1 | |
---|
2 | |
---|
3 | \section[Multiple Scattering]{Multiple Scattering} |
---|
4 | |
---|
5 | Geant4 uses a new multiple scattering (MSC) model to simulate the |
---|
6 | multiple scattering of charged particles in matter. This model does not use |
---|
7 | the Moliere formalism \cite{msc.moliere}, but is based on the more complete |
---|
8 | Lewis theory \cite{msc.lewis}. The model simulates the scattering of the |
---|
9 | particle after a given step, and also computes the path length |
---|
10 | correction and the lateral displacement. |
---|
11 | |
---|
12 | \subsection{Introduction} |
---|
13 | |
---|
14 | MSC simulation algorithms can be classified as either {\em detailed} or |
---|
15 | {\em condensed}. In the detailed algorithms, all the collisions/interactions |
---|
16 | experienced by the particle are simulated. This simulation can be considered |
---|
17 | as exact; it gives the same results as the solution of the transport equation. |
---|
18 | However, it can be used only if the number of collisions is not too large, a |
---|
19 | condition fulfilled only for special geometries (such as thin foils), or low |
---|
20 | enough kinetic energies. For larger kinetic energies the average number of |
---|
21 | collisions is very large and the detailed simulation becomes very inefficient. |
---|
22 | High energy simulation codes use condensed simulation algorithms, in which |
---|
23 | the global effects of the collisions are simulated at the end of a track |
---|
24 | segment. The global effects generally computed in these codes are the net |
---|
25 | displacement, energy loss, and change of direction of the charged particle. |
---|
26 | These quantities are computed from the multiple scattering theories used in |
---|
27 | the codes. The accuracy of the condensed simulations is limited by the |
---|
28 | approximations of the multiple scattering theories. \\ |
---|
29 | |
---|
30 | \noindent |
---|
31 | Most particle physics simulation codes use the multiple scattering theories |
---|
32 | of Moli\`ere \cite{msc.moliere}, Goudsmit and Saunderson \cite{msc.goudsmit} |
---|
33 | and Lewis \cite{msc.lewis}. The theories of Moli\`ere and Goudsmit-Saunderson |
---|
34 | give only the angular distribution after a step, while the Lewis theory |
---|
35 | computes the moments of the spatial distribution as well. None of these |
---|
36 | MSC theories gives the probability distribution of the spatial displacement. |
---|
37 | Therefore each of the MSC simulation codes incorporates its own algorithm to |
---|
38 | determine the spatial displacement of the charged particle after a given step. |
---|
39 | These algorithms are not exact, of course, and are responsible for most of the |
---|
40 | uncertainties in the MSC codes. Therefore the simulation results can depend |
---|
41 | on the value of the step length and generally one has to select the value of |
---|
42 | the step length carefully. \\ |
---|
43 | |
---|
44 | \noindent |
---|
45 | A new class of MSC simulation, the {\em mixed} simulation algorithms (see |
---|
46 | e.g.\cite{msc.fernandez}), appeared in the literature recently. The mixed |
---|
47 | algorithm simulates the {\em hard} collisions one by one and uses a MSC theory to |
---|
48 | treat the effects of the {\em soft} collisions at the end of a given step. Such |
---|
49 | algorithms can prevent the number of steps from becoming too large and also |
---|
50 | reduce the dependence on the step length. \\ |
---|
51 | |
---|
52 | \noindent |
---|
53 | The MSC model used in Geant4 belongs to the class of condensed |
---|
54 | simulations. The model is based on Lewis' MSC theory and uses model functions |
---|
55 | to determine the angular and spatial distributions after a step. The |
---|
56 | functions have been chosen in such a way as to give the same moments of the |
---|
57 | (angular and spatial) distributions as the Lewis theory. |
---|
58 | |
---|
59 | \subsection{Definition of Terms} |
---|
60 | |
---|
61 | In simulation, a particle is transported by steps through the detector |
---|
62 | geometry. The shortest distance between the endpoints of a step is called |
---|
63 | the {\em geometrical path length}, $z$. |
---|
64 | In the absence of a magnetic field, this is a straight line. |
---|
65 | For non-zero fields, $z$ is the shortest distance along |
---|
66 | a curved trajectory. Constraints on $z$ are imposed when particle tracks |
---|
67 | cross volume boundaries. The path length of an actual particle, however, is |
---|
68 | usually longer than the geometrical path length, due to physical interactions |
---|
69 | like multiple scattering. |
---|
70 | This distance is called the {\em true path length}, $t$. |
---|
71 | Constraints on $t$ are imposed by the physical processes acting on the |
---|
72 | particle. \\ |
---|
73 | |
---|
74 | \noindent |
---|
75 | The properties of the multiple scattering process are completely determined |
---|
76 | by the {\em transport mean free paths}, $\lambda_k$, which are functions of the |
---|
77 | energy in a given material. The $k$-th transport mean free path is defined as |
---|
78 | |
---|
79 | \begin{equation} |
---|
80 | \frac {1}{\lambda_k} = 2 \pi n_a \int_{-1}^{1} \left[1 - P_k(cos\chi) \right] |
---|
81 | \frac{d\sigma(\chi)}{d\Omega} d(cos\chi) |
---|
82 | \label{msc.a1} |
---|
83 | \end{equation} |
---|
84 | |
---|
85 | \noindent |
---|
86 | where $d\sigma(\chi)/d\Omega$ is the differential cross section of the |
---|
87 | scattering, $P_k(cos\chi)$ is the $k$-th Legendre polynomial, and $n_a$ is the |
---|
88 | number of atoms per volume. \\ |
---|
89 | |
---|
90 | \noindent |
---|
91 | Most of the mean properties of MSC computed in the simulation codes depend |
---|
92 | only on the first and second transport mean free paths. The mean value of |
---|
93 | the geometrical path length (first moment) corresponding to a given true path |
---|
94 | length $t$ is given by |
---|
95 | \begin{equation} |
---|
96 | \langle z \rangle = |
---|
97 | \lambda_1 \left[ 1-\exp \left(-\frac{t}{\lambda_1} \right)\right] |
---|
98 | \label{msc.a} |
---|
99 | \end{equation} |
---|
100 | |
---|
101 | \noindent |
---|
102 | Eq.~\ref{msc.a} is an exact result for the mean value of $z$ if the |
---|
103 | differential cross section has axial symmetry and the energy loss can be |
---|
104 | neglected. The transformation between true and geometrical path lengths is |
---|
105 | called the {\em path length correction}. This formula and other expressions for |
---|
106 | the first moments of the spatial distribution were taken from either |
---|
107 | \cite{msc.fernandez} or \cite{msc.kawrakow}, but were originally |
---|
108 | calculated by Goudsmit and Saunderson \cite{msc.goudsmit} and Lewis |
---|
109 | \cite{msc.lewis}. \\ |
---|
110 | |
---|
111 | \noindent |
---|
112 | At the end of the true step length, $t$, the scattering angle is $\theta$. |
---|
113 | The mean value of $cos\theta$ is |
---|
114 | \begin{equation} |
---|
115 | \langle cos\theta \rangle = \exp \left[-\frac{t}{\lambda_1} \right] |
---|
116 | \label{msc.c} |
---|
117 | \end{equation} |
---|
118 | The variance of $cos\theta$ can be written as |
---|
119 | \begin{equation} |
---|
120 | \sigma^2 = \langle cos^2\theta \rangle - \langle cos\theta \rangle ^2 = |
---|
121 | \frac{1 + 2 e^{- 2 \kappa \tau}} {3} - e^{-2 \tau} |
---|
122 | \label{msc.c1} |
---|
123 | \end{equation} |
---|
124 | |
---|
125 | \noindent |
---|
126 | where $\tau = t/\lambda_1$ and $\kappa = \lambda_1/\lambda_2$. \\ |
---|
127 | |
---|
128 | \noindent |
---|
129 | The mean lateral displacement is given by a more complicated formula |
---|
130 | \cite{msc.fernandez}, but this quantity can also be calculated relatively |
---|
131 | easily and accurately. The square of the {\em mean lateral displacement} is |
---|
132 | \begin{equation} |
---|
133 | \langle x^2 + y^2 \rangle = \frac{4 \lambda_1^2}{3} \ \left[\tau |
---|
134 | - \frac{\kappa+1}{\kappa} |
---|
135 | + \frac{\kappa}{\kappa-1} e^{-\tau} - |
---|
136 | \frac{1}{\kappa (\kappa -1)} e^{-\kappa \tau} \right] |
---|
137 | \label{msc.e1} |
---|
138 | \end{equation} |
---|
139 | |
---|
140 | \noindent |
---|
141 | Here it is assumed that the initial particle direction is parallel to the |
---|
142 | the $z$ axis.\\ |
---|
143 | |
---|
144 | \noindent |
---|
145 | The lateral correlation is determined by the equation |
---|
146 | \begin{equation} |
---|
147 | \langle x v_x+y v_y \rangle = \frac{2 \lambda_1}{3} \ \left[1 |
---|
148 | - \frac{\kappa}{\kappa-1} e^{-\tau} + |
---|
149 | \frac{1}{\kappa-1} e^{-\kappa \tau} \right] |
---|
150 | \label{msc.e2} |
---|
151 | \end{equation} |
---|
152 | |
---|
153 | \noindent |
---|
154 | where $v_x$ and $v_y$ are the x and y components of the direction unit |
---|
155 | vector. |
---|
156 | This equation gives the correlation strength between the final lateral |
---|
157 | position and final direction.\\ |
---|
158 | |
---|
159 | \noindent |
---|
160 | The transport mean free path values have been calculated by Liljequist et al. |
---|
161 | \cite{msc.liljequist1}, \cite{msc.liljequist2} for electrons and positrons in |
---|
162 | the kinetic energy range \mbox{100 eV - 20 MeV} in 15 materials. The MSC |
---|
163 | model in Geant4 uses these values for kinetic energies below 10 MeV. |
---|
164 | For high energy particles (above 10 MeV) the transport mean free path |
---|
165 | values have been taken from a paper of R.Mayol and F.Salvat (\cite{msc.mayol}). |
---|
166 | When necessary, the model linearly interpolates |
---|
167 | or extrapolates the transport cross section, $\sigma_1 = 1 / \lambda_1$, in |
---|
168 | atomic number $Z$ and in the square of the particle velocity, $\beta^2$. |
---|
169 | The ratio $\kappa$ is a very slowly varying function of the energy: |
---|
170 | $\kappa > 2$ for $T >$ a few keV, and $\kappa \rightarrow 3$ for very high |
---|
171 | energies (see \cite{msc.kawrakow}). Hence, a constant value of 2.5 is used |
---|
172 | in the model.\\ |
---|
173 | |
---|
174 | \noindent |
---|
175 | Nuclear size effects are negligible for low energy particles and |
---|
176 | they are accounted for in the Born approximation in \cite{msc.mayol}, |
---|
177 | so there is no need for extra corrections of this kind in the model. |
---|
178 | |
---|
179 | \subsection{Path Length Correction} |
---|
180 | |
---|
181 | As mentioned above, the path length correction refers to the transformation |
---|
182 | true path length $\longrightarrow$ geometrical path length and its inverse. |
---|
183 | The true path length $\longrightarrow$ geometrical path length transformation |
---|
184 | is given by eq. \ref{msc.a} if the step is small and the energy loss can be |
---|
185 | neglected. If the step is not small the energy dependence makes the |
---|
186 | transformation more complicated. For this case Eqs. \ref{msc.c},\ref{msc.a} |
---|
187 | should be modified as |
---|
188 | \begin{equation} |
---|
189 | \langle cos\theta \rangle = \exp \left[-\int_0^t \frac{du}{\lambda_1 (u)} |
---|
190 | \right] |
---|
191 | \label{msc.ax} |
---|
192 | \end{equation} |
---|
193 | |
---|
194 | \begin{equation} |
---|
195 | \langle z \rangle = \int_0^t \langle cos\theta \rangle_u \ du |
---|
196 | \label{msc.bx} |
---|
197 | \end{equation} |
---|
198 | |
---|
199 | \noindent |
---|
200 | where $\theta$ is the scattering angle, $t$ and $z$ are the true and |
---|
201 | geometrical path lengths, and $\lambda_1$ is the transport mean free path. \\ |
---|
202 | |
---|
203 | \noindent |
---|
204 | In order to compute Eqs. \ref{msc.ax},\ref{msc.bx} the $t$ dependence of the |
---|
205 | transport mean free path must be known. $\lambda_1$ depends on the kinetic |
---|
206 | energy of the particle which decreases along the step. All computations in |
---|
207 | the model use a linear approximation for this $t$ dependence: |
---|
208 | \begin{equation} |
---|
209 | \lambda_1(t) = \lambda_{10} ( 1- \alpha t) \label{msc.cx} |
---|
210 | \end{equation} |
---|
211 | |
---|
212 | \noindent |
---|
213 | Here $\lambda_{10}$ denotes the value of $\lambda_1$ at the start of the step, |
---|
214 | and $\alpha $ is a constant. It is worth noting that Eq. \ref{msc.cx} is |
---|
215 | \emph{not} |
---|
216 | a crude approximation. It is rather good at low ($ < $ 1 MeV) energy. At |
---|
217 | higher energies the step is generally much smaller than the range of the |
---|
218 | particle, so the change in energy is small and so is the change in $\lambda_1$. |
---|
219 | Using Eqs. \ref{msc.ax} - \ref{msc.cx} the explicit formula for |
---|
220 | $\langle cos\theta \rangle$ and $\langle z \rangle$ are : |
---|
221 | \begin{equation} |
---|
222 | \langle cos\theta \rangle = |
---|
223 | \left(1 - \alpha t \right)^{\frac{1}{ \alpha \lambda_{10}}} |
---|
224 | \label{msc.ff} |
---|
225 | \end{equation} |
---|
226 | |
---|
227 | \begin{equation} |
---|
228 | \langle z \rangle = \frac{1} { \alpha (1 + \frac{1}{\alpha \lambda_{10}})} |
---|
229 | \left[ 1 - (1 - \alpha t)^{1+ \frac{1}{ \alpha \lambda_{10}}} \right] |
---|
230 | \label{msc.dx} |
---|
231 | \end{equation} |
---|
232 | B |
---|
233 | |
---|
234 | \noindent |
---|
235 | The value of the constant $\alpha $ can be expressed using $\lambda_{10}$ and |
---|
236 | $\lambda_{11}$ where $\lambda_{11}$ is the value of the transport mean free |
---|
237 | path at the end of the step |
---|
238 | \begin{equation} |
---|
239 | \alpha = \frac{\lambda_{10} - \lambda_{11}} {t \lambda_{10}} |
---|
240 | \label{msc.ex} |
---|
241 | \end{equation} |
---|
242 | |
---|
243 | \noindent |
---|
244 | At low energies ( $T_{kin} < M$ , M - particle mass) $\alpha $ has a simpler |
---|
245 | form: |
---|
246 | \begin{equation} |
---|
247 | \alpha = \frac{1} { r_0} \label{msc.fx} |
---|
248 | \end{equation} |
---|
249 | where $r_0$ denotes the range of the particle at the start of the step. |
---|
250 | |
---|
251 | \noindent |
---|
252 | It can easily be seen that for a small step (i.e. for a step with small |
---|
253 | relative energy loss) the formula of $\langle z \rangle$ is |
---|
254 | \begin{equation} |
---|
255 | \langle z \rangle = |
---|
256 | \lambda_{10} \left[ 1-\exp{\left( -\frac{t}{\lambda_{10}}\right)}\right] |
---|
257 | \label{msc.gx} |
---|
258 | \end{equation} |
---|
259 | |
---|
260 | \noindent |
---|
261 | Eq. \ref{msc.dx} or \ref{msc.gx} gives the mean value of the geometrical step |
---|
262 | length for a given true step length. \\ |
---|
263 | |
---|
264 | \noindent |
---|
265 | The actual geometrical path length is sampled in the |
---|
266 | model according to the simple probability density function defined for |
---|
267 | $v = z/t \in [0 , 1]$ : |
---|
268 | \begin{equation} |
---|
269 | f(v) = (k+1)(k+2) v^k (1-v) |
---|
270 | \label{msc.d2} |
---|
271 | \end{equation} |
---|
272 | |
---|
273 | \noindent |
---|
274 | The value of the exponent $k$ is computed from the requirement that $f(v)$ |
---|
275 | must give the same mean value for $z = v t$ as eq. \ref{msc.dx} or |
---|
276 | \ref{msc.gx}. Hence |
---|
277 | \begin{equation} |
---|
278 | k = \frac{3 \langle z \rangle - t}{t - \langle z \rangle} |
---|
279 | \label{msc.d3} |
---|
280 | \end{equation} |
---|
281 | |
---|
282 | \noindent |
---|
283 | The value of $z = v t$ is sampled using $f(v)$ if $k > 0$, otherwise |
---|
284 | $z = \langle z \rangle$ is used. \\ |
---|
285 | |
---|
286 | \noindent |
---|
287 | The geometrical path length $\longrightarrow$ true path length transformation |
---|
288 | is performed using the mean values. The transformation can be written as |
---|
289 | \begin{equation} |
---|
290 | t(z) = \langle t \rangle = -\lambda_1 \log\left(1-\frac{z}{\lambda_1}\right) |
---|
291 | \label{msc.d4} |
---|
292 | \end{equation} |
---|
293 | if the geometrical step is small and |
---|
294 | \begin{equation} |
---|
295 | t(z) = \frac{1}{\alpha} \left[ 1 - (1 - \alpha w z)^{\frac{1}{w}} \right] |
---|
296 | \label{msc.hx} |
---|
297 | \end{equation} |
---|
298 | where $$w = 1 + \frac{1}{\alpha \lambda_{10}}$$ if the step is not small, i.e. |
---|
299 | the energy loss should be taken into account. \\ |
---|
300 | |
---|
301 | \noindent |
---|
302 | This transformation is needed when the particle arrives at a volume boundary, |
---|
303 | causing the step to be geometry-limited. In this case the true path length |
---|
304 | should be computed in order to have the correct energy loss of the particle |
---|
305 | after the step. |
---|
306 | |
---|
307 | \subsection{Angular Distribution} |
---|
308 | |
---|
309 | The quantity $u = cos\theta$ is sampled according to a model function |
---|
310 | $g(u)$. The shape of this function has been chosen such that |
---|
311 | Eqs. \ref{msc.c} and \ref{msc.c1} are satisfied. The functional form of $g$ is |
---|
312 | \begin{equation} |
---|
313 | g(u) = p [q g_1(u) + (1-q) g_3(u)] + (1-p) g_2(u) |
---|
314 | \label{msc.d} |
---|
315 | \end{equation} |
---|
316 | |
---|
317 | \noindent |
---|
318 | where $ 0 \leq p,q \leq 1 $, and the $g_i$ are simple functions of |
---|
319 | $u = cos\theta$, normalized over the range $ u \in [-1,\ 1] $. The functions |
---|
320 | $g_i$ have been chosen as |
---|
321 | \begin{equation} |
---|
322 | g_1(u) = C_{1}\hspace{3mm} e^{-a (1-u)} \hspace{2cm} -1 \leq u_0 \leq u \leq 1 |
---|
323 | \label{msc.d5} |
---|
324 | \end{equation} |
---|
325 | \begin{equation} |
---|
326 | g_2(u) = C_{2}\hspace{3mm} \frac{1} { (b-u)^d} |
---|
327 | \hspace{2cm} -1 \leq u \leq u_0 \leq 1 |
---|
328 | \label{msc.d6} |
---|
329 | \end{equation} |
---|
330 | \begin{equation} |
---|
331 | g_3(u) = C_{3} \hspace{4.8cm} -1 \leq u \leq 1 |
---|
332 | \label{msc.d7} |
---|
333 | \end{equation} |
---|
334 | where $a > 0$, $b > 0$, $d > 0$ and $u_0$ are model parameters, and the $C_{i}$ |
---|
335 | are normalization constants. It is worth noting that for small scattering |
---|
336 | angles, $\theta$, $g_1(u)$ is nearly Gaussian ($exp(-\theta^2/2 \theta_0^2)$) |
---|
337 | if $\theta_0^2 \approx 1 / a$, while $g_2(u)$ has a Rutherford-like tail for |
---|
338 | large $\theta$, if $b \approx 1$ and $d$ is not far from 2 . |
---|
339 | |
---|
340 | \subsection{Determination of the Model Parameters} |
---|
341 | |
---|
342 | The parameters $a$, $b$, $d$, $u_0$ and $p$, $q$ are not independent. The |
---|
343 | requirement that the angular distribution function $g(u)$ and its first |
---|
344 | derivative be continuous at $u = u_0$ imposes two constraints on the |
---|
345 | parameters: |
---|
346 | \begin{equation} |
---|
347 | p\hspace{1mm} g_1(u_0) = (1-p)\hspace{1mm} g_2(u_0) |
---|
348 | \label{msc.p1} |
---|
349 | \end{equation} |
---|
350 | \begin{equation} |
---|
351 | p\hspace{1mm} a\hspace{1mm} g_1(u_0) = (1-p)\hspace{1mm} |
---|
352 | \frac{d}{b-u_0}\hspace{1mm} g_2(u_0) |
---|
353 | \label{msc.p2} |
---|
354 | \end{equation} |
---|
355 | |
---|
356 | \noindent |
---|
357 | A third constraint comes from Eq. \ref{msc.ax} : $g(u)$ must give the same |
---|
358 | mean value for $u$ as the theory. \\ |
---|
359 | |
---|
360 | \noindent |
---|
361 | It follows from Eqs. \ref{msc.ff} and \ref{msc.d} that |
---|
362 | \begin{equation} |
---|
363 | q \{ p \langle u \rangle_1 + (1-p) \langle u \rangle_2 \} = |
---|
364 | [ 1 - \alpha \ t ]\ ^{\frac{1}{\alpha \lambda_{10}}} |
---|
365 | \label{msc.par1} |
---|
366 | \end{equation} |
---|
367 | where $\langle u \rangle_i$ denotes the mean value of $u$ computed from the |
---|
368 | distribution $g_i(u)$. \\ |
---|
369 | |
---|
370 | \noindent |
---|
371 | The parameter $a$ was chosen according to a modified Highland-Lynch-Dahl formula |
---|
372 | for the width of the angular distribution \cite{msc.highland}, |
---|
373 | \cite{msc.lynch}. |
---|
374 | \begin{equation} |
---|
375 | a = \frac {0.5} {1 -cos(\theta_0)} |
---|
376 | \end{equation} |
---|
377 | where $\theta_0$ is |
---|
378 | \begin{equation} |
---|
379 | \theta_0 = \frac {13.6 MeV}{ \beta c p} z_{ch} \sqrt{\frac{t}{X_0}} |
---|
380 | \ \left[ 1 + 0.038 \ln \left(\frac{t}{X_0} \right)\ \right] |
---|
381 | \end{equation} |
---|
382 | |
---|
383 | \noindent |
---|
384 | when the original Highland-Lynch-Dahl formula is used. |
---|
385 | Here $\theta_0 = \theta^{rms}_{plane}$ is the width of the approximate |
---|
386 | Gaussian projected angle distribution, $p$, $\beta c$ and $z_{ch}$ are the |
---|
387 | momentum, |
---|
388 | velocity and charge number of the incident particle, and $t/X_0$ is the |
---|
389 | true path length in radiation length unit. This value of |
---|
390 | $\theta_0$ is from a fit to the Moli\`ere distribution for singly charged |
---|
391 | particles with $\beta = 1$ for all Z, and is accurate to 11 $\%$ or better |
---|
392 | for $ 10^{-3} \leq t/X_0 \leq 100$ (see e.g. Rev. of Particle Properties, |
---|
393 | section 23.3). \\ |
---|
394 | |
---|
395 | \noindent |
---|
396 | The modified formula for $\theta_0$ is |
---|
397 | \begin{equation} |
---|
398 | \theta_0 = \frac {13.6 MeV}{\beta c p} z_{ch} \sqrt{\frac{t}{X_0} } |
---|
399 | \left[ 1 + 0.105 \ln \left(\frac{t}{X_0}\right) |
---|
400 | + 0.0035 \left(\ln \left(\frac{t}{X_0}\right)\right)^2 |
---|
401 | \right] ^{\frac{1}{2}} f \left(Z \right) |
---|
402 | \end{equation} |
---|
403 | |
---|
404 | \noindent |
---|
405 | where |
---|
406 | \begin{equation} |
---|
407 | f \left(Z \right) = 1 - \frac{0.24}{ Z \left(Z + 1 \right) } |
---|
408 | \end{equation} |
---|
409 | \noindent |
---|
410 | is an empirical correction factor based on high energy proton scattering |
---|
411 | data \cite{msc.shen}. |
---|
412 | This formula gives a much smaller step dependence in the angular |
---|
413 | distribution and describes the available |
---|
414 | electron scattering data better than the Highland form. \\ |
---|
415 | |
---|
416 | \noindent |
---|
417 | The value of the parameter $u_0$ has been chosen as |
---|
418 | \begin{equation} |
---|
419 | u_0 \hspace{4mm} = \hspace{4mm} 1 - \frac{\xi}{a} |
---|
420 | \end{equation} |
---|
421 | where $\xi$ is a constant ($\xi = 3$). \\ |
---|
422 | |
---|
423 | \noindent |
---|
424 | The parameter $d$ is set to |
---|
425 | \begin{equation} |
---|
426 | d \hspace{4mm} = \hspace{4mm} 2.40 - 0.027 \ Z^ \frac{2}{3} |
---|
427 | \end{equation} |
---|
428 | This (empirical) expression is obtained comparing the simulation |
---|
429 | results to the data of the MuScat experiment \cite{msc.attwood}. \\ |
---|
430 | |
---|
431 | \noindent |
---|
432 | The remaining three parameters can be computed from |
---|
433 | Eqs. \ref{msc.p1} - \ref{msc.par1}. |
---|
434 | The numerical value of the parameters can be found in the |
---|
435 | code. \\ |
---|
436 | |
---|
437 | \noindent |
---|
438 | It should be noted that in this model there is no step limitation |
---|
439 | originating from the multiple scattering process. Another important feature |
---|
440 | of this model is that the sum of the 'true' step lengths of the particle, that |
---|
441 | is, the total true path length, does not depend on the length of the steps. |
---|
442 | Most algorithms used in simulations do not have these properties. \\ |
---|
443 | |
---|
444 | \noindent |
---|
445 | In the case of heavy charged particles ($\mu$, $\pi$, $p$, etc.) the mean |
---|
446 | transport free path is calculated from the electron or positron $\lambda_1$ |
---|
447 | values with a 'scaling' applied. This is possible because the transport |
---|
448 | mean free path $\lambda_1$ depends only on the variable $P \beta c$, where |
---|
449 | $P$ is the momentum, and $\beta c$ is the velocity of the particle. \\ |
---|
450 | |
---|
451 | \noindent |
---|
452 | In its present form the model samples the path length correction and angular |
---|
453 | distribution from model functions, while for the lateral displacement and |
---|
454 | the lateral correlation only |
---|
455 | the mean values are used and all the other correlations are neglected. |
---|
456 | However, the model |
---|
457 | is general enough to incorporate other random quantities and correlations in |
---|
458 | the future. |
---|
459 | |
---|
460 | \subsection{The MSC Process in Geant4} |
---|
461 | |
---|
462 | The step length of the particles is determined by the physics processes or |
---|
463 | the geometry of the detectors. The tracking/stepping algorithm checks all the |
---|
464 | step lengths demanded by the (continuous or discrete) physics processes and |
---|
465 | determines the minimum of these step lengths. \\ |
---|
466 | |
---|
467 | \noindent |
---|
468 | Then, this minimum step length |
---|
469 | must be compared with the length determined by the geometry of the detectors |
---|
470 | and one has to select the minimum of the 'physics step length' and the |
---|
471 | 'geometrical step length' as the actual step length. \\ |
---|
472 | |
---|
473 | \noindent |
---|
474 | This is the point where the MSC model comes into the game. All the |
---|
475 | physics processes use the true path length $t$ to sample the interaction point, |
---|
476 | while the step limitation originated from the geometry is a |
---|
477 | geometrical path length $z$. The MSC algorithm transforms the 'physics step |
---|
478 | length' into a 'geometrical step length' before the comparison of the two |
---|
479 | lengths. This 't'\(\rightarrow\)'z' transformation can be called the |
---|
480 | inverse of the path length correction. \\ |
---|
481 | |
---|
482 | \noindent |
---|
483 | After the actual step length has been determined and the particle relocation |
---|
484 | has been performed the MSC performs the transformation 'z'\(\rightarrow\)'t', |
---|
485 | because the energy loss and scattering computation need the true step length |
---|
486 | 't'. \\ |
---|
487 | |
---|
488 | \noindent |
---|
489 | The scattering angle $\theta$ of the particle after the step of length 't' is |
---|
490 | sampled according to the model function given in eq. \ref{msc.d} . |
---|
491 | The azimuthal angle $\phi$ is generated uniformly in the range $[0, 2 \pi]$. \\ |
---|
492 | |
---|
493 | \noindent |
---|
494 | After the simulation of the scattering angle, the lateral displacement is |
---|
495 | computed using eq. \ref{msc.e1}. Then the correlation given by eq. \ref{msc.e2} |
---|
496 | is used to determine the direction of the lateral displacement. |
---|
497 | Before 'moving' the particle according to the displacement a check is performed |
---|
498 | to ensure that the relocation of the particle with the lateral displacement |
---|
499 | does not take the particle beyond the volume boundary. |
---|
500 | |
---|
501 | \subsection{Step Limitation Algorithm} |
---|
502 | |
---|
503 | In Geant4 the boundary crossing is treated by the transportation process. |
---|
504 | The transportation ensures that the |
---|
505 | particle does not penetrate in a new volume without stopping at the boundary, |
---|
506 | it restricts the step size when the particle leaves a volume. However, |
---|
507 | this step restriction can be rather weak in big volumes and this fact |
---|
508 | can result a not very good angular distribution after the volume. |
---|
509 | At the same time, there is no similar |
---|
510 | step limitation when a particle enters a volume and this fact does not allow |
---|
511 | a good backscattering simulation for low energy particles. Low energy particles |
---|
512 | penetrate too deeply into the volume in the first step and then - because |
---|
513 | of energy loss - they are not able to reach again the boundary in backward |
---|
514 | direction.\\ |
---|
515 | |
---|
516 | \noindent |
---|
517 | A very simple step limitation algorithm has been implemented in the |
---|
518 | MSC code to cure this situation. At the start of a track or after |
---|
519 | entering in a new volume, the algorithm restricts |
---|
520 | the step size to a value |
---|
521 | \begin{equation} |
---|
522 | f_r \cdot max\{r,\lambda_1\} |
---|
523 | \end{equation} |
---|
524 | where $r$ is the range of the particle, $f_r$ is a constant ($f_r \in [0, 1]$); |
---|
525 | taking the max of $r$ and $\lambda_1$ is an empirical choice. In order not to |
---|
526 | use very small - unphysical - step sizes a lower limit is given for the step |
---|
527 | size as |
---|
528 | \begin{equation} |
---|
529 | tlimitmin = max\left[ \frac{\lambda_1}{nstepmax}, \lambda_{elastic} \right] |
---|
530 | \end{equation} |
---|
531 | with $nstepmax = 25$ and $\lambda_{elastic}$ is the elastic mean free path |
---|
532 | of the particle (see later).\\ |
---|
533 | |
---|
534 | \noindent |
---|
535 | It can be easily seen that this kind of step limitation poses a real constraint |
---|
536 | only for low energy particles. \\ |
---|
537 | |
---|
538 | \noindent |
---|
539 | In order to prevent a particle from crossing |
---|
540 | a volume in just one step, an additional limitation is imposed: |
---|
541 | after entering a volume |
---|
542 | the step size cannot be bigger than |
---|
543 | \begin{equation} |
---|
544 | \frac {d_{geom}}{f_g} |
---|
545 | \end{equation} |
---|
546 | where $d_{geom}$ is the distance to the next boundary (in the direction |
---|
547 | of the particle) and $f_g$ is a constant parameter. A similar restriction |
---|
548 | at the start of a track is |
---|
549 | \begin{equation} |
---|
550 | \frac {2 d_{geom}}{f_g} |
---|
551 | \end{equation} |
---|
552 | |
---|
553 | \noindent |
---|
554 | The choice of the parameters $f_r$ and $f_g$ is also |
---|
555 | related to performance. By default $f_r = 0.02$ and $f_g = 2.5$ |
---|
556 | are used, but these may |
---|
557 | be set to any other value in a simple way. One can get an |
---|
558 | approximate simulation of the backscattering with the default value, while |
---|
559 | if a better backscattering simulation is needed it is possible to get it |
---|
560 | using a smaller value for $f_r$. However, this model is very simple and |
---|
561 | it can only approximately reproduce the backscattering data. |
---|
562 | |
---|
563 | \subsection{Boundary Crossing Algorithm} |
---|
564 | |
---|
565 | A special stepping algorithm has been implemented recently (Autumn 2006) |
---|
566 | in order to improve the simulation around interfaces. |
---|
567 | This algorithm does not allow 'big' last steps in a volume and 'big' first steps |
---|
568 | in the next volume. The step length of these steps around a boundary crossing |
---|
569 | can not be bigger than the mean free path of the elastic scattering |
---|
570 | of the particle in the given volume (material). After these small steps the particle |
---|
571 | scattered according to a single scattering law (i.e. there is no multiple scattering |
---|
572 | very close to the boundary or at the boundary). \\ |
---|
573 | |
---|
574 | \noindent |
---|
575 | The key parameter of the algorithm is the variable called $skin$. The algorithm is |
---|
576 | not active for $skin \leq 0$, while for $skin > 0$ it is active in layers |
---|
577 | of thickness $skin \cdot \lambda_{elastic}$ before boundary crossing |
---|
578 | and of thickness $(skin-1) \cdot \lambda_{elastic}$ after |
---|
579 | boundary crossing (for $skin=1$ there is only one small step just before |
---|
580 | the boundary). In this active area the particle |
---|
581 | performs steps of length $\lambda_{elastic}$ (or smaller if the particle reaches the |
---|
582 | boundary traversing a smaller distance than this value). \\ |
---|
583 | |
---|
584 | \noindent |
---|
585 | The scattering at the end of a small step is single or |
---|
586 | plural and for these small steps there are no path length correction |
---|
587 | and lateral displacement computation. In other words the program works in this |
---|
588 | thin layer in 'microscopic mode'. \\ |
---|
589 | |
---|
590 | \noindent |
---|
591 | The elastic mean free path can be estimated as |
---|
592 | \begin{equation} |
---|
593 | \lambda_{elastic} = \lambda_1 \cdot rat \left( T_{kin} \right) |
---|
594 | \end{equation} |
---|
595 | where $rat(T_{kin})$ a simple empirical function computed from the elastic and |
---|
596 | first transport cross section values of Mayol and Salvat \cite{msc.mayol} |
---|
597 | |
---|
598 | \begin{equation} |
---|
599 | rat \left( T_{kin} \right) = \frac{0.001 (MeV)^2} |
---|
600 | { T_{kin} \left(T_{kin} + 10 MeV \right)} |
---|
601 | \end{equation} |
---|
602 | $T_{kin}$ is the kinetic energy of the particle. \\ |
---|
603 | |
---|
604 | \noindent |
---|
605 | At the end of a small step the number of scatterings is sampled according to |
---|
606 | the Poissonian distribution with a mean value $t/\lambda_{elastic}$ and in the |
---|
607 | case of plural scattering the final scattering angle is computed by summing |
---|
608 | the contributions of the individual scatterings. \\ |
---|
609 | \noindent |
---|
610 | The single scattering is determined by the distribution |
---|
611 | \begin{equation} |
---|
612 | g(u) = C \frac{1} {(1 + 0.5 a^2 -u)^2} |
---|
613 | \end{equation} |
---|
614 | where $u = \cos(\theta)$ , $a$ is the screening parameter, |
---|
615 | $C$ is a normalization constant. The form of the screening parameter is |
---|
616 | \begin{equation} |
---|
617 | a = \frac{\alpha Z^{1/3}}{\sqrt{(\tau (\tau+2))}} |
---|
618 | \end{equation} |
---|
619 | where $Z$ is the atomic number, $\tau$ is the kinetic energy measured in |
---|
620 | particle mass units, $\alpha$ is a constant. |
---|
621 | It can be shown easily that the function $g(u)$ for small scattering angle $\theta$ |
---|
622 | is equivalent to the well known screened Rutherford scattering formula : |
---|
623 | \begin{equation} |
---|
624 | \tilde{g}(\theta) \ d\Omega = \tilde{C} |
---|
625 | \frac{\theta \ d\theta} { (a^2 + \theta^2)^2} |
---|
626 | \end{equation} |
---|
627 | |
---|
628 | |
---|
629 | \subsection{Implementation Details} |
---|
630 | |
---|
631 | The {\sf G4MultipleScattering} process contains some intialization |
---|
632 | functions which allow the values of stepping and model parameters |
---|
633 | to be defined. \\ |
---|
634 | |
---|
635 | \noindent |
---|
636 | The stepping/boundary crossing algorithm is governed by the data members |
---|
637 | $f_r$ and $f_g$ (facrange and facgeom in the code) and $skin$. The default values |
---|
638 | of $f_r$ and $f_g$ parameters are $0.02$ and $2.5$ resp. but these values can be changed |
---|
639 | via set functions : |
---|
640 | \begin{itemize} |
---|
641 | \item |
---|
642 | {\tt MscStepLimitation(bool, double)} - activate/deactivate stepping algorithm |
---|
643 | and set facrange value; |
---|
644 | \item |
---|
645 | {\tt SetFacrange(double)} |
---|
646 | \item |
---|
647 | {\tt SetFacgeom(double)} |
---|
648 | \end{itemize} |
---|
649 | The default value of $skin$ is $1$ for electrons and positrons and $0$ for all |
---|
650 | the other charged particles. These default values can be changed using the set function |
---|
651 | {\sf SetSkin(double)}. \\ |
---|
652 | |
---|
653 | \noindent |
---|
654 | The concrete physics model is implemented in the class {\sf G4UrbanMscModel}. |
---|
655 | As the boundary crossing/stepping algorithm is a part of the model, this class |
---|
656 | contains it in the {\tt TruePathLengthLimit} method. \\ |
---|
657 | |
---|
658 | \noindent |
---|
659 | Because multiple scattering is very similar for different particles the base |
---|
660 | class {\sf G4VMultipleScattering} was created to collect and provide significant |
---|
661 | features of the calculations which are common to different particle types. \\ |
---|
662 | |
---|
663 | \noindent |
---|
664 | In the {\tt AlongStepGetPhysicalInteractionLength} method the minimum step |
---|
665 | size due to the physics processes is compared with the step size constraints |
---|
666 | imposed by the transportation process and the geometry. In order to do this, |
---|
667 | the \mbox{'t' step $\rightarrow$ 'z' step} transformation must be performed. |
---|
668 | Therefore, the method should be invoked after the |
---|
669 | {\tt GetPhysicalInteractionLength} methods of other physics processes, |
---|
670 | but before the same method of the transportation process. |
---|
671 | The reason for this ordering is that the physics processes 'feel' |
---|
672 | the true path length $t$ traveled by the particle, while the transportation |
---|
673 | process (geometry) uses the $z$ step length.\\ |
---|
674 | |
---|
675 | \noindent |
---|
676 | At this point the program also checks whether the particle has entered a |
---|
677 | new volume. If it has, the particle steps cannot be bigger than |
---|
678 | $t_{lim} = f_r\hspace{2mm} max( r, \lambda )$. This step limitation is |
---|
679 | governed by the physics, because $t_{lim}$ depends on the particle energy |
---|
680 | and the material. \\ |
---|
681 | |
---|
682 | \noindent |
---|
683 | The {\tt PostStepGetPhysicalInteractionLength} method of the multiple |
---|
684 | scattering process simply sets the force flag to 'Forced' in order to |
---|
685 | ensure that {\tt PostStepDoIt} is called at every step. It also returns a |
---|
686 | large value for the interaction length so that there is no step limitation |
---|
687 | at this level. \\ |
---|
688 | |
---|
689 | \noindent |
---|
690 | The {\tt AlongStepDoIt} function of the process performs the inverse,\linebreak |
---|
691 | \mbox{'z' $\rightarrow$ 't'} transformation. This function should be |
---|
692 | invoked after the \linebreak {\tt AlongStepDoIt} mehtod of the transportation |
---|
693 | process, that is, after the particle relocation is determined by the |
---|
694 | geometrical step length, but before applying any other physics |
---|
695 | {\tt AlongStepDoIt}. \\ |
---|
696 | |
---|
697 | \noindent |
---|
698 | The {\tt PostStepDoIt} method of the process samples the scattering angle |
---|
699 | and performs the lateral displacement when the particle is not near a boundary. |
---|
700 | |
---|
701 | \subsection{Status of this document} |
---|
702 | 09.10.98 created by L. Urb\'an. \\ |
---|
703 | 15.11.01 major revision by L. Urb\'an.\\ |
---|
704 | 18.04.02 updated by L. Urb\'an. \\ |
---|
705 | 25.04.02 re-worded by D.H. Wright \\ |
---|
706 | 07.06.02 major revision by L. Urb\'an. \\ |
---|
707 | 18.11.02 updated by L. Urb\'an, now it describes the new angle distribution. \\ |
---|
708 | 05.12.02 grammar check and parts re-written by D.H. Wright \\ |
---|
709 | 13.11.03 revision by L. Urb\'an. \\ |
---|
710 | 01.12.03 revision by V. Ivanchenko. \\ |
---|
711 | 17.05.04 revision by L.Urb\'an. \\ |
---|
712 | 01.12.04 updated by L.Urb\'an. \\ |
---|
713 | 18.03.05 sampling z + mistyping corrections (mma) \\ |
---|
714 | 22.06.05 grammar, spelling check by D.H. Wright \\ |
---|
715 | 12.12.05 revised by L. Urb\'an, according Msc version in Geant4 V8.0 \\ |
---|
716 | 14.12.05 updated Implementation Details (mma) \\ |
---|
717 | 08.06.06 revised by L. Urb\'an, according Msc version in Geant4 V8.1 \\ |
---|
718 | 25.11.06 revised by L. Urb\'an, according Msc version in Geant4 V8.2 \\ |
---|
719 | 29.03.07 revised by L. Urb\'an \\ |
---|
720 | |
---|
721 | \begin{latexonly} |
---|
722 | |
---|
723 | \begin{thebibliography}{99} |
---|
724 | |
---|
725 | \bibitem{msc.moliere} G. Z. Moli\`ere |
---|
726 | {\em Z. Naturforsch. 3a (1948) 78. } |
---|
727 | \bibitem{msc.lewis} H. W. Lewis. |
---|
728 | {\em Phys. Rev. 78 (1950) 526. } |
---|
729 | \bibitem{msc.goudsmit}S. Goudsmit and J. L. Saunderson. |
---|
730 | {\em Phys. Rev. 57 (1940) 24. } |
---|
731 | \bibitem{msc.fernandez}J. M. Fernandez-Varea et al. |
---|
732 | {\em NIM B73 (1993) 447.} |
---|
733 | \bibitem{msc.kawrakow} I. Kawrakow and Alex F. Bielajew |
---|
734 | {\em NIM B 142 (1998) 253. } |
---|
735 | \bibitem{msc.liljequist1} D. Liljequist and M. Ismail. |
---|
736 | {\em J.Appl.Phys. 62 (1987) 342. } |
---|
737 | \bibitem{msc.liljequist2} D. Liljequist et al. |
---|
738 | {\em J.Appl.Phys. 68 (1990) 3061. } |
---|
739 | \bibitem{msc.mayol} R.Mayol and F.Salvat |
---|
740 | {\em At.Data and Nucl.Data Tables} {\bf 65}, p. 55 (1997). |
---|
741 | \bibitem{msc.highland} V.L.Highland |
---|
742 | {\em NIM 129 (1975) 497. } |
---|
743 | \bibitem{msc.lynch} G.R. Lynch and O.I. Dahl |
---|
744 | {\em NIM B58 (1991) 6. } |
---|
745 | \bibitem{msc.shen} G.Shen et al. |
---|
746 | {\em Phys. Rev. D 20 (1979) 1584.} |
---|
747 | \bibitem{msc.attwood} D. Attwood et al. |
---|
748 | {\em NIM B 251 (2006) 41.} |
---|
749 | \end{thebibliography} |
---|
750 | |
---|
751 | \end{latexonly} |
---|
752 | |
---|
753 | \begin{htmlonly} |
---|
754 | |
---|
755 | \subsection{Bibliography} |
---|
756 | |
---|
757 | \begin{enumerate} |
---|
758 | |
---|
759 | \item G. Z. Moli\`ere |
---|
760 | {\em Z. Naturforsch. 3a (1948) 78. } |
---|
761 | \item H. W. Lewis |
---|
762 | {\em Phys. Rev. 78 (1950) 526. } |
---|
763 | \item S. Goudsmit and J. L. Saunderson. |
---|
764 | {\em Phys. Rev. 57 (1940) 24. } |
---|
765 | \item J. M. Fernandez-Varea et al. |
---|
766 | {\em NIM B73 (1993) 447.} |
---|
767 | \item I. Kawrakow and Alex F. Bielajew |
---|
768 | {\em NIM B 142 (1998) 253. } |
---|
769 | \item D. Liljequist and M. Ismail. |
---|
770 | {\em J.Appl.Phys. 62 (1987) 342. } |
---|
771 | \item D. Liljequist et al. |
---|
772 | {\em J.Appl.Phys. 68 (1990) 3061. } |
---|
773 | \item R.Mayol and F.Salvat |
---|
774 | {\em At.Data and Nucl.Data Tables 65 (1997) 55.} |
---|
775 | \item V.L.Highland |
---|
776 | {\em NIM 129 (1975) 497. } |
---|
777 | \item G.R. Lynch and O.I. Dahl |
---|
778 | {\em NIM B58 (1991) 6. } |
---|
779 | \item{msc.shen} G.Shen et al. |
---|
780 | {\em Phys. Rev. D 20 (1979) 1584.} |
---|
781 | \item{msc.attwood} D. Attwood et al. |
---|
782 | {\em NIM B 251 (2006) 41.} |
---|
783 | \end{enumerate} |
---|
784 | |
---|
785 | \end{htmlonly} |
---|
786 | |
---|
787 | |
---|
788 | |
---|