Language selection


Research report summaries 2015–2016

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

RSP-617.1 – Expert review of “Technical Basis for Probabilistic Assessments of Pressure Tube to Calandria Tube Contact and Blister Susceptibility”

Until 2012, the CANDU industry used a probabilistic methodology developed by AMEC-NSS to assess potential contact between the pressure tube (PT) and calandria tube (CT) and associated PT failure. The methodology developed by AMEC-NSS utilized a logistic model for the spacer movement and conservative assumptions for the PT creep ratio and endslopes. The AMEC-NSS methodology used the C-DEPTH code to assess the potential contact between the PT and CT for a given channel. It should be noted that during that time, the CANDU industry was developing capability to physically measure the gap between the PT and CT; however, the AMEC-NSS methodology is not based on the measured gap, as the measurement technology had not matured in 2006, so AMEC-NSS methodology assumed conservative values for the PT creep ratio and endslopes.

However, as the PT to CT gap measurement technology matured, the CANDU industry developed a new methodology based on Candu Energy Inc’s (CEI) approach to assess potential contact between the PT and CT and associated PT failure. This methodology used the measured and normalized PT to CT gap to derive PT creep ratio and endslopes for a given reactor core and derived a distribution for these two variables. This derived distribution is used in the probabilistic assessment to assess the PT failure due to potential contact between PT and CT.

The Canadian Nuclear Safety Commission initiated this research project to complete an expert review on the new methodology and associated technical basis. The new methodology utilized the following newly developed techniques:

  • Statistical techniques to normalize/modify the measured PT-CT gap data
  • Benchmarking techniques to derive the PT creep and endslopes from the measured gap data
  • Garter spring spacer movement model exclusively based on statistical treatment of the available limited spacer movement data

The scope of work involved:

  • providing a scientific review of the methods used in the probabilistic assessment methodology proposed by the CANDU industry
  • comparing the proposed model calibration methodology with established methods of statistical inference
  • examining the adequacy of the interpretation of probabilistic assessment
  • identifying any gaps and recommending potential improvements in the proposed methodology

Read the RSP-617.1-Final Report (PDF)

RSP-614.1 – Flood hazard assessment for nuclear power plants in Canada

The 2011 Fukushima Daiichi Nuclear Power Plant accident in Japan, which was caused by a large earthquake and associated tsunami, demonstrated the destructive power of flooding. It urged operators and regulators around the world to review design and operational procedures to enhance flood protection at nuclear facilities.

Before considering the need to enhance the regulatory framework for assessing flood risk, a survey of current best practices is required. This study compares best practices in flood hazard assessment nationally and internationally, with the objective of identifying potential gaps and areas for potential improvement, if necessary.

The report provides a literature review of national and international state-of-the-art approaches to external flood hazard assessments, with an emphasis on nuclear power plants. A gap analysis of flood hazard assessment approaches taken by CNSC licensees was completed by reviewing publicly available licensees’ reports. All causes of flood relevant to Canadian facilities, including extreme precipitation, ice jams, snowmelt and seiche, were considered. Based on the literature review and gap analysis, an assessment of the current state of the CNSC’s regulatory framework on flood hazard assessment for existing and potential new power plants was made.

With respect to continuing improvements, the report identifies inconsistencies in the way licensees carry out flood assessments and further recommends regulatory guidance for harmonizing the assessment of various flood risks that may be encountered at Canadian facilities.

Read the RSP-614.1-Final Report (PDF)

RSP-612.1 – Application of Bayes method in evaluation of ROP/NOP trip setpoint

Regulatory independent evaluation of the shutdown systems’ neutron overpower protection (NOP) trip setpoint under aging conditions is necessary for verification and confirmation of the adequacy of licensees’ proposed values of the installed trip setpoints. The intent of the NOP in-core detector system is to initiate a reactor shutdown whenever the neutron flux reaches a level that is considered to be too high at any place in the reactor core. Regional overpower protection (ROP) has the same meaning as NOP.

A multi-phase research project has been undertaken to develop a Bayesian statistical framework (SF) and software to support CNSC staff’s assessment of licensees’ proposed ROP/NOP trip setpoints under aging conditions and support risk-informed decision-making (RIDM) processes, when necessary.

CNSC staff decided to use a Bayesian approach because the NOP/ROP methodology is based on a probabilistic model. By design, Bayesian methods natively consider the uncertainty associated with the parameters of a probability model (even if those uncertain parameters are believed to be fixed numbers) and are often recommended and used (as proven practice) in areas of reliability and probabilistic assessment as the proper way to make formal use of subjective information.

In the context of this project, the Bayesian SF is proposed to estimate functional failure distributions for a NOP system with fixed design control variables (number of detectors, location of detectors and value of trip setpoint).

Specifically, the Bayesian SF should answer the question: What is the estimated frequency of NOP system (functional) failure and the uncertainty around that estimate?

During this first phase of the project a theoretical framework using the Bayes method was developed. The report includes:

  • the theory and model of proposed Bayesian SF
  • the software specification
  • guidelines for the development of benchmark cases and selection of performance criteria for testing and qualification 

Software using this framework is to be developed in a second phase.

Licensees’ staff also reviewed the contractor’s final report and provided their comments. The comments and the disposition of those comments are included in the appendix of the final report.

Read the RSP-612.1-Final Report (PDF)

RSP-606.1 – Synchrotron X-ray microspectroscopy for detecting Ra-226 and its daughters in tailings and mill waste from uranium mining operations

Understanding the stability and form of long-lived radionuclides is a key element of the CNSC’s oversight of the long-term behaviour of tailings management facilities.

This report provides an independent evaluation of initial CNSC studies on the use of synchrotron techniques for characterizing the fine structure and forms of key radionuclides (radium (Ra) and its parents, uranium (U) and thorium (Th)) in representative tailings from a modern operating mine in Canada.

Thin sections were prepared from samples provided by AREVA from the McClean Lake Tailings Management Facility in 2008. These were analyzed by CNSC staff at the 13-ID-E hard X-ray microprobe at the Advanced Photon Source in Argonne, Illinois, USA. Application of these techniques to these materials was successful and informative for uranium, but not for radium and thorium. Attempts to detect radium X-ray L emission lines in these samples were unsuccessful. This was due to the much lower Ra concentrations coupled with difficulties in resolving the Ra emission lines from certain other elements (lead in particular). The technique may be viable for Th for tailings from other mines with higher amounts of Th.

Mapping of U distribution by µXRF (X-ray fluorescence) showed that U was spatially dispersed and not just localized in high U particulates. Characterization of particles with high U by µXRD (X-ray diffraction) did not identify the presence of distinct, crystalline U minerals. The µXRD analyses did show, however, that tailings contain abundant illite, hematite and gypsum. Uranium L3 XANES results (µX-ray absorption near edge spectroscopy) showed that U was dominantly in the hexavalent state.

Read the RSP-606.1-Final Report (PDF)

RSP-598.1 – Feasibility study of an integrated framework for characterization of uncertainties with application to CANDU steady state and transient reactor physics simulation

The current CNSC REGDOC-2.4.1, Deterministic Safety Analysis allows for use of best estimate (BE) reactor analysis simulations with consideration of uncertainties. Quantification and understanding of uncertainty sources is an essential requirement of BE analysis, as it provides a reliable metric by which the quality of the predictions can be assessed. Although direct comparison against measurements provides the ultimate evidence that simulation predictions are reliable, the true value of any BE simulation lies in its ability to analyze reactor conditions for which measurements are unavailable. Therefore, there is a clear need to characterize, i.e., to propagate and prioritize, all sources of uncertainties in order to reliably use the results of BE calculations in the various aspects of reactor design, operation and safety.

In preparation for this, the CNSC initiated this study to investigate the feasibility of the development of a first-of-a-kind integrated framework for uncertainty characterization (UC) with primary application to CANDU neutronics calculations. The goal was to provide a comprehensive and scientifically defendable methodology for characterizing uncertainties in all BE reactor analysis calculations, including both steady state and transient simulations, to be used in independent verification of licensees’ safety case. The framework was based on open-source libraries for standard UC process algorithms as well as novel algorithms, to be developed and implemented.

The scope of this project included identification of needs and key challenges in developing and coding the UC framework (UCF). The UCF was developed to accomplish four primary functions: first, to identify all sources of uncertainties resulting from modeling assumptions, numerical approximations, nuclear data uncertainties, and technological parameters uncertainties; second, to propagate the identified uncertainties to the responses of interest such as the core eigenvalue, power distribution, bundle enthalpy rise, etc.; third, to map the propagated uncertainties to the wide range of operating conditions; and lastly, to generate a priority identification and ranking table (PIRT) which identifies and ranks according to importance the dominant sources of uncertainties.

Read the RSP-598.1-Final Report (PDF)

RSP-587.1 – Mortality risks in the pooled analysis of the Canadian and German uranium processing workers

Epidemiological studies have been used to examine potential health risks for workers involved in the nuclear fuel cycle. Studies of underground uranium miners, in particular, have shown increases in lung cancer risk from exposure to radon decay products (RDP). Workers engaged in uranium processing (i.e., milling, refining, etc.) can receive radioactive and non-radioactive exposures from the uranium ore dust but are less exposed to RDP typical of uranium mining workers. The potential exposures of workers in the uranium processing industry are substantially different from those of uranium underground miners or other workers in the nuclear fuel cycle and should be carefully evaluated in separate studies. Only a few studies have examined the risks of exposures from uranium processing and have had contradictory results, necessitating further research in this area.

The purpose of this study was to develop an improved understanding of the long-term health risks of occupational exposures in the uranium processing industry. The study objectives were:

  • to examine the radiation-related risks of mortality from site-specific cancers, with special attention to cancers of the lung and bronchi, bone, liver and kidney as well as non-malignant respiratory, and to leukemia and lymphoma and renal and cardiovascular disease (CVD)
  • to determine the effects of exposure to radium and silica dust on the radiation-related risks of mortality
  • to investigate the radiation-related risks of mortality in women

Data from previously published studies on Canadian (Port Hope) and German (Wismut) uranium processing workers were pooled to create one of the largest cohort analyses for workers in this industry. Small, non-significant increases in risks of lung cancer due to RDP exposures and of CVD due to both RDP and gamma-ray exposures were observed among males. Risks of solid cancers, excluding lung cancer, were increased both for RPD exposures and for gamma-ray doses, significantly so among women, although this latter finding should be interpreted with caution. Continued follow-up of the Canadian and German cohorts, and pooling with other cohorts of workers in the uranium processing industry, could provide valuable insight into radiation-related risks for these workers and into suspected differences in risks with other nuclear fuel cycle workers.

Read the RSP-587.1 Final Report (PDF)

RSP-584.1 – Minimum staff complement: A review of regulatory requirements, industry practices, scientific literature and experience of stakeholders

The CNSC initiated this research project as a precursor to reviewing and updating Regulatory Guide G-323 entitled “Ensuring the Presence of Sufficient Qualified Staff at Class I Nuclear Facilities – Minimum Staff Complement”.

CAE was tasked to review regulatory requirements, industry practices and scientific literature related to minimum staffing from a range of industries. Additionally, interviews were conducted with internal and external stakeholders to obtain feedback about their experience using G-323.  The review’s scope included practices for Beyond Design Basis Accidents (BDBA).

The literature review identified several similarities and differences between nuclear regulators and industries with traits similar to nuclear (e.g., aviation, marine, chemical). Multiple industries rely heavily on human factors standards to support the determination and validation of the minimum staff complement (MSC). Some industries focus on MSC during design and procurement of equipment, but attend less to MSC during operations. Some industries, including nuclear, require extensive justification to support requests to reduce staffing levels. In comparison to other requirements, strengths of the CNSC’s approach are that G-323 is a document focused on staffing level that explicitly addresses periodic reviews, MSC change requests and licensed and non-licensed workers.

Many tools exist to assist in setting the MSC with all industries using some form of task analysis. The use of simulation, especially elements of constructive and virtual simulation, hold promise for the nuclear industry as part of a systematic approach to setting and validating staffing levels for accidents within and beyond the design basis.

Although the review of literature shows that the term “Beyond Design Basis Accident” is primarily used in the nuclear industry, lessons can be drawn from events that extended beyond the design basis in other industries. Such lessons include the advantage of having additional expert staff available and integrating them effectively within the “core” staff, the value of staff having strong mental models of the system, and the importance of clear lines of accountability.

Based on feedback from interviews, stakeholders indicated that the MSC is an important defence against accidents and that G-323 supports a more rigorous, scientific-based approach to determining and managing MSC. Stakeholder interviews support the following high-level recommendations:

  • move G-323 from a guideline to a regulatory document with clear requirements and guidance.
  • provide more direction regarding the level of fidelity, selection of scenarios, and metrics for confirming the MSC is valid
  • include guidance specific to decommissioning
  • integrate BDBA considerations

The report includes many best practices and recommendations arising from the literature review and interviews.

Read the RSP-584.1-Final Report (PDF)

RSP-557.1 – Assessing regulatory requirements and guidelines for the single failure criterion

This report describes and compares approaches to applications of the single failure criterion (SFC) principle in nuclear power plant (NPP) design and operation internationally. The SFC requires that NPP safety systems preserve their safety functions in the presence of any single and resultant failures. This report outlines benefits and disadvantages of different national implementations of the SFC, as well as suggested improvements for the application of the SFC principle. The main focus is on the Canadian practice, as the project was initiated by the Canadian Nuclear Safety Commission for Canadian regulatory framework development. The presented analysis reflects the results of the authors’ research and understanding of current national and international practices and views for improvements in the described area.

The report provides an overview of the regulatory design requirements for new reactors addressing the SFC in accordance with international best practices, particularly considering the SFC relation to in-service testing, maintenance, repair, inspection and monitoring of systems, structures and components important to safety.

The scope of the work includes:

  • a review and comparison of the current SFC requirements and guidelines published by the International Atomic Energy Agency, the Western European Nuclear Regulators Association, the European Utility Requirements and nuclear regulators in the United States, United Kingdom, Russia, Korea, Japan, China and Finland (this review addresses the application of SFC requirements in design; considerations for testing, maintenance, repair, inspection and monitoring; allowable equipment outage times; exemptions to SFC requirements; and analysis for SFC application to  two-, three- and four-train systems)
  • identification and analysis of how Canada and the above-mentioned countries differ with respect to SFC requirements and how the SFC is applied

Read the RSP-557.1-Final Report (PDF)

RSP-531.1 – Characterization of alpha radiation hazards: biosolubility of radionuclides within CANDU reactor aerosols and implications for internal dosimetry

Experience at Canadian CANDU facilities has shown that workers may be exposed to aerosols contaminated by alpha emitting transuranic radionuclides. Intakes of transuranics – mainly plutonium, americium and curium isotopes – could occur during refit and maintenance operations. Dose calculations for inhalation intakes of transuranics are generally accomplished by measuring a major component (e.g., 239Pu) in excreta and then inferring the contribution of other transuranics based on the relative abundance of these radionuclides in the workplace. The underlying assumption in this approach is that all transuranic radionuclides within the inhaled aerosol are leached from particles at the same rate in the respiratory tract. This had yet to be verified for transuranic-bearing aerosols of CANDU origin. Without such data, urinary and faecal excretion rates of transuranics in an exposed worker are compared, over several weeks or months of monitoring, in order to infer the clearance rates of transuranics in the respiratory tract. The dose from inhaled transuranics hinges on this latter parameter.

The objective of this study is to characterize the biosolubility of radionuclides associated with irradiated CANDU fuel. Radionuclides investigated in the study include transuranics and fission products.

The biosolubility of particles of un-irradiated and irradiated uranium dioxide, as well as transuranics and fission products was measured using in vitro and in vivo techniques. The biosolubility of irradiated uranium dioxide, transuranics and fission products was measured for various fuel burnups to characterize the effect of burnup on the solubility of such materials. In vitro techniques examined the dissolution rates of various materials in simulated lung fluid and include i) a sequential extraction method developed by the National Institute of Standards and Technology, ii) a static solubility test, and iii) a flow-through solubility test. The in vivo technique involved exposing rats to aerosols by intratracheal instillation.

The study sought to explore the following areas: the solubility of fuel as a function of burnup in a CANDU reactor; the leach rates of radionuclides from CANDU fuel particles; the solubility of radionuclide contaminated pressure tube and feeder pipe residues; and the inhalation solubility class of the materials investigated compared with those defined by the International Commission on Radiological Protection.

Read the RSP-531.1-Final Report (PDF)

RSP-493.1 – Permafrost degradation within continuous permafrost zones due to mining disturbances in Canadian northern regions

Uranium mines generate vast quantities of waste – tens of millions of cubic metres of mine tailings and mine waste rock. Successful management of these wastes is important for environmental protection. This success relies on the integrity of the tailings pits and the proper design of the tailings management facilities (TMFs).

Areva Resources Canada Inc. has submitted a project proposal for mining and milling uranium at Kiggavik, NU. This would be the first uranium mine and mill facility within the continuous permafrost zone of Canada, should the project be licensed under the Nuclear Safety and Control Act. Because permafrost degradation (thawing) affects the integrity of the tailings pits and TMFs, predicting this degradation is important for both environmental assessment and licensing decisions. The success of the Kiggavik project would rely on the integrity of the permafrost.

Currently, it is not clear how the permafrost beneath the TMFs would be changed by the mining and waste management activities. The tailings ponds will act as a heat source to the permafrost below and may cause its thawing. Global warming could also further exacerbate these disturbances. The understanding and prediction of permafrost behaviour (degradation and its rate) beneath the tailings ponds, as well as the long-term predictions of the thermal regime around the TMFs under climate changes, are fundamental for assessing the potential adverse environmental effects of projects in arctic and sub-arctic regions. This research project will provide independent knowledge to assess the potential long-term impact of mining and tailings management in northern environments, specifically the proposed Kiggavik project.

The study provides numerical analysis of permafrost degradation due to mining activities within the continuous permafrost zone in Canadian northern regions. The study targeted the proposed Kiggavik project where four open pits and one underground mine would be excavated into permafrost. The stability of permafrost underlying the TMFs is examined in both the short term (mine operation period, around 20 years), and long term (both mine decommissioning and the climate change effect over more than 2,000 years).

The numerical computational work follows a series of incrementally complex hydro-thermal models. In these models, various physical processes such as phase change and unsaturated flow are considered, so that the proper response of permafrost and tailings materials in a scientifically sound framework can be addressed. All numerical computations were successfully completed using COMSOL commercial software in both two-dimensional and three-dimensional settings.

Read the RSP-493.1-Final Report (PDF)

RSP-444.2 – Development of Analytical Tools for Soil-Structure Analysis

The study of the impact of earthquakes on nuclear facilities is an important part of a safety assessment. The Earthquake Soil-Structure Interaction (ESSI) is a key issue in seismic analysis of nuclear facilities. In some cases the soil-structure interaction can be a source of overly conservative seismic margins, while in other cases the effects may be underestimated. It is important to find a balanced and comprehensive regulatory approach which will take into account the seismic requirements both for new builds and existing nuclear facilities.

The objective of this project was to develop tools to simulate realistic, non-linear, seismic soil-structure interaction in order to assess the margins due to ESSI. The ultimate objective is to develop a regulatory approach to assessing ESSI.

During this project, new methodologies were developed and existing methodologies improved for modeling and simulation of the ESSI problem. In particular, high-fidelity methods and models were developed for:

  • analysis of 3D (6D) earthquake motions (including incoherence modeling and assessment)
  • methods and models for elastic-plastic soils
  • contact problems (between soil and foundation concrete, for both dry and saturated conditions, which lends itself to buoyancy modeling)
  • timing and location of energy dissipation
  • deep foundation effects for pile groups

Verification and validation programs were also developed. Verification was fully developed, while validation was developed for existing data. In addition, the importance of all analyzed phenomena was assessed with respect to safety of nuclear facilities.

This forward-looking project has sparked significant interest. A large-scale project that will use developments from this CNSC project was recently initiated in the US (supported by the US Department of Energy).

Read the RSP-444.2-Final Report (PDF)

RSP-413.9 – Evaluation of safety assessment code for used fuel disposal facilities

Computer models (or codes) are used to calculate the potential exposure to humans from radionuclides in nuclear wastes emplaced in a deep geological repository. Establishing the reliability of these codes is essential; one way to assess their validity is to compare the results from independently developed codes. 

The Nuclear Waste Management Organization’s (NWMO) Fifth Case Study (2013) report documents the post-closure safety assessment of a generic deep geological repository for nuclear fuel wastes in sedimentary rocks. This assessment modelled the groundwater flow and transport of radionuclides to calculate the potential total dose to humans over a period of one million years. In the study presented by the NWMO, one dimensional modelling was conducted using computer code SYVAC3-CC4.

The purpose of this research was to independently model the NWMO’s Fifth Case Study dose calculations using the SOAR (Scoping of Options and Analyzing Risk) model developed by the United States Nuclear Regulatory Commission, and to compare the results.

This report summarizes the independent modelling work in which the SOAR model was applied to the reference case presented in the NWMO’s Fifth Case Study report. Both deterministic and probabilistic approaches were used. Overall, the results between the two models agreed well, providing additional confidence in their use as modeling tools. This report also presents a general review of the licensing status and of the models and codes used internationally for the assessment of deep geological disposal facilities for nuclear wastes.

Read the RSP-413.9-Final Report (PDF)

RSP-413.8 – Laboratory tri-axial and permeability tests on Cobourg limestone and Tournemire shale

Deep geological disposal is currently being considered for Canadian radioactive wastes. Deep geological disposal relies on the surrounding host rock and engineered barriers to contain and isolate waste from the biosphere for hundreds of thousands to millions of years.

Following the excavation of a repository, the surrounding rock along its galleries and shafts will contain numerous fractures and micro-cracks. This zone with created fractures and cracks, called the excavation damage zone, could act as a pathway for the migration of radionuclides and compromise the long-term safety of the repository.

The objective of this research was to enhance our knowledge and ability to assess the long-term performance of a geological repository. The focus of this work was on sedimentary rocks such as limestone and shale, with special attention being paid to the changes in their hydro-mechanical properties in the excavation damage zone. The experimental data collected during this project will be used to develop the mathematical model of the host rock to assess the long-term performance of the repository.

The University of Toronto was contracted to experimentally determine the hydro-mechanical characteristics of Tournemire shale and Cobourg limestone using the Rock Fracture Dynamic Facility. The facility operates a tri-axial geophysical imaging cell capable of quantifying the coupled hydro-mechanical response of rock samples under triaxial loads. Cobourg limestone is one of the candidate rock types for a proposed deep geological repository for radioactive wastes in Canada. The Tournemire shale samples were cored from the Tournemire underground research laboratory in France.

Tri-axial and permeability tests were carried out on the rock samples to obtain their mechanical and hydraulic properties and responses simultaneously. Geophysical methods such as P-wave and S-wave measurements were used to detect damage to the samples and to assess their anisotropy.

Specimens of the Tournemire shale and Cobourg limestone tested at various angles with respect to their foliation planes show two different failure modes – axial splitting and shearing failure. The permeability of Cobourg limestone measured post-failure is about 2 to 3 orders of magnitude higher than that of intact rock. This means that the permeability of the highly damaged limestone due to excavation could be at least three orders of magnitude higher than that of the undamaged rock.

Read the RSP-413.8-Final Report (PDF)

RSP-0310 – Natural and anthropogenic analogues for high-level nuclear waste disposal repositories: A review

This report provides a review of natural and anthropogenic analogues for high-level nuclear waste in a deep geological repository (DGR). We also occasionally highlight analogues that have been used for low and intermediate waste.

Most studies define natural analogues as either naturally occurring or anthropogenic systems. In this report, we distinguish between natural analogues and anthropogenic analogues because anthropogenic analogues generally provide non-technical (anecdotal) illustration of concepts and processes for the safety case, whereas natural analogues can provide technical and quantitative information. In addition, natural analogues can provide information over geological time (millions of years) and spatial (kilometres) scales whereas anthropogenic analogues provide information over a much more limited time scale (hundreds or thousands of years). Regardless of the definition, analogue studies provide one of the multiple lines of evidence intended to increase confidence in the safe geologic disposal of high-level radioactive waste. They are deemed necessary because they complement the experiments that are carried out over a period of months or years. They also provide a way to validate numerical long-term safety assessment models with information and data covering geological time and spatial scales.

The first part of this report describes the analogue concept. The second and third parts provide examples of natural and anthropogenic analogues for engineered barrier systems and natural barriers. Part four describes analogues for complex coupled transport processes general and specific recommendations for future research.

A key recommendation is that a concerted effort should be made to ensure that there is a transfer of data from the complex, natural analogue field studies to simplistic models which, by necessity, are used in performance assessment. Field analogue studies should be planned to interface with laboratory experiments and, ultimately, with in situ field experiments when the final repository site is selected. This will provide a more quantitative use of natural analogue data in support of a DGR concept.

Read the RSP-0310-Final Report (PDF)

Page details

Date modified: