Selenium and Light
Retired from Dept. of Physics, University of Leicester, U.K.
The element selenium is sensitive to light, varying both its electrical resistance and generating a small photovoltaic current proportional to the intensity of illumination falling upon it. The latter property was employed in the ‘Megatron’ cell for measuring the intensity of light in a range of situations. The cell is described here, together with other applications of selenium to monitoring light.
Sulphur, selenium and tellurium fall into Group VI of the periodic table, and share a number of chemical properties (ref. 1). Sulphur has been known since ancient times since it occurs in the free state around the vents of active volcanoes, such as those in Italy. Selenium, however, was not discovered until 1817, when Berzelius separated it from a deposit formed in a lead chamber used for making sulphuric acid from roasted iron pyrites. Tellurium was isolated from the same source in 1832. Nowadays, both elements are obtained from the ‘anode slimes’ that are by-products of electrolytic copper refining.
Like sulphur, selenium exists in several allotropic modifications. Black or red amorphous forms are produced on rapidly cooling molten selenium, but above 80o change gradually into a grey metallic state. Conversion is complete by 180o, the element taking up a complex polymeric ring structure. It then exhibits an electrical conductivity intermediate between metals and non-metals, and used to be known as a semiconductor.
The photoelectric properties of this form of selenium were glimpsed by Willoughby Smith (ref. 2) in 1873, when he observed that its resistivity was affected by the intensity of illumination falling on the specimen. This led to Siemens in the mid-1870s constructing the first cell deliberately designed for sensing light: he simply wound two separate helices of thin platinum wire side-by-side around a mica rectangle, filled the spaces between the windings with molten selenium, and then annealed the assembly in an oven at 180o. The selenium was protected from moisture and mechanical damage by a thin glass plate or envelope. A large surface-to-volume ratio was necessary because photoconductivity is a surface effect, while the bulk resistivity of selenium is quite high.
This ‘selenium cell’ was produced commercially by a number of instrument makers (Figure 1). Versions were used in Alexander Graham Bell’s ‘photophone’ of 1879 (ref. 3) and in early experimental television transmissions by Baird around 1927. It was still being recommended in 1938 (ref. 4).
However, a frustrating problem was to obtain reproducible results from a comparatively thick layer of selenium in an ill-defined allotropic state (ref. 5).
The breakthrough in the application of selenium to measuring illumination was the discovery (ref. 6) that a thin film of the element on a steel plate (with which it reacted to form a selenide) generated a small current across its thickness proportional to the intensity of light falling upon it. This direct conversion of light energy into electricity became known as the photovoltaic effect. To a lesser extent it is also displayed by sulphur, tellurium, silicon, and a number of complex semiconductors, but for a long time selenium gave the best match to the response curve of the human eye.
Selenium cells utilising the photovoltaic phenomenon were produced commercially by Megatron Ltd (later Salford Electrical Instruments Ltd, Eccles, Manchester; finally closing down in February 2010). These cells employed the construction shown diagrammatically in Figure 2.
The assembly was commonly potted within a protective black plastic casing. An example is shown in Figure 3.
The output of the Megatron selenium cell is close to linear when the illumination is low and the load resistance is small. Light meters therefore normally employed a fairly large cell directly connected to a high quality microammeter. Type B cells were general purpose products, with high sensitivity and stability. They showed a wavelength response with a maximum that was a fair approximation to that of the human eye, but the type M cells incorporated a filter that gave a close match to the CIE standard observer (Figure 4). Type B produced about 0.07 A/lux/cm2; type M about 0.03 A/lux/cm2.
A Megatron selenium cell was combined with a battery-powered transistorised amplifier and range-changing switch in the Minilux light meter (Figure 5). This was calibrated from 10 – 2500 lux, and was intended to cover the range of illumination that might be encountered in ordinary domestic and industrial situations. As a contemporary example of such an application, the sensor shown in Figure 3 was placed in a light-tight box with a modern Chinese-made 1360 lumen ‘energy saving’ lamp (consuming 35 W) containing a helical phosphor tube within a conventional frosted-glass envelope. Figure 6 shows how the lamp took 3 minutes to reach the rated output. No mention of this on the package, of course!
Another example of the Megatron selenium cell being used to determine the percentage of polarized light from a blue sky is given in reference no.7. A digital microammeter was used to measure its output.
A major application of the selenium photovoltaic cell was to compact cameras and their associated light meters – either built-in or separate – in the 1950s. An example is illustrated in Figure 7. The construction of the cell followed Figure 2, and it was connected to a small but robust mechanical microammeter. The face of the cell was covered by a characteristic plastic window embodying a honeycomb of convex lenses. This produced an approximate match with the field of view of the camera.
The Xerox process
Electrophotography was a unique photocopying technique invented by Chester Carlson (ref. 8) in 1938, for which he was awarded a patent in 1942. Unlike all other photographic processes of the period it was completely dry, being based on electrostatics rather than chemistry. It was therefore given the proprietary name of ‘xerography‘. The key component was a cylindrical aluminium drum coated with a thin film of vacuum-deposited selenium. It was sprayed (in the dark) with a static charge of chosen polarity by means of a corona discharge, and then exposed to light imaged from a black-and-white master. (Initially the tonal range was very limited, so was best suited to copying printed type.) The charge leaked away where the image was bright, so that when the roller was dusted with a charged black powder it adhered only to the dark areas. Rolling on to a clean sheet of paper, and neutralizing the charge, transferred the powder image. It was finally fused in place.
It took 18 years for the process to be successfully refined (it was rejected by a dozen companies) but the Xerox process (soon to show tones and colour) eventually took control of the commercial and domestic photocopying industry and became vastly lucrative. However, organic semiconducting elastomers have now replaced selenium-covered rollers, and lasers may be employed to charge and discharge them.
- J.R. Partington, General and Inorganic Chemistry, (London: Macmillan, 1951).
- W. Smith, ‘The action of light on selenium’, Soc.Telegraph Eng.J., 2 (1873), 31-33.
- T.B. Greenslade, ‘The photophone’, Physics Teacher, 17 (1979), p.382.
- Richard M. Sutton, Demonstration Experiments in Physics, (New York: McGraw-Hill, 1938) p. 318.
- George P. Barnard, The Selenium Cell: Its Properties and Applications (London: Constable, 1930).
- A. Christy, Selenium Cell, US Patent 2,066,611, Jan.5, 1937. Assigned to G-M Laboratories Inc., so Christy was probably an employee.
- Allan Mills, ‘An electronic polarization sundial and photometer’, British Sundial Soc. Bull. , 21 (2009), 14-16.
- David Owen, Copies in Seconds, (New York, NY, Simon & Schuster, 2004).