Elsevier

Annals of Nuclear Energy

Volume 59, September 2013, Pages 126-138
Annals of Nuclear Energy

The experimental determination and evaluation of the spectral indices of the IPEN/MB-01 reactor for the IRPhE project

https://doi.org/10.1016/j.anucene.2013.03.049Get rights and content

Highlights

  • A new experimental method for the spectral index measurements was developed.

  • The method eliminates the need of calculated correction factors for the case of 28ρ.

  • The experimental results were approved for inclusion in the IRPhE handbook.

  • MCNP5 with ENDF/B-VII.0 described well the experimental data.

Abstract

New experimental results for the spectral indices of the IPEN/MB-01 reactor are presented in this work. The experimental approach considers a new technique for the 28ρ case which does not require any sort of calculated correction factors. This aspect gave to the IPEN/MB-01 experiment an excellent quality and free of possible bias due to these calculated correction factors. The uncertainty analysis show, even considering the uncertainties of the geometric and material data of the facility, that the final uncertainties are small enough and well understood for a benchmark problem. The experiment was evaluated and included in the IRPhE handbook. The theory/experiment comparison reveals that there is considerable progress in the 238U nuclear data for application in thermal reactors. The theory/experiment comparison employs the nuclear data libraries ENDF/B-VII.0, ENDF/B-VI.8, JEF3.1, and JENDL3.3 and reveals an excellent agreement for the spectral index 28ρ* independently of the nuclear data library considered; aspect never found before in several other comparisons. The long term overprediction of the 238U neutron epithermal capture appears to be eliminated with the improvements in the recent nuclear data libraries. The experimental performed at the IPEN/MB-01 reactor supports the changes in the 238U nuclear data incorporated in ENDF/B-VII.0 as well as in the other libraries studied in this work. The theory/experiment comparison of 25δ* and (C8/F)ept show that these spectral indices are in general slightly overpredicted, thus suggesting that the thermal fission cross section of 235U might be a little bit underestimated. The overall analysis of the theory/experiment comparisons show the excellent applicability of ENDF/B-VII.0 for thermal reactors fuelled with slightly enriched uranium.

Introduction

Experiments involving determination of the reaction rates in the fuel pellets are of fundamental importance to correlate theory and experiment mainly concerning mathematical methods and related nuclear data libraries. These experiments are normally performed through the irradiation of bare and Cadmium covered fertile and/or fissile foils. Typical examples are the spectral indices 28ρ and 25δ which provide the ratio of the epithermal to thermal neutron captures in 238U and the ratio of the epithermal to thermal fission in 235U respectively. Highly enriched foils are used for the measurements of 25δ and depleted uranium foils are used for 28ρ. The method basically consists in the determination of the Cadmium ratio and the transformation of the perturbed system into a non-perturbed by means of calculated correction factors (Bitelli, 2001, Sher and Fiarman, 1976). Furthermore, the corrections are also applied to transform the thermal cutoff to 0.625 eV (Sher and Fiarman, 1976). Two major problems occur in the measurements of these spectral indices. The first is the maintenance of same reactor power in the irradiation of the bare and Cadmium covered foils. Usually they are performed in two distinct irradiations. The second problem refers to the introduction of the calculated correction factors. Their uncertainties and the validation of the applied calculation methods are extremely complex and do not have any experimental support for that. The most famous spectral indices measurements for thermal reactor applications are the ones performed in the TRX and BAPL critical facilities selected by the Cross Section Evaluation Working Group (CSEWG, 1974) as benchmarks.

Table 1 (Yudkevich et al., 1994) shows the status of art related to the spectral indices. It can be noted that the discrepancies between theory and experiment are as high as 25% for the NORA reactor. Currently the level of discrepancy runs from −2.0% to +3.0% for the critical facilities TRX (Rahman et al., 2004), BAPL (Rahman et al., 2004), MISTRAL (Courcelle et al., 2006), and ERASME (Courcelle et al., 2006). However, the experimental uncertainties are over 1.5% which makes difficult the comparison between theory and experiment. Besides that, the reported uncertainties do not take into consideration the uncertainties due to the geometric and material data of the facility. A clear aspect seen in Table 1 is that the spectral index 28ρ is systematically overpredicted in practically all experiments reported. This overprediction in the spectral index 28ρ has been historically mostly credited to the overprediction of the epithermal cross section of 238U. A direct consequence of this effect observed in several facilities was the underprediction of keff in several critical experiments.

The purpose of this work is to show the experimental determination and evaluation of several spectral indices of the core of the IPEN/MB-01 reactor. The spectral indices considered in this work are the ones based on the ratio of epithermal to thermal reaction rates and on the ratio of specific reaction rates. The first category will consider the very classical spectral indices 28ρ and 25δ. The second category will consider the ratios of the neutron capture rates in 238U to the total fission rates (C8/F). The proposed method is based on a fuel rod gamma scanning technique. The apparatus available at the IPEN/MB-01 facility has an opening collimator of 1.0 cm. These experiments were mainly proposed to fulfill a specific need to verify the new 238U nuclear data with spectral indices that does not require any sort of calculated correction factors. The final whole set of experimental data was recently evaluated and approved as benchmark data for inclusion in the IRPhE Handbook (Dos Santos et al., 2012).

The IPEN/MB-01 reactor is a zero power-critical facility especially designed for measurement of a wide variety of reactor physics parameters to be used as a benchmark experimental data for checking the calculation methodologies and related nuclear data libraries commonly used in the field of reactor physics. This facility consists of an array of 28 × 26 UO2 fuel rods, 4.3% enriched and clad by stainless steel (type 304) inside of a light water tank. A complete description of the IPEN/MB-01 reactor may be found elsewhere (Dos Santos et al., 2004).

Section snippets

Geometry of the experiment configuration and measurement procedure

The experiments employed the standard 28 × 26-fuel-rod configuration as shown in Fig. 1. Experimental fuel rods with the same description as the IPEN/MB-01 core fuel rods were used throughout the measurements. The reactivity was controlled by two control banks (BC1 and BC2 in Fig. 1) of Ag–In–Cd alloy. The safety banks of B4C are kept at its completely removal position of 135% (the absorber is at 35% of the active core length above the active core) during the whole set of measurements. Therefore,

Method to measure the spectral indices

The essence of the proposed method to measure the spectral indices starts with the works of Nakajima (Nakajima et al., 1994a, Nakajima et al., 1994b). According to Nakajima the 238U capture rate (C8) and total fission rate (F) inferred from the scanning detector countings are given respectively by Eqs. (1), (2) as:C8=λU9-λNpλU9·λNp·NNpfγNp·gNp·ηNp·[1-exp(-λNptI)]·exp(-λNptE)·[1-exp(-λNptC)],F=λCe·NCeY¯Ce·fγCe·gCe·ηCe·[1-exp(-λCetI)]·exp(-λCetW)·[1-exp(-λCetC)],where λi is the decay constant of

Experimental results

The first part of the experimental process was to calibrate both the detector systems (disk and fuel rod) employing 152Eu and 133Ba standard sources. This calibration was accomplished in a straightforward fashion and the uncertainty analysis was performed employing the procedure described in the previous section. The quantities of interest for this work are presented in Table 2.

The basic experimental quantity measured in this work is the gamma energy spectra emitted by the fission products,

Effect of parameter uncertainties on the spectral indices

All parameter uncertainties and how they were derived are described in LEU-COMP-THERM-077 (Dos Santos et al., 2004). Here only the propagation of these uncertainties to the spectral indices is considered. The approach follows basically the same one used in LEU-COMP-THERM-077.

The uncertainty due to the geometrical and material composition data was obtained in the companion HAMMER-TECHNION/ CITATION codes. HAMMER-TECHNION (Barhen et al., 1978) is used for the few-group cross-section generation

Theoretical analysis of spectral index calculations

The theoretical analysis of the spectral index experiments realized at the IPEN/MB-01 research reactor facility was performed employing the Monte Carlo MCNP-5 (X-5 Monte Carlo Team, 2003). The nuclear data libraries used for this analysis were ENDF/B-VII.0 (Oblozinsky and Herman, 2006), ENDF/B-VI.8 (ENDF/B-VI Summary Documentation, 2000), JEF3.1 (NEA Data Bank, 2006), and JENDL3.3 (Shibata et al., 2002).

The quantities to be calculated are expressed mathematically as:RRx=Σx(r,E)ϕ(r,Ω,E)drdΩ

Conclusions

The experiment performed at the IPEN/MB-01 reactor was successfully designed, executed and evaluated. The experiment and the corresponding results are well documented and with uncertainties small enough and well understood suitable for a benchmark problem. The complete evaluation reported in the IRPhE handbook has shown to be very useful to test and verify the current and new versions of the nuclear data libraries for thermal reactor application. Specifically for the 28ρ* case, the experimental

References (28)

  • M. Rahman et al.

    Validation study of the cell code WIMS-D and a 69 group library based on JENDL-3.2

    Ann. Nucl. Energy

    (2004)
  • Barhen, J., Rhotenstein, W., Taviv, E., 1978. The Hammer Code System. Technion-Israel Institute of Technology, Haifa,...
  • P.R. Bevington

    Data Reduction and Error Analysis for the Physical Sciences

    (1969)
  • Bitelli, U.d’U., 2001. Medida de Parâmetros Integrais no Reator IPEN/MB-01. Tese de Doutorado –...
  • Bitelli, U.d’U., Dos Santos, A., 2002. The Spectral Indices of the IPEN/MB-01 Reactor: Measurements and Calculations....
  • Briggs, J.B. (Ed.), 2012. International Handbook of Evaluated Criticality Safety Benchmark Project. NEA/NSC/DOC (95)03,...
  • Courcelle, A. et al., 2006. Nuclear Data for Improved LEU-LWR Reactivity Predictions. OECD Nuclear Energy Agency,...
  • Cross Section Evaluation Working Group Benchmark Specification, 1974. ENDF-202/BNL-19302. Brookhaven National...
  • Dos Santos, A. et al., 2012. IPEN (MB01)-LWR-CRIT-SPEC-REAC-COEF-KIN-RRATE-POWDIS-001: Reactor Physics Experiments in...
  • Dos Santos, A., Fanaro, L.C.C.B., Yamaguchi, M., Jerez, R., Andrade e Silva, G.S., Siqueira, P.T.D., Abe, A.Y., Fuga,...
  • ENDF/B-VI Summary Documentation, 2000. In: Rose, P.F. (Ed.), BNL-NCS-17451 (ENDF-201), 4th ed. (ENDF/B-VI). National...
  • Fowler, T.B., Vondy, D.R., Cunninghan, G.W., 1971. Nuclear Reactor Core Analysis Code: CITATION. ORNL-2496, Version...
  • Hardy Jr., J., 1983. ENDF/B Data Testing results for Thermal Reactor Benchmark. Thermal Reactor Benchmark Calculation,...
  • Koppel, J.U., Houston, D.H., 1978. Reference Manual for ENDF Thermal Neutron Scattering Data. Report General Atomics:...
  • Cited by (2)

    View full text