The Homestake Mine started yielding gold in 1876. If you had asked George Hearst, the operator at the time, if the mine would someday yield the secrets of the universe I bet he would have laughed you out of the room. But sure enough, by 1960 a laboratory deep in the mine started doing just that. Many experiments have been conducted there in the five and a half decades since. The Large Underground Xenon (LUX) experiment is one of them, and has been running is what is now called the Sanford Underground Research Facility (SURF) for about four years. LUX’s first round of data was collected in 2013, with the experiment and the rest of the data slated to conclude in 2016. The method, hardware, and results wrapped up in LUX are utterly fascinating.
It’s All About Noise Filtering
Detecting dark matter is exceedingly hard. So much so that we’ve never actually done it. A sufficiently large detector like the LUX can hope to catch just a few interactions per year. To make sure that these aren’t missed, it’s important to filter out as many non-events as possible. This is the reason behind the underground location; 4,850 feet of earth stand between the detector and open air to reduce cosmic rays by a factor of 1 million. That doesn’t get rid of everything but it becomes possible to discern false readings.
The ultra pure Xenon further filters this noise in the system. What is left is a very dark environment waiting for Weakly Interactive Massive Particles (WIMPs) to pass through it. Theory tells us that somewhere between millions and billions of the WIMP dark matter particles move through one square centimeter of space every second. But one square centimeter on a subatomic scale is a vast and empty space. With the noise filtered out, researchers are just waiting for a WIMP to collide with a Xenon nucleus.
Xenon is a scintillator; when the nucleus is struck by a fast-moving subatomic particle, it gives off a photon. Ionizing electrons are also a result of the interaction and they in turn create secondary scintillation. The pattern of primary and secondary light is specific to the type of interaction and, if measured properly, can be used to distinguish a dark matter interaction from other events. The good news is that we’re really good at measuring light. In fact, you can probably already guess what mechanism is used in the measurement: a Photomultiplier Tube (PMT).
PMTs are used in all kinds of scientific measurement equipment, and also appear often in medical devices and photography equipment. We seen this last example as a source for a PMT that Kerry Wong used to demonstrate the speed of light.
LUX has 122 PMTs in its sensor array. A fair amount of signal conditioning sets the levels before being fed into the custom triggering system. Then things really start to get interesting. Readings are summed into 16 groups of PMTs which are then processed by the triggering system to establish if the pattern is that of a possible dark matter interaction. If you need to do a lot of very fast, parallel processing, what kind of hardware do you choose? You’re right, you reach for an FPGA and build up a system around it.
If you’re on the edge of your seat for details, the research team has come through in a big way. In November they published a paper entitled FPGA-based Trigger System for the LUX Dark Matter Experiment. And even if you don’t want to dig that deep, Professor Frank L.H. Wolfs at the University of Rochester has a concise set of pages dedicated to the triggering hardware which was designed by Wojtek Skulski.
The current iteration of the trigger pulse digitizer board goes by the part number DDC-8DSP. Two of these boards use a Xilinx Spartan-3A to capture the signals using 14-bit resolution at 64 MHz. Each of these boards collects and processes the signals, then send the data to a Trigger Builder board via HDMI cables. This board gets its name because it is responsible for inspecting for a characteristic pattern of primary and secondary scintillation that indicates a WIMP interaction (and differentiates from other interactions). The diagram on the left is from the published paper and illustrates the signal pattern that the hardware is looking for.
Did it Work?
Ah, be careful what you ask. Yes, LUX worked, delivering the first round of data in 2013 as anticipated. The real question is did it detect dark matter? Evidence of dark matter has not been proven in that data. But this, the most sensitive detector ever build for this purpose, hasn’t gone to waste. The research team has been able to further establish what properties WIMPs do not possess.
The experiment is currently running another round, having been further configured based on the 2013 data. This data set is expected to be ready in June of this year and will be the last run for this iteration of LUX. In the works is the LUX-ZEPLIN project which will use a much larger detector apparatus with vastly more liquid xenon, 488 PMTs in the sensor array, and several other improvements.
Experiments that Deserve Celebrity Status
When first hearing about LUX, you might be reminded of similar experiments to detect neutrinos because both experiments need to be located underground to filter out cosmic noise. Neutrino experiments have a higher public profile, although they’re probably not at celebrity status like the Large Hadron Collider. And there will be an exciting neutrino detection experiment sharing he SURF facility with LUX before too long.
With one billion dollars committed to build the new Deep Underground Neutrino Experiment (DUNE), planned to start searching for neutrinos in 2022, it is easy to look at these types of experiments as the new space race. Surely, investment in experimentation will yield technological advancements similarly transformative as those which can be attributed to mankind making our way into space. The missing piece of the puzzle is widespread public awareness and excitement for scientific discovery.