US1001105A - Manufacture of electric filaments. - Google Patents

Description

UNITED STATES PATENT OFFICE.

CARL A'UEB VON WELSBACH, OF VIENNA, AUSTRIA-HUNGARY.

MANUFACTURE OF ELECTRIC FILAMENTS.

Specification of Letters Patent. No Drawing. Application filed August 24, 1899, Serial No. 728,356. Renewed August 21, 1909. Serial No.

Patented Aug. 22, 1911.

To all whom it may concern:

Be it known that I, CARL Anna VON WELS- EACH, :1 subject of the Emperor of Austria- Hungary, residing at Vienna, Austria-Hun- 5 .gary, have invented or discovered a certain new and useful Improvement in the Manufacture of Electric Filaments, of which the following is a specification.

Incertain applications for Letters Patent of the United States filed by me August 9,

1898; Nos. 688,201; 688,202; 688,203; 688,204; 688,205; 688,206; and 688,207 Ihave described the production of filaments for electric vacuum lamps, consisting entirely of osmium or of osmium in alloy with another metal or metals, or incorporated with or coated with refractory oxids. In the manufacture of said filaments, according to one of the methods described, the carbon present in the filamentary body at one stage of the operation, is finally. eliminated in the presence of a gas of such'a character as to combine with and remove 'the carbon without oxidizing the metal, or reducing the metallic oxid or oxids. f The gas that I prefer for this purpose may be derived, for instance, from the mixture'of gases vapors incident to the incomplete combustion of illuminating gas by the ignition thereof at the air inlet openings of a'Bunsen burner, the said mixture of gasesLand'i-vapors being collected from the burnertube. They consist mainly ofhydrog'en, carbon monoirid, and carbon dioxid, together with a quantity of watery vapor.-- I have ascertained that filaments so prepared contain, in the body of the osmium, aicon siderable quantity of occluded gases, fa's, for instance, hydrogen, carbonic oxid, carbon di-oxid and hydrocarbon gases, and

40 that these gases arenot readily given 01f the osmium when heated, its behavior in this respect being in strong contrast to that of many other gas-absorbing metals, which readily part with their occluded gases when subjected to heat. In fact the osmium retains certain gases, as, for instance, acetylene, even at high temperatures w1th such I tenacity that we are almost justified in speaking of the combination, asan osmium- 50. gas alloy. The filaments (of osmium, os-

Ifium-alloy, or composite osmium and ofxid), retaming their occluded gases at the tiemperatures' to which the filaments have been subjected before being sealed into the vvacuum bulbs, thereafter, at the much higher temperature incident to their actual use as lllummants, glve off the said gases,

which thereupon issue into the space sur-! rounding the filaments, interfering with the vacuum, or otherwise disturbing the normal conditions that should prevail for uniform and reliable service. 7

The purpose of my present invention is to substantially rid the filaments of these occluded gases, before the bulbs have been finally sealed, so that thereafter their otherwise disturbing influence shall be eliminated. I will first describe the application of my invention to the elimination of occluded gases from a filament of pure osmium: .When the filament for the bulb is to be of pure osmium, I mount it therein by fusing the bulb to the portion of the lamp which carries the filament and leading-in wires taking care to first carefullyclean the bulb itself. I thenfill the bulb with a dry gas 4 such as nitrogen which will not attack the osmium while hot, and then, while highly heating the bulb, I exhaust the gas therefrom. Thereupon, I pass a current of such a dry gas through the bulb and again exhaust, repeating'Tthis latter alternate gasrinsing and exhausting operation several times. Finally, after the last gas-rinsing and while the pressure in the bulb is low, I

pass an electric current of say 2 to 4 watts through the filament, sufiicient per candle, L 0

to a corresponding luminosity,

to bring. it

and then I'complete thevacuum. The filamentbegins to slowly give off its occluded gases, and its luminosity increases until t attains a dazzling whiteness. As soon as it com letely ceases to give oli gases, I seal the ulb, whereupon it will be found that the exterior appearance of the filament has not been affected "'n any able manner, and a substantially perfect vacuum obtained which will besubstantially maintained throughout the life of the lamp.

If the lamp is to be used in an inclined position, or under circumstances" sub ecting it to jolts or jars, the filament mayv be attached at any place'within the bulb, in manner similar to the way in which carbon-filamentsfor high tension'lamps are attached; To eliminate the occluded gases (without particularly noticethorium, oxid of zirconium, oxid of yttrium I proceed as follows: I first mount the said.

' filaments in the bulb, .as before, by fusion.

I then rinse the bulb by passing a current of. dry gas such as nitrogen through it and thereupon exhaust it. I then pass a current of electricity through the filament suflicient to first raise the filament to a red heat, gra dually increasing the current until the filament is raised to a yellowish white heat, in the meantime continuously, or from time to time, increasing the vacuum. During this operation, the filament gives off its occluded gases hereinbefore specified and a faint 'glimmermg takes place within the bulbfor a considerable period. As soon as the development of gas ceases permanently, but not until then, I increase the strength of the current sufliciently to bring the filament to a bright white heat, and, as soon. as the resistance becomes constant, I seal the bulb. Care must be taken, in this process, to prevent energetic glimmering, as otherwise the interior of the bulb will be covered with a brownish coating.

The process just described for obtaining said filaments of associated osmium and oxid free from occluded gases maybe replaced by the following process which requires less time: An osmium filament first fastened inthe lamp-base with osmium cement, or preferably iridium-osmium cement, as contemplated in one of my applications hereinbefore referred to z'. a. Serial No. 688,204. It. is then inserted within a closed receptacle anda' vacuum is produced within the receptacle. The filament is then raised to a yellowish white heat by the electric ,cur-

rent, and, as soon as glimmering begins, nltrogen w1th some oxldlzmg agent, as, for mstance, small quantities of air, (or, preferably, steam or carbonicacid gas,) or s1m1-' lar gases or vapors, are admitted, for the purpose of uniting with the occluded gases as they are given off by the filament. As soon as this union has taken place, the vessel is filled with dry' nitrogen. The filament is then coated with oxids described in my former applications referred to and dried quickly, and the lamp base isunited with'the bulb, whereupon, to finish the lamp,

gases from the osmium but not suflicient to produce a temperature at which the gases would combine with-and reduce the oxid. Thus, according to the first method above described for treating the said composite filaments,'the filament is first raised to a red heat and finally to a yellowish white heat, and the occluded gases given otf are continuously exhausted by the vacuum pump; and it is not until the development of gas has entirely ceased that a higher temperature is employed. So also in the second method described for producing a con'iposite osmium and oxid filament devoid of occluded gases, the occluded gases are first eliminated from the osmium filament before the oxids are applied to said filament. The resulting filament in both cases is, therefore a filament consisting substantially of osmium either admixed with or coated with an oxid practically unreduced. I may so conduct the operation, however, as 'to'produce a filament wherein the osmium is associated with or coated with a combination of varying composition, varying according to the amount of. oxygen withdrawn from the oxid of the original'oxid and osmium filament treated. These combinations are produced'either by a partial reduction of the oxid with which the osmium is associated br coated, or by a partial oxidation of a metal-osmium-- filament. these filaments exhibit a nearly metallic appearance and appear to be entirely homogeneous. Their conductibility is in general greater than that of the original filaments of oxid and osmium, but less than that of the metallic osmium alloys.- They occlude gases in aboutthe same degree as do the osmium filaments themselves, and in the gas mixture of the Bunsen flame, which contains watery vaporor steam, they are ap- Externally patently not appreciably altered even when and from the bulb-itself by the action of the vacuum pump, whereby the vacuum in the bulb is attained with corresponding rapidity and with the assurance that it will be permanent,'but also that the degree of the reduction may be predetermined and controlled with great accuracy.

The process of producing the said filaments by obtaining a partial reduction of the oxids by the occluded gases is quite similar to that for the production of the oxid. and osmium filaments. Ordinary 0x1 and osmium filaments from which they artr produced, for instance filaments consisting of osmlum associated with thormm oxld are first mounted, by fusion, in the lamp bulb.

' The bulb is then rinsed with. dry gases as,

for instance, nitrogen and filled therewith, whereupon it is connected with the vacuum pump and the exhausting operation is ben: The filament is then broughtto a bright incandescence, the rather high resistance of the filament remaining-constant.

The development of gas begins and e\'entually the lamp begins to glimmer. When, a fter a considerable interval, the development of gas slackens and an almost complete vacuum exists. in the bulb, the strength of the current is increased until the filament is brought to a dazzling white heat. Presently the resistance of the filament suddenly sinks,

the temperature required for reduction having been reached, whereupon a correspondmg reslstance must be switched 111 or the tension of-the current must be diminished.

The development of gas' then ceases and the ter some time becomes constant.

glimmer becomes extinguished. The resistance gradually sinks still lower, and af- The filament contracts and the vacuum in the lamp being completed it is then to be sealed. Ac-

cording to the. quantity of the gases here.-.

inbefore referred to-present in the osmium filament at the time of thebeginning of the reaction, and according to-the quantity of oxid present in thefila ment itself, the resulting conductibility of the" filament. is

higher or lower.v Inasmuch as in practice, it is often desirable to have a high resistance, I preferably give to the filament as high a percentage of oxid as possible. But,

in View of the fact that filaments with a high percentage of oxid require for their production high tension currents and inasmuch as this percentage in many cases '(particularly for thorium oxid) should not materially exceed fifty per cent. because the conductibility of the original dxid and osmium filaments in the cold diminishes as the percentage of oxid increases and with even greater rapidity, it is therefore advisable 1 not to give a filament too" high an original percentage ,ofoxidsfbut to terward saturate it with the ox'id and t' coat it, Itwill 7 ofcourse be understood that osmium maybe .advantage of making the filaments denser,

commingled with the oxid in the saturating or coating solution or emulsion. When this process 13 correctly. carried out it has the so that they will not glimmer materially during the subsequent formation period i the lam I The oxid or mixture ofosmiu and oxi applied to the filament for the pure, pose of saturating and coating the same is preferably, in the finest subdivision, made into an emulsion: The filament is'thereupon dipped into the emulsionthroughout the en-- tire length of the filament excepting the ter- The filament is:-

ament within at protecting gas mixture and to bring it to a white heattherein. In this manner we obtain a core of better conducting oxid and osmium 'and a coating of less conductivity; Instead of using oxid and osmium filaments, it will. of course be understood that filaments of pure osmium may be used; in the latter case, however, it

is advisable not to make the filaments too dense which maybe readily guarded against by making the filament out of lime and osmiumpaste and by not heating the filament -too high subsequently when it is being consolidated. In this instance, the saturation and coating of the filament takes place in the manner just described, and the further treatment of the saturated and coated filament, up to the completion of the lamp, is

' likewiseithe same.

I have found that thorium oxid is the oxid best adapted for the production of these filaments of partly reduced oxid and osmium or partly oxidized osmium, It may,

however, be entirely or partially replaced by OXlCl of zirconium, the ytterite earths and similar refractory oxuls whose partially reduced oxid when associated with osmium resists the high temperatures employed in practice and these substitute materials may be employed either alone -or in admixture with each other in place of the. thorium oxid,

although, as I have intimated, withlessadvantage. IVhe'n oxids of the highest capacthe filament may likewise contain oxide which of't-hemselves would melt or volatilize at= white incandescence, as for instance, lime; the greater the percentage of the oxid which fully resists volatilization, the more firmly will be held the-oxid which resists this volatilization to a lesser degree. In all of these filaments, osmium may be replaced in part by other platinum metals, as ruthenium, iridium or rhodium, without essentially. altering the operation; 1 Such' filaments, however, in case they contain considerable quantities of the metals referred to, have not a very high capacity for resisting high emperatures. Nevertheless, they are chardescence in. the openair.

Filaments" produced inaccordance withthe processes hereinbefore set forth are, by reason of the practical elimination of the cterized by the remarkable property that, n spite of a considerable percentage in 0s 1 mium, they do masher a time, give off os- -mium vapors when brought to white in'can ity to resist high temperatures are employed,

occluded gases wholly or substantially free from glimmering when brought to white incandescence.

Having thus described my invention, what I claim is:

1; The process of eliminating occluded gases from a filament containing osmium and at the same time forming an osmiumalloy filament, which consists in coat-ing the filament with a magma containing metallic oxid in such proportion as to be reduced by the occluded gases, then expelling the occluded gases by passing through the filament a current of electricity sufficient, for that purpose, and finally raising the temperature of the filament to a point at which the oxid will be reduced by the gases and the metal of the oxid caused to alloy with the osmium.

2. The process of eliminating occluded gases from a filament containing osmium and a metallicoxid, which consists in raising the temperature of the filament to a degree sufiicient toyexpel a portion of the gases therefrom, but lower than the reduction temperature of the oxid, and thereafter raising the temperature of the filament still higher and effecting the desired reduction thereby. 7 j

3. The process of eliminating occluded gases from a filament containing osmium and a metallic oxid, which consists in-passing a current of electricity through said filament of a strength sufiicient to gradually expelthe gases but below the reduction temperature of the oxid, progressively withdrawing the gases thus developed, and finally, before the development of gases has ceased, increasing the strength of the cur rent to the reduction temperature of the oxid.;.and thereby effecting zupai-tiah educ tion of the oxid by the remaining gases.

4. The process of making an incandescent the gases butbelow the reduction temperalamp having a filament containing osmium from. which the occluded gases are to be eliminated, which consists in first drying and 4 rinsing the bulb by alternately admitting dry reducing gases therein and exhausting said gases from the bulb, then admitting a further quantity of reducing gases into the bulb and exhausting the same, passing a current of electricity through the filament during the last mentioned period of exhaustion and thereby expelling the occluded gases from the filament, and finally increasing the strength of the current so as to bring the 5 filament to white incandescence and sealing the bulb when a substantially complete vacuum has been produced within it.

5. The process otsubstantia-lly eliminat-' ing the gases from a filament containing a metallic oxid and osmium wherein such gases are occluded, which consists in disengaging them by slowly bringing the filament up to a temperature sutticient to drive ofi ture of the oxid, and finally increasing the temperature to white 1ncandescence,-thereby 7 partially reducing the oxid.

6. The process of substantially eliminatmetallic oxid and osmium wherein such gases areoccluded, which consists in slowly bringing the filament up to a temperature at which the gases will exercise a reducing action on the oxid, thereby forming gaseous combinations with the oxygen thereof, and removing said gaseous combinations.

In witness whereof I hereunto attach my signature, in the presence of two subscribing witnesses.

CARL AUER- vox \VELSBACH.

Witnesses: r

Anonr G.\LLI.-\, L. Hammer-2R.