Edison tried carbon fiber in 1879, and it took hundreds of hours. Although carbon has a very high melting point (3550℃), it has a low sublimation temperature. Directly sublimes from solid to gas at low temperature, so it is easy to consume and has a short service life. And it must be completely isolated from the air (it will burn in the air). Almost all tungsten wires with melting point of (34 10℃) were used before the project. The advantage is that its "sublimation" rate is lower below the melting point. So it can be heated to a higher temperature than "carbon wire". Tungsten wire will also burn in air, so it is necessary to vacuum the bulb.
In order to avoid sublimation of the filament, inert gas is injected into the bulb, which is mainly argon and does not contain oxygen. The partially evaporated tungsten atoms can return to the filament by collision. Although inert gas increases the service life of filament, it also pays some price. Because of the existence of inert gas in the original vacuum bubble, the conduction and convection of heat are increased and energy is taken away, thus reducing the equilibrium temperature. The sublimated tungsten gas forms weak particles in inert gas and forms black spots on the inner surface of the bulb by convection.
1903, according to the patents of A. Just and F. Hannaman, Hungary made tungsten wire for the first time. In this method, carbon filaments are heated to a high temperature by current in halogen oxide vapor of tungsten containing free hydrogen, so that carbon is completely replaced by tungsten. The incandescent filament made in this way contains carbon more or less, which is not only very brittle, but also the filament of the bulb is constantly dense during use, so the electrical parameters of the filament will change.
In 1904, Jester and Hannaman realized the influence of carbon on brittleness, and used a carbon-free binder to mix with tungsten compounds, then extruded into filaments, and then heated and reduced to metal in hydrogen. The tungsten wire made by this method is very brittle, but because of its much better light efficiency, it can still replace carbon wire, osmium wire and tantalum wire to make light bulbs.
None of the above methods can prepare fine tungsten wire. In order to solve this problem, 1907 developed a tungsten alloy with low nickel content. It is prepared by machining, but its serious brittleness hinders its application. Until 1909, W.D. Coolidge of American General Electric Company manufactured tungsten blank by powder metallurgy, and then produced tungsten wire with ductility at room temperature by mechanical processing, thus laying the foundation for tungsten wire processing industry and powder metallurgy.
However, this "tough" tungsten wire shows obvious brittleness after the bulb is ignited. 19 13 years, Pinch invented thorium tungsten wire (ThO2 content is 1% ~ 2%), which greatly reduced the brittleness of incandescent filament. At first, the filament drooping (see the anti-drooping performance of tungsten filament) was not a problem, because the filament was straight at this time, but after 19 13, Langmuir changed the straight wire into a spiral wire, so that when the bulb was used, the filament drooped due to high working temperature and self-weight, and pure tungsten and thorium tungsten could hardly meet the use requirements.
19 17 years, in order to solve the problems of drooping and short life of tungsten wire, A. Pacz invented the high-temperature "non-deformation" tungsten wire. At first, when he was preparing pure tungsten, he baked WO3 in a refractory crucible. Unexpectedly, he found that the tungsten wire spiral made of this WO3-reduced tungsten powder was mysteriously no longer drooping after recrystallization. Then, after repeated experiments of 2 18, he finally found that adding potassium and sodium silicate to tungstic acid (WO3·H2O), after reduction, pressing, sintering, processing and recrystallization, a tungsten wire with a fairly coarse grain structure was formed, which was neither soft nor sag-resistant. This is the earliest sag-free tungsten wire. The discovery of Perth laid the foundation for the production of non-drooping tungsten wire. Until now, the United States still calls the non-drooping tungsten wire "2 18 tungsten wire" to commemorate this important discovery in Perth.
However, the brittleness of the earliest non-drooping tungsten wire is more serious than that of thorium tungsten wire, so that some bulb factories insist on using thorium tungsten wire as filament. However, with the continuous development and improvement of the production technology of sag-free tungsten wire, people gradually realize that adding compounds of K, Si and Al into tungsten oxide can make tungsten wire have good sag resistance at high temperature and satisfactory room temperature ductility after recrystallization. This is what people often call "AKS tungsten wire", that is, "non-drooping tungsten wire" or "doped tungsten wire". In 193 1, T. Millner called this improved non-drooping effect "GK effect".