scispace - formally typeset

ReportDOI

Developing Spent Fuel Assembly for Advanced NDA Instrument Calibration - NGSI Spent Fuel Project

01 Mar 2014-

AboutThe article was published on 2014-03-01 and is currently open access. It has received 4 citation(s) till now. The article focuses on the topic(s): Spent nuclear fuel.

Topics: Spent nuclear fuel (57%)

Summary (11 min read)

Jump to: [GE Hitachi Nuclear Energy][3.1 EXPERIMENTAL BENCHMARK DATA][3.2 SCALE VALIDATION STUDIES][3.3 DATA REQUIRED FOR MODELING][3.3.1 Fuel Assembly Design Data][3.3.2 Reactor Operating History][3.3.3 Assembly Average Burnup][3.3.4 Axial and Radial Burnup Distributions][3.3.5 Reactivity Control Exposure][3.3.6 Neighbor Assemblies][3.3.7 Deformation of the Fuel][3.3.8 Nuclear Data][3.4.1 Uncertainties in Operator-Estimated Assembly Burnup][3.4.2 Uncertainties in Operator-Estimated Axial Burnup][3.5 IMPACTS OF UNCERTAINTIES IN OPERATOR DATA][4.1 CONVENTIONAL NDA TECHNIQUES][4.1.1 Passive Neutron][4.1.2 Passive Gamma][4.2 NDA INSTRUMENTS][4.2.1 High-Purity Germanium (HPGe)][4.2.2 Cadmium-Zinc-Telluride (CZT)][4.2.3 Cerenkov Viewing Devices (CVD)][4.2.4 Fork Detector][4.2.5 Spent Fuel Attribute Tester (SFAT)][4.2.6 Partial defect DETector (PDET)][4.3 DETERMINATION OF BURNUP USING GAMMA NDA][4.4 ACCURACY OF NDA MEASUREMENT OF BURNUP][4.5 APPLICATIONS OF PDET][5.1 MODELING REQUIREMENTS][5.2 DESCRIPTION OF THE NEW NODAL DEPLETION CAPABILITY][5.2.1 Radial Representation][5.2.2 Axial Representation][5.2.3 Output][6.1 DESCRIPTION OF THE TMI-1 ASSEMBLIES][6.2 EVALUATION OF OPERATOR AND NDA DATA FOR TMI-1 FUEL][6.2.1 Axial Burnup Profile of TMI-1 Fuel][6.3.1 The Burnup Models][6.3.3 Nuclide Concentrations][7.1 UNCERTAINTIES IN NUCLEAR DATA][7.2 UNCERTAINTY ANALYSIS METHODS][7.3 IMPACT OF NUCLEAR DATA UNCERTAINTIES ON NUCLIDE CONCENTRATIONS][8.1 DESCRIPTION OF CIPN][8.2 PROPAGATION OF UNCERTAINTIES TO CIPN][8.2.1 Case 1][8.2.2 Case 2][8.2.3 Case 3] and [8.2.4 Case 4]

GE Hitachi Nuclear Energy

  • Steven E. Skutnik University of Tennessee DOCUMENT AVAILABILITY.
  • The uncertainty in a calibration standard is required for the analysis of overall uncertainties associated with the complete instrument calibration process.

3.1 EXPERIMENTAL BENCHMARK DATA

  • The purpose of this subsection is to provide an overview of the existing DA measurement data that can be used for burnup code validation.
  • A large amount of spent fuel DA measurement data has been reported over the years, for both pressurized water reactors (PWRs) and boiling water reactors (BWRs), by different experimental programs, both domestic and international, that were designed to provide data for the purposes of code validation.
  • As shown, the experimental data includes a broad range of assembly designs, enrichments (from 2.63 to 4.657%), burnups (from 7.2 to 54 GWd/tU), and cooling 6 times that are representative of the commercial fuel inventory.
  • The list of measured nuclides varies from one experiment to another, depending on the primary interest of the programs, but the list usually includes uranium, plutonium and other higher actinides, major fission products (e.g., 137Cs and 154Eu), and sometimes the main neutron-absorbing fission products (e.g., 149Sm and 155Gd).
  • When all uncertainties related to sample handling, dissolution, and radiochemistry are included, the total uncertainties of 1–2% can be achieved using TIMS for nuclides with high isotopic abundance, a value established by independent laboratory cross-check measurements [10, 11].

3.2 SCALE VALIDATION STUDIES

  • The SCALE code system is a nuclear systems modeling and simulation package developed at Oak Ridge National Laboratory (ORNL) [6], and it has been used for the assembly calculations described in this report.
  • (There is a 3-D burnup module under development in SCALE using KENO.).
  • The results for some nuclides have been found to be very sensitive to the source of the cross section data [9].
  • The burnup of a calibration assembly will likely rely on the operator calculated value that has larger uncertainty than experimentally determined values.
  • Validation studies for BWR fuels have also been performed [14].

3.3 DATA REQUIRED FOR MODELING

  • Also included is the accessibility to that information, and general comments.
  • This list is based on the most important modeling parameters based on international experience [10].
  • These items are discussed in more detail below.
  • The modeling requirements for BWR assembly designs are more complex and are not addressed in this report.

3.3.1 Fuel Assembly Design Data

  • Fuel design information is generally available, although for modern designs detailed information may be deemed proprietary by fuel vendors (e.g., IFBA patterns and dimensions, enrichment variation, etc.) and could be more difficult to obtain.
  • Reference 15 summarizes the types of fuel assemblies used in both PWRs and BWRs in the United States.

3.3.2 Reactor Operating History

  • The reactor operating history information is important because the local power level and duration of irradiation and decay can have significant impacts on many nuclides.
  • This information includes the start and end date of each cycle and the power of the assemblies as provided by the operator.
  • This information is usually easy to obtain.
  • Detailed day-to-day power variation could be more difficult to obtain unless provided by the operator.

3.3.3 Assembly Average Burnup

  • The assembly average burnup is routinely provided by the operator and included in fuel declarations.
  • Burnup of the fuel is one of the more important factors that affect the nuclide compositions, and the uncertainties in operatorprovided burnup are discussed in subsection 3.4.

3.3.4 Axial and Radial Burnup Distributions

  • The burnup varies significantly along the axial length of the assembly due to the power distributions in the reactor core, power depression around spacer grids, axial variation in the moderator density, insertion of control rods, and even operating anomalies (e.g., boron deposition on the fuel rods).
  • The burnup can also vary radially due to design and operating factors, such as guide tube configuration, usage of burnable poison rods (BPRs), insertion of control rods, and asymmetric neighbor assembly conditions.
  • Commercial full-core analysis codes can estimate both the axial and radial burnup distributions, generally within a few percent of measurements [16].
  • In PWRs, neutron flux is routinely measured in the IT of some of the incore assemblies (one IT per assembly) at different axial elevations, and these measurements can be used to benchmark code calculations.
  • Concerns regarding accuracy (not directly measured) and accessibility are the two main issues for this category.

3.3.5 Reactivity Control Exposure

  • Reactor reactivity-control measures can have a very significant impact on assembly nuclide compositions.
  • Studies using SCALE have found that using a computed cycle-average boron concentration, instead of the variation in boron concentration vs. burnup, makes only a minor impact on nuclide compositions [17].
  • Control rods are sometimes partially inserted and therefore affect only axial regions near the top of the assembly.
  • Exposure to BPRs can have a significant impact on nuclide compositions as they are fully inserted in guide tubes of the assembly when they are used and introduce neutron-absorbing poison in the assembly and displace water.

3.3.6 Neighbor Assemblies

  • Adjacent assemblies can impact the burnup gradient within an assembly, especially when the burnup of neighbor assemblies vary significantly or the host assembly resides at the edge of the core.
  • The information about neighbor assemblies may be difficult to obtain and requires the core loading patterns showing the assembly locations and properties.
  • This effect is generally limited to the outer row of fuel rods.
  • In many cases the results obtained by neglecting the neighbor assemblies were comparable to more detailed neighbor models.
  • Asymmetric neighbor configuration may introduce significant gradients in burnup and thus on the distribution of nuclide compositions in the assembly (even when the total quantity of certain nuclides is similar in the assembly).

3.3.7 Deformation of the Fuel

  • Fuel rods usually experience deformations under the high-temperature, high-pressure environment during irradiation.
  • The fuel swells due to fission gas accumulation and thermal expansion.
  • The fuel rods may also bow in some direction, which may alter the rod pitch and thus impact the local neutron spectrum.
  • Information about the degree of bowing is not usually available, and is therefore not considered in this work.
  • Changes from the as-built geometry will contribute to the overall uncertainty in the results from computational models.

3.3.8 Nuclear Data

  • Modeling and simulation requires high-quality nuclear data, and this is particularly true for depletion calculations.
  • Uncertainties in nuclear cross-section data, decay data, and fission yield data contribute to the uncertainties in the calculations.
  • Significant improvements in the accuracy of calculations have been observed for several fission product nuclides using the latest ENDF/B-VII nuclear data [9].
  • The impacts of nuclear data uncertainties on nuclide compositions are evaluated in Section 7, and the impacts of these uncertainties on advanced NDA responses are summarized in Section 8.

3.4.1 Uncertainties in Operator-Estimated Assembly Burnup

  • There are several publicly available reports that quantify the accuracy of the assembly average burnup provided by a reactor operator.
  • In the AREVA study, a large number of assemblies (over 10,000) from nine different plants were analyzed.
  • The conversions of reaction rates to burnup (units of GWd/tU) are complex and must include the relative contribution to time-dependent fission from each major nuclide, depending on multiple factors including enrichment, moderator densities, etc.
  • Therefore, after including the uncertainties stemming from the conversions, the estimated uncertainties of the operator-provided assembly average burnup are expected to be larger than the reported values in these two studies, but are usually less than 5% [17].

3.4.2 Uncertainties in Operator-Estimated Axial Burnup

  • Commercial core simulator codes, such as the Studsvik Scandpower codes CASMO and SIMULATE [20], are used to calculate the burnup distribution of assemblies.
  • Because these codes have been extensively benchmarked against DA data and include in-core measurement data, they usually predict the axial burnup well.
  • Figure 3 shows a comparison of operator-estimated and NDA measured nodal burnup on a single fuel rod [17].
  • The estimated burnup was calculated using SIMULATE, and the experimental NDA burnup was based on a 137Cs gamma scan.
  • As shown, the difference is generally within 2% except at the upper region of the fuel rod where deviations increase to about 3%.

3.5 IMPACTS OF UNCERTAINTIES IN OPERATOR DATA

  • The operator data required to construct an assembly depletion model has inherent uncertainties.
  • As summarized in Table 4, the exposure to BPRs had the largest impacts on 239Pu, total plutonium content, and total fissile content (combined mass of 235U, 239Pu, and 241Pu).
  • The impacts from other factors studied are observed to be relatively small, including the four Gd rods.
  • Application of NDA to spent fuel safeguards has been more challenging due to the complexity of nuclear compositions and radiation emissions in irradiated fuel, and the need to perform assembly measurements in water pools which restricts the types of instruments and detectors that can be used.
  • This would require the NDA data to have smaller uncertainties than data provided by the operator.

4.1 CONVENTIONAL NDA TECHNIQUES

  • The focus of the NDA application for spent fuel characterization presented in this section is on the conventional passive NDA techniques currently in use.
  • The advanced detectors developed under the NGSI program are not considered for this purpose, because they have not been calibrated.
  • The reliance on conventional NDA instruments limits the information that can be obtained from spent fuel, because these instruments are limited to gross neutron, gross gamma, and gamma-ray spectroscopy.
  • NDA is used most frequently to obtain estimates of burnup, but has also been used to derive other properties of the fuel including cooling time.
  • Passive neutron and gamma stand for the neutron and gamma radiation, respectively, emitted directly from the irradiated fuel without the presence of external radiation sources.

4.1.1 Passive Neutron

  • Spent nuclear fuel emits neutrons primarily through the spontaneous fission of higher actinides, such as curium, that grows with burnup roughly to the fourth power.
  • The accumulation of curium is extremely sensitive to the fuel burnup.
  • Fission chambers are commonly used to measure neutrons from spent fuel.
  • The measured neutron signal may be used to infer the fuel burnup.

4.1.2 Passive Gamma

  • Spent fuel emits gamma-ray radiation, primarily through the decay of fission products.
  • The intensity of these lines can then be used to infer the amount of the source nuclides in the fuel.
  • In addition to the attenuation problem, gross gamma has other complications for the purpose of spent fuel characterization; for spent fuel with relatively short cooling time, gross gamma is very sensitive to the power level at which the fuel was last irradiated.
  • Gross gamma is often measured by ion chambers, while the gamma energy spectrum is usually measured with High-Purity Germanium (HPGe), and cadmium-zinc-telluride (CZT) detectors.

4.2 NDA INSTRUMENTS

  • Table 5 summarizes the conventional NDA instruments used to characterize or measure spent nuclear fuel.
  • Also included are the measured signals, the advantages, and the disadvantages (in the context of spent fuel measurement or characterization) of each NDA technique.
  • The details of each NDA instrument are further discussed below.

4.2.1 High-Purity Germanium (HPGe)

  • HPGe detectors are a type of semiconductor diode with a high-energy resolution, which allows excellent gamma energy peak distinction in measured gamma spectra (from either solid fuel or solution).
  • Major contributors to the gamma spectrum in spent fuel include 134Cs, 137Cs, and 154Eu, with 134Cs and 154Eu having multiple major peaks.
  • While different configurations are available, all HPGe detectors basically consist of a detector crystal with a preamplifier which is mounted under vacuum typically in an aluminum cap.
  • The difficulty in using the detector is the necessity to cool it to around -196 °C with either liquid nitrogen or an electrically powered cryogenic refrigerator [10, 22], which generally prohibits the usage of HPGe for in-pool measurements frequently required for spent fuel assemblies.

4.2.2 Cadmium-Zinc-Telluride (CZT)

  • The CZT detector is a room-temperature semiconductor detector that can measure the gamma spectrum of spent fuel with moderate energy resolution.
  • Because it does not require active cooling, CZT can be used for in-pool measurement.
  • A CZT detector has been used in combination with the fission and ionization chambers in a Fork detector [22] for spent fuel measurements, enabling simultaneous measurements of passive neutron, gross gamma, and gamma energy spectra.
  • The ability to measure gamma spectrum, rather than just gross gamma, is a significant improvement over the conventional Fork detector.
  • The CZT detector has more flexibility than HPGe for spent fuel measurement because it does not have strict cooling requirements.

4.2.3 Cerenkov Viewing Devices (CVD)

  • Cerenkov viewing devices typically measure the light produced in the water from gamma-ray excitations.
  • It does not provide a quantitative measure of the fuel.
  • It can also be challenged by low-burnup and poor water clarity.
  • Therefore, it is not considered as useful for the current study.

4.2.4 Fork Detector

  • The Fork detector is widely used in operational measurements of spent fuel safeguards to verify operator declaration and to detect partial defects (e.g., missing nuclear material).
  • This device is usually equipped with an ionization chamber and two fission chambers on each arm to measure gross gamma and total neutron respectively.
  • Past experience with the Fork detector suggests that it can only detect partial defect when 50% or greater amounts of fuel rods are removed or replaced [23].
  • In some cases, the Fork detector 20 is also equipped with a CZT detector, adding the capability of measuring the gamma spectrum.
  • Recent work has shown that by coupling with advanced modeling and simulation, the Fork detector can be used to verify operator declarations of burnup with relatively low uncertainties (e.g., ~5%) [24].

4.2.5 Spent Fuel Attribute Tester (SFAT)

  • It typically measures the upper fraction of the assembly.
  • Given the significant attenuation in the system, it is difficult to quantitatively relate the measured signal to source nuclide concentrations in the fuel assembly.
  • This device is intended to provide qualitative measurement of one of the dominant fission products.

4.2.6 Partial defect DETector (PDET)

  • PDET was developed with the goal to detect missing fuel rods in spent fuel assemblies [26].
  • There is an opportunity to explore the application of PDET results for developing calibration assemblies.
  • It resembles the PWR control rod cluster that can be inserted into empty guide tube locations within the assembly to measure the local neutron and gamma intensities within the assembly.
  • PDET is unique because it is the only NDA instrument that measures the interior of an assembly, thereby providing information on the spatial distribution of radionuclide in different fuel rods of the assembly.
  • In this work it is found that the gamma measurement of PDET may be used to infer the burnup distribution of the assembly when distributions are not available from the operator.

4.3 DETERMINATION OF BURNUP USING GAMMA NDA

  • Fuel burnup, both the assembly average and the distribution within the assembly, is an important factor that influences the fuel compositions and detector response.
  • Because 134Cs and 154Eu are produced through neutron capture reaction of 133Cs and 153Eu, respectively, their production rates are dependent on the neutron spectrum in the fuel.
  • Figure 5(a) shows the correlation between the 134Cs/137Cs ratio and burnup as measured using DA nuclide data from Takahama-3 spent fuel samples [27].
  • (The gamma ratios can also be used as a relative tool on similar types of fuel assemblies.).

4.4 ACCURACY OF NDA MEASUREMENT OF BURNUP

  • Usage of NDA methods to estimate spent fuel burnup has been studied in the past, and the reported accuracies (compared to DA results) vary widely from 4% to 20% [22, 29].
  • The selected experiments include measurements for more than 50 different fuel samples from the Trino Vercellese [5], Obrigheim [5], and Vandellós [16] reactors.
  • The NDA measured burnup was determined based on the 137Cs content, measured using gamma spectrometry on individual fuel rods.
  • The differences are greater on both ends of the fuel rods, especially for Obrigheim.

4.5 APPLICATIONS OF PDET

  • It could potentially be used to measure radial burnup distribution within an assembly in situations where such information is not available from the reactor operator.
  • Because the PDET does not measure all 179 fuel rods in the assembly, interpolation and extrapolation are needed to reconstruct the pin-by-pin burnup.
  • As previously discussed, spent fuel assembly nuclide compositions are needed for advanced NDA instrument calibration and to support detector performance simulations.
  • The accuracy of the nuclide compositions will play an important role in defining the accuracy that can be achieved using advanced NDA instruments.
  • SCALE includes several modules to perform the neutron transport calculations necessary to simulate the assembly during irradiation in the reactor and uses the ORIGEN code [33] to calculate the isotopic evolution and decay.

5.1 MODELING REQUIREMENTS

  • An important requirement of the modeling and simulation codes for this project is the need to accurately represent the three-dimensional (3-D) variation of the nuclide compositions of the assembly, radially (pinby-pin) and axially.
  • Variations in the compositions within an assembly will directly impact the measured signals of advanced NDA instruments.
  • These variations can be caused by many factors including asymmetric power gradients from neighbor assemblies, leakage at the core boundaries, control rod exposure, layout of guide tubes, BPRs, and other operating conditions.

5.2 DESCRIPTION OF THE NEW NODAL DEPLETION CAPABILITY

  • Under this task of the NGSI project, a new 3-D assembly depletion capability, named ORIGAMI (ORIgen AsseMbly Isotopics), has been developed at ORNL to generate nuclide compositions for an entire fuel assembly using a user-defined pin-by-pin burnup map and axial burnup profile.
  • The axial distribution is represented using any number of discrete axial zones as required.
  • For a typical 14×14 PWR assembly (179 fuel rods), represented axially using 25 zones (typical of data provided by the reactor 30 operator), there are approximately 4,300 separate fuel regions of the assembly to be calculated.
  • This function is performed using the depletion analysis sequence TRITON in SCALE, which performs a neutron transport analysis of the assembly (using either 2-D or 3-D models), to generate spatially (pin-by-pin) and time dependent cross-section libraries for ORIGEN.
  • Because the transport calculations used to generate cross sections are performed in advance, a typical single fuel region calculation using ORIGEN alone requires only seconds on a typical computer.

5.2.1 Radial Representation

  • The nuclide compositions for each fuel pin of the assembly are calculated explicitly using ORIGEN with the burnup distribution provided for each rod.
  • These distributions may be provided by the reactor operator or measured using NDA instruments (e.g., passive gamma).
  • These burnup codes do not allow the user to define the actual burnup distribution of the fuel rods, but rather this distribution is calculated automatically by the transport code.
  • Introducing large burnup gradients in an 32 assembly, such as that shown in Fig. 12, would require that the model be adjusted to reduce the power on one side (or sides) of the assembly by either introducing an absorbing material or otherwise changing the boundary conditions of the problem.
  • For this study, separate libraries were generated for groups of similar fuel rods: (1) corner rods, (2) periphery (edge) rods, (3) rods adjacent to guide tubes, and (4) other interior fuel rods.

5.2.2 Axial Representation

  • The axial burnup distribution is represented in the model in the same way as the radial (X-Y) distribution.
  • The axial distribution is discretized, and the relative burnup value is assigned to each axial zone.
  • These axial values are normalized to unity and become multipliers of the X-Y map burnup values to obtain the power for each axial and radial fuel region of the assembly.
  • Axial burnups may also be measured by NDA, most frequently using an axial gamma scan of the fuel rod or assembly using 137Cs as the burnup indicator.
  • The slight increase in burnup near both ends of the fuel rod is caused by increased moderation in this region, and the depressions (five) are caused by the fuel rod grid spacers (displacement of moderator and neutron absorption by the grid).

5.2.3 Output

  • ORIGAMI has been used to generate the detailed isotopic results in each axial and radial fuel region in each ROK assembly destined to be tested with the NGSI advanced NDAs.
  • The calculation procedure also generates plots of the distributions for key nuclides of the assembly.
  • As shown in Fig. 15(a), the plutonium content correlates well to the radial burnup gradient as depicted in Fig. 12.
  • This methodology has been used to generate calibration standards for the ROK assemblies where pin-by-pin burnup maps are available from the operator.
  • This report has discussed the application of fuel design information, operator data, and NDA measurements to predict the compositions for a spent fuel assembly used for instrument calibration.

6.1 DESCRIPTION OF THE TMI-1 ASSEMBLIES

  • Several fuel rods from two assemblies, NJ070G and NJ05YU, were removed from the reactor after Cycle 10 for examinations.
  • For fuel temperature and soluble boron, though only average values are cited here, time-varying values are available in Ref. 38, and those detailed values are used in the models in this work.
  • Figure 16 illustrates the configuration of assembly NJ070G [38].
  • Fuel rods D5 and H6 were extracted from the assembly for examinations after Cycle 10 (the second irradiation cycle of the assembly).

6.2 EVALUATION OF OPERATOR AND NDA DATA FOR TMI-1 FUEL

  • Under the EPRI investigation of the root cause of fuel failure in TMI-1, extensive examinations have been performed on the fuel rods from these two assemblies (NJ070G and NJ05YU), including gamma scanning, neutron radiography, and destructive radiochemical assay [4].
  • Of most interest to this work are the gamma scanning and radiochemical measurements, because (a) the axial gamma scanning provides the axial burnup profile of the fuel rods, and (b) the radiochemical measurements provide the nuclide compositions of the fuel samples.
  • Significant amounts of operator data on the TMI-1 fuel were also collected by the EPRI and other studies [4, 38, 39].
  • Before using the operator and NDA data for burnup calculations, it is prudent to evaluate the quality of these data.

6.2.1 Axial Burnup Profile of TMI-1 Fuel

  • The axial burnup profile of TMI-1 fuel was measured using NDA methods (gamma scanning) and can also be derived from operator data.
  • Details are discussed in the following subsections.

6.3.1 The Burnup Models

  • Both assemblies NJ070G and NJ05YU were modeled using SCALE/TRITON in the burnup calculations in this study, and the models are described later in this subsection.
  • As shown, the previously irradiated assemblies were colored in yellow, for which the batch ID and the BOC (beginning of cycle) burnup (GWd/tU) are listed.
  • As shown, the neighbor configuration is not symmetric in Cycle 9 (notice the difference between the southwest and northeast corner), but is symmetric in Cycle 10.
  • 44 Figure 22 shows the TRITON model for assembly NJ070G and its eight neighbors.
  • To study the impacts of neighbor assemblies, both assembly NJ070G and NJ05YU were also modeled as stand-alone (i.e., no neighbor assemblies included) assemblies with reflective boundary conditions and these models are referred to as “without-neighbor” model.

6.3.3 Nuclide Concentrations

  • In addition to the radial burnup distributions, the nuclide concentrations in these two assemblies were also calculated and analyzed.
  • Some of the results are presented in this subsection.

7.1 UNCERTAINTIES IN NUCLEAR DATA

  • There are three main types of nuclear data involved in burnup calculations: 1) neutron cross sections (e.g., fission and absorption cross sections); 2) fission product yields (e.g., fission product generation due to the fission of an actinide); and 3) decay data (e.g., half-lives, branching ratios).
  • In addition, many of the data are correlated, and accurate representations of these data correlations (covariance files) are necessary for rigorous uncertainty analysis.
  • The neutron cross-section covariance data used in this work were developed prior to the release of ENDF/B-VII.1 and are distributed with the SCALE code system.
  • Selected covariance evaluations were taken from the pre-release of ENDF.
  • To support uncertainty analysis for fission products, correlation matrices for direct fission yields have recently been developed by ORNL using the nuclear data and uncertainties in the ENDF/B-VII.0 evaluations, developed by England and Rider [47], and these covariance files have been implemented for use in SCALE.

7.2 UNCERTAINTY ANALYSIS METHODS

  • A newly developed uncertainty analysis tool within SCALE, named Sampler [43], was applied to the burnup calculations used to support NGSI spent fuel analysis in this work.
  • Sampler repeatedly calls the SCALE sequence to perform the calculation, each time using a different set of perturbed nuclear data libraries, and then post-processes the results to obtain the distribution and statistical parameters on the calculated quantities.
  • Figure 28 shows the flowchart of Sampler.
  • For each set of the perturbed data libraries, an individual TRITON calculation was executed and the responses (e.g., nuclide concentrations in this case) due to the different data libraries were obtained.
  • The variance in the responses attributed to the nuclear data uncertainties can thus be assessed.

7.3 IMPACT OF NUCLEAR DATA UNCERTAINTIES ON NUCLIDE CONCENTRATIONS

  • A simplified assembly model of a typical 15×15 PWR design with 16 guide tubes and 1 central instrument tube was developed for this work, and it is shown in Figure 29.
  • As shown, the relative standard deviations caused by the uncertainties in nuclear data are generally within 2% for most actinides, and they vary from one nuclide to another because their production paths are different.
  • Many of these nuclides (e.g., 133Cs, 143Nd, 149Sm, 154Eu) are major neutron absorbers in spent fuel, and they have important impacts on NDA instrument neutron signals.
  • In the previous sections, the authors discussed how uncertainties in design and operating history information, NDA data, and nuclear data affect the nuclide compositions in a spent fuel assembly.
  • It is therefore important to not only quantify the uncertainties in the compositions of a calibration assembly but also the net impacts of those uncertainties on the signals measured by advanced NDA instruments.

8.1 DESCRIPTION OF CIPN

  • CIPN is a relatively low-cost and lightweight instrument that resembles a Fork detector, except that CIPN has an active interrogation source (252Cf).
  • CIPN currently does not measure gamma spectrum, but such capability could be added.
  • As shown, there are four fission chambers in the instrument to detect total neutrons and two ion chambers to detect total photons.
  • In addition to the passive neutrons, the neutrons emitted from the californium source will induce fissions in the fuel, and these fission neutrons will add to the neutron signal.
  • The difference of neutron counts between the active and passive mode, or the net neutron 60 count, is related to the neutron multiplication factor of the assembly and thus the fissile content [48].

8.2 PROPAGATION OF UNCERTAINTIES TO CIPN

  • Several scenarios were studied to assess how uncertainties in the nuclide contents for a reference calibration assembly affect the CIPN signals.
  • The following cases were analyzed: 1) Case 1, modeling an assembly with or without the neighbor assembly information;.
  • The purpose of this case is to assess the impact of neighbor assemblies on the detector response, because the information about neighbor assemblies may be difficult to obtain from the operator [10].
  • In Case 4, 20 sets of assembly nuclide concentrations were generated with perturbed nuclear data libraries using Sampler, and then these concentrations were imported into the CIPN MCNPX model to simulate the CIPN count rates.

8.2.1 Case 1

  • Figure 35 shows the relative percent difference in CIPN neutron count rate due to the two different sets of nuclide compositions generated by two different depletion models for assembly NJ05YU; one modeled the actual neighbor assemblies during irradiation and the other did not.
  • (Curium-244 is the primary source of neutrons in the passive mode because the fuel is cooled for more than 15 years.).
  • Also shown is the relative difference in 137Cs in each fuel rod between the two depletion models after the first cycle.
  • The photons usually cannot travel far because of the high attenuation within a fuel assembly; thus, the gamma signals are primarily correlated to outer fuel rods that are closest to the detectors.
  • The differences in nuclide compositions of assembly NJ05YU after the second cycle are much smaller than the first cycle, and thus the differences in CIPN responses are small as well (less than 1% in neutron signals and about 1.7% in gamma signals).

8.2.2 Case 2

  • The nuclide compositions are sensitive to assembly average burnup, as previously shown in Table 4 (Section 3.5).
  • This case evaluates the impacts due to uncertainties in assembly average burnup using two cases that have a 5% difference in average burnup based on assembly NJ05YU.
  • This is due to the fact that the generation of 244Cm is highly sensitive to burnup (to the fourth power).
  • The average difference in the CIPN passive neutron signals is around 17%, and around 5% for the average gamma signals.

8.2.3 Case 3

  • The impacts of different burnup distributions within an assembly on CIPN gamma and neutron count rates were also studied using three ROK fuel assemblies.
  • Assembly 1 (FA1) is a low-burnup assembly (~17 GWd) that was not subjected to significant cross assembly power gradients.
  • The lowest burnup occurred for fuel on the four corners.
  • In order to quantify the impact of potential uncertainties in the fuel burnup distribution in an assembly, a comparison was done using these three assemblies that had operator-estimated burnup distributions.
  • Table 16 shows the relative difference (%) in neutron count rates of CIPN due to different burnup distributions within the assembly (both passive and net counts).

8.2.4 Case 4

  • As described in Sect. 7, different assembly nuclide concentrations can be generated with different sets of nuclear data libraries, with each set reflecting the uncertainties in the underlying nuclear data, using Sampler.
  • The nuclear data uncertainties have a larger impact on passive neutron count rates than gamma count rates because 244Cm is more sensitive to nuclear data uncertainties than 137Cs.
  • O Evaluated the impact of neighbor assemblies on CIPN response, the impact of uncertainties in assembly average burnup, and uncertainties in radial burnup profiles on CIPN signals.

Did you find this useful? Give us your feedback

...read more

Content maybe subject to copyright    Report

ORNL/TM-2013/576
Developing Spent Fuel Assembly Standards for
Advanced NDA Instrument Calibration –
NGSI Spent Fuel Project
February 2014
Prepared by:
Jianwei Hu
Ian C. Gauld
Oak Ridge National Laboratory
James E. Banfield
GE Hitachi Nuclear Energy
Steven E. Skutnik
University of Tennessee

DOCUMENT AVAILABILITY
Reports produced after January 1, 1996, are generally available free via US Department of Energy
(DOE) SciTech Connect.
Website http://www.osti.gov/scitech/
Reports produced before January 1, 1996, may be purchased by members of the public from the
following source:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone 703-605-6000 (1-800-553-6847)
TDD 703-487-4639
Fax 703-605-6900
E-mail info@ntis.gov
Website http://www.ntis.gov/support/ordernowabout.htm
Reports are available to DOE employees, DOE contractors, Energy Technology Data Exchange
representatives, and International Nuclear Information System representatives from the following
source:
Office of Scientific and Technical Information
PO Box 62
Oak Ridge, TN 37831
Telephone 865-576-8401
Fax 865-576-5728
E-mail reports@osti.gov
Website http://www.osti.gov/contact.html
This report was prepared as an account of work sponsored by an
agency of the United States Government. Neither the United States
Government nor any agency thereof, nor any of their employees,
makes any warranty, express or implied, or assumes any legal
liability or responsibility for the accuracy, completeness, or
usefulness of any information, apparatus, product, or process
disclosed, or represents that its use would not infringe privately
owned rights. Reference herein to any specific commercial product,
process, or service by trade name, trademark, manufacturer, or
otherwise, does not necessarily constitute or imply its endorsement,
recommendation, or favoring by the United States Government or
any agency thereof. The views and opinions of authors expressed
herein do not necessarily state or reflect those of the United States
Government or any agency thereof.

ORNL/TM-2013/576
Reactor and Nuclear Systems Division
DEVELOPING SPENT FUEL ASSEMBLY STANDARDS FOR ADVANCED NDA
INSTRUMENT CALIBRATION – NGSI SPENT FUEL PROJECT
Jianwei Hu
Ian C. Gauld
James E. Banfield
Steven E. Skutnik
∗∗
_______________
GE Hitachi Nuclear Energy
∗∗
University of Tennessee
Date published: February 2014
Prepared by
OAK RIDGE NATIONAL LABORATORY
Oak Ridge, Tennessee 37831-6285
managed by
UT-BATTELLE, LLC
for the
U.S. DEPARTMENT OF ENERGY
under contract DE-AC05-00OR22725


iii
CONTENTS
Page
LIST OF FIGURES ...................................................................................................................................... v
LIST OF TABLES ...................................................................................................................................... vii
ACKNOWLEDGMENTS ........................................................................................................................... ix
EXECUTIVE SUMMARY ......................................................................................................................... xi
LIST OF ACRONYMS ............................................................................................................................. xiii
1. INTRODUCTION ................................................................................................................................ 1
2. APPROACHES TO CALIBRATION STANDARDS DEVELOPMENT ........................................... 3
3. DATA REQUIREMENTS AND UNCERTAINTIES IN SPENT FUEL MODELING AND
SIMULATION ..................................................................................................................................... 5
3.1 EXPERIMENTAL BENCHMARK DATA ............................................................................... 5
3.2 SCALE VALIDATION STUDIES ............................................................................................. 7
3.3 DATA REQUIRED FOR MODELING ................................................................................... 10
3.3.1 Fuel Assembly Design Data ......................................................................................... 10
3.3.2 Reactor Operating History ........................................................................................... 10
3.3.3 Assembly Average Burnup .......................................................................................... 10
3.3.4 Axial and Radial Burnup Distributions ........................................................................ 11
3.3.5 Reactivity Control Exposure ........................................................................................ 12
3.3.6 Neighbor Assemblies ................................................................................................... 12
3.3.7 Deformation of the Fuel ............................................................................................... 12
3.3.8 Nuclear Data ................................................................................................................ 13
3.4 UNCERTAINTIES IN FUEL BURNUP .................................................................................. 13
3.4.1 Uncertainties in Operator-Estimated Assembly Burnup .............................................. 13
3.4.2 Uncertainties in Operator-Estimated Axial Burnup ..................................................... 13
3.5 IMPACTS OF UNCERTAINTIES IN OPERATOR DATA ................................................... 14
4. APPLICATION OF NDA FOR SPENT FUEL CHARACTERIZATION ........................................ 17
4.1 CONVENTIONAL NDA TECHNIQUES ............................................................................... 17
4.1.1 Passive Neutron ........................................................................................................... 17
4.1.2 Passive Gamma ............................................................................................................ 18
4.2 NDA INSTRUMENTS ............................................................................................................. 18
4.2.1 High-Purity Germanium (HPGe) ................................................................................. 18
4.2.2 Cadmium-Zinc-Telluride (CZT) .................................................................................. 19
4.2.3 Cerenkov Viewing Devices (CVD) ............................................................................. 19
4.2.4 Fork Detector ............................................................................................................... 19
4.2.5 Spent Fuel Attribute Tester (SFAT) ............................................................................. 20
4.2.6 Partial defect DETector (PDET) .................................................................................. 20
4.3 DETERMINATION OF BURNUP USING GAMMA NDA ................................................... 20
4.4 ACCURACY OF NDA MEASUREMENT OF BURNUP ...................................................... 24
4.5 APPLICATIONS OF PDET ..................................................................................................... 26
5. SPENT FUEL ASSEMBLY MODELING AND SIMULATION METHODS ................................. 29
5.1 MODELING REQUIREMENTS ............................................................................................. 29
5.2 DESCRIPTION OF THE NEW NODAL DEPLETION CAPABILITY ................................. 29
5.2.1 Radial Representation .................................................................................................. 30
5.2.2 Axial Representation .................................................................................................... 32
5.2.3 Output .......................................................................................................................... 33
6. A SPENT FUEL CALIBRATION STANDARD – A CASE STUDY WITH THE THREE
MILE ISLAND FUEL ........................................................................................................................ 35

Citations
More filters

01 Jan 2014
Abstract: The Swedish Nuclear Fuel and Waste Management Company (SKB), European Atomic Energy Community (Euratom), two universities and several U.S. Department of Energy Laboratories have joined in a collaborative research effort to determine the capability of non-destructive assay (NDA) techniques to meet the combined needs of the safeguards community and the Swedish encapsulation and repository facilities operator SKB. These needs include partial defect detection, heat quantification, assembly identification (initial enrichment, burnup and cooling time), and Pu mass and reactivity determination. The experimental component of this research effort involves the measurement of 50 assemblies at the Central Storage of Spent Nuclear Fuel (Clab) facility in Sweden, 25 of which were irradiated in Pressurized Water Reactors and 25 in Boiling Water Reactors. The experimental signatures being measured for all assemblies include spectral resolved gammas (HPGe and LaBr3), time correlated neutrons (Differential Die-away Self Interrogation), time-varying and continuous active neutron interrogation (Differential Die-away and an approximation of Californium Interrogation Prompt Neutron), total neutron and total gamma fluxes (Fork Detector), total heat (assembly length calorimeter) and possibly the Cerenkov light emission (Digital Cerenkov Viewing Device). This paper fits into the IAEA’s Department of Safeguards Long-Term R&D Plan in the context of developing “more sensitive and less intrusive alternatives to existing NDA instruments to perform partial defect test on spent fuel assembly prior to transfer to difficult to access storage,” as well as potentially supporting pyrochemical processing. The work describes the specific measured signatures, the uniqueness of the information contained in these signatures and why a data mining approach is being used to combine the various signatures to optimally satisfy the various needs of the collaboration. This paper will address efficient and effective verification strategies particularly in the context of encapsulation and repository facilities.

10 citations


Journal ArticleDOI
Abstract: The performance of a passive neutron albedo reactivity (PNAR) instrument to measure neutron multiplication of spent nuclear fuel in borated water is investigated as part of an integrated non-destructive assay safeguards system. To measure the PNAR Ratio, which is proportional to the neutron multiplication, the total neutron count rate is measured in high- and low-multiplying environments by the PNAR instrument. The integrated system also contains a load cell and a passive gamma emission tomograph, and as such meets all the recommendations of the IAEA’s recent ASTOR Experts Group report. A virtual spent fuel library for VVER-440 fuel was used in conjunction with MCNP simulations of the PNAR instrument to estimate the measurement uncertainties from (1) variation in the water boron content, (2) assembly positioning in the detector and (3) counting statistics. The estimated aggregate measurement uncertainty on the PNAR Ratio measurement is 0.008, to put this uncertainty in context, the difference in the PNAR Ratio between a fully irradiated assembly and this same assembly when fissile isotopes only absorb neutrons, but do not emit neutrons, is 0.106, a 13-sigma effect. The 1-sigma variation of 0.008 in the PNAR Ratio is estimated to correspond to a 3.2 GWd/tU change in assembly burnup.

5 citations


Journal ArticleDOI
M. Preston1, A. Borella, Erik Branger1, Sophie Grape1, Riccardo Rossa 
Abstract: The radionuclide composition of, and emitted radiation in, spent nuclear fuel from the future MYRRHA facility have been studied using depletion simulations to understand potential consequences for safeguards verification using non-destructive assay. The simulations show that both the gamma-ray and neutron emission rates in spent MYRRHA assemblies are lower than in spent PWR UO2 and MOX assemblies. In addition, gamma-ray emission rates from 134Cs and 154Eu are considerably lower, and the total neutron emission rate in MYRRHA fuel is much less sensitive to fuel burnup and cooling time. The main reason is that the fast neutron spectrum in MYRRHA affects the radionuclide production in the fuel. One result is that 244Cm, the main contributor to the neutron emission in spent light water reactor fuel, has a limited production in MYRRHA. Consequently, neutron-detection techniques could be used to more directly assay the plutonium content of spent MYRRHA fuel.

2 citations


Journal ArticleDOI
Han Eol Lee1, Man-Sung Yim1
TL;DR: The feasibility of developing an efficient and cost-effective method for detecting partial defects of spent fuel assemblies by using the proposed SPDD method for screening purposes and the cost associated with SPDD was estimated in comparison to that of existing partial defect technologies.
Abstract: As the number of countries with their spent fuel inventory in storage increases, spent fuel verification for nuclear safeguards becomes important. In particular, detection of partial defect, which is the result of local diversion of fuel rods, deserves special attention. This is because accumulation of small scale fuel rod diversions could lead to undesirable consequences. While the necessary detection technologies are available, the cost and detection time associated make it difficult for the partial defect detection technologies to be applied to all spent fuel assemblies. This research investigated the feasibility of developing an efficient and cost-effective method for detecting partial defects of spent fuel assemblies. The approach is based on using the gamma radiation emitted from spent fuel and converting its energy to electric energy. Such approach was incorporated into a new detector concept called, “scintillator based partial defect detector (SPDD)”. SPDD detects the intensity of passive gamma by converting gamma radiation into photons using a CdWO4 scintillator and then into electric current by using amorphous silicon photodiode. Along with the use of detector measurements, a parallel approach of estimating generated electric current by using computation models using the declared spent fuel information is implemented. Detection of partial defects is based on comparing the differences in the electric current between measurements and the expected results. The proposed method was tested using the scenario of detecting 1 SQ (significant quantity) of Pu missing among spent fuel assemblies loaded for a shipment in a typical transportation cask. Results from selected test cases indicated the feasibility of as well as limitations in detecting partial defects by using the proposed method for screening purposes. The irradiation damage to the CdWO4 scintillator of SPDD was also examined for the periods of in-reactor application. In addition, the cost associated with SPDD was estimated in comparison to that of existing partial defect technologies. Purpose of the research To design a cost-effective partial defect detector with fast screening capability. To demonstrate the feasibility of applying the detector in high radiation environment. To setup partial defect detection criterion. To examine the feasibility of applying the developed detector to partial defect detection. Approaches Design a scintillator – photodiode based gamma detector. Develop a method for detecting partial defects. Setup detection criterion for partial defect detection. Analyze radiation damage issue and cost – effectiveness of the proposed detection method. Demonstrate the feasibility of partial defect detection base on examining test case assemblies.

1 citations