Radioactive Waste Management 2000
This page intentionally left blank
IMechE Conference Transactions
Radioactive Wa...
42 downloads
2157 Views
10MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Radioactive Waste Management 2000
This page intentionally left blank
IMechE Conference Transactions
Radioactive Waste Management 2000 Challenges, Solutions, and Opportunities
Organized by The Nuclear Energy Committee of the Power Industries Division of the Institution of Mechanical Engineers (IMechE) The American Society of Mechanical Engineers The Japan Society of Mechanical Engineers
Co-sponsored by BNES - British Nuclear Energy Society ICE - The Institution of Civil Engineers IEE - Institution of Electrical Engineers IChemE - Institution of Chemical Engineers
IMechE Conference Transactions 2001-1
Published by Professional Engineering Publishing Limited for The Institution of Mechanical Engineers, Bury St Edmunds and London, UK.
First Published 2001 This publication is copyright under the Berne Convention and the International Copyright Convention. All rights reserved. Apart from any fair dealing for the purpose of private study, research, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, no part may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, electrical, chemical, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owners. Unlicensed multiple copying of the contents of this publication is illegal. Inquiries should be addressed to: The Publishing Editor, Professional Engineering Publishing Limited, Northgate Avenue, Bury St Edmunds, Suffolk, IP32 6BW, UK. Fax: +44 (0) 1284 705271.
© 2000 The Institution of Mechanical Engineers, unless otherwise stated.
ISSN 1356-1488 ISBN 1 86058 276 1
A CIP catalogue record for this book is available from the British Library.
Printed by The Cromwell Press, Trowbridge, Wiltshire, UK
The Publishers are not responsible for any statement made in this publication. Data, discussion, and conclusions developed by authors are for information only and are not intended for use without independent substantiating investigation on the part of potential users. Opinions expressed are those of the Author and are not necessarily those of the Institution of Mechanical Engineers or its Publishers.
Conference Organizing Committee D Bonser (Chairman) BNFL M Brewin AEA Technology Nuclear Engineering B Bryce Mitsui Babcock Energy I Critchley BNFL
A Goddard Imperial College of Science, Technology, and Medicine S Harnwell SDP Commissioning T Lawrence NNC C Waker Health and Safety Executive
C Ealing ALSTEC
Held 18-19 October 2000, at IMechE Headquarters, London, UK
This page intentionally left blank
Related Titles of Interest Title
Editor/Author
ISBN
IMechE Engineers' Data Book — Second Edition
C Matthews
1 86058 248 6
Practical Guide to Engineering Failure Investigation
C Matthews
1 86058 086 6
Integrity of High-temperature Welds
IOM/IMechE
1 86058 149 8
Assuring Its Safe: Integrating Structural Integrity, IMechE Conference Inspection Monitoring and Monitoring Safety, and Risk Assessment
1 86058 147 1
Nuclear Decommissioning '98
IMechE Conference
1 86058 151 X
Boiler Shell Weld Repair Sizewell 'A' Nuclear Power Station
IMechE Seminar
1 86058 244 3
Fire and Explosions — Recent Advances in Modelling and Analysis
IMechE Seminar
1 86058 193 5
Remanent Life Prediction
IMechE Seminar
1 86058 154 4
Plant Monitoring and Maintenance Routines
IMechE Seminar
1 86058 087 4
For the full range of titles published by Professional Engineering Publishing contact: Sales Department Professional Engineering Publishing Limited Northgate Avenue Bury St Edmunds Suffolk IP32 6BW UK Tel: +44 (0)1284 724384 Fax: +44 (0)1284 718692 Website: www.pepublishing.com
This page intentionally left blank
Contents Treatment Issues C584/004/2000
C584/006/2000
C584/015/2000
The new Dounreay low-level liquid effluent treatment plant P F G Thomson
3
The disposal of a radioactive cell M Harrison
21
Technical and operational risk management strategies for the Sellafield Drypac Plant (SDP) G McCracken
31
Waste Management Practice C584/026/2000
C584/001/2000
C584/022/2000
C584/025/2000
C584/009/2000
Radiation inheritance of Russian nuclear fleet and ecological safety problems relating to utilization of nuclear submarines and rehabilitation of other facilities in the Navy A P Vasiljev, V A Mazokin, M E Netecha, Yu V Orlov, and V A Shishkin
43
Decontamination and waste minimization techniques in nuclear decommissioning K F Langley and J Williams
47
Transuranic waste management at Los Alamos National Laboratory J J Balkcy and R E Wieneke
57
Disposition of Russian nuclear submarines - outlines of the concept and implementation problems B A Gabaraev, V A Shishkin, and V A Mazokin
67
Management of accumulated operational wastes at BNFL's decommissioning reactor sites A T Ellis, L McTagget, and R I Hey
73
Transport and Storage C584/013/2000
C584/032/2000
Transportation of spent fuel in Japan M Nakajima
85
Engineering considerations associated with plant used for storage of intermediate level waste - a regulator's view W Seddon
95
C584/018/2000
C584/014/2000
C584/016/2000
C584/002/2000
The packaging of waste for safe long-term management S V Barlow and J D Palmer
105
Independent monitoring of solid low-level radioactive waste disposals in the UK S Newstead, N A Leech, and S R Daish
117
Round robin test for the non-destructive assay of 220 litre radioactive waste packages L P M Van Velzen
129
The feasibility of surface storage for high-level waste LCave
141
Environmental and Regulation C584/012/2000
C584/017/2000
C584/020/2000
C584/027/2000
C584/010/2000
C584/019/2000
Authors' Index
Application of in-line monitoring to waste minimization during soil remediation T J Miller
153
Contained water management within the Chernobyl 'shelter object' A A Kornyeyev, C R Wilding, T H Green, and A P Krinitsyn
161
ALARP as applied to high-level waste - the regulatory approach at Sellafield C H Waker
171
Radiation safety problems arising with damaged nuclear submarines utilization V A Mazokin, M E Netecha, YU V Orlov, G A Stanislavski, G A Vasilicv, and V V Borisov
183
Experience in nuclear decommissioning and waste management G R Edler, D Bradbury, and C J Wood
191
Disposal of radioactive waste - a puzzle in four dimensions I J Duncan
201 211
Treatment Issues
This page intentionally left blank
C584/004/2000 The new Dounreay low-level liquid effluent treatment plant P F G THOMSON NNC Limited, Knulsford, UK
SYNOPSIS In February 1997, NNC Ltd were awarded the contract by the UK Atomic Energy Authority (UKAEA) to design, build and carry out the inactive commissioning of a new Low Level Liquid Effluent Treatment Plant (LLLETP) at Dounreay. The purpose of the LLLETP is to collect and treat all the low level liquid effluent from the Dounreay site. The new plant will replace an existing facility which is nearing the end of its operational life. When operational, the LLLETP will enable the UKAEA to meet the more stringent sea discharge requirements specified by the regulatory authorities. This paper considers some of the significant aspects associated with the design and construction of the plant. This includes the key design issues and safety requirements associated with building a facility of this type at Dounreay.
1.
PURPOSE, LOCATION AND KEY FEATURES
The primary purpose of the LLLETP is to collect and treat low level liquid effluent from the Dounreay site. The design ensures that discharges of effluent to sea will be within the pH range of 5 to 9, as required by the Scottish Environmental Protection Agency, (SEPA), compared with the current limits of 2 to 11. In addition, the new plant meets modern standards to reduce radioactive doses to operators to As Low As Reasonably Practicable (ALARP). Prior to awarding the contract to NNC, a number of option studies had been commissioned by UKAEA to determine the best approach to meet the new pH requirement and to minimise the discharge of particulates and solvent to sea.
C584/004/2000
© IMechE 2000
3
The option studies identified a number of key requirements: two stage pH adjustment two large sea discharge tanks with solids settling capacity a sludge removal system a solvent removal capability. To achieve these requirements it was considered necessary to install a new plant which will replace the 40 year old existing facility on site. The new plant is located adjacent to the existing effluent treatment plant. A pictorial view is shown on Figure 1 and the location is shown on Figure 2. The main features of the plant are illustrated on Figure 3, which is a schematic diagram of the plant. Figure 4 shows the building and equipment arrangement. The main features of the LLLETP include: • A below-ground gravity fed receipt tank which collects the low level effluent from the site drains. The tank includes a solvent separation feature to remove solvent from the effluent stream and collect it for disposal. • A buffer tank, which provides sufficient volume to contain 15 hours of average effluent inflow from the gravity receipt tank prior to neutralisation. • Two staged neutralisation for continuous pH adjustment of the effluent stream by the addition of acid or alkali solutions. • Two sea discharge tanks, each able to hold in excess of the expected daily effluent flow to the plant. The tanks are designed to enable solids to precipitate out and for the accumulated sludge to be removed to a collection system. The design life for the plant, support structures and building housing is fifty years. Material selection, design detail and plant layout have addressed the requirement to provide a cost effective lifetime maintenance regime for the facility.
2.
DESIGN SAFETY PRINCIPLES
The design of the LLLETP satisfies the Design Safety Principles applicable to nuclear facilities. These are embodied in the HM Nuclear Installations Inspectorate (NII) Safety Assessment Principles (SAPs). The Safety Assessment Principles are addressed in the design as follows: • A mainly automated plant minimising the operator exposure time and capable of being operated remotely.
4
C584/004/2000
© IMechE 2000
Location of the control room outside the radiologically controlled area. Filtered vessel extract ventilation system to prevent releases to the building atmosphere. Filtered building extract ventilation system to continuously remove any potential airborne activity. A flushing water system to wash-down the inside of the tanks and pipework to reduce activity levels. An installed shield wall between the main tank area and the bulk of equipment requiring maintenance ensures reduced radiological exposure. Provision of uncomplicated equipment and controls. A combination of Hazard Assessment and HAZOP studies throughout the design process. A number of further specific design safety principles as identified in the Preliminary Safety Report (PSR), to assist in the detailed design of the new facility.
3.
WASTE ASSESSMENT
3.1 Incoming Effluent The LLLETP receives effluent from a number of diverse systems. Generally, all effluent is sampled and sentenced by the donor plants before authorisation is sought to discharge to the LLLETP. Each of the donor plants also generally includes buffer storage to enable scheduling of routine discharges. Solids will settle out of the effluent streams in the new plant. The mass of retained solids is expected to be considerably greater than at present as a result of precipitation resulting from pH adjustment. There is also a potential for solvent in the effluent. Solvent arisings are only anticipated to occur during fault conditions. If solvents were to arise, then they can be effectively removed in the gravity receipt tank. 3.2 Outgoing Wastes On average, the LLLETP discharges 450 m3 of liquid effluent daily through the sea discharge pipelines. In addition there are the following waste arisings: • Solids in the form of a sludge for transfer to an on-site sludge processing plant. • Small quantities of solvent to be transferred to an-site disposal facility. • Potentially contaminated ventilation filters • Miscellaneous clothing, wipes etc.
C584/004/2000
© IMechE 2000
5
4.
HAZARD ASSESSMENT
4.1 Introduction The purpose of the hazard assessment was to assess the hazards associated with both the construction and the subsequent operation of the new plant. As the construction involves new plant and the work was carried out in an essentially radiologically clean area, few hazards would have been expected other than conventional industrial hazards. 4.2 During Construction Prior to any construction activities taking place, procedures were developed to ensure that appropriate ground surveys and soil strategies were utilised should the ground on which the LLLETP was to be constructed be found to have contained any small areas of sub-surface contamination. 4.3 Normal Operation The hazards during normal operation could include: Radiation dose to the operator during normal operational tasks. Radiation dose to the operator and maintenance staff during programmed examination, maintenance, inspection and testing (EMIT). Radiation dose to other workers on the site not within the LLLETP building. Radiation dose to members of the public outside the plant. Industrial hazards to the operator or maintenance staff. Environmental effects. The design contains a number of features which ensures that in normal operations the doses are ALARP, these include: Automatic operation so the LLLETP is not normally manned. Routine manual activities take place in a controlled environment. Vessel and building ventilation with specific ventilation in sampling boxes. Health physics surveys to ensure maintenance activities are conducted under controlled and monitored conditions. The location of maintained equipment has been chosen to provide shielding from the main tanks even though the activity levels are low. As the LLLETP is not normally manned, duplicate instruments and controls are located both in the LLLETP control room and in an existing permanently manned control room approximately 700 m from the LLLETP. If a fault were to occur, the operator will determine the cause and the recovery procedure from the remote control room. The operator is thus protected from the initial consequences of a fault. At least one hour is available before operator intervention is essential in the case of the failure of the automatic control system or failure of power supplies. For all other faults several hours are available.
6
C584/004/2000
© IMechE 2000
5.
CONSTRUCTION AND OPERATION OF THE PLANT
5.1 Construction The LLLETP plant is housed in a building 35m x l6 m x l3m high with an annexe 25 m x 6 m x 4 m high. The building is a steel framed construction with part height brick walls and the remainder clad in profiled coated aluminium sheeting. The annexe contains the control room, change areas, switchgear and ventilation fan and filter rooms. The gravity receipt tank is located below the east end of the annexe within a concrete bunded pit. The building is designed to comply with the Scottish Building Regulations and relevant current British Standards. Foundation design is based on ground information obtained from a soil investigation. The building is supported on concrete piles and the foundations are arranged such that excavation and soil removal is minimised. Before construction commenced, the area was swept to check for any underground services and contamination in accordance with site procedures. Excavation was then carried out in pre-defined depths within strips with a radiation walk over taking place at each stage. Where contamination greater than background was found the material was segregated. Material at background or less was set aside and where suitable, was used as fill. Excavation for the receipt tank pit was been carried out within a sheet piled cofferdam following the same procedures. One of the most demanding elements of the construction programme was the fabrication of the Sea Discharge and Buffer Tanks. From the early conceptual stages of the plant it was recognised that it would be preferable to construct the tanks in the completed building. This would avoid chloride contamination of the stainless steel components and minimise disruption of fabrication activities due to the weather conditions at the exposed coastal location had the tanks been constructed outdoors. For construction of the tanks, a prepared access way was formed on the south side of the tank locations by placing fill material on the new slab up to the level of the nominally 300 mm high bunds. Access into the building was through a temporary door in the east gable. The plates from which the tanks were constructed were brought into the building piece-small and the tanks were fabricated in-situ. The stainless steel plates were welded to form strakes. When each strake was complete a series of jacks was used to raise the tank structure to allow another strake to be built beneath the first. This process was repeated until the tank structure was complete. A schematic of this procedure is shown in Figure 5. 5.2 Operations All the liquid effluent will reach the plant via the existing site drain system with the existing pipework extended to the new plant location. All these drain lines have been provided with secondary containment to prevent the possibility of uncontrolled releases to the environment.
C584/004/2000
© IMechE 2000
7
The incoming effluent volumes have been estimated at 150,000 m3 per year at an average flow rate of 20 m3 per hour with a maximum of 50 m3 per hour. The effluent is routed directly to an underground receipt tank with a volume of 35 m3. The tank contains a constant level solvent separator and a partitioned area which acts as a solvent collection tank. The purpose of the solvent separator is to enable the removal and recovery of light solvents from the site effluent. The collected solvent can then be pumped to a terminal point for disposal. The receipt tank is provided with a sludge removal pump and a flushing water system to mobilise the settled solids before they are pumped out. A level control system on the receipt tank starts the variable speed transfer pump as required to maintain a generally constant low level in the tank. This pump transfers the effluent to the 300 m3 buffer tank which has sufficient capacity to accommodate up to 15 hours of average effluent inflow. Because the role of the transfer pump is important in maintaining the availability of the LLLETP, a standby pump is provided. A diesel generator back-up power supply is also provided to each pump in case of loss of supplies. The buffer tank provides capacity to allow for downstream hold-ups and receipt fluctuations. The design aim in normal operation is to operate this tank with a constant low effluent level to maximise the buffer capacity. The tank is provided with sludge removal pumps and flushing water connections to mobilise settled solids if required before they are pumped out. The buffer tank pumps transfer the effluent to the first of two neutralisation tanks located at a high level in the building. The two neutralisation tanks are operated in series as a continuous process, the effluent gravitates from the first tank to the second tank. The pH of the effluent is measured and the control system adjusts the rate of effluent flow and alkali or acid addition to achieve the required pH. An agitator is provided in each tank to ensure thorough mixing of the contents. The alkali or acid is supplied to each tank using the chemical dosing pumps from bulk storage tanks; these are located outside the main building within their own bunded area. The effluent flows from the second neutralisation tank by gravity to the two sea discharge tanks. The effluent is collected in one of the sea discharge tanks. During the transfer to the sea discharge tanks, the effluent is sampled on a continuous basis to confirm that the effluent meets the discharge authorisation requirements. The tank contents are then recirculated using one of the two sea discharge pumps and a recirculation line to ensure homogeneous mixing of the tank contents. After a period, the recirculation of the tank is stopped and sludge allowed to settle to minimise the amount of solids discharged to sea.
8
C584/004/2000
© IMechE 2000
Following settling, and subject to sample analysis results, tidal conditions and the plant manager's authorisation, the key controlled bypass valve is closed, the sea discharge valve opened and the sea discharge pump stalled. During discharge, the effluent is sampled again and the discharge volume is recorded. Each tank is provided with connections to the sludge discharge pumps and water flushing connections to assist in sludge removal as required. In addition to handling effluent from the site drains, the LLLETP is also designed to handle effluent arising from the PFR Sodium Disposal Plant (SDP) located on the Dounreay site. Because of its high chloride content, the SDP effluent forms a separate stream within the LLLETP building. The SDP releases effluent in batches with a total daily release transferred to the 30 m3 SDP hold-up tank located in the LLLETP building. The total daily volume can be accommodated in the hold-up tank. The effluent is sampled at the SDP before transfer to the LLLETP and confirmed to meet the sea discharge requirements, the effluent can then be discharged directly to sea, subject to tidal conditions and the plant manager's authorisation. As with discharge from the sea discharge tanks, samples of effluent are taken prior to discharge. The LLLETP is generally automatically controlled by a Distributed Control System (DCS). Sufficient, independent, hard wired instruments and controls are also provided to ensure the plant can continue to be operated manually if the DCS were to fail. Each vessel is connected to a vessel ventilation system with a HEPA filtered extract to avoid active aerosol discharge to the building. A HEPA filtered building extract and a dedicated sample box extract system are also provided.
6.
DECOMMISSIONING DESIGN
The LLLETP is designed to have an operating life of fifty years. Decommissioning of the plant at the end of life has been considered during the design of the LLLETP. Equipment has been designed to avoid potential contamination traps and be readily decontaminable. This is to be achieved by the use of stainless steel for the majority of the equipment and the use of decontaminable finishes on surfaces. A flushing water system is provided to enable regular removal of sludge and contamination from within tanks and pumps and thus minimise activity build-up. Access is provided to the interior of all tanks to enable regular maintenance and inspection and thorough decontamination prior to dismantling. Tanks and equipment have been designed and located to ensure easy access whenever practical.
C584/004/2000
© IMechE 2000
9
7.
CONCLUSIONS
In the summer of 1999, NNC completed the final inactive commissioning and functional trials of the LLLETP. This has subsequently been followed by the construction of a new dedicated sludge handling facility to manage the solid waste arisings from the plant. Active commissioning trials will now be carried out together with the phased transition from the old facility to the new plant. On completion, the UKAEA will have a state-of-the-art facility with the capability of treating all active liquid effluent from the Dounreay site for the next fifty years.
10
C584/004/2000
© IMechE 2000
Dounreay low level liquid effluent treatment plant Figure 1 Low level liquid effluent treatment plant - pictorial representation
Figure 2
Location of Low Level Liquid Effluent Treatment Plant
Figure 3
Schematic Diagram - Low Level Liquid Effluent Treatment Plant
Figure 4
Low Level Liquid Effluent Treatment Plant - Building and Equipment Layout
Dounreay low level liquid effluent treatment plant Figure 5a Sea discharge and buffer tanks schematic construction sequence - sheet 1
Dounreay low level liquid effluent treatment plant Figure 5b Sea discharge and buffer tanks schematic construction sequence - sheet 2
Dounreay low level liquid effluent treatment plant Figure 5c Sea discharge and buffer tanks schematic construction sequence - sheet 3
Dounreay low level liquid effluent treatment plant Figure 5d Sea discharge and buffer tanks schematic construction sequence - sheet 4
Dounreay low level liquid effluent treatment plant Figure 5e Sea discharge and buffer tanks schematic construction sequence - sheet 5
This page intentionally left blank
C584/006/2000 The disposal of a radioactive cell M HARRISON NUKEM Nuclear Limited, Dorchester, UK
SYNOPSIS Cave 9 in the Active Handling Building on the UKAEA site at Winfrith was an active cell used for experimental work on irradiated fuel during the 1970s and 1980s. During 1998 and 1999 the cave was stripped out, decontaminated and demolished. The challenge was to complete the decontamination of the structure to achieve free release disposal of the concrete and to minimise the volume of low level waste.
1.
CAVE 9 CONSTRUCTION
Cave 9 was the smallest and newest of the three cave lines in the Active Handling Building on the UKAEA site at Winfrith. It was designed and constructed over the period 1972-1974 as an active cell for inspection work on non-fissile items for UKAEA. Later a ventilation system was added to enable a broader range of studies to be carried out for the CEGB on more active fissile and non-fissile items. Figure 1 shows the cave in its operational state. The main cave structure, which was a free standing structure, was based on the availability of five large 40ton shielding blocks that had been displaced from use in the north cave line by an earlier modification. These five shielding blocks were approximately 2.7m wide, 1.5m deep and 3.7m high and were located within the structure, three forming part of the walls and two forming part of the roof. In addition thirteen blocks, each weighing approximately 2.5ton, were used in the construction of the cave supplemented by 36m3 of in-situ reinforced concrete. Figure 2 is a sectional elevation showing the arrangement of four of the 40ton blocks and nine of the 2.5ton blocks, the remaining blocks making up part of the front wall of the cave. The overall dimensions of the cave were 8.3 metres long, 4.8 metres deep and 5.6 metres high with walls 1.5 metres thick. The internal size provided by the construction was approximately
C584/006/2000
© IMechE 2000
21
5.2m in length, 1.8m deep and 4.0m high, and the internal walls were part lined with stainless steel. The operating face was fitted with two zinc bromide windows, one full size and one half size, and four master slave manipulators together with service plugs. Operation of the cave ceased in the early 1990s when plans were developed for its decommissioning and demolition.
2.
CAVE DECOMMISSIONING
Decommissioning of the cave began in 1998, at this time the contamination levels on the walls and on the bench surface were typically in the range 2-6mSv/hr, whilst at floor level contamination up to l00mSv/hr was recorded. The decommissioning of the cave was undertaken in two stages, the first by remote techniques to reduce the radiation levels within the cave, and the second by a series of man-entries. Use was made of the cave equipment, such as the manipulators and waste posting port, to remove contamination from the cave. The internal features of the cave were decontaminated using vacuum cleaners and swabs. Contaminated equipment left over from the last operations in the cave was size reduced in the cave using remote tooling before being posted out for disposal as intermediate or low level waste. This first stage of decommissioning was continued until man-entries to the cave could be justified on ALARP grounds. At this time the radiation levels in the cave were in the range 150 to 300 microsieverts per hour, though on the floor of the cave the levels were several times these figures. The strategy adopted for the second stage of decommissioning involving man-entries was to remove, from one end of the cave, the 40ton shielding block that formed the cave wall. A temporary enclosure was constructed at this end of the cave to allow this process to take place and to act as the controlled entry point into the cave. The enclosure provided a location with relatively low radiation levels from which further decontamination of the cave could be performed using extended tooling. In particular the higher levels of radiation on the floor including some very high point sources could be dealt with whilst incurring acceptably low dose uptake. The majority of the entries were made wearing air fed suits or half-suits as a precaution against the disturbance of contamination during the operations. In all 68 controlled entries were made, involving some 30 staff, with a total whole body dose of 7.24mSv being recorded, the highest individual dose being 0.95mSv. This stage of the process was completed over a four month period. During this phase the cave was stripped of all its equipment including its steel cladding, its windows and manipulators. Loose contamination was removed using HEPA filtered vacuum cleaners and damp swabbing with decontamination reagents. Sprayed on peelable coatings were also used to both temporarily fix the contamination and to remove it with the coating. It was found that care had to be taken with the application of the peelable coatings as they are difficult to remove from pitted surfaces such as bare concrete. A number of high radiation fixed contamination spots were found, these were dealt with by more vigorous mechanical techniques. The decontamination achieved a reduction in internal surface contact dose rates down to an average of less than 20uSv/hr with isolated spots up to l00uSv/hr. The surfaces of the cave were then coated with a water based masonry paint to temporarily fix any remaining loose contamination, achieving the required free breathing conditions within the
22
C584/006/2000
© IMechE 2000
cave. Once the levels in the cave had been satisfactorily reduced the two 40ton shielding blocks making up the majority of the roof were lifted down to ground level. A final decontamination exercise resulted in a cave carcase with low levels of fixed contamination ready for demolition, plus the three 40ton shielding blocks with similarly low fixed contamination levels. The project strategy was that the bulk of the cave structure should be decontaminated to free release levels rather than being despatched as low level waste. However, the project was also required to remove the structure from its location at the earliest opportunity. This drove the decision to demolish the structure in a controlled manner whilst it still had the low levels of fixed contamination, rather than continue with final decontamination in-situ. The demolition was achieved by a combination of dismantling the pre-cast sections (the two remaining 40ton shielding blocks and the thirteen 2.5ton blocks) and cutting the remaining walls using a variety of techniques. Much of the cutting was undertaken using diamond tipped barrel drills and a diamond wire saw, this technique proving to be the most efficient of those tried. Lesser use was made of a diamond tipped road saw and hydraulic bursting. The result was a collection of concrete blocks of between 40 and 2 tonnes in weight, the total weight being approximately 350 tonnes.
3.
DECONTAMINATION FOR FREE RELEASE
The challenge was to complete the decontamination of the concrete blocks to achieve free release of the concrete and to minimise the volume of low level waste. A number of abrasive techniques and equipment were used to remove contamination from all surfaces of each block. Containment tents, fitted with HEPA filtered extract, were erected within the Active Handling Building in which to undertake the decontamination. A shot blasting device supplied by USF Blastrac was effective on the larger plane surfaces of the blocks. This device uses 1mm diameter hardened steel shot that is contained by the blasting head and recovered for recycling. The device incorporates a vacuum system and a HEPA filtered exhaust to recover the debris. Two arrangements of the equipment's blasting head were utilised, the first uses a support beam and winch for blasting vertical surfaces in vertical passes by operation of the winch. The second arrangement is for blasting horizontal surfaces uses an electrically propelled carriage. More localised decontamination was undertaken using hand held scabblers and needle guns.
4.
FREE RELEASE PROTOCOL
The driving force for free release disposal of the cave structure arises from the large volume and weight of the material against the modest and essentially surface nature of the residual contamination. The alternative was to classify it all as low level waste and to transport it to the UK Low Level waste disposal site at Drigg in Cumbria.
C584/006/2000
© IMechE 2000
23
The protocol for defining the free release disposal requirements was established with the involvement of the contractor (NUKEM), the client (UKAEA) and the Environment Agency. The protocol for the disposal of the cave addressed a number of aspects; the measured surface contamination, the presence of surface coatings (paint) as a contamination trap, the activity within the bulk material and, penetrations within the blocks. The measurement of surface contamination was undertaken using standard health physics monitoring equipment by both smear and direct probe. The protocol required 100% surface monitoring of all surfaces by both methods with the limits set at 4Bq/cm2 beta-gamma and 0.4Bq/cm2 alpha, i.e. less than the levels required for removal from a designated area. Large area smears were taken for the detection of loose contamination. If any positive readings were obtained the area in question was subject to a more focused smear survey to find the source of the contamination. For the probe survey a grid was marked on the surface of the block in order to ensure that the whole surface was monitored. Although the contamination was expected to be mostly 137Cs and 60Co, the surveys also included monitoring for alpha activity. The presence of paint on the surfaces was considered as having the potential to mask contamination, the history of the cave and the blocks was uncertain therefore repainting to seal contamination may have occurred in their past. The protocol therefore required the removal of all surface coatings back to the original material surface. It was necessary to demonstrate that the bulk material, i.e. the concrete, was within the regulatory limits for free release disposal. The concern was that the bulk concrete had become radioactive by some means, be that migration into the concrete matrix or activation of the concrete by the high radiation levels that has existed in the cave during its operation. Whilst both of these processes were thought unlikely to have occurred, the protocol required that this be demonstrated. To provide this demonstration the chosen method was to remove core samples from the blocks for radiochemical analysis. Cores were removed by dry diamond drilling using a 50mm diameter coring bit, to a depth of 100mm. The core was then split along its length into four approximately equal 25mm long sections. Each of these sections was identified uniquely and analysed individually. The protocol aimed to limit the number of core samples taken for analysis by recognising the similarity in design and operational history of the blocks making up the structure. For this, the concrete blocks and cave structure were divided into four groups; the five 40ton blocks, the thirteen 2.5ton blocks, the in-situ east wall of the cave and, the in-situ west wall of the cave. The number of cores to be taken was then defined for each group. For the 40ton and 2.5ton blocks the sampling regime required that the first block have cores taken from approximately the centre of each vertical face, plus one from either the base or top face. If these cores proved to be within the free release criteria then the second block only required three of the vertical faces to be cored, though one of these had to be taken from what had been the inside face. If these cores proved to be within the free release criteria then only one core was taken from the inside face of the remaining blocks. The same regime was adopted for the 2.5ton blocks. The sections of the east and west walls of the cave were each sampled once, again at approximately the centre of the face that had been nearest to the inside face of the cave.
24
C584/006/2000
©IMechE2000
The surface of the blocks at the location where the cores were to be taken was decontaminated and monitored prior to removing the cores. This ensured that contamination was not driven into the block by the drilling operation, and that the cores could be removed from the designated area and sent off the site for analysis.
5.
TREATMENT OF PENETRATIONS
The concrete blocks contained numerous penetrations ranging from small bolt holes just a few centimetres deep to en-cast liners through the full depth of the blocks and up to 300mm diameter. The monitoring of the smaller penetrations by smear or probe was clearly impractical and an alternative method was required. The method selected for the small penetrations was to remove any surfaces that had the potential to be contaminated, by over-drilling. The 'rule of thumb' adopted for this method was to drill out a core of twice the diameter of the penetration and one and a half times its depth. The oversize penetration was taped over to prevent recontamination. For the larger penetrations the original proposal was to decontaminate these using a grit blasting lance and then monitor them using extended reach tooling. However a detailed examination of the penetrations and the block revealed that many of them were constructed of concentric tubes where the inner tube did not extend the full depth of the block. To compound this problem it was also found that the tubes were not sealed together leaving an inaccessible gap between the two. The potential for this gap to be a contamination trap was considered to be too significant to ignore, surface decontamination and monitoring was therefore ruled out for these larger penetrations. In order to free release these blocks with the larger penetrations an alternative method was required. The method chosen was to seal up the penetrations at both ends by welding on metal caps, effectively sealing in any contamination. Once the remaining free release exercise was complete, the block was broken up using a hydraulic breaker and the sealed steel penetrations recovered for disposal as low level waste. In addition to the designed penetrations in the blocks there were also a variety of penetrations or gaps between the concrete mass and any steel features, tubes, plates etc. These small gaps were both potential contamination traps and were inaccessible for monitoring. In these cases the concrete was cut back until it was observed that the gap between the concrete and the steel had closed up. A further 10mm of concrete was then removed to ensure that any potentially contaminated material had been removed. Alternatively the steel plate was cut back to reveal a minimum of 100mm of the concrete face to allow this to be thoroughly monitored.
6.
CORE ANALYSIS
The core sample analysis was undertaken by Southampton Oceanography Centre. The objective was to determine the alpha and beta-gamma activity for each of the concrete samples and compare the results against the free release criteria of 0.4Bq/g. This, together with the surface monitoring, would determine whether the blocks could be free released for disposal or whether they would have to be disposed as low level waste.
C584/006/2000
© IMechE 2000
25
The concrete samples were ground, thoroughly homogenised and a sample was then removed for gamma spectroscopy. A further sample was removed and digested in aqua regia, the resulting leachate was measured for total alpha and beta activity. All anthropogenic radioisotopes identified were reported. In addition, limits of detection were calculated for the isotopes specified by NUKEM, namely 54Mn, 60Co, 137Cs and 241 Am. A typical result from the analysis of one sample was as follows: Sample reference:
A59/C9/23B1
54
Mn I 60Co I 137Cs I 241Am I Total alpha I Total beta 5 cps above background (C3) were noted and identified as 'hot spots'. These 'hot spots' were then surveyed in INT mode (100 s count), by setting up the IS 610 on its tripod so that the detector was 30 cm above the centre of the m2 square containing the 'hot spot', and collecting and recording the cps above background.
C584/012/2000
157
2.3 "Hot-Spot" treatment All "hot-spots" were removed (shovel and bucket) for assay at the monitoring station. The IS 610 was used in standard INT mode and the soil was spread out evenly in a m2 tray to a depth of 2.5 cm. All trayloads below 2.5 Bq/g (above background) were returned to their point of origin. Those above 2.5 Bq/g were packaged for controlled disposal.
3 RESULTS 3.1 IS 610 performance Given the standard counting time of 100 s, the standard IS 610 counting geometry and a sample mass of 30 kg, spread to a depth of 2.5 cm in m2 trays, it was possible to achieve sub Bq/g detection levels for DU in soil. Efficiency factors were improved by using larger masses of sample (Table I). Table I IS 610 efficiency factors for bulk soil assay Soil Mass (kg)
10 3 10 30
Soil Dimensions (cm) 25 Diameter x17 height 48 x 57 x 0.9 height 48 x 57 x 3 height 100 x 100 x 2.5 height
Efficiency (cps/Bq/g)
1.7 0.6 1.7 2.7
3.2 Field survey and assay A few "hot spot" areas were located in places where samples had also given high results. Removal of the topsoil from the "hot spot" areas and assay at the side of the field showed that only one 30 Kg tray in 17 was contaminated above the end-point criterion of 2.5 Bq/g (Table II). This was disposed of as controlled waste and the other 16 trays were returned to their point of origin. Subsequently, the field was resurveyed with the IS 610 and found to be uncontaminated. This was confirmed by conventional sampling and laboratory analysis.
158
C584/012/2000
Table II IS 610 field assay of 30 Kg trays of soil Tray
cps
Bq/g
1 2 3 4 5 6
4.3 4.5 3.6 3.1 4.9 3.3 2.0 1.9 3.4 2.4 3.4 0.7 2.1 1.4 1.3 8.0 3.8
1.6+/-0.1 1.7+/-0.1 1.3+/-0.1 1.1+/-0.1 1.8+/-0.1 1.2+/-0.1 0.7+/-0.1 0.7+/-0.1 1.3+/-0.1 0.9+/-0.1 1.3+/-0.1 0.3+/-0.1 0.8+/-0.1 0.5+/-0.1
7
8 9 10 11 12 13 14 15 16 17
0.5+ /-0.1 3.0+/-0.1 1.4+/-0.1
4. COMPARISON OF TRADITIONAL AND IN-LINE TECHNIQUES The costs, timescales and effectiveness of the in-line monitoring approach are all far superior to the traditional approach. Table III gives a simple comparison of costs. Table III Comparative costs for remediation of a small area of DU contaminated ground Operation
Traditional procedure (£)
DG procedure (£)
Site characterisation Excavation of contamination Confirmatory monitoring Waste disposal Totals
3,000
100 300 50 30 480
300 3,000 6,000 12,300
4.1 Site characterisation A conventional survey for a small area would cost around £3,000 and it would be several months before the results were known (5). Also, grid sampling in the presence of "hot spots" is a hit and miss method of determining the overall contamination distribution and its boundaries. This can lead to the excavation of much clean soil, along with the contamination, or missing contaminated areas altogether and removing no soil. By contrast, the IS 610 is able to rapidly home in on "hot spots" in RUN mode. The contaminated areas may then be monitored more accurately in INT mode. The whole operation takes only a few hours and would cost only £100.
C584/012/2000
159
4.2 Excavation of contamination On the basis of the conventional survey alone it would have been recommended that an area of 60 m2 be excavated to a depth of 0.1 m, generating some 6 tonnes of waste. This operation would cost around £300 and may need to be repeated several times until subsequent surveys indicate that radiological end-points have been met. By contrast, the in-line monitoring technique generates only a small quantity of waste, since only contaminated surface soil is excavated. This is then assayed at the side of the field and only removed if it is above the end-point criterion. Clean soil is returned. The net cost for this operation would be around £300 for the small "hotspot" areas requiring treatment. 4.3 Confirmatory monitoring A second conventional survey would be required to confirm that radiological end-points had been met and would cost a further £3,000 and have the drawbacks noted for the initial characterisation. If the site were still contaminated, further cycles of excavation and monitoring would lead to rapidly escalating costs. However, a second IS 610 survey would be more rapid than the first, since only the excavated areas, where the "hot spots" were, would need to be examined. This would only cost around £50. 4.4 Waste disposal The traditional technique would generate around 6 tonnes of waste which would cost £6,000 if sent to Drigg as low level waste. By contrast the in-line monitoring technique ensures that only the contamination is removed with the absolute minimum of associated soil. This was only 30 Kg, with minimal disposal costs. This paper was published in the WM'00 Conference Proceedings, February 27-March 2 (2000), Tucson, Arizona, USA, on CD-ROM, 31-1.
5 REFERENCES 1. Radioactive Substances Act 1960 (RSA 1960) in conjunction with Exemption Order SI1002 (1986) (exempts natural uranium isotopes from controlled disposal below 11.1 Bq/g). 2. N.Harris, IS 610 x-ray monitor user manual, AWE, Aldermaston, Reading, Berkshire, RG7 4PR, UK, March 1994. 3. L.W.Hensman, use of the IS 610 for ground contamination measurement, Safety Division Technical Note 21/92, AWE, Aldermaston, Reading, Berkshire, RG7 4PR, UK, December 1992. 4. B.B.Warren, Environmental Monitoring Group Report no.35, AWE, Aldermaston, Reading, Berkshire, RG7 4PR, UK, 1981. 5. D.Urquhart, Environmental Monitoring Group Manager, costs of analysing environmental samples, AWE, Aldermaston, Reading, Berkshire, RG7 4PR, UK, 1994. © British Crown Copyright 2000/MoD
160
C584/012/2000
C584/017/2000 Contained water management within the Chernobyl 'shelter object' A A KORNYEYEV Energoatom, Ukraine C R WILDING and T H GREEN AEA Technology, Didcot, UK A P KRINITSYN National Academy of Sciences of Ukraine, Chernobyl, Ukraine
1
INTRODUCTION
On April 26th, 1986, the biggest accident in the history of the nuclear industry occurred inside Unit 4 of the Chernobyl Nuclear Power Plant (ChNPP) in the Ukraine. Within six months, a containment was built over the remains of the reactor. It was not feasible to construct a leaktight containment in such a short timescale and under such difficult radiation conditions. Consequently, the "Shelter Object" contains a number of construction defects. A "Shelter Implementation Plan" (SIP) for stabilisation of the Shelter, and conversion into an environmentally safe site was subsequently formulated and funded by the G7 countries. The SIP Plan is administered by the European Bank of Reconstruction and Development. One aspect of this Plan is management of the water contained inside the Shelter. The scope of the paper includes water contained inside the Shelter and does not address the water inside the Turbine Hall. The paper discusses the sources of water inside the Shelter, and water losses. It summarises the results from characterisation of the water and presents the current status of studies conducted for its future management.
2
SOURCES OF WATER INSIDE THE SHELTER
Precipitation Rainfall and snow-melt leak through the roof of the Shelter. About 2,200 m3 of water per year could enter the Shelter by this mechanism. Condensation Condensation occurs of moisture vapour in the air flowing through the Shelter. During the period May to August, when the temperature of the air entering the Shelter is greater than that of the walls, structures and pools within the Shelter, moisture vapour condenses from the
C584/017/2000
© IMechE 2000
161
incoming air. About 1,650 m3 of water per year could condense on the walls, structures and pools by such mechanisms. Since 1996 a scheme to minimise condensation of water has been implemented in the Shelter. This has involved heating of the air and suppression of natural ventilation mechanisms. Dust suppression liquids Dust suppression systems periodically release water (containing a variety of chemicals) from sprinklers at various locations within the Shelter. The dust suppression system in the Central Hall has been operating since January 1990. From January 1990 to 1999 over 1000 tonnes of dust suppression compounds have been introduced into the Shelter. On average, about 270 m3 of water per year is released from the dust suppression system.
3
WATER LOSSES FROM THE SHELTER
Evaporation Water evaporates inside the Shelter. During the period January to April and September to December, when the air entering the Shelter is cooler than the walls and structures and pools within the Shelter, moisture vapour evaporates into the air. About 2,100 m3 of water per year could evaporate by such mechanisms and is carried out of or redistributed within the Shelter as moisture vapour in the air. Surrounding structures It has been experimentally determined (by visual observations and tracer studies) that water from the Unit 4 continuously moves through the dividing wall between the Unit 4 Nuclear Island Auxiliary Systems Room 001/3 into Unit 3. It then finds its way into the Unit 3 drainage water collection system. Surrounding ground Since 1990, investigations of contaminated ground in areas close to the Shelter have been made. This has involved drilling a network of bore-holes round the Shelter and sampling and analysing the water. Trace levels of Cs-137, Sr-90, uranium and plutonium have been found in the water.
4
WATER BALANCE
About 4,120 m3 of water could enter the Shelter each year by the mechanisms discussed above. The levels of the pools of water have been observed to remain fairly constant over a long period (i.e. recent years). The permanent water volumes within the Shelter are about 400 to 700 cubic metres, depending on the time of year. About 70% of the water accumulations are from the Nuclear Island Auxiliary Systems Room 001/3. If it is assumed that the evaporated water is carried out of the Shelter, the remainder of the water, 2,020 m3 (i.e. 4,120 m3 2,100m 3 ) could flow out of the Shelter each year, by leakage to groundwater and/or to neighbouring structures. The water flow is illustrated in Figure 1 and the balance in Figure 2.
162
C584/017/2000
© IMechE 2000
Figure 1: Flow of water through the shelter
C584/017/2000
© IMechE 2000
16s
Figure 2: Summary of shelter water balance Recent observations and tracer study results indicate that the water ingress and egress values discussed above could be overestimates by a factor of about two.
5
WATER CHARACTERISATION
Sampling Investigations of water within the Shelter began in 1991. To establish the composition of the water, 31 sampling points were identified that would provide the most information on principal water flowpaths. At these points samples of water are taken at regulator monthly intervals. The concentrations of radioactive isotopes of caesium, strontium, uranium and transuranic elements, heavy metals and the macrochemical composition were/are determined. For selected samples the activity on the suspended solid phase (from 0.03 to 1 |J,m) of collected samples has been determined. Water analysis The radiochemical concentration of the Shelter water is summarised in Table 1.
164
C584/017/2000
© IMechE 2000
Table 1: Radiochemical concentration of the shelter water
Radiochemical Cs-137 Sr-90 Plutonium Isotopes Uranium Isotopes
Minimum Value from all water samples taken 0.1MBq/l 0.2 kBq/1 2Bq/l 0.007 mg/1
Maximum Value from all water samples taken 100 MBq/1 10 MBq/1 l,300Bq/l 57 mg/1
Averages Values in Room 001/3
6.8 MBq/1 0.47 MBq/1 80 Bq/1 3.2 mg/1
Data on average concentrations of radionuclides dissolved in the Shelter water as a function of time are shown in Figure 3.
Figure 3: Variations of average nuclide concentrations of shelter water with time Sludge analyses When colloidal particles enter the accumulations of stagnant waters at lower levels of the Shelter, they precipitate and form sediments. The volumes of these sediments (sludges) has been estimated at approximately 100 m3 in room 001/3 (Level -1.0 metres). The radiochemical composition of the sludges are shown in Table 2 below.
C584/017/2000
© IMechE 2000
165
Table 2: Radiochemical composition of the sludges Property/Content Sr-90 concentration Cs-134 concentration Cs-137 concentration U concentration Pu concentration Mass of Uranium Radiologically significant radionuclides
Averages Values in Room 001/3 5.8E7 Bq/kg of dry precipitate 6.3E7 Bq/kg of dry precipitate 3.7E9 Bq/kg of dry precipitate 550 mg/kg of dry precipitate 6.8E5 Bq/kg of dry precipitate >50kg >2.7 TBq
Colloid analyses There are only limited data available on the radioactivity associated with the colloidal (0.1 to 1 micron) and the ultracolloidal (0.1 to 0.01 micron) phases. The samples were ultrafiltered, using nitrogen gas to pressurise the liquid through the filter medium. The results indicate that Cs-137, Sr-90, Pu and U are present in the solid phase particles with colloidal and ultracolloidal dimensions. It has been speculated that the alkaline pH of the water causes plutonium species with oxidation states (III) and (IV) to precipitate from solution in colloidal form. The amount of plutonium associated with the colloidal phase can be similar to that found in solution. It cannot be ruled out that the majority of the Pu in some cases might not be in solution. There are no data available for colloidal plutonium in Room 001/3, the room which contains the largest water accumulations.
6
WATER MANAGEMENT STRATEGY
In order to determine an optimum water management strategy it was necessary to identify the options available for each stage of the process, from collection through to disposal. These process stage options were then accepted or rejected by qualitative arguments. The acceptable process stage options were then combined to produce a number of strategic options. These were then assessed using semi-quantitative arguments to determine the preferred strategic option. Technologies which involved a considerable amount of work in the Shelter were rejected, since implementation could result in unacceptable construction and operational doses. For example, sludge collection by filtering the water in the Shelter was rejected. Not only was this a high dose rate option, but it would also generate an intermediate activity solid waste stream which would require provision of additional containers and premises for safe storage. The preferred strategy is to collect the Shelter water and sludges, then pre-treat them in a Shelter Water Pre-Treatment Facility (SWPTF) for removal of the organics and transuranics. The objective is to pre-treat the water to meet the ChNPP evaporator acceptance criteria. The pre-treated water will be evaporated and the ChNPP evaporator concentrates processed in the Liquid Radwaste Treatment Plant (LRTP). The sludges which arise from the SWPTF will be
166
C584/017/2000
© IMechE 2000
temporarily stored and eventually conditioned and disposed of as long lived waste. This is illustrated in the process flow diagram shown in Figure 4.
Figure 4: Shelter water pre-treatment facility flow diagram The main functions fulfilled by the SWPTF are thus: •
Hydraulically connect to Room 001/3 those rooms within the Shelter which contain water.
•
Collect the available water and sludges in Room 001/3.
•
Transfer the water and sludges to a sludge storage tank.
•
Store the sludge for an interim period.
•
Transfer the water to a treatment tank.
•
Reduce the organic content of the water by addition of oxidising chemicals (e.g. hydrogen peroxide).
•
Reduce the transuranic content of the water by adding sodium hydroxide to increase the pH of the water and thus precipitate the transuranics.
•
Transfer the liquor to a settling tank.
•
Transfer the pre-treated water to a sentencing tank and from there to the ChNPP water collection system. Evaporate the water using the ChNPP evaporator. The sludges from the evaporation process will be sent at the discretion of ChNPP to the Liquid Radwaste Treatment Plant for further treatment, including encapsulation in cement and then disposal.
•
Interim storage of the solid residues.
C584/017/2000
© IMechE 2000
167
•
Eventual encapsulation of the solid residues.
•
Eventual storage and disposal of the solid residues.
One major question was the possibility of a criticality. At an early stage, a criticality assessment was carried out on the assumption that all of the Shelter fissile material in the water, colloids and sludges was located inside one tank. The results indicated that there was not enough fissile material under any circumstances to cause a criticality. Therefore the strategy of collecting the water and sludges in one tank was sound from a criticality point of view. Arguments for location of the Facility outside the Shelter include those shown below. •
In the event of Shelter collapse, water will enter the Shelter and will require to be treated by the SWPTF. If the Facility was inside the Shelter it could be damaged by roof collapse.
•
The dose rates inside the Shelter rooms are higher than the ambient levels outside the Shelter. It would be a violation of the ALARA principle to expose operators to these higher dose rates on a regular basis.
7
FACILITY THROUGHPUT
The assumption was made that the current rate of water ingress would remain constant and that no measures would be taken in the short term to reduce water ingress. There are a number of possible scenarios which would alter the water volumes to be treated, and these are discussed below. 1. Work could be undertaken to block off some of the holes in the roof. This would reduce the amount of rain and snow which ingresses the Shelter. 2. A containment is built over the Shelter. This would reduce the amount of rain and snow which ingresses the Shelter to zero levels. There would still be a small amount of condensation water, but this has not been quantified. A containment system will not be built for practical reasons for at least 7 years or longer. 3. Dust Suppression Liquids: There could be a decrease or increase in the quantity of dust suppression liquids used in the Shelter. 4. Decontamination Waste: Liquid decontamination reagents could be used in the Shelter. These will need to be treated prior to discharge. 5. SIP Package D wastes: Package D are currently evaluating the use of wet chemical techniques for removal of some of the FCM. If this is technique is adopted, some contaminated water will require treatment. The amount of water is unknown. Figure 5 shows a hypothetical profile illustrating the volume of water which could be treated in the Facility over a twenty year period. Note that this analyses using more recent data
168
C584/017/2000
© IMechE 2000
which indicate that less water flows through the Shelter than that shown in Figure 2. A summary of the main points in Figure 5 is given below. Years 1-6: The Shelter water is treated for 6 years at a rate of 2,000 cubic meters per year. Years 7-11: Some repairs are made to the Shelter roof. There is a decrease in the volume of water to be treated. Years 11-15: In year 11, small amounts of decontamination waste are treated in the facility. This causes a slight increase in the facility throughput. Year 16: The Shelter containment is completed. About 400 cubic metros of water remains in the Shelter and this is treated in year 16. Water ingress is zero, but some condensation water enters the Shelter every year. Years 17-20: Dust suppression solutions, condensation water, decontamination water and Package D waste are processed from year 16 to year 20.
Figure 5: Illustration of water volumes which could be treated at the shelter
8
PRELIMINARY SAFETY ANALYSIS REPORT
A Preliminary Safety Analysis Report (PSAR) has been carried out on the conceptual design. The radiation doses and risks during construction, operations and under accident conditions were found acceptable, provided ALARA considerations were applied where relevant.
C584/017/2000
© IMechE 2000
169
A number of improvements to safety were identified and these will be addressed during detailed design.
9
CONCLUSIONS
There has been an extensive amount of work done to characterise the water within the Shelter and to determine the origins and volumes of water. The results indicate that the water compositions within the major water locations have stabilised and the chemical and radioachemical compositions are well characterised. Based on data generated the preferred water management strategy is to remove the water from the Shelter and pre-treat before sending to ChNPP for evaporation. The pre-treatment will remove transuranics and organics. The sludges will also be removed and together with sludges arising from pre-treatment will be temporarily stored prior to conditioning for future disposal as long-lived radioactive waste.
170
C584/017/2000
© IMechE 2000
C584/020/2000 ALARP as applied to high-level waste - the regulatory approach at Sellafield C H WAKER Nuclear Safety Directorate, Health and Safety Executive, Bootle, UK
SYNOPSIS Traditional risk management in parts of the nuclear industry assumes that if numerical risk levels, in terms of probability times consequence, are at or below numerical criteria then the requirements of ALARP are satisfied. However, risk in terms of health and safety law is to do with the potential to harm people, and if it is reasonably practical to reduce that potential then that is what the law requires, regardless of numerical risk figures. Furthermore, Nil's SAPs also set out as the basis for its safety analysis a fundamental hierarchy of deterministic principles. The first of these is to avoid the hazards and maintain safety by inherent and, where possible, passive design features. Where this is not reasonably practicable the design should be such that the sensitivity to faults is minimised. This hierarchy is applied with a rigour proportionate to the potential to do harm. These principles have been applied to NII's regulation of the storage of liquid HLW at Sellafield. They are the basis for a regulatory strategy which has the objective of reducing the hazard progressively by requiring the reduction of the stock to all but a minimum buffer quantity by around 2015. Some of the considerations NII has required of BNFL in its studies for defining a minimum buffer stock require a wider holistic view of the upstream and downstream processes to determine ALARP. This has caused NII to look at operational options itself, in order to benchmark BNFL's proposals and to constructively challenge assertions and assumptions.
C584/020/2000
© IMechE 2000
171
1. INTRODUCTION 1.1 In February 2000 NII published a report on the storage of liquid high level waste (HLW) at Sellafield (1). In it the safety issues related to the storage of HLW were discussed and it was explained why NII accepts that the operation of the plant is acceptably safe. 1.2 Liquid High Level Waste or Highly Active Liquor (HAL), is derived from the reprocessing of irradiated nuclear fuel at the Sellafield Works of British Nuclear Fuels plc (BNFL); it is stored and vitrified on the site. This waste has accumulated since the 1950s from earlier reprocessing and is also being produced from current operations. The HAL, which is a concentrated solution of fission products in nitric acid, is stored in a number of water-cooled Highly Active Storage Tanks (HASTs) housed in the Highly Active Liquid Evaporator and Storage Plant known as "B215". 1.3 HSE and BNFL have the mutually agreed policy that the HASTs should be emptied and the HAL converted into a solid form as soon as is reasonably practicable. The solid form of HLW adopted in the UK is borosilicate glass in which the fission products are incorporated, a process known as vitrification. Our policy is that this should be achieved by a target of around 2015, based on a judgement of what can reasonably be achieved. 1.4 A characteristic of HAL is that it generates heat so that, if not adequately cooled, it has the potential to reach a significantly increased temperature and, in the extreme, could boil. Overheating could lead to a reduction in the effectiveness of the containment system by, for example, over loading the ventilation clean up and filtration system. The need for high reliability cooling systems is therefore essential. By contrast the storage of HLW as a solid avoids any dependence on the continued availability of installed services such as electricity and water, or the need for operator control, because the fission products are immobilised in the solid matrix and the glass is cooled by natural air circulation. It is therefore passively safe. 1.5 Thus, despite the finding that B215 is acceptably safe NII believes that safety can and should be systematically and progressively improved by reducing the amount of HAL stored to a buffer stock, thus reducing the hazard potential associated with its storage. It also believes that the current operational mode should be reviewed to see if further improvements to safety can be achieved. It intends that these issues are controlled in accordance with the regulatory framework described below. 2. THE UK REGULATORY FRAMEWORK 2.1 The basis for the HSE policy that improvements to safety should be undertaken is the requirement in the Health and Safety at Work etc. Act 1974 (HSW Act) that employers must take measures to avert risks arising from their activities so far as is reasonably practicable. This requirement can be reformulated as requiring an employer to reduce the risk posed by its facility to As Low As Reasonably Practicable (ALARP) and is embodied in HSE's document on the Tolerability of Risk (2). 2.2 In considering safety issues it is useful to distinguish between the terms "hazard" and "risk". "Hazard" is usually defined as the intrinsic property of something which provides it with the potential to cause harm. "Risk" is the combination of the probability of the harm
172
C584/020/2000
© IMechE 2000
being realised and the consequences of such harm. In the case of HAL the hazard results from its high radioactive inventory, its mobility and its heat generating properties. 2.3 The implication, as set out in HSE's discussion document on reducing risk and protecting people (3), is that a successful approach to managing risks must ensure that hazards are properly addressed. It is also worth noting that the ACOP for the Management of Health and Safety at Work Regulations 1999 (4) states (paragraph 27) "it is always best if possible to avoid a risk altogether". 2.4 The requirement to reduce hazard further and to strive towards inherent or passive safety is reflected in the Nil's Safety Assessment Principles (SAPs) (5). The major part of the SAPs comprise engineering principles the first two of which (P61 and P62) are key. These form a hierarchy, the first of which is that hazards should be avoided by design and that safety should be maintained by passive means. Where this cannot be fully achieved, the next key principle requires the sensitivity of the design to faults to be minimized. It is these hierarchical safety principles, whose basis is in general Health and Safety law, that form the driver for HSE's policy that the HASTs should be reduced to a buffer stock as soon as reasonably practicable. A further requirement of these principles is that the design basis accidents must be assessed in the safety case using robust deterministic arguments with appropriately conservative assumptions. 2.5 To be compliant with the SAPs, a safety case should have at least three "legs" - a demonstration of sound engineering that the design is fit for purpose (based on fundamental safety principles and good practice), a conservative deterministic analysis of the design basis to show robust tolerance to more frequent faults, and a probabilistic (or quantified) risk analysis based on best estimate data for comparison with risk criteria and to search out weaknesses in the plant and its operations. The relative "weight" of these legs will depend on the specific technical considerations and the hazard. Probabilistic arguments are also important in order to ensure that the overall plant risk and the balance of risk across the plant are acceptable. Paraphasing from reference 4, it may be said that risk assessment must be suitable and sufficient for the purpose of identifying the measures needed to control the hazard. 2.6 A regulatory body must be aware of public and political concerns. Recent research for HSE into public perception suggests that there is more interest in large accidents irrespective of their likelihood (6). It is concluded from the above that intrinsic hazard is appropriate as a measure of the threat to safety and provides a meaningful concept for communication with the public. The policy on B215, that the hazard should be reduced, will address many of the concerns held by the public and appropriately focuses on the scale of the potential consequences rather than the risk in probabilistic terms. This is consistent with Reference 3. 2.7 The primary tool for nuclear regulation in the UK are the conditions attached to the nuclear site licence which HSE grants intend to use a nuclear site for defined purposes. This comes from from the Nuclear Installations Act (as amended) 1965, a relevant the HSW Act.
C584/020/2000
© IMechE 2000
powers derived from the to corporate bodies who delegated powers derived statutory provision under
173
2.8 Licence condition 32[4] relates to the accumulation of radioactive waste on the site and allows the NII to "specify" "limitations as to quantity, type and form" of the accumulated waste. NII intends to utilise this condition in controlling the amount of HAL in B215. 3. HIGH LEVEL WASTE PRODUCTION AND MANAGEMENT AT SELLAFIELD1 3.1 HAL was first stored in B215 in 1955. These early arisings were generated from the reprocessing of irradiated nuclear fuel from the Windscale Piles and the early Magnox reactors. Some of this early HAL is still stored in the so called "Old Side" of B215. HAL is currently being produced at Sellafield as a result of the reprocessing of both Magnox and Oxide spent nuclear fuels and is stored in the "New Side" of B215. 3.2 A liquid waste stream, known as highly active raffmate (HAR), is produced in the first solvent extraction stages of the reprocessing plants. This raffmate contains more than 99.9% of the radioactive fission products present in the spent fuel. The HAR from both Magnox reprocessing and the Thermal Oxide Reprocessing Plant (THORP) is transferred separately to B215 via pipe bridges but is in volumes which are impracticable to store without treatment 3.3 The function of B215 is to receive, concentrate, blend and condition the HAR and thus store as HAL prior to vitrification. On receipt the HAR is fed on a semi-continuous basis to one of three evaporators and treated until the required HAL condition is achieved. The HAL is then transferred to the HASTs designated to hold the HAL derived from that particular source (i.e. Magnox or THORP) where further in-situ evaporation is carried out. 3.4 The HAL is subsequently blended and treated to the required composition by mixing the different types of HAL and the addition of additives prior to being exported to the Waste Vitrification Plant (WVP) for vitrification. The blending programme is intended to minimise the number of vitrified product containers which have to be made in WVP by ensuring that higher burn up material from oxide fuels is blended with Magnox material (or historic wastes from the "Old Side") to optimise the amount of waste oxide incorporation in the glass and avoid it being limited by constraints such as the heat rating. The planning of the associated liquor movements and the subsequent blending operation is thus complex. 3.5 The B215 facility currently consists of 3 evaporators and 21 HASTs. It has evolved into its present form over a period of 45 years with individual HASTs which have been in operation for very different lengths of time, ranging from only 10 years for the newest tanks, to 45 years for the oldest. The "Old Side" tanks, HASTs 1-8 ( Fig 1), were commissioned between 1955 and 1968 and are each of 70 m3 nominal capacity. These tanks have either one or three internal cooling coils, through which water is circulated to cool the HAL. 3.6 HASTs 9-21 (Fig. 2 and 3) - the "New Side" - are all of 150 m3 nominal capacity and were commissioned between 1970 and 1990. These HASTs each have seven internal cooling coils, and one or more external cooling jackets and include systems for agitating the contents. The design of these larger HASTs evolved from HAST 9 onwards. The main areas of development were the extension of the cooling jackets to the full operational height of the vessel from Tank 12 onwards, and modifications to the HAST emptying systems. 1
A more detailed description may be found in Reference 1.
174
C584/020/2000
© IMechE 2000
3.7 The HAL is maintained within a temperature range of 50-60°C requiring the active involvement of the operator, as does the maintenance of other properties of the HAL Cooling water is supplied to the HASTs via an open loop cooling water system from one of two sets of forced draft cooling towers. The cooling system and its associated pumps and power supply system have a high reliability with a large degree of redundancy. 38 The HASTs provide the primary containment for the HAL and are housed in the cells, (heavily shielded enclosures) to provide a secondary containment. The gases ventilated from the HASTs are routed to a dedicated vessel ventilation system which removes entrained radioactivity from the off-gases prior to them being discharged to the environment. 3.9 In the event of the need to take a HAST out of service, spare capacity is available to allow the contents of the HAST in question to be transferred to another tank. The spare capacity is equivalent to one spare tank for every three working HASTs, and is known as the "one-in-four" spares policy. This policy has been a requirement which NII has consistently maintained over many years. 3.10 The characteristics of the waste vary from tank to tank. The degree of radioactive decay of the fission products is related to the storage time. Consequently, the isotopic composition of the waste changes and the heat generation rate decreases as the storage time increases. The current heat generation rate of the HAL varies from about 20 - 460 kW per tank from Magnox fuel and can be up to twice the higher figure for oxide fuel. HAL from oxide fuel also has a higher specific radioactivity as a result of higher burn-up in reactors. 3.11 WVP came into active operation in 1991 and comprises two vitrification lines known as Lines 1 and 2. A number of operational problems have resulted in the throughput of WVP being significantly lower than originally intended. As a result BNFL has constructed a third vitrification line which is being commissioned. A discussion of the operational experience of Lines 1 and 2, and the improvements carried out to improve plant throughput is contained in Ref 1 and 7. 3.12 In WVP the HAL is first evaporated, dried and partially denitrated to produce a fine powder known as calcine. The calcine is then fed with crushed glass into a melter, in which the glass melts and the calcine dissolves, to create a molten mass. This is then poured into a stainless steel container, allowed to cool, and a lid is welded on. The container is transferred in a shielded flask to the adjacent Vitrified Product Store which includes an "export facility". 4. THE REGULATORY POSITION 4.1 Background 4.1.1 Subsequent to the publication by NII of reference 1 BNFL were required to respond to the report by 18 August 2000 and agree emptying curves with NII or NII would impose them. BNFL responded comprehensively to all 22 recommendations in NII's report before the deadline. In relation to the requirement to empty the tanks to a buffer stock by 2015, an emptying curve in terms of volume has been offered based on realistic assumptions about the ramp up of vitrification capacity and using the most up to date business plans for Magnox and THORP business. The submission also outlines areas for further development with milestones
C584/020/2000
© IMechE 2000
175
which should ensure continuous improvement to vitrification performance and tank operational regime. 4.2 Key Issues 4.2.1 Although quantified risk analysis in the BNFL safety case shows that the overall risks arising as a result of the operation in B215 are consistent with the requirements in Nil's SAPs, the P61/62 drivers raise the issue as to how the principle of ALARP is to be judged and the need for the harm potential to be further reduced. That debate is still ongoing as a result of the responses received by HSE to its discussion document at reference 3. However, Nil's approach has been to expect the licensee to establish the "modern standard" - i.e. what would it look like if designed today. This can be done by a process of "optioneering" to identify what is possible from a safety perspective and then to argue on the grounds of reasonableness and practicability how close the ideal can be achieved. 4.2.2 Whilst reprocessing continues there will always be a need for a buffer stock of liquid HAL to be held in order to allow operational flexibility between the reprocessing and vitrification plants and to enable the necessary conditioning of the feed stock to the vitrification plant. Thus achieving more inherent safety means reaching the point where there the storage of HAL is minimised , and in the meantime operating the existing plant such that it is inherently safer. Judging what this should "look like" must be done in a holistic manner. It will include ensuring the best use of the vitrification capacity and be clearly seen to be minimizing the harm potential whilst enabling the licensee to meet its reasonable and legitimate commercial aspirations. 4.2.3 There are a number of related issues in the case of HAL storage. Firstly there is the question of defining what is meant by a minimum buffer stock. The minimization of the amount stored will reduce the hazard potential, most obviously by reducing the consequences of any accident and should reduce the potential for harm to as low as reasonably practicable. The current design of tank and the operational mode effectively limits this to at least one tank. Our own assessment studies indicate that in the ideal small volumes may be achievable, a situation we have termed "near real time" vitrification. Nil's report required BNFL to challenge the present operational limits, including the consideration of smaller intrinsically safe tank designs for housing the post 2015 buffer stock. BNFL's response to the NII report has identified a number of key development studies with milestones towards this. 4.2.4 Secondly there is the question of the emptying strategy and the time and rate of approach to achieving the buffer stock. NII wishes to see the vitrification capacity exceed the rate of arisings so that the accumulation over the years can be reduced. This state has yet to be achieved and depends on the vitrification performance improving. NII welcomes the milestone programme that BNFL has put forward in its submission aimed at tackling the technical issues associated with WVP performance. We believe this is the key to success with its proposal. 4.2.5 The emptying strategy is technically complex and, in addition to vitrification performance, depends upon the assumed reprocessing inputs. Another considerations is the options for blending the different types of HAL so as to maximise the incorporation of oxides
176
C584/020/2000
© IMechE 2000
per unit volume of glass. This is important whilst vitrification capacity is limiting since it maximises volume reduction for a given production rate of vitrified product containers. 4.2.6 Thirdly there is the question of the parameters that should be used for monitoring the systematic improvement in safety, which is the whole purpose of the emptying strategy. Emptying discussions to date have used volume as the monitoring parameter. However, it has to be recognised that in the context of safety this has to be broken down to the constituent parts (i.e. THORP, Magnox - including relative age - and historic). This is because the hazard and risk are, to a first order approximation, strongly related to the oxide component, due to its radioactive inventory which dominates the potential dose in an accident, whilst the propensity to overheat is related to the material's age. For example, the historic HAL in the "Old Side" has cooled such that it cannot boil. 4.2.7 In its public report (1), NII identified a number of physical parameters in discussion with BNFL, but these parameters are only potential surrogates for measuring hazard or risk. It is recognised that none of the identified monitoring parameters alone is a satisfactory surrogate for the "risk". For example, another problem with volume that it is sensitive to both the performance of the evaporators and in tank evaporation. Whilst useful for monitoring improvements in safety as the plant moves towards the buffer storage state, and for informing decisions on options for that state, none of the parameters provide a substitute for a full safety analysis. BNFL has proposed that a safety index be developed and has included this in the milestone programme with its submission. The development of a meaningful index, which will show how the hazard changes with time, is a an important requirement to which NII is asking BNFL to give some urgency. 4.3 Current Status 4.3.1 NII has welcomed the response and recognises the considerable effort by BNFL to put it together on the time scale. The issues and the modelling by NII of options, in order to make a technically sound and quality assessment, are complex.'Discussion with BNFL to reach a technically feasible and acceptable position is not yet complete, so outcomes cannot be reported here.2 5. CONCLUSION 5.1 BNFL has made a large number of commitments which NII welcomes. A final decision on the tank emptying curves which will be used for regulatory purposes has yet to be reached. When achieved NII will use a Specification under Licence Condition 32[4] to ensure that BNFL meets its commitments. 5.2 A monitoring programme is being developed which will enable progress in the longer term to be reported upon as part of normal regulatory business via Nil's reports to the Sellafield Local Liaison Committee. 5.3 This short paper has of necessity touched upon only the more significant aspects of long and complex process of assessment and determination of an ALARP position. In the 2
However, NII promised in its report to publish an addendum in 18 months (i.e. mid 2001) explaining the outcome of its assessment and the final regulatory position.
C584/020/2000
© IMechE 2000
177
search for an agreed ALARP solution NII has sought to achieve a technically robust position that meets ALARP in all its facets, both deterministic and probabilistic. ACKNOWLEDGEMENT Although the author is leading for NII on the regulation of the B215 HAST emptying, it goes without saying that he is strongly supported by a team of inspectors that are working together in the assessment of the complex issues and their resolution. The author wishes to acknowledge their contribution to the thinking and analysis that is reflected in this paper. The paper is, however, the view of the author on a developing situation and therefore it does not represent Nil's formal view. The author is however grateful to NII for permission to publish. REFERENCES 1. The Storage of liquid High Level Waste at BNFL Sellafield, Health and Safety Executive (www.open.gov.uk/hse/nsd), February 2000. 2. The Tolerability of Risk from Nuclear Power Stations. HMSO Publications. ISBNO 11 886368 1, 1992. 3.
Reducing Risks, Protecting People, A Discussion Document, HSE Books, May 1999.
4. Management of Health and Safety at Work: Approved Code of Practice for management of Health and safety at Work Regulations 1992: HMSO Publications. 5.
Safety Assessment Principles for Nuclear Plants. HMSO Publications.
ISBNO 11 882043 5. 1992. 6. Public Perception of Risks Associated with Major Accident Hazards, HSE Contract Research Report 194/1998. 7. Progress with highly active waste vitrification at BNFL Sellafield. The Nuclear Engineer, Volume 36, No. 2, March/April 1995.
178
C584/020/2000
© IMechE 2000
S
Figure t
Cut away section of a smaller (70m3) HAST
Figure 2 Principal components of the high active waste storage tank
Figure
C584/020/2000
3 Internal arrangement of a larger (150m3) HAST Note: The figure is based on HASTs 18 - 21
© IMechE 2000
181
This page intentionally left blank
C584/027/2000 Radiation safety problems arising with damaged nuclear submarines utilization V A MAZOKIN, M E NETECHA, YU V ORLOV, G A STANISLAVSKI, G A VASILIEV, and V V BORISOV Research and Development Institute of Power Engineering, Moscow, Russia
Reactor compartments (RC), which have lost their primary circuit tightness due to accident, involving the coolant ingress in the reactor compartment sections, while the reactors have (had) the cores with damaged fuel elements, are referred to RC with damaged nuclear steam supplied systems (NSSS). Radiation situation in the compartments after accident is characterized as hazardous and intolerable. Additional safety activities are required for the work in the compartments. The presence of nuclear fuel in such NSSS reactors means that the compartments are potential nuclear hazard. Before 1999 there were seven nuclear submarines (NS), their reactor compartment state which can be classified as the damaged one. They are NS of project 675 serial No. 175, 533, 541, project 671 serial No. 610, project 705 serial No. 900, 910, project 705k serial No. 105. Late in 1999 - early in 2000 irradiated fuel was unloaded from reactors of a damaged NS. It was the first complicated enough effort aimed at removing fuel from damaged cores, which permitted transfer of the NS mentioned to the category of conventional NS subject to utilization. The second damaged compartment has been removed from operation by now and is at a standstill in Gremikha. To formulate the main problems arising from utilization of NS with damaged NSSS, let us consider the technical state and radiation conditions in the reactor compartment in the postaccident period (see Table 1) - slide 1.
C584/027/2000
© IMechE 2000
183
TECHNICAL STATE AND RADIATION SITUATION IN THE AREA OF RC OF DAMAGED NS Nos.
1.
NSSS type VM-A
Accident character Date of accident 1979 Primary circuit coolant leakage and loss of integrity of fuel cladding
Radiation situation after 20 years of cooling Reactor screened section, port side - 0.5-2.3 mSv/h Reactor screened section, starboard side - 0.025-0.12 mSv/h Galleries - 0.005-0.049 mSv/h Light-weight vessel, above RC - 0.003-0.03 mSv/h In adjacent sections - 0.002-0.0025 mSv/h Beta contamination Reactor screened section, port side - 6500 particle/cm2min The rest RC sections - 40-190 particle/cm2min Major contaminant is Cs-137
Accident consequences The primary circuit lost its tightness. Reactor partition and the hold were contaminated by radioactive products. Starboard side NSSS was not damaged. Spent fuel can be unloaded from the reactors using the additional protection means for the personnel
TECHNICAL STATE AND RADIATION SITUATION IN THE AREA OF RC OF DAMAGED NS Nos. 2.
NSSS type OK300
Accident character Date of accident Primary circuit coolant leakage, 1985 accompanied by loss of integrity of a portion of fuel elements in the core
Radiation situation after 15 years of cooling Equipment screened section, reactor head, port side - up to 75 mSv/h Passage gallery - 7 mSv/h Pump screened section of floors I-III - 18-60 mSv/h Partitions of adjacent sections - 0.3-30 mSv/h Light-weight vessel above the reactor - 0.8 mSv/h Contamination of screened equipment section - 36000 particle/cm2 Major contaminant is Cs-137
Accident consequences The primary circuit lost its tightness. Reactor core was seriously damaged, inner premises of RC are severely contaminated by radionuclides. Radiation situation inside RC does not permit unloading of the cores from the reactors.
TECHNICAL STATE AND RADIATION SITUATION IN THE AREA OF RC OF DAMAGED NS Nos.
3.
NSSS type VM-A
Date of accident 1985
Accident character Self-sustained chain reaction during reactor refueling, fire in RC and ingress of sea water into
Accident consequences Reactor internals and reactor core were destroyed as well as reactor partitions and strong hull.
it
Radiation situation after 15 years of cooling Dead-end gallery - 14 - 60 mSv/h Passage gallery - 3 - 100 mSv/h Pump screened section (floor III) -1-3 mSv/h Partition of adjacent sections - 0.3 - 0.4 mSv/h Above RC at the light-weight vessel level - 3 mSv/h Major contaminant is Co-60
The technical state and radiation situation do not permit any operations in RC.
TECHNICAL STATE AND RADIATION SITUATION IN THE AREA OF RC OF DAMAGED NS Accident character Date of accident Loss of tightness of the CPS shim 4. 1989 rod claddings; radionuclide ingress in the upper cavities of the actuators shrouds Radiation situation after 10 years of cooling Streaming radiation over CPS shrouds reaches up to 30 Sv/h in a local area above the head Nos.
5.
NSSS type OK550
VM-A
1989
Primary coolant leak and loss of fuel cladding integrity
Radiation situation after 10 years of cooling Above the reactor head - 3.1-14.5 mR/h On light-weight vessel above the reactor - 0.4 mR/h Dead-end gallery - 1.5 mR/h Passage gallery - 100 mR/h Pump enclosure (floor III) - 0.9 mR/h Partition of the adjacent compartments - 0.5-1.5 mR/h
Accident consequences Radiation situation deterioration in the reactor head area. Unloading of spent fuel is possible if the additional means are applied for personnel protection.
The primary circuit lost its tightness, the hold and some premises in the RC were contaminated by radionuclides. RC can be removed if the additional means for personnel protection are applied.
The results provided in Table 1 (slide 1) suggest that, in the first place, the radiation safety problems in the NS are closely interlaced with necessity to prevent nuclear hazard and, in the second place, with the need to restore the destroyed or to create additional ecological barriers preventing radioactivity ingress to the environment. Thus, all the problems are concentrated around two major ones:
• nuclear fuel unloading from the reactors; • • bringing of RC with the damaged NSSS into radiation and ecologically safe state, bearing in mind subsequent transportation and storage in long-term storage places. The analysis of the damaged RC state and radiation situation indicates that under certain conditions at some projects (items 1, 4 and 5 in Table above) it proved possible to unload fuel from reactors and to localize the radioactivity remaining within RC, i.e. using welding to seal primary circuit nozzles, remove transition pipelines and plug their attachment points, install the additional ecological barriers (non-standard partitions along ends of RC, insulation of the lower part of the strong hull below the reactor, filling the reactor with a solidifying conserving agent) and isolate and seal the inner space of RC. The scheme of the protective barriers for RC is shown in slide 2. As was stated above, this technology has been implemented in part in 1999 at the nuclear submarine given as item 5 in Table 1. After 10 years following the accident the submarine has been brought into the ecologically safe state, namely: •
nuclear fuel was unloaded from the reactors;
•
standard heads were put onto the reactor vessels;
• primary coolant and liquid radwaste were removed from RC hold and transported for disposal; •
inner cavities of the reactor were filled with the solidifying conserving agent;
• the operations on preparing this submarine for short-term storage afloat are close to completion at present; • as soon as the local depository for RC from NS to be disposed is put in service, RC will be cut off (removed) from this submarine and subject to cooling procedure according to the standard technology. For the damaged RC, from which unloading of SNF is currently impossible (see items 2 and 3 in Table 1), it is suggested that the additional barriers are made not only inside RC, but outside them as well. Specifically, each of the above-mentioned damaged RC can be placed in a cylindrical steel casing, one of the RC larger diameter of the NS subject to utilization being used as such, for one example. For this variant of handling the damaged RC of NS (item 3, Table 1) the Central Design Office of Marine Technologies (Ts KBMT) «Rubin», RDIPE and other organizations have performed design and technological developments for bringing the
188
C584/027/2000
© IMechE 2000
RC in nuclear, radiation and ecologically safe state, the project being named «Sarkofag» (Sarcophagus) - slide 3. Specifically, nuclear safety of the reactor with unloaded core can be assured by removal of moderator (water) from the reactor, by its filling with solidifying conserving agent with absorbing additives, which provides a deep subriticality of the core, ruling out reactivity compensation members movement and water ingress inside the reactor, emergency external impacts of technogenic and natural character inclusively. Radiation safety is assured by providing an additional biological shield inside and outside of RC, as well as on the «sarcophagus» body. The points of installation and the thickness of the additional biological shield (made of concrete) provide gamma radiation exposure reduction to a value of 10 mR/h at most at a distance of 1 m from the «sarcophagus» body, which meets the requirements specified by the sanitary regulations for radioactive waste management SPORO-85. Ecological safety is provided by placing the damaged RC in a strong leak-tight shell made of three leak-tight missile compartments forming the «sarcophagus» and taken from the Pr. 677B NS PC which are to be utilized. The damaged RC is to be installed in the middle section of the «sarcophagus», being thus reliably isolated from the environment by its strong hull and two inter-sectional partitions on each side. In 50-60 years the radiation situation in the RC will be normalized as a result of natural decay of Co-60 dominating radionuclide, the conditions for conducting the activities of spent fuel unloading will emerge and the «sarcophagus» structures, including the damaged RC, could be utilized. The reactor vessels and iron-water shielding tanks will play the role of protective containers for remaining radioactive internals and they will be subject to subsequent arrangement in RW storage. As for the damaged NS (item 2, Table 1), where radiation situation is determined by Cs-137 radionuclide, its half life being 30 years, one should expect the normalization of radiation situation not earlier than 350-370 years later. So, for this NS PC the «sarcophagus» type package can not assure ecological safety for so long period owing to insufficient corrosion resistance. It seems that setting the RC in concrete, wrapping it by sand, gravel and rocks of the necessary thickness with water proofing and controlled leakage arrangement is a more reliable and feasible way to isolate the damaged section of this NS from the environment - slide 4. Hence, solution of nuclear and radiation safety problems during management of RC containing damaged NSSS shall proceed along the following directions: 1. Reactor compartments with damaged NSSS, their technical state and radiation situation permitting unloading of nuclear fuel from the reactors, shall be brought to nuclear-safe state by SNF removal from them; procedure for their further handling shall include the necessary measures of radiation and ecological safety. 2. In the cases when it proved impossible to unload SNF for reasons of technical state or radiation situation, the moderator (primary coolant) shall be removed from the reactor, shim rods shall be fixed in their bottom position thereby excluding their reciprocal
C584/027/2000
© IMechE 2000
189
movement, and the protective «container» (package) should be formed using the standard strong hull, partitions and additional protective barriers. If for some reasons the package could not provide ecological safety during long-term storage of RC, then more complex and expensive projects shall be used such as «sarcophagus» or «shelten> packages which have to be developed specifically for each damaged NS taking into account the particular circumstances of its technical state.
190
C584/027/2000
© IMechE 2000
C584/010/2000 Experience in nuclear decommissioning and waste management G R EDLER and D BRADBURY Bradtec Decon Technologies Limited, University of the West of England, Bristol, UK C J WOOD EPRI, Palo Alto, California, USA
ABSTRACT Chemical decontamination of radioactive components is a now a well-established procedure. Many components, particularly metallic ones, become contaminated with radioactive materials during normal operation of a nuclear power plant or other nuclear facility. In an earlier conference paper(1) a report was given on the application of the EPRI DFD Process to full reactor systems. At a recent EPRI Decommissioning Workshop'2' a number of papers were presented giving details of the benefits realised in the last two years. A summary of the decommissioning experience showed that techniques such as decontamination can offer savings in radiation exposure, waste disposal and dismantling times. New decommissioning challenges can be met with the development of new technologies. This is best achieved by close cooperation between the end user and the process developer.
1.
INTRODUCTION
Chemical decontamination of radioactive components is a now a well-established procedure. Many components, particularly metallic ones, become contaminated with radioactive materials during normal operation of a nuclear power plant or other nuclear facility. However, unless the component has been activated by a neutron flux, this contamination commonly occurs on the surface of the component only. The benefits of chemical decontamination include the reduction of radiation dose to people working on or close to, the component in question. More recently there are examples where the efficient decontamination of redundant components during decommissioning can allow the cleaned components to be released from radioactive materials controls so that they can be recycled or disposed of in conventional industry. This not only has economic advantages but
C584/010/2000
© IMechE 2000
191
benefits the environment as well through the recycling of valuable materials and the reduction in the volume of radioactive waste requiring disposal. Physical decontamination methods (eg such as shot blasting) are often preferable to chemical methods for cleaning externally contaminated surfaces which are easily accessible. However, these methods cannot normally be applied conveniently to complex components, structures or systems where the contamination is on the inside of the component or system. Chemical decontamination is usually preferable for such tasks. 2.
THE EPRI DFD PROCESS - RECENT APPLICATIONS
2.1 Full System Decontamination In an earlier conference paper(1) a report was given on the application of the EPRI DFD Process. This paper described the application of the process to full reactor systems after final shut down, and to a variety of components. Since then dismantling of the primary circuits has begun and information is now available on the benefit accrued due to decontamination. At a recent EPRI Decommissioning Workshop (2) a number of papers were presented giving details of the benefits. Ken Pallagi (3), speaking on the Big Rock Point decontamination(1), stated that two key issues in the decommissioning were dose reduction and alpha contamination control. As a result of the decontamination no airborne alpha was identified, no uptake or ingestion occurred, no dose assessments were required and only two positive smears in the RCP room piping were found. These measured 23 dpm/100 cm2 and 41 dpm/100 cm2. Immediate benefits identified were dose reduction and relaxing of job coverage requirements (two technician coverage versus four to six technicians). Due to the lower backgrounds, hot spots were easier to locate and evaluate. Removal or shielding of hot spots greatly reduced the dose received by workers. Further benefits from the radiation protection and ALARA perspective were minimal shielding requirements, minimal interference with work, no additional engineering controls required on the ventilation system and that during dismantlement of the primary circuit the containment building remained clean with no interference of other work. Table 1 is a summary of the dose savings in the recirculation pump room. Table 1 - Summary of Dose Savings in the Big Rock Point Recirculation Pump Room YEAR
1998 1999 April 2000
RECIRC PUMP ROOM WITH DECON 670 mSv (67 Rem) 340 mSv (34 Rem) 130mSv(13Rem)
ESTIMATED WITHOUT DECON 1150mSv(115Rem) 1630 mSv (163 Rem) 870 mSv (87 Rem)
ESTIMATED DOSE SAVINGS 480 mSv (48 Rem) 1290 mSv (129 Rem) 740 mSv (74 Rem)
Similarly Paul Plante and Glenn Collins(4) reported the benefits realised at Maine Yankee(1) as a result of the primary circuit decontamination. Table 2 is a summary of the dose benefits determined for the RCS system only.
192
C584/010/2000
© IMechE 2000
Table 2 - Benefit Determination at Maine Yankee for the RCS System Only Year
1998 1999 2000 Total
Actual Dose 560 mSv (56 Rem) 940 mSv (94 Rem) 520 mSv (52 Rem) (April) 2020 mSv (202 Rem)
Dose Estimate w/o Decon 1550 mSv (155 Rem)
Dose Savings 980 mSv (98 Rem)
2900 mSv (290 Rem)
1960mSv(196Rem)
870 mSv (87 Rem)
350 mSv (35 Rem)
5320 mSv (532 Rem)
3300 mSv (330 Rem)
Major Work Accomplished Asbestos Abatement RCP Removal Large Bore pipe, S/G Preps Loop Clean-up S/G & PZR Removal
The 3300 mSv (330 Rems) saved by the EPRI DFD treatment of the RCS system compared favourably with the original estimates of 2710 mSv (271 Rems) for a DF of 100 and 2560 mSv (256 Rems) for a DF of 15. Other benefits described included savings in packaging of radioactive materials from containment, minimising of dose to the public during on-site storage of radioactive materials, minimising of dose at the waste reprocessers, and lowered dose to the public during transportation of radioactive materials. Non technical aspects such as a positive impression by stakeholders and regulators were also reported. Lessons learned from these projects included •
"expect the unexpected"
•
flexibility in procedures and planning reduce delays
•
secondary waste volumes are not always determined by the decontamination process alone.
There was general agreement by the engineers that decontamination is best applied as soon as possible after final closure because of •
the use of station equipment, eg. reactant coolant pumps
•
having expert staff available on site
•
maximising the exposure saving benefits.
One trend associated with the removal of components is the diminishing benefits from the decontamination. The benefits are therefore most pronounced with high dose removal jobs early in decommissioning.
C584/010/2000
© IMechE 2000
193
2.2 Process Pumps The process has recently been used with great success in the USA for the decontamination of 400 series stainless steel pump components. A particular new variant of the process called "EPRI DFD Lite" is used for these components. The EPRI DFD Lite Process is applied to the pump components so that the components can be released to a workshop which has no radioactive materials controls for refurbishment and repair. This is considerably less expensive than the alternatives of discarding the components, or repairing in a radioactive workshop. In a recent paper the Southern Company(5) stated that application of the EPRI DFD Process had saved over $2M in the refurbishment of pumps in the period up to 1999 . The application of EPRI DFD Lite is performed by ALARON Corporation at their facility in Pennsylvania USA. The equipment used is an inexpensive recirculating system reported to have been assembled for approximately $5,000(6). Figure 1 shows a pump impeller being removed from the process tank.
2.3 US Department of Energy Trials have taken place at the Oak Ridge Reservation near Knoxville, Tennessee. Part of the accumulated redundant plant waste is the discarded aluminium compressor blades which have failed in service and subsequently been removed during maintenance outages. Several drums of blades were received by Decontamination Recovery Services (DRS) to be used in a decontamination trial at their Oak Ridge facility. DRS supplied a secure facility and project support such as health physics. The EPRI DFD Process was applied and engineered by a combined team from Practical Machine engineering (PMe) and Bradtec Decon Technologies. The blades tested were chosen at random from each of the four drums provided, but care was taken to ensure that a mixture of sizes was included. The isotopes of primary concern were uranium and technetium, technetium being the more abundant of the two. Over 70% of the blades were able to be released with some of the remainder being volumetrically contaminated due to re-casting in previous recycling campaigns. After the process trials both the cation and anion resins were successfully regenerated and a neutralised sludge produced. The sludge was successfully incorporated into a cement matrix. Samples of resin taken before regeneration were subjected to a Toxic Characteristic Leach Procedure (TCLP) test by an independent laboratory. The resin successfully passed the test. Security considerations did not allow the blades to be released from the DOE compound. However it was agreed that small cored samples could be removed from the blades for further laboratory investigation. The samples were tested at Bradtec's facility in Bristol. Figure 2 shows some samples before and after decontamination. The laboratory results supported those already recorded in the pilot scale work. Although the trials were technically very successful, a significant slowdown in progress has occurred as a result of the recent moratorium on recycling of materials announced by the Secretary Richardson in the United States on July 13, 2000. However, the main reason for concern leading to the moratorium is the potential for contamination in the public domain. There are many opportunities to recycle material which do not lead to the material being released into the public domain, and applications of decontamination to achieve unrestricted
194
C584/010/2000
© IMechE 2000
release standards are still likely to form an important future business, albeit with more restrictions placed on the outlet materials. 2.4 Feasibility Study Information released recently by the Ministry of Defence'7' indicates that after a study by them, they are considering the possibility of land storage for redundant submarine reactors. In a separate move Babcock Rosyth Defence Limited have been given approval to start feasibility and planning work on a proposal they have put forward to dismantle the reactor compartment from one of the decommissioned nuclear submarines, HMS Renown. Studsvik AB and Babcock International Group plc have recently formed a joint venture company known as Studsvik Rosyth Nuclear Services Ltd (SRNS). The new company is based at Rosyth Dockyard in Scotland. The function of the company is to provide services to the UK nuclear industry, particularly in the field of decontamination and radioactive waste management in decommissioning of nuclear facilities. Decontamination technologies are provided to SRNS by Bradtec Decon Technologies. The EPRI DFD Process is a central part of the Renown feasibility study. After decontamination one possible option for the metal components is that they could be sent to the Studsvik facility in Sweden where metal is melted to be released for recycling. Melting provides a further layer of assurance when measuring materials for unrestricted release. Internal surfaces and complex components are homogenised in the melt and processed into metal ingots. This process affords the opportunity for the precise determination of radioactivity and the production of archive samples. Melting also reduces the need for expensive space in waste repositories because only the slag has to be disposed of. Radioactive components with low radioactive content such as heat exchangers and moisture separators are routinely treated at the facility, which has an induction furnace with a capacity of 1500kg per hour. Aluminium is treated in a crucible furnace of about 250kg capacity. The facility has recently been upgraded to achieve a greater routine throughput. Materials are brought to the facility for processing not only from within Sweden but also from other countries such as Germany. Ingots from the melting process which achieve free release standards can be released in Sweden, but all secondary radioactive waste is returned to the originator. 3.
FOAM DECONTAMINATION
Although the preliminary step of filling a component or system with water for decontamination can normally be achieved, it is not always convenient. Some nuclear components are not designed to be filled with water, and the weight of the system when full would be in excess of the structural limits. An example of such a system is the Magnox gascooled reactor boiler. These items are carbon steel heat exchangers which (during operation) exchange heat from the reactor coolant gas to the water/steam circuit. The gas side of these boilers is contaminated, and was not originally designed to be filled with water. Another disadvantage of using water as the decontamination medium is that in systems with a large internal volume it is difficult to avoid stagnant pockets during the decontamination. If the decontamination chemical reagent is not efficiently circulated through the whole system
C584/010/2000
© IMechE 2000
195
volume during decontamination, then that part of the system in the stagnant pocket may not be efficiently cleaned. Such a problem would be encountered, for example in attempting to decontaminate a Boiling Water Reactor Steam Turbine. The restrictions on the weight of water could in principle be overcome by applying the chemical reagent in the form of foam, in which the liquid volume is highly expanded by entraining a gas. A new procedure has been developed for applying foam decontamination'8'. A foam decontamination reagent is placed inside the system or component to be decontaminated in an appropriate quantity to occupy a small proportion of the overall system volume. This proportion may be any proportion between about 1% and 10% of the system volume. Gas is introduced through a suitable inlet or inlets into the liquid volume at the bottom of the system. The gas becomes entrained to expand the liquid and thereby cause it to fill the entire volume of the system. When the decontamination reagent capacity is used up, the gas flow is ceased and the liquid is allowed to collect at the bottom of the system. The foam decontamination liquid is then removed from the system by pumping out or by gravity drain. The system surfaces are rinsed with clean water and, if necessary, returned to a dry condition thereafter. The radioactive waste management of the combined foam and rinsing solution employs traditional methods and principles. A filter may be used to remove insoluble particulate material from the waste solution. The waste solution may then be routed to a waste hold-up tank. In this tank the solution may then be mixed with chemicals added to achieve pH neutral conditions. The liquid may then be routed to an evaporator. For evaporation to take place efficiently it may then be desirable to add small amount of a suitable anti-foaming chemical. The condensate from the evaporation process can be recycled for use as rinse water or for further reagent make-up. The residue may be routed to waste drums for in-situ grouting with cement. The waste drums can thereafter be sealed and transported away for burial. Some trials of this new process have taken place with samples removed from a Magnox Boiler (see figure 3).
4.
FUTURE CHALLENGES
Nuclear decommissioning continues to raise new technical challenges and also continues to be a subject of great interest to the environmental lobby. Techniques such as decontamination can offer savings in radiation exposure, waste disposal and dismantling times. The decommissioning of nuclear reactors to a green field site in no more than ten years (Big Rock Point plans 8 years) is a measure by which the public can judge the maturity of our industry to cope with the aftermath of electricity generation by nuclear power. The American Light Water Reactors (LWRs) lend themselves more easily to early decommissioning. The need to develop suitable decommissioning technologies requires that engineering expertise and also a commercial incentive exist within the industry. We have at present the ability to develop and engineer decommissioning processes. This paper refers to one such process, EPRI DFD, which was initiated as a research programme in 1994 and by the end of
196
C584/010/2000
© IMechE 2000
February 1998 had been applied to two full reactor systems. This was only possible with the support of the US generating utilities via the research funding from EPRI. New fledging technologies, foam decontamination, and ideas, graphite pyrolysis'9' exist today. These technologies are best developed in conjunction with the end user to ensure a focussed and successful development programme. Ultimately everyone benefits from good process design and implementation when taking on the challenges of nuclear decommissioning. 5.
CONCLUSIONS
•
In the last two years of decommissioning at Big Rock Point and Maine Yankee, the benefits of the initial application of a full system decontamination using the EPRI DFD Process have been assessed.
•
It was agreed that considerable benefits were gained as regards reduction in personnel dose, reduction in waste generation and reduction in dismantling times.
•
Non technical benefits such as public acceptance and stakeholder confidence were also reported.
•
New decommissioning challenges can be met with the development of new technologies. This is best achieved with cooperation between the end user and the process developer.
REFERENCES 1.
Elder, G R, Bradbury, D, Wood, C J, "The application of EPRI DFD process for full reactor system decontamination post operational shut down", IMechE Conference Nuclear Decommissioning '98, Paper no: C539/016/98, page 113.
2.
Plant, P, Collins G, "System chemical Decontamination: Assessing its Benefit on Decommissioning Maine Yankee", EPRI Decommissioning Workshop, 13 June 2000.
3.
Pallagi, K, "Decontamination Benefit Assessment", Decontamination, ALARA and Worker Safety Workshop, June 2000.
4.
EPRI Innovators 2000, IN-114722
5.
Harverson, J, "ALARON's Experience using EPRI DFD Free Release Contaminated Components", EPRI Chemical Decontamination Conference, Greenville, SC. May 18-19, 1998.
7.
MoD Contracts Bulletin, 31 May 2000, p.20
8.
Foam Decontamination - British Patent Pending
9.
Mason, J B, Bradbury, D, "Pyrolysis and its Potential use in Nuclear Graphite Disposal", IAEA Technical Committee Meeting on Nuclear Graphite Disposal, 18-20 October 1999, Manchester, UK.
C584/010/2000
© IMechE 2000
197
Figure 1 - Pump Impeller treated by the EPRI DFD Lite Process
198
C584/010/2000
© IMechE 2000
Figure 2 - Aluminium Samples taken from Compressor Blades Top Row Untreated, Bottom Row Treated with the EPRI DFD Process
C584/010/2000
© IMechE 2000
199
Figure 3 - Foam Trials on Tube Samples from a Magnox Boiler
200
C584/010/2000
© IMechE 2000
C584/019/2000 Disposal of radioactive waste - a puzzle in four dimensions I J DUNCAN School of Geography, University of Oxford, UK
SYNOPSIS Society accepts that its demands for materials, energy, food and transport all create wastes some of it hazardous - and yet the public resists the disposal of such wastes, particularly if the intended site for a repository is in their locale. Waste disposal can be categorised into one of two possible regimes, firstly dilution and dispersal into the biosphere, or secondly concentration, containment and isolation from the biosphere. Waste can be gaseous, liquid or solid and varies greatly as to the degree of chemical and physical hazard. In the case of radioactive waste, the radiotoxicity decays with time and there is chemical change, two of the parameters that illustrate that waste disposal is indeed a dynamic process. 1. INTRODUCTION Some might argue that the public perception of the disposal of long-lived solid waste is a matter of placing it into a repository and trusting that it will remain there forever - a static situation. However research has shown that the public does not trust such a proposal and fully expects the system to fail and the waste, while still hazardous, to re-connect with the biosphere. Waste disposal experts counter this view with scientific expressions of degree of hazard and probability such as the risk that an individual human being will suffer a serious health effect (fatal or genetic) from any releases of radioactive material from a sealed repository should be less than one in a million (10-6) per year. It is seldom realised in the industry that such a statement is unintelligible to the public (5, p. 194). This paper develops a hypothesis that takes into account public perceptions of time and trust and considers the disposal of solid contained Intermediate Level Waste (ILW) as an example. The outcome could however be equally applicable to the disposal of all hazardous waste including Low Level Waste (LLW) and High Level Waste (HLW). Specific issues addressed include the questions: • Is the disposal of radioactive waste a static or dynamic process? • Does the public trust geological disposal? • If there is a failure in the system can it be remedied? • How can science and engineering develop a disposal method that will not only isolate the waste from the biosphere for the necessary period but will also win public acceptance and confidence?
C584/019/2000
©IMechE2000
201
The traditional classification of radioactive waste has, at least to experts in the field, inferred the degree of hazard and thereby implied the isolation period necessary for each category. The isolation period for LLW is say 300 years, ILW about 5000 years and HLW at least 100,000 years. However the newly proposed categories of exempt wastes (EW), very low level waste (VLLW), short-lived low and intermediate level waste (SL-LILW), long-lived low and intermediate level waste (LL-LILW) and high level waste (HLW) should be promoted as the temporal requirements will then be more apparent to the public, although not quantified. Clearly the volumes and isolation periods for each category of waste are known in the industry, illustrating that disposal is indeed a puzzle in four dimensions. The temporal dimension of the disposal is discussed, the dynamism of the process is established and public perceptions considered in 2. DISPOSAL OF RADIOACTIVE WASTE. The perception that waste in a geological repository will re-connect with the biosphere in a relatively short period of time (less than the time required for isolation) is a commonly held belief. Is this due to the public having sufficient knowledge to be able to predict a failure of the multi-barrier system or is there another cause for this belief? This is discussed in 3. TEMPORAL DIMENSIONS. For sometime the disposal industry has toyed with the concept of retrievability of wastes. This is discussed in 4. REVERSIBILITY, RETRIEVABILITY AND REMEDIABILITY. Taking into account the issues considered in this paper a hypothesis is constructed that may assist in resolving the apparent impasse to public acceptance of geological disposal of wastes. This is presented in 5. A GENERAL HYPOTHESIS FOR WASTE DISPOSAL and leads to 6. THE HYPOTHESIS APPLIED TO ILW. The impact of this hypothesis on the science and engineering which must underwrite the safe disposal of radioactive waste is drawn together in 7. CONCLUSION. In presenting this paper to the Institution of Mechanical Engineers in a conference on Radioactive Waste the underlying expertise of the audience is acknowledged however generic background material can be found in the publications referenced in (1), (2), (3) and (4). 2. DISPOSAL OF RADIOACTIVE WASTE The regime for disposal of all waste falls into one of two categories, firstly dilution and dispersion into the biosphere or secondly, containment and isolation from the biosphere as shown in Figure 1. Waste Disposal Concepts. The biosphere is that part of the atmosphere, hydrosphere or lithosphere that supports life. Radioactive wastes are disposed of by both methods with gases and dilute liquids being dispersed into the air, sewer and soils and solid higher-level wastes being conditioned, compacted, contained and isolated from the biosphere probably within a multi-barrier deep geological repository. Radioactive waste exists and awaits disposal in considerable quantities, particularly in countries having nuclear power and a weapons program. It will continue to accumulate irrespective of the future for nuclear power and yet it could be argued that each generation should put in place the regime for the disposal of the wastes that it generates. The author's current research into public attitudes to time, space, risk and trust relevant to the disposal of radioactive waste has highlighted the dislocation between the public's perception
202
C584/019/2000
© IMechE 2000
of forward time horizons and the required isolation period. Siting issues are not pursued here as these have been addressed by others and by the author in papers referenced (5) and (6). There is a need to go back to the fundamentals to develop a waste disposal regime that must be more acceptable to the public and thereby advance the disposal of these wastes. Such a hypothesis needs to take into account that waste disposal is a dynamic process and for public acceptance of the scheme, their perspectives on space, time, risk and trust must be taken into account. Experts in this field believe that the multi-barrier geological disposal system envisaged for the higher level wastes provides not only a retardation of possible contact with the near-field, but would modulate the rate of reaction between the waste and the host material. It would also provide a level of redundancy should one or more of the barriers default. In essence, the system must provide effective barriers to leakage but equally importantly it must provide the time necessary for the decay of radiotoxicity. The barriers must provide the temporal buffer for this dynamic process. The following research will show that people generally accept or reject a disposal method based upon their time perspective and as that is much shorter than the required isolation period, they will usually reject the concept. 3. TEMPORAL DIMENSIONS Extensive polling of UK University students (n=688) and omnibus polling of German speaking Swiss (n=1057) and Japanese (n=2203) has confirmed that a majority of the public believes that geological disposal (at 500 metres) will fail and the wastes will re-connect with the biosphere in less than 1000 years. Question Q8. in the Swiss poll and the results obtained illustrate this point. Q8. If hazardous waste is buried in solid rock 500 metres below the Earth's surface, do you believe that it will be isolated from the living environment for: (a) 100 years 58% (b) 1000 years 19% (c) 10 000 years 5% (d) forever 13% These results are consistent with the UK outcome (not polled in Japan). When compared to the proposals for ILW disposal at 250-500 metres depth and requiring isolation for up to 5000 years it becomes clear that the public do not trust the concept with 58% inferring that the system will fail in less than 1000 years. Further polling results on public perceptions of forward time relative to self, family and community, show that a majority of people have an outer time horizon not further forward than the life on their grandchildren or say 100 years. This work has been fully described in references (5) and (6) and is illustrated by the following questions that were polled in the UK, Switzerland and in Japan. Q5. When considering the future welfare of yourself and your family, how far forward do you think? UK 92%, (Japan 91%, Swiss 87%) selected an outer time horizon of 100 years or less. Q6. When considering the environmental welfare of your home township, how far forward do you think? UK 54% (Japan 64%, Swiss 64%) selected 50 years or less; UK 90% (Japan 89%, Swiss 84%) selected an outer time horizon of 100 years or less.
C584/019/2000
©IMechE2000
203
While the results of question Q8. suggest that the public has an understanding of the concept of geological disposal but still believes that it will fail in a relatively short time, questions Q5. and Q6. provide scope for another interpretation. Perhaps the relatively short expected life of the repository is not necessarily due to a critical understanding of the geotechnical aspects but is a manifestation of personal beliefs that are limited to less than 100 years for some 55-65% of the population and less than 1000 years for some 85-90%. It is the author's view that these beliefs are so entrenched in the population that discussions about isolation for 5000 years, or even more so for 100,000 years, are incomprehensible and therefore will be rejected. From these data, the outer time horizons of the population are clear and in democracies subject to plebiscites, it should be remembered that a simple majority decision is sufficient to decide the outcome of a siting issue. In the case of quantifying the forward horizon in the categories of family and community, 50% of the people have a horizon of less than 50 years! This research has led to the realisation that the time spans necessary for the isolation of radioactive waste are far beyond the time horizons of the 'common man'. This is a very significant finding for the nuclear industry and it may be the underlying factor as to why the responses to question Q8. have a relatively short horizon when compared to the scientific requirements. That is, perhaps the cause is not that there is a knowledgeable belief that the isolating barriers will break down in such a short time (58% in less than 1000 years), but that the natural outer limit for analytical thought of the 'common man' when considering self, family and community, impacts on the response. Glossy brochures depicting smiling operators, mountain streams, tree covered hills and decay diagrams on a log/log basis will not alter public opinion as to the future environmental security of a geological repository. So where does this take industry?
4. REVERSIBILITY, RETRIEVABILITY AND REMEDIABILITY The waste disposal industry has considered the option of retrievability for a decade or so. For example the concept appears in the publications of the Swedish, Finnish and UK industry, but never prominently. The principal concerns of industry are that if the concept of retrievability were adopted then there would be added cost and technical complication whereas the management firmly believes that the currently proposed systems are good enough to meet all contingencies. There is also the possibility that 'retrievability' could imply to the public a lack of confidence in the system. For example, if the wastes can be retrieved is this not tantamount to just another form of interim storage, could the waste be retrieved by 'unauthorised others' and does it not infer that it would be better to leave the wastes on the surface until a truly final disposal system can be developed and accepted by all. The origins of retrievability are probably based on the disposal of HLW in the form of spent nuclear fuel (not reprocessed to extract the reusable uranium and plutonium) where, with a change of circumstance there could be a resource, environment or security reason to recover the material. It was not envisaged that the HLW arising from reprocessing would ever need to be retrieved but as the UK now needs to consider the disposal of surplus plutonium as a waste, it brings the prospect of retrievability into focus. Likewise with ILW, whether shortlived or long-lived, there was no foreseeable need to ever retrieve these wastes. The need to reassess the concept of retrievability has arisen from the results of this research, which has identified the relatively short time horizons of the public and their distrust of long term geological disposal.
204
C584/019/2000
© IMechE 2000
Reversibility is a term now being used to describe the option of recovering ILW and HLW from repositories before they are closed. The possible needs to reverse the placement while the repository remains open could arise from technical, environmental, health or resource (in the case of spent fuel and plutonium) reasons. Reversal prior to closure, whilst unlikely, is technically more simple than retrieval after the repository is closed although it has a cost and once again puts the wastes back onto the surface. However if there is a just cause then surely reversal is a preferred action when compared to retrieval after closure. ILW and HLW systems incorporate multiple barriers such as the primary form of the waste (metallic, glass or ceramic), metal container, geological overpack (modified clays), near-field strata (salt, clay, rock) and distance from the biosphere. In most systems there is an in-built redundancy such that if a barrier fails prematurely, other barriers will compensate. One concept not yet fully explored is the issue of remediation, that is, a form of geotechnical correction if there is a premature failure of the barriers. Industry could be loath to discuss this option as it, like reversibility and retrievability, might infer from the outset that the system could be flawed. It could also infer that if there is a premature failure then industry proposes a set of band-aids that hopefully will fix the problem: a poor image. However we should consider for example the environmental corrections taking place in and around uranium mines and processing plants in the Former Soviet Union, particularly in the former East Germany and Czechoslovakia where there has been significant 'remediation'. Mine effluents have been cleaned and are now safely allowed to flow into the river systems, underground acidity is being corrected and surface tailings consolidated and covered with clean material. Perhaps we should think in terms of what would we do today faced with a radioactive leak from an underground repository. If such a leak is water born then technologies such as pressure grouting of incoming aquifers and the repository, bypass hydraulics, water quality modification, retrievability of source, down stream environmental and health monitoring could all be applicable. If the contaminants are gas born, then a system of management also needs to be prescribed. Perhaps this could be based upon underground collection manifolds, filtering, conditioning of gases, dilution and release. Any recovered gas born particles should be treated as for any other solid waste. There are geo-technical systems available for the remediation of closed coal, salt and metal mines. The proposed sequestration of carbon dioxide and other gases also illustrates some of the technologies available for remediation. There is probably a strongly held public view that all manmade artefacts will need repair sooner or later, therefore why not prescribe that from the outset. It can be shown that with today's technology remediation of a premature failure in a repository can be addressed. This is not dependent upon new and as yet to be invented technologies but of course there will be evolution in these geo-technical areas. Would the public fear less the prospect of remediation if required than they do of irreversible and irreparable placement?
5. A GENERAL HYPOTHESIS FOR WASTE DISPOSAL It has been established that: • There is a clear discontinuity between the time to the forward horizon for the public and the temporal needs for the isolation of radioactive wastes.
C584/019/2000
©IMechE2000
205
•
• •
Waste disposal is a dynamic process that incorporates chemical and physical change, allowance for exhaustion of the primary containment, probable interaction with the nearfield and finally the concept that buried wastes will alter and become part of the earth. The issue of radioactive waste disposal needs to be resolved, and Reversibility, retrievability and remediability are candidate ancillary concepts.
It falls to scientists and engineers to take these parameters into consideration when designing a viable system for the disposal of radioactive waste and to win the confidence of the public. Schemes that are technically valid but cannot engender public confidence will become the deLoreans of this industry: expensive, glitzy, technically smart but unsaleable. To stimulate discussion on the evolution of an acceptable system, a general hypothesis is now proposed as shown in Figure 2. The General Hypothesis for the Geological Disposal of Waste: the Hazard-Time Relationship, where the relative hazard (to the biosphere) of a waste is plotted against time. Line 'A-E' represents the Hazard-Time (H-T) Relationship. The yaxis is divided about a line above which there is increasing hazard and below which it is proposed that there is no discernible hazard. 'A' represents the opening of the repository to receive wastes and period 'A-B' on the H-T line corresponds to the period that a repository receives waste. 'B' represents the closure of the repository, 'C' the limit of the design life of the primary containment and 'D' the minimum lifetime of the artificial and natural barriers. After 'D' it is assumed that the waste has become an inert part of the earth and constitutes no risk to the biosphere. 'E' represents the full absorption of the waste into the earth. In the event that there is an unlikely premature failure of the system, then the suggested remedies are: • ' A-B': reversibility with recovery of the waste in its primary container. • 'B-C': retrievability of the waste in its primary container or remediation • 'C-D': remediation • 'D-E': no action required as the waste is now altered and possibly absorbed into the substrate materials. This model hopes to demonstrate a technical resolution for geological disposal that engenders public acceptance. For all future generations there is a prescribed remedy to any possible default of the system during the designated isolation period. Additionally there would be a scientific and engineering certainty that the risk of a fatality in the population would never exceed 1x10"6 p. a. although this is not an expression that inspires confidence in the public. Reversibility, retrievability and remediability however may give confidence to all concerned. 6. THE HYPOTHESIS APPLIED TO ILW Figure 3. The Hazard-Time Relationship for the Geological Disposal of ILW is based on the general hypothesis but ascribes time values to the x-axis. The time value for each point is: 'A' = present, 'B' = 100 years, 'C' = 1000 years and 'D' = 5000 years although these are indicative only and can be adjusted to suit a specific classification of ILW. It follows that for HLW, the values would be in the order of A = present, B = 100 years, C = 3-5000 years and D = 100,000 years. Conceptionally, the case for public acceptance of a remedial system to contain ILW for up to 5000 years seems achievable. Public acceptance of a system to isolate
206
C584/019/2000
© IMechE 2000
HLW for up to 100,000 years is more problematic but is perhaps achievable with increases in geological isolation and in the design life of the primary containment. 7. CONCLUSION This paper establishes that the disposal of radioactive waste is a dynamic process encompassed in the four dimensions of space and time. It demonstrates that public perception of future time horizons is significantly shorter than the periods required for the isolation of radioactive waste. A General Hypothesis has been developed that should give the public a greater confidence in geological disposal systems. This hypothesis is not based on the traditional stated objectives that the repository will not fail or if it does then the consequences to the biosphere will be minimal, but rather that should it fail, then the placement is reversible, the wastes retrievable or the system remediable. It is hoped that these views will stimulate a fresh approach to the development of disposal systems taking into account public perceptions of space, time, risk and trust. Without public acceptance, the most elite of technical systems will not be adopted.
ACKNOWLEDGEMENTS The author's doctoral study at the School of Geography, University of Oxford is supervised by Professor Gordon Clark, Halford Mackinder Professor of Geography and is materially supported by British Nuclear Fuels plc, Tokyo Electric Power Company, Steering Committee on High-Level-Radioactive-Waste Project, Japan and Nationale Genossenschaft for die Lagerung radioaktiver Abfalle (NAGRA), Switzerland. Comments on drafts of this paper by Emma Cornish, Gerald Clark, Brian Eyre and Chris Ealing are greatly appreciated.
REFERENCES (1) Parliamentary Office of Science and Technology (November 1997), Radioactive Waste Where Next? London. (2) The House of Lords, Session 1997-98. Select Committee on Science and Technology, Management of Nuclear Waste: Written Evidence, (April 1998). London: The Stationery Office. (3) The House of Lords, Session 1998-99. Select Committee on Science and Technology, 3"" Report: Management of Nuclear Waste, (March 1999). London: The Stationery Office. (4) National Radiological Protection Board, (1989) Living With Radiation. Didcot, Oxon. (5) Duncan, I. J. 1999a. A Community that Accepts Risk Should be Rewarded. Decision Risk and Policy 4 (3): 191-199. (6) Duncan, I. J. 1999b. Some Aspects of the Relationship between Society and the Disposal of Radioactive Waste. Paper read at The Uranium Institute: Twenty-Fourth Annual Symposium, September 1999, at London.
C584/019/2000
© IMechE 2000
207
Figure 1. Waste Disposal Concepts
Figure 2. The General Hypothesis for the Geological Disposal of Waste: the Hazard - Time Relationship
Figure 3. The Hazard - Time Relationship for the Geological Disposal of lLW
Authors' Index B Balkey, J J Barlow, S V Borisov, V V Bradbury, D
N 57-66 105-116 183-190 191-200
P
Cave, L
Nakajima, M Netecha, M E Newstead, S
85-94 43-46, 183-190 117-128
o Orlov, Yu V
43-46, 183-190
141-150 P
Palmer, J D Daish, S R Duncan, I J
117-128 201-210
E
Edler, G R Ellis, AT
191-200 73-82
105-116
S Seddon, W Shishkin, V A Stanislavski, G A
95-104 43-46, 67-72 183-190
T
G Gabaraev, B A Green, T H
Thomson, P F G 67-72 161-170
H
Harrison, M Hey,R I
21-30 73-82
161-170 161-170
L Langley, KF Leech, N A
47-56 117-128
M Mazokin, V A McCracken, G McTagget, L Miller, T J
V Van Velzen, L P M Vasiliev, G A Vasiljev, A P
129-140 183-190 43-46
W
K
Kornyeyev, A A Krinitsyn, A P
3-20
43-46, 67-72, 183-190 31-40 73-82 153-160
Waker, C H Wieneke, R E Wilding, C R Williams, J Wood, C J
171-182 57-66 161-170 47-56 191-200