Recently, first candidates for low-lying isovector states in the octupole sector were suggested. This class of states is predicted in the frame of the sdf-IBM-2. Due to the isovector character the unambiguous identification of those states will contribute to the decomposition of the octupole-octupole residual interaction in an isoscalar and isovector part.

In Nd-144 the 3_3^- level at 2778 keV is a good candidate for such a "mixed-symmetry" octupole state. For a transition from a "mixed-symmetry" octupole state to the symmetric octupole phonon a strong M1 component is expected. In order to clarify the nature of this state, a Nd-143(n, g)-experiment was conducted with the EXILL-setup at the ILL, Grenoble, in the frame of the $(n, g)$ campaign. Following neutron capture the 3^- states are populated and EXILL provides the opportunity to determine the multipole-mixing ratios delta of the 3_i^- --> 3_1^- transitions as well as the branching ratios b precisely and, therefore, to determine the nature of the 3_3^- state. Also the lifetime of the 3_3^- state was determined with GAMS6, a high-resulution spectrometer at ILL, to be able to calculate the B(M1;3_3^- --> 3_1^-) value using the results of both experiments.

In the frame of this project, additionally, first candidates for [3^- x 3^-]_{2^+}$ and $[3^- x 3^-]_{4^+}$ in Nd-144 can be suggested. This class of states is up to the present day identified in the semimagic nucleus Gd-146 and Pb-208 only.