US1739234A - Process of making metal filaments - Google Patents

Description

Patented es. ifi,

T TATES PATENT OFFICE RICHARD JACOBY AND FRITZ KOREF, OF BERLIN, GERMANY, ASSIGNORS TO GENERAL ELECTRIC COMPANY, A CORPORATION OF NEW YORK PROCESS F MAKING METAL FILAMENTS No Drawing. Application filed June 1, 1922, Serial No. 565,269, and in Germany June 18, 1921.

If a tungsten filament of an incandescent electric lamp isobserved throughout its life, one notes that the original inherent fibrous structure disintegrates and an irregular crystal structure takes its place. As a consequence, the original ductility and tenacity are lost.

Inorder to overcome this disadvantage, which can easily result in rendering the lamp ,useless, two methods have previously been I used.

' grains are consequently, in spite of their By the first one, it has been sought to eliminate the process of crystallization and to this end, foreign bodies, as for. example, thorium oxide, have been added. By this precaution the destructive process of recrystallization was, it is true, delayed but not prevented so changed. Filaments of that kind retain throughout their life their ductility. Their manufacture however is difiicult and expensive.

The present invention 'ofi'ers a new solution of the problem of permanently ductile filaments, The object of the invention is to provide a wire of the highest melting metal such as tungsten which consists throughout its entire length of crystal grains which have linear dimensions amounting to a multiple, for example, 10 to 20 fold, of their diameter. The transverse dimension of a crystal grain may be very different at diflerent places it amounts to a considerable part of the diameter of the wireand may indeed at some parts of the'diameter of a crystal grain equal. that diameter of the wire.

The contact surfaces of an elongated crystal with its neighboring individuals are as a rule only slightly inclined toward the axis of the wire so that the crystal grains, which in the longitudinal direction of the wire follow one upon the other, overlap to a great extent. The crystal considerable size, well interlocked or ancess will serve.

chored one to the other. The joints between the ends of adjoining crystal grains are muchlike scarf joints, and are so strong that the wire is nearly as strong against bending at the joints as at other points. The transverse section of the wire may, therefore, either be filled throughout with a few, for example, two to ten crystal grains or in some cases with a less number, or else a single crystal or two. Metal wires which consist of crystal grains of this kind are completely ductile and do not change their structure and ductility during burning.

For their manufacture the following pro- Drawn wires for example of tungsten are subjected to a heat treatment of such kind that the long drawn out fibres of the wire produced by the previous working or drawing are brought to the point where destruction'begins to a slight extent and then the wire is heated so high and so quickly that the products resulting from destruction of the fibres are combined by the road of recrystallization into new crystal grains of the kind above portrayed. The heating must not be carried out with such slowly increas: ing temperature that they can form a coarsely crystalline intermediate stage.

A suitable way of carrying out the invention by means of which one can obtain the desired crystal structure is, for example, the following.

First, the drawn wire is led continuously at a suitable speed through a zone of elevated temperature in whichtthe velocity and temperature are so chosen that the fibrous strucv ture due to drawing at once begins to break up, a fact which can be ascertained by microphotographic tests. The correctness of the conditions can further be determined by the fact that the wire still remains really ductile, but by the choice of a little higher temper ature becomes brittle, which can be ascertained through testing. The conditions are difl'erentfor different wires according tothe previous mechanical treatment and the chemical nature of the materials, since the metals used in technical work are, as well known, never absolutely the same, and must in each case be found out by trial,

Y ways.

In the second stage the continuous temperature treatment is then carried out at the highest possible temperature. The upper limit of this temperature is, determined only by the melting point of the material and its mechanical tenacity, regard to which must be given by suitable construction of the feed: ing devices.

It is of course also possible to carry out only one of the two treating steps in a continuous manner.

With tungsten wire particularly good and uniform results are obtained through the addition of-small amounts of difiicultly fusible and diflicultly reducible oxides to the raw material of the drawn wire. This addition may for example be made in well known For example tungsten oxide before its reduction to metal is brought in refractory Battersea crucibles to temperatures which are in the neighborhood of the melting point of the oxide,'w'hereby crucible material is taken up by the oxide.

It has further been found that the contact surfaces of the crystal grains contacting one with the other are enlarged to a great extent and the interlocking or anchoring of these surfaces therefore considerably im-.

proved if'the' wire between the two steps of treatment is subjected to mechanical deformation. as for example by being drawn down one or more steps. An example is given in the following detailed description:

- One starts with drawn wire which is'made according to U. S. patent, Coolidge, 1,082,933,

- December 30, 1913, or according to German Thereupon the wire is again, as in the previous treatment, heated by passing it through two mercury contacts during which the rate of feed amounts to 41 centimeters per -minute, the length of the incandescent portion 2 centimeters, and the strength of the current 2.5 amps.

What we claim as new and desire to secure by Letters Patent of the United States, is

lQThe process of making from tungsten drawn ductile wires composed of elongated crystal grains overlapping along the length of the wire, which consists in first bringing a drawn wire with fibrous structure to the temperature at which destruction of the fibres starts but does not render the wire brittle and then heating it to nearly the melting point quickly enough to produce new crystals having a length many times their diameter andstrongly interlocked.

2. The process of'making from tungsten drawn ductile wires composed of elongated crystal grains of a size to fill all or a considerable part of the cross section of the wire and overlapping along the length of the wire, which consists in first bringing a drawn wire with fibrous structure to the temperature at which destruction of the fibres begins and then heating it to nearly the melting point quickly enough to produce new .crystals having a length manytimes their diameter and strongly interlocked, one or both of said heat treatments being given the wire by passing it continuously through a zone of suitable temperature. v

3. The process of making from tungsten drawn ductile wires composed of elongated crystal grains each of a length many times its diameter and on the average of a. size to fill all or a considerable part of the cross section-of the wire and overlapping along the length of vthe wire, which consists in first bringing a drawn wire with fibrous structure to a temperature at which destruction of the fibers'begins, then subjecting the wire to mechanical treatment to deform it slightly; and then passing it continuously through a zone of suitable temperature to form new'crystals.

4. In the manufacture of tungsten filaments, the improved process which comprises annealing a wire, applying a suitable strain tothe wire, and then passing the wire through a zone of steep temperature gradient.

In witness whereof, we have hereunto set our hands this eleventh day of May, 1922.

v RICHARD JACOBY.

, FRITZ KOREF.