Tuesday 22 February 2011

Evaporography - home brew thermal imaging

There are a few indirect methods for imaging far-infrared radiation, relying on a visible effect caused by the infrared which can then be seen or photographed. One of these is known as evaporography.

Dr Marianus Czerny from the University of Frankfurt, writing in Zeitschrift fur Physik in 1929 (this is a big file of translated papers ... go to page 1), proposed coating a celluloid membrane with a thin layer of white napthalene and putting this in an enclosure saturated with napthalene vapour. Any areas of the coating that heated up due to infrared radiation falling on it would evaporate (sublime) and it would be possible to photograph the membrane to 'fix' a record of the patterns produced. This became known as an evaporograph.

This technique was largely unknown to the public until May 1956, when a newly-declassified device called Eva was demonstrated by Baird Associates of Cambridge Mass. The images were unmistakably thermal images and, in this case, were formed on a thin sheet of plastic coated with silicon oil. This was held in a small vacuum chamber with a salt window through which the infrared passed. The thickness of the oil film varied with the amount of radiation falling on it and interference effects resulted in this being shown as different colours with a resolution of one degree (presumably Fahrenheit). Baird's evaporographs (or sometimes evapographs) appeared in several magazines, including Time, Popular Science, Popular Mechanics and Life.

Although Baird's applications, at least as promoted, were of thermal imaging, the evaporograph principle can be applied to shorter wavelengths. Pol F Swings, who wrote a paper in 1945 on astronomical applications of evaporography, compared this to near infrared photography using Agfa 1050 plates (without ammonia hypersensitisation) and concluded that at wavelengths shorter than 1200 nm the Agfa was faster where as at wavelengths longer than 1300 the evaporograph was faster.

The process is relatively easy to reproduce, probably well within the range of a school laboratory, and a recipe currently available on the internet (scan down the page), but dating from 1972, claims to be able to record the heat from a human body after dark in ten seconds.