Reactor Dosimetry in the 21st Century

Foreword.

Organising Committee.

Contents.

The key component of WWER is the Reactor Pressure Vessel (RPV). The evaluation and prognosis of RPV material embrittlement and the allowable period of its safe operation are performed on the basis of impact test results on irradiated surveillance specimens (SS). The main problem is that the SS irradiation conditions (temperature of SS, neutron flux and neutron spectrum) have not been determined yet with the necessary accuracy. These conditions could differ from the actual RPV component conditions. In particular, the key issue is the possible difference between the irradiation temperature of the SS and the actual RPV temperature. It is recognized that the direct measurement of temperature by thermocouples during reactor operation is the only way for receiving reliable information. However, this is a complex engineering task. In addition, the neutron field’s parameters for surveillance specimens have not been determined yet with the necessary accuracy, and these (neutron flux and spectrum) are different from the RPV irradiation parameters. The use of state of the art dosimeters (such as the AMES common reference dosimeters) can provide high accuracy in the determination of the neutron exposure level.

In order to solve the above-mentioned problems, a joint project (COBRA project) is being carried out in the frame of the European Programme Copernicus 2, with the participation of organizations and researchers of the European Union, the CEECs and the NIS. In this project surveillance capsules were manufactured which contain state of art dosimeters and temperature monitors (melting alloys). In addition, thermocouples were installed throughout the instrumentation channels of the vessel head to measure directly the irradiation temperature in the surveillance position during reactor operation. The selected reactor was the Unit 3 of Kola NPP situated in the arctic area of Russia. Irradiation of the capsules and online temperature measurements started in September 2001.

In this paper preliminary project results of the irradiation temperature measurements with thermocouples will be presented, together with a description of the experimental set up and the equipment developed for the project, including state of the art dosimeters.

The role and problems of neutron dosimetry have been analysed with regard to RPV embrittlement and lifetime prolongation. The status of Russian regulatory documents concerning neutron dosimetry has been considered. The ways of neutron dosimetry improvement have been discussed and recommended.

In order to improve the assessment of the post-annealing neutron exposure an ex-vessel (cavity) irradiation was carried out during cycle 22 (1998-99) at unit 1 of the Loviisa NPP (type VVER 440) in Finland. A dosimeter holder rack was designed and built by Škoda JS a.s., Czech Republic. Dosimeter sets were provided by VTT, Škoda and JRC/NRG Petten, the Netherlands. Some of the dosimeters were counted at Škoda and NRG Petten in addition to VTT. The large number of reactions used provided a good basis for spectrum adjustment. The results of the intercomparison are presented as well as a critical assessment of the materials and methods used. A similar irradiation is planned for Loviisa unit 2 during 2002-03.

The neutron fluence associated with each material in the pressure vessel beltline region is determined on a plant specific basis at each surveillance capsule withdrawal. Based on an assumed mode of operation, fluence projections to account for future operation are then made for use in vessel integrity evaluations. The applicability of these assumed projections is normally verified and updated, if necessary, at each subsequent surveillance capsule withdrawal. However, following the last scheduled withdrawal of a surveillance capsule, there is generally no formal mechanism in place to assure that fluence projections for the remainder of plant operating lifetime remain valid. This paper provides a review of a methodology that can be efficiently used in conjunction with future fuel loading patterns or on-line core power distribution monitoring systems to track the actual fluence accrued by each of the pressure vessel beltline materials in the operating period following the last capsule withdrawal.

Neutron fluences can be measured and radiation damage parameters determined by analyzing the neutron reaction products in very small samples removed from components of an operating power or research reactor. This process, known as retrospective reactor dosimetry, provides precise neutron exposure parameters for establishing or validating calculations of neutron fluences, helium generation, and radiation damage to reactor materials. Correlation of the neutron fluence and helium data helps to establish and validate models of radiation damage and helium production that are needed to address important issues such as irradiation assisted stress corrosion cracking, void swelling, and weld repair of cracks. Results are presented for samples recently obtained from several operating reactors.

International Reactor Innovative and Secure (IRIS) is a medium-power (~300 MWe) advanced light water reactor that features an integral primary system configuration to enhance safety. Steam generators are located inside the pressure vessel above the core, forming a thick (~1.68 m) annular region, that extends into an equally thick downcomer surrounding the core. As a result, neutron fluence at the pressure vessel and in the cavity is reduced by 5-6 orders of magnitude relative to present loop-type Pressurized Water Reactors (PWRs). Reduction of the RPV fluence eliminates embrittlement concerns, but introduces new challenges for the ex-core flux monitors. This paper proposes using advanced flux monitors, such as SiC semiconductor neutron detectors, and examines their optimum placement in the downcomer region. Furthermore, the requirements on neutron dosimetry/monitors considered for the IRIS-reactor are common to Generation-IV Integral Primary System Reactors (IPSRs).

Calculations of fast neutron fluence to structures in boiling water reactor plant geometries have been performed using two-dimensional transport and a synthesis method. The calculations have been benchmarked using NRC recommended standard benchmarks and using measured data from operating BWR nuclear power plants. Comparisons of the calculations with measurements indicate that the calculations produce fluence estimates with acceptable accuracy.

Three dimensional (3D) neutron and gamma calculations for structural materials inside the reactor vessel in a commercial PWR were performed using the 3D transport code TORT and the Monte Carlo code MCNP to assess the accuracy of calculations using these codes and libraries. Comparisons with two dimensional DORT calculations with various libraries and surveillance dosimetry measurements indicated that TORT and MCNP calculations give similar agreements with surveillance measurements to DORT calculations. Influences of the cross section data, ENDF/B-IV, ENDF/B-VI and JENDL3.2 on attenuation of the fast flux and dpa rate in the reactor vessel, relative contributions of gamma-rays and thermal neutrons to dpa were discussed.

To estimate the applicability of the TORT code, a benchmark calculation was performed using the measured neutron flux in a 375MWe BWR in Switzerland. The calculated neutron flux was compared with the measured neutron fluxes at 27 locations between the shroud and the RPV. The reaction rates of thermal and fast dosimeters calculated by TORT agreed well with the measured data.

As a next step, the TORT code was applied to estimate the neutron flux distribution in Japanese 800MWe BWR plants and compared with the measured radioactivity of a few pieces of the top guide beam, shroud and in-core monitor guide tube. Because a reasonable C/M value was obtained, we conclude that we can obtain reasonable neutron distribution profiles with TORT.

In preparation for the eventual decommissioning of the High Flux Beam Reactor after the permanent removal of its fuel elements from the Brookhaven National Laboratory, both measurements and calculations of the decay gamma-ray dose rate have been performed for the reactor pressure vessel and vessel internal structures which included the upper and lower thermal shields, the Transition Plate, and the Control Rod blades. The measurements were made using Red Perspex™ polymethyl methacrylate high-level film dosimeters, a Radcal “peanut” ion chamber, and Eberline's high-range ion chamber. To compare with measured gamma-ray dose rates, the Monte Carlo MCNP code and geometric progressive MicroShield code were used to model the gamma-ray transport and dose buildup.

For a quantitative identification of radiation damage in materials under neutron irradiation, a new computer code, NPRIM, has been developed to calculate several important quantities such as dpa rate and helium production. Various neutron spectra of typical neutron fields in power reactors, material testing reactors and fusion reactors are bundled in the code. Neutron cross section sets can be selected from the JENDL3.3, ASTM Standard E693 and ENDF/B-V of the SPECTER code. The information from the code provides important correlation parameters and indices to estimate material degradation and plant life in power reactors.

The damage energy analysis of Fe and W irradiated by 3 GeV protons is reported. The subcascade formation energies for Fe and W are estimated to be 10 keV and 100 keV, respectively using binary collision approximation while systematically varying the energy of the incident atom. The result calculated by Dementyev is used for the number of isotopes and their average recoil energy formed by 3 GeV proton irradiation in Fe and W. The damage energy is obtained using the LSS and the Robinson formulations, and the number of subcascades is estimated by the Kinchin-Pease model. The recoil of heavy isotopes gives large damage energy and forms a lot of Frenkel pairs and subcascades. The total number of subcascades formed by a 3 GeV proton is 1.0 in Fe and 0.55 in W. Two-thirds of the subcascades are formed by the recoil of isotopes heavier than V in Fe and than Sm in W.

A new methodology is developed for the prediction of RPV embrittlement that utilizes a combination of domain models and nonlinear estimators including neural networks and nearest neighbor regressions. The Power Reactor Embrittlement Database is used in this study. The results from newly developed nearest neighbor projective fuser indicate that the combined embrittlement predictor achieved about 67.3% and 52.4% reductions in the uncertainties for General Electric Boiling Water Reactor plate and weld data compared to Regulatory Guide 1.99, Revision 2, respectively. The implications of irradiation temperature effects to the development of radiation embrittlement models are then discussed.

Several Russian WWER units are to be removed from service in the near future. To study the main calculation problems concerned with decommissioning, the typical WWER-440 unit was selected. The 1D & 2D models of a core, vessel and shielding were designed to apply in transport and inventory calculations. The 2D KASKAD code based on the discrete ordinates technique was applied in criticality and transport calculations. To confirm the results at the mid-plane, the 1D ROZ-6 discrete ordinates code was used as well as the MCNP Monte-Carlo code. The most important inventory calculations were performed with the ORIGEN-S code.

The Heavy Water Neutron Irradiation Facility of the Kyoto University Reactor can supply neutron energy spectra from almost pure thermal to mainly epi-thermal, using a spectrum shifter and thermal neutron filters. We will report about the measurement of the neutron and gamma-ray doses using a twin-chamber. The used twin-chamber is the combination of a tissue-equivalent ionization-chamber and a graphite ionization chamber, with detecting volumes of 80 cc. From the comparisons between the chamber-measured dose rates and the nominal values, it was confirmed that the relative dependencies of the neutron and gamma-ray doses on the heavy water thickness, were almost the same, excepting the smaller heavy-water-thickness mode, such as CB-0000-F.

In order to get useful information about neutron energy spectrum and neutron dose, a versatile and accurate reactor model of the Kinki University Reactor (UTR-KINKI) was developed under the three-dimensional continuous-energy MCNP Monte Carlo code. The agreement between MCNP predictions and the experimentally determined values was very good. This paper describes characteristics of neutron fields at the Kinki University Reactor calculated with the present MCNP model of the UTR-KINKI. From the results obtained it was clear that these neutron fields are applicable to development and performance evaluation of personnel dosimeters and experimental studies on biological effects of low levels of radiation.

This paper presents the results of the evaluation of the gamma-field parameters in the KORPUS facility capsules and irradiation channel #9 of the RBT-6 reactor. These results were obtained using both calculation and experimental techniques. The paper describes the technique for measurement of energy release values by means of miniature detectors of calorimetric type (gamma-thermometers). The comparison of measured values to calculation results is given.

The neutron and photon fields present at the Fast Neutron Beam of RPI were simulated with MCNP-4C and measured with activation foils, TLDs and ionisation chambers. In general, there is a good agreement between calculations and measurements, although the model overestimates the thermal neutron component. Aluminum oxide TLDs were found to be promising for monitoring the photon dose in actual irradiations of circuits.

A helium accumulation fluence monitor (HAFM) has been developed at JOYO for precise and reliable neutron dosimetry. Calibration tests using helium ion implanted samples showed that the HAFM measurement system can measure the number of helium atoms with an uncertainty of about 5%. To verify the measurement accuracy of the neutron fluence, the HAFMs were irradiated in YAYOI. It was found that the HAFM measurement system can measure neutron fluence with an uncertainty of about 7%. The HAFMs were then irradiated in JOYO and the neutron fluence was measured. The measured results of enriched boron type HAFMs are in good agreement with those from the existing foil activation method within the experimental uncertainty and the reliability of the HAFM method was verified. It was confirmed that the HAFM could be applicable for fast reactor dosimetry.

Solid state track recorders (SSTR) were deployed in two high level waste storage tanks and were exposed to a neutron field resulting from an unknown quantity of actinide nuclides producing neutrons via spontaneous fission and (α,n) reactions with light nuclei present in the tank. Discrete ordinates transport theory calculations of the tank and its contents were used to infer the amount of radioactive material present.

A liquid scintillation spectrometer whose detector was composed of a NE213 liquid scintillator covered with a bismuth shield was constructed to estimate energy spectra of low intensity and relatively low energy neutrons around a nuclear reactor. A response matrix used for unfolding the pulse-height distributions was obtained from the matrices representing modulations of neutron energy spectra by the bismuth shield and response functions of the liquid scintillation detector. Measurements of reactor neutrons were carried out with the aid of the pulse-shape discrimination method in the vicinity of the Kinki University Reactor, UTR-KINKI. It became clear from the experiments that the pulse-height distributions by fission neutrons could be successfully obtained by the scintillation spectrometer constructed here in spite of the measurements in the low n/γratio fields around nuclear facility and neutron energy spectra could be determined by the unfolding technique with the response matrix in the energy range from about 1 MeV to tens of MeV.

Based on the experimental observation of the OH vibration band growth in neutron-irradiated polymer coated fibres, we present a feasibility study of a fast neutron optical fibre detector. The principle of the system is explained and the design parameters are identified. Potential applications of the system are briefly discussed.

In order to measure the general spatial distribution of the thermal neutron fluence during the so called “weak” irradiation (less than 10¹⁷ n/m²) of HTGR nuclear fuel for subsequent high temperature tests including fission products release, we apply local (0.3 cm rings) and distributed (long rods up to 65 cm) accumulative detectors of neutrons and gamma with results' reading by the electron spin resonance method (ESR-sensors). Sensors materials are: silicate ceramic (glass) containing B₂O₃ (neutron sensor) and quartz with Al₂O₃ addition (gamma sensor). The new possibilities of nontraditional ESR-sensors, a new type of nuclear radiation detectors are discussed.

A comparison has been performed between the Co(Cd) method and the Co/Ag method for the determination of the thermal and epithermal neutron flux. Significant differences between the two methods have been observed. It is concluded that especially the results for the epithermal flux should be considered with care.

A multiparameter spectrometry system for neutron and gamma spectra measurement is described with the organic scintillator stilbene or NE-213 scintillator. The control logic has been realized with the Field Programmable Gate Array. The spectrometer was tested at the Nuclear Research Institute Rez. Measurements in the VVER - 1000 type reactor pressure vessel dosimetry benchmark in the LR-0 experimental reactor have been performed.

The development of MOSFET devices towards reactor dosimetry is described. The feasibility of large-geometry MOSFET devices has been assessed for both active and passive neutron dosimetry. Neutron activation has been observed in these devices. A rationale for its minimisation has been prepared to exclude susceptible materials via novel packaging of single devices including Chip-on-Board and Flip-Chip methods. Both single and stacked MOSFET architectures have been exposed to γ-rays (⁶⁰Co) and neutrons (²⁵²Cf) to provide neutron sensitivity estimates for these devices. That of the former has been enhanced by hydrogenous material.

The High Temperature Engineering Test Reactor (HTTR) can provide very large spaces at high temperatures for various irradiation tests. The first test rig, the I-I type irradiation equipment, was developed for an in-pile creep test on a stainless steel with standard size specimens, not miniature ones generally used at other testing reactors. Prior to the creep test, the irradiation conditions are to be measured by the rig as the first irradiation test. Self-Powered Neutron Detectors (SPNDs), neutron fluence monitors, thermocouples and melting temperature monitors are used for the dosimetry. The dosimetry plan and the subsequent data assessment procedure are described in this paper.

A set of ten Self-Powered Neutron Detectors with Co, Rh and Ag emitters has been irradiated in several channels of the BR2 research reactor at SCK•CEN aiming at a comparison of their performance as thermal neutron flux detectors under various conditions. To allow for a correct interpretation of their signals, all detector sensitivity contributions (prompt and delayed) were calculated using a dedicated Monte Carlo model. The various contributions were also measured separately; the agreement between calculated and experimental data, including data from activation dosimetry, was excellent. Detailed neutron flux profiles were obtained from the SPND data, after correction for the finite detector lengths and for the slow response of delayed SPNDs.

The paper presents the peculiar features of the neutron fluence evaluation on the vessel steel specimens irradiated in the capsules of the KORPUS facility of the RBT-6 reactor. The conditions were formulated for the joint location of material science specimens and neutron detectors for metrological support. On the basis of the MCU code a calculation model was developed and the neutron spectra and attenuation value of fast neutrons of different threshold energy inside the steel bulk up to ~200mm thick were calculated. A good correlation between the calculated and experimental data is noted.

Detailed time-history calculations of the radiation environment (neutron exposure and gamma ray heat generation rates) experienced by the baffle-former bolts at two operating nuclear power plants were calculated. This data was, in turn, used to calculate the operating temperature history of the bolts. The radiation and thermal environment data fed into the interpretation of the hot cell testing of the removed bolts.

A three-dimensional computational method based on Monte Carlo radiation transport techniques was developed to calculate thermal and fast neutron fields in the downcomer region of a Boiling Water Reactor (BWR). This methodology was validated using measured data obtained from an operating BWR. The helium production was measured in stainless steel at locations near the shroud and compared with values from the Monte Carlo calculations. The methodology produced results that were in agreement with measurements, thereby providing a useful tool for the determination of helium levels in shroud components.

Fourteen chains containing experimental surveillance material specimens of the VVER 440/213 nuclear power reactor pressure vessels were irradiated in the surveillance channels of the Nuclear Power Plant Dukovany in the Czech Republic. The irradiation periods were one, two or three cycles. For the absolute fluence values evaluation account was taken of the time history of the reactor power and of local changes of the neutron flux along the reactor core height, and of correction factors due to the orientation of monitors with respect to the reactor core centre. Neutron fluence values above 0.5 MeV energy and above 1.0 MeV energy in the container axis at the axial positions of the sample centres and fluence values in the geometric centre of the samples were calculated making use the exponential attenuation model of the incident neutron beam.

Two experimental surveillance assemblies (full copies of standard ones), equipped with the activation monitors ⁹³Nb(n,n'), ⁵⁴Fe(n,p), ⁵⁸Ni(n,p), ⁴⁶Ti(n,p), ⁶³Cu(n,α) were placed in standard positions at Balakovo-1 WWER-1000 and were irradiated during one fuel cycle. The goal of this experiment was to obtain the experimental information for validation of calculation methods, which are used in dosimetry of WWER-1000 surveillance specimens. RRC KI and SCK/CEN delivered the activation monitors, and both laboratories carried out measurements. RRC KI provided the calculation analysis. The monitors were located inside notches of Charpy specimens, in the Aluminum filling plates in surveillance containers, and along the outside surfaces of Charpy specimens (wire detectors). Special measurements and calculations were carried out for the determination of each assembly orientation during irradiation.

This paper reviews the reactor dosimetry program that has supported steel pressure vessel integrity assessments for magnox power plants over the last ten years. The dosimetry program has aimed to achieve consistent:.

• calculated and measured fast and thermal neutron doses.

• data for surveillance specimens and reactor pressure vessels.

Throughout the program, the flux measurements on the plants have been judged essential for any doses where a high degree of confidence is required. The work to support operation is now largely complete and the dosimetry is being extended to assess radioactive inventories as part of the decommissioning process.

Irradiation growth is a dimensional change at constant volume experienced by anisotropic materials such as Zr alloys when exposed to neutron irradiation. Growth rates at 350 K, although smaller than at 550 K, are significant, and models based on migration of point defects have difficulty rationalizing its characteristics. We propose an irradiation growth mechanism based on the direct athermal deposition of point defects onto extended sinks. Self-interstitial atoms are deposited preferentially on crystallographically aligned sinks, either because of the cascade anisotropy in the Zr hcp structure, or because of their highly anisotropic diffusion through the lattice. The model is non-saturating, and the strain rate is linearly proportional to the flux as observed. Preliminary calculations show cascades are anisotropic, which can contribute to the observed growth strains.

This paper deals with the effect of gamma irradiation on light water reactor pressure vessel steel embrittlement, on fission detector dosimetry, and with the accuracy of the neutron/gamma transport calculations. Four Russian and German types of reactors are considered. The neutron/gamma-transport calculations were performed according to the synthesis method using the ANISN and DORT codes with the BUGLE-96 cross-section set and the Monte Carlo code MCNP with the ENDF/B-VI evaluated data library. The calculations of the γ-dpa were performed using different dpa-cross-sections including a new calculation by means of the Monte Carlo code EGS4.

The activations of the reactor internal structures, of the Pressure Vessel, and of the concrete walls of the PWR units Doel 1-2 and Tihange 1, operated by the Belgian utility Electrabel, have been calculated for updating of decommissioning costs. Neutron fluxes and 44 activation reactions have been computed in 224 regions, ranging from core plates and baffle to concrete walls around the reactor cavity and around the reactor pool. The code MCBEND (Answers Service, Serco Assurance, U.K.) has been used in both Monte Carlo theory and diffusion theory with the ADC (Adjusted Diffusion Coefficient) method.

CHOOZ-A, a French-Belgian power plant was the first PWR in operation in France; it was definitively shut down in 1991. A large feedback program has been realized during the decommissioning. In this framework, several core samples have been machined from the pressure vessel. They have been used for the determination of the attenuation of the cobalt and iron reaction rates inside the steel. The first one follows an exponential(x²) type shape. ⁵⁴Fe reaction rate decreases throughout the vessel according to an exponential law with a coefficient of -0.0165 ±0.0011 mm^-1 (for a coverage factor of 2). The results of a sample located near the inner surface are in agreement with previous measurement and calculation.

Clinical trials of Boron Neutron Capture Therapy for patients with malignant brain tumor had been carried out for half a decade, using an epithermal neutron beam at the Brookhaven Medical Reactor. The decision to permanently close this reactor in 2000 cut short the efforts to implement a new conceptual design to optimize this beam in preparation for use with possible new protocols. Details of the conceptual design to produce a higher intensity, more forward-directed neutron beam with less contamination from gamma rays, fast and thermal neutrons are presented here for their potential applicability to other reactor facilities. Monte Carlo calculations were used to predict the flux and absorbed dose produced by the proposed design. The results were benchmarked by the dose rate and flux measurements taken at the facility then in use.

A high fidelity characterization of the ACRR reference neutron field, a water moderated test reactor, is presented. This characterization includes neutron spectrum uncertainty, covariance matrices, and measured activities for 44 reactions. A least-squares spectrum adjustment was done and showed a χ² per degree of freedom of 1.68 when all dosimetry-quality activities were used. This high fidelity spectrum characterization is well documented and is available for testing the consistency of dosimetry-quality cross sections.

The neutron field in the core and reflector regions of the Portuguese Research Reactor (RPI) was simulated with the Monte Carlo code MCNP-4C, using criticality and fixed-source calculations. A comparison of the results obtained with MCNP and the deterministic codes WIMSD-5 and CITATION was performed, showing consistence between them for the effective multiplication constant and the neutron fluxes. The calculations and measurements agree within 15%. Therefore the model can be used to predict the neutron field in the region studied and in other locations and irradiation facilities of the RPI.

In this work, the thermal neutron field at the Heavy Water Neutron Irradiation Facility of the Kyoto University Reactor, which had been used for boron neutron capture therapies for several years, was characterized. The spatial distributions of the thermal neutrons were measured with a small thermal neutron probe that was put on a tip of an optical fiber with a diameter of 1 mm. The probe consisted of ZnS(Ag) and LiF as a neutron converter. The probe was very small so that the detector could measure the spatial distribution with an excellent spatial resolution and without disturbing the thermal neutron field. The results were compared with that obtained with a conventional gold wire activation method and Monte Carlo calculations.

The responses of silicon carbide semiconductor Schottky diodes to several fast-neutron sources have been measured. The pulse height spectra that are produced in the diodes by neutron-induced charged particles contain features that can be related to the characteristics of the incident neutron energy spectra. Calculational methods can be developed to unfold these response spectra to reveal information on the incident neutron energy spectra. Several potential reactor neutron dosimetry and characterization applications can be addressed with these silicon carbide fast neutron spectrometers.

A series of basic experiments for an accelerator driven subcritical reactor (ADSR) has been performed at the Kyoto University Critical Assembly (KUCA) by combining a critical assembly with a Cockcroft-Walton type accelerator in view of a future plan to establish a new neutron source for research. By injecting 14 MeV neutrons into the subcritical assembly, the neutron multiplication and the prompt neutron decay constant were measured mainly by an optical fiber detector system as a function of subcriticality. Through the present study, it was strongly recognized that the present tools for the neutronics design calculation are not accurate enough to predict the nuclear parameters in the ADSR.

Three innovative sub-miniature fission chambers (SMFC), designed and manufactured at the Nuclear Measurement Systems Laboratory (LSMN) of CEA/Cadarache, were extensively tested in the BR2 research reactor at SCK•CEN, Mol. We present the experimental results for the (thermal) neutron sensitivity, the gamma-induced signal, the signal due to activation, the current picked up by the signal cable, the global current/voltage characteristics and the long term behaviour up to a thermal neutron fluence of 2.7·10²¹ n/cm². We also compare the data with results from calculations with our FCD computer code. The onset of the saturation domain is well predicted by FCD; the neutron sensitivities can be accounted for perfectly after a refinement of the FCD model.

Certified reference materials are distributed by the European Commission through the BCR^® programme (over 500 CRMs) including a series of activation and fission monitor materials originally proposed by the Euratom Working Group on Reactor Dosimetry. The current range (18 CRMs) includes materials to cover the complete energy spectrum, and suitable for different irradiation times. Fission monitors are ²³⁸UO₂ or ²³⁷NpO₂ in the form of microspheres. Activation monitors are high purity metals (Ni, Cu, Al, Fe, Nb, Rh, or Ti), certified for interfering trace impurities, or dilute aluminium-based alloys. Reference materials newly certified are IRMM-530R A1-0.1%Au, replacing the exhausted IRMM-530 material, used as comparator for k₀- standardisation, and three new Al-Co alloys (0.01, 0.1 and 1.0%Co). Others in the process of certification are A1-0.1%Ag and A1-2%Sc for thermal and epithermal fluence rate measurements and two uranium-doped glass materials intended for dosimetry by the fission-track technique. Various alloy compositions have been prepared for use as melt-wire temperature monitors with melting points ranging from 198 to 327ºC.

In order to assure the reliability and accuracy of neutron flux and related characteristics such as dpa, helium production and fuel power in the irradiation test of JOYO, reactor dosimetry and neutronic calculation have been developed. The detailed calculation was conducted using transport and Monte Carlo codes with the core subassembly composition obtained by three dimensional diffusion theory. Helium Accumulation Fluence Monitor (HAFM) were also used to measure the neutron fluence. The calculation method was verified by the comparison of measured fuel power based on the PIE data and adjusted neutron flux using measured reaction rates. As a result, it was confirmed that the calculation with experimental correction can characterize the JOYO neutron field precisely and meet the specified accuracy set for each irradiation test.

In Monju, shielding measurements were made around the reactor core as a part of the system start-up tests in order to evaluate the design margins of the shielding performance, to demonstrate the validity of the shielding analysis method, and to acquire basic data for use in future FBR design. The measured reaction rates have been obtained radially from the core to the in-vessel storage rack and axially to the reactor vessel upper plenum. The measured values (E) were compared with the calculated values (C) obtained with the FBR shielding analysis system on the basis of the nuclear data library JENDL-3.2. Based upon these results, the design margins around the reactor core have been examined.

Spallation-neutron fluence distributions and integral yields for protons of E_p = 300-590MeV incident on thick Pb/Bi targets immersed in a water bath have been determined. The results constitute a data-base pertinent to a well-defined, representative configuration for practicable, high-intensity spallation neutron sources at these intermediate beam energies. They include neutron-fluence distributions based on Au and Mn reaction rates around a thick Pb/Bi target at E_p = 300, 420 and 590 MeV and total n/p values at 420 and 590 MeV, respectively. The latter agree well with previously-measured values, but are of higher precision. Such a database permits accurate benchmarking and detailed inter-comparison of particle-transport-code predictions.

The WWER-1000 Mock-up investigations were carried out on the LR-0 reactor to support the dosimetry of the special irradiation rigs, which were used in Novo Voronezh 5 WWER-1000 for irradiation of RPV steel specimens. The neutron spectrum was measured by the proton recoil method (using stilbene scintillators and hydrogen proportional counters) inside the baffle channel without irradiation rig, on the RPV inner surface and inside the RPV. The activation and track measurements of various threshold reaction rates were carried out inside the irradiation rig - on the surfaces of the specimen simulators. The spectra adjustment was carried out using measured reaction rates. Experimental data are compared with calculational ones.

The adjustment of neutron-fluence spectra at dosimetry surveillance and vessel positions and of the neutron cross sections (parameters) used to calculate them may be carried out in separate steps, provided the dosimetry and the neutron-transport parameters can be appropriately partitioned. We present the prescriptions for carrying out these steps. The purpose is to allow the use of widely available least-squares adjustment codes as modules of a consistent and comprehensive package for the adjustment of all relevant quantities in reactor dosimetry analyses.

The main cause of difference between point-wise Monte Carlo and multigroup deterministic transport calculations is the multigroup cross sections. Therefore, it is of most interest to generate multigroup libraries that can represent point-wise cross sections effectively. In this paper, we generated cross-section libraries considering different methodologies to identify and reduce the differences between multigroup deterministic and point-wise Monte Carlo calculations. We demonstrated that differences can be reduced by the use of appropriate methodologies for selection of group structure, and cross-section collapsing. We suspect that the remaining difference of ~6-7% may be caused by the differences in scattering treatment and self-shielding.

A new application called SIGUEVIVA has been created to estimate the fast neutron flux and fluence in the vessel and internals of a BWR, based on the real core power distribution. In this paper its development for the Cofrentes NPP is described and the results of comparison with dosimetric results are presented.

The application is based on an extensive set of three-dimensional Monte-Carlo calculations performed with MCNP4B [1] which constitute a three-dimensional library of contribution factors of each fuel assembly to each azimuthal angle in the different radial layers of shroud and vessel. These contribution factors are combined with the assembly-to-assembly power distribution obtained with SIMULATE-3 [2] to obtain the absolute fluxes in all the points of interest.

Adjustment of the neutron fluence at the VVER-1000 RPV inner wall has been carried out. For the purpose of this adjustment the neutron flux response sensitivity to the main parameters of calculation uncertainty has been calculated. The obtained sensitivities, the parameters uncertainty and activity measurement data of iron, copper and niobium detectors positioned behind the RPV of Kozloduy NPP Unit 5 have been used in this adjustment.

The JOYO MK-II core was operated from 1983 to 2000 as fast neutron irradiation bed. In order to meet various requirements for irradiation tests for development of FBRs, the JOYO upgrading project named MK-III program was initiated. The irradiation capability in the MK-III core will be about four times larger than that of the MK-II core. Advanced irradiation test subassemblies such as capsule type subassembly and on-line instrumentation rig are planned. As an innovative reactor safety system, the irradiation test of Self-Actuated Shutdown System (SASS) will be conducted. In order to improve the accuracy of neutron fluence, the core management code system was upgraded, and the Monte Carlo code and Helium Accumulation Fluence Monitor (HAFM) were applied. The MK-III core is planned to achieve initial criticality in July 2003.

Pohang Neutron Facility (PNF) is a pulsed neutron facility based on the 100-MeV electron linear accelerator. It was constructed for nuclear data production in Korea, and it consists of an electron linear accelerator, a water-cooled Ta target with a water moderator and a time-of-flight path with an 11 m length. The 100-MeV electron linac uses a thermionic RF-gun, an alpha magnet, four quadrupole magnets, two SLAC-type accelerating sections, a quadrupole triplet, and a beam-analyzing magnet. It has been equipped with a new four-position sample changer controlled remotely by a CAMAC data acquisition system, which allows the simultaneous accumulation of the neutron time-of-flight spectra from 4 different samples. The neutron total cross sections of natural In and Cu have been measured in the neutron energy range from 0.1 eV to 100 eV by the neutron time-of-flight method.

Multi-pass pebble bed reactor concepts are characterized by circulating fuel systems that cycle the pebbles in and out of the core until the burnup limit is reached. Currently modular designs of such reactors, with nominal powers of approximately 300 MW-thermal, are under consideration for deployment internationally. A concern of the proposed designs is the ability to perform online measurements of the fuel burnup to determine whether a pebble has reached its end-of-life burnup limit (~ 80,000 MWD/MTU). In this work, computational simulations were performed to assess the utilization of a passive gamma ray spectrometric approach to perform this task. However, in addition to using the inherent signatures of the irradiated fuel, the use of the ⁵⁹Co(n,γ)⁶⁰Co reaction as a burnup indicator is considered. The results show that the activity ratio of ¹³⁴Cs/⁶⁰Co can provide an indicator that is accurate to within 5% at burnup greater than 20,000 MWD/MTU as the power is varied between 50% and 200% of the reactor's thermal power.

Thermoluminescence Dosimeter (TLD) distributions are usually assumed to be normally distributed. Analysis of large TLD data sets shows this assumption to be false, with distributions that exhibit significant low-side tailing. The arithmetic mean therefore may not be the best estimator for the determination of absorbed dose from these measurements. This work suggests that the use of simple robust estimators may substantially reduce measurement uncertainties.

The main inconsistency between the results based on the multigroup approximation and those based on continuous energy representation of neutron cross sections is obtained for the VVER-1000 downcomer. The attenuation of the neutron flux through the downcomer calculated with continuous energy representation is about 10% less than the one calculated with multigroup representation. The application of Legendre expansions with higher order up to P7S16 has improved the agreement but not very effectively. The good consistency of results for the attenuation through the VVER-1000 RPV shows that the application of multigroup neutron cross sections representation for iron does not make notable impact on the neutron flux attenuation results in comparison with those obtained by continuous representation.

The Brookhaven National Laboratory Sigma Pile is a Radium-Beryllium neutron source imbedded in a cube of graphite blocks. The pile is approximately 2.13 m on four sides and is 3.07 m high. Thermo-luminescent dosimeters were used to determine the neutron and gamma-ray dose rates in the pile. Gamma-ray dose rate measurements have also been made in the air outside of the pile, while the Radium-Beryllium neutron source was being withdrawn from the pile. The Monte Carlo code has been used to calculate the coupled neutron-photon transport. Measured dose rates at various locations agreed with the calculated values within 5% to 15%.

This paper describes the results of analyzing the VENUS-1 Benchmark in X, Y, Z and R, Θ, Z geometry using TORT, the BUGLE-96 transport cross-section library, and the SNLRML neutron dosimetry cross-section library. The three-dimensional TORT calculations show significant improvement over prior two-dimensional calculations in comparison to measurements for both neutron dosimetry and gamma ray dose rates.

Neutron Excitation Function Guide for Reactor Dosimetry (NEFGRD) contains the graphic and text information for 56 nuclides (81 dosimetry reactions). The graphic information is compiled of the dosimetry reaction cross sections as the function of neutron energy taken from all available specialized dosimetry, ENDF and CSISRS libraries. The text information consists of four blocks each of them contains different information about reaction, nuclide, etc. In this paper we cite as an example only one dosimetry reaction - ⁵⁴Fe (n,p)⁵⁴Mn. Total size of NEFGRD is about 28 Mbytes. The data can be retrieved through WEB or obtained on CD-ROM or hard copy as the IAEA report INDC(UKR)-005/EL.

The Nuclear Data Section of the International Atomic Energy Agency started a project in 2001 for updating the old reactor dosimetry file IRDF-90. The data for the new library (entitled IRDF-2002) should derive from tested, up-to-date reactor dosimetry files or evaluations. Two new reactor dosimetry libraries have been published from the time of releasing IRDF-90: JENDL/D-99 (Japan) and RRDF-98 (Russia). As the uncertainties of the cross sections in the new libraries are significantly smaller for a number of reactions, and as, furthermore, several new reactions can be found in these files, as compared with IRDF-90, they can be considered as potential sources for the up-dating procedure. Therefore, the content of these files has been analysed, tested and, intercompared with the corresponding data of IRDF-90. This paper presents the results from this comparison and, gives a proposal concerning the cross sections to be included in IRDF-2002.

The space – energy distribution of the mixed neutron – photon radiation field has been measured over the Reactor Pressure Vessel (RPV) simulator thickness in the WWER-1000 engineering benchmark assembly in the LR-0 experimental reactor (in Nuclear Research Institute Řež plc) with a scintillation spectrometer. The spectra have been measured before the RPV in one quarter, one half, and three quarters of its thickness and behind the RPV in the energy range 0.5 – 10 MeV. The evaluated integral fluxes above 1 MeV and their ratio are compared with the MCNP and DORT calculation, the measured spectra are presented graphically. The measurements are being performed in the frame of the project REDOS [1], 5^th Framwork Programme of the European Community 1998 – 2002.

Iron and water are two main shielding materials used in nuclear reactors, therefore these two materials have been studied from the point of view of neutron and gamma radiation transport. WWER-440 and WWER-1000 mock-up arrangements have been assembled at the LR-0 experimental reactor in NRI Rez plc before with the purpose to study neutron and gamma flux impinging the reactor pressure vessel. In parallel, the neutron and gamma radiation transport have been studied on the simplest spherical geometry, i.e. on benchmark assemblies located in the LR-0 reactor hall.

Iron spheres of 30, 50 and 100 cm in diameter and water spheres of 30 and 50 cm in diameter were used for that benchmark. A Cf-252 neutron source with emission of 3×10⁸ n/s was each time positioned in the centre of a sphere. The leakage neutron and gamma spectra were measured by independent neutron and gamma spectrometers of the different laboratories. Neutron spectra were measured by proportional counters filled with hydrogen and by scintillation detectors of stilbene type in the energy range of 0.01-20 MeV. Gamma spectra were measured by detectors of stilbene, germanium and silicon type, in the energy range of 0.4-12 MeV.

All the important parameters of the experimental assemblies, necessary for the computation (geometry, material composition, neutron source input spectra), were summarised. The calculations of neutron and gamma spectra were performed with the MCNP code using the ENDF/B-VI data library. The intercomparison of measured and calculated data has been done.

In view of the good reproducibility of mixed neutron and gamma (n,g) field parameters and exact data setting for calculation the experimental assembly can be used for the following purposes:.

- testing of spectrometry and dosimetry equipment in mixed n,g fields.

- testing of calculation methods and small-group data libraries for Fe and H₂O.

In addition, upon a careful analysis of the obtained results, the neutron spectrum of the iron sphere with a diameter of 50 cm is proposed for an international standard of the moderated neutron field.

The Laboratory of Studies and Research in External Dosimetry (LRDE) associated to the National Office for Metrology (BNM) has to maintain the traceability of the French references for the calibration of neutron dosimeters. The LRDE owns a facility which provides some conventional neutron spectra from sources of ²⁴¹Am-Be, ²⁵²Cf, and (²⁵²Cf + D₂O)_/Cd recommended by ISO standards. These ISO spectra appear not appropriated to simulate some kind of workplace spectra. In order to have similar radiation conditions between the calibration and the use of the device, LRDE has built facilities (“SIGMA” and “CANEL”) providing some neutron spectra from thermal to fast energies reproducing those encountered in workplaces.

An expert system for preparing an effective mesh distribution for the S_N method has been developed. It consists of two main parts: 1) an algorithm for generating an effective mesh distribution in a serial environment, and 2) an algorithm for selecting an effective domain decomposition strategy for parallel computing. For the first part, the algorithm consists of four steps: creation of a geometric model and coarse meshes, calculation of uncollided flux, selection of differencing scheme, and generation of fine mesh distribution. For the second part, the algorithm accounts for available computing resources, load balance, granularity and degree of coupling among processors. This expert system has been successfully tested for simulating a real-life shielding problem.

Coupled neutron and gamma multigroup (broad-group) libraries used for Light Water Reactor shielding and dosimetry commonly include 47-neutron and 20-gamma groups. These libraries are derived from the 199-neutron, 42-gamma fine-group VITAMIN-B6 library. In this paper, we introduce modifications to the generation procedure of the broad-group libraries. Among these modifications, we show that the fine-group structure and collapsing technique have the largest impact. We demonstrate that a more refined fine-group library and the bi-linear adjoint weighting collapsing technique can improve the accuracy of transport calculation results.

Taking account of the radial source negative gradient in the periphery of reactor core leads to diminishing the evaluation of the neutron fluence onto the reactor vessel in comparison with the calculated one with an assembly-wise source. In the case of VVER-440 in the direction of maximum exposure this diminishing is about 10%. In the case of VVER-1000 the neutron fluence evaluation diminishes by about 20%. The results obtained give a base for reduction of the neutron fluence evaluation without diminishing the conservatism.

In the case of surveillance specimens of VVER-1000/320 taking account of the radial gradient of the neutron source does not make a significant impact on the neutron fluence evaluation.

A study of the influence of the multigroup representation of the neutron cross sections for the purpose of improving the VVER-440 reactor vessel fluence determination was carried out. To assess this influence a comparison of calculational results obtained on the base of the DORT Discrete Ordinates' code and the multigroup cross-section library BGL440, and results obtained by the MCNP Monte Carlo code and the continuous representation of neutron cross sections was carried out. The MCNP results for the fluence onto the vessel, after neutron transmission through the downcomer, in direction of maximum exposure were about 8% higher than the DORT ones. The results for the flux attenuation through the vessel were practically consistent between the two methods.

Monte Carlo calculations were performed for a selected cycle of the Paks NPP Unit II to test a computational model. In the model the source term was calculated by the core design code KARATE and the neutron transport calculations were performed by the MCNP. Different forms of the source specification were examined. The calculated results were compared with measurements and in most cases fairly good agreement was found.

The ability of transport codes ANISN, DORT, ROZ-6, MCNP and TRAMO, as well as nuclear data libraries BUGLE-96, ABBN-93, VITAMIN-B6 and ENDF/B-6 to deliver consistent gamma and neutron flux results was tested in the calculation of a one-dimensional cylindrical model consisting of a homogeneous core and an outer zone with a single material. Model variants with H₂O, Fe, Cr and Ni in the outer zones were investigated. The results are compared with MCNP-ENDF/B-6 results. Discrepancies are discussed. The specified test model is proposed as a computational benchmark for testing calculation codes and data libraries.

In the novel MultiTrans SP₃ radiation transport code the advanced tree multigrid technique is applied to the simplified P₃ (SP₃) transport approximation. The tree multigrid is generated directly from stereolitography (STL) files exported by computer-aided design (CAD) systems, thus allowing an easy interface for construction and upgrading of the geometry. The deterministic MultiTrans code allows fast solution of complicated three-dimensional transport problems in detail, offering a new tool for calculation of quantities of dosimetric interest. In order to determine the feasibility of a new code, computational benchmarks need to be carried out. In this paper, an application of the MultiTrans code to criticality problem is for the first time reported.

The neutron capture cross sections of ¹²⁹I, ¹³³Cs and ¹⁴¹Pr have been measured relative to the ¹⁰B(n,α) standard cross section at energies below several tens of keV by the neutron time-of-flight (TOF) method with an electron linear accelerator (linac) at the Research Reactor Institute, Kyoto University (KURRI). The capture gamma-ray measurements have been made with a pair of C₆D₆ liquid scintillators or twelve Bi₄Ge₃O₁₂ (BGO) scintillators. The relative measurements of ¹²⁹I and ¹⁴¹Pr have been normalized to the reference cross section at 0.0253 eV. The capture cross section of ¹³³Cs has been absolutely obtained with the BGO scintillation assembly as a total energy absorption detector. The results of these cross section measurements have been compared with the existing experimental values and the evaluated data in ENDF/B-VI, JENDL-3.2 and JEF-2.2.

We irradiated 230 and 100 MeV/nucleon Ne, C, He, p and 230 MeV/nucleon Ar ions onto a Cu target to investigate the projectile dependency of induced radioactivities of spallation products. We obtained the variation of the residual activities of nuclides produced in activation samples inserted in the Cu target and the mass-yield distribution of nuclides produced in activation samples put on the surface of the Cu target. The results show that the projectile dependency of the mass-yield distribution and the variation of the residual activities varies with the mass number difference between the sample nuclide and the produced nuclide. These experimental results were compared with the Monte Carlo calculation.

The neutron capture cross sections of Dy isotopes (¹⁶¹Dy, ¹⁶²Dy, ¹⁶³Dy, and ¹⁶⁴Dy) have been measured in the neutron energy range from 10 to 90 keV using the 3-MV Pelletron accelerator of the Research Laboratory for Nuclear Reactors at the Tokyo Institute of Technology. Pulsed keV neutrons were produced from the ⁷Li(p,n)⁷Be reaction by bombarding the lithium target with the 1.5-ns bunched proton beam from the Pelletron accelerator. The incident neutron spectrum on a capture sample was measured by means of a TOF method with a ⁶Li-glass detector. Capture γ-rays were detected with a large anti-Compton NaI(T1) spectrometer, employing a TOF method. A pulse-height weighting technique was applied to observed capture γ-ray pulse-height spectra to derive capture yields. The capture cross sections were obtained by using the standard capture cross sections of ¹⁹⁷Au. The present results were compared with the previous measurements and the evaluated values of ENDF/B-VI.

The neutron capture cross-sections of ¹⁶¹Dy and ¹⁶³Dy have been measured in the energy region from 0.003 eV to several tens of keV by using the neutron time-of-flight (TOF) method with a 46 MeV electron linear accelerator (linac) at the Research Reactor Institute, Kyoto University (KURRI). An assembly of twelve Bi₄Ge₃O₁₂ (BGO) scintillators (size of each: 5 cm × 5 cm, 7.5 cm long), which was placed at a distance of 12.70±0.02 m from the neutron source, was employed as a total energy absorption detector for the prompt capture gammaray measurement, to obtain the absolute capture cross-sections. The enriched samples of ¹⁶¹Dy and ¹⁶³Dy are in the form of metallic plates (20 mm in diameter, 0.21 mm in thickness). A ¹⁰B sample (18 mm × 18 mm, 4.54 g/cm²) was employed to monitor the neutron flux of the TOF beam using the standard reference cross-section of the ¹⁰B(n,α) reaction. The Dy sample or B-10 capture sample was set at the center of the through hole in the BGO assembly. The current measurement has been compared with the previous measurements and the evaluated data in ENDF/B-VI.

Prompt neutron spectra for Cm-isotopes (²⁴²Cm, ²⁴³Cm, ²⁴⁴Cm, ²⁴⁵Cm, ²⁴⁶Cm, ²⁴⁸Cm) were calculated on the basis of a modified version of the Madland-Nix model combined with a multimodal fission model. The predicted spectra were found to be in fair agreement with recent data. A slight enhancement of the low-energy component of the spectrum was interpreted in terms of neutron emission during fragment acceleration.

We have developed a simple method to estimate covariances for resolved and unresolved resonance parameters. The covariance matrices of the resolved and unresolved resonance parameters are estimated so as to reproduce accuracy of the total, capture, and fission cross sections. Energy-averaged cross sections and their covariances are used to obtain these matrices. An example calculation is shown for the ²³⁹Pu resolved resonance and ²³⁵U unresolved resonance regions.

International evaluations are planned for the H(n,n), ³He(n,p), ⁶Li(n,t), ¹⁰B(n,I), ¹⁰B(n,α₁γ), Au(n,γ), and ²³⁵U(n,f) cross section standards. Some additional important cross sections, including those for ²³⁸U(n,f), ²³⁸U(n,γ) and ²³⁹Pu(n,f) will also be evaluated. Most of these cross sections are actively used in neutron dosimetry for fluence determination. Work on the standards evaluation has begun with an investigation of the experiments to be used in the evaluation.

Review articles are in preparation for the 2003 edition of the CRC’s Handbook of Chemistry and Physics dealing with both non-neutron and neutron nuclear data. Highlights include: withdrawal of the claim for discovery of element 118; new measurements of isotopic abundances for many elements; new recommended standards for calibration of γ-ray energies; new half-life measurements for very short lived, long-lived nuclides and ββ decay measurements for quasi-stable nuclides; new reassessment of spontaneous fission (sf) half-lives, distinguishing sf decay and cluster decay half-lives and the new cluster-fission decay; charged particle cross sections, (n,p) and (n,α) measurements for thermal neutrons incident on light nuclides; new thermal (n,γ) cross sections and neutron resonance integrals measured.

The last tested version of the reactor dosimetry file IRDF-90 V2 was released in 1993. Since then large amounts of new experimental data have been measured, and two new national reactor dosimetry libraries have been produced. The Nuclear Data Section (NDS) of the International Atomic Energy Agency (IAEA) is coordinating the efforts of several specialists in the field of reactor dosimetry in order to prepare a new, updated International Reactor Dosimetry File IRDF-2002. The status of this work is described and reviewed in this paper.

The second step of the VVER-1000 Balakovo-3 Ex-vessel Exercise comprises an analysis of calculation results, their intercomparison and comparison with reference experimental data. A good agreement has been found between different calculated fast neutron data as well as between calculated (C) and experimental (E) activation reaction rates (C/E deviates less than 12 % from 1). C/E values for cadmium covered (n,γ) detectors are generally reasonable. The presented results show that the Balakovo-3 Exercise is suitable for benchmarking 3D ex-vessel fast neutron fluence calculations.

Monte Carlo simulations (MCNP4C2) were performed to assess the ability to benchmark neutron transport calculations in iron using a pulsed-neutron slowing-down experiment. Specifically, calculations were performed to obtain the time dependent neutron energy spectra inside a 1 × 1 × 1 m natural iron moderator that is driven by a 14-MeV pulsed neutron source (simulating a pulsed D-T neutron generator). At various time intervals after the pulse, the energy spectrum was tallied and used to estimate the integral time-dependent reaction rates in ²³⁵U, ²³⁸U, ²³⁷Np, and ²³⁹Pu fission detectors that were located inside the moderator. The results show that within 0.05 □s after the pulse, the average energy of the neutrons drops below 800 keV. Therefore, the threshold detectors (²³⁷Np, and ²³⁸U) can be useful at early times, while the fissile detectors (²³⁵U and ²³⁹Pu) can be utilized throughout the experiment. For these detectors, the time dependent reaction rates and spectral indexes (²³⁵U/²³⁹Pu, ²³⁷Np/²³⁹Pu, and ²³⁸U/²³⁹Pu) are developed and discussed.

The US National Institute of Standards and Technology operates a premier calibration laboratory for certifying the neutron emission rate of radioisotopic neutron sources with rigorous quality assurance over a wide range of emission rates. Calibrations are performed using the manganous sulfate bath technique, and are performed with a relative expanded uncertainty of approximately 3.5% (2σ). Recently, an improvement to the calibration procedure has been implemented whereby, in addition to the national standard neutron source, sources are regularly calibrated against one of three (international) standard sources formerly maintained by the Bureau International des Poids et Mesures. This procedure helps ensure that the fidelity of NIST neutron source calibrations is maintained at the highest level.

In accordance with the objective of the Project REDOS [1] available experimental data has been reviewed. The paper is dealing with the data obtained in the frame of various projects which utilize the full scale engineering models of WWER type reactors assembled in the LR-0 experimental reactor, i.e. these ones where it was possible to perform the measurements from the core boundary over the reactor internals and reactor pressure vessel (RPV) to the biological shielding simulator. The aim of the evaluation of the available experimental data set is their qualification for benchmarking of RPV dosimetry (exposure) data.

We have recently developed and implemented a Monte Carlo method for the calculation of sensitivities to the angular distribution of secondary neutrons. In the current work, a sensitivity analysis of iron spheres experiments is presented, using this novel method. The experiments were performed at NRI, Rez, Czech Republic, at NIST, Gaithersburg (both with a ²⁵²Cf source), and at LLNL, Livermore (with a 14 MeV source). Sensitivities of the responses to the angular distribution of neutrons scattered elastically or inelastically to selected levels were calculated and reported. It was found that in leakage experiments the sensitivities to angular distributions are of similar magnitude as the sensitivities to the corresponding cross sections. The sensitivity to the first Legendre moment has usually the opposite sign of the sensitivity to the corresponding cross section. Lack of covariance data in cross section libraries could seriously impede realistic estimation of nuclear-data-induced calculated uncertainties when analyzing experimental results.

No abstract received.

Participants of the 11th International Symposium on Reactor Dosimetry.

Author Index.