Photoneutron reaction cross sections for 75 As and 181 Ta : Sytematic uncertainties and data reliability

There is well–known problem of significant systematic disagreements between data for reactions (γ, 1n), (γ, 2n), obtained at Livermore (USA) and Saclay (France) using the method of photoneutron multiplicity sorting. The averaged ratios Rint of integrated cross sections obtained at Saclay and Livermore for 19 nuclei S/L from 51V to 239U are equal to 0.84 for (γ, 2n) and 1.07 for (γ, 1n) reactions. For 75As Rint ratios for both partial S/L reactions are very close (1.22 and 1.21) but for 181Ta – are quite different (0.89 and 1.25). Using the objective physical data reliability criteria it was found that there are serious doubts in reliability of Saclay and Livermore data. The newly evaluated reliable cross sections disagree with experimental data. In addition to unreliable sorting of many neutrons between both partial reactions many neutrons were lost – on the case of 181Ta in 1n channel, in the case of 75As in both 1n and 2n channels.

In figure 1 one can see two very interesting cases: 75 As, for which R int S /L ratios for both partial reactions are very close (1.22 and 1.21) and 181 Ta, for which R int S /L ratios are significantly different (0.89 and 1.25). This article is devoted to the comparative investigation in * e-mail: VVVarlamov@gmail.com detail experimental data for partial photoneutron reaction cross sections for 75 As and 181 Ta.

The experimental-theoretical method for reliable cross-section evaluation
The ratios, where (γ, in) is the definite partial reaction and (γ, S n) is the neutron yield reaction (1), were proposed [6] as the objective physical criteria of partial reaction cross-section reliability. The definitely positive Fi must not have values higher than 1.00, 0.50, 0.33 for i = 1, 2, 3. Larger F exp values mean that cross sections contains significant systematic uncertainties. The experimental-theoretical method was developed for evaluating the partial reaction cross sections satisfying reliability criteria [6]. The experimental σ exp (γ, S n) rather independent from the experimental neutron multiplicity sorting problems was decomposed into the partial reaction cross sections, using the functions F theor i (2) calculated in the Combined model of photonucleon reactions (CMPNR) [16,17]. For 181 Ta,197 Au and 209 Bi it was shown that the newly evaluated partial reaction cross sections agree with the data obtained using the activation method [18,19] and therefore are reliable. It was found that for many nuclei the experimental partial reaction cross sections do not satisfy the proposed data reliability criteria and are noticeably different from evaluated cross sections [6][7][8][9][10][11][12][13][14][15]20]. It was shown that the main reason of that is unreliable transmission of many neutrons from one partial reaction to another because of shortcomings of procedures used to separate counts into 1n and 2n events.

Photoneutron reaction cross sections for 181 Ta -strange features
Experimental [21,22] and evaluated [10] cross sections for 181 Ta are compared in Figure 2. In the (γ, S n) obtained at Livermore [21] one see the neutrons at photon energies only up to ∼ 17.5 MeV though at Saclay [22] -up to ∼ 25 MeV. Table 1 gives the respective ratios σ int S (γ, 1n)/σ int L (γ, 1n) of integrated cross sections obtained up to energy E int = 25 MeV. Evaluated data are enough close to Saclay [22] but not for Livermore data [21] the larger the fraction of the σ(γ, 1n) reaction in the cross-section: for the total reactions, the higher the degree to which the latter is underestimated (1.24 → 1.30→ 1.46). For σ(γ, 2n), in which the fraction of the σ(γ, 1n) is equal to zero, the σ int eval /σ int L = 1.05. Those data and correspondingly the differences ∆σ = σ eval − σ exp presented in figure 3 confirm that experimental cross section σ(γ, 2n) [22] in general is reliable, but σ(γ, 1n) is absolutely unreliable because many neutrons were lost. Incorrect behavior of the σ(γ, 1n) is due to a very large (46%) underestimation of the number of multiplicity 1 neutrons in the σ(γ, S n) cross-section rather than to unjustifiably associating extra neutrons of multiplicity 2 with the (γ, 2n) reaction.
In figures 4 and 5 one can see that there is noticeable difference between σ(γ, 1n) obtained at Livermore and Saclay already at energies below the threshold B2n of σ(γ, 2n). It means that not only uncertainties of neutron multiplicity sorting exist. The differences ∆σ between the evaluated and experimental cross sections (Fig. 5a) look  as "reflected in a mirror". For both reactions ∆σ is about 3 -5 %. That means that the main reasons for the systematic uncertainties are shortcomings of sorting a certain number of neutrons between 1n and 2n channels. For Livermore [23] data the situation is completely different. The differences ∆σ are significantly large (Fig. 5b) in comparison with those for Saclay data (Fig. 5a) and both not naturally are positive. The ratios σ int eval /σ int exp of integrated cross sections obtained up to energy E int = 26.2 MeV for various reactions are presented in Table 3. In analogy to the case for 181 Ta the larger the fraction of the σ(γ, 1n) in the cross sectiona for the total reactions, the higher the degree to which the latter is underestimated (1.27 → 1.30 →1.34). But for σ(γ, 2n) σ int eval /σ int L = 1.15 is noticeably larger than the correspondent value for 181 Ta (1.05). It means that in the case of 75 As many neutrons were lost in both (γ, 1n) and (γ, 2n) reactions.
Because the ratios between Saclay and Livermore data for 75 As are approximately the same (1.21-1.22) for all reactions, it was supposed that the disagreements could be because of errors in normalization. All Livermore [23] cross sections were multiplied by 1.22. The normalized σ exp (γ, S n) become very close to that of [24]. But the differences ∆σ are significantly different again ( Figure 5). ∆σ(γ, 1n) and ∆σ(γ, 2n) look absolutely different in comparison with previous once (Fig. 5b). Both looked as "reflected in a mirror", but the values were noticeably large: ∆σ(γ, 1n) ∼ 12 mb, ∆σ(γ, 2n) ∼ 7 mb. In figure 5 lines present the calculated [16,17] σ(γ, 1n1p). The sharing of nuclear excitation energy between neutron and proton is similar to that for two neutrons in the reaction (γ, 2n) but the multiplicity of outgoing neutron is 1 not 2. So one is forced to conclude that experimental data [23] are really unreliable because of significant systematic uncertainties of three types: i) unreliable sorting of neutrons between (γ, 1n) and (γ, 2n) reactions, ii) unreliable sorting of neutrons between (γ, 1n1p) and (γ, 2n) reactions, iii) the lost of many neutrons from both (γ, 1n) and (γ, 2n) reactions.