Lowrey’s reassurance came amidst reports that Edison had been having difficulties with his experiments using platinum as a filament. Like previous experimenters, Edison had focused his initial lamp experiments on two materials—carbon and platinum. Charred paper or cardboard was a commonly used and cheap carbon material for lamps and Edison used these in his earliest experimental lamps. However, the hand-operated vacuum pump in his laboratory did not create a sufficient vacuum to prevent the carbon from burning up in the atmosphere that remained in the bulb. Edison instead focused his research on platinum filament lamps that used a regulator to prevent the metal from reaching its melting point. Although platinum was rare and expensive, it had the highest melting point of any metal other than tungsten, which is used in modern incandescent bulbs. However, at the time the technology did not exist to draw tungsten into a thin filament.
While Edison’s early platinum-lamp experiments showed promise, he found that his filaments were melting at too low a temperature. Thus, in January 1879 he conducted basic research on the heating of platinum to find out why this was happening. His experiments showed that air was absorbed into the pores of the metal as it was heating, causing it to weaken and melt at lower temperatures. To overcome this problem Edison decided to place his metal filament in a vacuum bulb. Later in the year, Edison presented a paper on this research at the annual meeting of the American Association for the Advancement of Science.
Edison and his assistants had used a mechanical hand pump in some of the early lamp experiments and turned to it again for the new experiments. However, such pumps provided an inadequate vacuum and the staff began to investigate the state of the art of vacuum technology by searching through scientific and technical journals. The best pumps of the time were Sprengel and Geissler mercury pumps. After an unsuccessful effort to acquire a Sprengel pump from nearby universities, Edison had a Geissler pump made by the New York glassblowing firm of Reinmann & Baetz. William Baetz also came to Menlo Park periodically to assist in further development of vacuum pumps for the laboratory. By mid-August he had assembled several different arrangements of Geissler and Sprengel pumps, including combinations of the two based on an article Edison had read; it included a McLeod gauge, which was added to the Menlo Park pump design later in the year. The advantage of combining the two designs arose from the fact that the Geissler pump produced a vacuum more quickly while the Sprengel pump provided a higher vacuum. When Baetz turned down an offer to join the laboratory staff, Edison hired Ludwig Boehm, a young glassblower who had worked with Geissler in Germany before immigrating to the United States. An intense series of experiments in early October led to the development of better and better vacuum pumps.
Using a vacuum improved the performance of Edison’s lamps and he no longer needed to use a regulator to prevent the filaments from melting. However, platinum continued to present problems that were not easily solved. Not only was it a very expensive metal, but platinum also had a relatively low resistance to the electric current. Edison could increase the resistance of his filaments by coiling them into spirals, thus increasing the amount of metal offering resistance to the current. But he had not yet produced low-cost lamps with sufficiently high resistance to make his electrical disribution system economical for widespread use.