On this page
Spread sunlight into a spectrum and, at the ultraviolet end, it stops abruptly a little below 300 nanometres. Something in the air is absorbing everything shorter. In 1913 Charles Fabry and H. Buisson measured ozone in the laboratory, compared it with sunlight, and showed that ozone does the job and that it must sit high in the atmosphere. This post walks through their method, works one of their own measurements, and ends with what their approach can teach you.
The later story of how man-made chemicals thin this layer is in our post on Mario Molina and the ozone hole. This one is about how the layer was first measured.
Who were Fabry and Buisson?
Charles Fabry was born in Marseille on 11 June 1867, as the Royal Society obituary of him records, and died in Paris on 11 December 1945. Britannica says he joined Marseille University in 1894 and in 1896, with Alfred Pérot, invented the Fabry–Pérot interferometer, an instrument for measuring wavelengths very precisely. He later became professor of physics at the Sorbonne (1921) and the first director of the Institut d'Optique in Paris.
His co-author, who signed his papers simply as H. Buisson, worked with him for years. Crossref lists a run of joint Fabry–Buisson papers on the solar spectrum from 1910 to 1921, including the 1913 paper this post is about.
What was the puzzle?
The paper, L'absorption de l'ultra-violet par l'ozone et la limite du spectre solaire (read here in the open HAL archive copy), opens with the problem. The solar spectrum ends a little below 3000 ångströms (300 nm), certainly because of the atmosphere. The British chemist Walter Hartley had suggested ozone, writing in 1881 on the absorption of solar rays by atmospheric ozone. But, Fabry and Buisson wrote, the idea rested only on ozone having a strong ultraviolet absorption band. To test it you needed numbers, and the existing ozone measurements disagreed with each other by a factor of up to four.
How did they measure ozone's absorption?
They shone light from a quartz mercury-vapour lamp through tubes of ozonised oxygen with quartz windows. The tubes were 0.92, 7.08, 49.7 and 194.6 cm long, so that strongly and weakly absorbed wavelengths could each be measured well. They measured how much ozone was in the gas chemically, by bubbling it through potassium iodide and titrating the iodine released.
Ultraviolet is invisible, so they used photographic plates. On the same plate they made one exposure through the ozone and several without it, dimmed by known amounts. Matching the darkening told them what fraction of light got through.
They expressed the ozone as an "equivalent thickness": how thick a layer it would make if it were pure ozone at 0 °C and 76 cm of mercury.
Worked example: their measurement at 2536 ångströms
Their rule was I = I₀ × 10^(−αx), where x is the equivalent thickness of pure ozone in cm and α is the absorption constant.
Their two exposures gave log₁₀(I/I₀) = −1.01 and −1.03; they took the mean, −1.02.
So the transmission was 10^(−1.02) = 0.0955, or 9.55%, as the paper states.
The tube held x = 0.0083 cm of ozone, so α = 1.02 ÷ 0.0083 ≈ 123 per cm.
Halving thickness: x = log₁₀ 2 ÷ 123 = 0.301 ÷ 123 ≈ 0.0024 cm, about 25 microns. The paper says the same: near 2550 Å, 25 microns of pure ozone would halve the light.
Now put today's ozone overhead into their formula. NASA Ozone Watch gives an average of about 300 Dobson units, which is a layer 3 mm (0.3 cm) thick. Then αx = 123 × 0.3 ≈ 37, so only about 1 part in 10³⁷ of 254 nm sunlight gets through.
Slips to avoid: using ln instead of log₁₀ with their α; forgetting that x is in cm; and confusing equivalent thickness with altitude. The 3 mm is how thick the ozone would be if you squeezed it to ground pressure, not where it sits.
How much ozone did they find, and where?
From provisional sunlight measurements made in the poor summer of 1912, they estimated the atmosphere holds ozone equivalent to a 5 mm layer of pure ozone. Chemical tests of air near the ground gave about 75 times too little ozone to explain this. Their conclusion: the ozone must exist mainly in the very high atmosphere, where it is made by the Sun's most extreme ultraviolet.
They also checked their result against older solar observations by Alfred Cornu, whose formula for how the spectrum's end shifts with the Sun's height had the same form as theirs.
What can a student take from their way of working?
Turn a hunch into a number. Hartley's idea was reasonable, but it became science only when someone measured absorption and compared it with the sky.
Design for the range you need. Four tube lengths let them measure both strong and weak absorption accurately.
Check against independent data. Agreement with Cornu's older observations made their case stronger.
The ozone layer appears in Class 10 science, covered in our guide to Our Environment for Class 10. If absorption, wavelength and the electromagnetic spectrum still feel separate in your head, the Fermi Concept Library covers every concept in physics, chemistry, biology, mathematics and computing.
Next step: use α = 123 per cm to find what thickness of ozone lets through exactly 1% of 254 nm light.
