Impact of molecular bonding on therm. neutron cross-sections

[KarenRei]

Hi everyone. Most of what I've read on thermal neutron cross sections talks only about what isotope makes up the neutron target. Likewise, all of the cross sections listed on Sigma:

http://www.nndc.bnl.gov/sigma/index.jsp?as=9&lib=endfb7.1&nsub=10

... are broken down just by isotope. However, in Geant4's (very limited) thermal neutron scattering functionality, it wants to know what sort of bonding structure the isotope is in. Their available materials are:

TS_Aluminium_Metal
TS_Beryllium_Metal
TS_Be_of_Beryllium_Oxide
TS_C_of_Graphite
TS_D_of_Heavy_Water
TS_H_of_Water
TS_H_of_Zirconium_Hydride
TS_H_of_Polyethylene
TS_Iron_Metal
TS_f_Uranium_Dioxide
TS_f_Beryllium_Oxide
TS_U_of_Uranium_Dioxide
TS_Zr_of_Zirconium_Hydride
TS_H_of_Para_Hydrogen
TS_H_of_Ortho_Hydrogen
TS_D_of_Para_Deuterium
TS_D_of_Ortho_Deuterium
TS_H_of_Liquid_Methane
TS_H_of_Solid_Methane

This sort of information isn't needed when not using thermal scattering, one simply lists what isotopes are in the material and at what concentrations.

At least in the case of hydrogen, there indeed seems to be a difference. For example, this paper:

http://journals.aps.org/pr/abstract/10.1103/PhysRev.54.266

...says:

"The following values were obtained for the neutron scattering cross sections, σ, of liquid ortho- and parahydrogen: for ∼300°K neutrons (paraffin at room temperatures), σ (ortho)=56×10-24 cm2 per molecule, σ (para)=29×10-24; for ∼120°K neutrons (paraffin cooled with liquid air), σ (ortho)=79×10-24 and σ (para)=18×10-24. We were able to prove conclusively: (1) that the interaction between neutrons and protons is dependent upon the relative alignment of their spins, and (2) that the energy of the singlet state of the deuteron, in which the spins of the proton and neutron are antiparallel, is greater than the energy of these particles when far apart ... The scattering cross sections of the protons in water, (12)[σ(H2O)-σ(O)], were 43.6×10-24 cm2 per proton for ∼300°K neutrons and 56×10-24 cm2 per proton for ∼120°K neutrons."

That is to say, for 120K neutrons, hydrogen's cross sections are, in barns:

Orthohydrogen: 79
Parahydrogen: 18
Water: 56

For 300K neutrons:

Orthohydrogen: 56
Parahydrogen: 29 (*more* than at 120K?)
Water: 43.6

That's some serious reported differences, and it's no wonder that Geant4 would want to know it. But then of what use are Sigma's cross sections that don't take into account molecular bonding?

Or, contrarily, is this something that's predominantly only a pronounced effect in hydrogen, due to its light mass and single valence electron?

 

[DEvens]

The effect is most important for hydrogen and thermal neutrons. The thermal velocity of a hydrogen nucleus, a proton, can be a significant fraction of the velocity of a thermal neutron. Thus the motion of the proton can have an effect on the cross section. Recall that in this region the cross section curve has this 1/v part. So if the proton happens to be moving towards or away from the neutron, it can change the cross section quite a bit.

 

[KarenRei]

Ah, that makes good sense. Thanks!