Photovoltaics from Milliwatts to Gigawatts. Tim Bruton

of an electric current when gold or platinum electrodes were immersed into an electrolyte (acidic or alkaline) and exposed to uneven solar radiation. Some ten years later, Alfred Smee in London observed a current in an electrochemical cell on exposure to intense light, which he called a ‘photo‐voltaic’ circuit – linking the Greek word for light phos and the name ‘Volta’, the original inventor of the galvanic cell [11].

The next step was the observation of photoconductivity in a solid material. A British engineer, Willoughby Smith, in search of a high‐resistance metal for use in testing the trans‐Atlantic telegraph cable, was recommended selenium. He purchased some selenium rods of between 5 and 10 cm in length and 1 and 1.5 mm in diameter [12]. These were hermetically sealed in glass cylinders, with leads to the outside. They worked well at night, but in bright daylight they became too conducting. Smith concluded that there was no heating effect and that the change in resistance was purely due to the action of light [13]. This stimulated further research into the properties of selenium. The British scientists William Grylls Adams and Richard Evans Day observed current flowing in their selenium sample when no external voltage was applied and were able to show that ‘a current could be started in the selenium by the action of light alone’ [14]. They had demonstrated for the first time that light caused the flow of electricity in a solid material. They used the term ‘photoelectric’ to describe their device, and Adams believed it could be used as a means of measuring light intensity [15].

The narrative now switches to America, where Charles Fritts made the first working solar module by covering a copper plate with a layer of selenium and applying a semitransparent gold layer as the top electrode [16]. An example is shown in Figure 1.2. Fritts described the module as producing a ‘current that is constant and of considerable force … not only by exposure to sunlight but also to dim diffused light and even to lamplight.’ He supplied samples to the German electricity pioneer Werner von Siemens, who greeted them enthusiastically, announcing Fritts’ module to be ‘scientifically of the most far‐reaching importance’. However, its low efficiency – below 1% – made it of little commercial importance. Indeed, there was considerable scepticism at the time, with solar cells viewed as some kind of perpetual‐motion machine. The principles of their operation were not understood. One of the leading physicists of the day, James Clerk Maxwell, while welcoming photoelectrcity as ‘a very valuable contribution to science’, wondered ‘is the radiation the immediate cause or does it act by producing some change in the chemical state’ [16].

Figure 1.2 Charles Fritts’ first photovoltaic array, produced in New York City in 1884 [16]

      (Courtesy New World Library)

      The underlying science of photovoltaics was given a big boost by the parallel discoveries and developments in photoemission. Hertz observed in 1887 that ultraviolet light caused a significant increase in the sparks in an air gap between electrodes and that it was a function of the wavelength of the light rather than its intensity [17]. While a number of physicists worked on the effect, it was Albert Einstein in 1905 who explained it in terms of different wavelengths behaving as particles of energy, which he called ‘quanta’ but which were later renamed ‘photons’. These quanta had different energies depending on their wavelength. Einstein was awarded the Nobel Prize in 1921 for this work [15]. While these discoveries and other advances in quantum mechanics at the start for the twentieth century did not directly explain photovoltaic effects, they did provide a scientific basis for understanding the interaction of light and materials.

      Although research continued on developing solar cells, little progress was made. However, photovoltaics still had its advocates in the 1930s. Ludwig Lange, a German physicist, predicted in 1931 that ‘in the distant future huge plants will employ thousands of these plates to transform sunlight into electric power … that can compete with hydroelectricity and steam driven generators in running factories and lighting homes’ [15]. A more pragmatic view was taken by E.D. Wilson at Westinghouse Electric, who stated that the efficiency of the photovoltaic cell would need to be increased by a factor of 50 in order for them to be of practical use, and this was unlikely to happen [15]. Actually, as will be shown in later chapters, a factor of 20 was achieved, and this was sufficient to create the current global markets.

      While progress in other areas of technology was immense in the nineteenth and early twentieth centuries, little real advancement in photovoltaics had been made since Becquerel’s discovery a hundred years previously. Entering into the second half of the twentieth century, everything would change.

      1.2.2 The Breakthrough to Commercial Photovoltaic Cells

It is well known that the birth of the commercially successful photovoltaic cell dates back to April 1954, when Pearson, Chapin, and Fuller demonstrated the first 6% efficient cell using a p/n junction in silicon. It is no surprise that this discovery occurred at Bell Telephone Laboratories, which was one of the world’s premier research laboratories until its forced break‐up in 1984. As the research arm of the American Telephone and Telegraph Company, it had a long history of successful innovation, with nine Nobel Prizes awarded over time for work done there. Perhaps its most notable success was the demonstration in 1948 of the point‐contact germanium transistor. This illustrates the strength and depth of both the theoretical understanding and expertise in semiconductor processing at Bell labs [18].

Figure 1.3 Ohl’s patented solar cell structure [20]

      Source: R.S. Ohl: US Patent Application filed 27th May 1941

Russel Ohl, a Bell Labs scientist interested in exploring the crystallisation of silicon, is recognised as the discoverer of the p/n junction in this material, in 1941 [19]. In directionally solidifying 99.85% pure silicon, Ohl noted a change in the structure of the solidified ingot, with the upper portion becoming columnar and the lower portion showing no structure; a striated region appeared between the two, forming a barrier to conduction. The upper zone was p type while the lower zone was n type [20]. This can be easily understood as a result of the segregation of dopants during the crystallisation process. While measuring the resistance of rods containing the barrier, Ohl noted a sensitivity to light, which he termed a ‘photo electromotive force’. He proceeded to patent this as a solar cell, although its efficiency was similar to that of the selenium cells, at about 1% [20]. Figure 1.3 shows Ohl’s silicon structure, the n type region being fine‐grained crystallites and the cell contacts plated rhodium. The low efficiency is not surprising given the relatively impure starting material, its multicrystalline nature, and the fact that the n type region was 0.5 mm thick. The relatively low efficiency meant little further work was done until a new approach at Bell Labs.

Success came in the 1950s. The first transistor had been demonstrated at Bell in 1948 using germanium, and had entered commercial production in 1951 [21]. However, germanium had some disadvantages in its fragility and stability, and silicon offered a better option – although a working silicon transistor was not demonstrated until 1954. Two scientists working on this were Calvin S. Fuller and Gerald L. Pearson. Fuller was an expert in doping silicon, while Pearson was an experimentalist. There were three iterations before a good working solar cell was demonstrated [22]. Initially, while not looking for a solar cell, Fuller produced a p type gallium‐doped silicon sample, which Pearson dipped into a lithium bath to form a shallow n type region. When Pearson exposed the sample to light, he found to his surprise that a current was generated. At the same time, in a different department, another scientist, Daryl M. Chapin, was looking for a power source for telecommunications repeaters