Towards an Understanding of Zirconium Alloy Corrosion

v V..

AECL-5548

ATOMIC ENERGY L'ENERGIE ATOMIQUE OF CANADA LIMITED DU CANADA LIMITEE

TOWARDS AN UNDERSTANDING OF ZIRCONIUM ALLOY CORROSION

by B. COX

Presented on the occasion of the award of the William J, Kroll Medal, Quebec City, P.Q. August 1976

Chalk River Nuclear Laboratories Chalk River, Ontario August 1976 Cover Photograph: Large columnar oxide grains on zirconium oxidised in air at 650°C, magnification x20,000. TOWARDS AW UNDERSTANDING OF ZIRCONIUM ALLOY CORROSION

B. Cox (M.A., Ph.D., Cantab.) Head of Materials Science Branch Atomic Energy of Canada Limited Chalk River Nuclear Laboratories Chalk River, Ontario KOJ 1J0

on the occaiZon o & the

William J. Kroll Medal Presentation Quebec City, P.Q. August 11, 19 76

AECL-5548 ^^mjrrejid_re_lji corrosion des alliages de zi rconium

par

B. Cox

à l'occasion de la présentation de la Médaille William J. Krol 1 Québec, P.Q. 11 août 1976

Résumé

On donne un bref historique du développement d'un programme visant

à mieux comprendre les mécanismes de corrosion qui jouent dans les alliages de zirconium. Un sommaire général indique les progrès réalisés jusqu'à présent dans la mise en oeuvre de ce programme.

L'Energie Atomique du Canada, Limitée Laboratoires Nucléaires de Chalk River Chalk River, Ontario

AoQt 1976

AECL-5548 TOWARDS AN UNDERSTANDING OF ZIRCONIUM ALLOY CORROSION

B . Cox

on the occasion of th(. William J. Kroll Medal Presentation Quebec City, P.Q. August 11, 1976

ABSTRACT

A brief historical summary is given of the development of a programme for understanding the corrosion mechanisms operating for zirconium alloys. A general summary is given of the progress made, so far, in carrying through this programme.

Chalk River Nuclear Laboratories Chalk River, Ontario August 19 76

AECL-5548 1.

I feel myself greatly honoured to be the second recipient of the William J. Kroll Medal for Zirconium, and very humble at being considered worthy to follow so major a personality in the nuclear field as Admiral H.G. Rickover USN. I was a relative late- comer to the zirconium alloy field, having joined the U.K. Atomic Energy Authority's research establishment at Harwell only in September 19 55. When I arrived the first Geneva Conference on the Peaceful Uses of Atomic Energy had already taken place the previous month, and one of my first duties on arrival at Harwell was to search through a stack of preprints of the conference papers for the ones relevant to the area I had been assigned to work in - zirconium alloy corrosion.

For a new Ph.D. from Cambridge, whose work had been mainly in the area of structural inorganic chemistry, this was quite a change in field, and I eagerly sought a copy of the new bib]e of the zirconium aficionado (The Metallurgy of Zirconium, edited by Lustman and Kerze), which had been published simultaneously with the Geneva Conference. At that time, copies were rare and prized possessions, and it is a token of the amount and quality of information that had already been generated on zirconium alloys that it remains even now the basic reference book of the zirconium field. Considering the picture as it appeared at that time, I miaht have wondered how I could carve out a niche for myself in this field. The alloy Zircaloy-2 (still the "reference alloy" for the industry against which others are compared) was already developed and in use. The first U.S. nuclear submarine, the Nautilus, was already at sea; what more was there still to do in this field? I conclude that it has been the approach I have adopted to the study of corrosion mechanisms, and more recently to the mechanistic aspects of other areas of the physical metallurgy of zirconium, which has lead to this award. I hope my listeners will find it informative, therefore, if I delve firstly into those factors which have conditioned me to this approach, and then summa- rize briefly where this approach has led me. Undoubtedly one of the primary influences was that of my supervisor at Cambridge, Dr. A.G. Sharpe, who, himself something of an iconoclast, insisted on cultivating a healthy scepticism towards currently assumed theories in all who worked for him. I think such a questioning attitude to received information is a prerequisite for any scientist hoping to solve complicated mechanistic problems, and is epitomised for me by the English north country expression "believe nowt ye read, and only 'alf ye see".

The second major influence on the subsequent direction of my studies was to have been assigned to the U.K. Homogeneous Aqueous Reactor (HAR) Project, rather than to the parallel light water reactor project (LEO). In the parallel U.S. Homogeneous Reactor Project enormous increases in the corrosion rate of zirconium alloys 2.

(thousands of times the laboratory rate) were being observed under irradiation conditions. This large irradiation-induced increase in corrosion showed that all was not ideal with the behaviour of zirconium alloys in all situations. Thus, there were some major factors affecting the corrosion process which could not be ignored and needed to be understood. It was many years before workers in the light water reactor area became convinced that irradiation universally affected the corrosion rate of zirconium alloys, and that it was merely- the chemistry of the aqueous environment which determined whether these effects were so large as to kill a reactor project or so small that they required many years of observation to reveal their presence.

The third major factor determining my course of action has been my personal predilection for developing techniques to look at specific aspects of an overall problem, rather than the more commonly adopted approach of specialising in one, or a few related, techniques, and then seeing what problems can be tackled using these techniques. This line of attack has enabled me to pursue a predetermined attack on the remaining mechanistic problems in the corrosion field, at least over the last twelve years or so, by first identifying a sequence of questions needing to be answered if the process was to be understood and then trying to find, or develop, techniques capable of answering these questions; an example is given in Table 1.

The detailed development of this attack has been summarised in the reviews which I have written recently (1,2). I do not propose to try to do a further such review here, but I will restrict myself to a more anecdotal approach, covering first the early years, where I was in the process of deciding what the questions were that really needed asking; and will then give a fairly general summary of the later development of the programme. In learning what were the important questions there were inevitably many false starts. However, perhaps we learn most from our mistakes, so I hope you will bear with me if; in the next few minutes, I seem to be relating a series of failures rather than successes.

Again serendipity played its part in the early days, both as a result of the specific task I was set on arrival at Harwell - to develop a technique for in situ measurement of the corrosion rate of zirconium alloys in high temperature and high pressure (irradiated) solutions - and of the commercial zirconium alloy materials with which we had to work at that time in the U.K. In attacking the main task, the availability of large laboratory autoclaves led us to study techniques for monitoring the corrosion rate of individual specimens within a large autoclave, rather than adopt the ORNL approach of monitoring the corrosion rate of a small autoclave itself (by following the decrease with time ir the oxygen over-pressure). Thus we were led into the field of electrical and electrochemical corrosion monitoring techniques, and the vagaries of high-pressure electrical seals. The techniques we used were of two types, firstly measurements of the polarisation curves of zirconium alloy electrodes, using early models of potentiostat, which were already available (constructed by the Harwell electronics division). Ultimately, this type of d.c. electrochemical measurement was abandoned after we succeeded in repeatedly and unexplainedly electrodepositing gold (from the counter-electrode) on the specimens. Gold, rather than platinum, was used at this time because of its supposed greater stability in these solutions. Thus, it was obvious that the polarisation currents we were measuring contained sizeable components representing electrochemical dissolution and deposition of supposedly inert parts of the system.

Our attention was next directed towards a.c. methods/ especially measurements of oxide film capacitance, for continuous monitoring of the corrosion process. We started off using a commercial capacitance bridge, and using this we were soon able to demonstrate the major effect which the rectification phenomenon, discovered by John Wanklyn, had on the specimen corrosion rate. Using this technique, all zirconium alloy specimens were soon covered with thick white oxide films which indicated that rectification of the a.c. current had caused a large increase in the amount of corrosion occurring. We were not the last investigators to fall into this trap. We changed to a homemade bridge using a transformer to reduce the a.c. applied voltage, and also studied the use of the square wave polarograph (an instrument then having only recently been developed by George Barker). It was possible to derive capacitance figures using this instrument and we thought that the small square wave pulses it used might cause less of a problem than the continuous a.c. applied by the capacitance bridge. In fact, both instruments seemed to give reasonable plots of 1/C versus thickness when applied to anodised zirconium at room temperature, and we got similar apparently convincing plots versus time during autoclave tests at ?50°C.

I was on the verge of publishing this work, when I began to have doubts after studying the capacitance behaviour of the system during autoclave cooldown - the capacitance of the samples apparently increased (i.e. film thickness apparently decreased). This led me to study in more detail than previously, the double layer capacitance on the gold counter-electrode. We had established its variation with temperature, but had assumed that at any given temperature, it would remain roughly constant with time - it didn't. So we had again been measuring some peculiarities of the electrochemistry of the counter-electrode rather than the specimen. These experiments had led to no usable results, but they did convince me of the importance of the electrical properties of the oxide in the oxidation mechanism, and disillusioned me suffi- ciently about the interpretation of high temperature aqueous electrochemistry that I subsequently decided, when I returned to studies in this area, to use a fused salt as the electrolyte , an approach which was to prove far more fruitful. 4.

The materials we had to work with at that time also affected my approach to the subject. Although we had some arc-melted Zircaloy-2 manufactured by I.C.I. Metals Division (as it was then), most of it was still melted in graphite crucibles The zirconium carbide particles in such alloys were obviously playing a major part in the corrosion process, as could be seen from an optical metallographic examination of specimens before and after corrosion. However, I was unconvinced that this was simply a case of prefeiren- tial attack on the carbide particle: itself, as was commonly assumed. In order to examine the process in more detail, I was led into the field of electron microscopy, and particularly the interpretation of electron microscope replicas, which had not up to that time been used extensively in the study of corrosion processes.

This foray into the field of electron microscopy opened up a new world to me. I was able to see that the corrosion attack did not start at the carbide particle, but in the matrix adjacent to the matrix-particle interface. Many other interesting things also appeared, such as localised oxidation along what was originally a metal grain or twin boundary. The formation of large pores in the oxide at prior metal grain boundaries in pre-oxidised specimens bombarded with fission fragments and subsequently re-oxidised was a phenomenon observed, but still not fully understood - why were these pores not randomly distributed? This early work on the application of the electron microscope led to a number of papers, for one of which I was proud to receive in 1961 the A.B. Campbell Award from NACE. It also developed my continuing interest in the application of all types of electron optical techniques to corrosion studies , an interest which is still evident in the programme of the Materials Science Branch at Chalk River.

A third area of techniques, which I entered in the early days to see whether or not they could contribute to our understanding of zirconium corrosion, was that which can be broadly classified under the heading autoradiographic. These comprised not only the true autoradiographic techniques using radioactive isotopes of the species concerned (of the ones we tried at that time, iron, chromium, nickel and tritium,only the last led to any success), but also techniques such as "Bitter Figures" which depend on imaging magnetic domain boundaries, and by which we tried unsuccessfully to locate iron-containing second-phase particles. However, a continuing interest in this area ensued, and following my move to Atomic Energy of Canada Limited at Chalk River Nuclear Laboratories in 1963, I was able to pursue the tritium autoradio- graphy further and also to extend my interest into the area of nuclear reactions for identifying the location of particular species (there being already a strong group at CRNL working in this area).

Thus, by the time I left AERE I had already developed my interests in the three areas of experimentation (electrical properties of oxides, electron microscopy, autoradiographic techniques) which were to form the base from which I could attempt to •. :-.>-_ai!i the oxiuatioa mechanism. Other experiments at AERE on the corrosion properties of numbers of different alloys, and in the reactors there provided the background from which to decide those questions which played an important part in the corrosion mechanism. Soon after arriving at CRNL these ideas were formalized into a list of questions to be answered. While this list has seen some modification (and some answers inserted in it) it still remains much the same (Table 1). Similar sets of questions could be listed for other phenomena, such as hydrogen absorption during oxidation, or stress corrosion cracking. But for illustrative purposes the one related to oxidation mechanism will be used.

With Table 1 as the basis for discussion, I will now try to summarise the progress that has been made in elucidating the oxidation mechanism of zirconium alloys. Strictly speaking, a separate set of answers should be prepared for each alloy, but for a general discussion it will probably be sufficient to orient the answers towards the Zircaloy type of alloy (i.e. one containing iron, chromium, and nickel second phase particles). For a more detailed exposition of the problem,readers should consult references 1 and 2. In order to decide whether the oxidation process is controlled by diffusion through a protective oxide film or not, it is seldom sufficient to rely alone on the kinetic form of the oxidation curve. There will always be the possibility of misinterpretation since, for instance, parabolic kinetics imply only that the oxidation rate is inversely proportional to the oxide thickness. Such kinetics can be achieved equally by diffusion through a barrier oxide, or by gaseous flow through an increasing thickness of porous oxide. A technique which can distinguish between these possibili- ties is to instantaneously change the environmental gas pressure during a continuous microbalance experiment, and look for instanta- neous changes in oxidation rate. This technique proved very useful in identifying the onset of porosity during the oxidation of Zircaloy--2 with the onset of the kinetic transition, and showed that this porosity extended essentially up to the oxide-metal interface. Prior to the rate transition in the oxidation kinetics therefore, the transport process consists of diffusion through a barrier film, except for the influence of local cracking at any sharp edges on the specimen.

In the pre-transition period of the oxidation kinetics we have to be concerned with both ionic and electronic transport through the oxide, since the net charge transfer must be zero, and we cannot assume that any species migrates with such ease that it can be ignored as a factor in the overall transport process. The evidence available on the ionic transport process suggests that oxygen is by far the most mobile species, while, as we will see later, the defect structure of the oxide lattice is of relatively little concern for oxidation at normal reactor temperatures, because of the microcrystalline nature of the oxide film. 6.

By using a nuclear reaction of oxygen (0!7 + i!e] -*• 0: ' + Hesj we were able to show that the oxygen diffusion process during oxidation was essentially a line diffusion process, and that the bulk diffusion of oxygen in the oxide was several orders of magnitude too slow to account for the amount of oxidation which occurred. Incidentally, I have been criticised by other experimental ists ir. the field of applications of nuclear reactions, for employing a rare and expensive isotope of oxygen in a reaction that required a large and even more expensive accelerator. However, as I have already said I regard techniques as means to an end, and in chis instance we were approached by physicists at Chalk River to make some stable 017 targets for their studies cf transition states in the neon nucleus. This coincided with our interest in using a nuclear reaction to study oxygen migration. So by a happy conso- nance of aims we agreed to provide targets for their study, if thc-y in turn would do some measurements en our diffusion specimens. The- narrow bulk diffusion profile at the specimen surface proved too thin for accurate study by this technique, but usinc the same targets Paul Pemsler and I were able to measure it successfully, using an ion bombardment mass spectrometer.

Thus, the deduction from this work is thjt the line diffusion process represents diffusion along crystalline boundaries in the oxide, and that this is the primary route for oxygen migration during oxidation, diffusion through the bulk being too slow by several orders of magnitude. Transmission electron microscopy has been used extensively to study the development and epitaxy of crystallites as the oxide thickens, but the limited transparency of the oxide to electrons at up to 200 keV, has meant that a different technique had to be used to study the crystallite morphology near to the kinetic transition. It was found that a technique for rapid fracturing of thick oxide films, coupled with replication of the fracture surface, gave very good impressions of the large crystallites which developed in thick oxide films. By the technique we were able to observe in unalloyed zirconium (which does not show a kinetic transition in 600°C oxygen at film thickness up to 10 microns) the large columnar crystallites formed, whereas in Zircaloy--2 the crystal growth process was interrupted, so that (at the transition?) renucleation of small crystallites occurred. This process of growth and renucleation appeared to continue repetitively.

Thus the ionic transport process is apparently one in which oxvgen ions diffuse along crystallite boundaries in the oxide, the precise kinetics being determined by the manner in which the crystallites change size in any particular oxide film.

To study the electron transport process we needed a method of forming an electrical contact with the outer surface of the oxide which would satisfy the following criteria:

1. It should not affect the oxidation rate L'f the surface beneath the contact. 2. It shoulei contact the whole specimen area, or a known fraction or that area so that the current density could bo measured.

3. It should be as near as possible to an ohmic reversible contact" Fn. both ionic a^d electronic processes.

-1 . y.c add i t i on 31 contributions to the measured current (i.e. fror proccssos other than specimen oxidation) should be present:.

Various J orms of metallic or porous conducting oxide were el i r.'imtod, nn.i our previous experience with hiqh temperature 3qi;eoi:Fi e] ectrnciiemi stry suggested that it did not meet some of the above conditions. We finally settled on a conducting, oxirfisinr fused salt as the best solution to the problem.

"y using this technique to measure the current-voltage (I-V) characteristics of growing oxide films, and by separating these curves '. ;;to their ionic and electronic components we have been able to show that the electronic transport process is quite variable, although it can often be approximated to a Schottky emission process. It appears that the electron transport process may differ at different locations ovi the specimen surface with the measured charactertistic being a weighted mean of the processes contributing over the whole a r e a .

rsoth electron and hole conduction have been identified under different conditions, again showing the variability of the electron conduction processes which occur. Using an imaging technique for the electron curie.*t it was possible to show that, at least at room temperature, conduction was mainly localised at intermetallic particles in Zircaloy type alloys. It is inferred that this is also the main route for electron conduction during oxidation. From observations in a number of oxidising media of differing conductivity, and from direct measurement we have also been able to deduce that surface resistivity is a major part of the overall electronic resistance, especially for unalloyed zirconium.

Putting this evidence together we find that in zirconium and the Zircaloys the electronic component of the oxidation current flows primarily at a few sites, commonly identified with the intermetallic particles. Since the reduction of oxygen and its diffusion through the oxide are occurring more uniformly over the surface (because of the small crystallite size) the electrical circuit must be closed by surface conduction. The oxidation field across the oxide is set by that voltage needed to equalize the two components of the oxidation current. In practice, because of the steeply rising nature of the electronic current as a function of increasing voltage, this oxidation field is determined largely by the electronic conduction. The ionic flux is controlled by this field, and hence the oxidation rate is fixed, within limits set by the oxygen diffusion process, ,jy tne electronic properties of che oxide. In such ^ situation neither process can be said to be uniquely controlling the oxidation. Thus a change in conditions which causes a large change in only one of the two components of the I-V curve will not result in a big change in oxidation rate. Both processes must be changed significantly before any large changes in oxidation rate can be observed. This argument formed the basis of an hypothesis I proposed to explain effects of irradiation on the oxidation rate. So far this hypothesis seems to have withstood the test of time and further experimental observations.

We have still to consider the processes occurring during the post-transition oxidation period when the whole oxide does not represent a diffusion barrier. Here we have made extensive use of electron microscopy to characterise the size, location and development of pores and cracks in the oxide. We have also developed techniques based on the rate effect of capillary rise of an electrolyte in these flaws on the impedence of the oxide; and a mercury porosimeter, where the flow of mercury in the flaws was monitored by impedance measurements, to give information o.. the size, depth and frequency of these defects.

Studies of the stress generated in the oxide, and the recrystallization processes proceeding in the oxide have also been important in reaching conclusions about the mode of oxide breakdown.

I have concluded that, because small pores are always generated at the oxidation transition, whereas cracks in the oxide, although common, are not univerally observed, the development of small pores in the oxide is the primary cause of the loss of the protective nature of the pre-transition oxide. These pores are thought to develop at crystallites boundaries by virtue of the recrystallization processes occurring during oxide growth. The precise details of crystallite growth from specimen to specimen and alloy to alloy may be quite variable; however, the common end effect is that pores develop which permit oxygen gas to flow through most of the oxide thickness. For Zircaloy-2 any remaining barrier at the oxide/metal interface is exceedingly thin, only a few nano- meters at most, whereas for some other alloys such as Zr-2.5% Nb the pores appear only to penetrate through a much smaller fraction of the oxide thickness. Thus, unlike Zircaloy-2, where a pressure change during oxidation results in a proportional change in oxidation rate, for Zr-2.5% Nb little effect of a pressure change is seen. By the combined application of the above techniques to studies of the effects of variables such as heat treatment, alloying and irradiation it should be possible to predict the behaviour of alloys under any given set of conditions. However, the effort involved in such an approach would be massive, and it remains probable that in most instances "ad hoc" solutions will remain the preferred approach, unless such a solution is found to be too elusive.

In recent years my interests have expanded to cover the environmentally induced cracking of zirconium alloys, and their 9.

delayed hydride cracking. Whenever possible I have tried to adopt a similar philosophy to that described above in tackling these problems. Broader responsibilities have also permitted me to apply this approach to other aspects of the physical metallurgy of zirconium. T^ particular, I have been able to encourage other members of the ivterials Science Branch at Chalk River to develop new techniques wi anever possible to aid in the understanding of in-reactor creep, the defect structure of zirconium and ordered alloys. I expect to continue working in these areas in the future, and finish by thanking you for your attention, and the Kroll Award Committee for the honour bestowed on me.

Ac h i:c(<:Cc dg CIT.V i> t$ : It would take too long to list all the people who have collaborated in, assisted with, or commented on this programme over the years. The names of iay principal collaborators can be obtained from the co-authors listed in the bibliography. To them and the others acknowledged individually in individual papers I offer again my sincerest thanks for the help I have received.

Bj_b£ iognav'nij: [in •tei'&'VSc c hnonclcgij) •• 1. B. Cox, "Oxidation of Zirconium and its Alloys", Adv. in Corr. Sci. and Tech., Plenum, N.Y., Vol. V, p.173, 1976. 2. B. Cox, "Zirconium Alloys in High Temperature Water", Int. Conf. on High Temp., High Press Electrochem., Univ. of Surrey, Guildford, Jan. 1973, Pub. NACE 1976. 3. B. Cox and R.A. Ploc, Comments on the Origin of the Cubic Rate Law in Zirconium Alloy Oxidation, J.E.C.S., 1975, 122, 1744. 4. B. Cox, "Environmentally Induced Cracking of Zirconium Alloys", in Revs, in Coatings and Corrosion, Vol. 1, No. 4, p.366, Freund, Tel Aviv 1975. 5. B. Cox, "Techniques for Studying the Rate Controlling Processes during Metal Oxidation',' Proc. of 5th Int. Cong, on Met. Corr. Tokyo, 1972 (publ. NACE 1974) p.682. 6. B. Cox and J.C. Wood, "Iodine Induced Cracking of Zircaloy Fuel Cladding - A Review", Proc. of Symp. on Corrosion Problems in Energy Conversion and Generation, Electrochem. Soc, N.Y., Oct. 1974, p.275. 7. B. Cox, "Stress Corrosion Cracking of Zircaloys in Iodine Containing Environments" ASTM-STP-551 (1974) p.419. 10.

8. B. Cox, "A Correlation between Acoustic Emission during SCC and Fractography of Cracking of Zircaloys" Corrosion, 1974, 30_, 191.

9. B. Cox and A. Donner, "The Morphology of Thick Oxide Films en Zircaloy-2", J. Nucl. Mat., 1973, 47_', 1972.

10. B. Cox, "The Effect of Surface Films on the Initiation of Stress Corrosion Cracking of Zircaloy-2", Presented at the 56th Annual C.I.C. meeting, June 1973 (AECL-4589).

11. B. Cox, "Stress Corrosion Cracking of Zircaloy-2 in Neutral Aqueous Chloride Solutions at 25 C", Corrosion, 1973, 29^, 157.

12. B. CJX, "Accelerated Oxidation of Zircaloy-2 in Supercritical Steam" Atomic Energy of Canada Limited, Report AECL-4448 (1073)

13. W. Iliibner and B. Cox, "Electrochemical Properties and Oxida- tion of some Zirconium Alloys in Molten Salt at 300'-500 C", Atomic Energy of Canada Limited, Report AECL-4431 (1973).

14. B. Cox, "Stress-Corrosion Cracking of Zirconium Alloys" in Materials Research at AECL, Fall 1972, Atomic Energy of Canada Limited, Report AECL-4 353.

15. B. Cox, "Environmentally Induced Cracking of Zirconium Alloys", Corrosion, 1972, 2_8_, 207.

16. B. Cox, "Comments on the Effect of an Applied Electric Field on the Oxidation of Aluminum in the Temperature Range 50-400°C" Oxid. of Met., 1971, 3^ 529- 17. B. Cox, "Catastrophic Oxidation of Zircaloys in Fused Salts at 300°C", Oxid. of Met., 1971, 3_, 399.

18. B. Cox, "Comments on the Influence of Oxide Stress on the Breakaway Oxidation of Zircaloy-2", J. Nucl. Mat., 1971, 41, 96.

19. B. Cox, "Environmentally Induced Cracking of Zirconium Alloys. Ill, Cracking in Hot and Fused Salts", Atomic Energy of Canada Limited, Report AECL-3799 (1971) .

20. F.J. Shirvington and B. Cox, "A Study of Charge Transport Processes during the Oxidation of Zirconium Alloys", J. Nucl. Mat., 1970, 35, 211.

21. B. Cox, "Factors Affecting the Growth of Porous Anodic Oxide Films on Zirconium", J.E.C.S., 1970, 117, 654.

22. B. Cox, "Environmentally Induced Cracking of Zirconium Alloys. II, Liquid Metal Embrittlement", Atomic Energy of Canada Limited, Report AECL-3612 (1970). 11.

2J. B. COX, "Scanning Electron Microscopy", in Materials Research a; AECL, Sprinq 1970, Atomic Energy of Canada Limited, Report A1XL-36 0 4 .

24. B. Cox, "Environmentally Induced Cracking of Zirconium All°vs. I, Topography of Stress Corrosion Cracking in Methanolic Solu- tions" Atomic Energy of Canada Limited, Report AECL-3551 (1970).

25. B. Cox, "Comments on the Use of the Nuclear Reactions 1EO(d,p)17O to Study Oxygen Diffusion in Solids and Its Application to Zirconium", J. Appl . Phys . , 1969, 4_0, 4669.

2f: . H. Cox, "Thermal Oxidation of Metals" in Materials Research at Chalk River, Fall 1969, Atomic Energy of Canada Limited, Report AECL-3478.

27. B. Cox, "The Zirconium-Zirconia Interface", J. Aust. Inst. Met., 1969, 14, 123.

28. B. Cox, "Rate Controlling Processes during the Pre-Transition Oxidation of Zirconium Alloys", J. Nucl. Mat., 1969, 31, 48.

29. B. Cox, "Comments on Aqueous Corrosion of the Zircaloys at Low Temperatures", J. Nucl. Mat., 19 69, ^0, 3 51.

30. B. Cox, "Processes Occurring during the Breakdown of Oxide Films on Zirconium Alloys", J. Nucl. Mat., 1969, 329, 50.

31. B. Cox, "The Morphology of Zirconia Films and its Relation to the Oxidation Kinetics", Atomic Energy of Canada Limited., Report AECL-3285 (1969) .

32. B. Cox, "Comments on the Influence of Thin Noble Metal Films on Zirconium Oxidation", J.E.C.S., 1968, 115, 1259.

33. B. Cox, and A.R. Mclntosh, "The Oxide Topography on Crystal-bar Zirconium and Reactor-Grade Sponge Zirconium", Atomic Energy of Canada Limited, Report AECL-3223 (1968) .

34. B. Cox and J.P. Pemsler, "Diffusion of Oxygen in Growing Zirco- nia Films", J. Nucl. Mat., 1968, 2Q_, 73.

35. B. Cox, "Effects of Irradiation on the Oxidation of Zirconium Alloys in High Temperature Aqueous Environments", J. Nucl. Mat., 1968, 2S_, 1.

36. B. Cox, "A Porosimeter for Determining the Sizes of Flaws in Zirconia or Other Insulating Films in situ", J. Nucl. Mat., 1968, 2J7_, 1.

37. B. Cox, "Comments on the Dielectric Constant of Zirconia" Brit. J. Appl. Phys. (J. Phys. D.) 1968, Ser. 2, .L, 671.

38. J.S. Sheasby and B. Cox, "Oxygen Diffusion in Alpha-Niobium Pentoxide", J. Less Comm. Met., 1968, 15, 129. 12.

39. B. Cox, "Low Temperature (<300°C) Oxidation of Zircaloy-2 in Water", J. Nucl. Mat., 1968, 2_5, 310.

40. B. Cox, "Rate Controlling Processes during the Oxidation of Zirconium Alloys", Atomic Energy of Canada Limited, Report AECL-2777 (1967) .

41. B. Cox, "Oxidation cf Zirconium-Aluminum Alloys", Atomic Energy of Canada Limited, ReportAECL-2776 (1967).

42. B. Cox and D.L. Speirs, "Rectification by Oxide Films on Zirconium Alloys", Atomic Energy of Canada Limited, Report AECL-2690 (19G7).

43. B. Cox, "The Use of Electrical Methods for Investigating the Growth and Breakdown of Oxide Films on Zirconium Alloys", Atomic Energy of Canada Limited, Report AECL-2668 (1967).

44. B. Cox, "Porosity in Oxide Films on Zirconium Alloys", Proceed- ings of 3rd Int. Cong, of Met. Corr., Moscow, 1966, Vol. IV, p. 341 (Izdat., Moscow, 1969).

45. B. Cox and C. Roy, "Transport of Oxygen in Oxide Films on Zirconium Determined by the Nuclear Reactor O17(He3,a)O16", Electrochem. Tech. 1966, 4_, 122, and Atomic Energy of Canada Limited, Report AECL-2350~(1965).

46. B. Cox and C. Roy, "The Use of Tritium as a Tracer in Studies of Hydrogen Uptake by Zirconium Alloys", Atomic Energy of Canada Limited, Report AECL-2519 (1965).

47. J.A.L. Robertson, and B. Cox, "Comparison of Water-Ccoled Fuels", Nuclear News, 1964, Oct., p. 34.

48. B. Cox and R.W. Ball, "A Study of Oxide Film Breakdown on Zirconium Alloys by Capacitance Measurements", Atomic Energy of Canada Limited, Report AECL-2144 (1964) .

49. J.H. Chute (ed. B. Cox), "An Electron Microscope Study of the Oxidation of Some Zirconium Alloys in Steam", Atomic Energy of Canada Limited, Report AECL-1999 (1964) .

50. J. Adam and B. Cox, "The Irradiation-Induced Phase Transfor- mation in Zirconia", Reactor Sci. & Tech. (J. Nucl. Eng. A/B), 1963, r7, 435.

51. B. Cox and Mrs. J.A. Read, "Oxidation of a Zr-2.5% Nb alloy in Steam and Air", UKAEA Report, AERE-R4459 (1963) .

52. B. Cox, "Some Effects of Pressure on the Oxidation of Zircaloy-2 in Steam and Oxygen", J. Less Comm. Metals, 1963, 5_, 325; and Proc. of Conf. on Zr Alloy Tech., Castlewood, Calif"., Nov. 1962, GEAP-4089. 13.

53. B. Cox and B.R. Harder, "The Effect of Dissolved Oxygen on the Oxidation of Zircaloy-2 by Steam", J.E.C.S., 1963, 110, 1110. 54. B. Coy "The Effect of some Alloying Additions on the Oxidation of Zirconium in Steam", U.K.A.E.A. Report, AERE-R4458 (1963). 55. B. Cox, "Some Factors which Affect the Rate of Oxidation and Hydrogen Absorption of Zircalcy-2 in Steam", U.K.A.E.A. Report AERE-4348 (1963). 56. B. Cox and T. Johnston, "The Oxidation and Corrosion of Niobium (Columbium)", Trans. Met. Soc. AIME, 1963, 227, 36. 57. R.C. Asher and B. Cox, "The Effects of Irradiation on the Oxi- dation of Zirconium Alloys", Proceeding of Conf. on Corr. of Reactor Materials, Salzburg, 1962, p.209 (IAEA, Vienna). 58. B. Cox, P.G. Chadd and J.F. Short, "The Oxidation and Corrosion of Zirconium and its Alloys. XV, Further Studies of Zirconium- Niobium Alloys" U.K.A.E.A. Report, AERE-R4134 (1962). 59. B. Cox, "Hydrogen Absorption by Zircaloy-2 and Some Other Alloys during Corrosion in Steam", J.E.C.S., 1962, 109, 6; and U.K.A.L'.A. Report, AERE-R3556 (1961). 60. B. Cox and T. Johnston, "The Oxidation and Corrosion of Zirco- nium and its Alloys. XIII, Some Observations of Hydride in Zirconium and Zircaloy-2 and its Subsequent Effect on Corrosion" U.K.A.E.A. Report, AERE-R3881 (1962). 61. B. Cox, "Recent Developments in Zirconium Alloy Corrosion Tech- nology" in Progress in Nucl. Energy, Ser. IV, Vol. 4 Pergamon London, 1962, p.166. 62. B. Cox, "Causes of a Second Transition Point Occurring during Oxidation of Zirconium Alloys", Corrosion, 1962, 18_, 336. 63. B. Cox and B.R. Harder, "An Attempt to Locate Intermetallic Particles in Zirconium Alloys using a Bitter Figure Technique", U.K.A.E.A. Report AERE-R3845 (1961). 64. R.C. Asher, B. Cox and J.K. Dawson, "Investigations Relating to the Use of Zirconium Alloys in Steam-Cooled Reactors", Proc. of I.A.E.A. Symposium on Power Reactor Experiments, Vol. II, p.135, Vienna, 1961. 65. B. Cox, K. Alcock and F.W. Derrick, "The Oxidation and Corro- sion of Zirconium and its Alloys. VI, The Mechanism of the Fission Fragment Induced Corrosion of Zircaloy-2", J.E.C.S. 1961, 108, 129; and U.K.A.E.A. Report, AERE-R2932, (1959). 66. B. Cox, "The Oxidation and Corrosion of Zirconium and its Alloys. V, Mechanism of Oxide Film Growth and Breakdown 14 .

on Zirconium and Zircaloy-2", J.E.C.S., 1961,108, 24? and U.K.A.E.A. Report AERE-R2931 (1959). 67. B. Cox, "Heavy Particle Irradiation Effects during th3 Reaction of Solids with Gases or Liquids", U.K.A.E.A. Report AERE-M742. 68. B. Cox, M.J. Davies and T. Johnston, "The Oxidation and Corrosion of Zirconium and its Alloys. XI, The Oxidation Kinetics of Zirconium-Niobium Binary Alloys in Steam at 300-500°C", U.K.A.E.A. Report, AERE-R3257 (1968). 69. B. Cox, "The Effect of Fission Fragment and Neutron Irradiation on the Kinetics of Zirconium Oxidation", Proceedings of 4th Int. Symp. on the Reactivity of Solids, Amsterdam, 1960, p.425, eds. J.H. de Boer et al. (Elsevier). 70. B. Cox, "The Oxidation and Corrosion of Zirconium and its Alloys. III. Oxide Film Breakdown in Arc-Melted Sponge Zirconium", Corrosion, i960, 16_, 3806; and U.K.A.E.A. Report, AERE-R2874 (1959). 71. B. Cox, "The Oxidation and Corrosion of Zirconium and its Alloys. II, The Effect of Carbide Inclusions on Oxide Film Failure", Corrosion, 1960, 1_6_, 188t, and U.K.A.E.A. Report, AERE-R2873 (1969). 72. B. Cox, M.J. Davies and A.D. Dent, "The Oxidation and Corro- sion of Zirconium and its Alloys. X, Hydrogen Absorption During Oxidation in Steam and Aqueous Solutions", U.K.A.E.A., AERE-M621 (1960). 73. B. Cox and T. Johnston, "The Oxidation and Corrosion of Zirco- nium and its Alloys. IX, Observation of a Second Transition Point during the Oxidation of Zirconium Alloys", U.K.A.E.A. Report, AERE-R3256 (1960). 74. K. Alcock and B. Cox, "The Oxidation and Corrosion of Zirconium and its Alloys. VIII, The Effect of Neutron Irradiation on the Dissolution Rate of ZrO2", U.K.A.E.A. Report, AERE-R3114 (1959) . 75. B. Cox, "An Investigation of the Mechanism of Oxide Film Growth and Failure on Zirconium and Zircaloy-2", Proceedings of AEC-EURATOM Conf. on Aqueous Corrosion of Reactor Materials, Brussels, Oct. 1969, U.S.A.E.C. Report TID-7587. 76. J. Adam and B. Cox, "Neutron and Fission Fragment Damage in Zirconia',' Phys. Rev. Lett., 1959, 2_, 543. 77. J. Adam and B. Cox, "The Irradiation-Induced Phase Transfor- mation in Zirconium Solid Solutions", J. Nucl. En. A, Reactor Science, 1959, 11, 31; and U.K.A.E.A. Report, AERE-M415 (1959). 15.

78. K. Alcock and B. Cox, "The Oxidation and Corrosion of Zirco- nium and its Alloys. VII, Experience with High Pressure Equipment for Use with Aqueous Solutions at High Temperatures" U.K.A.R.A. Report, AERE-R3029 (1959). 79. B. Cox, "The Oxidation and Corrosion of Zirconium and its Alloys. IV, The Effect of Fission Fragment Irradiation on the Subsequent Corrosion of Two Zirconium Alloys", U.K.A.E.A. Report, AERE-R2875 (1959). 80. K. Alcock and B. Cox, "The Oxidation and Corrosion of Zirconium and its Alloys. I, Destructive Internal Oxidation of Zircaloy-2 in Steam Under Irradiation at 300°C",U.K.A.E.A. Report, AERE-C/R 2826 (1959). 81. B. Cox and K. Alcock, "The Effect of Surface Preparation and Neutron Irradiation on the Corrosion of Zircaloy-2 in High Temperature Aqueous Corrosion", U.K.A.E.A. Report, AERE-C/M 353 (1958). P.-.. J.K. Dawson, B. Cox, R.Murdoch and R.G. Sowden, "Some Chemical Problems of Homogeneous Aqueous Reactors", Proc. of 2nd U.N. Conf. on the Peaceful Uses of Atomic Energy, Geneva, 1958, P.46. 16 .

TABLE I. Understanding the Thermal Oxidation of Zirconium Alloys

Is the oxidation controlled by diffusion through a protective oxide film? (deduce from kinetics and pressure dependence)

If NO,

Study transport of ions and electrons through oxide film Study the migration of molecular species in the oxide filn Ions Electrons

1. Which is mobile ion? 1. What is electron transport process? 1. Study visual occurrence of cracks, (metal, oxygen, both) (diffusion, tunneling, emission) pores, etc., in oxide.

2. What is oxide defect structure? 2. What is mobile species? 2- Determine size, depth, and frequency (p-type, n-type, junction) of these defects.

3. What is macroscopic diffusion 3. What is microscopic transport route 3. Study processing generating defects. coefficient? for electron current? a. Stress in oxide b. Recrystallization of oxide 4. What is microscopic diffusion process? 4. What is contribution of surface conduction to overall electron transport process?

5. Identify diffusion route from studies of oxide morphology.

Define ionic transport process Define electron transport process Define oxide film breakdown process

Determine rate-controlling process

Assess effect of variables (e.g. heat treatment, irradiation) on all processes"

Predict behaviour of Alloy under given Conditions f he ln(t-riKi(i')ii;il St;imJ;ii'<] Sm;il Number

ISSN 0067 <).?6"?

h:is been as^iiiiu'd to this series of reports.

In iflentif) individual dociimonls in the series ue li;»o jssignetl iin A!:( I. mimher.

I'leii.H" refer to the Al-.t'L iitimher when requeslini; adililiiinal eojiies of this doeiiment from

Scientific Document Distribution Office \toniie [nergv of C;HI;KI;I Limited Ch;ilk Kiu'r. ()iil:iri'>. (':HKK1;I

KOI 1.10

Price S2.00 per cop\

1222-76