On this page
Solar cells existed before 1954, but they were too weak to power anything useful. Then a physicist, a chemist and an engineer at Bell Labs combined what each knew and made a silicon cell that could. This post covers who they were, how the cell came about, a worked example built on the numbers in their own patent, and what a student can take from how they worked together.
Why did Bell Labs want a solar battery?
The American Physical Society's history of the first practical silicon solar cell says that in 1953 Daryl Chapin was looking for a power source for telephone equipment in remote, humid places, where dry cell batteries degraded too quickly. He tried selenium solar cells and found them too inefficient.
Their patent, Solar Energy Converting Apparatus, filed on 5 March 1954 and granted on 5 February 1957, gives the target. Earlier devices had not reached overall efficiencies "significantly greater than one percent". A telephone repeater station needs about 0.2 watt, and a converter giving about one watt in usual sunlight should keep a storage battery charged for such a station.
Who were Chapin, Fuller and Pearson?
The National Inventors Hall of Fame, which inducted all three in 2008, gives their backgrounds.
Daryl Chapin (21 July 1906 to 19 January 1995) was born in Ellensburg, Washington, studied at Willamette University and the University of Washington, and joined Bell Labs in 1930. He worked on magnetic materials and, during the Second World War, on underwater sound.
Calvin Fuller (25 May 1902 to 28 October 1994) was a chemist with a PhD in physical chemistry from the University of Chicago. From 1948 he worked on semiconductors and developed a method of diffusing impurities into the surface of silicon.
Gerald Pearson (31 March 1905 to 25 October 1987) was a physicist from Salem, Oregon, with a master's degree from Stanford. He joined Bell Labs in 1927 and later set up a compound semiconductor research programme at Stanford.
How did the cell come about?
APS tells the story. Fuller gave Pearson a piece of silicon containing gallium. Pearson dipped it in lithium, making a p-n junction, then shone a light on it while it was connected to an ammeter. The reading jumped. Pearson told Chapin to stop working on selenium, and Chapin switched to silicon.
Over several months the three solved two problems. Good electrical contacts were hard to make, and lithium slowly moved through the silicon at room temperature, carrying the junction away from the surface where light arrives. They switched to arsenic and boron, which kept the junction near the surface and allowed good contacts. Bell Labs announced the "solar battery" on 25 April 1954 in Murray Hill, New Jersey, using it to run a toy Ferris wheel and a radio transmitter. The cells were about 6 percent efficient.
The team's paper, A New Silicon p-n Junction Photocell for Converting Solar Radiation into Electrical Power, appeared in the Journal of Applied Physics in May 1954. The junction idea itself goes back to Russell Ohl's 1940 discovery, told in our post on Russell Ohl and the p-n junction.
Worked example: sizing a solar battery from the patent
The patent describes a cell that delivered power "at the rate of more than 55 watts per square meter of cell surface area", with "an open circuit voltage of approximately 0.52 volt per cell".
Area for one watt. A = P ÷ (power per area) = 1 W ÷ 55 W/m² = 0.018 m². Since 1 m² = 10,000 cm², that is about 182 cm², a square roughly 13.5 cm on each side.
Area for the repeater alone. 0.2 W ÷ 55 W/m² = 0.0036 m², about 36 cm². The patent asks for five times this power; the sun does not shine at night, so the battery must be charged by day.
Cells in series. For a sample 6 V battery, the number of cells is 6 ÷ 0.52 = 11.5, so at least 12 cells, giving 12 × 0.52 = 6.24 V on open circuit.
Which light can it use? The NCERT Class 12 semiconductor chapter gives silicon's band gap as about 1.1 eV. With hc ≈ 1240 eV nm, the longest useful wavelength is 1240 ÷ 1.1 ≈ 1130 nm. Light of longer wavelength does not have enough energy per photon to free an electron.
Common mistakes:
Mixing cm² and m². Divide by 10,000 to go from cm² to m², not by 100.
Treating open-circuit voltage as working voltage. Under load a cell gives less, so real designs add cells.
Calling efficiency a power. Efficiency is power out divided by light power in, a ratio with no unit.
The NCERT chapter covers the p-n junction itself; solar cells are no longer in the rationalised book, but this kind of question uses exactly the junction ideas it teaches. Our guide to p-n junction diode characteristics is a good next read. If you want to check each step of a sizing problem like this, the Fermi Problem Library has a step-by-step solver that explains the reasoning rather than printing an answer.
What happened next?
APS records that the first cells were expensive, but within a few years solar cells were powering satellites. In 1959, the Hall of Fame notes, Chapin simplified the process enough that it became a Bell System science experiment for high school students.
What can a student take from how they worked?
Mix disciplines on purpose. A chemist's diffusion method and a physicist's junction met an engineer's practical problem.
Act on a surprise. One jumping ammeter reading changed Chapin's whole plan.
Fix the failure you can see. Drifting lithium was a clear problem; changing the impurities solved it.
Try it now: using 55 W/m², work out the cell area needed to run a 5 W lamp, and how many 0.52 V cells give at least 12 V.
