What properties of Germanium make it suitable for Dark Matter detectors? I tried googling but there was too many results describing the use of Germanium Chrystals at low tempretures for Dark Matter detaction but not the why of it.

Or alternatively what are the properties that make a good dark matter detector? in this case Germanium just gets to fit the requirement and have a look at other possible candidates.

Assuming Germanium is rare and hard to get I prompted this question, on the other hand if it was copper that was being used for such a task, it would have gone completely unnoticed.

  • $\begingroup$ This is something that I should know, but I am left with guessing. Presumably the same properties that make it suitable for a high-precision, low-background gamma counter: very low noise (approaching zero even in commercial cells), high time and energy resolution, and fairly high quantum efficiency. $\endgroup$ May 25, 2011 at 23:16
  • $\begingroup$ Most likely the germanium is used as a calorimeter. $\endgroup$
    – Georg
    May 27, 2011 at 9:42

1 Answer 1


WIMP's are hypothesized to interact only by the weak force, thus will interact only with the nucleus. The WIMP will hit the nucleus, slow down and change direction. The nucleus will recoil, and introduce vibrations (phonons) into the lattice. Those vibrations create heat which is measured. The crystal must me kept at 10 mK in order to reduce thermal noise. Additionally, the recoil of the nucleus lifts some electrons into the conduction band where they too can be measured. Interactions by alpha, beta and gamma rays generate more ionization than temperature rise. WIMP interactions will generate a large nuclear recoil and many phonons but will not interact directly with the shell electrons thus very few electrons will enter the conduction band. By looking for a large phonon signal and a relatively small ionization signal, WIMPS can be sorted out. But why Germanium? Because Ge has the smallest band gap of the semiconductors. An incident particle will lift more electrons out of the ground state and put them into the conduction state. The signal reaching the preamp is larger with Ge than any other semiconductor. Germanium is the gold standard for the detection of gamma rays. Silicon has a larger band gap thus is not as sensitive.

  • 3
    $\begingroup$ It's important to note that WIMP's are hypothetical particles, and, even if they are found to exist, are only a potential candidate for Dark Matter as found around galaxies and throughout the universe. Germanium detectors are useful for detecting WIMP's - if they exist. But if dark matter is in fact not weakly-interacting or if WIMP's are not the dark matter we observe, they are patently useless for dark matter studies. $\endgroup$
    – crasic
    May 26, 2011 at 4:24
  • $\begingroup$ Something doesn't add up here - band gap of germanium is ~800meV, the deBye T of Ge is 360K, making the highest energy phonons around 40 meV, the melting T of Ge is ~1200K, or 104 meV - so for a WIMP to excite the nucleus to drive electrons into the conduction band, it puts enough energy into the lattice to locally dissociate the Ge atom in the detector, thus reducing the crystallinity of the Ge, and doing who knows what to the band structure. This would make the detectors worse over time for the detection scheme above. Is this how the detectors are expected to operate? $\endgroup$
    – Jen
    May 26, 2011 at 11:07
  • $\begingroup$ Given that they only expect one WIMP per year, lattice damage is not going to be a problem. But, if there are local single atom dislocations to the lattice then annealing should put them right again. I have read in the literature (Nuclear Instrumentation Methods) of the machinations they went through in the early sixties to deal with noise in the nascent low purity Ge crystal detectors. I read many instances of annealing being used to quiet a crystal. $\endgroup$
    – Bill Slugg
    May 27, 2011 at 4:07
  • $\begingroup$ Just a brief commercial. If there are any retired GeLi operators out there, I need to find out about redrifting. Have you ever seen it done/paid to have it done? I own 3 ancient PGT jellies and would love to redissolve the Li/anneal them/drift them a bit/cool them and have an operable detector (albeit at a lower count rate than a new one). I would be thrilled to get one of my 20% units operable at even one percent. Right now I am going to follow the literature (NIM) and wing it. $\endgroup$
    – Bill Slugg
    May 28, 2011 at 15:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.