Research report summaries 2012–2013

Contractors' reports are only available in the language in which they are submitted to CNSC.

• RSP-0283 – Ontario uranium miners cohort mortality and cancer incidence study
• RSP-0284 – Ageing management of cable in nuclear generating stations
• RSP-0285 – Review of INFO-0286 (1988); Doses from portable gauges
• RSP-0286 – Irradiation effects on material properties for 304L stainless steel base metal and welds
• RSP-0287 – Laboratory characterization, modelling, and numerical simulation of an excavation-damaged zone around deep geologic repositories in sedimentary rocks
• RSP-0288 – Technical report UOME-RBB-2012-04 (Grid spacer simulations)
• RSP-0289 – Review of criteria for assessing shift schedules in the nuclear industry
• RSP-0290 – Estimation of the range of radiation dose for a radon progeny working level due to physical parameters
• RSP-0298 – International CFD benchmark problem

RSP-0283 – Summary report of the Ontario uranium miners cohort mortality and cancer incidence study

This report describes the record linkage work carried out at Statistics Canada for an epidemiological mortality and cancer study of 30,914 Ontario uranium miners who worked for at least one week between 1954 and 2004.

The following record linkages have been carried out: (1) a linkage to the Historical Summary Tax File (HSTF) for the event years 1984 to 2009 inclusive; (2) a linkage to the Canadian Mortality Database (CMDB) for event years 1954 to 2007 inclusive; and (3) a linkage to the Canadian Cancer Database (CCDB) for event years 1969 to 2005 inclusive.

The cohort file was linked to the HSTF to determine the vital status of the individuals at the end of the follow-up period and to enhance the cohort file to facilitate the linkages to the mortality and cancer files. This was possible for 75 percent of individuals who had a social insurance number (SIN) either supplied or found through the use of a composite key generated as part of the work process. On the original cohort file, 63 percent of the records had a valid SIN supplied, while the "exact key" match generated a SIN for another 12 percent of individuals. Records that matched on SIN were then confirmed using other identifying information, such as names and birthdates.

Some 386 records were excluded from further processing because of insufficient identifiers or duplication. Therefore, 30,528 records were linked to the CMDB and the CCDB. Approximately 12,000 potential links to the 1954–2007 CMDB were reviewed. During the manual resolution phase, the team consulted detailed reports of common identifiers from the Ontario Uranium Miners Cohort and the CMDB.

In total, 8,794 (28.8) mortality links were found. There were 97 deaths identified on the HSTF and not in the mortality linkage. These could potentially be individuals who died outside of Canada. Some, 16,455 individuals (53.9 percent) were determined to be alive at the end of the follow-up period and 5,182 (16.9 percent) were lost from follow-up prior to the end of 2007.

Finally, a cancer linkage was carried out using the CCDB for the years 1969 to 2005. The pockets and rules used for this linkage were similar to those used for the mortality linkage. After manually resolving the potential links, there remained 4,320 accepted links to the CCDB.

Read the RSP-0283 Final Report (PDF)

RSP-0284 – Ageing management of cable in nuclear generating stations

Power, control and instrument cables are intrinsically important to safe and reliable nuclear power plant (NPP) operation. International experience has indicated that, as operating plants have aged, the number and rate of cable failures has increased, implying that degradation due to ageing may be a contributing factor. With extension of plant life beyond the original design life of cables, the risk of failure rates increasing due to ageing degradation also increases and this may introduce new challenges to long-term plant safety and reliability. Based on analysis of the available international and Canadian NPP cable failure history, the majority of cable failures have been related to random installation/maintenance damage or age-induced degradation due to adverse service conditions, particularly medium voltage (MV) cables in submerged environments. There has been no indication that cables in typical low-stress environments are prematurely failing due to ageing.

The nuclear power industry has recognized that a comprehensive and coordinated cable ageing management program is needed to provide reasonable assurance that age-related degradation does not significantly affect the long-term safe and reliable operation of NPPs [13.4][13.7][13.17][13.35].  Much effort and resources have been dedicated over the past 20 years to establishing the bases for such a program.

Information was gathered from representatives of the operating Canadian NPPs related to their experience with cable reliability and current programs or practices employed to address cable ageing [13.112][13.113][13.114][13.115][13.116]. Canadian NPP configurations and cable failure experience was assessed against international research, experience and regulatory guidance. It was concluded that the current international guidance on cable ageing management is generally appropriate and recommended for Canadian NPPs. 

Recommendations for Canadian regulatory guidance were developed on this basis, and core program elements were established, consistent with USNRC regulatory guide RG1.218 [13.35]. The following 11 core elements were defined as essential to an effective cable ageing management program (Section 12).

• Definition of “scope of cable” to be addressed by program.
• Develop and maintain a database of all cables to be monitored.
• Characterize and monitor service environments.
• Identify stressors and expected ageing mechanisms.
• Select condition monitoring techniques suitable to monitored cables.
• Establish baseline condition of monitored cables.
• Identify cable characteristics/ageing effects monitored by each CM technique.
• Periodically perform CM tests and inspections on cables.
• Periodically review and incorporate plant and industry experience.
• Periodically review, assess, and trend the condition of monitored cables.
• Identify degraded conditions and define/take corrective actions
  

Section 6 through 10 provide a review of Canadian NPP cable ageing management practices against the above CAMP core elements. While information from the NPPs was limited, it could be determined that none of the NPPs currently have a comprehensive CAMP that addresses all of the above core element requirements. Program gaps vary between NPPs but further information and analysis is required to accurately define the full extent of these gaps.

Read the RSP-0284 Final Report (PDF)

RSP-0285 – Review of INFO-0286 (1988); Doses from portable gauges

Stantec had been experimenting with calculated exposure since 2002. We had been using multipliers to calculate the exposure of our employees by counting three parameters and by assigning a multiplier value for each. Our results were balanced against the dosimetry used by some of the employees to adjust the multipliers. Our pool of data was minimal and we knew we needed a larger pool of data to enhance our performance. This limited tool allowed us to calculate a likely exposure for our employees especially those with little use of portable gauges; it was most useful when trying to estimate the exposure to an employee following an incident or an overexposed dosimetry. The CNSC was looking for a data pool backed by a record of activities carried out by the employees. They were seeking data to answer a burning question. What was the neutron dose contribution to a portable gauge operator? Previous research had only a limited data set because of a small sample size and no modeling was possible to study the exposure because there was no activity log available to perform the task. During our attendance at a Radiation Conference someone had noted that we had common interests and introduced us to one another.

A research project was initiated and the data was gathered from field personnel wearing a Landauer® dosimetry combo with type TA1 Luxel® for gamma and CR39 for neutrons. The one year assessment project, with deployment of radiation monitors, was initiated on June 1st, 2011, and was terminated on May 31st, 2012.

The employees submitted monthly exposure reports detailing key personal exposure information parameters as per the study terms which included:

• the hours transporting a gauge
• the number of shots taken
• the time taken to perform maintenance on the gauge as per manufacturer’s instructions

Once the assessment portion of the project was completed, the exposure report was matched to the totals for the three parameters listed above, to identify the most likely contributor of gamma/neutron exposure to density gauge operators.

Read the RSP-0285 Final Report (PDF)

RSP-0286 – Irradiation effects on material properties for 304L stainless steel base metal and welds

A literature survey was conducted to understand the irradiation effects on the mechanical properties of 304L stainless steel, base, and weld metals. The objective was to identify the pieces of information that will allow for the assessment of suitability of data for the estimation of the end-of-life properties of CANDU calandria vessels after 60 years in service. The focus was on the magnitude of the fracture toughness of irradiated 304L stainless steel. Most of the data reported in the literature are for high irradiation and test temperatures that are conditions pertinent to fast reactors and light water reactors (in which operating temperatures are above 350○C and 280○C, respectively). Based on the surveyed literature, one may infer that valid fracture toughness testing of 304 stainless steel specimens irradiated to dose of 4.1 dpa at temperatures 25 to 125○C, which are conditions that may be deemed close to those prevailing in the calandria, indicates that the heat affected zone (HAZ) of welds has a minimum fracture toughness KIC ≈ 108 MPa√m. Fracture toughness tests not meeting all validity size requirements indicate that the HAZ may be the weakest part of the structure. 

Read the RSP-0286 Final Report (PDF)

RSP-0287 – Laboratory characterization, modelling, and numerical simulation of an excavation-damaged zone around deep geologic repositories in sedimentary rocks

This report contains the following: a literature review; experimental data for the mechanical behavior of Tournemire argillite; a constitutive model for the mechanical behavior of Tournemire argillite; a numerical simulation of the excavation-damaged zone (EDZ) around an existing gallery at the site of Tournemire Underground Research Laboratory, France; and a numerical simulation of an EDZ around a proposed DGR in Cobourg limestone at the Bruce site, Canada.

The experimental program includes the measurements of the physical properties of the transversely isotropic material of Tournemire argillite and its mechanical response under uniaxial tests, triaxial tests with different confining pressures, unconfined and confined cyclic tests, Brazilian tests, and creep tests.

Based on the experimental data, an elastoplastic model has been developed to simulate the mechanical behavior of the Tournemire argillite. The model is formulated using the classical theory of plasticity and continuum damage mechanics concepts. The effects of inherent anisotropy on the deformation and strength properties are incorporated in the model. All model parameters are estimated from the experimental data.

The numerical simulations consisted of plain strain finite element analyses of the coupled hydro-mechanical processes around the gallery at the Tournemire site and a hypothetical emplacement room at Bruce site. The influence of the inherent anisotropy on the development of the excavation damaged zone is investigated.

Read the RSP-0287 Final Report (PDF)

RSP-0288 – Technical report UOME-RBB-2012-04 (Grid spacer simulations)

Numerical simulations of turbulent flows in a rod bundle with two types of spacer grids (split-vane type and swirl-vane type) have been performed within the framework of an international blind CFD benchmark exercise. Details of the test geometry and conditions were provided by the Korea Atomic Energy Research Institute (KAERI), South Korea. The model of the computational geometry was based on the provided CAD files, with some parts simplified in order to avoid deterioration of the mesh quality.

In the present analysis, we used the segregated turbulence model, consisting of a combination of SAS (Scale Adaptive Simulations, which are second generation solutions of the unsteady Reynolds-averaged Navier-Stokes equations) and LES (Large Eddy Simulations). This hybrid approach achieves the accuracy of LES at a computational cost that is lower than that of full LES. The simulations employed the second-order central difference scheme for the discretization of the convection terms of the transport equations, and the second-order implicit Euler scheme for temporal discretization. As inlet boundary condition for the velocity distribution, we used results of steady, separate simulations of flow in the bare rod bundle.

On May 8, 2012, our numerical results for the split-vane spacer grid, using 22M cells, and the swirl-vane spacer grid, using 20M cells, were submitted to the organizer (KAERI) in the form of time-averaged velocity and rms velocity fluctuation profiles in quadrant sections of planes located 0.5Dh, 1.0Dh, 4.0Dh, and 10.0Dh downstream of the spacer grid tips. Contours of streamwise vorticity and circulation values on these planes were also provided.

KAERI received 21 submissions for the split-vane spacer grid case, and ranked our time-averaged velocity data at z = 1.0Dh in the 11th place and those at z = 4.0Dh in the 3rd place. KAERI received 14 submissions for the swirl-vane spacer grid case, and ranked our time-averaged velocity data at z = 1.0Dh in the 1st place and those at z = 4.0Dh in the 4th place. For the split-vane spacer configuration, fair agreement was obtained between the predicted lateral and spanwise velocities and the experimental ones; however, the predicted streamwise velocity was far from the measured one. This difference is attributed to the low mesh quality. For the swirl-vane spacer configuration, the predicted velocity showed excellent agreement with the experimental data; however, the predicted turbulence stress values for both spacer grid simulations were significantly lower than the experimental results. This is attributed to the mesh in the downstream rod bundle zone being too coarse.

To improve the prediction of the turbulence level, we conducted a refined simulation of the split-vane spacer configuration, using 40M cells after refining the mesh of the 22M cell case in the downstream rod bundle zone. The simulation results were in good agreement with the experimental data. This indicates the importance of the mesh quality and mesh size, especially in the downstream rod bundle zones, where LES was applied.

Read the RSP-0288 Final Report (PDF)

RSP-0289 – Review of criteria for assessing shift schedules in the nuclear industry

The CNSC is planning to formalize their fatigue management and hours of work requirements in a regulatory document intended for use by licensees of nuclear power plants and facilities. The CNSC currently specifies mandatory rest periods and limits to hours of work that are based on criteria originally developed in 1989 and later updated in 2005. Accordingly, it is important to review the existing 2005 criteria and to identify any opportunities for improvement, in terms of current research, evidence-based best practices, and benchmarks.

Human Factors North Inc. (HFN) was contracted by the CNSC to conduct research to identify gaps between the 2005 criteria and current scientific evidence and benchmarking related to hours of work and managing fatigue. The specific project objectives were to: update the review of the literature; compare the existing 2005 criteria to current science and benchmarking on hours of work and managing fatigue; identify opportunities for improving the existing 2005 criteria; and make recommendations for improving the existing 2005 criteria that are supported by findings from benchmarking and the review of the literature.

Research and unpublished "grey" literature, including regulatory documents, for the period 1995 to November 2012 were searched. The search strategy included a variety of search terms relating to shift work, hours of work and rest, and scheduling. Some of the literature reviewed referred to recommendations made by various industry organizations regarding hours of work and rest. Comparisons were made between the 2005 criteria and these benchmarks in order to generate best practice guidelines and recommendations for the Canadian nuclear industry. Project objectives were to answer five broad research questions. The questions and their answers are summarized below:

1. Are the hours of work limits and mandatory rest periods in the 2005 criteria (see Appendix in full report - 2005 Hours of Work Limits and Mandatory Rest Periods) aligned with current science and benchmarking?
   
   Yes, with the exception of the following:
   1. The maximum night shift (i.e. a shift including the period between midnight and 5:00 a.m.) duration should be 12 hours; otherwise, a day shift could be extended to 16 hours in a 24-hour period. (Original limit allowed 16 hours in 24 hours on rare occasions with no restriction to day shifts.)
   2. The number of hours worked in a 48-hour period shall not exceed 26. (Original limit was 28 hours.)
   3. The maximum number of hours worked in a 7-day rolling period shall not exceed 60. (Original weekly limit was not a rolling limit.)
   4. Work hours should be limited to 260 hours for a 5-week cycle. (Original limit was 268 hours for a 5-week cycle.)
   5. For the purpose of determining compliance with the limits, all time should be included from the time that the worker reports to work until the time that the worker is relieved from all responsibility for work, including unpaid lunch or rest breaks, with the exception of restorative naps (with guidance as to appropriate conditions for naps to be established). (Original limit was not to make an exception of naps.)
   6. For blocks of 12-hour shifts,
      1. A minimum recovery period of 48 hours shall follow a block of 5 consecutive day shifts, excluding shift turnover time (original limit was to have 48 hours recovery after 3 consecutive day shifts.)
      2. Due to lack of specific studies supporting such a requirement, no minimum recovery period was specified after a block of 3 or 4 consecutive day shifts or after a block of 2 consecutive night shifts.
   A block, sometimes called a "workset", is defined as a set of consecutive shifts with the same start and end times, that is followed by a minimum recovery period and a subsequent set of consecutive shifts.
2. The current limits on hours of work apply to a day, week, shift cycle, and year. Should limits be set for other time frames?
   
   At this time, there is no scientific evidence to suggest that limits for time frames other than those mentioned above are required.
3. What mandatory rest periods and limits to consecutive shifts should be applied to those working 8-hour or 10-hour day, evening and/or night shifts?
   
   For 10-hour shifts, the recommended maximum is 5 consecutive day shifts or 4 consecutive night shifts. (Original limits did not include mandatory rest periods following a block of 10-hour shifts.)
   1. A minimum recovery period of 48 hours shall follow a block of 5 consecutive day shifts, excluding shift turnover time.
   2. A minimum recovery period of 72 hours shall follow a block of 3 or 4 consecutive night shifts, excluding shift turnover time.
   For 8-hour shifts, the recommended maximum is 6 consecutive day or evening shifts or 5 consecutive night shifts. (Original limits did not include mandatory rest periods following a block of 8-hour shifts.)
   1. A minimum recovery period of 36 hours should follow blocks of either 5 or 6 consecutive day or evening shifts, excluding shift turnover time.
   2. A minimum recovery period of 48 hours should follow blocks of either 4 or 5 consecutive night shifts, excluding shift turnover time.
   3. A forward direction of shift rotation should be used for those individuals working 8-hour shifts.
4. Is there any basis for granting exceptions to the hours of work limits or rest periods for short durations at times of peak demand? (If yes, recommend evidence-based, permissible exceptions to the hours of work limits or rest periods and the duration of these exceptions.)
   
   It should be noted that working at the limits for hours of work and rest is not sustainable and may lead to chronic sleep debt. Without decreasing safety, there is no scientific basis for allowing exceptions to the hours of work limits or rest periods.
5. Using evidence from benchmarking and research, what fatigue management provisions, including hours of work limits and mandatory rest periods, are appropriate during construction of a facility that will require high reliability operations, such as a nuclear power plant?

Individuals who perform construction work on safety relevant facilities should be covered under the same CNSC hours of work regulations as power plant operators or others with safety sensitive roles.

Read the RSP-0289-Final Report (PDF)

RSP-0290 – Estimation of the range of radiation dose for a radon progeny working level due to physical parameters

The traditional metric for exposure to uranium miners has been the working level month (WLM), a measure of exposure to radon progeny. This has been the approach used internationally to manage and regulate exposures of miners to radon and its short-lived decay products.

For the purpose of developing a system of radiological protection, it has been necessary to compare doses from radon and its short-lived decay product to doses from other sources of radiation exposure; hence, a dose conversion convention (DCC) is required. The International Commission on Radiological Protection (ICRP) developed guidance in this regard. ICRP 65 (1993) proposed a convention to convert radon exposure (in WLM) to effective dose by dividing the risk of lung cancer derived from epidemiological studies of miners by the detriment from external radiation based primarily on data from follow-up on Japanese atomic bomb survivors.

In addition to the epidemiological approach, a dosimetric approach (i.e., one relying on
biokinetic and dosimetric modelling) can also be used to estimate dose to the lungs (or regions of the lungs) and with the help of subjective weighting factors, the corresponding effective dose can be estimated and used for radiation protection purposes. In its 2009 statement on radon, the ICRP proposed to move from its current epidemiologically based approach (i.e., DCCs) to treat radon and progeny in the same way as other radionuclides and to publish dose coefficients calculated using dosimetric models for use within the ICRP’s system of protection.

The primary objective of the study is to understand “how” and “to what degree” the environmental factors within a mine can affect the conversion of airborne concentrations in radon decay products (RDP) in working levels to dose estimation.

The dosimetry-based dose from radon decay products depends on the characteristics of the mine environment including, combinations of several aerosol inputs including alpha energy (typically measured via WL), the particle size distribution of the mine workplace aerosols, including the unattached fraction, and if radon gas is measured instead of radon progeny, the equilibrium factor (F) between radon and its progeny is important. Thus, this study

• looked for dosimetrically relevant information on mine environments and  concluded that relevant data is very limited
• looked at currently available methods and equipment for measuring
• dosimetrically relevant parameters and concluded that at this time, off-the-shelf equipment is not available
• estimated the range of dosimetrically relevant parameters for conditions in modern underground uranium mines; radon progeny doses per WLM (provided by the Health Protection Agency (HPA) using the HPA’s implementation of the ICRP’s HRTM model)
• simulations were performed to provide an estimation of the variation in absorbed dose conversion factor for hypothetical mine environments and exposure scenarios.

In mines, all the parameters mentioned above vary widely within just a single mine. Parameters vary from place to place with mining activity and ventilation and over time. Due to the changing characteristics of a mine’s atmosphere, the parameters can not only vary between mines, but also between different workstations, and as operations and conditions change within a single mine site.

Coming up with standard values of atmospheric parameters that can be applied across all jobs in uranium mines is therefore extremely difficult. The simulations described in this report consider picking an average WL and associated particle size distribution for each of three hypothetical workplaces. The particle size determines the workplace-specific dose conversion factors. The dosimetric factors used in the current simulations are presented in units of mGy/WLM and the effect of radiation and tissue weighting factors (used to convert from absorbed dose to lung to effective dose) are considered separately.

The DCFs (as a function of particle size) were combined with workplace WL and the time a miner in each workplace to predict the dose received (WLM) in each workplace. The WL and WLM are summed across the workplaces and a weighted DCF is determined. For the hypothetical mine environments and worker exposure scenarios considered in this report, the DCFs for annual doses were estimated to range from about 6 to 10 mGy/WLM for both of the two different worker types considered.

Our overall conclusions are that:

• dosimetrically relevant data for modern uranium mines is very limited
• suitable commercial (off-the shelf) equipment for measuring dosimetrically relevant parameters required for modelling is not currently available
• there are considerable uncertainties associated with the implementation of a fully dosimetric approach

In broad terms, further research, advancements in current measuring practices and relevant data on modern mine environments is needed and thus, until such time as such data is available, we recommend continuing with current practice for monitoring, reporting and regulating miner’s exposure to radon progeny.

Read the RSP-0290 Final Report (PDF)

RSP-0298 – International CFD benchmark problem

Computational Fluid Dynamics (CFD) is being increasingly utilized in the design, analysis and licensing of nuclear power stations. Internationally CFD has been used to assess mixing vane spacer designs, to examine boron dilution problems, t-junction induced aging and other localized or 3-dimensional phenomena. In Canada, CFD has been applied in header geometries to study header pressure and temperature gradients, flow and turbulence generation inside fuel bundles and in safety and licensing applications related to moderator flow and temperature distributions. This report documents the scope and results for a CFD study and comparison to experimental results from the latest OECD-NEA computational fluid dynamics benchmark.

Prediction of subchannel flows, even in isothermal conditions, is very challenging. Complicated flow structures, mixing in the gap region, and even unsteady pulsing type behavior all produce uncertainties in any predictive models. This is made more complicated by the presence of grid spacers in LWR assemblies, or endplates in CANDU fuel, which cause rigorous mixing as well as greatly increasing the local turbulence levels. In many historical studies, subchannel thermalhydraulic codes such as COBRA or ASSERT-PV have been used to predict subchannel flows within fuel bundles. However these subchannel codes rely on empirically derived mixing coefficients, hydraulic loss factors, and empirical shear stress relationships in closing the system of equations. The advantage of a CFD code for subchannel predictions is that it does not rely so heavily on geometrically dependent mixing factors and empiricisms. Hence CFD results have the potential for wider applicability, notwithstanding their needs for adequate validation and testing. The study of CFD code applicability and accuracy has been the topic of a large number of validation exercises as well as international benchmarks.

The OECD-NEA has organized a new benchmark to study CFD applications to bundle flows and turbulence levels. This benchmark involved taking new experimental data in a rod-bundle geometry including grid spacers and taking detailed point-wise time-average and rms velocity measurements.  These data were hidden from the benchmark participants until after the deadline for submitting contributions.  Thus the CFD predictions were “blind” from this respect. The benchmark team released the specifications, geometries and boundary conditions for this experiment¹ in 2011, and participants in the CFD benchmark were asked to submit their results in May 2012. After submissions, the OECD performed an assessment of the submitted CFD results against the experimental data and also released the results to the CFD participants so they could do their own comparisons.

Read the RSP-0298 Final Report (PDF)