We have analyzed primary γ-ray spectra of the odd-odd 238 Np nucleus extracted from 237 Np(d, pγ) 238 Np coincidence data measured at the Oslo Cyclotron Laboratory. The primary γ spectra cover an excitation-energy region of 0 ≤ E i ≤ 5.4 MeV, and allowed us to perform a detailed study of the γ-ray strength as function of excitation energy. Hence, we could test the validity of the generalized Brink-Axel hypothesis, which, in its strictest form, claims no excitation-energy dependence on the γ strength. In this work, using the available highquality 238 Np data, we show that the γ-ray strength function is to a very large extent independent on the initial and final states. Thus, for the first time, the generalized Brink-Axel hypothesis has been experimentally verified for γ transitions between states in the quasi-continuum region, not only for specific collective resonances, but also for the full strength below the neutron separation energy. Based on our findings, the necessary criteria for the generalized Brink-Axel hypothesis to be fulfilled are outlined.PACS numbers: 24.30. Gd, 21.10.Ma, 25.40.Hs Sixty years ago, David M. Brink proposed in his PhD thesis [1] that the photoabsorption cross section of the giant electric dipole resonance (GDR) is independent of the detailed structure of the initial state. In his thesis, he expressed his hypothesis as follows: "If it were possible to perform the photo effect on an excited state, the cross section for absorption of a photon of energy E would still have an energy dependence given by (15)", where equation (15) refers to a Lorentzian shape of the photoabsorption cross section. Brink's original idea, the Brink hypothesis, was first intended for the photoabsorption process on the GDR, but has been further generalized, applying the principle of detailed balance, to include absorption and emission of γ rays between resonant states [2,3]. In addition to assuming independence of excitation energy, there is no explicit dependence of initial and final spins except the obvious dipole selection rules, implying that all levels exhibit the same dipole strength regardless of their initial spin quantum number. We will refer to this as the generalized Brink-Axel (gBA) hypothesis. A review of the history of the hypothesis was given by Brink in Ref. [4].The gBA hypothesis has implications for almost any situation where nuclei are brought to an excited state above ≈ 2∆, where ∆ ≈ 1 MeV is the pair-gap parameter. Here, the nucleus will typically de-excite via γ-ray emission and/or by emission of particles. In this context, it is usual to translate the γ-ray cross section σ (E γ ) into γ-ray strength function (γSF) by f (E γ ) = (3π 2h2 c 2 ) −1 σ (E γ )/E γ .To describe and model the electric dipole part of the γ-decay channel, the gBA hypothesis is frequently used, applying in particular the assumption of spin independence [5]. For example, a rather standard approach to calculating E1 strength is to apply some variant of the quasi-particle random-phase approximation (QRPA) to obtai...
