OFFICERS OF SECTION D.

Vice President.

C. J. H. WOODBURY of Boston, Mass.

Secretary.

ARTHUR BEARDSLEY of Swarthmore, Pa.

Member of Council.

OLIN H. LANDRETH of Nashville, Tenn.

Members of Sectional Committee.

WILLIAM HARKNESS of Washington, WM. A. ROGERS of Waterville, Me. OLIN H. LANDRETH of Nashville, Tenn.

Member of Nominating Committee.

JAMES E. DENTON of Hoboken, N. J.

Members of Sub-committee on Nominations.

JEROME L. BOYER of Reading, Pa., W. R. WARNER of Cleveland, Ohio, ALBERT A. CARY of New York.

PAPERS READ.

THE INFLUENCE OF ALUMINUM UPON CAST IRON. By W. J. KEEP, C. E., Prof. C. F. MABERY and L. D. VORCE of Cleveland. Read by W. J. KEEP, Detroit, Mich.

ALUMINUM is a metal obtained from its oxide, alumina. It is white in color and very tenacious, and it alloys readily with iron.

Cast iron, ordinarily used, is iron which contains all the carbon that it could absorb during its reduction in the blast furnace.

This carbon, when found in chemical union with the iron, is called combined carbon. In this state it cannot be seen. It is also found mechanically mixed with the iron in the form of graphitic carbon, when it becomes visible.

Other elements commonly found in cast iron are phosphorus, sulphur, manganese and silicon.

The natural condition of carbon in iron is the combined state. The presence of silicon drives a portion of the carbon into the graphitic state. Sulphur, manganese and phosphorus do not cause the carbon to leave its natural combined state, and if silicon be present, these elements either drive it out or overpower it.

Carbon is therefore a passive element and is made to change its form by the presence of other elements. It is this change of carbon which indicates, to the eye, the influence of any element upon the cast iron.

Iron and combined carbon, or carburetted iron, is called white iron, and the grain is generally very fine, and often even, and the metal is very hard. Graphite darkens the fracture until it becomes a very dark gray, and the grain is coarse and irregular. With increase of graphite the metal

becomes soft.

We shall confine ourselves in this paper to the influence of aluminum upon cast iron.

Let us for a moment review the present knowledge on this subject. It is known that fused wrought iron, a mixture of cast iron and steel, or steel alone, either of which would make castings which would be full of blow holes, will make solid and homogeneous castings if as small a quantity of aluminum as one-tenth of one per cent is added just before pouring. Also, that such addition causes the iron to remain fluid long enough to allow its being cast into moulds.

It seems to be the general opinion that the aluminum does not remain in the metal, but that it exerts its influence between the time of its introduction and the time of its departure.

This seems to be the sum total of the present information regarding the influence of aluminum upon iron.

We propose in this paper to give the results of a series of very carefully conducted tests to substantiate further the statements just made, and to settle the question as to whether aluminum remains in the casting; also to determine the influence of this metal upon the physical structure and upon the composition of iron.

The physical tests that we have employed are what are known as "Keep's tests;" and by them we are enabled to make apparent to the eye the influence of any element upon cast iron.

When it was understood that we were to undertake this examination, the Cowles Electric Smelting and Aluminum Company of this city (Cleveland) kindly furnished us with what ferro-aluminum we needed, and Prof. C. F. Mabery and L. D. Vorce, of the Case School of Applied Sciences, of this city, volunteered to undertake the chemical examination of the test bars. The results of these investigations will be appreciated, when it is understood that we began without the least expectation of the very important results we have obtained, and that the methods for the determination of minute quantities of aluminum were so imperfect, that the small quantities used in the "Mitis" process could not be determined, if they still remained in the castings.

Regarding the physical tests, we should state that we use two bases: one, a white iron, with composition Si .186, P .263, S .0307, Mn .092; the other, a gray Swedish iron marked FLM, with composition, Si 1.249, P .084, S .04, Mn .187. The ferro-aluminum contained Si 3.86, Al 11.42.

The melting was done in a covered plumbago crucible, in a coke furnace driven by a blast of two and one-fourth ounces. The test bars were one foot long and cast in pairs; one, one-half an inch square and its mate one-tenth of an inch thick and one inch wide.

We started with thirty pounds of the base in the crucible; at the first heat there were cast four pairs of bars from the base alone, which took five pounds of metal. After allowing the remaining metal to become solid, we returned the runners of the first cast, and added four pounds of the base, and returned the crucible to the furnace. When nearly melted, we added enough ferro-aluminum to bring the percentage of aluminum in the whole to where we wished it for the second set of bars. We proceeded in like manner through the entire series of heats. To arrive at the influence of the aluminum, we made another series of heats, with the same base with exactly the same conditions, only we did not add the aluminum. The difference between the two series of tests gives the effect of the aluminum.

We shall consider this subject under the following heads:

1. The solidity of castings and the prevention of blow holes.

2. Does the aluminum remain in the iron to exert an influence when the iron is remelted?

3. The effect of aluminum upon the grain, or the changing of the carbon from the combined to the graphitic state.

4. Taking away the tendency to chill.

5. The prevention of sand scale.

6. The effect upon hardness.

7. The resistance to a load gradually applied, or a dead weight.

8. The resistance to a load suddenly applied, or impact.

(1) The solidity of castings and the prevention of blow holes.

All of our tests bear upon this subject, but we have made one test, using the white base iron and one-tenth of one per cent of aluminum. It is almost impossible to get a solid casting of the white base alone, and its resistance to weight is generally about 175 pounds for the one-half inch square bars, and its resistance to impact is about 100 pounds. We have obtained, however, exceptionally sound castings of this base, and we shall use the strength of such castings for comparison.

These sound castings of the white base alone resisted a weight of 379 pounds. With one-tenth of one per cent of aluminum added it resisted 545 pounds, a gain of 166 pounds, or about forty-four per cent from this small addition.

Measuring the resistance to impact, the white alone was 239 pounds; with aluminum 254 pounds, or about six per cent gain.

The castings appear of slightly finer grain, and the character of the crystallization is somewhat different, but the secret of the strength lies in the closing of spaces between the grains, or in other words, in the increased solidity of the casting. No other change is noticeable in the metal. A graphic representation of this test is not needed.

(2) Does the aluminum remain in the iron to exert an influence when the iron is remelted? CHART 1.

To determine this, we made a series of six heats from the white base, and added to the first heat one-fourth of one per cent of aluminum. This amount alters the grain very perceptibly, making it whiter and finer, and removing the tendency of the base to a slight specular appearance, and giving a homogeneous fracture. It increases the strength above the base about twenty per cent to resist weight, and for impact, an increase of over seventy per cent.

The next heat was a remelt of the first, with the runners of the first cast put back, and enough white base added to reduce the aluminum to twotenths of one per cent when the second cast was made.

Our comparisons will now be made between this series and the comparison series of the base alone. Looking at the chart we see that the effect of the aluminum in this second heat is greater than it was in the first heat to which heat the aluminum was added. This is due to the increasing porosity at each heat of the base when melted alone, and to the solidity of

the series with aluminum. At the third and subsequent heats the same result is apparent, the remaining aluminum causing more solid castings, though the continued additions of white iron at each heat and the consequent lessening of aluminum render the castings less strong at each remelting. Yet the effect of the aluminum is so constantly apparent at each melt, as to leave no doubt as to its presence even in the sixth remelting. The chart, which we have prepared, shows these effects both as to weight and impact.

As we proceed with the description of other tests it will be noticed that we add but a small quantity of aluminum at each heat, and depend upon the additions made at previous heats to bring up the required percentage.

The results of the tests show conclusively that the aluminum remains and exerts its influence in subsequent casts as fully as would be expected. (3) The effect of the aluminum upon the grain, or the changing of carbon from the combined to the graphitic state.

Let us say a few words in regard to the way in which, and the reason why, carbon takes on the graphitic form.

All of the carbon, both combined and graphitic, which the iron is capable of holding when solid, must be dissolved and exist as combined carbon in the melted iron. Cast iron made in the usual way contains all of the carbon that it can hold.

Very often cast iron, when melted, contains more carbon than it can hold in combination when at a lower temperature; if so, as the iron cools down, such excess of carbon will separate as graphite and rise to the surface.

In any case, when a melted iron contains more carbon than the iron can hold in com

bination when cold, all of the excess will not be able to reach the surface, though it may not be visible to the eye, in the casting.

The introduction of other elements into the melted metal may alter its ability to hold the carbon. Sulphur causes it to let some go, while manganese enables it to hold more carbon in solution. Silicon also somewhat diminishes the capacity of the molten metal to retain carbon while it is liquid.

Aluminum allows most of the carbon to retain its natural combined form until the metal is too thick for the separated carbon to escape, but at the