150 pages
English

A review of the aqueous chemistry and partitioning of inorganic iodine under LSW severe accident conditions

-

Obtenez un accès à la bibliothèque pour le consulter en ligne
En savoir plus

Description

Industrial research and development

Sujets

Informations

Publié par
Nombre de lectures 10
Langue English
Poids de l'ouvrage 6 Mo

ISSN: 0379-4229
CODEN: EARRDF 6(4) 631-1002 (1985)
REPRINTED FROM
VOLUME 6
NUMBER 4
1985 EUROPEAN
APPLIED RESE ARCH
REPORTS
A Journal off European Science
and Technology
NUCLEAR SCIENCE AND TECHNOLOGY SECTION
Published for the Commission of the European Communities,
Directorate-General Information Market and Innovation
A REVIEW OF THE AQUEOUS CHEMISTRY AND
PARTITIONING OF INORGANIC IODINE UNDER LWR
SEVERE ACCIDENT CONDITIONS
By P. N. Clough and H. C. Starkie (EUR 9408 EN)
G + B/harwood European Appi. Res. Rep t.-Nucí. Sci. Technol.
Vol. 6, No. 4 (1985), pp. 631-776
0379-4229/85/0604-0631 $12.00
Printed in Great Britain
A REVIEW OF THE AQUEOUS CHEMISTRY AND PARTITIONING OF
INORGANIC IODINE UNDER LWR SEVERE ACCIDENT CONDITIONS
P. N. CLOUGH and H. C. STARKIE
Safety and Reliability Directorate,
UKAEA,
Culcheth,
United Kingdom
© This report was prepared for the Commission of the European
Communities under Contract ECI.987.B7210.83.UK.
EUR 9408 EN
ECSC, EEC, EAEC, Brussels and Luxembourg, 1985 632
SUMMARY
Fission product iodine released from the core of an LWR into the reactor building
during a severe accident will become distributed between water pools, where it is
largely immobilised, and the gas phase, from which it may readily escape to the
environment· Iodine in aqueous solution can undergo a complex series of oxidation
and reduction reactions involving many molecular species, only a few of which are
capable of partitioning into the gas phase· This review is concerned with assessing
the chemistry and partitioning behaviour of inorganic iodine relevant to severe LWR
accidents-
Iodine chemical behaviour is solution is sensitive to water chemistry conditions,
particularly pH and temperature, and also to the presence of radiation fields· The
likely conditions in the reactor building are first identified· A detailed analysis
of the redox chemistry of iodine is then given, and pertinent equilibrium and kinetic
data are reviewed with particular emphasis on data uncertainties. Sensitivity
studies are performed of the dependence of the total partitioning of iodine on
uncertainties in individual data, concentrating on temperatures > 100"C, with a view
to identifying key data on which improved experimental measurements appear to be
needed. Finally, recommendations are made of priorities for future work on aqueous
iodine chemistry. 633
CONTENTS
Page
1 INTRODUCTION 634
2 WATER CHEMISTRY CONDITIONS IN REACTOR ACCIDENTS 639
2.1 PUR
2.2 BWR 645
2.3 Effects of Extraneous Materials 647
3 CHEMISTRY OF THE AQUEOUS IODINE SYSTEM 650
3.1 Background 65
3.2 Equilibrium and Kinetics of the Reactions of Iodine Species
in Aqueous Solution 66
3.3 Partitioning of Iodine Species 686
3.4 Effects of Radiolysis on Aqueous Iodine Chemistry 69
3.5 Organic Iodide Formation 712
4 SENSITIVITY STUDIES OF THE UNCERTAINTIES IN THERMODYNAMICS
AND KINETIC DATA 721
4.1 Roles of Thermodynamics and Kinetics
4.2 The IOKIN Code 730
4.3 Selection of Water Chemistry Conditions and Reaction Rate
Constants used in the IOKIN Runs3
4.4 Results and Discussion of Sensitivity Studies 738
5 DISCUSSION AND RECOMMENDATIONS 75
APPENDIX I Summary of Aqueous Iodine Chemistry Research Known
to be in Progress throughout the World 774 634
1 INTRODUCTION
Radioiodine has long been identified as a major contributor to the health effects and
environmental damage assessed to be likely in consequence of postulated severe
accidents in nuclear reactors (1). The radio-isotopes from 1131 to 1135 inclusive
which are produced with appreciable yields in the thermal fission process are all of
importance with respect to the early post-accident hazard. Of main concern, because
of their longer half-lives, are 1131 (t. - 8.04 d) and 1133 (t. = 0.866 d). In
common with the other halogen elements, iodine is very chemically reactive, and this
property enhances the radiological significance of accidental releases since the
element is readily absorbed by the human body, and by vegetation. However, in
compensation for this adverse effect, the high reactivity increases the possibility
that iodine will be trapped and retained along its transport path from the reactor
core before it reaches the environment. In the case of water reactors, an important
process which can play a large part in such trapping is solution in water which is
accidentally released in the course of the accident or intentionally introduced for
accident control purposes. The objective of this report is to review and assess the
fundamental chemistry of the iodine-water system in the context of iodine retention
in LWR accident situations.
Historically, iodine has probably been the most studied of all fission products in
nuclear reactor safety assessment. Because of the high volatility of the elemental
form, a simple and conservative assumption that the release behaviour would resemble
that of a gas was adopted early on in the development of safety analysis. Analogies
were made with the noble gas fission products. The chief evidence for this idea came
from the 1957 Windscale accident (2), where a fire in the graphite moderator of an
air-cooled, metallic fueled, nuclear pile led to the release of some 20,000-30,000 Ci
of 1131 to the environment. There is no doubt that in this oxidizing situation, the 635
release form of iodine was predominantly the elemental vapour. The Windscale
evidence was influential in the framing of US Regulatory Guides 1.3 and 1.4 for
consequence assessment of LWR loss-of-coolant accidents (3), which formalise the view
that 85% of the released iodine will be in elemental form. The remainder is
stipulated to be treated as 10% organic and 57, particulate-borne. The emphasis on
the elemental form was reinforced in the conclusions of the USNRC Reactor Safety
Study, WASH-1400 (4). A detailed review of the evidence available up to the mid-
1970s on release rates and chemical interactions of fission products in LWR core-melt
accident situations was presented in Appendix VII of this study. The possibility
that iodine behaviour might be influenced by reaction with caesium and other metallic
fission products within the fuel matrix to form iodides was specifically considered
in this review (5). However, rather because of a lack of conclusive evidence for
iodide formation than due to any contrary evidence, the authors felt it necessary on
grounds of conservatism to adhere to the view that free elemental iodine should be
considered the dominant release form from the molten core. This assumption had
strong repercussions in the analysis of the subsequent transport of iodine through
the reactor buildings and the role of iodine-water interactions in retention. In
particular, the main effects of spray removal in FWR containments and of pool-
scrubbing attenuation in BWR wetwells were treated in terms of the dissolution of
elemental iodine vapour in aqueous solutions, with a minor role attributed to organic
iodides.
In the aftermath of the accident at Three Mile Island, Unit 2 (TMI-2) in 1979, papers
appeared by Levenson and Rahn (6) and by Campbell, Malinauskas and Stratton (7)
reopening the question of iodine chemical behaviour in LWR accidents. The insignifi­
cant releases of radio-iodine to the environment in this accident demonstrated the
extremely effective role of water in bringing about retention. Both of the cited 636
papers reviewed evidence from past reactor incidents and laboratory experiments
pointing to fission product iodine being released in core melt accidents as metal
iodides, predominantly as caesium iodide. The high solubility of Csl would certainly
favour the very high retention observed at TMI-2, although it only provides indirect
evidence on the matter. New and more direct experimental evidence on the chemical
form of iodine released from overheated LWR fuel was becoming available about the
same time from the experiments of Lorenz and co-workers at Oak Ridge National
Laboratory (8). This and all of the other information up to early in 1981 was
thoroughly reviewed and assessed in the USNRC report, NUREG-0772, 'Technical Bases
for Estimating Fission Product Behaviour During LWR Accidents' (9). A major
conclusion of this exercise was that 'The current data base suggests that caesium
iodide will be the expected predominant iodine chemical form under most postulated
light water reactor accident conditions'. However, the qualifying rider is added
that 'The formation of some more volatile iodine species (eg elemental iodine and
organic iodines) cannot be precluded under certain accident conditions'. A starting
assumption of the present report will be that iodine released from fuel in accident
conditions enters the aqueous phase initially as iodide ion. As such, it is
essentially involatile in the temperature range of interest here, and its transport
properties can be taken as entirely associated with the aqueous phase. However,
oxidation can generate more volatile iodine species which will be partially released
from solution into the gas phase. Assessment of the mechanisms and extents of
formation of these species, and their likely importance in specific accident
circumstances, forms the bulk of this report.
There exists a long history of both experimental investigation and theoretical
analysis of aqueous iodine chemistry, and a substantial contribution to this has been
made in the context of nuclear safety analysis. An important early contribution was 637
made by Eggleton (10) in his modelling analysis of the solution species and overall
partitioning properties for solutions of elemental iodine. This model was re­
examined and extended by Parsly (11), with particular reference to the effectiveness
of spray removal of iodine from the atmospheres of reactor containment buildings
Parsly's work provided much of the basis of the treatment of iodine in the Reactor
Safety Study (4). The recent resurgence of interest in the subject has resulted in a
number of detailed reviews. The US report NUREG-0772 (9) includes an account of
aqueous iodine chemistry, and this has been expanded upon in two subsequent ORNL
reports. The first of these (12) is concerned mainly with assessing the equilibrium
aspects of reactions in solution and partitioning into the gas phase, whilst the
second (13) deals with reaction kinetics. An independent examination of the
equilibrium thermodynamic data base and some aspects of the reaction kinetics has
been made by Canadian researchers at Whiteshell (14). Vinson (15) has also produced
a review of iodine solution chemistry with emphasis on reaction mechanisms. In the
UK, a comprehensive attempt has been made to apply results of these modelling and
data analyses to a range of design basis accidents for a modern PWR (16,17). This
has highlighted the large uncertainties which result from present inadequacies in the
available information.
In view of this extensive recent activity, the present work must have a number of
aims which attempt to build upon and develop the earlier studies if it is to be of
real value. As a first step, we examine the range of conditions which can arise in
LWR severe accidents up to and including full core melt, in order to define all the
possible influences which must be taken into acount in analysing iodine chemical
behaviour. The fundamental chemistry of the iodine-water system is then reviewed
with a view to establishing uncertainly limits in the relevant thermodynamic and
kinetic data, and the impact of these uncertainties on the prediction of iodine 638
transport behaviour. As an aid to this work, a computer code has been written in
order to perform sensitivity studies directed to identifying the chief contributors
to overall uncertainty. Temperatures in the vicinity of 100°C, which are prevalent
in reactor containments during the critical stages of many LWR accident scenarios,
are of particular concern. The situation with respect to new experimental data has
developed rapidly within the past two years, and an assessment of the impact of these
new results is included in our review. Finally, in the light of this analysis,
proposals are made as to the most valuable areas for further experimental work
designed to produce improved data.