HISTORICAL NOTE Materials for Incandescent Lighting: 110 Years for the Light Bulb In 1879 Thomas Edison gave a gala exhibi­ In 1850 Edward G. Shepard invented a while, introduced extruded cellulose fila­ tion that showed the world not just his lamp that used only incandescent char­ ments for light bulbs in 1883, but by then electric light bulb but his complete electri­ coal, and about the same time English Edison had made his name. cal system for making it a practical, com­ physicist Sir Joseph Wilson Swan created At his laboratory in Menlo Park, New mercially viable way to illuminate entire carbon filaments out of paper. In 1856 C. de Jersey, Edison gave the first public demon­ dties. Edison's invention (and others at Chagny, a French inventor, patented an stration of his lighting system on Decem­ about the same time) of heating a thin fila­ unusual lamp for mining which ber 21, 1879. The steamship "Columbia" ment to incandescence inside an evacu­ functioned with an incandescent platinum became the first commercial installation to ated glass bulb showed the first filament. use Edison's light bulb in May 1880, install­ fundamental advance in lighting technol­ Using a \'-shaped piece of graphite for ing 115 bulbs in a system that would oper­ ogy since direct flames. his filament, Russian physicist Alexandre ate successfully for 15 years. In 1881 a New In order for incandescent filaments to be de Lodyguine in 1872 created a light source York Gty factory used Edison's lighting put to use in light bulbs, suitable new ma­ inside a sealed glass globe filled with nitro­ system, and in the next two years more terials had to be developed. The optimum gen. Two hundred of de Lodyguine's bulbs than 150 other places installed the new material needed to be durable, inexpen­ were installed at the Admiralty dock in St. lighting, including the Savoy Theatre in sive, and capable of being heated to incan­ Petersburg, but maintenance costs and the London, which had been built for the Gil­ descence over and over again. bulbs' unreliability made the system im­ bert and Sullivan operas. Most urban ar­ In 1802 Sir Humphrey Davy discovered practical. In the same year the Russian eas, however, still used gas lighting that he could use electridty to heat strips of S.A. Kosloff used multiple rods of graph­ facilities which had been installed decades platinum or other metals to incandescence, ite, also in a nitrogen-filled chamber, but before. which would then give off light for some this, too, yielded poor results. In 1886 Austrian chemist Karl Auer von time. This proved impractical, though, be­ By 1878 Sir Joseph Wilson Swan, who 28 Welsbach studied materials to find a way to cause no satisfactory source of electridty years earlier had created carbon filaments create incandescent lighting with a gas was available to power the system, and out of paper, gave a large-scale exhibition burner. Welsbach developed a stiff spheri­ also because the incandescent metal of public lighting with bulbs containing a cal or cylindrical "mantle" of woven strands in air rapidly burned away. Seven filament of cotton thread treated with sul­ threads that glowed brilliantly when years later, in 1809, Davy used a 2,()()(kell furic add and mounted in an evacuated heated by the Bunsen gas burner in stand­ battery to pass current through two char­ glass bulb. Swan called his invention the ard use. Originally, the threads were made coal sticks separated by four inches. The "electric glow lamp:' of sea-island and Egyptian cotton soaked electricity arced across the gap, creating in a solution of 99% and 1% salts of tho­ the first arc lamp. Swan worked concurrently with rium and cerium, respectively. Later, he re­ By 1820 De la Rue had made the first at­ Thomas A. Edison, and much controversy placed them with a skeleton of thorium tempt at an actual incandescent lamp, en­ ensued over who deserved the credit for oxide threads. The mantles were treated to closing a platinum coil inside a section of inventing the modem light bulb. Edison make them strong enough to withstand glass tubing under a vacuum. In 1840 Sir had begun working on the problem in 1877 transport and commercial sale. After the William Robert Grove lit an English audito­ and reportedly conducted over 1,200 ex­ mantle had been lit on a gas burner, rium with similar electric lamps-platinum periments over the next two years. On Oc­ though, the stabilizing material burned off coils heated to incandescence and covered tober 21, 1879, Edison succeeded in and left a fragile web that could easily be by inverted glass tumblers in dishes partly lighting a bulb that used a carbonized sew­ damaged or destroyed by any shock. The filled with water. This light was feeble and ing thread filament in an evacuated glass incandescent mantle cast a garish impractical, but the cost of the current con­ globe. The carbon filament was strung greenish-white glow (characteristic of the sumed in this one instance came to several across the inside of the glass globe and gaslight era) because of the thorium/ hundred pounds per kilowatt-hour! brought out through sealed holes in the cerium salts. The Welsbach gas mantle be­ A year later the very first patent for an sides, where the ends of the thread were came the main source of city lighting until incandescent lamp went to Frederick de attached to an electric supply. His lamp it was superseded by electric lighting; it re­ Moleyns in England. His lamp differed burned steadily for two days, producing mained a standard for rural lighting up un­ from those used by De la Rue and Grove, 1.4 lumens per watt, or about the light til1930. though, in that powdered charcoal bridged equivalent of 16 candles. (A modem 60 W Eventually, continued improvements in the gap between two platinum filaments tungsten-filament bulb produces 13.9 lu­ filament materials made the electric incan­ inside an evacuated glass globe. A current mens per watt.) descent bulb a more practical option than passing through the platinum filaments With further experiments Edison found even the Welsbach gas mantle. Also crucial heated the charcoal powder to incandes­ that with filaments of carbonized Bristol to the functioning and long life of the light cence. Unfortunately, the inside of the board ("visiting card paper") he could bulb was the attainment of a good vacuum globe blackened quickly from the pow­ make a lamp bum for several hundred inside the glass, which kept the filament dered carbon, and the lamp didn't last hours. Edison also began to use filaments material from oxidizing. Even small im­ long. made of carbonized bamboo. Swan, mean- provements in filament materials allowed a 52 MRS BULLETIN/JANUARY 1990 HISTORICAL NOTE bulb either to be operated at a higher tem­ perature (thereby producing brighter, whiter light) or to have a longer lifetime. In 1905 Willis R. Whitney introduced the "metallized carbon filament;' a thread heated to a very high temperature in an electric furnace before being installed in a bulb. Intense heating changed the carbon and gave it electrical properties more like that of a metal. The metallized carbon fila­ ment increased the output from Edison's 1.4lumens per watt to about 4lumens per watt. The most important advance in fila­ ments, though, was the search for materi­ als other than carbon, culminating in the Today's newest lamps use modern engineered materials for lamp envelopes, phosphors, optical introduction of metallic filaments. Wels­ coatings, and containers for high-temperature gas discharges. Shown are compact halogen bach himself developed an electric lamp incandescent lamps, biaxial and linear fluorescent lamps with good tri-phosphor coatings, and using a filament of incandescent osmium new law-wattage metal-halide and high-pressure sodium high-intensity discharge lamps. The metal. Osmium-filament lamps were used new 60-watt Halogen-JR'M lamp has the life and performance features of a 150-watt conventional from about 1899 to 1907, but osmium itself lamp thanks to optical coatings that trap infrared emissions fram the filament inside a quartz was so rare that the filament material was filament tube. This lamp represents a 40% improvement in lamp efficiency aver conventional prohibitively expensive. incandescent lamps. (Photo courtesy of GE Lighting.) In 1901, the German W.H. Nernst pat­ ented the "Nernst glower;' an electrically heated filament made of rare-earth oxides bulb and lower light output in addition to a bulb wall, where it reacts with the halogen. (a standard commercial formula was 85% thinning of the filament itself until it broke. The tungsten halide then migrates back to­ zirconium oxide and 15% yttrium oxide). Introducing inert argon gas into the bulb ward the incandescent filament, where it In 1902, Russian chemist Werner von Bol­ reduced this migration and also allowed dissociates, depositing the tungsten back ton invented the tantalum-filament lamp, the filament to be heated to an even higher onto the filament and leaving the halogen which produced 5 lumens per watt. The temperature. free to repeat the cycle. Iodine is most com­ tantalum-filament lamp was introduced in The tungsten filament was further im­ monly used in these types of bulbs because the United States in 1906, but was replaced proved by the creation of a coiled version, the temperatures needed to maintain the a year later by a pressed-tungsten filament then the coiled-coil filament in 1937, which regenerative cycle fit well with other light yielding 8lumens per watt. further increased the effective size of the bulb designs, and iodine is the least diffi­ Tungsten was found to be the ideal mate­ filament.