US1082933A - Tungsten and method of making the same for use as filaments of incandescent electric lamps and for other purposes. - Google Patents

Description

'W. D. COOLIDGE. TUNGSTEN AND METHOD OF MAKING THE SAME FOR USE AS FILAMENIS OI INCANDESCENT ELECTRIC LAMPS AND FOR OTHER PURPOSES.

APPLICATION FILED JUNE 19, 1912.

Patented Dec 30, 1913.

Inventor: William. D. Coolidge,

, Hls .fittorney.

20 producing the same.

UNITED STATES. EEENT oration.

WILLIAM D.

coonnon, or 'scnnnnc'ranv, new Yank, assmnoa roonnnm nnpc'rnm comm, A 0032034131011 01; NEW max;

'rurms'rnn nn'rnon or unique 'rnn amnion trsn AS murmurs or INCANDESOENT Ennomc name no: son ornnn runrosns.

Specification autumn. p te t De ;30,1913,

Application filed June 19, 1912. Serial in. 704,530.

To all whom it mag concern Be it known that I, WILLIAM D. Coo- LIDGE, a citizen of the United States, residing at Schenectady, county of Schenectady,

5 State of New York, have invented certain new and useful Improvements in Tungsten and Methods of Making the Same for Use as Filaments of Incandescent Electric Lamps and for other Purposes, of which 19 the following is a specification.

her 545,274, filed February 23, 1910; and Serial Number-577,353 filed August 15, 1910.

My invention comprises a new incandescent lamp filament of drawn wire made from the metal tungsten, and a process of I have been able to produce from this refractory, brittle, nonductile and generally unworkable metal a wire which is tough, fibrous and elastic and of high tensile strength, and have thus been '25 able to facilitate the manufacture and improve the quality of tungsten incandescentelectric lamps.

The wire produced by my invention has found various other applications, and may 80 be used wherever high tensile strength, freedom from corrosion, or ability to withstand high-temperatures are of value. So also the newmaterial of which the wire consists,

fully worked ductile tungsten, has in other mechanical forms a wide variety of useful;

applications Further, an incidental but valuable new product is found at an inter mediate stage of my process, in tungsten which is wrought, but in which the working 40 process has been interrupted before the above described properties have fully developed but at a point'where desirable properties not found in any tungsten of the prior art have been developed to,an amount sufficient for the purposes for which the material inque'stion is to be used.

The metal tungsten'is described in stand- ,ard works on chemistry and rretallurgy as obtained in the form of a steel-gray to black powder, having a fusing point of about 3200 degrees C. It is stated to be entirely non-ductile. Thus, for example, Roscoe and Schorle'mmers Treatise on Chemistry, vol.

'11, p. 1058, published in 1907, cites the;

work of Moissan, but states that: The purest forms of tungsten at pres'ent obtainable are hard and brittle, and are not ductile either at ordinary temperatures or when heated. De Lodyguines Patent No. 575,002, referring to tungsten, states (page 1) that it is almost absolutely non-ductile and cannot be drawn out into wires in the sten by means ofk'a sintering process, that is,

by pressing up tungsten powder with-or without a binder into the desired shape, and then consolidating it at a high temperature by passing electric currentthrough it-.- Filaments so formed have been of-exceedingly low strength and very brittle.

I .have discovered a process by which tungsten bodies, when prepared under certain conditions, as will be hereinafter more fully described, can be mechanically worked as by hammering, swaging, rolling and drawing, and have further discovered that Incandeswhen this mechanical Working is carried on while the metal is heated to temperatures within certain maximum and minimum limits and is continued long enough, such bodies will be converted from their original crystalline character to a condition having all the characteristics of a ductile metal. ed hotworking so changes the metal that it acquires te'nsile strength and also becomes pliable andductile at ordinary or room temperatures, and if the mechanical working is In otherwords, the repeatcarried sufiiciently far, and 'undenproper conditions, the tensile strength of the metal may become equal to or greater than that of the best steel I have thus not only devised a new process by which it has becomepossible to work tungsten, as by hammering it or rolling it into the desired form, and by drawing it into fine strong wire, but-I have also produced by the operation of this process a new product, v1z: wrou ht tungsten, and,,if the process be carried far enough, ductile tungsten, a material having properties and utilities difierent from those of any previously known substance. When it is desired to produce from this material an incandescent lamp filament, or any other body which in normal operation is to be operated at highjempertures, I use a special mode of preparation in order to minimize the tendency of the tungsten to revert to a .brittle crystalline ,form. When. this crystallization' becomes excessive the crystals may, in the case of a filament, become so large as to extend across the entire section of the filament and thereupon the sections may move laterally upon each other and produce the condition known as offsetting I shall hereinafter describe more in detail the specialmethod which I employ for minimizin the loss in ductility and for preventing this offsetting effect.

My process divides itself into two stages, first the preparation of an ingot or billet of tungsten, and second the working of the ingot, and I find that the more successfully the ingot is prepared the more quickly and readily it can be worked. Also, by using certain precautions in working I am able to handle ingots which have not been prepared in the best possible way. I have no doubt that with the progress of the art improvements will be made in my process whereby the ingot will be so prepared that its working becomes much easier than it now is, or whereby, through improvements in the working, less care and skill will be required in the preparation of the ingots. I will, however, first describe in detail a process which I have worked successfully on a large scale for the production of drawn tungsten wire of the highest quality suitable for filaments'for incandescent lamps and by means of which a single tungsten wire seven miles long has been produced, and will thereafter describe certain of the numerous modifications of the process,

1 within the J scope of my invention, which may be resorted to as occasion requires. In describingthis process I will refer to the accompanying drawings in which-- Figure 1 shows a swaging machine, Figs.

2 and 3 are detailed views showing different forms of swaging dies; and Fig. 4 is a' view of one form of wire-drawing apparatus, and Fig. 5 shows an electric furnace, and Fig. 6 is a view of an incandescent lamp made in accordance with my invention.

Starting with tungstie oxid (W0 as pure as commercially obtainable, I dissolve itin ammonia water and thus form ammo nium tungstate, which ispurified by crystallization over a steam bath and is then,

substances thus added to the oxid of tungsten, which for convenience I may designate as additions, consist largely of alumina, and silica. I have obtained best results when these additions occurred in the proportion of from 0.8 to 1.5 of one per cent.

If on analysis it 'is found that less than say 0.8 of one per cent. of additions is present, the operation of firing in the Hessian or Battersea crucible is repeated as many times as may be necessary, a new crucible being used each time. The material, which has somewhat melted and agglomerated into a mass, is'ground after each firing.

The next operation is performed in a tube, for example of porcelain or fused quartz, having a diameter of one and onehalf inches and a length of two feet. This tube is packed nearly full of tungstic oxid dry hydrogen passes through the tuheduring this heating operation for a period of say six to fifteen hours, and is preferably introduced througha small passage formed in the upper part of the charge next to the containing tube; which arran ement leaves a thick body-of the charge to e acted upon by the hydrogen. In this operation the tungstic oxid, oras it is called the yellow oxid, is progressively reduced to the blue oxid and 'then to the brown oxid and then to the pure metal. The powdered tungsten thus formed is inserted, :in, weighed quantities, into a steel mold and is by means of a hydraulic. press pressed into sticks or rods which may for example be i of an inch square and "six inches long, though good results have also been obtained with octagonal rods.

The mold is of extreme strength and rigidity and is arranged to apply the pressure transversely of the stick to be formed. The powder is evenly placed in the mold so that the sticks shall be of uniform density from end to end.

W hen the pressure is removed the mold is split or opened and the pressed tungsten stick is gently slid on to a strip for example of molybdenum or tungsten and introduced into. a tube of some such material as porcelain or alundum electrically heated by a winding of tungsten or platinum wire, in which apparatus the tun sten stick is fired in an atmosphere of hydrogen to a temperature of say 1200 degrees (l, for from ten minutes to an hour, or long enough to permit the sticksto become strong, so that they may easily. be handled. The molybdenum support may be withdrawn when the stick has been brought to a bright red heat.

The pressing operation which I have just described is performed with ,dry tungsten powder but I may if desired use a binder which must afterward be removed. If the binder is carbonaceous, the carbon may be removed by heating the sticks in moist hydrogen or if desired in hydrogen containing a small amount of air.

If the-tube in which the heating or firing operation is performed is made of porcelain, I fire the stick on a boat of suitable material such for example as molybdenum.

The fired sticks are then placed, one at a time, in a closed treating vessel, ordinarily of copper and similar to those commonly used in slnteriug tungsten filaments, but

larger and of more rugged construction,

through which hydrogen or other reducing atmosphere is constantly passed, and an electrical, current is passed through them sufliciently intense to raise them to a blinding white heat. This sinters the stick and converts it into a coherent slug of metal. While in the treating vessel the sticks are held in water-cooled spring clamps, one of which is carried by a fixed support while the lower one dips into a water cooled pool of mercury so as to adjust its position as the p with a mercury seal. i

stick shortens up during sintering. In some cases I water cool the bottle itself, and I find it best to 'cover the lower edges of the bottle I prefer with'the i inch rod six inches long to sinter at 1700 to, 1800 amperes (alternating) for fifteen'minutes, the current being gradually raised to this point in say about five minutes, and then to reduce the current by degrees until at the end of one minute or more it is cut off altogether, whereupon the rod may be withdrawnas soon as it cools sufiiciently to allow the treating vessel to be opened. This slow reduction of the current is intended to prevent a sudden redution of gas pressure in the treating or sintering vessel, whichwould tend to draw in air and produce-an explosion. After the sinterin operation, the ends of the slug held by the c amps and therefore but partially sintered, are broken off. This completes the preparation of the slug or billet.

The next step is the mechanical working, which, in the particular process I am now describing, is carried on as follows: In. Fig. 1 A is a swaging machine which may be of the general type shown in United States patent to Dayton, No, 515,576, February 27, 1894. In this machine the head B is rotated and a pair of dies each like the dies shown for example in Fig. 2 or Fig. 3 are brought together and then separated several times in each revolution. The head B carries the swaging dies and their operating mechanism and through the machine extends a hori-' zontal passage D for receiving the work as it passes through the dies. The dies G, unlike the ordinary swaging dies, have very short faces E, about equal in length to the diameter of the tungsten piece passing between I them. The edges C in the working surfaces are rounded or backed off to keep the die from digging into the rod. In some cases a die may be used having a peculiar form of face, shown in Fig. 3, in which the edges of the striking portion or face ofthe die are rounded off gradually. '7

In Fig. 5, F is a tube furnace of the type described in the Winne and 'Dantsizen article above referred to. It consists of a tube G of alunclum or porcelain; a winding of Wire H of platinum or preferably tungsten or molybdenum through which an electric current may be passed, and a heat insulating cover K of magnesite, fine silica or other suitable material held by a suitable shell or casing of asbestos, firebrick, sheet iron or the like.' Hydrogen is passed through the tube G in any suitable way as by the pipe I. The operator grasps with a pair of tongs one of the tungsten bill'ets or slugs and inserts it in the furnace, which is in some convenient location near the swager, and leaves it there until it attains a high temperature, say 1300 to 1700 degrees (1, and at which, when'the billet is removed from the 'furnace into the open air, it gives off a cloud of volatilized oxid of tungsten. He then takes the smoking hotbillet withthe tongs and quickly thrusts it as far as he can into the swaging machine and as quickly withdraws it; Y The swaging machine runs at high speed and care should be taken to keep the slug in rapid motion into and out of the dies so as not to coolitmaterially. The dies are chosen so that each shall produce only aslightqdee crease in the cross-section of the billet and if 1 the billet is square it is necessary to use care in passing from the square to the circular form, which form is of course produced by the action of the dies, and ordinarily the reduced to thirty mils. diameter.

' dies should be so adjusted that several passes through different dies occur in the production of the circular form.

The billet after bein inserte'din the first dies is reheated in the urnace, then. grasped by the other end and, without changin the dies, quickly thrust into the swaging mac inc, and quickly withdrawn as before. As many billets as may be ready for treatment are thus manipulated and then the dies are removed and other dies are inserted arranged to produce a somewhat smaller'section, and

SO OH.

Starting with the inch'square billet, I

am able with five (5) passes .through the machine to obtain a circular rod of 210 mils. in

I diameter. J From 210 mils. to 60 mils. I ordinarily proceed by ten ste s. As the billet becomes longer and smal er in diameter I may use a gas furnace or heater instead of the electric furnace and a smaller size swaging machine, and the billet, which is now a wire or rod is drawn by the action-of mechanically riven feed rollers throu h the gas flame of a gas furnace, hich is t en located in front ofthe swager, and through the swaging dies, passingcompletely through the openin corresponding to. D in Fig. 1, and when t is arrangementfis adopted only one'pass is necessary withv each die settin The swaging operation is continued,. w't continual reduction of the dies, until in six or seven more operations the wire has been The temperature-to which the rod is heated, which was originally at white heat, is reduced at .end infused potassium nitrite, as is disclosed in the Whitney Patent No. 1,008,762, and the pointed end is then inserted in the die and grasped by the clamp '0, which may be heated. at the pointwhere it holds the wire by a gas jet P, or the operator may merely grasp by a pair of pliers the end of the'wire where it emerges from the die and I pull. 'a'few feet of it through. the die and attach the end to a spool or spindle which is revolved by power. In this manner. the whole length of the wire resultin from the swaged billet is drawn through t e die, but before it passes through the die it should .pass through a bath of a material known as aqua dag, which understand .is a defloc;

culated graphite suspended in water, or of oil dag, which is a similar preparation in which oil is used instead of water as a suspending medium. This is done in the manner set forthin the Dempster Patent No.

1,022,182. The die should be at a temperai tureof 600 to 650 degrees C. and the wire should be red hot both when oing into the .die'and when emerging there rom.

. Further reduction is accomplished in the same way, the dies varying by one mil. steps down to 14 vmilsf, makin 16 drawing operations from 30 to 14ml s. At 14 mils. or

thereabout the temperature of the. dies may 0.2 of.a mil. drafts; and from 2 mils. to 1.4

mils. there'are six operations with steps of of amil.;- and from 1.4 mils. to 0.5 mil.

-thereare 18 operations with one twentieth of a mil. draft.

'As the wire reaches smaller sizes the method of pointing described in the Whitcurrent is passed through the wire in such a direction as to make it an anode.

Of course the wire drawing may be stopped at any I stage accordin to the size of wire .desired.

The metal 0 tainable as the ultimate result of the operations above described is a new product differing radically from any form of tungsten heretofore obtained. Not only is it pliable and ductile at room temperature, but it is stron and'tough. It is substantlally non-oxidizin in-moist air at ordinary temperatures an resists the action of acids to an unusual degree. Its specific gravity is higher than that of sintered tungs sten and has been found in certain samples to run from 19.15 to 19.3. It has an exceedingly high tensile strength, as high in particular samples of .00 diameter as 600,000 to 650,000 lbs. per square inch, is non-magnetic and not easily corroded, and is elastic and resilient.

The -'material has a wide variety of tech- I nical uses. If used for-filaments of incandescent lamps. it maybe mounted as described in the Feuerlein Patent, No. 819,009, April-25, 1906, as shown in Fig. 6 of the drawings hereto attached. Before this is done, however, the .wiremay if desired bev freed from impurities on its surface in some suitable manner as by passing it through a bath of hot caustic'soda and then through a uitable wiper. When .run in the lamp the drawntungsten filam'ent made as described memes loses much its strength but is then, if carefully repared in accordance with the directions ereingiven, nevertheless a much strongerfilament than that made by ordinary squirting processes. To guard a ainst this loss in strength, the filament shou d not berun at a very high temperature in the lamp until. the lamp has been sealed off from. the evacuating pump.

If 111 new material is'used as the target of an -ray tube it gives-the tube a life and constancy of vacuum tar superior to that possible by the use of a target of any other known material. It presents many advanta ot or devices where making and breaking of a circuit occurs, and is for such purposes far preferable to platinum and other materials vnow in use. 1

For certain applications the complete working process resulting in an entirely ductile metal of great tensile strength is not required. For example," after a certain amount of rolling or swaging, dependin on the amount of softness and strength w ich the metal should have'for the purpose for which it is to be used, or without any swagworking process.

ing or rolling whatever if great strength is not necessary in. the'finished product, I can heat the metal, for example in an alundum tube furnace, and hammer it into any desired shape, shear it, or roll it into plates, and from these latcs' I can punch parts of any desired con guration provided that the material has been prepared in accordance with the directions above given, and that the operation,

except in cases where the fiber has been very.

thoroughly developed by long-repeated mechanical working, at asufiiciently high temperature. When I have desired to produce a number of identical pieces I have resorted to a drop-forging operation.

The metal produced even by incomplete working difiers markedly from any form known, for example, it is stronger and tougher and less liable to crackin. on unequal heating and cooling. When t e working is fairly complete, the crystals characteristic of'sintered tungsten have been deformed, broken up and elongated to such an extent that the structure is When the working/is less complete the crystals are broken up, though to a less extent, and are deformed as by elongation or flattening, according to the nature of the When, for example, the

" billet is hammered intoa plate or sheet, the

crystals are found to have become flattened. An important point "in practising the above described method of carryin out my invention is the temperature to w ich the metal is raised and at which the several operations are carried on. Where it 1s desired perature be highenough to for use as contacts for magnetos and whatever it may be, is performed,

well fused superficial portions,

"that the resulting metal shall have the properties-oft wrought metal more or less pliable and ductile at ordinary temperatures,

it is not onl necessary to repeat the me- I chanical wor ing a large numberof times, but it is necessary to keep at all times withm a certain range of temperature at least during the later stages 'of working. It is necessary, on theone hand, that the temmake it possible to mechanically work the body, but, on the other hand, too high; a temperature is obj ectionable because if the temperature is too high a tendency appears to lose fiber already acquired and to pass to a state of brittleness from. a state of ductility. The range of temperatures within which properly re pared tungsten bodies may be works to produce ductility at room temperature in accordance with my invention varies with the particular material used, but according to my experlence heating the metal to or wor ing it at a temperature correspondin to white heatv and materially in excess of a out 1500 C. will tend to revent, with any tungsten metal which I- ave been able to produce, the formation of fiber and will tend to destroy fiber which has been produced by previous working. On the. other hand,

an excessive amount of mechanical working 7 tends to cause the bodies to become so fibrous and hard as to split up and break. I have found that this result may be obviated by carrying on the mechanical working during the early stages at such temperatures as enable the body to be worked and reduced in size but without causing any special change inthe structure of the metal. These temperatures may correspond to a white heat or thereabout but-should be above a temperature which I estimate at about 1500 degrees C. At the proper stages of the process the temperature is reduced to 1200 degrees (3., or lower, whereupon as the mechanical working progresses the improvement in the structure of the metal begins.

The preparation of a tungsten body which is to be mechanically worked 'is an important feature of my invention, for it is not every tungsten body which is. susceptible of mechanical working. For example, Moissan states (Uomptes Bend-us, vol.123,-\pp. 13-16, 1896-7) that he obtained by heating tungstic acid and sugar carbon in a carbon crucible an ingot or melt which presented but which. was porous in its interior portions. Of this product Moissan says that when it is in a porous condition it has, like iron, the property that it may be consolidated (la prop'm't de se suder d Zui-mme) by hammer- 111%!!!11011 below its point of fusion.

ubsequent investigators have expressed the opinion that the material which Mo-issan obtained was notgpure tungsten, but tungsten contamina by carbon. there is no indication that by hammering Moissan was able to do an l i more than v to close more or the pores the porous material, whatever it may have been, to which he refers; nor is there any indication that when he had gone thus far he had roduced a material which could be work I have found thatin order that the ewe ing'process maybe carried on su y it 1s necessa that the tungsten should be in a certain p ysical or molecular condition, such as is brought about b the treatments which I have discussed 9. 've. I find it particularly advantageous 'to work with tungsten powder which is coarse as distinguished from the exceedingly fine and impalpable powder such as has been used in the manufacture of lamp filaments. The 0 ration of heating in the Battersea cruci b e coarsens the powder, as does also the reduction operati'omparticularly if it be carried on in the manner described, and if the hydr en be passed rather slowly through the reduction furnace. While my invention has made it possible with precautions to work fine grained meta-l (especially if it contains oxids, which delaysintering), as by excessively small drafts and by surrounding the rod with a sheath or casing of some other metal, the condition of thetungsten which is characteristic of the coarse powder is one which ispeculiarly adapted to the successful practiceoof my invention.

. Though the operativeness of my invention obviously does not depend upon any theory of operation, I may state that I am at present of the o inion that the difliculty which is observe in swaging the rods d from the fine powder may well be c?- to the presence'in such rods of ce impurities, probably gaseous impurities, as

- for example oxy t which would probably occur as 1: of, tungsten. The atomic weight of is so high (18%) that evcn a small pe ex st as the oxid, It seems p the sinter 1;; p the ties are more t a. 1

in ets characteristic of coarse grained powder are produced by any process by which a is formed which is coarse at the time it -n' out. 4

cessfully carrie. have stated above that certain additions are tags of w willcause a yery large-pentage of th hesis is 3:4

names is exposed to thesintering operations For example, I have produced rods from exceedingly fine tungsten powder which,'after sinterin'g, swage very well, by carrying on the baking operation in an iron tube furnace. The rods are packed in the tube in powdered silica and hydrogen is passed through the tube. The tube is heated by gas flames for-8 or 10 hours to a rather low temperature. This baking operation produces a rod' of coarse grained structure, probabl because of the act that the oxid, natura y existing in considerable quantities in so fine a powder, requires prolonged heating to effect complete reduction and coa'rsens during this pro onged heating in the oxid stalge. I

- t is desirable in all cases that, the smtering operation should take place slowl in or e r to give time for the action 0 the hydrogen to penetrate through the entire mass, and to facilitate the escape of im urities,as well as-to revent the center 0 the rod, by a too rapi increase of the heating current, from becomin much hotter than the outside and actua y melting and run- T e ultimate temperaturereachedin the sintering operation should be very h, a dazzlin white heat probably not far om 2800 Itisdiflicu1ttog1veany, eral rule,except to state that if an ins ciently' high temperature is reached the rod will break up in the swaging machine. Such a rod ordmaril shows a very finely fracture, w e my experience is t at the rods which are capable -of being ;worked easily'show a somewhat coarser and more 'c stalline fracture.

t. is my belief that certain impurities in thesintered rod, such for example as oxygen and carbon, are highly detrimental and prevent the sw r operation from being sucon. On the other hand I absorbed from the Hessian or Battersea crucible operation, and I find that these additions do' not revent the 1. 1; while they exerciw a highly beneficial action where the tungsten is to be exposedfor long periods of time to high temperatures, particularly in the form of incandescent lamp filaments. It may be that ofisetting, a he" ined nomenon which asabove explained is c aracteristic of some drawn wirelamp filaments, is due to the crystallization of the metal and the segregation of im urities, whereby ot s are developed at-which this side slipping or offsetting occurs, and that the presence of the additions introduced from the Hessian or Battersea crucible retards or interferes with the diffusion of such impurities and thus retards crystallization Y 5 materials suc as t oxide of thorium, zirconium, yttrium, erbigm, didymium, or ytterbium. I have use with especial success thorium nitrate, which gives thoria when decomposed. A small amountof the above mentioned oxids, 'or oflc'om'pounds or solutions which upon heating or other treatment will yield these oxids, maybe mixed with the tungstic oxid'before reduction or with the metal powder afterreduction, or may 5 even lie-introduced after the metal has been made into a stickfeither by soaking the stick in a solution of a suitablesubstance orbyheating the stick in the presence of a suitable material in the vaporous condition.

When thorium nitrate is used, for-example,

- I employ an amount which is equivalent to 2% or less of thoria referred to tungsten.

The essential point seems to be,that the material should be introduced before the 5 tungsten is sinter'ed. ,Suchadditions, particularly in the case of a filamentary body, may actually roduce a certain amount/of ductility and exibility in the sintered filament without mechanical working. Usually,

) however, instead of directly adding thesesubstances, I prefer to employ the treatment above described 9f simply firing the tungstic oxid in a Hessian or Battersea crucible, for this treatment is simple, successfuland reliable.

While I have referred to a considerable extent to the manufacture of ductile tungsten in which so-called additions are introduced at some stage of the process in order that the resulting product may be especially valuable when used for filaments of incandescent lamps, and in connection with other uses where the metal is to 'be exposed forlon'g periods of time to high temperatures, nevertheless operative and useful incandescent-lamp filaments may be made by carrying out my process as herein described, omitting the additions. whether produced by direct introduction of additional material, as above described, or by the Battersea-crucible process; and where it is sought 'to. obtain ductile tungsten for uses in which it is not likely to be exposed to high temperatures, the introduction of such additional, substances may. be entirely unnecessary. I do not however wish it -to be understood that metal prepared with the additions can'only housed in incandescent lamps and for similar purposes, as such metal can,-'a's far as my present knowledge extends,be used for any purpose to which the ductile tungsten prepared without the use of the additions is applicable.

Incon'nection with the swaging operation t should be noted that I have departed radiout using:

cally from the ordinary swa 'n rocess, in which the dies are long and ti e 513511 is held a considerable time in the machine or assed very slowly through it, so that a num r of blows are struck at the same place, on the metal, and a smooth surface 18 produced,"

I find it highly desirable, particularly in the earlier stages of the swagingproce'ss, to use a die with a very short working face such as I have shown in the drawings, and

to .pass\ the metal through the machine at high speed and so that it is still highly heated when it emerges from the dies. Indeed, it is best not to allow. two successive blows to be struck at too closely overlapping places,

because each impact of the die abstracts a certain amount of heat from the material and unless this precaution be observed there is great danger that a blow will be struck upon metal which is too cold to endure it without cracking. Preferably, then, the metal emerges from the swagin'g machine, not smooth, but showing every distinct'blow of the hammer, and therefore with an irregular wavy surface.

The swaging operation, though the most successful at the moment, is .not the only operation by which my invention can be practised. For example, I have succeeded, without 4 any swaging operation whatever and using only wire-drawing dies, in producin perfectly ductile tungsten. In this case have begun the operation with sintered conductors of relatively small size. For example, I have used ordinary sintered series tungsten incandescent lamp filaments of the diameter of 25 mils. Where the original body is so small in diameter it isi'elatively easy to expel injurious impurities in the sintering process, and the devices above described to facilitate this end, which are of importance in'the case of the larger. bodies whlch are required for economical largescale production of lamp filaments, or for the preparation of larger articles, such as Xray tube targets, projectiles, contacts, etc., become relatively unimportant wherethe original sintered body is of su'fliciently small diameter. I pointthese filaments by the process of the Whitney Patent, No; 1,008,762, above referred to, and carefully insert the point into a die which is heated to 600 to .the filament is then with great care and with .a regular steady die, the diameter 0 than A; a mil. (.0005 inch less than the maximum diameter of the filament at any point. The filament is then in a similar way drawn pull drawn through the through a die having a diameter of A; a mil. (.0005 inch) less than that and so on down to 13' mils. Below this size the dies'should vary in'steps of one-quarter of a mil. until a size of say four "mils. is

which should be not less of the first die,

led

C; at mils. the tem erature is reduced to 400 C. The diesarelu rlcatedwith aquadag or oil da as in the wire-drawing process first described. After such a mil. filament has been drawn to a diameter of 7 mils it can,

at room temperature, be bent around the finger without breaking, and will readily take a v permanent set. As the tungsten is stillffurt-her worked by drawing down to smaller sizes, the fibrous condition becomes more and more developed and the wire becomes more and more pliable, until, at a diameter of about 4 mils., it is in every sense ductile material, and could if desired be manipulated as? any .other ductile metal, and drawn through dies not heated above room temperature, though I ordinarily heat the dies during the last stages. of the drawing process as well as during the first stages, since 26 a rendered somewhat easier.

by the application .of heat the drawing is As a convenient means for keeping the tungsten hot during the drawing operation, I may, if desired, pass current through the material under treatment. The die and the portion-of the wire which is hot may be inclosed in an evacuated space ,or in a hydro- .gen or other reducing atmosphere, although I do not find it necessary to do so. The lubricants mentioned herein themselves materiallyassist in protecting the wire from oxidation. So also .I have succeeded in producing ductile tungsten by a rollingprocess. In carrying out such a process, at least until the metal has acquired a considerable amount of fiber, the-rolling operation should be carried on with the metal at a high temperature, preferably between-what is known as bright red heat and white heat, and

that the metal should retain this temperature as long as the rolls are acting upon it. Excessive temperatures, such as temperatures above 1500 C., are objectionable for the reasonsand to the extent already explained. The rolls naturally abstract heat rapidly and therefore the metal should pass through the rolls at veryhigh s eed; it is also. essential that the draft -0 the rolls should not be too great. Also,.at least during the later stages of rolling, the temperature should not be higher than about 1200 C; and preferablyin the neighborhood of 1000 C. When a sufliciently fibrous condition has been obtained by rolling, the metal may be run through dies as above described, or the working may be continued exclusively by rolling.

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

1. The process of producing tungsten havthe desired shape:

assa 3. The process of produc' ductile tungsten wire which consists in orming a sintered body of tungsten, hot swaging said body repeatedly until it becomes fibrous in structure, and then further vreducing it by hot drawing.

4. The process of producingductile tungsten wire, which consists in first forming a sintered body vof tungsten free from oxygen carbon and other impurities which would render the body unworkable mechanically, then hot swaging such sintered-and purified body repeatedly until it becomes fibrous in structure and then further reducing it by hot drawing.

5. The rocess which consists mechanically wor ing while hot-tungsten which is brittle at room temperatureuntil it becomes pliable at room temperature.

6. The process-of forming bodies of coherent homogeneous tungsten which consists in agglomerating tungsten powder, sintering the body thus formed, and then subjecting it to hot mechanical working to such an extent as to improve the internal structure and to bring the bodyinto a shape adapted for use in the arts. a

7. The method which consists in first producing a body of tungsten sintered or com-' pacted throughout and free from such impurities as would render the body unworkable mechanically, and the'nfhot working said body-to such an extent as to deform its crystalline structure and to bring it into a shape adapted for use-in the arts.

8. The method of producing pliable tungsten wire for lamp filaments, which consists in first producing a body of tungsten in a sintered state and free from oxygen carbon and other impurities whichxwould render the body unworkable mechanically and then subjecting this sintered and purified body to-mechanicalworking-many times repeated, with gradual reduction in diameter and elongation in length. x

9. The method which consists in agglomerating relativelycoarse tungsten powder into a porous mass, eliminating harmful-impurities and compacting the mass by a sinterin operation, and then mechanlcally working the resulting body while hot into 10. The method of producing ductile tungsten which consists in sinteringan agglomerated massv of 1 tungsten: particles or powder into a' dense coherent and homogeneous form in such manner as to allow elimination of harmful impurities,,and then mechamcally workmg'said body until it is and then subjecting the body to hotmechanical working.

13. The rocess which consists in firing tungsten oxid in a refractory crucible of the .Hessian or, Battersea. type, reducing the oxid to coarse metallic p'owder,'a glomerating the powder into a coherent ody, sintering the body, and then subjectingit to hot mechanical working.

14. The process which consists in forming a body oftungsten'powder containing add1- tional material which will prevent coarse crystalli'zation of the tungsten athigh temperature, sintering' the body, and then sub jecting it to hot mechanical, working."

15. The step inkthe preparation of tung sten bodies which consists in firing tungsten oxid in a refractory crucible of the HessianorfBattersea type, and subsequently Eeducing the oxid to metallic'tungsten pow- 16. The process which consists in producing tungsten containing beneficial additions, formlng it into a billet, sintering it at high temperature, and mechanically working it. a great number of times at high temperature, and; reducing the temperature during the workin v 17. Tie method of producing bodies of tungsten pliable and ductile at ordinary-V or room' temperatures from bodies of tungsten which are brittle at ordinary or room temperatures; which consists in heating said brittle bodies to a temperature suflicientto render them susceptible of mechanical working, repeatedl working the bodies while so heated, and uring at least the last stages of, mechanical working carrying on the workin attemperatures such that the mechanica working has the effect of produc-.

ing ductility in the finished product.

ductile tungsten from bodies of tungsten which are non-ductile, which consists in mechanically working the non-ductile bodies of tungsten within a temperature range at the lower limit of which the body to be worked is soft enough to be susceptible of.

such working while the up or limit, at least during the ater stages 0 theworking is s'ufliciently low tocause the body t us worked to retain to a greater or less degree the eflect of the mechanical working.

19. The method of producing ductile tun sten which consists in repeatedly hot wor ing a tungsten body within a range of temperature in which the body is suificie'ntly soft to'be susceptibleof such working, but maintaining'the temperatures below those at which the efl'ect of the mechanical working islost. v

20. The method of producing a body of ductile tungsten from a body of tungsten which is brittle when cold, which consists in mechanically working said bod -while heated to a temperature at which "the body is, softened and finishing the mechanical working at temperatures below those at whichno ductility is imparted to the metal.

by the mechanical working.

21. The method which consists in mechanically' working at a temperature above approximately 1500 degrees C. a tungsten body which is brittle, cold, interrupting the operation before the body is' brought into the form ultimately desired, and "completing the mechanical working at a temperature below 1500 degrees C. a

1 22. The method of securing a ductile metal body without previous melting of the metal, which consists in producing a'compact body of metal from divided metalliferous material, consolidating said body by heating to a temperature less than the melting temperature. and such as to produce a body of metal non-ductile when cold, subjecting the body to repeated mechanical working while in a heated condition, and

continuing the working until the body is ductile without the application of heat.

23. The method of securing a ductile metal body without previous melting of the divided met'alliferous material a compact body of metal, sintering said body and thereby producing a crystalline body brittle at room temperatures, and then subjecting metal, which consists in producing from the body to repeated mechanical working while hot, and continuing the working until the metalremains ductile when cooled.

24. A wire formed of ductile tungsten.

25. An incandescent electric lamp having a filament of drawn tungsten wire.

25. Substantially pure tungsten having ductility and high tensile'strength. 18. The method of producing bodies of 27. A ductile tungsten wire having a 29. An incandescent electric lamp having a filament formed of ductile tungsten.

30. A lamp filamentof tungsten which 3.10 1,osa,eae

resists I I stlglngthoiiifsettmg and has high tensile gofirzgter, and capable of being forged and A lamentary conductor of drawn 34. Wrou ht tun ten a soli c tungsten so constituted as to be free from material chiracteriz e d by the 265223322 15 e oflsettmg when operated as a filament of crystals deformed by mechanical working. an incandescent lamp. 3 y In witness whereof, I have hereunto set 32. A filamentfor electric ncandescent my hand this 18th da of June 1912. lamfis conslstlng of tungsten w1re which is WILLI D. CdOLIDGE.

plia 1e at ordinary or room temperature. Witnesses: 10 33; The maternal wrought tungsten; ha'v- NEWTON A. Buxenss mga specific gravity of approximately 19 HELEN Onrono.

DISOLAlM'EZR.

1,082',933.;William D. Coolidge, Schenectady, N. Y. TUNGs'rnN' AND METHODS OF MAKING THE SAME FOR UsE as FILAMENTS or INCANDESCENT ELECTRIC LAMPS \-A-ND FOR OTHER PURPosEs. Patent dated December 30, 1913. Disclaimerfiled J anuary 25, 1929, by the assignee, General Electric Company. Herelby disclaims the subject matter of claims 14,- 16, 24, 25, 2'6, 28, 29, 30, 3'1, 32, an 33. y Your petitioner further hereby disclaim the subject matter of claim 27, which" reads as follows: V

A ductile tungsten Wire having a fibrous structure. exce t when theproduct defined in said claim 27 has been produced by the process disc osed in the specification of said patent, as more particularly defined in claim 5 of the said patent. 1

, [Qflieial Gazette February 5, 1929.1 1

Images