Evidence of delayed light emission of tetraphenyl-butadiene excited by liquid-argon scintillation light

Abstract

Tetraphenyl-butadiene is the wavelength shifter most widely used in combination with liquid argon. The latter emits scintillation photons with a wavelength of 127 nm that need to be downshifted to be detected by photomultipliers with glass or quartz windows. Tetraphenyl-butadiene has been demonstrated to have an extremely high conversion efficiency, possibly higher than 100% for 127 nm photons, while there is no precise information about the time dependence of its emission. It is usually assumed to be exponentially decaying with a characteristic time of the order of one ns, as an extrapolation from measurements with exciting radiation in the near UV. This work shows that tetraphenyl-butadiene, when excited by 127 nm photons, re-emits photons not only with a very short decay time, but also with slower ones due to triplet states de-excitations. This fact can strongly contribute to clarifying the anomalies in liquid-argon scintillation light reported in the literature since the 1970s, namely, the inconsistency in the measured values of the long decay time constant and the appearance of an intermediate component. Similar effects should be also expected when the TPB is used in combination with helium and neon, which emit scintillation photons with wavelengths shorter than 127 nm.

Calorimetry for low-energy electrons using charge and light in liquid argon

“…The prompt singlet-state Ar * 2 component of a typical optical pulse in LArIAT is characterized by a rise time of ∼10 ns and a width of ∼50 ns at half-max, with pulse amplitudes in the range of ∼50-500 ADC counts. Each prompt pulse is followed by an exponential tail of late light from tripletstate Ar * 2 decays (τ t = 1.3 µs) and delayed TPB fluorescence [33]. To prevent a systematic mis-measurement of pulse areas due to a negative undershoot observed in the HMM PMT signal, a simple correction is crafted by exploiting an observed linear relationship between the integral around the peak of each pulse and the amplitude of the undershoot that follows it.…”

“…Late light from the muon will naturally overlap with the Michel candidate integration window. This contamination is exaggerated by the time structure of TPB fluourescence [33]. Attempts are made to correct for this by fitting the muon light in the range 0.4 µs to 1.8 µs to a function approximating the distribution of late light,…”

“…IV A). The second term in S late is meant to ac- count for the component of TPB fluorescence with the longest lifetime, τ * = 3.55 µs, which comprises 8% of the re-emitted visible light [33]. The normalization factor B is fixed to values based on fits of Eq.…”

“…Since LAr scintillation time structure is convolved with the four components of TPB fluorescence, with lifetimes ranging from ≈ 1 ns to as high as 3.55 µs [33], a conversion is needed between the measured late-light lifetime from waveforms to the "true" τ t . To accomplish this, a Monte Carlo (MC) study was performed in which photon arrival-time distributions were generated, incorporating LAr physics and TPB reemission effects (as described in the following subsection), with different levels of quenching applied.…”

Simulation of argon response and light detection in the DarkSide-50 dual phase TPC

“…Segreto [18] found delayed light emission components from TPB. As this is a recent result it has not been included in G4DS.…”