Calibration and characterization of the IceCube photomultiplier tube

Abstract

Over 5000 PMTs are being deployed at the South Pole to compose the IceCube neutrino observatory. Many are placed deep in the ice to detect Cherenkov light emitted by the products of high-energy neutrino interactions, and others are frozen into tanks on the surface to detect particles from atmospheric cosmic ray showers. IceCube is using the 10-in. diameter R7081-02 made by Hamamatsu Photonics. This paper describes the laboratory characterization and calibration of these PMTs before deployment. PMTs were illuminated with pulses ranging from single photons to saturation level. Parameterizations are given for the single photoelectron charge spectrum and the saturation behavior. Time resolution, late pulses and afterpulses are characterized. Because the PMTs are relatively large, the cathode sensitivity uniformity was measured. The absolute photon detection efficiency was calibrated using Rayleigh-scattered photons from a nitrogen laser. Measured characteristics are discussed in the context of their relevance to IceCube event reconstruction and simulation efforts

Search for a diffuse flux of astrophysical muon neutrinos with the IceCube 59-string configuration

“…The PMT quantum efficiency as well as the transparency of the glass and the optical gel make the module most sensitive to wavelengths in the ultraviolet and blue regions [20]. This is optimal for the Cherenkov radiation filtered through Antarctic ice.…”

IceTop: The surface component of IceCube

“…Each DOM [10] consists of a 10" Hamamatsu photomultiplier tube (PMT) [11] and electronic circuitry, contained inside a 33 cm glass pressure sphere, as shown in Fig. 10 a.…”

“…The charge distribution obtained from these pulses can be described by a Gaussian single-photoelectron peak and an exponential contribution at low charges [11]. The relation between high voltage setting and single-photoelectron peak is determined by varying the bias voltage between 1200 and 1900 V. These voltages are higher than those used for the low gain PMTs in IceTop (750 V and 1250 V for low gain and high-gain PMTs, respectively, Table 4) so that an extrapolation is necessary.…”

“…The absorption length in ice was taken to be 100 m from measurements in the deep-ice (this value is not critical). For the wavelength dependent PMT efficiencies averages of sample measurements are used [11]. Another study concerned the implementation of the snow cover of the tanks into the simulation.…”

“…To simulate the photomultipliers, single photoelectrons with Gaussian shaped waveforms (σ = 3.2 ns) and a charge randomly chosen from the measured single photoelectron spectrum [11] are superimposed. Pulse saturation is simulated depending on the instantaneous current according to the gain dependent functions depicted in Fig.…”