betaFeedback

WorksPhysics and Astronomy

Detector Response Matrices, Effective Areas, and Flash-Effective Areas for Radiation Detectors

Abstract

A Detector Response Matrix (DRM) is a discrete representation of an instrument's Detector Response Function (DRF), which quantifies how many discrete energy depositions occur in a detector volume for a given distribution of particles incident on the detector. For simple radiation detectors that can count such energy depositions (such as scintillators, Proportional Counter Tubes (PCTs), etc), we consider the ideal counting DRF, Gφ(Ein,Edep)\mathbf{G}_φ(E_\mathrm{in}, E_\mathrm{dep}), which relates the detector's counting histogram (number of energy depositions within a given channel) to an incident particles characterization, φφ (e.g. incident flux, fluence, intensity). From the counting DRF we can derive the counting DRM, the effective area, and the flash effective area (which measures the total energy deposited in the detector from a large, instantaneous fluence).

LA-UR-25-31843

Approved for public release; distribution is unlimited.

Intended for: Report

Los Alamos National Laboratory, an affirmative action/equal opportunity employer, is operated by Triad National Security, LLC for the National Nuclear Security Administration of U.S. Department of Energy under contract 89233218CNA000001. By approving this article, the publisher recognizes that the U.S. Government retains nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or to allow others to do so, for U.S. Government purposes. Los Alamos National Laboratory requests that the publisher identify this article as work performed under the auspices of the U.S. Department of Energy. Los Alamos National Laboratory strongly supports academic freedom and a researcher's right to publish; as an institution, however, the Laboratory does not endorse the viewpoint of a publication or guarantee its technical correctness.

Overview

A Detector Response Matrix (DRM) is a discrete representation of an instrument’s Detector Response Function (DRF), which quantifies how many discrete energy depositions occur in a detector volume for a given distribution of particles incident on the detector. For simple radiation detectors that can count such energy depositions (such as scintillators, Proportional Counter Tubes (PCTs), etc), we consider the ideal counting DRF, Gφ(Ein,Edep)G_{\varphi}(E_{\mathrm{in}}, E_{\mathrm{dep}}), which relates the detector’s counting histogram (number of energy depositions within a given channel) to an incident particles characterization, φ\varphi (e.g. incident flux, fluence, intensity). From the counting DRF we can derive the counting DRM, the effective area, and the flash effective area (which measures the total energy deposited in the detector from a large, instantaneous fluence).

This work is a simplification of the detector efficiency defined in Sullivan 1971 [2]; here we assume a fixed incident angular distribution. While this approach does not consider a fully generalized detector efficiency, Gφ(Ein,Edep,Ω)G_{\varphi}(E_{\mathrm{in}}, E_{\mathrm{dep}}, \Omega), and is therefore not suitable to particle telescopes (where one wants to discriminate response based on variable look direction, see [2]), it illustrates the fundamental concepts and intuition behind these response functions, their discretization, and provides insight for how to extend to more generalized response functions.

Finally, this work assumes ideal detectors (able to perfectly resolve energy depositions), and ignores many practical aspects of detector characterization, such as pulse shaping, dead-time, quenching, pulse shape discrimination, etc.

Introduction

Particle detectors do not measure particles, they measure energy depositions. The basic operating principle of a radiation detector that characterizes energy depositions is illustrated in figure 2-1 for a scintillation detector. This concept can be applied to other types of detectors that measure, for example, charge instead of scintillation light, such as a neutron proportional counter tube. Ideally a detector is constructed in such a way that it is only sensitive to the particle type to be detected, e.g., a neutron detector being constructed in such a way that only neutrons create energy depositions. However, often other incident particles can create energy depositions in that sensitive volume as well but techniques like pulse-shape discrimination (PSD) may be used to disambiguate them, but again, this is looking at variations in energy depositions.

The most general characterization of NinN_{\mathrm{in}} (number of particles in the environment incident on a detector) is given by the particle radiance spectrum Φ˙E,Ω\dot{\Phi}_{E,\Omega}, sometimes referred to particle intensity JJ, according to

Φ˙E,Ω=dNindt dA dEin dΩ.(1)\dot{\Phi}_{E,\Omega} = \frac{\mathrm{d}N_{\mathrm{in}}}{\mathrm{d}t\,\mathrm{d}A\,\mathrm{d}E_{\mathrm{in}}\,\mathrm{d}\Omega}. \tag*{(1)}

For example, consider the electron background in the Earth’s radiation belts. These electrons will gyrate, bounce, and drift along magnetic field lines. For a given point in the radiation belts, one can ask the question: how many electrons traverse through a small surface area dA\mathrm{d}A along a direction θ\theta relative to the surface normal, confined within a solid angle dΩ\mathrm{d}\Omega per unit time dt\mathrm{d}t per unit particle energy dEin\mathrm{d}E_{\mathrm{in}}, as shown in figure 2-2.

Different integrations or moments of this relation yield different quantities defined in table 2-1. For example, for a given angular spectral flux, the number of particles over ranges of time, tt, incident surface area, SS, solid angle, Ω\Omega, and incident energy, EinE_{\mathrm{in}}, are

Introduction

Image not extracted

Figure 2-1. Basic scintillation detector. An incident particle of ionizing radiation characterized by its incident energy, EinE_{\mathrm{in}} (1), deposits an amount of energy EdepE_{\mathrm{dep}}, in a volume of scintillator (cyan) and releases scintillation light (2). Some of this light travels through the scintillator and causes the emission of a photo-electron (3) at the cathode of an attached photomultiplier tube (PMT). This photo-electron cascades and multiplies through the high voltage dynode chain (4) and produces an analog current signal at the output (5), which can be input into a multichannel analyser (MCA). The size of this analog pulse is proportional to the energy deposition (EdepE_{\mathrm{dep}}) in the scintillator at (2).

Image not extracted

Figure 2-2. Geometry for describing intensity, JJ of particles at a point in space. Adapted from “Radiative Processes in Astrophysics" by Rybicki and Lightman [1].

Introduction

Nin=∫dNin(2)N_{\mathrm{in}} = \int\mathrm{d}N_{\mathrm{in}} \tag*{(2)}
=∫t∫S∫Ω′∫Ein=0∞Φ˙E,Ω dt n^⋅r^ dA dΩ dE(3)= \int_t \int_S \int_{\Omega'} \int_{E_{\mathrm{in}}=0}^{\infty} \dot{\Phi}_{E,\Omega}\,\mathrm{d}t\,\hat{n}\cdot\hat{r}\,\mathrm{d}A\,\mathrm{d}\Omega\,\mathrm{d}E \tag*{(3)}

where n^⋅r^=cos⁡θ\hat{n}\cdot\hat{r}=\cos\theta, n^\hat{n} is the direction of the surface area, and r^\hat{r} is the direction alone the solid angle dΩ\mathrm{d}\Omega.

It is important to distinguish counts of particles in the background environment (i.e., radiation belt electrons, solar energetic particles, etc) from counts of things recorded by a detector (i.e. particles entering a detector volume, energy depositions within that volume, etc). Typically, particle environments are characterized by the number of incident particles NinN_{\mathrm{in}}, and their associated incident energies, EinE_{\mathrm{in}}; counting detectors count the number of energy depositions, NdepN_{\mathrm{dep}}, within a given energy range, or channel, around EdepE_{\mathrm{dep}}. It is the purpose of the Detector Response Function or Detector Response Matrix to relate these quantities for a given incident particle characterization, φ\varphi.

Introduction

N=N =Number of particles in environment
N˙=dNdt\dot{N} = \frac{\mathrm{d}N}{\mathrm{d}t}Particle rate
number of particles per unit time (s−1\mathrm{s}^{-1})
Φ=dNdA\Phi= \frac{\mathrm{d}N}{\mathrm{d}A}Particle Fluence
number of particles crossing a unit area (cm−2\mathrm{cm}^{-2})
Φ˙=dΦdt\dot{\Phi} = \frac{\mathrm{d}\Phi}{\mathrm{d}t}Particle Fluence Rate (Flux)
number of particles crossing a unit area per unit time (cm−2 s−1\mathrm{cm}^{-2}\ \mathrm{s}^{-1})
ΦE=dΦdE\Phi_E = \frac{\mathrm{d}\Phi}{\mathrm{d}E}Particle Fluence Spectrum (Spectral Fluence)
number of particles per incident energy per unit area (cm−2 keV−1\mathrm{cm}^{-2}\ \mathrm{keV}^{-1})
Φ˙E=dΦ˙dE\dot{\Phi}_E = \frac{\mathrm{d}\dot{\Phi}}{\mathrm{d}E}Particle Fluence Rate Spectrum (Spectral Flux)
number of particles per incident energy per unit area per unit time (cm−2 keV−1 s−1\mathrm{cm}^{-2}\ \mathrm{keV}^{-1}\ \mathrm{s}^{-1})
Φ˙E,Ω=dΦ˙EdΩ\dot{\Phi}_{E,\Omega} = \frac{\mathrm{d}\dot{\Phi}_E}{\mathrm{d}\Omega}Particle Radiance Spectrum (Angular Spectral Flux)
particles per incident energy per unit area per unit time per unit
solid angle (cm−2 keV−1 s−1 Sr−1\mathrm{cm}^{-2}\ \mathrm{keV}^{-1}\ \mathrm{s}^{-1}\ \mathrm{Sr}^{-1})
Also called JJ, “particle intensity” [1]
φ\varphiIncident particle quantification;
Generic variable indicating any incident particle quantification,
(e.g. fluence, flux, spectral flux, angular spectral flux, intensity, etc).

Table 2-1. Definitions for specifying particle environment, from ICRU Report 85[3]

Counting Response Function, DRF

A detector response function, Gφ(Ein,Edep)\mathbf{G}_{\varphi}(E_{\mathrm{in}}, E_{\mathrm{dep}}), is a differentiable quantity that relates the number of energy depositions occurring within a detector volume to some quantification of the incident particle environment, φ\varphi; the particle environment can be quantified in many ways, according to either the incident fluence, Φ\Phi, fluence rate, Φ˙\dot{\Phi}, fluence spectrum, ΦEin\Phi_{E_{\mathrm{in}}}, etc (see table 2-1).

The units of Gφ(Ein,Edep)\mathbf{G}_{\varphi}(E_{\mathrm{in}}, E_{\mathrm{dep}}) will depend on the incident particle quantification φ\varphi. In practice, the incident particle quantification should be explicitly used and stated when discussing the response; for example, if a detector response is constructed for the incident particle fluence spectrum, ΦE\Phi_E, the corresponding response should be written as GΦE(Ein,Edep)\mathbf{G}_{\Phi_E}(E_{\mathrm{in}}, E_{\mathrm{dep}}), and referred to as the ‘particle fluence spectrum detector response’. This will make it clear as to both how the detector response has been defined, and how it should be used in practice, as discussed below.

The detector response function, Gφ(Ein,Edep)\mathbf{G}_{\varphi}(E_{\mathrm{in}}, E_{\mathrm{dep}}), for the number of energy depositions, NdepN_{\mathrm{dep}}, around energy EdepE_{\mathrm{dep}}, within a detector volume, per unit incident particle quantification, φ^\hat{\varphi}, per unit incident energy, EinE_{\mathrm{in}}, per unit deposited energy, EdepE_{\mathrm{dep}}, is defined as

Gφ(Ein,Edep′)≡1φ^(Ein′)∂2Ndep∂Ein∂Edep∣Ein′,Edep′′(4)\mathbf{G}_{\varphi}(E'_{\mathrm{in},E'_{\mathrm{dep}}) \equiv\left.\frac{1}{\hat{\varphi}(E'_{\mathrm{in}})}\frac{\partial^2 N_{\mathrm{dep}}}{\partial E_{\mathrm{in}}\partial E_{\mathrm{dep}}}\right|_{E'_{\mathrm{in}},E'_{\mathrm{dep}}}} \tag*{(4)}

The quantity φ^\hat{\varphi} is the unit value of φ\varphi. So for example, if the incident particle characterization is flux, φ≡Φ\varphi\equiv\Phi, and the value of the flux at Ein=E0E_{\mathrm{in}} = E_0 is 50 particle/cm2^2, then the ratio Φ(E0)/Φ^(E0)=50\Phi(E_0)/\hat{\Phi}(E_0) = 50.

Given an incident particle quantification, the total number of energy depositions, NdepN_{\mathrm{dep}}, occurring in the detector volume with energies between EAE_A and EBE_B can then be calculated according to the integral expression

Ndep([EA,EB]dep)=∫EAEB∫0∞Gφ(Ein,Edep)[φ(Ein)] dEindEdep(5)N_{\mathrm{dep}([E_A,E_B]_{\mathrm{dep}}) = \int_{E_A}^{E_B}\int_{0}^{\infty}\mathbf{G}_{\varphi}(E_{\mathrm{in}},E_{\mathrm{dep}})\left[\varphi(E_{\mathrm{in}})\right]\,\mathrm{d}E_{\mathrm{in}}\mathrm{d}E_{\mathrm{dep}}} \tag*{(5)}

where we have used a shorthand notation to indicate the energy deposition channel, [EA,EB]dep[E_A,E_B]_{\mathrm{dep}}, defined by

[EA,EB]dep≡EA<Edep<EB(6)[E_A,E_B]_{\mathrm{dep}} \equiv E_A < E_{\mathrm{dep}} < E_B \tag*{(6)}

We can think of integral equations like 5 as consisting of a kernel operator (a function inside an integral, in this case Gφ(Ein,Edep)\mathbf{G}_{\varphi}(E_{\mathrm{in}},E_{\mathrm{dep}})), acting on, or simply multiplying a kernel argument (the thing inside the square brackets, [ ][\,], in this case, φ(Ein)\varphi(E_{\mathrm{in}})).1

Note that equation 5 is a simplification of the definition for the counting rate spectrum found in Sullivan 1971[2], where here we have integrated out the angular dependence.

Counting Response Matrix, DRM

To derive the Detector Response Matrix, we discretize the integral expression for the total number of energy depositions in equation 5, and look to express it in compact matrix notation.

Counting Response Matrix, Gφ(j,k)G_{\varphi}^{(j,k)}

Image not extracted

Figure 4-1. Tally Matrix, nj,kn_{j,k}, of number of energy depositions occurring in detector volume within Edep(j)≤Edep<Edep(j+1)E_{\mathrm{dep}}^{(j)} \le E_{\mathrm{dep}} < E_{\mathrm{dep}}^{(j+1)}, for incident particles with energy within Ein(k)≤Ein<Ein(k+1)E_{\mathrm{in}}^{(k)} \le E_{\mathrm{in}} < E_{\mathrm{in}}^{(k+1)}. Red bins track number of incident particles that resulted in no energy deposition, Edep(j)=0E_{\mathrm{dep}}^{(j)} = 0 (which can also be thought of as ‘underflow’ bins).

Intuitively, a matrix product is a compact way of expressing a discrete integral expression over a single integral or summation2. We note that equation 5 is a double integral. We consider an infinitesimal range of energy depositions in the channel [Edep(j),Edep(j)+ΔEdep(j)]dep[E_{\mathrm{dep}}^{(j)}, E_{\mathrm{dep}}^{(j)} + \Delta E_{\mathrm{dep}}^{(j)}]_{\mathrm{dep}}, and define the energy deposition histogram, Hdep(Edep)H_{\mathrm{dep}}(E_{\mathrm{dep}}) according to

Hdep(Edep(j))≡Ndep([Edep(j),Edep(j)+ΔEdep(j)]dep)(7)H_{\mathrm{dep}}(E_{\mathrm{dep}}^{(j)}) \equiv N_{\mathrm{dep}}([E_{\mathrm{dep}}^{(j)}, E_{\mathrm{dep}}^{(j)} + \Delta E_{\mathrm{dep}}^{(j)}]_{\mathrm{dep})} \tag*{(7)}
≡∫Edep(j)Edep(j)+ΔEdep(j)∫0∞Gφ(Ein,Edep)φ(Ein) dEindEdep(8)\equiv\int_{E_{\mathrm{dep}}^{(j)}}^{E_{\mathrm{dep}}^{(j)}+\Delta E_{\mathrm{dep}}^{(j)}} \int_{0}^{\infty} G_{\varphi}(E_{\mathrm{in}}, E_{\mathrm{dep}})\varphi(E_{\mathrm{in}})\,\mathrm{d}E_{\mathrm{in}}\mathrm{d}E_{\mathrm{dep}} \tag*{(8)}
≈∫0∞Gφ(Ein,Edep)φ(Ein) dEinΔEdep(j)(9)\approx\int_{0}^{\infty} G_{\varphi}(E_{\mathrm{in}}, E_{\mathrm{dep}})\varphi(E_{\mathrm{in}})\,\mathrm{d}E_{\mathrm{in}}\Delta E_{\mathrm{dep}}^{(j)} \tag*{(9)}

Equation 10 is now a single integral equation, and now amenable to expression as a matrix product.

We define a discrete coordinate grid over the Energy Deposition, EdepE_{\mathrm{dep}}, and Incident Energy, EinE_{\mathrm{in}}, according to figure 4-1; specifically we use jj and kk to index into the discrete energy binnings, Edep(j)E_{\mathrm{dep}}^{(j)}, and Ein(k)E_{\mathrm{in}}^{(k)}, respectively.

We can express the integral form of 10 in discrete form as

y(j)=∑k=1nA(j,k)x(k)(10)y^{(j)} = \sum_{k=1}^{n} A^{(j,k)}x^{(k)} \tag*{(10)}
Hdep(Edep(j))=∑kGφ(Ein(k),Edep(j))φ(Ein(k))ΔEin(k)ΔEdep(j)(11)H_{\mathrm{dep}}(E_{\mathrm{dep}}^{(j)}) = \sum_{k} G_{\varphi}(E_{\mathrm{in}}^{(k)}, E_{\mathrm{dep}}^{(j)})\varphi(E_{\mathrm{in}}^{(k)})\Delta E_{\mathrm{in}}^{(k)}\Delta E_{\mathrm{dep}}^{(j)} \tag*{(11)}

where the number of discrete energy depositions per energy deposition bin, or counting histogram, is defined as

Hdep(j)=Hdep(Edep(j))(12)\mathbf{H}_{\mathrm{dep}}^{(j)} = H_{\mathrm{dep}}(E_{\mathrm{dep}}^{(j)}) \tag*{(12)}

and can be expressed in compact matrix notation according to

Hdep=Gφφ(13)\mathbf{H}_{\mathrm{dep}} = \mathbf{G}_{\varphi}\varphi \tag*{(13)}

where the detector response matrix, Gφ\mathbf{G}_{\varphi} is defined to be

Gφ(j,k)≡Gφ(Ein(k),Edep(j))ΔEdep(j)ΔEin(k)(14)\mathbf{G}_{\varphi}^{(j,k)} \equiv G_{\varphi}(E_{\mathrm{in}}^{(k)}, E_{\mathrm{dep}}^{(j)})\Delta E_{\mathrm{dep}}^{(j)}\Delta E_{\mathrm{in}}^{(k)} \tag*{(14)}

and the incident particle quantification vector, φ\varphi is given by

φ(k)≡φ(Ein(k)).(15)\varphi^{(k)} \equiv\varphi(E_{\mathrm{in}}^{(k)}) . \tag*{(15)}

Counting Effective Area, Aφ(Ein,[EA,EB]dep)A_{\varphi}(E_{\mathrm{in}}, [E_A, E_B]_{\mathrm{dep}})

We can evaluate the EdepE_{\mathrm{dep}} integral in equation 5 to reduce an integration, and in the process define a new function, or kernel operator, called the effective area, Aφ\mathbf{A}_{\varphi}, that is the contribution per incident particle quantification within given energy deposition channel to the total instrument’s count rate.

Ndep([EA,EB]dep)=∫0∞Aφ(Ein,[EA,EB]dep)φ(Ein)dEin(16)N_{\mathrm{dep}}([E_A, E_B]_{\mathrm{dep}}) = \int_{0}^{\infty} A_{\varphi}(E_{\mathrm{in}}, [E_A, E_B]_{\mathrm{dep}})\varphi(E_{\mathrm{in}})\mathrm{d}E_{\mathrm{in}} \tag*{(16)}

where

Aφ(Ein,[EA,EB]dep)≡∫EAEBGφ(Ein,Edep)dEdep.(17)A_{\varphi}(E_{\mathrm{in}}, [E_A, E_B]_{\mathrm{dep}}) \equiv\int_{E_A}^{E_B} G_{\varphi}(E_{\mathrm{in}}, E_{\mathrm{dep}})\mathrm{d}E_{\mathrm{dep}} . \tag*{(17)}

By reducing equation 16 to a single integration, we can discretize it and reduce it’s computation to a compact vector dot product calculation.

Discrete Effective Area, Aφ,[EA,EB]dep(k)A^{(k)}_{\varphi,[E_A,E_B]_{\mathrm{dep}}}

Discretizing equation (16) and 17, and using the definition of the DRM in equation (14), we can express the integral counts, as

Ndep([EA,EB]dep)=∑k(∑j∈EA<Edep(j)<EBGφ(Ein(k),Edep(j))ΔEdep(j)ΔEin(k))φ(Ein(k))(18)N_{\mathrm{dep}}([E_A,E_B]_{\mathrm{dep}})=\sum_k\left(\sum_{\substack{j\in\\E_A<E_{\mathrm{dep}}^{(j)}<E_B}}\mathbf{G}_{\varphi}(E_{\mathrm{in}}^{(k)},E_{\mathrm{dep}}^{(j)})\Delta E_{\mathrm{dep}}^{(j)}\Delta E_{\mathrm{in}}^{(k)}\right)\varphi(E_{\mathrm{in}}^{(k)}) \tag*{(18)}
=∑k(∑j∈EA<Edep(j)<EBGφ(j,k))φ(Ein(k))(19)=\sum_k\left(\sum_{\substack{j\in\\E_A<E_{\mathrm{dep}}^{(j)}<E_B}}\mathbf{G}_{\varphi}^{(j,k)}\right)\varphi(E_{\mathrm{in}}^{(k)}) \tag*{(19)}
=∑kAφ,[EA,EB]dep(k)φ(Ein(k))(20)=\sum_k A_{\varphi,[E_A,E_B]_{\mathrm{dep}}}^{(k)}\varphi(E_{\mathrm{in}}^{(k)}) \tag*{(20)}

or recalling the definition of the vector dot product 3,

Ndep([EA,EB]dep)=Aφ,[EA,EB]dep(k)⋅φ(21)N_{\mathrm{dep}}([E_A,E_B]_{\mathrm{dep}})=A_{\varphi,[E_A,E_B]_{\mathrm{dep}}}^{(k)}\cdot\varphi \tag*{(21)}

where

Aφ,[EA,EB]dep(k)≡∑j∈EA<Edep(j)<EBGφ(j,k).(22)A_{\varphi,[E_A,E_B]_{\mathrm{dep}}}^{(k)}\equiv\sum_{\substack{j\in\\E_A<E_{\mathrm{dep}}^{(j)}<E_B}}\mathbf{G}_{\varphi}^{(j,k)}. \tag*{(22)}

Flash Effective Area, Fφ(Ein)F_{\varphi}(E_{\mathrm{in}})

In this section we will develop the concept of a flash effective area, which measures the total energy deposition in our instrument, and is functionally analogous to the counting effective area developed in the previous section.

We start by writing the integral representation for the total energy deposited, TdepT_{\mathrm{dep}}, for an incident particle characterization, φ(Ein)\varphi(E_{\mathrm{in}}), given by

Tdep=∫0∞∫0∞Gφ(Ein,Edep)[φ(Ein)Edep] dEindEdep.(23)T_{\mathrm{dep}}=\int_{0}^{\infty}\int_{0}^{\infty}\mathbf{G}_{\varphi}(E_{\mathrm{in}},E_{\mathrm{dep}})[\varphi(E_{\mathrm{in}})E_{\mathrm{dep}}]\,\mathrm{d}E_{\mathrm{in}}\mathrm{d}E_{\mathrm{dep}}. \tag*{(23)}

Previously, we defined our effective area as the integral over EdepE_{\mathrm{dep}} within the expression for the total number of counts (see equation (17)); this produced a kernel operator that could be convolved with the incident particle characterization, φ(Ein)\varphi(E_{\mathrm{in}}) (the kernel argument of 16). For the flash response, we wish to create a kernel operator that can be convolved with the incident energy-characterization, φ(Ein)Ein≡φ(Ein)×Ein\varphi(E_{\mathrm{in}})E_{\mathrm{in}} \equiv\varphi(E_{\mathrm{in}}) \times E_{\mathrm{in}}. We see in equation 23 that we don’t quite have the kernel argument we want; we have φ(Ein)Edep≡φ(Ein)×Edep\varphi(E_{\mathrm{in}})E_{\mathrm{dep}} \equiv\varphi(E_{\mathrm{in}}) \times E_{\mathrm{dep}}, which is not a well defined physical quantity. We can get to an incident-energy characterization as the kernel argument of 23 if we introduce the factor Ein/Ein=1E_{\mathrm{in}}/E_{\mathrm{in}} = 1 within the integral, and group the factor Edep/EinE_{\mathrm{dep}}/E_{\mathrm{in}} with the detector response function. We therefore define the the total energy deposition with the following functional form

Tdep=∫0∞Fφ(Ein)[φ(Ein)Ein] dEin(24)T_{\mathrm{dep}} = \int_{0}^{\infty} F_{\varphi}(E_{\mathrm{in}}) [\varphi(E_{\mathrm{in}})E_{\mathrm{in}}]\,\mathrm{d}E_{\mathrm{in}} \tag*{(24)}

then comparing equations24 with 23, it follows that

Fφ(Ein)≡∫0∞Gφ(Ein,Edep)(EdepEin) dEdep.(25)F_{\varphi}(E_{\mathrm{in}}) \equiv\int_{0}^{\infty} G_{\varphi}(E_{\mathrm{in}},E_{\mathrm{dep}}) \left(\frac{E_{\mathrm{dep}}}{E_{\mathrm{in}}}\right)\,\mathrm{d}E_{\mathrm{dep}} . \tag*{(25)}

By this definition, the flash effective area will have the same units as the counting effective area, and we have created a kernel operator which can be convolved with a physically well defined quantity, the incident-energy characterization, φ(Ein)Ein\varphi(E_{\mathrm{in}})E_{\mathrm{in}}.

Discrete Flash Effective Area, Fφ(k)F_{\varphi}^{(k)}

Discretizing equation 24, we have that the

Tdep=∑kFφ(Ein(k)) φ(k)Ein(k)ΔEin(k)(26)T_{\mathrm{dep}} = \sum_{k} \mathbf{F}_{\varphi}(E_{\mathrm{in}}^{(k)})\,\varphi^{(k)}E_{\mathrm{in}}^{(k)}\Delta E_{\mathrm{in}}^{(k)} \tag*{(26)}

which can be expressed in compact matrix notation (using the dot product) according to

Tdep=Fφ⋅(φEin)(27)T_{\mathrm{dep}} = \mathbf{F}_{\varphi} \cdot(\varphi E_{\mathrm{in}}) \tag*{(27)}

where

Fφ(k)≡Fφ(Ein(k))ΔEin(k)(28)\mathbf{F}_{\varphi}^{(k)} \equiv\mathbf{F}_{\varphi}(E_{\mathrm{in}}^{(k)})\Delta E_{\mathrm{in}}^{(k)} \tag*{(28)}

and

(φEin)(k)≡φ(Ein(k))Ein(k).(29)(\varphi E_{\mathrm{in}})^{(k)} \equiv\varphi(E_{\mathrm{in}}^{(k)})E_{\mathrm{in}}^{(k)} . \tag*{(29)}

A detector flash response, Fφ(Ein)\mathbf{F}_{\varphi}(E_{\mathrm{in}}) estimates the average fractional energy deposition per unit incident particle characterization and is computed according to

Fφ(k)≡1φ(k)∑i=1Nin(k)(Ei,dep(k)Ein(k))(30)\mathbf{F}_{\varphi}^{(k)} \equiv\frac{1}{\varphi^{(k)}} \sum_{i=1}^{N_{\mathrm{in}}^{(k)}} \left(\frac{\mathcal{E}_{i,\mathrm{dep}}^{(k)}}{E_{\mathrm{in}}^{(k)}}\right) \tag*{(30)}

where the summation is over the discrete Monte Carlo events, indicated by the index ii, where Ei,dep(k)\mathcal{E}_{i,\mathrm{dep}}^{(k)} is the energy deposition in the detector resulting from the ithi^{\mathrm{th}} Monte Carlo throw of incident particles with energy in bin Ein(k)E_{\mathrm{in}}^{(k)}.

The Flash response in 30 can be computed exactly using a counting DRM, Gj,k\mathbf{G}_{j,k} according to

Fφ(k)≡1φ(k)∑j(nj,k⟨Edep(j)⟩Ein(k))(31)\mathbf{F}_{\varphi}^{(k)} \equiv\frac{1}{\varphi^{(k)}} \sum_j \left(\frac{n_{j,k}\langle E_{\mathrm{dep}}^{(j)}\rangle}{E_{\mathrm{in}}^{(k)}}\right) \tag*{(31)}
=∑jGφ(j,k)(⟨Edep(j)⟩Ein(k))(32)= \sum_j \mathbf{G}_{\varphi}^{(j,k)}\left(\frac{\langle E_{\mathrm{dep}}^{(j)}\rangle}{E_{\mathrm{in}}^{(k)}}\right) \tag*{(32)}

where ⟨Edep(j)⟩\langle E_{\mathrm{dep}}^{(j)}\rangle is the mean energy deposition within bin jj. We can then estimate the total energy deposition in 26 according to

Tdep=∑j∑kGφ(j,k)φ(k)⟨Edep(j)⟩.(33)T_{\mathrm{dep}} = \sum_j \sum_k \mathbf{G}_{\varphi}^{(j,k)}\varphi^{(k)}\langle E_{\mathrm{dep}}^{(j)}\rangle. \tag*{(33)}

If we assume that ⟨Edep(j)⟩≈12(Edep(j)+Edep(j+1))\langle E_{\mathrm{dep}}^{(j)}\rangle\approx\frac{1}{2}(E_{\mathrm{dep}}^{(j)}+E_{\mathrm{dep}}^{(j+1)}), i.e. that the average Energy deposted in bin jj is the average energy of the bin, jj, then the total energy deposition is approximated by

Tdep≈∑j∑kGφ(j,k)φ(k)12(Edep(j)+Edep(j+1)).(34)T_{\mathrm{dep}} \approx\sum_j \sum_k \mathbf{G}_{\varphi}^{(j,k)}\varphi^{(k)}\frac{1}{2}(E_{\mathrm{dep}}^{(j)}+E_{\mathrm{dep}}^{(j+1)}) . \tag*{(34)}

The Tally Matrix, nj,kn_{j,k}

We define a tally matrix on a grid of incident and deposited energies according to figure 4-1 where jj is an index into a discrete array of energy deposition bin edges, Edep(j)E_{\mathrm{dep}}^{(j)}; kk is an index into a discrete array of incident particle energy bin edges, Ein(k)E_{\mathrm{in}}^{(k)}; the bin widths ΔE(i)≡E(i+1)−E(i)\Delta E^{(i)} \equiv E^{(i+1)}-E^{(i)}; and φ(Ein(k))\varphi(E_{\mathrm{in}}^{(k)}) is the incident particle characterization within the interval Ein(k)<Ein<Ein(k)+ΔEin(k)E_{\mathrm{in}}^{(k)} < E_{\mathrm{in}} < E_{\mathrm{in}}^{(k)}+\Delta E_{\mathrm{in}}^{(k)}.

The detector response matrix, Gφ(j,k)\mathbf{G}_{\varphi}^{(j,k)}, follows from substitution of the definition of Gφ(j,k)\mathbf{G}_{\varphi}^{(j,k)} given by equation 14 into the discretized form of equation 4

Gφ(j,k)=nj,kφ^(Ein(k))(35)\mathbf{G}_{\varphi}^{(j,k)} = \frac{n_{j,k}}{\hat{\varphi}(E_{\mathrm{in}}^{(k)})} \tag*{(35)}

where

φ^(Ein(k))≡1ΔE(k)∫Ein(k)Ein(k+1)φ(E) dE\hat{\varphi}(E_{\mathrm{in}}^{(k)}) \equiv\frac{1}{\Delta E^{(k)}}\int_{E_{\mathrm{in}}^{(k)}}^{E_{\mathrm{in}}^{(k+1)}}\varphi(E)\,\mathrm{d}E

and nj,kn_{j,k} is a tally matrix defined as

The Tally Matrix, nj,kn_{j,k}

nj,k≡Total number of Energy depositionsn_{j,k} \equiv\text{Total number of Energy depositions}

with Edep(j)≤Edep<Edep(j+1)E_{\mathrm{dep}}^{(j)} \le E_{\mathrm{dep}} < E_{\mathrm{dep}}^{(j+1)} from incident particles with energies Ein(k)≤Ein<Ein(k+1)E_{\mathrm{in}}^{(k)} \le E_{\mathrm{in}} < E_{\mathrm{in}}^{(k+1)}

≡∑i=1Nin(k)[Edep(j)≤Edep(i)<Edep(j+1)]\equiv\sum_{i=1}^{N_{\mathrm{in}}^{(k)}} [E_{\mathrm{dep}}^{(j)} \le E_{\mathrm{dep}}^{(i)} < E_{\mathrm{dep}}^{(j+1)}]

where [P][P] is the Iverson bracket, evaluating to 1 if P is true, and 0 otherwise. Gϕ(j,k)G_{\phi}^{(j,k)} is the number of energy depositions within the range [Edep(j),Edep(j+1)][E_{\mathrm{dep}}^{(j)}, E_{\mathrm{dep}}^{(j+1)}] per unit ϕ\phi of incident particles with energies in the range of [Ein(k),Ein(k+1)][E_{\mathrm{in}}^{(k)}, E_{\mathrm{in}}^{(k+1)}], per unit energy deposition.

Normalizing the Tally Matrix

To finish the calculation of the detector response matrix, one needs to normalize the tally matrix nj,kn_{j,k} to the simulations specific input geometry, and the number of particles thrown, which amounts to calculating the quantities ϕ(Ein(j),Ein(j+1))\phi(E_{\mathrm{in}}^{(j)}, E_{\mathrm{in}}^{(j+1)}).

Except for point sources, the simulated quantification of interest for the incident particle environment ϕ\phi will typically depend on the surface from which the incident particles are thrown at the detector (i.e. when ϕ∼dN/dA\phi\sim dN/dA)

In general, the detector response matrix will look like

Gϕ(j,k)∼nj,kNjA(36)G_{\phi}^{(j,k)} \sim\frac{n_{j,k}}{N_j} A \tag*{(36)}

where AA is an effective input area dependent on the particle input geometry of the simulation. We consider two typical cases: 1.) a incident plane wave, 2.) an isotropic (in solid angle) particle field.

Simulated Plane Wave

If particles are thrown from a planer surface with area A∗A^*, in the direction of θin\theta_{\mathrm{in}} with respect to the planar surface normal, then the effective input area is

Ain=A∗cos⁡θin(37)A_{\mathrm{in}} = A^* \cos\theta_{\mathrm{in}} \tag*{(37)}

If the number of particles thrown with energies in the range of Ej<Ein<Ej+1E_j < E_{\mathrm{in}} < E_{j+1} is NjN_j, the simulated fluence in this energy range is

Φ(Ein(j),Ein(j+1))=NjAin(38)\Phi(E_{\mathrm{in}}^{(j)}, E_{\mathrm{in}}^{(j+1)}) = \frac{N_j}{A_{\mathrm{in}}} \tag*{(38)}

Then for a planewave, the detector response matrix GΦ(j,k)G_{\Phi}^{(j,k)} is

GΦ(j,k)=nj,kNjAin(39)G_{\Phi}^{(j,k)} = \frac{n_{j,k}}{N_j} A_{\mathrm{in}} \tag*{(39)}

Simulated Isotropic Flux

If the detector response to a directionally isotropic particle field is desired, then one typically throws particles from the surface of a sphere of radius RR surrounding the detector according to a cosine-law distribution in the solid angle around the input surface normal and constructs the tally matrix nj,kn_{j,k} as described previously.

First, we assume that the particle intensity separates into independent energy and angular components given by

ΦE,Ω(E,Ω)=ΦE(E)F(Ω)(40)\Phi_{E,\Omega}(E,\Omega) = \Phi_E(E)F(\Omega) \tag*{(40)}

A directionally isotropic source implies that F(Ω)=1/4πF(\Omega) = 1/4\pi.

To determine the proper normalization, we calculate the number of incident particles, NN, input on a sphere of radius RR, for a particle intensity ΦE,Ω(E,Ω)=ΦE(E)F(Ω)\Phi_{E,\Omega}(E,\Omega) = \Phi_E(E)F(\Omega) with F(Ω)=1/4πF(\Omega) = 1/4\pi. At the surface the particle intensity in the direction of θ\theta relative to the surface normal is reduced by the factor dσ⃗⋅r^=cos⁡θdσ\mathrm{d}\vec{\sigma} \cdot\hat{r} = \cos\theta\mathrm{d}\sigma. Then

N=∫S∫Ω′∫E=0∞ΦE(E)F(Ω)dEcos⁡θdΩdσ(41)N = \int_S \int_{\Omega'} \int_{E=0}^{\infty} \Phi_E(E)F(\Omega)\mathrm{d}E \cos\theta\mathrm{d}\Omega\mathrm{d}\sigma \tag*{(41)}
=[∫Sdσ][∫Ω′cos⁡θdΩ][14π∫E=0∞ΦE(E)dEin](42)= \left[\int_S \mathrm{d}\sigma\right]\left[\int_{\Omega'} \cos\theta\mathrm{d}\Omega\right]\left[\frac{1}{4\pi}\int_{E=0}^{\infty}\Phi_E(E)\mathrm{d}E_{\mathrm{in}}\right] \tag*{(42)}
=[4πR2][π][Φ4π](43)= \left[4\pi R^2\right][\pi]\left[\frac{\Phi}{4\pi}\right] \tag*{(43)}
=4π2R2j0(44)= 4\pi^2 R^2 j_0 \tag*{(44)}
=πR2Φ(45)= \pi R^2\Phi \tag*{(45)}

where j0≡Φ/4πj_0 \equiv\Phi/4\pi is sometimes referred to as the “directional flux", with Φ\Phi referred to as the “omni-directional flux” [4].

The omnidirectional flux for particles thrown with energies in the range of Ej<Ein<Ej+1E_j < E_{\mathrm{in}} < E_{j+1} is given by

Φ(Ein(j),Ein(j+1))=NjπR2(46)\Phi\left(E_{\mathrm{in}}^{(j)},E_{\mathrm{in}}^{(j+1)}\right) = \frac{N_j}{\pi R^2} \tag*{(46)}

and the detector response matrix for an isotropic or omni-directional flux is given by

GΦ(j,k)=nj,kNjπR2(47)G_{\Phi}^{(j,k)} = \frac{n_{j,k}}{N_j}\pi R^2 \tag*{(47)}

We emphasize that equation (47) as defined is the detector response matrix that relates the omnidirectional flux, Φ\Phi, (counts/cm2^2/s) to instrument count rate. If one wanted to relate the instrument count rate to a directional flux j0j_0, the geometric factor would be given by equation (47) times an additional factor of 4π4\pi, or Gj0(j,k)=4πGΦ(j,k)G_{j_0}^{(j,k)} = 4\pi G_{\Phi}^{(j,k)}.

Total Average Effective Area

The Total effective area, GG, is the average effective area [cm2^2] for any energy deposition (Ek>0E_k > 0) over all incident particle energies (Ej>0E_j > 0), and is given by

Total Average Effective Area

⟨Gϕ⟩=1EK−Ek=0∑j,kΔEkGϕ(j,k)(48)\langle\mathbf{G}_{\phi} \rangle= \frac{1}{E_{K} - E_{k=0}} \sum_{j,k} \Delta E_{k} \mathbf{G}_{\phi}^{(j,k)} \tag*{(48)}

where KK is the total number of incident energy bins, and, EK−Ek=0E_{K} - E_{k=0} is the span of incident energies. It is also useful to define the total effective area for energy depositions above some threshold energy as G(Edep>Ethres)\mathbf{G}(E_{\mathrm{dep}} > E_{\mathrm{thres}}), where kk in equation (4) satisfies Ek>EthresE_{k} > E_{\mathrm{thres}}.

References

References

  1. [1]George B. Rybicki and Alan P. Lightman. Radiative processes in astrophysics. 1979.
  2. [2]J.D. Sullivan. “Geometrical factor and directional response of single and multi-element particle telescopes”. In: Nuc. Inst. and Meth. 95 (1971), pp. 5–11.DOI
  3. [3]David J. Thomas. “ICRU report 85: fundamental quantities and units for ionizing radiation”. In: Radiation Protection Dosimetry 150.4 (May 2012), pp. 550–552. ISSN: 0144-8420. DOI: 10.1093/rpd/ncs077. eprint: https://academic.oup.com/rpd/article-pdf/150/4/550/4558428/ncs077.pdf. URL: https://doi.org/10.1093/rpd/ncs077.
  4. [4]Karl Yando et al. “A Monte Carlo simulation of the NOAA POES medium energy proton and electron detector instrument”. In: Journal of Geophysical Research: Space Physics 116.A10 (2011).DOI

Paper details

vlnc:W_br4dhxdb
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents