No abstract

In the paper 'Neutron depolarization theory in the Larmor and the scattering approach' [1], it has been shown tlhat the extended Larmor and the scattering approach are basically equivalent but differ formally. However, both approaches are fully equivalent, which is shown in this comment. Furthermore, the equivalence of the Larmor and the extended Larmor approach in the case of infinitely elongated ellipsoids is shown in a more formal way than was done in [1]. For completeness, an inaccuracy and a printing error in two expressions of the original paper are corrected.In [1], it has been shown that the expressions for the depolarization matrix /5 in the scattering and the Larmor approach are equivalent, except for a term related to the vector qOr dO,Here, A V is a part of the medium considered, M(r) the local magnetization, r the scattering vector, k=r/lt~], q2 dO=4@ 2 dr and 2 the neutron wavelength. Obviously, q(r) is an anti-symmetric function of r (q(r)= -q(-~)). Therefore, the contribution of the term in a neutron transmission experiment, i.e. after integration over f2 under the approximation that ~c~=0, vanishes for any M(r) distribution. This implies the full equivalence of the extended Larrnor and the scattering approach. As the term discussed is not present in the formulas derived by Maleev and Ruban in the scattering approach [2, 3], they probably used the same argument.The equivalence of the Larmor and the extended Larmot approach in the case of infinitely elongated ellipsoids can be proved in a more formal and also simpler way than has been done in [-1] (31). The equivalence directly follows from a comparison of the general expressions for ~ij of ellipsoids with axial dimension 2b, radial dimension 2 a and a spontaneous magnetization M along the b-axis in both approaches. In the extended Larmor approach this expression is given by (see (29)

No abstract