NEST: a comprehensive model for scintillation yield in liquid xenon

Abstract

A comprehensive model for explaining scintillation yield in liquid xenon is introduced. We unify various definitions of work function which abound in the literature and incorporate all available data on electron recoil scintillation yield. This results in a better understanding of electron recoil, and facilitates an improved description of nuclear recoil. An incident gamma energy range of O(1 keV) to O(1 MeV) and electric fields between 0 and O(10 kV/cm) are incorporated into this heuristic model. We show results from a Geant4 implementation, but because the model has a few free parameters, implementation in any simulation package should be simple. We use a quasi-empirical approach, with an objective of improving detector calibrations and performance verification. The model will aid in the design and optimization of future detectors. This model is also easy to extend to other noble elements. In this paper we lay the foundation for an exhaustive simulation code which we call NEST (Noble Element Simulation Technique).

Search for two-neutrino double electron capture of Xe124 with XENON100

“…We employ γ-lines from neutron-activated xenon isotopes at 40 keV and 320 keV ( 129 Xe), 80 keV ( 131 Xe), 164 keV ( 131m Xe) and 236 keV ( 129m Xe). The energy E was obtained by a linear combination of the S1 and S2 signals measured in photoelectrons (PE), exploiting their anti-correlation [30,31] …”

Tritium calibration of the LUX dark matter experiment

“…We model recombination as follows [19,26,27]. Starting with a W -value of 13.7 eV, we assume that α, the initial ratio of excitons-to-ions prior to recombination, is 0.2 independent of energy and electric field [28,29].…”

First Results from the LUX Dark Matter Experiment at the Sanford Underground Research Facility

“…131m Xe and 129m Xe (164 and 236 keV de-excitations) afforded another internal calibration, providing a crosscheck of the photon detection and electron extraction efficiencies. To model these efficiencies, we employed field-and energy-dependent absolute scintillation and ionization yields from NEST [23][24][25], which provides an underlying physics model, not extrapolations, where only detector parameters such as photon detection efficiency, electron extraction efficiency and single electron response are inputs to the simulation. Using a Gaussian fit to the single phe area [26], together with the S1 spectrum of tritium events, the mean S1 photon detection efficiency was determined to be 0.14 ± 0.01, varying between 0.11 and 0.17 from the top to the bottom of the active region.…”

“…4), as this provides the dominant component of detector efficiency. We also show contours of approximated constant-energy [28], calculated from a linear combination of S1 and S2 [24,27,29] generated by converting the measured pulse areas into original photons and electrons (given their efficiencies).…”