Recently I got a special order for Atom Nano 15 gamma spectrometer equipped with BGO detector and here are some results on using BGO with SiPMs.
BGO is well-known for it's high (7.13 g/cm3) density but very poor light-yield, around 10 times less than light yield of CsI(Tl) at room temperature. SiPMs at RT are 3-6 orders of magnitude (!) noisier than PMTs, it makes them not a popular readout technique for BGO scintillators. Technical progress in SiPMs goes on increasing it's photon detection efficiencies and decreases noise, so why not to try?
Step by step...
1. BGO crystal with 55*18*15 mm size was ordered and detector assmbled.
Wrapped in Teflon tape, it reads 14% Cs137 resolution with Hamamatsu PMT at room temperatire.
2. Atom Nano spectrometer assembled with the detector bundle. Very heavy for it's size, the crystal itself is around 105 g!
3. Calibration is performed in temperature range between -15 and 57 degrees Centigrade, 4 degrees between calibration points. In Atom electronics all 3 compensations are used: SiPM bias voltage, baseline and gain.
The result is not too bad, not so many people using gamma spectrometers in sauna.
4. We lost some resolution and got large temperature drift to compensate in the internal spectrometer firmware, but what we got for it? Yes, you are right, we got efficiency, especially at high gamma energies. Here are some calculations made with limited version of LSRM monte-carlo simulation software for Marinelli geometry.
Dimensions are in centimeters.
Efficiency of BGO
Efficiency of CsI to compare
At Cesium's 662 keV and higher energies we've got around 3 times more efficient detector in the same size, if compared to more popular Cesium Iodide.
Also, one more confirmation of practical use of BGO with SiPMs readout.
Who want to have similar spectrometer - just ask me.
