Electronics Puzzler Solution: Measuring Ozone

How did Ozzie Measure Ozone in the Air Outside His Office?

By Forrest M. Mims III

As Ozzie watched the traffic light cycle though its colors of green, yellow and red, his boss's ozone challenge kept cycling through his mind. Suddenly he shouted, "Yes!" and immediately searched online for more information about the absorption of visible light by ozone.

Puzzler 20

Ozzie soon learned that James Chappuis had long ago discovered the ozone absorption feature named for him. The Chappuis ozone absorption band extends from the blue to the red region of the visual spectrum with twin peaks in the yellow at around 575 nm and the orange at around 603 nm, as shown in this chart:

Puzzler 20

This chart is a simplified version of measurements of the Chappuis band by many separate teams. References to all Chappuis and Wulf Bands listed here

Ozzie learned that the absorption by ozone in the visible spectrum is much less than in the UV, but with care, it is definitely measurable. He compared this new information with the colors emitted by the traffic light and found that the yellow light was closest to the peak absorption of the Chappuis band. Ozzie then designed a simple photometer that used a silicon solar cell as a photodiode.

Building the Prototype

Ozzie found that the entire traffic light and little else could be seen through a one-meter long aluminum tube. He dropped a string through the tube, tied the end to a cotton ball and darkened the inside of the tube by soaking the cotton ball in flat black enamel and pulling it through the tube.

After the paint dried, Ozzie inserted one end of the tube into a three-inch long second tube in which a solar cell was cemented at its closed end. He cemented the two tubes together, set the light detection assembly aside and went to work building a high-gain amplifier from a TLC271 operational amplifier with a seriously large feedback resistor (R1). He started with 10 megohms, knowing he could reduce or increase the value as needed. He installed the amplifier board with its 9-volt battery in a metal box for shielding and connected the solar cell leads to the amplifier's input via a length of shielded cable.

Puzzler 20

Ozzie next mounted the solar cell tube on a sturdy tripod and aligned it so that the solar cell was looking at the entire traffic light. He then connected the output from the amplifier to his trusty 16-bit A/D converter card on the old desktop computer he kept for experimenting.

As the traffic light cycled through its three colors, the output from the A/D converter produced a running sequence of three fixed amplitude bars on the screen of the computer. The peak response of silicon solar cells is in the near infrared, which explains why the red light elicited the highest bar on the running graph. The computer generated the visual graph and saved the date, time and the numerical data from each light in a simple five-column spreadsheet. Ozzie realized that changes in sunlight, background illumination, dust and slight movements of the traffic light would alter the data, but he reasoned the ratio of any two of the colors would cancel out such effects.

Over a period of several weeks he noticed a distinct diurnal cycle in the ratio of the yellow/red and yellow/green lights intensities. He plotted these ratios against the ambient ozone measured by the nearby monitor and derived an empirical equation that converted the yellow/red and green/red ratios into the ambient ozone. The correlation coefficient for the yellow/red lights was higher than the yellow/green pair, possibly because the green wavelengths were absorbed more by dust and exhaust than the red wavelengths.

The Hypothetical Result

In the end, Ozzie's ambient ozone measurements were not perfect, but they were reasonably close to those provided by the government instrument. He ignored minor day-to-day differences, which were probably caused by blowing dust and unusual traffic. One day he noticed what appeared to be an increasing difference between his data and that from the government monitor. Ozzie checked the alignment of his collimator and it was still on target. He looked for dust on his solar cell and there was none. After several weeks went by, the expensive government instrument was indicating 15 percent more ozone than Ozzie's ultra-low budget traffic light monitor made from Jameco parts and stuff lying around his shop.

Finally it was time to notify the regulators, who sent a crew of three technicians to check their expensive monitor. After an hour of testing, their frowning leader mumbled something about a trivial drift in the instrument as they were preparing to leave. The next day they arrived in a truck, picked up their instrument and moved it to a new site three miles away from Ozzie's office.

Going Further

While this Puzzler is mostly hypothetical, it has potential, for scientists have measured the ozone layer using Chappuis wavelengths and so have I. Maybe you can be first to determine if Ozzie's method really works or if his inspiration was modulated by too much wishful thinking.

Back to the Ozone Puzzler >>