Another interesting paper that popped up on the arXiv a few days ago is this one, which describes a prototype argon-based time projection chamber that would be used for measuring MeV to GeV scale gamma rays in astrophysics experiments. They hope to be able to reconstruct both the polarization and direction of incoming gamma rays by looking for electron-positron pair production. Certainly the direction should be easily measured by such a device: TPCs are commonly used as trackers for accelerator detectors as long as the event rate isn’t too high because of their excellent spatial resolution and minimal amount of mass in the active volume compared to solid-state detectors. Measuring polarization with a TPC, which sounds like a more difficult and maybe more interesting problem, is left to a couple other papers rather than this one.
While TPCs are standard particle detectors, I find this interesting because of the astrophysics connection. The paper mentions that one of the next tests would be a balloon-borne detector. We can then assume that a final version would be either on a balloon or in space. This would present many challenges that aren’t present in other detectors. The design would need to worry a lot more about things like weight, temperature, performance stability. Things like high voltage systems would need to be very robust and fully automated since the detector wouldn’t be accessible once it’s deployed. The design described in the paper uses both GEMs (gas electron multipliers) and Micromegas (a micromesh with pad or pixel readout) to achieve electron amplification. Both of these technologies can be quite fragile and probably aren’t very mechanically robust compared to something like a wire-based TPC. Even a single spark on one of these could destroy the detector. It would be interesting to see if such a device could survive a trip in a balloon or on a rocket. There are facilities available to shake objects to test their ability to handle being sent into space, so hopefully the people working on this project try something like that before going ahead with the enormously expensive step of actually deploying a prototype.