US2142694A - Process for reducing chromium compounds - Google Patents

Description

Patented Jan. 3, 1939 PROCESS FOR REDUCING CHROMIUM COMPOUNDS Application August 3, 1937, Serial No. 157,154

17 Claims.

'I'his invention relates to a method of reducing a. chromium chloride to chromium, and, more particularly, to a method of reducing chromium trichloride with hydrogen to provide chromium and especially chromium as sponge chromium.

A definition of sponge chromium The term sponge as applied to a metal has a very denite meaning. The term has been herem tofore loosely applied in the art as describing metals heretofore produced by low temperature operations without regard for the actual metal produced. Sponge metals have been properly dened as those reduced, usually by gaseous agents, from oxidic compounds of the metal under such low temperature conditions with respect to the melting point that grain'growth, cementation, or incipient fusion of the reduced metal does not take place. The material is thus macroscopically pseudomorphid with the particles of the source material, but under high magnication is seen to consist of aggregates of very minute metailic particles, which in general are of an individual grain size of a micron or less. Since ,D the macroscopic form is preserved, the whole structure is thus cellular or porous, whence the name sponge metal is applied.

While sponge iron is so produced commercially,

and is a material recognized as having metallurgical advantages for special purposes, sponge chromium has not been produced commercially, nor have previous investigations disclosed a workable method of manufacturing it, despite its obvlous interest for use in powder metallurgy.

The prior art The reason why sponge chromium can not be prepared by conventional methods, such as are lused for iron, resides, I have found, in the exceptionally high stability of chromium oxide (CrzOa). Usual reducing gases are only very weakly reducing on this oxide even at very high temperatures. Thus, taking the typical example of the reduction of CrzOs by hydrogen, the equilibrium constant for the reaction is at 1200 C., 1.3 .l04; at 1400 C., 5.5X104; and near the melting point of chromium (which has been given at 1530-1570 C.) about 1.6 103. Thus the amount of water vapor produced by passing pure hydrogen at one atmosphere pressure over chromium oxide is at 1200 C., only .01%; at 1400 C. about .06%, and at the melting point, about 0.16%. This shows the high sta- 55 bility of this oxide.

It has been proposed to reduce chromlc oxide by hydrogen at temperatures near 1500 C., by recirculating hydrogen gas repeatedly (obviously over a thousand cycles would be needed to use it up at 1500 0.), and each time removing the minute amount of water formed per cycle by freezing it out with liquid air. Such a process I deem uneconomical, and subject to great dimculty in adaptation to commercial use because of the very serious diiiiculties of nding materials gas tight to hydrogen at these temperatures, as well as to the low cyclic efliciency.

Other investigators have produced what they term as sponge metals,` including chromium. However, upon consideration of these, it will be readily seen that they do not have and could not possiblyhave produced sponge chromium. For example, Flodin in U. S. Patent 1,792,532 of February 17, 1931, produced an. alloy of iron and chromium. Flodin rst formed an ironchrome oxide briquet which contained silicon but 'which he claimed was substantially free of carbon. Thereafter, this briquet was heated to 1150-1300" C. to the end that the silicon reduced the chromium oxide. Since the material included iron, grain growth necessarily followed, since this begins in iron near 900 C. It is obvious that modins material could not answer the definition of a substantially pure, true sponge chromium.

It has been proposed to reduce chromic oxide with hydrogen at 900-1100' C. according to the reaction See Rich, U. S. Patent '1,741,955. Meyer has (Kaiser Wilhelm Inst. Eisenforch 13, 1931, p. 199) stated that he could not elect this reduction at 900 C. In any event, assuming Rich has an .operative process, the equilibrium constant for this reaction is 1.3X10-4 at 1200 C. My experimental observations have shown grain growth in chromium to begin below this temperature and to be appreciable at this temperature. From the low equilibrium constant it should be obvious that even at 1200 C. an enormous hydrogen to oxide ratio must beused. Rohn, in Patent 1,915,243 of June 20, 1933, proposed operation at 2000 C.

Baukloh and Henke (Zeit. fur Anorg. und All. Chemie 237 (1937), 307-310), give the percentages of reduction effected by hydrogen for various temperatures. According to them about eight hours are required at 1000 C. to eiect as much asv an 8% reduction and at 1200 C. only 40% reductionisachievedintliistimeperlod.

The non-equivalence of hydrogen and other reductants As will appear hereinafter,.I have found that the reduction of chromium chloride to produce sponge chromium must be conducted in a relatively narrow temperature range-between 775- 815 C., the uppertemperature being that at which chromous chloride melts. Preferably, -the reaction is carried out between 775 and 802 C. In Patent 1,373,038, Weber states that chromium chloride can be reduced by metallic iron to metallic chromium over a temperature range between 700 and 1200 C. I have determined that when the reducing agent is not a gaseous medium, a true sponge metal cannot result as a product.

In the case of Webers process, the reducing medium is a metal, particularly iron. Within the temperature range specified by Weber, iron is not a gaseous reducing medium. It should also be particularly noted that the reduction of chromlum oxide at temperatures near its melting point causes rapid grain growth in the metal, and such a product could not fall within the proper and accepted definition of a true sponge metal.

While carbonaceous reductants, such as carbon monoxide, or methane, are more powerful reducing agents than hydrogen at temperatures above about 900 C., these materials result in the formation of chromium carbide, which renders the product unsuited to the special requirements of powder metallurgy or the manufacture of stainless steels. Hydrogen therefore stands alone as a reducing agent for production of sponge chromium from a chromium chloride.

General considerations of the process I have discovered a method of reducing a chromium bearing material, as CrCl3, to metallic chromium at a temperature as low as 780-800 C. by hydrogen gas, and have succeeded in manufacturing a new and desired metallurgical material, sponge chromium of high purity.

My investigations of the reduction of chromous chloride by hydrogen CrCl2|H2=Cr+2HCl show that the amount of hydrochloric acid in equilibrium with hydrogen (when the total pressure was 1 atmosphere) was at 700 C., 1.05%; at 750 C., 1.87%; at 800 C., 3.12%; and at 815 C., 3.60%. At 815 C., pure chromous chloride melts, and, in the process of melting, easy access of hydrogen is hindered, so that reduction is slow at temperatures in the liquid range; 815 C. at one atmosphere therefore provides a limiting upper temperature. V

The first stage of reduction of chromium trichloride is, under normal circumstances, to the dichloride, rather than to metal. The reaction 2CrCla-l-H2=2CrCl2+2HCl is substantially irreversible, and for all practical purposes may be considered to go completely to the right, and at equilibrium results in virtually complete usage of the hydrogen. The reaction rate becomes appreldrogen .to metal is possible with a theoretical utilization of hydrogen corresponding to the limit set by equilibrium of 3.12% HC1 at 800 C., and in a continuous counter-current reduction apparatus the Hunting normal HClcontent is corresponding to the stoichiometric relationships Cr:CrCl2:CrCl3. Any concentration of HC1 above 3.12% will cause reversal in that part of the reduction unit where the metal is present, but by using a counter-current process I can secure higher HC1 concentrations. To avoid fusion and volatilization, and to ensure the macroscopic retention of form required to produce sponge chromium, temperatures above 802 C. should not be utilized, although where sponge chromium is not desired the temperature can vary between 775 C. and 850 C. Above 850 C. the' material vaporizes faster than it reduces and diiculties due toV vaporization are encountered.

Hydrogen purification Passing commercial hydrogen over the trichloride under these conditions, did indeed produce somechromium metal, but only in a form badly contaminated with oxide. Further, I found that ordinary methods of removing the hydrochloric acid gas from the reducing hydrogen by chemical absorption methods in water or basic solutions were entirely unsuited to the regeneration of the reductant gas. The reason for this was traced to the fact that chromium chloride has a very low tolerance for water vapor or any oxygen containing gas in fact. Thus, for example, I found that the reaction permits at 800 C. of a partial pressure of H2O of only 1.5 10*6 atmospheres when the pressure of HC1 is 1 atmosphere, but under conditions permitting the continuous reduction of CrClz, corresponding to 3.12% HC1, the tolerance drops and is only (1.5X10-6) (0.0312)2=1.5 10f9(approx.) as the partial pressure of water vapor in atmospheres. Any amount of water vapor in excess of this is completely reacted to chromium trioxide; Similarly, I found the tolerance of CrCla, corresponding to the reaction is a partial pressure of water vapor of 4.5X10-5 atmospheres, vat 800 C., when the partial pressure of hydrochloric acid is 0.0312 atm. and hydrogen .is practically 1 atmosphere.

Finally; in the oase of chromium metal itself, I found th'e tolerance corresponding to the reaction 2Cr+3H2O=Cr2Oa+3H2 is 2.7 10'I atmospheres partial pressure of water vapor at 800 C. when the hydrogen is near one atmosphere.

From the above it is clear that the impurities in commercial hydrogen, including oxygen from a sol Manner of practicing the invention I have discovered a presently related method of eliminating these diillculties successfully with respect to hydrogen and of setting up a continuous reductionand purication which produces substantially chemically pure sponge chromium which contains less than about 0.5% of oxide, with traces only of iron and carbon.

The manner in which the'invention is preferably practiced will be best disclosed and understood by reference to the drawing, wherein the single figure is a flow sheet indicating in a schematic manner the process of the present invention.

Commercial hydrogen is first purified of oxygen as by passing it through a drier 3, then over a nickel catalyst at about 600 C., and again over a drier 5, and thence passes into storage container 6, where it is stored over a dehydrating agent as concentrated sulphuric acid. (Suitable valves, pressure regulating devices, temperature and pressure indicating devices, are included in the operating system but are not shown.)

The hydrogen storage container 6 is connected to a closed gas circulating system consisting of a pair of sorption purifiers l and 8 containing active carbon, a circulating pump 9, a final purifier I containing active carbon, and the reducing furnace Il. The first sorption purifiers l and 8 are operated at ordinary temperatures, alternately. It has been found that commercial grades of active carbon are capable of preferentially absorbing hydrochloric acid gas up to 4 to 5% of the weight of carbon under these conditions, before appreciable HC1 passes through an absorber unadsorbed. When this point is reached. the alternate absorber is connected in the circulating system, and the saturated one is desorbed by first heating to about 200 C., Which drives off approximately 80% of the retained HC1, after which the temperature is further raised to about 300 C., and the purifier is flushed with about of the volume of gas which has been purified during the previous half cycle, using fresh hydrogen, admitted through line I2 from the storage container 6. This removes the remainder of the adsorbed HC1 and water vapor. It is then ready for further use. The HC1 is driven off and absorbed in water or any desired material suitable for the preparation of a commercial by-product in the absorber After leaving the pump, and before actually entering the reduction furnace, the gas is given a final purification in adsorber I0 containing active carbon and cooled to -70 to -40 with carbon dioxide snow. Small amounts of water vapor passing the larger absorbers (which are essentially the HCl adsorbers) are thus virtually completely removed. The regeneration of the accessory purifier need be accomplished only after relatively long periods of operation, since the amount of material retained is small. The regeneration is carried out in a manner similar to the larger` units.

The reduction unit is an electric tube furnace il of square or rectangular cross section through which trays containing CrCla crystals are passed countercurrent to the flow of hydrogen gas, the maximuml temperature being 802 C. Thegases in the furnace are not corrosive to ordinary steel, and the tube of the furnace may be either steel protected on the outside from oxidation by air, or of a non-scaling chrome-nickel steel. An especially satisfactory type of construction is to provide a double walled tube, the outer wall 2| being chrome nickel steel, and the inner wall 22 of ordinary low carbon steel. Hydrogen from line 23 is admitted to the space between the walls and is maintained at slightly above l atmosphere pressure. This permits the use of either a single, or multiple assembly of inner ducts, which may be made of low carbon steel of light gauge, and

which are readily replaceable. This construction Y gas, suitable gas locks 26 and. 21 are placed at the inlet and outlet end of the furnace. 'I'hese are operated in the usual manner, being purged and filled with hydrogen from 4lines 28 and 29 to ensure preservation of the oxygen free atmosphere in the furnace.

I have found that, while normally the reduction takes place in steps, flrst to CrCla and later to metal, resulting in concentrations near 4.7% HC1 maximum when using the continuous countercurrent apparatus, that, by using high gas velocities and causing the trays to enter rapidly the 800. zone of the furnace, I am able to obtain an exit gas as high as 8 to 10% HCl. This may be interpreted as showing that to some extent direct reduction of CrCla to metal is in part attained. I have also obtained hydrochloric acid concentrations above the CrCla-Hz-Cr-HCl limit when using slow gas rates and slow rates of tray admission, but in this case the result was presumably due to a preponderance of the reaction.

My invention isv applicable to the reduction of either of the solid chromium chlorides, whether stepwise or direct, and may obviously also be utilized for producing anhydrous chromium chloride, since the formation of metallic chromium can be` completely prevented at low gas rates by maintaining the entrant hydrogen with a content of HC1 of at least 3.12%.

In practicing my invention, it is desirable to remove all' traces of moisture which may be contained on or in the chromium chloride. The socalled violet anhydrous crystals of CrCla, which are virtually insoluble in water, nevertheless are capable of adsorbing several percent water when they have previously contacted this medium as liquid, or been stored in air of normal humidity. I have found it desirable, therefore, to dry the crystals by heating them, preferably to 200-300 C. and preferably in a vacuum, before introducing them into the reduction furnace, and to carry out the transfer quickly, or in a dried atmosphere, sothat recontamination is avoided. Temperatures higher than this must not be used not only because the chloride vaporizes and dissociates, but also because the crystals become reactive with oxygen at temperatures slightly above 300 C. This step is indicated at 3 l Thus it is seen that While one patentee found it necessary to utilize temperatures above 1600 C.

to reduce chromium successfully by hydrogen, by g5 carrying on the reduction from the CrCla form under the conditions, and in the manner specied above, I have been able to produce chromium in the true sponge form at temperatures of 800 C. and in a continuous counter-current unit, and have demonstrated the feasibility of producing a.

'new form of chromium having metallurgical advantages. The new product, sponge chromium, is claimed in a companion application, Serial No. 157,153 of even date.

I claim:

l. A process for the reduction of a chromium chloride to chromium 'comprising reducing said chloride with substantially only dry hydrogen at a temperature of about 775 C. in a reducing zone.

2. A process for reduction of a chromium chloride to chromium comprising'reducing said chloride with substantially dry pure hydrogen lat a temperature of 775-850 C. in a reducing zone, removing hydrogen containing hydrogen chloride from the reducing zone and separating hydrogen chloride and water vapor therefrom by a treatment including passage of the hydrogen over activated carbon, and returning the puriied hydrogen to the reducing zone.

3. A process for the reduction of a chromium chloride to chromium comprising reducing said chloride with substantially dry hydrogen ata temperature of about 775 C. in a reducing zone, removing hydrogen containing HC1 from the reducing zone, separating HCl and water Vapor from said removed hydrogen by contacting said removed hydrogen with activated carbon, and returning the purified hydrogen to the reducing zone.

4. In a process of reducing a chromium chlov ride yto chromium, the step of passing substantially dry pure hydrogen counter-current to said chloride in a reduction zone in a furnace at about 800 C. at such a rate that hydrogen leaving said l furnace carries about 4.7% HCl.

5. In a process of reducing chromium trichloride to chromium dichloride, the step of passing substantially dry oxygen free hydrogen containing about 3.12% HCl over said trichloride chloride in a reduction zone at a temperature of about 802 C.

6. In a process of reducing chromium trichloride to chromium dichloride, the step of passing substantially dry oxygen free hydrogen containingv direct contact therewith substantially dry pure hydrogen. I

9. A process for the production of sponge chromium comprising heating substantially dry chrobetween 775 and 815 mium trichloride to a temperature of 775-802 C. and passing dry pure hydrogen thereover to reduce said chloride to a chormium macroscopically pseudomorphic with said chloride and substantially free of grain growth.

10. Reducing chromic chloride to chrome by passing dry hydrogen over a mass consisting of substantially only said chloride at a temperature between 775 and 802 C. to reduce said chromic chloride directly to chromium, and continuing to pass said hydrogen until said mass .consists substantially only of chrome.

11. Reducing chromic chloride to chrome by passing dry hydrogen over a mass consisting of substantially only said chloride at a temperature C. to reduce said chromic, chloride directly to chromium, and continuing' to pass said hydrogen until said mass consists substantially only of chrome.

12. Reducing chromic chloride to chrome by passing dry hydrogen over a mass consisting of substantially only said chloride at a temperature between 775 and 850 C. to reduce said chromic chloride directly to chromium, and continuing to pass said hydrogen until said mass consists substantially only of chrome.

13. Reducing chromic chloride to chrome by passing dry hydrogen containing less than 4.7% HCl over a mass consisting of substantially only said chloride at a temperature between 775 and 802 C. to reduce said chromic chloride directly to chromium, and continuing to pass said hydrogen until said mass consists substantially only of chrome.

14. Reducing chromic chloride to chrome by passing dry hydrogen containing less than 4.7% HCl over a mass consisting of substantially only' said chloride at a temperature between 775 and 815 C. to reduce said chromic chloride directly to chromium, and continuing to pass said hydrogen until said mass consists substantially only of chrome.

15. Reducing chromic chloride to chrome by passing dryvhydrogen containing less than 4.7% HC1 over a mass consisting of substantially only said chloride at a temperature between 775 and 850 C. to reduce said chromic chloride directly to chromium, and continuing to pass said hydrogen until said mass consists substantially only of chrome.

16. In a process of reducing chromium trichloride with hydrogen to chrome metal whereby a hydrogen eilluent stream containing about 5% hydrogen chloride is produced, the steps of passing said eilluent stream into contact with activated carbon to remove said hydrogen chloride substantially entirely and simultaneously maintain the water content thereof at a point whereat the partial pressure of water vapor in the hydrogen is less than about 4;5 10-5 atmospheres, at a temperature of about 800 C.

17. In a process of reducing chromium trichloride tochrome metal utilizing hydrogen as the vreduc-tant, the steps which consist in employing as feed material chromium trichloride containing less than 0.05% wat-er while maintaining the reduction temperature between 775 and 850 C. CHARLES G. MAIER.

Images