B prev next
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

<< MC APPLICATIONS >>

Simulating Radioactive Decay

This is a truly random proceses. The probability of decay is constant. The probabilty that a nucleus undergoes radioactive decay in time dt is p:

p = L dt    (for L dt <<1)

where L (decay constant) is probability per unit time for the decay of each nucleus of a given nuclide.

Consider a system initially having N0 unstable nuclei. How does the number of parent nuclei, N, change with time?

Theoretically, the number of undecayed nuclei at time t is given by:

N = N 0 exp (- L t )

where N0 is the number of parent(undecayed) nuclei at t = 0.


Alogrithm:
 Determine N    (initial number of parents)
 Determine L    (decay constant)
 Determine Tmax (any time)
 Determine dt   (time step)

 LOOP from t=0 to Tmax, step dt

    LOOP over each remaining parent nucleus
      Generate a random number R from a uniform distribution
      Decide if the nucleus decays:
      Check IF( R < L dt ) THEN
         reduce the number of parents by 1
      END IF
    END LOOP over nuclei

    PLOT or Record N vs t

 END LOOP over time


The computer implementations of the algoritm: rd.f90 | rd.c
The plots N vs t (both linear and logaritmic): Gif image