Aluminum oxide abrasive and method of making the same

May 7, 1935.

J. S. MASIN

ALUMINUM OXIDE ABRASIVE AND METHOD OF

2,000,857

MAKING THE SAME

Filed Dec, 1, 1930

<img file="US2000857A_D0001.tif" />

<img file="US2000857A_D0002.tif" />

Patented May 7, 1935

2,000,857

UNITED STATES PATENT OFFICE

2,000,857 aluminum oxide abrasive and method OF MAKING THE SAME

Jacob S. Masin, Anniston, Ala., assignor to Swann Research, Incorporated, a corporation of Alabama

Application December 1

Claims.

My invention relates to aluminum oxide or alumina for abrasives and for other purposes, and to a method of preparing the same, and has for one of its objects the provision of an economical proc® ess whereby aluminum oxide containing a larger percentage of impurities than has heretofore been thought practicable, may be employed in the arts for the same purposes as pure aluminum oxide.

¹⁰ A further object of my invention is the production of an aluminum oxide containing from 1 to 5% of impurities, particularly TiCte, which shall be superior in its physical characteristics when used as an abrasive to pure aluminum oxide.

¹⁵ A still further object of my invention is the provision of a process for producing aluminum oxide containing from 1 to 5% of T1O2 whsrpin the finished grain will have a toughness comparable to or greater than that of pure aluminum oxide grain and which, when embodied in a grinding wheel, shall be free and cool cutting and which shall not smear or glaze the wheel in service or require its redressing.

Aluminum oxide, as heretofore produced for ²⁸ commercial purposes, may be divided into two general classes, namely, that used for more or less rough cutting grinding wheels where there are rather large cuts made in the metal to be ground, and aluminum oxide prepared for precision work 30 where small cuts are made with the grinding· wheel in which it is employed.

For the former purpose, a relatively tough and rugged grain material is required which may contain considerable amounts of impurities and this 35 is usually provided by smelting bauxite with coke in an electric furnace, followed by crushing separating and screening to the desired size. This treatment may also be followed by other treatments to increase the purity of the grain. In my 40 prior Patent, No. 1,77X570, issued October 7, 1930, there is described and claimed such a process for the production of aluminum oxide and in which I subject the separated grain, after furnace treatment, to leaching with a mixture of hydrofluoric 45 and sulphuric or hydrochloric acids to reduce the impurities to a minimum. After this treatment, the grain still contains from 1 to 5% of T1O2.

While aluminum oxide grain produced in accordance with my prior patent has proved emin50 ently successful in service and superior to impure aluminum oxide grain theretofore on the market, I have found that where it is embodied in a grinding wheel used for fine precision work, there is a tendency for the wheel to break down 55 too rapidly in service, resulting in a lower production per wheel.

In my' study of the action of grinding wheels, particularly as applied to precision work, I have determined that the major desiderata in a pre⁶⁰ cision grinding wheel grain are the following:— ., 1930, Serial No. 499,238 (CI. 51—279)

1. The grain must be sufficiently pure or else the impurities must be rendered innocuous in order to prevent glazing and burning of the wheel in service.

2. The grain must be sufficiently tough so 5 that a too rapid disintegration of the wheel does not occur in service.

3. The grain must be sufficiently friable so that a constant, uniform, though a not too rapid breaking down of the grain occurs in service 10 during which breaking down process new sharp cutting edges are uniformly generated to replace those dulled by contact with the steel being cut.

4. The particles of the grain formed by the 15 breaking down process should be sufficiently small arid friable so that the surface being ground is not deeply gouged as the loose particles pass out from under the revolving wheel.

5. The grain must have good bonding prop- 20 erties. Good bonding properties are manifested by chemical inertness of the grain surface.

Of the above mentioned desiderata, the proper balance between toughness and friability has hitherto been a difficult property to obtain, es- 25 pecially in grains made from alumina which contained considerable amounts of impurity.

It has heretofore been the practice, where an aluminum oxide abrasive was to be used for fine precision work, to employ a pure alumina which 30 does not smear or glaze over a grinding wheel because of the absence of associated impurities. Pure alumina, for this purpose, is usually produced by treating bauxite with caustic soda and heating it to bring about a reaction in which 35 sodium aluminate is formed. The latter is filtered off in solution and then treated with carbon dioxide or is merely diluted, whereby pure amorphous alumina is precipitated. After separating out the amorphous alumina, thus ob- 40 tained, and drying it, it is then fused or calcined to form pure aluminum oxide crystals. This product is characterized by being white in appearance and translucent. The product thus obtained is of a high degree of purity and when 45 bonded into grinding wheels, has substantially the physical characteristics pointed out above which are so desirable for precision work.

It is obvious that such a process is a very expensive one and there has resulted a real de- 50 mand in the art for a cheaper product having a suitable combination of toughness and friability. Although numerous efforts have heretofore been made to obtain such a cheaper product, none, so far as I am aware,. have hereto- 55 fore been successful.

In approaching the solution of the above mentioned problems I have found that when aluminum oxide prepared in accordance with my prior patent is placed in an oxidizing furnace and sub- ⁸⁰ jected to temperatures upward of 1200° C. for a sufficient length of time, a marked change occurs in the appearance, and in the physical characteristics of the product.

Furthermore, I have found that my improved product, when embodied in a grinding wheel, is free and cool cutting and does not smear or glaze the wheel, and that such a wheel does not re• quire frequent redressing.

in The treatment of the grain with heat after removing exposed, acid-soluble impurities should not be confused with the ordinary “roasting” step heretofore practiced in the preparation of aluminum oxide made from fused bauxite or from pure 15 amorphous alumina. Roasting, as heretofore practiced, has usually been at temperatures around 800° or 900° C., whereas I have found it necessary to go to considerably higher temperatures to bring about the required resetting of the 20 crystals, the development of a chemically inert surface, and the proper balance between toughness and friability desired. In general, it may be stated that these changes are evidenced in general appearance by the change in color to white 25 and in the change from translucency to opaqueWhen examined under the miscroscope, it will be noted that radical changes take place in grain treated in accordance with my improved process. 30 In order that these changes may be better understood reference will now be had to the accompanying drawing, in which:—

Fig. 1 is a view showing a well known pure alumina grain magnified 80 diameters.

Fig. 2 is a similar view of grain treated in accordance with my prior Patent No. 1,777,570.

Fig. 3 is a similar view showing grain treated in accordance with this application; and

Figs. 4, 5 and 6 are views of a grain of alu4e minum oxide treated in accordance with my present invention magnified over 700 diameters.

In Fig. 1 it will be noted that the grains are irregular in shape, presenting more or less irregular edges. This grain being pure alumina, is 45 translucent and white in appearance.

The grain shown in Fig. 2 has substantially the same outline as that shown in Fig. 1. This grain is made in accordance with my prior Patent No. 1,777,570. The shading of this grain rep50 resents impurities in the grain which is translucent, and the impurities show under the microscope to be dissolved in the grain, giving it its pinkish appearance.

In Fig. 3 I show the grain made in accordance 55 with my present invention and magnified 80 diameters, as in Figs. 1 and 2. It will be noted first that this grain, when crushed, breaks along far more irregular lines than the grain illustrated in Figs. X and 2. This is an important change in 60 that the grain presents to the work far more sharp cutting edges than either the pure alumina illustrated in Fig. 1 or the alumina prepared in accordance with my prior patent illustrated in Fig. 2. In addition to the more irregular con65 tour, the grain has become opaque rather than translucent as in Fig. 2 and is white in appearance rather than pink.

In Figs. 4, 5 and 6,1 show grains such as illustrated in Fig. 3 magnified to over 700 diameters. 70 When thus magnified I find that the impurities which before treatment at high temperature were dissolved in the crystals, have become segregated and appear in the treated crystals as “rods” extending through the crystals in a plurality of 75 parallel planes. Fig. 4 shows the appearance of

2,000,857 these rods when the crystal is observed from one direction, and Fig. 5 shows their appearance when viewed at right angles to Fig. 4. It will be seen that in this figure the ends of the “rods” appear, while in Fig. 4 they appear as straight 5 lines extending across the crystal. In some of the crystals which I have examined under the microscope, I find the rods sticking out of the crystal as in Fig. 6. This appearance is apparently brought about by viewing the crystal in a 10 position intermediate that shown in Figs. 4 and 5. -This segregation and more thorough oxidation of the impurities in the crystal are the apparent cause of the breakage of the crystals along the irregular lines heretofore mentioned and give it 15 its improved cutting qualities.

In carrying out my improved process, I have found that it is highly desirable to follow the steps outlined in my prior patent, hereinbefore mentioned, followed by the treatment at high 20 temperature as herein disclosed. I have made numerous efforts without success to produce an abrasive grain of comparable quality, for example, by omitting the step of treating the grain with an acid reagent to reduce the impurities 25 to from 1% to 5%.

The following is a specific example of a preferred method of carrying out my process:

A charge of bauxite and coke, to which iron may be added for the purpose of forming a mag- 3(¹netic alloy with the impurities contained therein, is smelted in an electric furnace so that complete fusion takes place. Time is allowed for the carbon to partly reduce the oxide impurities and segregate to * ’ e bottom of the melted zone in the 35 form of a magnetic iron, silicon and titanium alloy. This operation is referred to as fumacing. After the melt cools and solidifies, it is broken into lumps, and pieces of refuse and alloyed impurities are picked out. The lumps are then ₄₀crushed to 6 mesh size or smaller and passed over a magnetic separator to remove the magnetic impurities still remaining.

The material may then be treated according to the procedure outlined in my prior Patent No. 45 1,777,570, issued October 7, 1930, by which the exposed, acid-soluble impurities in the abrasive grain or crystals are reduced to an optimum amount by an acid treatment such as by treating in a bath of sulfuric acid and hydrofluoric acid. 50 As an illustration of the acid treatment, a lot of graded abrasive is placed in a lead vat containing a bath composed of a mixture of, approximately, 6% hydrofluoric and 16% sulfuric acid, and the grain allowed to remain in contact with 55 the acid for twenty-four hours. The grain is then washed with water and removed from the treating apparatus. It is then dried and screened into sizes in a manner similar to that employed for the other types of abrasive. A typical analy- 60 sis of the product at this stage of the process may be as follows:—

AI2O3------

TiO2------------------------------------ ²³³65

S1O2 Fe₂O3 Some variation in the product is permissible, for example, as much as 5% of T1O2 may be present and I find it preferable to have more than 1% of T1O2 present.

The grain thus prepared is heated with agitation in an oxidizing atmosphere to a temperature upward of 1200° C., the final temperature depending somewhat on the size of the grain and

2,000,857 the duration of exposure. I have found that for coarser grains a somewhat higher temperature and longer time is required. For the finer sizes of grain, say 240 mesh and under, a temperature .5 of between 1000° C. and 1200° C. may be used— the duration of exposure being somewhat longer than where higher temperatures are employed. For 240 mesh grain I prefer a temperature of about 1400° to 1700° C. for, say, 30 to 60 minutes. 10 I have found that agitation of the material is desirable and for this operation, I have found that a rocking type arc furnace is especially suited, although a rotating kiln, heated by means of oil or gas may also be employed. An oxidizing 15 atmosphere is maintained by admitting air or oxygen to the furnace, either continually during the heating operation, or after the material has been heated to the high temperature in a neutral or reducing atmosphere. The product 20 after treatment is removed from the furnace and cooled, after which it may be further sized if necessary. The composition of the grain after heating is approximately the same as before heating.

I have found it necessary to exercise considerable care during the heating operation since the heating may easily be carried too far. I have found that if the heating of the grains containing TiOa in the amount specified is carried to the 30 citron-yellow stage described by Mellor in Vol.

VII, page 33, line 6 from bottom of the page, that the oxidation and rearrangement of crystals has progressed to the proper extent and heating may be discontinued. When cold, the grain thus 35 treated is substantially white in appearance. If heating is carried beyond this stage, the color of the grains after cooling becomes more and more brownish, due probably to the formation of titanium nitride. Grains which have been too 40 strongly heated are rendered weak physically.

While I have described a particular acid mixture for treating the grains, other acids may be used. I have found that mineral acids in combination with hydrofluoric acid give good results, 45 in particular such acids as nitric, hydrochloric and phosphoric may be used. It is possible also to use other chemical reagents such as wet chlorine gas, hydrofluoric acid or fluorine gas and even solutions of sodium hydroxide or other also kali hydroxides may be used alone or in combination with other reagents. Hence, I do not desire to limit myself to the use of particular reagents, but contemplate the use of all reagents which will reduce the surface impurities and leave a grain 55 having associated with the alumina the impurities of the amounts and character above set out.

A product treated in accordance with the above process has a toughness factor of 44.95±2.00, whereas aluminum oxide grain tested before being subjected to oxidizing treatment at a high temperature, has a toughness factor of 35.30±2.00. This toughness factor was deter- 5 mined in the following manner:

One gram of aluminum oxide abrasive grain is placed in a steel mortar under a rotating disk or pestle 1 in diameter which weighs a total of 878.5 grams. The grain before testing is first 10 carefully screened through a 32 mesh screen and grains remaining on a 36 mesh screen placed in the mortar forming a layer substantially one grain deep. The pestle is now revolved on the grain at a uniform rate for 100 revolutions, the 15 time for this number of revolutions being about 15 seconds. The contents of the mortar is then screened and the percentage of the grain remaining on a 48 mesh screen expressed as the toughness factor. A variation of about 2 units 20 each way is to be expected and is expresed as a tolerance. The results of the test have been given above.

While I have described my invention in but one form, it will be obvious to those skilled in the 25 art that it is not so limited, but is susceptible of various changes and modifications, without departing from the spirit thereof, and I desire, therefore, that only such limitations shall be placed thereupon as are imposed by the prior 30 art or as are specifically set forth in the appended claims.