The TANDEM project is a European initiative funded under the EURATOM program. The project started on September 2022 and has a duration of 36 months. TANDEM stands for Small Modular ReacTor for a European sAfe aNd Decarbonized Energy Mix.Small Modular Reactors (SMRs) can be hybridized with other energy sources, storage systems and energy conversion applications to provide electricity, heat and hydrogen. Hybrid energy systems have the potential to strongly contribute to the energy decarbonization targeting carbon-neutrality in Europe by 2050. However, the integration of nuclear reactors, particularly SMRs, in hybrid energy systems, is a new R&D topic to be investigated. In this context, the TANDEM project aims to develop assessments and tools to facilitate the safe and efficient integration of SMRs into low-carbon hybrid energy systems. An open-source “TANDEM” model library of hybrid system components will be developed in Modelica language which, by coupling, will extend the capabilities of existing tools implemented in the project. The project proposes to specifically address the safety issues of SMRs related to their integration into hybrid energy systems, involving specific interactions between SMRs and the rest of the hybrid systems; new initiating events may have to be considered in the safety approach.TANDEM will study two hybrid systems covering the main trends of the European energy policy and market evolution at 2035's horizon: a district heating network and power supply in a large urban area, and an energy hub serving energy conversion systems, including hydrogen production; the energy hub is inspired from a harbor-like infrastructure. TANDEM will provide assessments on SMR safety, hybrid system operationality and techno-economics. Societal considerations will also be encased by analyzing European citizen engagement in SMR technology safety.The work will result in technical, economic and societal recommendations and policy briefs on the safety of SMRs and their integration into hybrid energy systems for industry, R&D teams, Technical Safety Organizations, regulators, Non-Governmental Organizations and policy makers. The TANDEM consortium will involve 17 partners from 8 European countries (Belgium, Czech Republic, Finland, France, Germany, Italy, Spain, Ukraine).The TANDEM project has the ambition to become a pioneer initiative in Europe in gathering efforts and expertise around development of SMRs integration into hybrid energy systems. The dissemination and the exploitation of the project outcomes as well as the proposed Education & Training activities shall serve as a basis for a number of new R&D and innovation projects addressing the safety issues of SMRs and their integration into hybrid energy systems.

The TANDEM project is a European initiative funded under the EURATOM program. The project started on September 2022 and has a duration of 36 months. TANDEM stands for Small Modular ReacTor for a European sAfe aNd Decarbonized Energy Mix. Small Modular Reactors (SMRs) can be hybridized with other energy sources, storage systems and energy conversion applications to provide electricity, heat and hydrogen. Hybrid energy systems have the potential to strongly contribute to the energy decarbonization targeting carbon-neutrality in Europe by 2050. However, the integration of nuclear reactors, particularly SMRs, in hybrid energy systems, is a new R&D topic to be investigated. In this context, the TANDEM project aims to develop assessments and tools to facilitate the safe and efficient integration of SMRs into low-carbon hybrid energy systems. An open-source “TANDEM” model library of hybrid system components will be developed in Modelica language which, by coupling, will extend the capabilities of existing tools implemented in the project. The project proposes to specifically address the safety issues of SMRs related to their integration into hybrid energy systems, involving specific interactions between SMRs and the rest of the hybrid systems; new initiating events may have to be considered in the safety approach. TANDEM will study two hybrid systems covering the main trends of the European energy policy and market evolution at 2035's horizon: a district heating network and power supply in a large urban area, and an energy hub serving energy conversion systems, including hydrogen production; the energy hub is inspired from a harbor-like infrastructure. TANDEM will provide assessments on SMR safety, hybrid system operationality and techno-economics. Societal considerations will also be encased by analyzing European citizen engagement in SMR technology safety. The work will result in technical, economic and societal recommendations and policy briefs on the safety of SMRs and their integration into hybrid energy systems for industry, R&D teams, Technical Safety Organizations, regulators, Non-Governmental Organizations and policy makers. The TANDEM consortium will involve 17 partners from 8 European countries (Belgium, Czech Republic, Finland, France, Germany, Italy, Spain, Ukraine). The TANDEM project has the ambition to become a pioneer initiative in Europe in gathering efforts and expertise around development of SMRs integration into hybrid energy systems. The dissemination and the exploitation of the project outcomes as well as the proposed Education & Training activities shall serve as a basis for a number of new R&D and innovation projects addressing the safety issues of SMRs and their integration into hybrid energy systems.

Severe accidents in nuclear power plants with loss of cooling or loss of coolant and core uncovery can lead to air ingress into fuel assemblies. If air or a steam-air-mixture comes into contact with cladding material, the exothermal character of zirconium oxidation and nitride formation leads to enhanced cladding oxidation, heatup and degradation up to core melt and fission product release. Two experiments, QUENCH-18 and CODEX-AIT3 were performed in the frame of EC projects ALISA and SAFEST to fill knowledge gaps on relevant phenomena. Both tests are accompanied by pre- and post-test benchmarks within the NUGENIA project QUESA. The pre-test benchmarks were completed, while the post-test benchmarks are currently running. Results of the pre-test benchmarks indicate that participating codes can qualitatively predict the experimental observations. Generally, the range of the predictions are wider for CODEX-AIT-3 while the trend is close except one for QUENCH-18. The specification of both test procedures was additionally supported by parameter studies. First post-test benchmark results for QUENCH-18 show that code predictions are in good agreement during the preoxidation and mixed steam/air phase. In the subsequent reflooding phase with intensive melt release, relocation and oxidation, notable differences between codes and simultaneously deviations from experimental results are observed. In the paper the main findings of the QUENCH-18 test are discussed, but the focus will be on the pre-test analyses of CODEX-AIT-3.

Verification and validation are basic quality assurance elements in code development and essential for code release. Therefore, the codes of AC(2) (ATHLET - ATHLET-CD - COCOSYS) are tested on separate effect tests, integral tests as well as plant scenarios to verify and validate the models after new implementation or updates. The verification assures that the models are implemented and working correctly while the validation checks if the models predict the right phenomena and combined with other models and modules. The selected experiments are summarized in GRS's validation matrices, which in turn are based on the CSNI validation matrices derived from OECD/WGAMA task groups as well as current activities on experimental test campaigns. For ATHLET several test series are used to cover a wide range of phenomena which can occur in PWR, BWR and VVER. Additionally, plant transients are considered for German LWR. The ATHLET-CD validation matrix contains experiments covering most phenomena which can occur during a severe accident. But due to the interaction of several effects even in small scale experiments mainly integral experimental campaigns are used for the validation. Over the last decades the validation of the AC(2) codeds ATHLET and ATHLET-CD has reached a high degree of fulfilment of GRS's validation matrices over all code versions. Innovative and advanced reactor concepts come with new or newly relevant phenomena, which AC(2) needs to provide models for. Extending the validation base of AC(2) for these models is one challenge for further code validation efforts besides the on-going update of the validation basis to recent code versions.

This paper illustrates the results obtained in the last phase of the NACIE-UP benchmark activity foreseen inside the EU SESAME Project. The purpose of this research activity, performed by system thermal-hydraulic (STH) codes, is finalized to the improvement, development and validation of existing STH codes for Heavy Liquid Metal (HLM) systems. All the participants improved their modelling of the NACIE-UP facility, respect to the initial blind simulation phase, adopting the actual experimental boundary conditions and reducing as much as possible sources of uncertainty in their numerical model. Four different STH codes were employed by the participants to the benchmark to model the NACIE-UP facility, namely: CATHARE for ENEA, ATHLET for GRS, RELAP5-3D (c) for the "Sapienza" University of Rome and RELAP5/Mod3.3(modified) for the University of Pisa. Three reference tests foreseen in the NACIE-UP benchmark and carried out at ENEA Brasimone Research Centre were analysed from four participants. The data from the post-test analyses, performed independently by the participant using different STH codes, were compared together and with the available experimental results and critically discussed.

The in-vessel melt retention by flooding the reactor vessel externally is regarded as an effective severe accident management (SAM) strategy. According to this strategy, the corium will be stabilized within the lower head, by transferring the decay heat through the wall into the containment via external cooling. One key question of this strategy is how the melt pool heat transfer reacts to different external cooling conditions. In this paper, the melt’s thermal–hydraulic behavior under different external cooling conditions is studied experimentally in two LIVE tests performed in the frame of the LIVE program investigating late in-vessel melt pool behavior and calculated with the lower head module AIDA of ATHLET-CD. One LIVE test was performed under nucleate boiling condition, the other under sub-cooling condition. Melt temperature, heat flux along the curved vessel wall and the crust behavior are described in transient and steady states. The simulation results have been compared with the experimental results. The results have been demonstrated the applicability of ATHLET-CD to investigate the SAM strategy in-vessel melt retention by external cooling. Furthermore, on the basis of the experimental results the modelling of heat transfer between corium and coolant has been improved.

Spent fuel pools (SFPs) are large structures equipped with storage racks designed to temporarily store irradiated nuclear fuel removed from the reactor. SFP severe accidents have long been considered as highly improbable since the accident progression is slow (in comparison with reactor core accidents) and let time to corrective operator actions. However, the accident at the Fukushima Dai-ichi Nuclear Power Plants has highlighted the vulnerability of nuclear fuels that are stored in SFPs in case of prolonged loss-of-cooling accidents and consequently renewed international interest in the safety of SFPs. In this context, the AIR-SFP project, funded by the Euratom 7th FP in the frame of the NUGENIA+ project, was launched in May 2015 with 15 participants. One of the objectives was to assess the applicability of Severe Accident (SA) codes, which were initially developed for reactor applications, to the calculation of transients in SFPs. To reach this objective, a benchmark, including a criticality risk assessment, was carried out. The degradation progression was computed by 14 participants with 6 different SA codes and 5 have participated to the criticality risk assessment. Main results are presented as well as conclusions that have been drawn concerning SA codes readiness to address these "beyond-scope" scenarios. (C) 2018 Elsevier Ltd. All rights reserved.

The accident at the Fukushima Dai-ichi Nuclear Power Plant has highlighted the vulnerability of nuclear fuels that are stored in spent fuel pools (SFP) before their evacuation and final disposal or possible reprocessing. This vulnerability is due to the potential loss of sufficient fuel cooling in case of internal events or of extreme external events such as earthquake or flooding. Further, the limited number of barriers containing the radioactive products, the fuel cladding is typically the only barrier in an SFP whereas there are three physical barriers when the fuel is in the reactor (the fuel cladding, the envelope of the primary circuit and the containment building), induces higher risks of radioactive releases to the environment if sufficient fuel cooling cannot be recovered.This paper presents the results of various assessments of SFP accidents performed with different severe accident codes for different SFP geometries. In these studies, both loss of cooling and loss of water transients were analyzed. These calculations have been performed in the framework of the SARNET2 project. The analysis of these different SFP scenarios was conducted to identify some limits and needs for improvement of the SA code developed and usually used for reactor applications. More specifically, some questions about the SA code capabilities to evaluate the impact of air flow on coolability of the fuel assemblies (FAs) and the rate and path of air/steam flow in these FAs, have arisen. This work has also enabled us to identify the need to improve the knowledge and the models related to the effect of mixed air/steam atmospheres, especially the role of nitrogen on the acceleration of the corrosion rate of Zircaloy claddings and on the degradation of their mechanical properties. (C) 2014 Elsevier Ltd. All rights reserved.

Boron carbide is used in many nuclear power plants like BWR, VVER, some PWR, and EPR as a neutron absorber material. Consequently, it is important to assess its role in the core degradation phenomena during a severe accident (SA), as well as that of the carbon gas released from its degradation on the fission products behaviour. This paper describes the progresses achieved in the frame of the Network of Excellence SARNET concerning the B4C control rod degradation modelling in Severe Accident codes, such as ATHLET-CD, ICARE2, ASTEC MELCOR and MAAP. These new developments complete improvements made during the European Union 5th Framework COLOSS project.Starting from basic modelling derived from available tests reported in the literature, large improvements of the kinetic correlation for B4C oxidation were obtained from analytical experiments performed at FzK (Germany) and IRSN (France), mostly in the temperature range above 1400 K.The new modelling was considered in the analysis of experiments involving a B4C control rod in small fuel rod assemblies, such as Phebus FPT3 in-pile experiment, as well as out of pile experiments Quench 07 and 09, aimed at studying the course of severe accidents. Regarding the hydrogen generation, the results given by different code simulations are consistent with the experimental values. Concerning the control rod degradation, SA codes such as ICARE2 and ATHLET-CD, using suitable modelling of B4C oxidation, predicted rather well the total carbon release. The results of the MELCOR code, involving initially a B4C oxidation model designed to be used for BWR control blades, have been largely improved in the most recently released version, with a model extended for PWR B4C control rods.Codes still have some difficulties to reproduce the final degradation of fuel bundles involving B4C rods. Spreading of molten materials from the control rods onto fuel rods of the bundle is suspected, suggesting that the main effect of the B4C control rod materials on the bundle behaviour during degradation is connected with B4C-Stainless Steel (SS) eutectics formation and B4C-SS-Zry liquid mixture relocation. These phenomena are not accounted for in the SA codes. The need for further code developments of the early phase of core degradation is recognized, involving the absorber rod material behaviour. The BECARRE experiments, on-going in the framework of the International Source Term Program, are designed to provide in-depth understanding of these phenomena and help improving their modelling. (C) 2009 Elsevier Ltd. All rights reserved.

In den QUENCH-Versuchen wird der Wasserstoffquellterm bei der Einspeisung von Notkuhlwasser in einen trockenen, uberhitzten Reaktorkern eines Leichtwasserreaktors untersucht. Ferner wird in den Versuchen das Verhalten von uberhitzten Brennelementen unter verschiedenen Flutbedingungen untersucht, eine Datenbasis zur Modellentwicklung und eine Weiterentwicklung von Rechenprogrammen zu Schweren Storfallen (engl. SFD – Severe Fuel Damage) erstellt. Der Ausdampf-Versuch QUENCH-11 wurde am 8. Dezember 2005 durchgefuhrt. Es war das zweite Experiment im Rahmen des EU-geforderten LACOMERA-Programms. Es sollte einen Kuhlmittelpumpenausfall wahrend eines Kuhlmittelverluststorfalls (hier ein sog. Small Break LOCA) oder einer plotzlichen Stromabschaltung (eng. „station blackout“) mit einer spaten Druckentlastung des Primarsystems simulieren. Verbunden mit dem Unfallszenario ist das Ausdampfen eines teilgefullten Reaktorkerns bzw. des Versuchsbundels. Das Ziel war die Untersuchung des Bundelverhaltens wahrend des Ausdampfens und des nachfolgenden Abschreckens mit reduzierter Wassereinspeiserate. Es war das erste Experiment, in dem der gesamte Unfallablauf von der Ausdampfphase bis zur verzogerten Flutung des Bundels bei einer geringen Wasser-Einspeiserate untersucht werden sollte. Das Ausmas der Wechselwirkungen von Thermalhydraulik und Materialien war in dem Experiment ausgepragter als in fruheren QUENCH-Versuchen. Das Experiment wurde von INRNE Sofia (Bulgarische Akademie der Wissenschaften) vorgeschlagen und zusammen mit dem Forschungszentrum Karlsruhe definiert. Nach dem Experiment wurde entschieden, die QUENCH-11-Daten fur ein Rechenprogramm-Benchmark, bei dem die Rechenergebnisse mit den experimentellen Daten verglichen werden, im Rahmen des Europaischen Exzellenz-Netzwerks SARNET anzubieten, um die Zuverlassigkeit der Rechnungen fur die verschiedenen Phasen von Unfall bzw. Experiment zu uberprufen. Die eingesetzten SFD-Rechenprogramme waren ASTEC, ATHLET-CD, ICARE-CATHARE, MELCOR, RATEG/SVECHA, RELAP/SCDAPSIM, und SCDAP/RELAP5. Die Koordination fur den Vergleich ubernahm INRNE. Als Grundlage fur den Vergleich dienten die zeitlichen Verlaufe von Temperaturen, Wasserstofferzeugung und anderer wichtiger Daten. Auserdem wurden Axialprofile, in erster Linie die der Temperatur zum Zeitpunkt des Flutbeginns und des Endstadiums, d. h. bei der Testzeit von 7000 s, verglichen. Fur die meisten Rechenergebnisse kann ein gemeinsamer Trendverlauf angegeben werden. Grosere Unterschiede zeigen die Ergebnisse fur die Wasserstofferzeugung und die zugehorige Oxidschichtdicke. Der Grad der Ubereinstimmung zwischen Rechnung und Experiment wird von den Schwachstellen der Rechnung und des Experiments gleichermasen mitbestimmt. SFD-Rechenprogramme sind zur Analyse von typischen Kernreaktorunfallen entwickelt worden. Einzelne Besonderheiten der experimentellen Anordnung integraler Experimente (wie auch QUENCH-11) sind bedingt durch das Vorhandensein von Dampffuhrungsrohr (Shroud) und Elektrodenmaterial fur die elektrische Stabheizung nicht reaktortypisch und konnen daher nicht in der gewunschten Einzelheit im Rechenprogramm nachgebildet werden. Hinzu kommen Effekte durch den Anwender. Da jedoch die Bandbreite der wesentlichen Rechenergebnisse einschlieslich der Wasserstofferzeugung nicht extrem gros ist, kann das Ergebnis des SFD-Rechenprogramm-Benchmarks insgesamt als positiv bewertet werden. Ein Vergleich mit anderen Experimenten zeigt einen weiteren Bedarf an Verbesserungen besonders im Hinblick auf die Oxidation stark zerstorter Bundelstrukturen wahrend des Flutens. Zusatzlich erwies sich das Rechenprogramm-Benchmark fur einige Programmanwender als wertvoll, um sich mit den physikalischen Problematiken und der Anwendung von grosen SFD-Rechenprogrammen vertraut zu machen. Es dient dem Erfahrungsaustausch mit jungeren Wissenschaftlern und Ingenieuren und der Aufrechterhaltung des Standards der nuklearen Sicherheit.

The QUENCH out-of-pile experiments at Forschungszentrum Karlsruhe (Karlsruhe Research Center) are set up to investigate the hydrogen source term that results from the water or steam injection into an uncovered core of a Light-Water Reactor, to examine the behavior of overheated fuel elements under different flooding conditions, and to create a database for model development and improvement of Severe Fuel Damage (SFD) code packages. The boil-off experiment QUENCH-11 was performed on December 8, 2005 as the second of two experiments in the frame of the EC-supported LACOMERA program. It was to simulate ceasing pumps in case of a small break LOCA or a station blackout with a late depressurization of the primary system, starting with boil-down of a test bundle that was partially filled with water. It is the first test to investigate the whole sequence of an anticipated reactor accident from the boil-off phase to delayed reflood of the bundle with a low water injection rate. The test is characterized by an interaction of thermal-hydraulics and material interactions that is even stronger than in previous QUENCH tests. It was proposed by INRNE Sofia (Bulgarian Academy of Sciences) and defined together with Forschungszentrum Karlsruhe. After the test, QUENCH-11 was chosen as a SARNET code benchmark exercise. Its task is a comparison between experimental data and analytical results to assess the reliability of the code prediction for different phases of an accident and the experiment. The SFD codes used were ASTEC, ATHLET-CD, ICARE-CATHARE, MELCOR, RATEG/SVECHA, RELAP/-SCDAPSIM, and SCDAP/RELAP5. The INRNE took responsibility as benchmark coordinator to compare the code results with the experimental data. As a basis of the present work, histories of temperatures, hydrogen production and other important variables were used. Besides, axial profiles at quench initiation and the final time of 7000 s, above all of temperatures, are presented. For most variables a mainstream of computational results can be defined. Larger discrepancies are seen in the hydrogen production and the related oxide scale thickness. Analysis shows that the agreement between calculated and experimental data is determined by both, limitations of severe accident codes and of the experiment. Severe accident codes are intended and developed to analyze typical accident situations in nuclear reactors. Special features of the experimental set-up of integral tests like QUENCH-11 as the presence of a shroud and electrode materials for the electric heating are irrelevant for reactors and cannot be simulated in the desirable detail. User effects add to the problems. However, a limited bandwidth of some calculated mainstream results, including hydrogen production, is a good outcome of the code benchmark. Taking in view other experiments, a further demand for an improvement concerning the oxidation of severe damaged structures during a reflood scenario is seen. Additionally, the benchmark proved to be valuable for a number of participants to become acquainted with the physical problems and with the application of large severe accident codes. For the transfer of knowledge and experience to younger scientists and engineers, this is an important issue to maintain the standard of nuclear safety.

Within SARNET, the corium topic covers all the behaviors of corium from early phase of core degradation to in or ex-vessel corium recovery with the exception of corium interaction with water, direct containment heating and fission product release. The corium topic regroups in three work packages the critical mass of competence required to improve significantly the corium behavior knowledge. The spirit of the SARNET networking is to share the knowledge, the facilities and the simulation tools for severe accidents, so to reach a better efficiency and to rationalize the R&D effort at European level. Extensive benchmarking has been launched in most of the areas of research. These benchmarks were mainly dedicated to the recalculation of experiments, while, in the next periods, a larger focus will be given to integral experiments or reactor applications. Eventually, all the knowledge will be accumulated in the ASTEC severe accident simulation code through physical model improvements and extension of validation database. This paper summarizes the progress that has been achieved in the frame of the networking activities. A special focus is placed on the melt pool and debris coolability and corium-concrete interaction, in which, the effects due to multidimensional geometries and heterogeneities has been shown, during SARNET, to play a crucial role and for which further research is still needed.

The QUENCH experiments at the Karlsruhe Research Center are set up to investigate hydrogen source term generation that results from the water or steam injection into an uncovered core of a Light-Water Reactor (LWR), to examine the behavior of overheated fuel elements under different flooding conditions, and to create a database for model development and improvement of Severe Fuel Damage code packages. The test QUENCH-11 was conducted at the Karlsruhe Research Center on 8 December 2005 as the second of two experiments in the frame of the EC-supported LACOMERA program. The experiment is focused on studying the bundle behavior during boil-off and subsequent quenching at low water injection rate. The test was proposed by INRNE - Sofia, and defined together with the Karlsruhe Research Center. A SARNET benchmark problem is defined as a comparison between QUENCH-11 experimental data and analytical results received from the different computer codes such as ASTEC, ICARE-CATHARE, MELCOR, RELAP5/SCDAP, RELAP/SCDAPSIM, RATEG/SVECHA, and ATHLET-CD. In total ten participants from seven countries participated in this exercise.

An assumed failure of the main and the emergency core cooling systems of a light water reactor in the course of a hypothetical loss-of-coolant accident can cause core meltdown. In a scenario of this kind, the reactor pressure vessel would be reflooded as an accident management measure in order to quench the reactor core components. The QUENCH series of experiments are carried out at the Karlsruhe Research Center to studybundle behavior during reflooding, with the main emphasis being put on the analysis of the H-2 source term. On December 8, 2005, the QUENCH-11 experiment was performed within the framework of the EU-funded LACOMERA project, an evaporation experiment with subsequent reflooding at a low water feed rate. QUENCH-11 was carried out as a "semi-blind" code benchmark in the European SARNET NoE. The article presents the work performed with the ATHLET-CD accident analysis code at the Chair for Energy Systems and Power Economy of the Bochum Ruhr University. A brief introduction is followed by a description of the facility and the experiment before an outline is presented of the modeling performed with ATHLET-CD. Next, the results of computations are discussed and evaluated. It is seen that the code allows the experiment to be simulated in a good approximation to the measured results. Further development work is seen to be necessary in the area of H-2 production during reflooding.

Experiment QUENCH-18 on air ingress and aerosol release was successfully conducted at KIT on 27 September 2017. This test was performed in the frame of the EC supported ALISA programme. It was proposed by XJTU Xi’an (China) and supported by PSI (Switzerland) and GRS (Germany). The primary aims were to examine the oxidation of M5® claddings (OD=9.5 mm, wall thickness 570 µm) in air/steam mixture following a limited pre-oxidation in steam, and to achieve a long period of oxygen and steam starvations to promote interaction with the nitrogen. QUENCH-18 was thus a companion test to the earlier air ingress experiments, QUENCH-10 and -16 (in contrast to QUENCH-18, these two bundle tests were performed without steam flow during the air ingress stage). Additionally, the QUENCH 18 experiment investigated the effects of the presence of two Ag/In/Cd control rods on early-phase bundle degradation (companion test to the QUENCH-13 experiment), and two pressured unheated rod simulators (60 bar, He). The low pressurised heater rods (2.3 bar, similar to the system pressure) were Kr-filled. In a first transient, the bundle was heated from the peak cladding temperature Tpct ≈ 900 K in an atmosphere of flowing argon (3 g/s) and superheated steam (3.3 g/s) by electrical power increase to the peak cladding temperature of Tpct ≈ 1400 K. During this heat-up (with the heat-up rate 0.3 K/s), claddings of the two pressurised rods were burst at temperature of 1045 K. The attainment of Tpct ≈ 1400 K marked the start of the pre-oxidation phase to achieve a maximum cladding oxide layer thickness of up to 120 µm. Then the power was reduced from 9 to 3.8 kW (simulation of decay heat) which effected a cooling of the bundle to Tpct ≈ 1080 K, as a preparation for the air ingress phase. In the subsequent air ingress stage, the steam flow was reduced to 0.3 g/s, the argon flow was reduced to 1 g/s, and air was injected with the flow rate of 0.2 g/s. The change in flow conditions had the immediate effect of reducing the heat transfer so that the temperatures began to rise again. After some time measurements demonstrated a gradual increasing consumption of oxygen. The first Ag/In/Cd aerosol release was registered at Tpct = 1350 K and was dominated by Cd bearing aerosols. Later in the transient, a significant release of Ag was observed along with continued Cd release, as well as a small amount of In. In contrast to the QUENCH-16 test (performed with the air ingress stage without steam flow), oxidation of bundle parts in steam caused release of additional chemical energy (power about 4 kW) and consequently acceleration of bundle heat-up. A strong temperature escalation started in the middle of the air ingress stage. Later a period of oxygen starvation was occurred and was followed by almost complete steam consumption and partial consumption of the nitrogen, indicating the possibility of bundle. Following this the temperatures continued to increase and stabilised at melting temperature of Zr bearing materials until water injection. The total uptakes of oxygen, steam and nitrogen were 100±3, 450±10 and 120±3 g, respectively. During the starvation period a noticeable production (about 25 mg/s, totally 45±1 g) of hydrogen was measured. Almost immediately after the start of reflood there was a temperature excursion in the mid to upper regions of the bundle, leading to maximum measured temperatures of about 2450 K. Reflood progressed rather slowly and final quench was achieved after about 800 s. A significant quantity of hydrogen was generated during the reflood (238±2 g). Nitrogen release (>54 g) due to re-oxidation of nitrides was also registered.