Determination of atmospheric ozone content using an integrating sphere: improving stability of light sources

Research Group: Derrick Green, Alyssa LaFleur, and Travis Widmer

Launch: Whitworth Spring 2016

''The goal of this experiment was to continue the ongoing project of constructing a low-cost integrating sphere based ozone detector, building off the work of previous groups. Low-cost methods of ozone detection are essential for getting detailed readings easily for a specific area without specialized, highly expensive equipment. The local ozone concentration sensor produced relied on a phototransistor housed in an integrating sphere, with a red LED and UV LED providing light for ozone gas to absorb, and a phototransistor and memory card to record variations in voltage as more or less light was absorbed by ozone to determine the concentration present. This version of the sensor paid special attention to the stability of the light sources, by housing them in a separate cap for better diffusion of entering light, and by implementing two fans to pull air through the sensor as it rose with the balloon. Unfortunately, the phototransistor came loose shortly after launching, and no usable data was gathered. Future recommendations for continuing this project are to focus on securing all sensor components, especially the batteries and phototransistor. ''

Importance of Ozone and History
Ozone, specifically ozone located above the earth in the stratosphere, plays a vital role in the defense of our planet from ultraviolet light(Gleason). More specifically, it shields mostly from UV-B radiation, in the range of 280-320 nm("Stratospheric Ozone Depletion"). These wavelengths are known to cause damage to DNA, as when high energy photons strike the molecule, it excites electrons and allows new bonds to form in DNA. This distorts the original structure of the molecule, creating a 'photoproduct' and causing a mutation. Photoproducts can interfere in typical cellular processes by making the affected portion of the DNA chain unreadable by the cell in that location(Allen). The two commonly formed photoproducts believed are cyclobutane pyrimidine dimer, and 6-4 pyrimidine –pyrimidone, where the bases of DNA bond together and distort the original structure (Allen).

DNA repair mechanisms exist in cells to deal with such damage, but overexposure to UV-B radiation has been shown to have adverse effects. For example, it is believed that by overtaxing cellular repair mechanisms from chronic exposure to sunlight, the human skin cells can become unable to find and repair all damage caused by the impacts of the excited ultraviolet photons. This can eventually lead to skin cancer(Allen). Estimations that a prolonged 1% decrease in stratospheric ozone will eventually lead to a 2-3% increase in non-melanoma skin cancer in humans have previously been put forth by the United Nations Environment Program (Gleason). It has also been shown that eyes are particularly at risk for damage from UV radiation, with cataracts, snow-blindness, and other vision impairments affecting both humans and other species after chronic overexposure. Overexposure to UV rays can also have detrimental effects in crop growth, phytoplankton activities, and the development of sea urchin embryos(Gleason).

The effects of UV-B radiation on phytoplankton is especially concerning, due to their important role in aquatic ecosystems. Of special importance is that in in the 1970’s, British scientists stationed at an Antarctic Survey station who monitor ozone discovered that it was thinning over the South Pole, dropping below the standard concentration of 220 Dobson’s Units (a unit of measurement to do with the concentration of ozone as if it was condensed into a thin layer of liquid)("The Montreal Protocol on..."). The decreasing concentration trend continued over the years, monitored by NASA and the Royal Netherlands Meteorological Institute, reaching a record low of 73 Dobson’s in 1994 ("Part 1: The History..."). It was eventually determined that these abnormal lows were caused by chlorofluorocarbons, a class of chemicals used in aerosols and refrigerators since 1930’s being released into the atmosphere, where they catalyzed the destruction of ozone in the upper atmosphere with chlorine gas (Gleason). In 1989 the Montreal Protocol, an international treaty to reduce the amount of ozone depleting substances produced and used, was enacted as it was realized globally how catastrophic the effects were of such substances (Lindsey).

The discovery of such a phenomenon years earlier could have averted some of the buildup of CFC’s, but as is, it is estimated that the ozone concentrations will not return to normal into the middle of this century (Gleason). This shows that monitoring of ozone is an important task, so as to avert any more unnoticed disasters, like the effects of CFC’s. Today, ozone concentration is still closely monitored by many agencies. There are two methods for measuring the concentration of ozone in the stratosphere, local and remote measurements ("How is Ozone..."). Remote measurements involve the detection of ozone from large distances by measuring the amount of UV rays it interacts with. Ground based stations detect how much radiation is actually reaching them, and satellite’s detect how much light is absorbed from the upper atmosphere by measuring light reflected by the earth’s surface to find ozone concentrations. Lasers and visible light are also commonly used for the same purpose by planes and ground bases, to measure concentration of ozone via its absorption of those wavelengths. Local measurements involve techniques which require air to be drawn directly into a measurement device, where ozone concentration is found by its absorption of light, or the electricity generated from an ozone chemical reaction. These local devices are usually used on planes, or in ‘ozonesondes’, modules were can be launched on weather balloons ("How is ozone...").

The purpose of this experiment was to attempt to improve on the design of the ongoing low cost integration sphere sensor ozonesonde which has been carried out in 2008, 2010, 2012, and 2014 through this project. This iteration of the sensor involved improvements to the local measurement device such as the use of two micro fans to give continuous air circulation from the outside of the pod through the sensor, and increased attention to the stability of electronic components relative to each other and the main sphere.

Absorption and Ozone
As altitude increases, the amount of ozone in the atmosphere increases until reaching the ozone layer. This ozone protects us from UV rays, as mentioned above. Ozone “shields” us from ultraviolet rays not by defecting them, but by absorbing them. This 'absorbing' happens during the chemical reaction in which ozone is formed, where the energy from UV light causes oxygen gas to break into two free oxygen atoms, which are very reactive. In this case, the oxygen atoms generated reacts with more oxygen gas (O2), and creates the tri-atomic (three atom) ozone molecule. This ozone absorbs light in the UV range, and reverts back to diatomic oxygen and another free oxygen molecule. This free oxygen then reacts with diatomic ozone again, to produce ozone in a exothermic reaction. An exothermic chemical reaction is one in which the end products of the reaction have a lower energy than the reactants, meaning that energy, in the form of heat in this case, is given off. Eventually this ozone molecule will react with a free oxygen, and revert back to two molecules of diatomic oxygen. Either way, the net reaction of all these steps involves the conversion of diatomic oxygen gas into ozone through the energy provided by UV rays, with the eventual release of said energy as heat, and the return of the diatomic oxygen molecules ("Chapter 5: Stratospheric..."). This process is known as the Chapman cycle.

A good way to test to see how much of a certain gas is contained in an area, otherwise known as concentration, is to see how much light that gas absorbs. Light is absorbed by a chemical compound when it is of the right energy level to 'excite' electrons to a state where they have higher energy, with the chemical 'taking in' energy from the photon to do so. Energy levels electrons can be in for an atom are quantized, meaning they can only exist at fixed levels. So, in order for absorption to take place, the energy imparted from the photon needs to be the size of the difference between the energy levels for the electrons in order to raise them to the next energy level. Certain wavelengths are better absorbed by chemical compounds than by other compounds, due to the different amounts of energy that electrons in compounds can take in to be excited to a higher energy state. So, by looking at how much light is absorbed by a gas in a wavelength already known to be readily absorbed by one chemical compound, it can be said that this absorbance is mainly due to that compound. Then, the concentration of said compound can be deduced from the absorbance by using the Beer-Lambert law, in which absorbance is directly proportional to concentration of a compound(Clark).

$$ A=cl\epsilon $$

The Beer Lambert law, where $$\epsilon$$ is a constant which varies from compound to compound, $$l$$ is path length, $$c$$ is concentration.

UV light is absorbed wonderfully by ozone, as mentioned previously in the Chapman Mechanism. However, several other wavelengths in the visible spectrum also are absorbed by ozone (. To measure the amount of ozone using the Beer Lambert Law, all that needs to be known is the amount of light being absorbed by the ozone, and the path length that said light travels. The longer the light has traveled; the more absorption takes place. So, a long path from the light source and the method of detection will result in better absorbance of light by ozone, and more accurate results for the determined ozone concentration. As the entire sensor for this project must be contained within a weather balloon pod, having the necessary path length for absorbance to take place is difficult.  To alleviate this problem, an integrating sphere was used.

Integrating spheres
Integrating spheres are devices designed to measure optical radiation though multiple reflections of light off of the diffuse surface inside of the integrating sphere. A diffuse surface is one which reflects light that hits it in such a way that it acts like another light source, scattering the rays into all different directions. This works as when light is reflected inside of the sphere, the fraction of electromagnetic radiation exchanged between two spots inside is the same at any point in the sphere, and is equal to the area of the point in question divided by the entire inner surface area ("Technical Guide.."). In other words, the light received at any point in the sphere is independent of both path length in the sphere and viewing angle, meaning every point receives the same amount of light to it. So, light can be bounced around inside of the sphere, to be measured by a detector, to find the radiant energy given off by the light as it is the same at all points. Integrating spheres work best when designed so that the ‘ports’, the holes were the light source and viewing/detection take place, are no more than 5% of the total surface area of the sphere, and the diffuse material on the inside has the highest reflectance possible. In the past, Barium Sulfate was typically used as the inner coating, but now spheres usually use some variant of an optical grade plastic ("Real Integrating Spheres"). Baffles, small shields coated in the same diffuse material used as the inner coating, are also needed to block the detection port’s field of view so that it does not include the initial spot where the light of the light source strikes, or else the light from that location and not the reflected light will be measured ("Technical Guide...").

For the purposes of this experiment, the integrating sphere’s ability to reflect light inside of it multiple times with a uniform aspect was of interest, instead of its ability to measure the optical radiation being put into it. It was used as a means of increasing the available path length for the absorbance of ozone, as the gas in need of measurement could be brought inside of the sphere to take advantage of the ‘increased’ path length. The light source for the sphere was chosen so that there were two wavelengths being used, one known to be absorbed and one not absorbed by ozone. That way, a phototransistor (an electrical component which allows different amounts of current to flow through it depending on how much light is striking it) could be used to measure the amount of light actually reaching it which was not being absorbed by ozone, by looking at the changes in voltage travelling through a detecting circuit as gas was moved through the pod.

Mechanical Design
An integrating sphere was constructed with two hemisphere Fat Daddio cake pans, which were painted white inside for the diffuse reflection of the light with Rust-Oleum Flat Protective Enamel spray paint. The paint was applied until a smooth, consistent white surface was obtained. Polishing sandpaper was used to sand the spheres until smooth to the touch between each painting to give a more consistent reflective surface. The pre-painted aluminum baffle was epoxied in after the paint applications. The two painted halves then had holes A, B, C, and D drilled through the metal (see figure of top view for locations and sizes).



Holes B and D had two half inch tubing brass nozzles epoxied on the outside of the sphere, which were inserted into vinyl tubing with a half inch inner diameter (15 mm). Plastic tubing was used to connect the sphere to two 3-D printed plastic fan housings containing micro fans. This was done to allow air to be pulled from the outside of the sphere to the inside, and then expelled, as the ozonesonde travelled upwards. This was done so that measurements taken would be actually from air where the pod was currently, instead of re-measuring air already inside of the sphere.

A diffusing ‘cap’ was constructed to externally house the LED lights (see electronic section for wavelengths and models used), so that the incident light would be diffuse when entering the sphere. The cap consisted of a Dura PVC Cap with holes drilled in on opposite sides to house the LED lights. The cap was positioned so that hole A was at its center, and it was then epoxied to the outer surface of the pan. Square holes were cut in the pod wall to inset the housings, which had the fans and housings inserted and secured with tape. Hole C had the phototransistor inserted in it. The phototransistor was wrapped in electrical tape to prevent conduction with the metal of the cake pans, and had to be wedged into place. All batteries and the three boards were bolted to particle board pieces, and placed inside of the pod. Due to limited space in the pod, movement of the boards was determined to be so limited that no additional method was used to secure them to the pod wall. The sphere was also so stable due to the tubing connecting the sphere to the fan housings, that no additional attachment method was used to secure it.



Parts Used:

Circuits and Parts
A simple phototransistor was used to capture light. This transistor lets a certain amount of voltage through the circuit based on the amount of light. The more light, the more voltage. A simple circuit was built to measure the voltage using the mbed. The two LEDs needed to be switched on and off back and forth. Two NPN transistors were used to switch these LEDs. When the transistor is set a high voltage, it completes the circuit internally and when it receives a low voltage (or no voltage) it disconnects the circuit. The two transistors were controlled by two separate digital outs from the mbed. Two 330 Ohm resistors were used to connect the circuit to ground. The circuits for the fans consisted of a L7805 voltage divider, used to divide the voltage from a 9 v battery to power the 5 volt microfans used for air circulation. The circuit was constructed using three 0.1 micro faraday capacitors in parallel, and one 0.1 micro faraday capacitor.



Coding
We used the ARM mbed site to build and share our code during the building and testing process. The mbed LPC1768 was used to receive voltage data from the phototransistor, store it to a micro SD card, and switch on and off the two LED lights. A simple SD Card program, using a public SD card repository, was used to take analog input from the transistor to the card. The file that the data was stored in was controlled by which LED light was on. The Red LED had it's own file and the Purple LED had it's own, too. These lights were controlled by separate 2N3904 NPN transistors which were turned on and off by high/low voltage sent from two digital outs of the mbed.

The entirety of the code used can be found at https://developer.mbed.org/users/Twidmer18/code/OzoneDetector/

Validation and Calibration
The completed ozone sensor was run in the test chamber, created by group before us, to ensure the voltage from the phototransistor was decreased as the amount of ozone in the tank increased. We ran multiple tests to validate this decrease by simply placing our detector in with an industrial ozone sensor from EcoSensors. We would read an increase in the ppm (parts per million) number displayed on industrial sensor they expect a drop in voltage from phototransistor. Problems began to occur when the industrial sensor would read" 0 ppm" when our detector was placed in the tank with the ozone maker on. Further investigation uncovered the material of the actual pod, a styrofoam type material, is a reactant with ozone causing the ozone in the tank to always be zero. This causes all previous validation data to be incorrect since all tests were completed with ozone detector placed in pod.

Data and Analysis
Sadly, due to problems securing the phototransistor, it popped out of the integrating sphere five minutes into launch. The amount of buffeting the pod received was going to receive was underestimated, and it turned out the immobility of the sensor due to the stiffness of the coiled wires and electrical tape used was not enough to keep the sensor in place.

Due to the limited amount of data collected, no real conclusions can be drawn from it. However, if the ozonesonde had functioned as planned, we would have expected to see concentration increase with altitude, peaking in the ozone layer. Then, on the return trip, we would expect more erratic data following the same general pattern as the pods descended.

Conclusion
This iteration of the ozone sensor project was a failure, as no usable data was gathered. While extra attention was given to the security of batteries, circuits, and the sphere itself, the phototransistor was not given a well-tested system to stay stationary. During testing in the lab, this problem was not apparent as the phototransistor's mobility made it quick and easy to prototype with as the construction of the pod progressed. When the pod was constructed, the coiled wire the phototransistor was connected to the mbed circuit with was stiff and relatively immobile, and the main body of the part was wrapped in a layer of electrical tape and had to be wedged into hole C. Because of the difficulty of getting the part into the proper place on the sphere, it was assumed that it would stay in. As an additional precaution, the corner of the pod containing the phototransistor wiring and transistor itself was kept clear of any parts which could have potentially come into contact with the transistor. However, these precautions were not enough, and as said above it did come out and no useful data was collected.

Future work on this project should be directed towards not just appropriately securing the phototransistor, but also to come up with more efficient methods of securing the batteries and other components in the pod. One such suggestion is the use of more 3-D printed mounts for all battery and circuit components, which could be inserted and secured externally through the walls of the pod. Also, the construction of the entire local sensor on a lightweight framework, of the appropriate size to be placed inside of the pod when completed, could be beneficial. That way, the problem with the material inside of the pod reacting to the ozone in the testing chamber could be avoided, and it would be easier to make sure that all parts and wires are completely immobile regardless of any amount of jostling, as they could be directly attached to the framework at all points. Making a primary attachment method and a backup attachment method in case of failure is also recommended, to decrease the probability of failure of the entire project due to parts coming loose. Whatever direction future work takes, the securing of the phototransistor should be the number one concern, followed by the securing of batteries as that was historically a problem. Also, more rigorous testing procedures after the final assembly of the pod should be implemented, to be sure that if will withstand the rigors of launch day.

References/Works Cited
Allen, Jeannie. “Ultraviolet Radiation: How it Affects Life on Earth”. NASA Earth Observatory. NASA. 6 Sept 2001. Web. 1 May 2016. .

“Chapter 5: Stratospheric Photochemistry”. Stratospheric Ozone: an Electronic Textbook. Center for Coastal Physical Oceanography. N.d. Web 19 May 2016. < http://www.ccpo.odu.edu/~lizsmith/SEES/ozone/class/Chap_5/index.htm>.

Clark, Jim. “USING UV-VISIBLE ABSORPTION SPECTRA”. Chemguide: Helping you to understand Chemistry. N.p. N.d. Web 15 May 2016. .

Gleason, Karin. “Science- Ozone Basics.” Stratospheric Ozone Monitoring and Research in NOAA. NOAA. 20 March 2008. Web. 1 May 2016. .

“How is Ozone Measured in the Atmosphere?”. Earth Systems Research Laboratory. NOAA. 2006. Web. 1 May 2016. < http://esrl.noaa.gov/csd/assessments/ozone/2006/chapters/Q5.pdf>.

Lindesy, Rebecca. “Antarctic Ozone Hole”. NASA Earth Observatory. NASA. 17 Sept. 1979. Web. 1 May 2016. .

“Part I: The History behind the Ozone Hole”. The Ozone Hole Tour. Centre for Atmospheric Science University of Cambridge. N.d. Web. 1 May 2016. .

“Real Integrating Spheres”. Gigahertz-Optik. Gigahertz-Optick Inc. N.d. Web 19 May 2016. .

“Stratospheric Ozone Depletion”. ACER. Association for Canadian Educational Resources. N.d. Web. 1 May 2016. .

“Technical Guide: Integrating Sphere Theory and Applications”. Labsphere. Labsphere. N.d. Web 19 May 2016. .

“Techniques used in the Nouspikel lab”. Medicine. Universite de Geneve. N.d. Web. 1 May 2016. .

“THE MONTREAL PROTOCOL ON SUBSTANCES THAT DEPLETE THE OZONE LAYER”. Ozone Secretariat. UNEP. N.d. Web 1 May 2016. .