Gabriele, John,..

As part of the Neutrino Factory/Muon Collider effort, we are still very much
interested in studying particle transport in Geant4, in presence of strong
electric field. As I am sure you are aware, if I apply an electric field, I
expect the particle to eventually change direction and *accelerate*.  This does
not occur if I use the following setup: (from my DetectorConstruction.cc).
:

      ExN01EFieldSpaceCharge* fullEMField  =
                        new  ExN01EFieldSpaceCharge();
      G4EqMagElectricField *fEquationE =
        new G4EqMagElectricField(fullEMField);
      G4MagIntegratorStepper *pStepper = new G4ClassicalRK4 (fEquationE, 6);
      fieldMgr->SetDetectorField(fullEMField);
      G4ChordFinder* pChordFinder =
             new G4ChordFinder(fullEMField, 10 * mm , pStepper);
      fieldMgr->SetChordFinder(pChordFinder);
      fullEMField->SetTimeForEfield(0.);

My private class ExN01EFieldSpaceCharge has as base class G4ElectroMagneticField
and has a concrete implementation of the required GetFieldValue method.
The codes runs fine, an Electric field change the trajctory, in the correct
direction. (if I expressed the Electric field in the adhoc units, which I
haven't quite figure it out, but that's a detail!).
However, for field corresponding to many MeV/m, the scalar velocity of the
particle does not chage.

Looking at the code, in various files, I indeed found code related to this
issue. But it commented out  (G4transportation.cc):

     // fTransportEndKineticEnergy= aFieldTrack.GetEnergy(); // Energy is wrong
#if 0
     G4ThreeVector endVelocity = aFieldTrack.GetVelocity();
     G4double  veloc_sq = endVelocity.mag2();
     fTransportEndKineticEnergy  = 0.5 * restMass * veloc_sq /
                        ( 1 - veloc_sq / c_squared );   // Lorentz correction
#endif

So my questions are:
a. Is there an option, or entire set of class(es) I missed?
b. if not, why can't we accelerate particles in G4 ?
c. How difficult it would be to implement? Is someone working on it?

So far, I have been very, very pleased with G4 product and the C++ language.
G4 is indeed superior to G3, no doubt about it, and I believe we need it for our
project. So we are willing to put some effort to solve this problem, however, it
will be a change in the core transportation algorithm..
Thank you!.

								Paul Lebrun.