Logo of the Physikalisch-Technische Bundesanstalt

Initiative for the improvement of the data set for electron transport in water


Biological effects which occur after living organisms have been exposed to radiation must largely be attributed to radiation interactions in the chromosomes and, in particular, to molecular changes in the DNA. These strongly depend on the character and on the spatial distribution of particle interactions which take place in the cell nucleus. Of special importance for the detrimental effect of ionizing radiation are primary double-strand breaks in the DNA, which can often not be repaired by the cell and may lead to considerable damage (such as genetic mutations or even cancer). If one assumes that at least two ionization events must have occurred for their initiation, decisive importance must be attributed to the generation of ionization clusters (number of ionization events in a defined target volume) in segments of the DNA. The physics of ionization cluster generation in nanometric volumes is, therefore, closely connected with radiation biology and represents a challenge to radiation metrology. A special part is played by low-energy electrons which are - independent of the radiation type - generated as secondary particles as a result of the ionization process in a medium and which are responsible for approx. 30 % of the deposited energy. The uncertainties in the calculation of the ionization cluster generation by low-energy electrons in a material important to dosimetry very strongly depend on the quality of the ionization cross sections. To determine their influence, the ionization cluster generation by electrons in water was simulated with the aid of two different Monte Carlo programs. The two programs (the Monte Carlo program of Department 6.6 [1] and GEANT4-DNA [2]) have partly different electron cross sections for inelastic processes and are, therefore, suited to this study. The selected geometry is a water cube with a lateral length of 160 nm in which the target volume for the calculation of the ionization cluster generation is contained. The target volume was assumed to be cylindrical, and its dimensions were selected in such a way that they corresponded to those of a DNA segment with ten base pairs (diameter D = 2.3 nm, height H = 3.4 nm). For the simulation calculations it was, in addition, assumed that monoenergetic electrons are emitted as a needle beam in the centre of the water cube in the direction of the target volume, orthogonal to the main axis.

As an example of the results, Figure 1 shows the mean size of the ionization clusters generated by electrons in water as a function of the electron energy T. The agreement in the case of energies larger than 500 eV is very good, whereas in the low-energy range the different cross sections, which are no longer based on the Bethe theory, cause deviations up to approx. 80 % at 50 eV. This can also be seen in Figure 2 which shows the distribution of the ionization cluster size for electrons with energies of 100 eV, 200 eV, 1 keV and 10 keV.

Figure 1 : Mean size M1 of the ionization clusters generated by electrons in water calculated with the aid of Monte Carlo simulations as a function of the electron energy T. The target volume is a cylinder with a diameter of D = 2.3 nm and a height of H = 3.4 nm. For the simulations, two programs were used: the package developed at Department 6.6 (PTB) and GEANT4-DNA (G4).

Figure 2 : Distribution Pν of the ionization cluster size ν for electrons with energies of 100 eV, 200 eV, 1 keV and 10 keV. See Figure 1 for further details.

This work was performed in cooperation with Mrs. Marion Bug (Centre for Medical Radiation Physics, University of Wollongong) [3] and represents the first step of an extensive initiative which will allow the data set for the electron transport in water be to improved and will make these data available to the scientific community.


  1. Grosswendt, B.:
    From macro to nanodosimetry: limits of the absorbed-dose concept and definition of new quantities,
    Proc. Workshop on Uncertainty Assessment in Computational Dosimetry (2007), Bologna.
  2. Chauvie, S.; Francis, Z.; Guatelli, S.; Incerti, S.; Mascialino, B.; Moretto, P.; Nieminen, P.; Pia, M. G.:
    Geant4 physics processes for microdosimetry simulation: Design foundation and implementation of the first set of models,
    IEEE Trans. Nucl. Sci. 54(6) (2007).
  3. Bug, M.; Gargioni, E.; Guatelli, S.; Incerti, S.; Bendall, L.; Kaplan, G.; El-Hajj, R.; Oktavia, S.; Wroe, A.; Schulte, R.; Rosenfeld, A.:
    Nanodosimetric modelling of low energy electrons in a magnetic field,
    Proc. 13th GEANT4 Collaboration Workshop, Kobe, Japan (2008).