BHSU Abstract-Temperature Control System for CRDS Laser Spectroscopy Cell
O. Smith (Spearfish High School), J. Weiland (Aberdeen High School)
Steven Gabriel (Spearfish High School)
Dr. Kara Keeter (Black Hills State University)
The purpose of our research was to create a device to control the temperature of the Tiger Optics cell in the Black Hills State University laser spectroscopy lab. Lab assistants requested a device that would offer temperature control up to ± .1ºC. The system was showing huge variations in data as the temperature rose and fell; the metal on the devices was expanding marginally as the temperature in the room rose fractions of a degree, which threw off the readings. One could look at their data and see when a person walked in the room. Clearly the cell was very sensitive.
Upon viewing the setup of the cell, which was bolted to a table along with the laser, mirrors, AOM, and detector, our first impression was to create an insulated container to put it all in. The other option was building a temperature containment chamber around just the cell, however we chose not to do so as we thought the extra glass pane would distort the laser excessively. Then, we began researching ways to contain temperature and maintain it. There were a variety of ways to do so, including insulation, heating, and cooling. We looked at many forms of each: water jackets, vacuum chambers, and space drapes as insulation, Tenny chambers, chilled beam cooling, heat lamps, and heat sinks. There were various pros and cons to all of these, one of the most influential being expense. Clearly, a large Tenny or vacuum chamber was out of the question, however, there was a possibility we could combine them into an effective device. Then, we stumbled across a scientific article entitled Design and Capabilities of the Temperature Control System for the Italian Experiment Based on Precision Laser Spectroscopy for a New Determination of the Boltzmann Constant (A. Merlone, F. Moro, A. Castrillo, L. Gianfrani). The paper described how to create an isothermal cell with phenomenal stability; the temperature could be maintained to within ± .1º mK. This was much, much more precise than we needed; however, the paper offered valuable information.
We were able to draft our first design soon after. It was a large vacuum chamber (120 cm by 50 cm) kept at a rough vacuum (20 torr) with the laser setup inside on a bolt-plate. Underneath the experiment but within the vacuum there were tubes that would be pumped full of chilled water and anti-freeze (We were aiming for a temperature of about 20º F as the photodiode in the detector works better under cold conditions.) We looked at a variety of other refrigerants such as CO2 , ammonia, liquid nitrogen, and some R11 mixtures, but the simplest and cheapest way to go was with water. There was also the possibility of using water vapor and compression to cool the system, which is a very efficient form of refrigeration; however, water’s high specific volume would be plain unwieldy (New, Natural and Alternative Refrigerants. Dr. S F Pearson). We decided we should just avoid the change of state and chill the liquid water. The cooled system would maintain a generally constant temperature; the more precise temperatures would be maintained using thermistors and a feedback loop (explained later).
The shape of the vacuum chamber was eventually made to be a cross section of an octagon (an irregular hexagon) with vacuum ports on the sides where the chords would enter and exit from. Originally, it was dome-shaped; the creases in the metal were added for stability later. The top would be removable, but the sides would be attached to the base so as not to disrupt the chords when editing the setup. (There is an illustration of the earliest model we had designed, back when it was a dome instead of a hexagon and the detector was also being kept at a steady temperature. Of course, things have changed since this was drawn. Drawn by John Weiland in MS Paint.) We had three options when it came to metal: steel, stainless steel, and aluminum. The consensus was aluminum because it was cheaper than stainless steel, easier to work with than stainless steel, lightweight unlike steel of any kind, and does not corrode or rust like regular steel. Also, aluminum is not magnetic at all, whereas steel is residually magnetic. The metal’s gage would be about 12.
At this point in our research, we still wanted to create a massive vacuum chamber. After talking to Dr. Brianna Mount, the leader of the spectroscopy project, we began to question that decision. It was unwieldy to create such a large chamber just for temperature containment when what we were really targeting was the cell. (The detector’s issue with Johnson noise was deemed irrelevant, as long as it was lower than that of the cell.) So, with the help of Dr. Mount and Dr. Keeter, we began revising our plan.
As of now, the new plan is not completely finished; we still do not know what method would be best for insulating the cell from the outside environment, however, we have worked to create a functional feedback loop. The feedback loop is created using a thermistor, a TEC, and a heating element. Our target temperature is 30 ºC (as it is much easier to heat than cool, and we have pretty much abandoned the idea, and so it would be best to use a 50KΩ thermistor. It is an NTC thermistor (Negative Temperature Coefficient) in which the resistance decreases as the temperature outside increases. The thermistor will be attached to the surface of the cell. The feedback loop will be moderated by the MPT2500 TEC from Wavelength Electronics; when the temperature goes below 30º, the TEC will turn on the heating element. Polyimide Thermofoil Heaters act as the heating element; they are thin and flexible devices from Minco.
Future plans include some type of insulation for the cell so that its heat is kept even more stable and possibly some form of temperature control for the detector. The system has not been tested yet, however, it will hopefully be implemented. This temperature control system will eliminate the variations in data caused by temperature fluxuations which would mean a great deal to the laser spectroscopy experiment and to the greater project at DarkSide.