it was found that the 100% cold reduction samples were composed of substantially 8% of ferrite.

These several samples were subjected to the- It was found that,

accelerated pit corrosion test. as an approximation, the degree or extent of pit corrosion is not dependent upon the amount of cold work, or more strictly, upon the amount of ferrite present, for reductions below approximately 40%. Above this, in the more severely worked samples, there appears to be a direct relationship between the amount of ferrite and the degree of pit corrosion.

This conclusion however could not immediately by a specially developed method. By using this direct comparison beof the colloid precipitated on the 2% reduction able.

specimen was substantially identical with that of the 50% reduction specimen. It appeared then, that while susceptibility to pit corrosion is a function of cold working, it is not directly or necessarily related to the amount of ferrite, as such, that is present on the surface of the steel.

To check this conclusion, a special series of experiments was conducted. In these, a number of samples were subjected to different degrees of cold work. Before submitting the specimens to the accelerated corrosion test, they were each pickled in asolutionof hydrochloric and sulphuric acid to largely etch away the immediate surface metal, After the specimens were thus treated, they were subjected to the corrosion .test. These tests showed a marked increase in pit susceptibility with increase in cold'working, particularly above 40% reductions. Since a considerable amount of surface ferrite had been removed prior to the corrosion tests, it is clear that the amount of ferrite does not, per se, determine the susceptibility to pitting.

It then became evident that the susceptibility of these steels topit corrosion was due to the mechanical effects of the cold working, that is, to the checks, cracks, fissures, or other surface irregularities or imperfections caused by the mill operation. It is evident that these macroscopic or submacroscopic depressions in the steel permit the accumulation or stagnation of the saline medium, thus permitting an'attack by the chlorine ions. In a series of additional experiments, the predominant'importance of such mechanical defects in the institution of pit corrosion was corroborated. It was further found that the susceptibility of the article to pit corrosion was not necessarily directly proportional to the degree of cold work, although with cold reductions above approximately 40% the susceptibility is markedly increased. Since such drastic workingconditions tend to the development-of surface imperfections, such findings are expect- It was then determined that if the specimens,

of whatever degree of cold work, were subjected to a. treatment-whereby the susceptibile surface was so modified as to remove these surface imperfections, or potential foci of pitting, the resistance of the article was remarkably increased.

With "the salient factor of pit susceptibility determined, it will be appreciated thata number of specifically different methods may be developed to render the steel substantially immune to pit corrosion.

In order to produce steel articles of optimum resistance to all types of corrosion, a special method of treatment was developed. In brief, this treatment comprises a high temperature vacuum anneal.

In carrying out the process, the plate, sheet, or preferably the fabricated articlein its final shape,

is placed in any suitable vacuum furnace. It is then heated to the austenitic temperature range,

that is, from 1050 C. to 1200 C., while under a vacuum preferably of the order from 10 mm. to 10* mm.- When such treatment is effected for 5 brief period of time, the treated article possesses a remarkable resistance to pit corrosion. When specimens are subjected to this vacuum anneal and are then tested by the accelerated corrosion method, there is no evidence of pit formation.

It is particularly to be observed that the improved vacuum anneal is effectuated at temperatures above the carbide forming range. With this method, as will be appreciated by those skilled in the art, the formation of the deleterious carbide inclusions is avoided and the treated article is rendered resistant not only to pit corrosion but also to intercrystalline corrosion.

The actual mechanism involved in the resurfacing and'consequent immunizing of the steel is not susceptible to precise definition. As can be readily appreciated, the temperature of the scratches, and other surface imperfections, The I sharper protuberances are thus physically removed by absorption as well as by evaporation. Within broad limits, the metal evaporated from the surface gives the same analysis ratio of metallic constituents as the alloy itself.

Such a high temperature treatment under the vacuum also effectively serves to withdraw ad; sorbed, occluded and dissolved gases, Itis quite likely that the aspiration of the gases from within the metal matrix also serves further to increase the mobility of the mass, thus enhancing the plasticity of the surface sections and facilitating the self-healing action. It is also quite likely that in their transpiration these occluded gases may chemically combine with and entrain some of the surface metal from the sharper'angles or spicules of the surface which compounds are sub-.

sequently sublimed or evaporated, thus tending to smooth out these areas, which in their nature would serve as foci of chemical or electrochemical action.

This improved heat treatment underyacuum also serves to break down the submacroscopic the thermal conditions of the treatment the mechanical strains, of course, arerelieved.

Whatever may be the precise reaction and the physical and physicochemical changes taking place, it is found as a matter of actual fact and test that this general class of potentially passive ferrous alloys are generally improved and their resistance to pit corrosion eminently increased by such a -vacuum annealing in the austenitic range.

It is to be observed that the improved efiects herein' discussed are not dependent upon ultra high vacuum. The amazing improvement in pit resistance may be secured by utilizing a lower vacuum, in other words, the vacuum may be varied over a relatively wide range.

The time of treatment, similar y, is not critical. While tosecure the best results, a vacuumizing' in the recrystallization range for a period of a half hour or more is recommended, a markedly improved product may be produced in a much shorter time. Thetime and vacuum conditions are interrelated and may be varied. as will be understood by those skilled in the art. The extent or duration of the treatment will, of course.

be established or controlled by the degree of surface healing or immunization that is required. When the cold reduction is drastic. or where the declivities, cracks or other surface blemishes are deep seated, by reason of imperfections in the rolls and the like, the treatment should be comobtained by producing stainless steels of the 18--8 type in which are included intercrystalline corrosion-inhibiting agents such as columbium and titanium. When such steels are treated according to the present invention, to markedly increase the resistance to corrosion, the resulting product presents a novel and improved article being characterized, as noted, by a striking resistance not only to intercrystalline corrosion but also to pit corrosion. When operating with such steels, which are inherently resistant to intercrystalline corrosion, i.e., those which contain columbium or an equivalent metal, a greater flexibility in the thermal conditions of the operation is permitted. With such a type of steel, the danger of chromium depletion by heating in the carbide forming range is substantially eliminated and hence the temperature of the anneal need not necessarily be maintained above the carbide forming range, but only at the temperature necessary to secure the desired resurfacing action.

It is particularly to be observed that while the present method of improving the pit corrosion resistance is intended to be employed in the fabrication cycle, it is also available for the improvement of articles already fabricated and in fact 7 articles which have been in use. Since the treatment requires the use only of standard metallurgical apparatus, the articles to be improved, if desired, may be subjected periodically to the treatment, particularly those articles which are submitted to the more drastic pit corrosion environment.

It is clearly to be understood that while the high temperature vacuum anneal has been described as one method of immunizing steel against pit corrosion, this has been given didactically to pointedly illustrate the principles involved and not as the sole method of efiectuating these principles. The improvements herein are conceived to reside in the discovery of the effective cause of pit corrosion and the development'of a generic method of minimizing it, and particularly in the production of novel types of steels characte: ized by a substantial immunity to pit corrosion.

I claim:

1. A method of treating chromium-containing ferrous alloys and steels to render such alloys substantially immune to pit corrosion which comprises, cold deforming the alloy to a desired ultimate shape and then heating the article while under a vacuum and above the carbide forming temperature for a period of time sufficiently prolonged to eliminate the imperfections on the surface which form potential loci for the initiation of pit corrosion.

2. A method of producing stainless steel articles of the 18 chromium, 8 nickel type, to impart thereto a marked resistance to intercrystalline and pit corrosion which comprises, forming an article from the steel, and after formation subjecting the article while maintained under a vacuum to an annealing treatment at a temperature of from substantially 1000 C. to 1200 C. to eliminate carbides while imparting to the metal sufficient mobility to substantially eliminate surface checks, cracks, and the like, and to establish a submacroscopically smooth surface on the article whereby pit corrosionengendering foci are eliminated.

3. An improved stainless steel article of the 18 chromium, 8 nickel type which has been simultaneously vacuumized and annealed at temperatures of from substantially 1000 C. to 1200 C., after the final forming step.

4. A method of producing stainless steel articles from steels of the type of chrome-nickel steels, which steels are characterized by a marked resistance to intercrystalline. and pit corrosion which comprises, forming an article from a stainless type of steel which contains an intercrystalline corrosion-inhibiting agent having the characteristics of columbium, and after formation, subjecting the article to an annealing treatment at a temperature of from substantially 1000 C. to 1200 C. while under a vacuum, to

establish a substantially submacroscopically smooth surface on the article.

JOHN C. WULFF.