We analyze the recent experimental data on photofissility for 237Np, 238U, and 232Th at incident photon energies above 200MeV . For this analysis, we developed a Monte Carlo algorithm for the nuclear evaporation process in photonuclear reactions. This code is used in association with the multi-collisional model for the photon-induced intranuclear cascade process. Our results show a good quantitative and qualitative agreement with the experimental data. It is shown that the emission of protons and alpha particles at the evaporation stage is an important component for the non-saturation of the actinides photofissility up to, at least, 1GeV . PACS25.85.Jg, 25.20.-x, 25.85.-w Key-words: Photonuclear reactions; PhotoÞssion; PhotoÞssility Permanent address: Instituto de Fisica da Universidade de Sao Paulo Sao PauloBrazil, P.O.Box 66318 CEP 05315-970 Also: Sociedade Educacional Sao Paulo ApostoloUniverCidade, 22710-260 Rio de Janeiro, Brazil. 1 CBPF-NF-048/01 It has been widely believed that the Þssility (W ) of actinide nuclei should saturate at 100% for energies above ∼ 100MeV [1, 2, 3, 4]. Such a possibility is so appealing and convincing that several groups have been proposing research projects devoted to the systematic investigation of the photoabsorption process at intermediate and high energies. In fact, since W = 1, photoÞssion cross section measurements would propitiate a good evaluation for the total nuclear photoabsorption cross section [4, 5, 6], and this would conÞgurate the easiest and most direct method of photoabsorption cross section measurement for the heavy nuclei. However, as pointed out elsewhere[5, 6, 7], the total nuclear photoabsorption cross section is an important and interesting source of information on the role played by the nuclear medium in the intrinsic properties and interaction aspects of the nucleons, as well as on the earlier stages of the shadowing effect, a manifestation of the hadronic nature of the photon. The Þrst disturbance in this optimistic scenario came, however, with the experimental results for the photoÞssility of Th. It was found that W is ∼ 60% to ∼ 80% of that for U in the energy interval 200− 1200MeV [4]. Because of some precedents associated with the photoprocess in Th, as the well-known thorium anomalies manifesting at low and intermediate energies[8, 9, 10], it could be conjectured that the non-saturation of the Th photoÞssility, at energies as high as 1.2GeV , is another sort of unexplained anomaly exhibited by this nuclide. In this regard, a phenomenological description of the photoÞssility[10], suggested that the non-saturation of photoÞssility in Th could be a consequence of its higher nuclear transparency comparatively to that of U, and a model based on the nuclear structure was proposed in Ref. [11] to explain these photoÞssility results. The second difficulty came from photoÞssion results for Np reported by the Novosibirsk group in the early nineties[3]. Quite disturbing at that time, the results revealed a photoÞssility for Np, in the energy interval 60− 240MeV , nearly 30% higher than that for U. These results were conÞrmed quite recently by Sanabria and collaborators[12], in a photoÞssion experiment carried out in Saskatoon. No convincing explanation has so far been presented for these Þndings. Finally, in a recent experiment performed at the Photon Tagging Facility in Hall B at the Thomas Jefferson Laboratory, Cetina et al.[13] thoroughly demonstrated that the photoÞssion cross section for U is about 80% of that for Np up to ∼ 4GeV . Again, neither qualitative nor quantitative explanation has been proposed. These Þndings claim for a convincing explanation given 2 CBPF-NF-048/01 their several implications on nuclear structure aspects[1, 4, 11], on compound nucleus formation mechanisms[10], and on the potentialities of the Þssion channel as a probe to infer new nuclear reactions characteristics[4, 5]. In this letter we present for the Þrst time a complete and detailed calculation of the photoÞssility for actinide nuclei. This is achieved by using a combination of the multicollisional Monte Carlo calculation (MCMC described in Ref. [14]) for the photoninduced intranuclear cascade process, and a new Monte Carlo algorithm developed by us for the evaporation-Þssion process, which includes not only the neutron evaporation vs. Þssion competition, but also takes into account the evaporation of protons and alpha particles. We have applied these calculational procedures to obtain the photoÞssility of Np, U and Th. As discussed below, our results provide a good description of the experimental absolute and relative photoÞssilities from 0.2GeV to 1GeV . We did not extend the calculation above 1GeV because a signiÞcant shadowing effect takes place at higher energies, starting below 1.5GeV [13, 15], and this effect is not yet included in our intranuclear cascade calculation. The MCMC method propitiates a more realistic description of the intranuclear cascade process, comparatively to the traditional methods[16, 17], since it gives a time-ordered evolution of the cascade by taking into account the nucleus conÞguration at each instant of time. The evaluation of the collisional probabilities among the nucleons, as well as the secondaries arising from these collisions, is carried out[14]. Such a realistic description results in a higher multiplicity of protons and neutrons leading, thus, to the formation of less massive compound nuclei as compared with those coming from traditional intranuclear cascade calculations. This aspect is the key to the photoÞssility non-saturation clue, because lighter nuclei have lower Þssion probabilities. We were, then, motivated to develop an algorithm for the evaporation process, which is a complement to the multicollisional algorithm. With the former we calculate the evaporation-Þssion competition taking place in the compound nuclei, which is obtained from the latter. The compound nucleus, (Ac, Zc), have excitation energy, Ec, which is in accordance with the results of a previous analysis on the subject [10]. The probability for the emission of a particle j with kinetic energy between Ek and Ek + dEk is calculated according to the Weisskopfs statistical model[18] as, 3 CBPF-NF-048/01 Pj (Ek) dEk = γjσjEk ! ρf ρi " dEk, (1) where σj is the nuclear capture cross section for particle j by the Þnal nucleus, γj = gm (π2h3) , where g denotes the number of spin states, and m is the particle mass. The level density for the initial and Þnal nuclei, ρi and ρf , respectively, are calculated from the Fermi gas expression ρ # E∗ j $ = exp % 2 # aE∗ j $ 1 2 & , (2) where a is the level density parameter, and E∗ j = E ∗ − (Bj + Vj) . (3) Here, E∗ is the nuclear excitation energy in the initial state, Bj is the particle separation energy, and Vj is the Coulomb potential barrier corrected for the nuclear temperature, τ , deÞned as E∗ = aτ . The particle emission width is calculated as