Why did the kinetic energy of a thermal neutron increase after an interaction?

I’m quite new to Geant4. Recently I was simulating neutron used the physics list QGSP_BIC_HP.
The material is shown below:

G4Element* elB = new G4Element(name = “Boron”,
symbol = “B”,
z = 5.0, a = 10.811 * g / mole);
G4Element* elN = new G4Element(name = “Nitrogen”,
symbol = “N”,
z = 7.0, a = 14.0067 * g / mole);

G4Material* hBN = new G4Material(name = “h-BN”,
density = 2.28 * g / cm3,
numberofelements = 2);
hBN->AddElement(elB, numberofatoms = 1);
hBN->AddElement(elN, numberofatoms = 1);

The kinetic energy of neutron increased after an interaction. And N14 appeared after a few interactions(also with increased kinetic energy of neutron). How is that possible?

particalname: neutron
PreKineticEnergy: 2.5e-05
PostKineticEnergy: 3.97562e-05

particalname: neutron
PreKineticEnergy: 3.97562e-05
PostKineticEnergy: 4.8027e-05

particalname: N14
trackID: 4
parentID: 1
PreKineticEnergy: 1.23255e-05
PostKineticEnergy: 0

Is is due to the thermal motion of the atoms in the material. The energy can be transferred to a neutron in a scattering. N14 is a knockout atom, Geant4 by default does not take into account the minimal kinetic energy needed for the atom displacement from the lattice. Usually the charged recoils with such low energy are not produced due to the tracking cut, unless the event is close to the geometry boundary.

Thank you so much your help. It seems that this phenomenon is not physical. Is there a way to avoid this or not? This is not helpful to my simulation.

It is physycal.
First, it affects the energy of the scattered neutrons. One can ignore this effect for the neutron energies much greater than the energy of the thermal motion of the atoms of the material. In your case they are of the same order of magnitude.
Second, it affects the effective cross section. In this paper https://iopscience.iop.org/article/10.1088/1742-6596/608/1/012050 you can see the formula for the elastic cross section. The capture cross section at thermal energies is unmodified due to the 1/v dependence. In the resonance region it leads to the broadening of the resonances. For some materials a more precise treatment is performed taking into account not only the temperature, but also the binding the atoms in the material https://inis.iaea.org/search/search.aspx?orig_q=RN:49103646