Simulated Environmental Tests for Selected Advanced Technology Fuel Cladding Solutions

The motivation behind the development work on advanced technology fuel (ATF), also sometimes called accident tolerant fuel, claddings is based on specific limitations associated with zirconium alloys under design-basis accident and beyond-design-basis accident scenarios. Zirconium fuel cladding in current light water reactors (LWR) provides adequate material performance while being relatively transparent to neutrons produced in a reactor core. However, hydrogen release due to the reaction of Zr with steam is one of the main contributors to serious loss of integrity scenarios in nuclear reactor accidents. The development of ATF cladding materials has been under consideration particularly after the events at Fukushima Daiichi nuclear power plant. In Fukushima Daiichi, a station black-out caused by an earthquake and subsequent tsunami resulted in failure of core cooling, temperature rise, and zirconium-steam reactions. The reaction produced hundreds of kilograms of H₂, which leaked out of the reactor pressure vessels in units 1–3. Subsequent hydrogen explosions resulted in severe damage to reactor buildings. After fuel rod cooling has been lost, failure mechanisms other than loss of integrity due to high temperature Zr-steam reactions are also possible, e.g., ballooning and bursting. Many of the failure mechanisms are expected to be LWR concept independent. On the other hand, many of the small modular reactor concepts have passive cooling, which reduces the risk of cladding overheating.

Several potential solutions have been proposed as replacements for present-day zirconium-based fuel cladding materials, such as improved Zr-based alloys, FeCrAl alloy, Mo-based alloys, SiC composites, and various coatings. Also, modification of fuel pellets has been proposed in order to reduce the risk of overheating [1–7]. In the development of the new materials, the understanding of the damage mechanisms of the fuel cladding is important, i.e., the oxidation behavior and mechanical properties in normal operating conditions and in loss of coolant scenarios in high temperature steam environments.

A number of high temperature steam and autoclave tests in normal LWR conditions on different materials and coating combinations were performed at VTT between 2017 and 2020. The tests were VTT's contribution to the IAEA coordinated research project analysis of options and experimental examination of fuels for water cooled reactors with increased accident tolerance (ACTOF). The results are reported in this paper.

In order to improve candidate materials' performance at high temperatures (e.g., in beyond design basis accidents), coatings were deposited on Zr-alloys. In this work, Zr coupon specimens were coated by using physical vapor deposition (PVD) technique. PVD is a process by which a thin film of material (typically a few microns only) is deposited on a substrate. PVD has the advantage over chemical vapor deposition (CVD) that the deposition temperature is lower, in the range of 500 °C or below [8,9]. Thus, undesired diffusion processes or reactions between substrate and coating are avoided. In contrast to plasma spraying, the PVD process leads to the formation of homogeneous coatings without pores or cracks. The coated layers are also significantly thinner.

The following coating material-Zircaloy combinations were provided by project partners: MAX phase coating (general formula: M_n+1AX_n, where n=1-3, M is an early transition metal, A is an A-group element, and X is either carbon or nitrogen) by KIT/Germany, Cr coating by CTU/Czech, and ZrSiCr by INCT/Poland. KIT and CTU provided also uncoated substrates for reference tests. The materials, test conditions, and durations are shown in Table 1. The nominal compositions of substrate materials are as following (in weight %):

• Zry-4 (1.5Sn 0.20Fe 0.1Cr 0.09-0.13O)

• Zry-2 (1.5Sn 0.12Fe 0.1Cr 0.12O 0.05Ni)

Tests were performed in flowing steam at temperatures between 1100 and 1300 °C and, in order to confirm candidate ATF cladding materials viability in normal LWR conditions, also in pressurized water reactor (PWR) water at 360 °C with [Li] = 2–2.2 ppm, [B] = 600–1000 ppm, and [H₂] = 3 ppm. The tested specimens were in a coupon form and had different coating compositions and coating thicknesses:

• Zry-4 with a ∼15 μm pure Cr coating,

• Zry-4 with a Cr/Cr₂AlC/Cr multilayer coating with sublayer thicknesses of 1.5 μm Cr, 4.5 μm Cr₂AlC, and 0.5 μm Cr,

• Zry-2 with a ∼2.5 μm ZrSiCr coating.

The multilayer coating called MAX phase is also an alumina-forming material in high-temperature oxidation [9].

The autoclave tests in PWR water conditions were performed in VTT's autoclave laboratory, see Fig. 1(a). The autoclave is connected to a water recirculation loop and the loop system consists of low- and high-pressure sections. The low-pressure section consists of a make-up water tank equipped with nitrogen, hydrogen, and mixed gas gasification systems, a low pressure recirculation pump, a mixed bed ion exchanger used for water purification before the test solution is prepared, and water chemistry measurement instrumentation for on-line monitoring of conductivity, pH, dissolved oxygen, and dissolved hydrogen. Water chemistry monitoring instruments can be switched either to measure autoclave inlet or outlet water.

Steam tests were performed by using a steam furnace system shown in Fig. 1(b). The furnace consists of an alumina tube (working tube), a specimen tube, and a sample holder, all manufactured from recrystallized alumina. The furnace has one zone, i.e., uniform heat is produced in the middle of test section. The desired temperature can be maintained in a 0.1 m long zone with silicon carbide heating elements. The maximum temperature of the steam furnace is 1600 °C.

An example of a typical specimen exposure sequence in a high temperature steam furnace is shown in Fig. 2. The curves represent specimen temperature (initially rising, then horizontal, and finally dropping line) and steam temperature in preheater (water vaporizer, line starting to drop at ca. 600 s and rising after 2500 s) in the case of MAX phase coated Zry-4 specimens at 1200 °C for 1800s. All coupons were tested one specimen at a time in the furnace. The coupon specimen was inserted into the furnace manually by using a pincer at a constant rate of 0.4 × 10⁻³ m/s. The same procedure was applied to remove coupon specimen from the furnace.

The steam was injected into the furnace after the coupon specimen reached the test section in the middle part of the furnace and the pump was switched on in order to generate steam atmosphere in the furnace (at around 600 s in Fig. 2). The furnace temperature increased from 30 °C to 1100 °C in ca. 800 s, being more than 2.5 °C/s between 500 and 800 °C. Contrary to this, the cooling rate of the furnace from 1100 °C to 150 °C took place in ca. 1080 s. The specimens were removed from the furnace after desired exposure times (3600 s, 1800s, and 300 s), and then photographed and weighed. Metallographic cross section specimens were prepared.

After autoclave and steam furnace tests, the specimens were characterized using a Zeiss Crossbeam 540 field emission gun-scanning electron microscope (FEG-SEM) equipped with a semiquantitative energy dispersive X-ray spectrometer (EDS). The SEM was used to take secondary electron and back scatter electron images. SEM-EDS was used in order to map and analyze elemental distributions and contents especially from the outer and inner oxide layers. Prior to testing, specimen dimensions were measured. The specimens were weighed before and after testing in order to calculate the weight change.

Five different specimens, including two reference specimens (uncoated Zry-4/KIT and Zry-4/CTU), were included in the PWR autoclave tests. The specimens were exposed for 21, 42, and 63 days except for specimens delivered by INCT. INCT specimens were installed into the autoclave near the end of the PWR test campaign and were exposed for 16.5 days.

Based on the weight change measurements, Fig. 3, the uncoated Zry-4 reference materials behaved as expected, i.e., the observed weight gain increased as a function of exposure time due to the growth of a protective oxide film. The Zry-4/KIT reference specimen exposed up to 63 days showed 1 μm–10 μm thick continuous oxide layer, Table 2, with some porosity and several cracks through the oxide layer in the post-test studies.

Cr coated Zry-4 showed only a small increase in weight gain after 21 and 42 days exposure. A somewhat lower weight gain (average of three measurements) is seen after the final exposure time of 63 days than after the 42 days exposure, although this may be a consequence of measurement accuracy (taking the measurement accuracy into account, many of the weight gains are overlapping in Fig. 3). SEM cross section and SEM-EDS spot analyses of Cr coated Zry-4 specimen after 42 days exposure are shown in Fig. 4 and Table 3. The coating + oxide film thickness ranged from about 20 μm to 40 μm. The coating was continuous but some cracks through the coating and cracks between the coating and the substrate were observed. A crack trough the coating can be seen in the SEM cross section. Based on the observations, it did not result in the oxidation of the underlying Zr substrate. The spot analyses do not reveal any considerable oxidation or mixing of the elements.

MAX phase-coated Zry-4 showed significant weight loss due to coating exfoliation after all exposure times. In all exposures, the coating layers indicated cracks through the coating as well as cracks between the coating and substrate. This is most probably due to different thermal expansion coefficients between formed oxide phases during heating and cooling periods. The observed coating/oxide layer thickness was from 0.5 μm to 5 μm after the 21 day exposure, from 3 μm to 10 μm after the 42 day exposure and from 2 μm to 40 μm after the 63 day exposure. The original unexposed coating layer thickness was around 6.5 μm. The observed coating/oxide layer thicknesses are rather inaccurate estimates as most of the coatings were lost most likely during the post-test specimen preparation. Most significant coating loss was observed after the 63 day exposure.

ZrSiCr-coated Zry-2 specimens showed also small weight gain after the short exposure period of 16.5 days. The ZrSiCr coated Zry-2 specimen had a 2 μm to 10 μm thick oxide layer after the exposure. The layer was continuous otherwise but parts of the oxide layer were lost during the post-test specimen preparation. Also, cracks through the oxide and cracks between the oxide layer and the substrate were observed. The increased crack density in the coating itself due to autoclave exposure resulted in oxygen enriched areas in the substrate under the coating, i.e., the protective feature of the coating layer was lost already after 16.5 days of exposure.

Three specimens per material/coating combination (Cr and MAX phase coated) were exposed to flowing steam at 1100 °C for 3600 s, 1200 °C for 1800s and 1300 °C for 300 s.

The weight changes of Cr-coated Zry-4 specimens are shown in Table 4. Moderate increasing in weight was observed with increasing temperature up to 1200 °C although exposure time decreased down to 1800s. The observed weight gain after exposure at 1300 °C was, on the other hand, relatively high being 6–20 times higher than at the lower temperatures although the exposure time was only 300 s.

The thickness of coating and the oxide layers in Cr-coated Zry-4 specimens increased with increasing temperature. At 1100 °C, the thickness was about 10 μm, at 1200 °C about 15 μm, and at 1300 °C the thickness ranged from 40 μm to 400 μm. The coating of the specimens after exposures at 1100 °C and 1200 °C was continuous. However, cracks between the coating and the oxide layer were found. The specimen exposed at 1200 °C also had cracks through the oxide layer. In 1100 °C and 1200 °C specimens, Fe enrichment was observed on the outer oxide layer as well as in the thin Cr and Zr-rich zone between Zr substrate and oxide layer. The coating of specimen exposed at 1300 °C was also continuous, but the coating and the substrate partly came detached during the post-test specimen preparation.

SEM-EDS spot analyses of the Cr coated Zry-4 specimen, which was exposed at 1300 °C for 300 s, are shown in Fig. 5 and Table 5. The strong oxidation of the Zr substrate (selected area 2 in Fig. 5 and Table 5) at 1300 °C indicates poor oxidation resistance of this type of coating at the highest temperatures.

In general, all specimens of MAX phase-coated Zry-4 oxidized heavily as shown in Fig. 6. Oxidation was very uneven, so quantitative measurements from oxide layer thicknesses were difficult to make, especially after the exposures to the higher temperatures of 1200 °C and 1300 °C. Every specimen had an area on both sides of the coupon where no oxidation had occurred, i.e., typically in the middle part of the coupon specimen. It seems that the rapid growth of the oxide is related to the edges of the specimens, i.e., from high surface energy sites. In the central regions, the coating is in relatively good condition. The weight changes after the steam exposures are shown in Table 6. The weight of the MAX phase-coated Zry-4 specimens increased as a function of test temperature although the exposure time decreased from 3600 s to 300 s as with the Cr-coated Zry-4 specimens. At 1300 °C, the weight gain was as high as 30% when all peeled off oxide was included.

Figure 7(a) shows (arrows in the image) that the MAX phase coating is still visible on top of the oxide layer in both sides of specimen after the exposure. Based on the observations, one can see that the oxide has grown fast toward the center from the edge of the specimen. However, almost no oxide formation under the intact coating is observed (right-hand-side image of Fig. 7). In an area with locally fractured coating, outlined with an ellipse, some oxide formation is seen, but on the arrow pointed area, where only the top layer of coating is breached, no oxide formation is visible. In addition, no fragile α-Zr is existing most probably due to the fact that there is no oxygen penetration. The observed coating condition underneath of specimen seems to be intact. There is no brittle α-Zr present, i.e., no oxidation.

The coating configuration of the MAX phase-coated Zry-4 is shown in Fig. 8(a) and SEM-EDS maps in Fig. 8(b) show elements of interest and grayscale image from the area where MAX phase coating was intact and no ZrO₂ was formed yet. It is noteworthy that aluminum has dispersed heavily from the assumed uniform initial middle layer, i.e., from original layer of Cr₂AlC. Due to this, there would be a clear need to characterize also an unexposed specimen in order to understand the behavior of aluminum at this temperature. However, this was not done due to limited number of specimens available.

It seems that a tin-rich layer develops into the oxide of MAX phase coated Zry-4 during high temperature oxidation as stated in Ref. [10]. A high magnification image from the tin-rich layer from the specimen held in the steam furnace at 1100 °C for 3600 s is shown in Fig. 9(a) and tin enriched areas between the oxide columns in Fig. 9(b). These areas do not exist above the tin-rich layer shown in Fig. 9(a). It was also noticed that the area above the tin-rich layer charges in the SEM and thus can be considered less conductive than the oxide below the tin-rich layer. In future, further measurements need to be done to define the exact composition of the tin-rich layer/precipitates in the tested specimens.

Remnants of MAX phase-coated Zry-4 coupon specimen after the exposure at 1300 °C for 300 s are shown in Fig. 10. Inside the rectangle shown in Fig. 10, different layers of the MAX phase coating can be seen; however, no further analysis of the specimen was done in this condition. A cross section was prepared approximately along the horizontal line shown in Fig. 10. SEM images from the almost thoroughly oxidized perimeters of the cross section are shown in Figs. 11 and 12. SEM images of the central area are shown in Figs. 13 and 14.

The bright areas in Figs. 11 and 12 show areas with only brittle α-Zr remaining from the initial Zr specimen. Only a thin layer of Zr is left in the perimeter of the specimen, Fig. 12. Tin from the Zry-4 matrix has flown into the crack and into the oxide below the α-Zr. The previously mentioned tin-rich layer, which forms also in lower temperatures as shown in Fig. 9, can also be seen above the α-Zr within the oxide as a white horizontal layer. Based on the substantial concentration of tin, it seems that the temperature has been above a threshold temperature, where tin decomposes from the initial β-Zr matrix completely. In general, in the areas where oxidation has occurred, similar development of microstructure is reported in literature [11,12].

SEM image from the middle of the cross section with evidently less oxidation is shown in Fig. 13 (full cross section from the middle of the same specimen as shown in Fig. 10). Both surfaces were observed to still have the MAX phase coating attached, but a ∼40 μm thick layer of brittle α-Zr was observed to have formed on both sides. The remaining part mostly consisted of the typical prior β-Zr phase consisting of acicular Widmanstätten plates described in Ref. [10]. Only limited oxidation has appeared on spots below the specimen where MAX phase coating has cracked (outlined with ellipses in Fig. 13). Figure 14 shows the MAX phase coating still attached on the surface, in the central region of the coupon specimen. SEM-EDS compositional analyses from this area showed roughly similar composition profile on the MAX phase and the near surface region as the specimen exposed to 1100 °C, Fig. 8. Thus, the coating remains attached, but is heavily leached of elements. No oxide layer formation can be observed in this area, but the ∼40 μm thick oxygen stabilized brittle α-Zr layer can be seen near the surface, below the MAX phase coating, containing transgranular cracks which turn parallel near the surface.

The MAX phase coating itself seems to be sustained in the steam environment. The brittleness of the specimen is more related to the β-Zr to α-Zr phase transformation (this is a coating independent phenomenon, not specific to the MAX phase coating). Tin diffuses into a layer within the oxide. Simultaneously the β-Zr transforms into a brittle α-Zr, containing an elevated amount (5–10%) of oxygen. This oxygen stabilized α-Zr layer seems to develop transgranular cracks mostly perpendicular to the surface of the specimen and spaced quite evenly. It cannot be defined based on this study if the cracking is related to increased oxygen or diffusion of tin out from the matrix. Literature would support the latter theory [11,12].

Characterization of the initial state of the coating materials should be made (preferably with the same equipment) and the test configuration should be updated to tubular and longer specimens so that the harsher condition at the peripheral areas of the specimen would not be highlighted as much as now. It can be observed from coupon specimen exposed at 1300 °C for 300 s that the oxidation progresses very quickly from specimen sides toward the central area. On the other hand, at the central area, the oxidation has progressed quite slowly, even on the areas, e.g., where the MAX phase coating has cracks. One could speculate here that the test arrangement with coupon-shaped specimens has distinct corner areas where the environment is far harsher than in the center of the specimen, and the center would represent the free surface of a cladding tube better. On the other hand, if the quick oxidation results from high surface energy in the edges, the situation is different. If a cladding tube, for example, breaks or cracks, the exposed edges would oxidize quickly. Based on these results, it is not possible to estimate what would be the situation if there would be a large scratch caused by a projectile etc. in the coolant flow.

In PWR water at 360 °C, the reference Zry-4 manages as well as could be expected due to the long and successful service in nuclear power plant normal operating conditions. Based on the weight change, the most stable coating was the Cr coating on the Zry-4 specimens, which did not change in any major way during the autoclave exposures. Although some coating cracking was observed, it did not result in oxidation of the underlying substrate. The MAX phase coating is not applicable in these conditions. The ZrSiCr coating also managed less well. The exposure in PWR water resulted in oxidation in the underlying Zry-2 substrate.

Neither Cr nor MAX phase coating protected the underlying Zry-4 substrate very well in steam at the temperatures from 1100 to 1300 °C encountered in design-basis accident or beyond design-basis accident.

The test materials were supplied by Czech Technical University in Prague, Institute of Nuclear Chemistry and Technology (Poland), and Karlsruhe Institute of Technology (Germany).

• The work was realized within GENXFIN (Safety of new reactor technologies) and INFLAME (Interdisciplinary fuels and materials) projects, which were subprojects in Finnish national research programs SAFIR2018 and SAFIR2022 (The Finnish Research Programs on Nuclear Power Plant Safety). The projects were funded by VTT Technical Research Centre of Finland and Finnish Nuclear Waste Management Fund.

ATF =

advanced technology fuel (sometimes accident tolerant fuel)

CTU =

Czech Technical University in Prague

CVD =

chemical vapor deposition

EDS =

energy dispersive X-ray spectrometer

FEG-SEM =

field emission gun-scanning electron microscope

IAEA =

International Atomic Energy Association

INCT =

Institute of Nuclear Chemistry and Technology (Poland)

KIT =

Karlsruhe Institute of Technology (Germany)

LWR =

light water reactor

ppm =

parts per million

PVD =

physical vapor deposition

PWR =

pressurized water reactor

SA =

selected area, an area selected for EDS analysis

VTT =

VTT Technical Research Centre Center of Finland

Zry =

zircaloy

Δw =

weight change