US2175850A - Powder metallurgy - Google Patents

Description

Patented Oct. 10, 1939 UNITED STATES POWDER METALLUBGY Raymond L. Patterson and Claus Guenter Goetzcl,

New York, N. Y., assignors to Hardy Metallurgical Company, New York, N. Y., a corporation of Delaware No Drawing. Application March 1'1, 1939, Serial No. 262,565

16 Claims. (Cl. '15- 22) This invention relates to the powder metallurgy of steel and contemplates improvements in the manufacture of steel objects by processes involving the compression and heat treatment of metal powders. The invention aims to provide a novel steel powder especially suited for such manufacture and a novel process for making such powder.

The manufacture of metal objects by comm pressing steel powder or a mixture of iron powder and carbon has been proposed heretofore, but has been attended by a variety of difficulties. Thus, the manufacture of steel objects by heating and compressing powders made by comminuting massive steel is uneconomic because of the high cost of the powders and unsatisfactory because such powders do not tend to cohere strongly and hence result in the formation of weak briquettes which crumble readily. Steel powders derived by subjecting iron powder to a carbonizing treatment are cheaper than those derived by comminution, but are also objectionable, in that upon compression they fail to cohere properly and form weak briquettes. The use of a mixture of iron powder and carbon instead of steel powder is also unsatisfactory in that a large but variable proportion of the carbon tends to be lost during heat treatment while the residual carbon tends to form graphitic lamina- 30 tions which constitute planes of weakness in the final product.

Another proposal for the manufacture of steel by powder metallurgical methods involves the compression and heat treatment of a mixture of 5 iron powder and finely divided high-carbon ferro-alloy combination, such as ferrochrome, obtained by comminution of a massive form of these brittle substance. The carbon is thus added as an intermetallic compound of iron, chromium 0 and carbon. With this process, however, there are three outstanding disadvantages: (1) Although some migration of the carbon from the ferrochrome particles into the iron could be observed by microscopic examination of compressed 45 and heat treated aggregates of specimens, this migration was neither sufliciently uniform or controllable to assure a homogeneous product; (2) the process is not applicable when steels containing no chromium are sought; and (3) the ferro- 50 chrome carbides in the product tend to decompose if heated to temperatures in excess of about 1135 C., forming ferrite, some pearlite, and free carbon inclusions. In order to secure adequate diffusion of chromium throughout the compressed to aggregate, a temperature in excess of 1250 C.

is necessary. Since free carbon is released at a lower temperature, it is apparent that adequate chromium diffusion is only attained at the expense of throwing down undesirable free carbon in the resulting steel object.

As a result of our investigations, we have succeeded in overcoming the difliculties attendant upon the above described prior proposals and have developed steel powder which coheres adequately upon being compressed and in which the 1 carbon is so combined that it does'not tend to form graphite upon heating to excessively high temperatures, say in excess of 1300 C. This product of our invention comprises finely divided steel particles having interior zones in which combined carbon is present in relatively high concentrations and exterior zones substantially free from carbon. Thus, the carbon in the interior of the particles is combined as pearlite or cementite or a mixture of the two, and the exterior of the particle is substantially ferritic in character. Iron carbide and its eutectic and eutectoid mixtures with iron are relatively brittle and glassy in character, and it is this quality which probably accounts for the fact that heretofore customary steel powders do not stick together adequately even when compressed under exceedingly high pressures. However, the glassy and brittle character of the iron carbide in steel particles made in accordance with our invention does not prevent strong coherence of the particles when these are compressed together, probably because the zones in the particles which contain the carbide are substantially enveloped by relatively soft and malleable ferrite. Whatever be the explanation, the fact remains that steel particles in which the combined carbon is concentrated within the particles and which have a substantial proportion of their exterior portions in the form of ferrite may be compressed into strong briquettes. Such briquettes withstand rough handling and upon heat treatment attain substantially better physical properties than have been attained heretofore in the powder metallurgy of steel. Moreover, the heat treatment of such powders may be carried to temperatures at which chromium and other difiicultly difiusible I alloy ingredients become thoroughly distributed throughout the resulting steel body without danger of free carbon deposition. Hence, the pow-" ders of our invention 'ofier outstanding advantages in the manufacture of alloy steel objects containing nickel, chromium, and the like.

We have discovered, further, that steel powders of the above described character can be made 'rupted. before the combined carbon to such partial 'decarburization may be made by subjecting ironpowder (derived by any conven-' lent method such as electrolysis or a sponge iron process) to the'action of a hot carburizing atrn'osa I bonized to a substantial; depth and so contain combined, carbon within their interiors.- Theresimply and easily bysubjecting steel powder par ticles having combined carbon distributed therethrough to the action of a decarburizing atmosphere at a relatively high temperature (but below the melting point ,of the particles) until theouter portions of the particles havebecome'substam' tially decarbonized, decarburizing being inter- I v 7 from the interior of the particles is removed.

The steel powder particles which are subjected phere (preferably gaseous and so controlled that substantially no free carbon or vsootis deposited in theiron) until the particles have been car-j after, the carboinizingatmosphere is replaced' by a decarbonizing atmosphere, such, for example, as

l hydrogen, and the, hot particles are maintained the following detailed description or the pro i duction and utilizationof the steel powder. r

o As indicated hereinbefore, iron powdermaybe' be more thoroughly understood in the light of obtained from any convenientsource, but ln the interests of thermal efliciency andcontinuity J of operation' it is desirable to derive it in a hot I condition by reduction offinely divided iron oxides The hot-iron thusobtained is preferably carburizing agent and subsequentlyto that'of a decarburizing; agent in the same vessel in which it was formed and subjected to the action of a without intermediate cooling, this operation and subsequent partial decarburization being substantially continuous.

Although reduction, carburizing and partial decarburizing may be carried on without agitating the powder, agitation is preferable and to this end the three operations may well be carried out in sequence in a rotatable muflle provided with means for supplying the various atmospheres required. Thus, means should be provided for supplying a reducing agent (say a current of reducing gas, such as hydrogen or carbon monoxide) to the muflle during the reduction of the iron oxide to iron powder, a carburlzing agent (say a current of gaseous hydrocarbon, such as methane, diluted with a gas that is substantially inert with respect to iron) during the transformation of the iron into a steel powder, and a decarburizing agent (such as a current of hydrogen containing a small proportion of water vapor) during the reformation of the shell of ferrite on the outside of the steel powder particles. These various atmospheres may be established by passing currents of the respective gases through the muille while the latter is rotated and while heat is supplied to maintain the powder at reaction temperature which may be about 900 C. in all three steps of the process. To consider the reduction of iron oxide in greater detail, Chataguey iron concentrate consisting principally of magnetite (F8304) and containing 67.7% Fe was subjected to dry hydrogen at three atmospheres pressure for about 4 hours imcsp i e is to 9-9%'Fe and suitable for steps of the process. The original iron concen- -l00 mesh (Tyler Scale);

Reduction of the oxide proceedsfrom the out side,- so the operation .must be sufiici-entlyprolonged to assure substantially complete reduction and the elimination of, cores of residual oxide from the particles. Other factors remaining equal, the larger the particles the longer the reactionoperation must be prolonged.

was. all minus about name was all minus vmesh (Tyler Scale) and. the resulting iron powder ,If iron, powder alreadyavai lablethe preliminary reduction step may be'largelyelimie ated, although. care shouldbe taken to remove I any rust from the surface of the particles. This may likewise be done in the rotating; mode in a a ph r of d y hydro en, preferably at a 'temperature'of about 900C. and with a treat merit-time of 30 minutes or less.

1 Followingthe reduction step, a oarburizing at-' 0 created in the mume. If initial re? 'duction' has been carried out with I suitable atmosphere is created by introducing a relatively small propcrtiono'f a-carbonaceous gas f (such as methane, ethane, propane; butane;ethy- I lene, or even city gas) into the entering. stream, of hydrogen, care being. taken not to increase the concentration of the carburizing agent to the pointy where freecarbon is deposited in the iron. Generally speaking, the hydrocarbon gas h dro e a should constitute less than half of the-total gas (by volume) and it is safer to employmuch lower concentrations. Good resultshave been attained .with mixturesof various carburizing gases and the following proportions at atdry hydrogen in Butane and h'ydrogenin ratios of; 1:18, 1:11, 1:8 ethane andhydrogen in ratlosof lzlS, 1:10 Ethylene and hydrogen in ratios of 1:18, 1:10

City gas in an undiluted condition or diluted with up to one volume of hydrogen and at a pressure of 2 to 3 atmospheres also gave good results without deposition of soot. A partially combusted natural gas containing a substantial proportion of diluent gas may also be employed.

If nitriding of the steel powder is desirable or at least not objectionable, nitrogen or cracked ammonia may be employed as a diluent instead of hydrogen.

The rate of carburization and the proportion of pearlite and cementite formed in the powder particles can be controlled by varying the ratio of carburlzing to diluent gas in the carburizing mixture.

In carburizing, a temperature of about 900 C. in the muflle is satisfactory, and permits the absorption of fixed carbon by the iron powder at a rapid rate even with relatively dilute carburizing gas mixtures. The duration of the carburizing treatment will depend upon the temperature, the

concentration of carbon in the carburizing mix-,

through. Otherwise, there is danger that subsequent decarburization may remove more carbon than is desirable. Thus, batches of "Nori iron powders (98-99% Fe) obtained by a sponge iron process were (after a preliminary deoxidation treatment in dry hydrogen at 900 C. for half an hour) subjected respectively to treatment in the above-described mixtures of butane, methane and ethylene with hydrogen for periods ranging from one to two hours at a temperature of about 900 C., and were found after this treatment to be thoroughly carburized throughout and without substantial proportions of residual ferrite on the interior of the particles. Depending upon the time of treatment and the concentration of carburizing agent employed, the powder particles were predominately pearlitic or cementitic in character. Free carbon was substantially absent in all cases.

Following carburization, the stream of carburizing mixture passing through the retort was replaced by a stream of commercial hydrogen, and this gas was passed in contact with the powders (which were maintained at about 900 C.) for periods ranging from half an hour with some batches to two hours with others. Thereafter, the powders were allowed to cool to room temperature in the hydrogen atmosphere. Some of the powders from each batch were made into slugs by compression under a force of 50 tons per sq. in. and subsequently heat treated. Slugs to which no alloying ingredients were added were maintained in hydrogen at 1000 C. (i. e., below the melting point of the ingredients) for two hours to bring about diffusion welding of the particles. Slugs to which nickel powder and lowcarbon ferrochrome powder were added were heat treated in dry hydrogen for eight hours at substantially constant temperatures ranging from 1175 C. to 1275 C.

The degree of decarburization and the thickness of the resulting shell of ferrite on the particles can be controlled by varying the time of treatment or by diluting the decarburizing gas (say hydrogen) with a relatively small proportion of a gas of opposing tendency (say methane).

Examination of the uncompressed powder, the compressed slugs, and the heat-treated slugs gave the following results: In every instance it was found that the powder particles from the decarburization treatment consisted essentially of cores containing combined carbon as pearlite or cementite or both, the outer portions of the particles being essentially ferritic. Those particles which had been subjected to more intense carburization had cores which were principally cementite; those which had been subjected to relatively mild carburizing treatment had cores that were essentially pearlitic. The amount of ferrite present and the thickness of the ferritic formation in the outer or surface portions of the particles varied in proportion to the duration of the decarburizing treatment.

The particles in each batch corresponded to each other in metallographic analysis. In some batches substantially all of the particles contained ferrite, cementite and pearlite. In other batches only ferrite and cementite were present and in still others the individual particles were substantially composed of ferrite and pearlite only. In many instances, ferrite zones were juxtaposed with cementite zones without intervening zones of pearlite. It will be recognized that this is an unusual metallographic structure in steel and one not found in ordinary steel products.

The particles were very irregular in shape, due to the fact that they were somewhat pseudomorphic and retained in some degree the irregular configuration of the original iron oxide particles from which they were formed. In some instances irregularity of particle shape was accentuated by the development of pores which projected deeply into the particles. The exposed surface of the particles including the pore surfaces were essentially ferritic, but the shape of the core portions containing the combined carbon was irregular, and the depth of the outer ferritic portion was not uniform. In some instances, where carbon had not been permitted to penetrate completely through the particles during carburization there were ferritic inclusions within the cores.

The total carbon content of various batches ranged from 0.1% to 2%, as determined by chemical analysis and checked by estimate of the relative areas occupied (in polished cross sections of the particles) by ferrite, pearlite (0.8% C.) and cementite (6.3% C.).

In all instances it was found that the decarburized powders cohered readily when subjected to compression, and the resulting slugs manifested no tendency to crumble. The powders in fact, behaved in compression very much like pure iron. For comparative purposes, some of the "un-decarburized powders from various batches were subjected to compression. These did not cohere properly and did not give briquettes which giould be safely handled in a commercial opera- The heat treated steel slugs to which no alloy ingredients had been added were subjected to microscopic examination and physical test. It was found in all cases that no trace of original grain boundaries remained, and that thorough dispersion of combined carbon had occurred so as to yield a fine grained massive steel comparable in structure to that obtained by processes involving fusion.

In cases where the total carbon content exceeded 0.8% the steel was an exceedingly intimate mixture of pearlite and cementite; in lower carbon specimens ferrite was present but the structure was equally fine in grain. Hardness of the slugs varied as the carbon content and ranged from 48 to Brinell, which is comparable to hardness of steels made by conventional fusion processes. Tensile strength of the slugs as taken from the heat treatment operation varied from 40,000 to 48,000 pounds per sq. in., but after oil hardening and back-drawing the tensile strengths of such slugs increased greatly and varied from 68,000 to 78,000 pounds per sq. in.

The batches of powder employed for making alloy steels contained combined carbon in proportions ranging from 0.2% to 0.5%. These were mixed with low carbon ferrochrome and nickel powders to match S. A. E. compositions for die steels (i. e., S. A. E. steels designated as No. 3250, No. 3325, No. 3340, No. 3415 and No. 3450) and (as indicated hereinbefore) were compressed at 50 tons per sq. in. and heated for eight hours in dry hydrogen to which was added 1% by weight of methane to counteract the decarburizing tendency of the hydrogen. After oil hardening and. backdrawing the resulting slugs or bars showed tensile strengths ranging from 90,000 to 137,000 pounds per sq. in. The majority of bars tested had tensile strengths exceeding 110,000 pounds per sq. in., the average S. A. E. specification for hardened Cr--Ni steel castings of similar chemical composition.

Examination of the microstructure of the alloy steel bars thus produced showed that 1175 C. is too low a temperature to permit adequate diifusion of chromium and nickel, but in the bars treated at 1275 C. the chromium and nickel were completely dissolved and the microstructure was fine grained and uniform. There was no evidence of free carbon inclusions, contrary to the case when carbon is added as high-carbon ferrochrome and the heat treatment is conducted at a temperature such as to permit adequate diffusion of the chromium. This is because the iron carbide in the steel powders of our invention is such that it merely becomes austenitic at high temperatures and suffers no decomposition, whereas the intermetallic chromium-iron-carbon composition of high-carbon ferrochrome is unstable at the temperature at which straight iron carbides change to austenitic form.

In preparing a steel powder for the manufacture of a steel of particular metallographic structure, regard should be had to the total combined carbon content of the steel powder without particular reference to whether it is peariitic or cementitic in character. In other words, if a pure pearlitic steel object containing substantially no free ferrite or cementite is desired, the powder should be prepared to have a total carbon content of 0.8%. Since there is a shell of ferrite on the powder particles, the cores must of necessity contain some cementite in order to raise the total carbon content to 0.8%. However, heat treatment of aggregates compressed from such powder causes the fixed carbon to diffuse so that the free ferrite and free cementite merge to form pearlite. Such difiusion occurs rapidly at a temperature of 1000 C. or more, i. e., while the aggregates remain in a solid condition.

It is, in fact, desirable to produce steel powders with purely cementitic cores, whether or not cementite is desired in the final steel, for in this way the combined carbon can be concentrated in the smallest possible space within the particles leaving a larger volume of the coherent ferrite to, further bonding of the particles under compression.

For optimum results, from the standpoint of coherence of the powder when compressed, the total carbon content of the powder particles should not exceed about 2%. Ferrite coated steel particles with cementite cores, however, are quite coherent even when they contain 5% by weight of combined carbon, but in general a lower carbon content is recommended. Coherence of the particles begins to decrease rapidly as their carbon content is increased above this figure, since such increase is accompanied by a decrease in the thickness of the ferrite on the outside of the particles. However, it is seldom necessary to produce steel objects containing even as much as 2% combined carbon, at least by powder metallurgical methods so that no serious practical limitation is imposed.

If desired, the steel powder of our invention may be made into alloy objects by mixing it with finely divided alloyingredients, such as vanadium, manganese, chromium, nickel, cobalt, molybdenum, and silicon, either in the elemental state or as ferrc-alloys, such as ferro-manganese, etc.

We claim:

1. A process for making steel powder which comprises subjecting finely divided iron oxide to the action of a reducing agent at an elevated temperature to produce hot iron powder. subjecting the resulting iron powder particles while maintaining them at an elevated temperature to the action of a carburizing agent until combined carbon has formed to a substantial depth below the surface of said particles, and partially decarburizing the resulting steel powder particles while maintaining them at an elevated temperature by subjecting them to the action of a decarburizing agent until a substantial proportion of carbonfree iron has been restored in the portions of the particles adjacent the surface thereof.

2. A process for making steel powder which comprises subjecting finely divided iron oxide to the action of a hot gaseous reducing agent while agitating the oxide until it has been reduced to finely divided iron particles, subjecting said iron powder particles while maintaining them in the hot condition in which they were formed to the action of a gaseous carburizing agent until combined carbon has formed to a substantial depth below the surface of the particles, and thereafter subjecting said particles while maintaining them in the hot condition to the action of a gaseous decarburizing agent until the particles are par tially decarburized.

3. A process for making steel powders which comprises subjecting hot iron powder particles to the action of a carburizing gas until carbon has been deposited on the interior of the particles in the form of a substance selected from the group consisting of pearlite and cementite, and thereafter subjecting said powder to the action of a gaseous decarburizing agent until the outer portions of the particles are substantially decarburized but combined carbon remains in the interior of the particles.

4. A process for making steel powders which comprises subjecting hot iron powder particles to the action of a carburizing gas diluted with a relatively large proportion of a diluent gas that is relatively inert with respect to iron until combined carbon has been deposited in the interior of the particles, and thereafter subjecting said particles to partial decarburization by exposing them in a hot condition to the action of a hot gaseous decarburizing agent.

5. Process according to claim 4 in which the ratio of pearlite to cementite formed in the particles is controlled by varying the ratio of carburizing gas to diluent gas.

6. Process according to claim 4 in which the diluent gas is selected from the group consisting of hydrogen and nitrogen.

7. Process according to claim 4 in which the decarburizing gas is hydrogen.

8. Process according to claim 4 in which carburizing of the iron particles and partial decarburizing are conducted at a temperature in the neighborhood of 900 C.

9. In a process for making steel objects, the improvement which comprises subjecting steel powder particles containing combined carbon substantially throughout to the action of a hot decarburizing atmosphere until a shell of substantially carbon-free iron has been formed on the outside of said particles, and compressing and heat treating the resulting powders to form a coherent mass.

10. In a process for making steel powders, the improvement which comprises subjecting iron powders to the action of a hot gaseous carburizing agent mixed with a diluent gas that is relatively inert with respect to iron until combined carbon has been formed to a substantial depth in the powder particles, controlling the ratio of carburizing agent to diluent gas to prevent the deposition of soot in the powder and to control the degree of carburization of the powder, and thereafter partially decarburizing the powder by subjecting it to the action of a hot gaseous decarburizing agent until shells of ferrite have been ticles leaving cores containing combined carbon within said particles.

11. In a process for making alloy steel articles containing difficultly diffusibie metallic a1- loy ingredients selected from the group consisting of chromium and nickel which involves compressing steel powders into a coherent mass, the improvement which comprises subjecting steel powders containing combined carbon to partial decarburization to DIOdllCBIShEHS of ferrite on the outside of the particles, mixing the powders thus treated with powders containing one or more finely divided alloy ingredients of said group, compressing the mixture to form a coherent mass and heating the mass to a temperature of at least 1200 C. but below the melting point thereof.

12. In a process for making steel objects, the improvement which comprises subjecting steel powder particles containing combined carbon substantially throughout to the action of a hot decarburizing atmosphere until shells of substantially carbon-free iron have been formed on the outside of said particles but leaving cores containing iron carbide on the inside of the particles, thereafter compressing the powders into a formed on said powder par-.

coherent mass and heat treating the coherent mass until combined carbon from the interior of the particles has diffused across. the original particle boundaries in the mass.

13. Steel powders comprising steel particles having iron substantially throughout and having a substantial proportion of carbon combined with the iron and concentrated in the cores of the particles, said particles having exterior ferritic zones substantially free from carbon.

14. In a process for making steel objects the improvement which comprises subjecting steel powder particles having interior portions containing combined carbon and exterior zones of substantially carbon-free iron to compression to form a coherent mass and heat treating the coherent mass to weld the particles together.

15. In a process for making steel objects the improvement which comprises subjecting to compression and heat treatment to form a coherent mass steel powder particles having interior portions containing combined carbon and exterior portions of substantially carbon-free iron.

16. In a process for making steel objects the improvement which comprises compressing and heat treating, to form a coherent mass, steel powder particles that have been partially decarburized so as to have cores containing combined carbon' and exterior portions substantially free of carbon.

RAYMOND L. PATTERSON. CLAUS GUENTER GOETZEL.