next up previous index
Next: Restrictions Up: PHYS351 Simulation of Previous: Subroutines

Method

The type of annihilation is sampled from the total cross-sections for the annihilation into two photons and into one photon (see section [PHYS350]).

Annihilation into two photons

The differential cross-section of the two-photon positron-electron annihilation can be written as [,]: {d σ(Z, ε)d ε}= m &sp;a [S(a ε) + S (a(1- ε))]

where m is the electron mass Z is the atomic number of the material. If we indicate with E the initial energy of the positron, with r0 the classical electron radius and with k the energy of the less energetic photon generated, we have:

γ = {Em} a = γ+1

ε = {kE+m} S(x) = C1[ -1 + {C2x}-{1x2}]

C1 = {Z πr02a(E-m)} C2 = a + {e γa}

The kinematical limits for the variable ε are:

ε0= {1a+ γ2-1}≤ε≤{12}

Due to the symmetry of the formula (gif) in ε , the range of ε can be expanded from (ε0,1/2 ) to (ε0,1-ε0 ) and the second function S can be eliminated from the formula. Having done this, the differential cross-section can be decomposed (apart from the normalisation) as: {d σd ε}={1ln{1 - ε0ε0}}{1ε}f(ε){(a2+2a-2)-a2ε-{1ε}a2-2ln{1 - ε0ε0}}g(ε)

 

 


Figure: Comparison between the K-shell binding energies given by the expression in the text and the tabulated values.

Using the expression (gif) with random numbers ri∈]0,1[, i=1,2 , the secondary photon energy is sampled by the following steps:

  1. sample ε from f(ε) : ε=ε0exp[ ln({1- ε0ε0}) r1]

  2. compute the rejection function g(ε) and
    1. if r2≤g(ε) accept ε

    2. if r1> g(ε) go back to 1.
After the successful sampling of ε , the photon energy is computed as k = (E+m)ε

and then GANNI generates the polar angles of the photon with respect to an axis defined by the momentum of the positron. The azimuthal angle Φ is generated isotropically and Θ is computed from the energy-momentum conservation. With this information, the momentum vector of both photons can be calculated and transformed into the GEANT coordinate system.

The routine GANNIR treates the special case when a positron falls below the cut-off energy ( CUTELE in common block /GCCUTS/) before annihilating. In this case, it is assumed that the positron comes to rest before annihilating. GANNIR generates two photons with energy k=m. The angular distribution is isotropic.

Annihilation into one photon

The generated photon is assumed to be collinear with the positron. Its energy will be k = E + me- Ebind

where Ebind is the binding energy of the K-shell electron. It can be estimated as follows Ebind= 0.5 (Z  α)2 me

where α is the fine stucture constant. The comparison of this expression with the experimental data from [] is shown in figure gif.



next up previous index
Next: Restrictions Up: PHYS351 Simulation of Previous: Subroutines


Janne Saarela
Mon Apr 3 12:46:29 METDST 1995