**2. SDF in the 235U(n,f) PFNS Measurements**

#### *2.1. Time Resolution and Bin Correction*

The PFNS is usually measured with Time of Flight (TOF) in Direct Beam (DB) experiments or in Pulsed Beam (PB) experiments. The DB means that these experiments were realized with a direct neutron beam, and a "stop" signal was produced from fission fragments. The PB experiments used a pulsed neutron beam for TOF.

The primary experimental data (counts registered by neutron detector) are collected at the time interval between the "start" event of a neutron detector and the "stop" event of a Fission Fragment (FF) detector. The PFNS is obtained after the transformation of a number of events from the time scale to the energy scale.

Several parameters and the accuracy of their determination are important for this type of experiment. We should examine additional measurements of the flight path, the time channel width, the neutron detector efficiency, possible shift of the timing position as a function of the neutron energy, the efficiency of the FF detector. In the case of the DB or the spontaneous fission measurements, the measured spectra should be corrected for the random coincidence of FF and neutron registration events.

We compare two experiments for PFNS measurements at 235U fission by thermal neutrons performed about 25 years apart. The most important parameters of the old DB experiment (1983) [3,4] and the new experiment (2008) [5,6] are presented in Table 1. The 252Cf(sf) PFNS was used as the standard for the determination of neutron detector efficiency in both cases.

The better time resolution for 235U in the experiment [5,6] is explained by a larger distance between the cathode and anode in the uranium section of the chamber (fission fragment detector) and, as a result, a higher amplitude of the FF pulse.

The uncertainties for channel width and flight path are similar, about 0.1%. In the 1983 experiment, the time resolution was better by a factor of about 2. Correction at finite time resolution and channel width (bin) was achieved using the Maxwellian shape of the spectrum and parameters from Table 1 (1983 data). The influence of these corrections at the PFNS was simulated with Monte Carlo calculations.


**Table 1.** Parameters of the different experiments. Full Width at Half Maximum (FWHM) is the time resolution.

The PFNS for 252Cf(SCf(E)) and 235U(SU(E)) were measured together in the same experiment. The influence of these corrections at the measured ratio of PFNS R(E) = SCf(E)/ SU(E) is very small, and we can conclude that we do not have any SDF connected with this part of data reduction if we measure relative to the 252Cf standard.

The analyses in [7] confirm this conclusion. Average energies of 235U PFNS estimated for these two experiments are very close <E> = 1.976 ± 0.002 MeV [3,4], <E> = 1.982 ± 0.004 MeV [5,6]. The comparison of spectra shown in Figure 1 is given as ratios to the spectrum predicted by the Scale Method [7]. The small difference of 0.006 MeV may be related to the SDF in the neutron angular-energy distribution relative to FF, which will be discussed in the following sections.

**Figure 1.** Ratio of the 235U PFNS for thermal neutrons [3–5] to the fitted SM function [7].
