Identifying radioactive backgrounds and their source, so that measures can be taken to reduce them, is the key to success of the COBRA experiment. This task is made difficult, however, when there are a number of backgrounds in different locations and many unknown factors, such as the width of the dead layer.
It is possible that more than one combination of contaminants and detector properties can produce very similar energy spectra. For example, gamma radiation has a large range and it is difficult to identify whether it originated inside a detector, in the paint or in the supporting structures.
Beta radiation and alpha radiation can provide more clues to where a radiation source is because they have much shorter ranges. However, beta spectra are not easy to identify because the Q-value is also shared with a neutrino and it is also difficult to predict
what an alpha spectrum would look like from an external source as this would require extensive knowledge about intricate details such as the passivation paint smoothness.
Internal backgrounds are any nuclear decays that occur inside the crystal itself that are not neutrinoless DBDs. It is important to use only pure materials for neutrinoless DBD searches. If we need to modify the production process, we must know as soon as possible. Identifying internal backgrounds can be done by finding unique signatures in the data that can only be produced by a known decay inside the detector. The easiest decays to search for are those that have the most striking signatures, because they are least likely to be mimicked by another background or random processes.
For isotopes in the natural radioactive decay chains, the best chance of tagging them is made with the alpha radiation they produce. Because the range of alphas in CZT is very short (s11 m) one can be certain that they do not escape outside the detectors and that their full energy is measured. The only exception to this would be for events that occur within microns of the sides of a detector or if an internal defect exists that reduces the efficiency of a detector. Alpha decays produced by short-lived isotopes can be easily singled out because the probability of two random events to occur in quick succession at the Q-values of the decays is small. No cosmogenically activated isotopes that decay via alpha emission are expected in the crystals. This is because the heaviest component of the detectors, tellurium, has a maximum of only 130 nucleons. The isotopes of CZT could be broken up through spallation from cosmic rays to unstable lighter nuclei, but there are only 16 alpha emitters from lower mass isotopes 11 and all have half-lives of less than 7 seconds. Furthermore, these radionuclides are all located on the table of isotopes where the imbalance of neutron to proton number gives rise to short half-lives
of less than a minute, so any daughter alpha emitters have long decayed by the time data taking started underground. Searches for cosmogenics must be made by looking for coincidences of beta and gamma radiation, which is difficult because betas share their energy with a neutrino, and gammas have only a small probability of depositing their full energy in an array of detectors. The isotopes 40K, 60Co and 137Cs do not form part of the uranium and thorium chains and it cannot be easily tested whether they are internal or external because they decay via beta radiation or election capture to stable isotopes. Consequently, the main focus of this section is on the achievable goal of setting a limit on the activity of uranium and thorium backgrounds inside the detectors of the 64 array.