Measurement of scintillation and ionization yield and scintillation pulse shape from nuclear recoils in liquid argon

H. Cao et al. (SCENE Collaboration)

Phys. Rev. D 91, 092007 – Published 26 May 2015

Abstract

We have measured the scintillation and ionization yield of recoiling nuclei in liquid argon as a function of applied electric field by exposing a dual-phase liquid argon time projection chamber (LAr-TPC) to a low energy pulsed narrow band neutron beam produced at the Notre Dame Institute for Structure and Nuclear Astrophysics. Liquid scintillation counters were arranged to detect and identify neutrons scattered in the TPC and to select the energy of the recoiling nuclei. We report measurements of the scintillation yields for nuclear recoils with energies from 10.3 to 57.3 keV and for median applied electric fields from 0 to $970 V / cm$ . For the ionization yields, we report measurements from 16.9 to 57.3 keV and for electric fields from 96.4 to $486 V / cm$ . We also report the observation of an anticorrelation between scintillation and ionization from nuclear recoils, which is similar to the anticorrelation between scintillation and ionization from electron recoils. Assuming that the energy loss partitions into excitons and ion pairs from ${}^{83 m}{Kr}$ internal conversion electrons is comparable to that from ${}^{207}{Bi}$ conversion electrons, we obtained the numbers of excitons ( $N_{ex}$ ) and ion pairs ( $N_{i}$ ) and their ratio ( $N_{ex} / N_{i}$ ) produced by nuclear recoils from 16.9 to 57.3 keV. Motivated by arguments suggesting direction sensitivity in LAr-TPC signals due to columnar recombination, a comparison of the light and charge yield of recoils parallel and perpendicular to the applied electric field is presented for the first time.

27 More

Received 16 July 2014

DOI:https://doi.org/10.1103/PhysRevD.91.092007

Authors & Affiliations

Click to Expand

Observation of the dependence on drift field of scintillation from nuclear recoils in liquid argon

T. Alexander et al. (SCENE Collaboration)

Phys. Rev. D 88, 092006 (2013)

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 91, Iss. 9 — 1 May 2015

Reuse & Permissions

Access Options

Article Available via CHORUS

Download Accepted Manuscript

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
A schematic of the experiment setup. $θ_{1}$ is the neutron production angle and $θ_{2}$ is the scattering angle. The inset shows a zoomed-in view of the TPC including the PMTs, field shaping rings and PTFE support structure. It does not include the inner reflector.
Reuse & Permissions
Figure 2
geant4-based simulation of the energy deposition in the LAr-TPC at the 10.3 keV setting. Black: All scatters that produced a coincidence between the TPC and the neutron detector and survived the timing cuts discussed in the text. Blue: From neutrons scattered more than once in any part of the entire TPC apparatus before reaching the neutron detector. About 25% of these events are very shallow scatters depositing minimal energy elsewhere in the apparatus. They look very much like single scatters and produce the peak in the multiple scattering distribution at 10 keV. Each setting is labeled according to the median of the simulated single scatter distribution.
Reuse & Permissions
Figure 3
Ar gas system used for continuous purification of the LAr and injection of ${}^{83 m}{Kr}$ source.
Reuse & Permissions
Figure 4
Efficiency for the two TPC trigger conditions described in the text. Black: OR of the two TPC PMT’s. Red: AND of the two TPC PMT’s. See text for description of the measurement of the efficiency.
Reuse & Permissions
Figure 5
Distributions of pulse shape discrimination vs time of flight for data taken in the 57.3 keV configuration described in the text. See the text for the definition of the variables. Red boxes outline the regions selected by analysis cuts. Panels (a) and (b) described the TPC response and panel (c) the neutron detectors response. In panel (a), the clusters of events with $f_{90} < 0.1$ have S2 signals that start before the termination of S1 signals. Panel (b) shows the distribution of events selected with the requirement that S1 and S2 signals are properly resolved.
Reuse & Permissions
Figure 6
(a) Surviving primary scintillation light (S1) distributions for 57.3 keV nuclear recoils as the neutron selection cuts described in the text were imposed sequentially. Data were collected with a drift field of $193 V / cm$ and an extraction field of $3 kV / cm$ . The high energy peak around 187 PE is due to the ${}^{83 m}{Kr}$ source used for continuous monitoring of the detector. (b) Surviving distributions of electroluminescence light from ionization (S2) for 57.3 keV nuclear recoils after the same cuts. (c) S2 vs S1 distribution for all events with resolved/nonoverlapping S1 and S2 before the neutron selection cuts. (d) S2 vs S1 distribution for the events surviving the neutron selection cuts.
Reuse & Permissions
Figure 7
Comparison of recoil S1 spectra taken with the TPC PMT’s OR (black) and AND (red) trigger for the 20.5 keV setting at $970 V / cm$ . The integral between 12 and 60 PE for each spectrum is normalized to 1. Use of the coincidence trigger had no significant effect on the spectral shape above $\sim 12 PE$ .
Reuse & Permissions
Figure 8
Mean S2 signal vs S1 for $E_{d} = 193 V / cm$ at all recoil energies. For this drift field, an S1 signal of 4 PE corresponds to an S2 signal of 123 PE, while an S1 signal of 12 PE corresponds to an S2 signal of 196 PE. These values form the lower bounds for the fit ranges used in the S2 analysis for data taken with the OR trigger and AND trigger respectively.
Reuse & Permissions
Figure 9
S1 yield as a function of nuclear recoil energy measured at five drift fields (0, 96.4, 193, 293 and $970 V / cm$ ) relative to the light yield of ${}^{83 m}{Kr}$ at zero field.
Reuse & Permissions
Figure 10
S1 yield as a function of nuclear recoil energy measured at zero field relative to the light yield of ${}^{83 m}{Kr}$ at zero field, compared to previous measurements [8, 9].
Reuse & Permissions
Figure 11
Systematic error induced by chemical impurities affecting the mean life of the triplet component of the S1 scintillation spectrum, as a function of mean life in the range of interest. The S1 time profile was simulated with two exponential decay terms. Each line represents the events with a given $f_{90}$ when the slow component lifetime is $1.45 μ s$ . Note that $f_{90}$ increases slightly with the decrease in the slow component lifetime.
Reuse & Permissions
Figure 12
Energy resolution, $σ_{1}$ , of the nuclear recoils extracted from the Monte Carlo fit, as a function of recoil energy for all drift field combinations. We separate the results from the June and October 2013 runs. The resolution for the nuclear recoils, $σ_{1}$ , is fit (black continuous curve) with the function described in the text and compared with the fit obtained for the ${}^{83 m}{Kr}$ (purple dashed curve) with the same function.
Reuse & Permissions
Figure 13
(a) Distribution of $f_{90}$ vs S1 for 20.5 keV recoil data taken at $E_{d} = 193 V / cm$ . The vertical dashed lines indicate the boundaries of the region where S1 is within $1 σ$ of the mean of the Gaussian fit $μ$ , as described in the text. (b) Black: $f_{90}$ distribution for the 20.5 keV nuclear recoil events with S1 falling in the region in [ $μ - σ$ , $μ + σ$ ] i.e., for the events fall in between the vertical dashed lines in panel (a). Red: $f_{90}$ distribution model prediction (not a fit, see text). (c) Simulated distribution of $f_{90}$ vs S1 with $\sim 30$ times the statistics present in the data. (d) Black: Same in (b). Red: $f_{90}$ distribution of the simulated events that fall in between the vertical dashed lines in panel (c).
Reuse & Permissions
Figure 14
Median $f_{90}$ of nuclear recoils as a function of energy at several drift fields. The median values and statistical errors for all energies explored are listed in Table 5. All points have a common systematic error of 0.01 (see text for discussion).
Reuse & Permissions
Figure 15
Measured S2 yield as a function of $E_{d}$ at four recoil energies. Extraction field is fixed at $3.0 kV / cm$ and multiplication field at $4.5 kV / cm$ . To quote S2 yield in [ $e^{-} / keV$ ], an additional 10% systematic uncertainty must be combined with each error bar shown, to take into account the uncertainty in the single-electron calibration. The dashed curve shows the best fit of the modified Thomas-Imel model (see text) to ${}^{83 m}{Kr}$ data. The solid curves show the best fits of the same model (see text) to the nuclear recoil data.
Reuse & Permissions
Figure 16
Best fit S2 yield as a function of recoil energy at four different drift fields (96.4, 193, 293 and $486 V / cm$ ), with a fixed extraction field of $3.0 kV / cm$ and multiplication field of $4.5 kV / cm$ . To quote S2 yield in [ $e^{-} / keV$ ], an additional 10% systematic uncertainty must be combined with each error bar shown, to take into account the uncertainty in the single-electron calibration.
Reuse & Permissions
Figure 17
Resolution vs S2 in PE at each recoil energy and drift field. The resolution is determined through the Monte Carlo fit. The resolutions of ${}^{83 m}{Kr}$ are shown in the same plot. The best overall fits of $R_{2}$ (indicated by the fit curves) are $0.19 \pm 0.01$ for nuclear recoils and $0.26 \pm 0.02$ for ${}^{83 m}{Kr}$ .
Reuse & Permissions
Figure 18
(a) S1 yield vs S2 yield for ${}^{83 m}{Kr}$ . The best fit results for Eq. (9) are shown. (b) S1 yield vs S2 yield for nuclear recoils. The data in (a) and (b) are fit simultaneously with the intercepts free and the slopes taken as a common parameter.
Reuse & Permissions
Figure 19
Total quenching factor $L$ (relative to ${}^{83 m}{Kr}$ events as defined in the text) compared to Lindhard’s theory and the Lindhard-Birk combined model proposed by Mei et al. The best fit curve to Mei’s model with Birk’s constant $k B$ as free parameter yields $k B = 5.0 \pm 0.2 \times 1 0^{- 4} {MeV}^{- 1} g {cm}^{- 2}$ .
Reuse & Permissions
Figure 20
The data points show S2 vs energy for $E_{d} = 193 V / cm$ , where the energy axis is evaluated using S1 and our measured $L_{eff, {}^{83 m}{Kr}}$ values. The line shows the logarithmic relationship [Eq. (4)] between S2 and energy as obtained from fitting S2 to the MC.
Reuse & Permissions
Figure 21
Scintillation yield relative to null field (left panels) and ionization yield with nonzero drift field (right panels) of nuclear recoils at 16.9, 36.1 and 57.3 keV. Black: Momentum of nuclear recoil is perpendicular to $E_{d}$ . Red: Momentum of nuclear recoil is parallel to $E_{d}$ . Sources of systematic uncertainties common to both field orientations are not included in the error bars.
Reuse & Permissions
Figure 22
All panels. Black: Experimental data collected for 10.3 keV nuclear recoils. Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines.
Reuse & Permissions
Figure 23
Top left panel. Black: geant4-based simulation of the energy deposition in the SCENE detector at the setting devised to produce 14.8 keV nuclear recoils. Blue: From neutrons scattered more than once in any part of the entire TPC apparatus before reaching the neutron detector. All other panels. Black: Experimental data collected for 14.8 keV nuclear recoils. Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines.
Reuse & Permissions
Figure 24
Top left panel. Black: geant4-based simulation of the energy deposition in the SCENE detector at the setting devised to produce 16.9 keV nuclear recoils. Blue: From neutrons scattered more than once in any part of the entire TPC apparatus before reaching the neutron detector. All other panels. Black: experimental data collected for 16.9 keV nuclear recoils. Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines.
Reuse & Permissions
Figure 25
Top left panel. Black: geant4-based simulation of the energy deposition in the SCENE detector at the setting devised to produce 20.5 keV nuclear recoils. Blue: From neutrons scattered more than once in any part of the entire TPC apparatus before reaching the neutron detector. All other panels. Black: Experimental data collected for 20.5 keV nuclear recoils. Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines.
Reuse & Permissions
Figure 26
Top left panel. Black: geant4-based simulation of the energy deposition in the SCENE detector at the setting devised to produce 25.4 keV nuclear recoils. Blue: From neutrons scattered more than once in any part of the entire TPC apparatus before reaching the neutron detector. All other panels. Black: Experimental data collected for 25.4 keV nuclear recoils. Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines.
Reuse & Permissions
Figure 27
Top left panel. Black: geant4-based simulation of the energy deposition in the SCENE detector at the setting devised to produce 28.7 keV nuclear recoils. Blue: From neutrons scattered more than once in any part of the entire TPC apparatus before reaching the neutron detector. All other panels. Black: Experimental data collected for 28.7 keV nuclear recoils. Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines.
Reuse & Permissions
Figure 28
Top left panel. Black: geant4-based simulation of the energy deposition in the SCENE detector at the setting devised to produce 36.1 keV nuclear recoils. Blue: From neutrons scattered more than once in any part of the entire TPC apparatus before reaching the neutron detector. All other panels. Black: Experimental data collected for 36.1 keV nuclear recoils. Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines.
Reuse & Permissions
Figure 29
Top left panel. Black: geant4-based simulation of the energy deposition in the SCENE detector at the setting devised to produce 49.7 keV nuclear recoils. Blue: From neutrons scattered more than once in any part of the entire TPC apparatus before reaching the neutron detector. All other panels. Black: Experimental data collected for 49.7 keV nuclear recoils. Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines.
Reuse & Permissions
Figure 30
Top left panel. Black: geant4-based simulation of the energy deposition in the SCENE detector at the setting devised to produce 57.3 keV nuclear recoils. Blue: From neutrons scattered more than once in any part of the entire TPC apparatus before reaching the neutron detector. All other panels. Black: Experimental data collected for 57.3 keV nuclear recoils. Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines.
Reuse & Permissions
Figure 31
Black: Experimental data collected with $E_{d} = 96.5 V / cm$ . Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines. The $χ^{2}$ (sum across all spectra as defined in the text) and the total number of degrees of freedom are shown in the last panel.
Reuse & Permissions
Figure 32
Black: Experimental data collected with $E_{d} = 193 V / cm$ . Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines. The $χ^{2}$ (sum across all spectra as defined in the text) and the total number of degrees of freedom are shown in the last panel.
Reuse & Permissions
Figure 33
Black: Experimental data collected with $E_{d} = 293 V / cm$ . Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines. The $χ^{2}$ (sum across all spectra as defined in the text) and the total number of degrees of freedom are shown in the last panel.
Reuse & Permissions
Figure 34
Black: Experimental data collected with $E_{d} = 486 V / cm$ . Red: Monte Carlo fit of the experimental data. The range used for each fit is indicated by the vertical blue dashed lines. The $χ^{2}$ (sum across all spectra as defined in the text) and the total number of degrees of freedom are shown in the last panel.
Reuse & Permissions

Physical Review D

covering particles, fields, gravitation, and cosmology