Publications Repository - Helmholtz-Zentrum Dresden-Rossendorf

We studied the influence of humic acid (HA) on the uranium(VI) sorption onto the rock material phyllite and onto its main mineralogical constituents quartz, muscovite, chlorite, and albite at an ionic strength of 0.1 M in the pH range of 3.5 to 9.5 under aerobic conditions. The uranium(VI) concentration was 1·10-6 M and the HA concentration was 5 and 60 mg/L, respectively. The mass loading was 12.5 g/L. Furthermore, we studied the uranium and HA sorption on ferrihydrite (3·10-4 M Fe) and compared the results to the sorption behavior of phyllite. The study showed that the uranium sorption onto phyllite and onto its mineralogical constituents is influenced by the pH-dependent sorption behavior of the HA. Due to high HA sorption onto the solids in the acidic pH range the uranium uptake is enhanced compared to the uranium uptake in the absence of HA. A high concentration of dissolved HA in the near-neutral pH range may reduce the uranium sorption due to formation of aqueous uranyl humate complexes. Furthermore, we could show that the high uranium and HA sorption on phyllite is primarily caused by minor amounts of the secondary mineral ferrihydrite that is formed due to weathering of phyllite. Thus, the ferrihydrite predominates the contributions of the main minerals quartz, muscovite, chlorite, and albite, that are naturally present in the rock material phyllite.

FOREWORD

The microstructural evolution of reactor pressure vessel (RPV) steel effected by neutron irradiation is not yet sufficiently understood. The issue, however, is highly relevant for a safe operation of nuclear power plants (NPP).
Investigations have been performed by the methods of analytical high-resolution transmission electron microscopy (AHRTEM), the small angle neutron scattering (SANS) and the anomalous small angle X-ray scattering (ASAXS). The materials selected represent two different state-of-the-art products of VVER-type RPV steels.
One material is typical for the first VVER-440/230 generation and has a quite high content of impurities. The second material consists of surveillance specimens manu factured by Skoda Pilsen for the VVER 440/213 NPP in Paks (Hungary). The materials were irradiated to fluences of 7 to 14x1023 n/m2 (E > 1 MeV) within a high flux channel of the VVER prototype VVER-2 in Rheinsberg (Germany), or in the surveillance position of the Paks NPP, respectively. Whereas the detection of irradiation-induced precipitates by AHRTEM was not successful in every case, SANS showed an increase in the scattering intensity after irradiation, which could be interpreted as caused by a bimodal size distribution of irradiation-induced precipitates. A high first peak of the distribution appears at a radius of the precipitates of 1 to 2 nm. This peak disappears after annealing and is especially high but also thermally unstable for charges with high copper content. ASAXS measu rements prove that these precipitates are vanadium rich. There is a clear correlation between the integral of the sieze distribution within the range of the first peak of the distribution curve and the hardness. The correlation is independent of the treatment or the fluence. This shows the fine-dispersed nano-scale precipitates are the reason of the irradiation hardening.

In dickwandigen Teilen aus den Druckbehälterstählen ASTM A533 B Cl. 1, 15CrMoV2 und 15CrNiMo2 wurde die Tiefenlagenabhängigkeit der Zähigkeit mit Kerbschlagbie geversuchen und mit quasistatischen 3-Punkt-Biegeversuchen an ermüdungsangeris senen Proben in Charpy-V-Geometrie nach dem ¿master curve¿-Konzept nach Wallin bestimmt. In bestimmten Tiefenbereichen treten Anomalien im Verlauf der Tiefenlagenabhängigkeit der mit den genannten Methoden ermittelten Übergangstemperaturen auf. Eine eingehendere Analyse zeigt, dass die Streuungen in den Zähigkeitskennwerten bimodal verteilt sind. Als Ursache werden Seigerungslinien ausgewiesen, in denen vorzugsweise der Kohlenstoffgehalt angereichert ist. Als Folge dessen entsteht in diesem Bereichen ein deutlich härteres, martensitisches Gefüge.
Die Lage der Linien relativ zur Rissspitze bestimmt die Bruchzähigkeit und damit die Referenztemperatur To nach dem ¿master curve¿-Konzept. Im Charpy-Kerbschlagbie geversuche ist eine derartige eindeutige Zuordnung nicht nachzuweisen.

The irradiation-induced microstructural changes in surveillance materials of the VVER
440-type weld metal Sv-10KhMFT were investigated by small angle neutron scattering
(SANS) and anomalous small angle X-ray scattering (SAXS). Due to the high fluence
a strong effect was found in the SANS experiment. No significant effect of the irradiationis detected by SAXS. The reason of this discrepancy is the different
scattering contrast of irradiation-induced defects for neutron and X-rays. An analysis
of the small angle X-ray scattering shows that the scattering intensity is mainly caused by vanadium-containing precipitates and grain boundaries. Both types of scattering defects are hardly changed by irradiation. Neutron irradiation rather produces additional scattering defects of few nanometers in size. Assuming these defects are clusters containing copper and other foreign atoms with a composition according to results of atom probe field ion microscopy investigations, both the high SANS and the low SAXS effect can be explained.

The knowledge about properties of the anisotropic turbulence in liquid metal flows exposed to an external magnetic field is essential for several technological applications such as the concept of a self-cooled liquid metal blanket for thermonuclear fusion reactors or the electromagnetic brake in metallurgical processes, but also of basic interest for turbulence research. Flow arameters like pressure drop and heat transfer rate are essentially determined by the interaction between the external magnetic field and the liquid metal flow. Several experimental studies [1-3] revealed that the application of a magnetic field leads not exclusively to a suppression of the turbulent perturbations. Velocity fluctuations remain and demonstrate a distinct anisotropy of the MHD turbulence showing a tendency to become two-dimensional. The properties of the local mass transfer are strongly
governed by the anisotropic character of the flow [4,5]. In this context a number of questions arises regarding the origin, the decay time or the size of the turbulent elements. A typical feature of the two-dimensional turbulence is that the energy becomes concentrated in organised large scale fluctuations. This fact ensures an intensive heat transfer on longer distances. Because of the anisotropy of the electromagnetic dissipation term vortices will be scarcely damped over long distances if their axes are aligned with the magnetic field lines [6]. For practical applications it is important to identify possibilities to promote the formation of such quasi-two-dimensional vortices in order to control the heat or mass transfer rate of the flow. It is well-known that the mentioned perturbations can be generated by rather specific means, for example, mechanical inserts or electrical currents between definitely arranged electrodes, etc..
In our experiments we force the turbulence intensity by mechanical means employing a grid of cylindrical bars or flat stripes. Due to the favourable material properties of the used liquid sodium we are able to extend the measurements into the region of high interaction parameters. In this way the obtained results complement the already existing knowledge about the two-dimensional MHD turbulence. The results concerning the measured turbulence intensity as a function of the interaction parameter N and the corresponding power spectral densities will be presented. Electrical potential probes were used to determine the velocity fluctuations. The measurements took place in two different test sections allowing either the variation of the axial distance between the turbulence promoter and the probe or the installation of special channel inserts in order to suppress the generation of the M-profile of the local velocity and to decrease the wall conductance ratio. The results obtained in both test sections will be compared and the influence of the altered boundary conditions onto the turbulent structure of the flow will be discussed.

In the direct melt extraction process a quickly rotating wheel draws fibres out of a liquid metal pool which may contain any pure metal or alloy. One common application of such fibres is the production of highly porous metallic substrates, which, in the case of Ni-Al is ideally suited to serve as a highly heat- and corrosion-resistant catalyst carrier. But this demands the fibres to have diameters in the range of 50 to 100 microns and a small width distribution of their cross section. The main limitations of this process are due to turbulence within the melt and a wavy motion on the surface which both inhibit to fulfil these requirements.
Many patents are concerned with the stabilisation of the melt pool or at least of the contact region between metal surface and the extraction wheel. They propose submersing mechanical parts directly into, or placing them at least in close vicinity to the melt to avoid the non-stationary conditions produced by the turbulence within the liquid metal pool. Even made of heat resistant ceramics they suffer from corrosion or cracks and do not work reliable at all. The present work describes a non invasive control mechanism by means of magnetic fields of different strength and orientation. They can be applied either globally to damp the flow within the whole melt volume or locally to the meniscus region where the fibre is formed, to reach higher Lorentz forces in this very sensitive region. Model experiments (In-Ga-Sn, liquid at room temperature; no extraction) with global stabilisation were carried out to study the calm down of the turbulent surface of an inductively stirred melt. In a second setup the influence of this globally applied field on the fibres have been investigated (low melting Sn-Pb). Though the otherwise wavy surface was damped down to a nearly mirror-like plane the goal of significant smaller fibre cross-section was not reached by the global stabilisation alone. The clearly observable positive tendency towards smaller fibres is presently investigated with two series of model experiments with local stabilisation. The smaller volume which needs to be magnetised allows for a 5 times stronger field which can be achieved either by rare earth permanent magnets or a concentration of the globally applied field with magnetic field guides.

Electromagnetic levitation is a well-known technique for containerless processing of metals and alloys both in the solid and in the molten state. In experiments the levitated bodies often times show different types of instabilities resulting in a rotating and oscillating motion of the sample. In the paper we analyze the physical reasons for such spontaneous instabilities, and conclude on possible measures against it. The first type of instability considered is that due to possible coupling between the electric current passing through the magnetic system and the variation of position of the levitated body. It is shown that oscillatory motion of the body may be unstable if the electric power supply regime permits the electric current passing through the magnetic system to vary depending on its effective inductance. Another type of instabilities may occur because of the coupling between the motion of the body and the electric currents induced in the body itself. This effect may cause a spontaneous spin-up of the spherical body, occuring if the magnetic field frequency exceeds a certain critical threshold depending on the configuration of the field. It is shown that this type of linear instability may be completely suppressed by imposing a steady magnetic field of a strength comparable to that of the oscillating one. As a third type of instability we consider small vibrations of the spherical body occuring due to the finite diffusion time of the magnetic field into the body resulting in a delay of the induced currents with respect to the body position. Again, such vibrations may occur if the non-dimensional field frequency exceeds some specific threshold. This threshold and the
underlaying mechanism will be explained. The theoretical predictions are verified by model experiments using solid Al or Mg spheres. These instabilities can obviously be avoided by system parameters below the corresponding non-dimensional reshold. If this is not possible in reality, an active damping method using DC magnetic fields has been developed, too. The DC field can either be produced by permanent magnets or by an electromagnetic superposition to the levitation coils. Experiments will be shown demonstrating the stabilization effect due to the DC magnetic fields.

Vertical two-phase flow is normally classified into the four basic flow regimes bubble, slug, churn and annular flow. The transition between the different flow regimes does not occur suddenly and many extensions of this classification can be found in literature. For the prediction of flow patterns empirical and theoretical flow pattern maps have been developed.
A new criterion is presented for the transition between bubble and slug flow which is based on local instantaneous conductivity measurements with a wire mesh sensor (1 kHz that means 1000 frames per second, 242 measuring points in a tube cross section area). The high resolution allows the calculation of particle size distributions.
The transition from homogeneous bubble flow to heterogeneous bubble flow is indicated by the appearance of a bimodal bubble size distribution. If the equivalent bubble diameter exceeds the tube diameter the transition from bubble to slug flow occurs. The new criterion is compared with different empirical (Govier & Aziz, Weisman & Kang) and theoretical flow maps (Taitel, Bornea & Dukler, Ishii & Mishima) and shows a good agreement.

The SWR1000 is a new innovative boiling water reactor concept, which is developed by Siemens AG. This concept is characterized in particular by passive safety systems (e.g. four emergency condensers, four building condensers, eight passive pressure pulse transmitters, six gravity-driven core flooding lines). In the framework of BWR Physics and Thermohydraulic Complementary Action (BWR-CA) to the EU BWR R&D Cluster emergency condenser tests were performed by Forschungszentrum Jülich at the NOKO test facility. In this paper post test calculations with ATHLET are presented, which aim at the determination of the removable power of the emergency condenser and its operation mode. The 1D thermal-hydraulic code ATHLET was extended by the module KONWAR for the calculation of the heat transfer coefficient during condensation in horizontal tubes. In addition, results of CFD-calculations using the code CFX-4 are presented, which investigate the natural convection during the heat up process at the secondary side of the NOKO test facility.

In many technologies such as the refinement of metallic melts the injection of gas bubbles is used to drive some liquid motion, enhance transport processes or to control the rate of chemical reactions. The resulting flow structure strongly depends on two phase flow parameters such as bubble size, bubble distribution or the local void fraction. Magnetic fields can be used to control the characteristics of a liquid metal bubbly flow.
We present experimental investigations of the bubble formation in heavy liquid metals as well as the influence of external magnetic fields on the turbulent dispersion of gas bubbles and the slip ratio in liquid metal bubbly flows, respectively.

a) Bubble formation
If gas bubbles are injected into a liquid metal characterised by a large surface tension one should be care to get a good wetting between the fluid and the surface of the gas injector. Otherwise, the gas would try to spread out along this interface to form gas layers. A control of the bubble size and formation rate becomes difficult. The comparison between experiment and theoretical models describing bubble formation processes requires an ideal wetted gas injector.
The bubble formation in mercury and the eutectic alloy InGaSn has been studied by means of several methods of gas injection, for instance through single orifices or injectors made from sintered metals with a mean porosity of a few microns. X-ray measurements have been used to directly observe the resulting gas bubbles rising in the liquid metal. In the case of an single orifce the influence of electromagnetic forces on the bubble frequency has been demonstrated.

b) Turbulent bubble dispersion, slip ratio
The transport properties of small argon bubbles have been studied in turbulent upwards channel flows of sodium and mercury. The bubbles were injected by a single orifice located in the centre of the channel cross section. After a distinct distance the local void fraction and the bubble velocity has been measured by means of electrical resistivity probes. The flow has been exposed to external magnetic fields directed transverse or longitudinal to the mean flow direction.
We will present and discuss measuring results showing the effect of the magnetic field strength and direction on the horizontal gas distribution and the ratio between gas and liquid velocity.

Model experiments are an important tool to understand the details of the flow structure and the transport properties of
flows occurring in real-scale metallurgical facilities as well as to validate the multitude of numerical codes for flow
simulation. Generally, water experiments are often performed to characterise the flow. Nevertheless, due to the large
differences of material properties like density, heat conductivity, surface tension or electrical conductivity the use of liquid
metals has clearly to be preferred if heat transfer phenomena, two-phase flows or the influence of electromagnetic fields
on the flow should be investigated. The application of suitable alloys with low melting points, for instance PbBi (Tm = 125
°C) or InGaSn (Tm = 5...10 °C), makes such kind of experiments very flexible and offers the ability to measure the
essential flow quantities like velocity, pressure or void fraction.

In this lecture we want to discuss a novel sensor which has been developed by the Rossendorf group to measure the local
velocities in opaque liquid flows. The measuring principle is based on the separation of a direct mechanical interaction
between flow and sensor tip and the optical acquisition and processing of the signal. In principle, this fact allows the
extension of the range of applicability to higher temperatures. Furthermore, the insensitivity of the system to electrical
noise and external magnetic fields can be considered as an important advantage. Until now, the sensor has been tested in
metallic melts up to temperatures of about 350 °C . In principle, an extension of the range of application up to
temperatures of about 1100 °C should be possible by the utilisation of quartz glass as material for the sensitive sensor tips.
The first sensors have been manufactured and tested with low temperature melts.

We present measurements of the local velocity obtained in an eutectic InGaSn melt driven by a rotating magnetic field.
The interest is focussed on geometry and parameters relevant for crystal growth technologies and mixing processes in
metallurgical applications, respectively. Measured profiles of the azimuthal velocity have been obtained at different
frequencies and field amplitudes.

Das Institut für Sicherheitsforschung ist eines von fünf wissenschaftlichen Instituten des Forschungszentrums Rossendorf e. V. (FZR). Die Rossendorfer Sicherheitsforschung baut auf langjährige Erfahrungen zur Analyse von Druckwasserreaktoren des russischen WWER-Types auf. WWER-Reaktoren nehmen bis heute einen wichtigen Platz im Spektrum des Institutes ein. So unterstützt das Institut im Rahmen eines Programmes des Bundesministeriums für Umwelt, Naturschutz, und Reaktorsicherheit die ukrainische Atomaufsichtsbehörde beim Aufbau einer modernen KKW-Überwachung. Inzwischen hat das Institut ebenso Kompetenz zur Sicherheitsforschung für westliche Leichtwasserreaktoren aufgebaut und kooperiert mit deutschen, europäischen und amerikanischen Forschungseinrichtungen und der Industrie.
Die Schwerpunkte der Sicherheitsforschung in Rossendorf liegen auf der Modellierung thermohydraulischer Phänomene bei unterstellten Störfallabläufen, der Reaktorphysik und dem Strahlungstransport und dem Verhalten von Komponenten und Materialien.
Der erste Teil des Forschungsberichtes beschäftigt sich ausschließlich mit den thermohydraulischen Arbeiten des Institutes wie den Vermischungsvorgängen im Primärkreis bei Borverdünnungs- und Kaltwassertransienten sowie der Modellentwicklung für transiente Zweiphasenströmungen. Für die thermohydraulischen Arbeiten hat das Institut ein 1:5 Modell eines deutschen DWR aufgebaut, die Experimente werden mit CFD-Codes (Computational Fluid Dynamics) nachgerechnet.

Model experiments are an important tool to understand the details of the flow structure and the transport properties of flows occurring in real-scale metallurgical facilities as well as to validate the multitude of numerical codes for flow simulation. The application of suitable alloys with low melting points, for instance PbBi (T = 125 °C) or InGaSn (T = 5...10 °C), guarantees realistic flow parameters and also offers the ability to measure essential flow quantities like velocity, pressure or void fraction. Two approaches will be discussed to measure the local velocities in opaque liquid flows such as liquid metals: a mechano-optical sensor and the ultrasonic Doppler method, respectively. Today, with respect to problems like high temperatures or material compatibility both techniques reveal severe limitations considering velocity measurements in liquid steel or aluminium, however, their reliability at moderate temperatures has been demonstrated. Here, we present velocity measurements obtained by means of both methods for the case of a cylinder filled with an eutectic InGaSn melt driven by a rotating magnetic field.

There has not yet been generally accepted single specimen technique in order to obtain dynamic fracture mechanical parameters. A potential candidate could be the measurement of the acoustic emission (AE). Previous investigations showed that characteristic AE signals appears in the vicinity of crack initiation for the first time. With a specially developed measuring arrangement the method could be compared with simultaneously measuring additional physical parameters at specimens of strongly different levels of toughness and strength. Furthermore the results could be verified by the results of multiple specimens technique. With this technique the dynamic crack resistance curve was determined and the physical crack initiation toughness was deduced by measurement of the stretch zone width. To realize the idea the inverted impact pendulum developed and installed by VTT Espoo/Finland was modified with an AE transducer and a magnetic emission (ME) probe. The inverted impact pendulum design reduces the inertia and oscillation effects in the recorded load curve and allows to measure the crack opening (COD) by a laser scattering light method. In this way, at the same specimen crack opening displacement (COD), AE, and ME could be measured.
Neither of the three single specimen methods can directly detect crack initiation. Except the lower shelf region, were only a tiny amount of ductile crack growth occurs before cleavage, the stable crack initiation could be proven by AE and COD. Good agreement between initiation parameters based on SZW and AE was found for all toughness levels. In some cases a strong scattering of toughness parameters was, however, observed and the ranking did not correspond. COD indicates crack initiation after an amount of macroscopic stable crack growth, which is small but cannot be ignored. The correlation to the SZW is not independent of the toughness; for steels with high toughness the correlation is unsatisfied. Eventually, ME does not reflect stable crack initiation and growth.
Obviously, the physical connection between crack initiation and AE cannot be revealed. The AE method has been proved as very sensible for microscopically active processes, which coincide spatially and chronologically with ductile crack initiation. Thus, it is appropriate to estimate the dynamic fracture mechanical parameter "crack initiation toughness" better than other recently applied methods.

Additional self-interstitials were introduced into the vacancy-rich depth range around half of the projected ion range of high energy ion implanted silicon in order to balance radiation induced excess vacancies. The trapping of Cu atoms was found to be not effected by this procedure. Additional interstitial loops were formed during annealing if the concentration of self-interstitials exceeded a certain threshold. These dislocation loops did not act as gettering centres for Cu atoms. The results obtained are not in agreement with the assumption that excess vacancies are the gettering centres for metal impurities like Cu. Instead, it may be assumed that small clusters of self-interstitials act as gettering centres.

A neural network approach has been developed for localizing leakages and estimating the leak rate in the VVER-440 pressure vessel head. Results are presented from experiments with simulated leaks. Three-layer perceptron networks were found to be best suited for leak localization and for the estimation of leak rates. However, the estimation of leak rates required an additional neural network because a different normalization procedure was necessary for extracting features from RMS values of the acoustic emission sensors. Perceptron networks with continuously valued outputs corresponding to the coordinates of the leak positions were useful for indentifying even leak positions which had not been offered during training.

Das Patent beschreibt ein Verfahren zur Minimierung des Einflusses von Resonanzfluktuationen des Meßmediums bei Ultraschall-Durchschallungsuntersuchungen an in Rohrleitungen geführten Zwei- und Mehrphasenströmungen durch eine schnelle Variation der Arbeitsfrequenz des Senders der Meßeinrichtung.

For the design, optimisation and safety analysis of nuclear plants, the use of 1D system codes is state of the art. However, for the modelling of passive equipment the range of validity of the 1D-codes is exceeded, because in these structures 3D-flow phenomena are often dominating. The fast progress of the computer technique and of Computational-Fluid-Dynamics (CFD-) codes gives the opportunity, to apply CFD-codes to the assessment of the efficiency of those components. Most of these codes are based either on the finite-volume or the finite-element method.
Integrated simulation of thermo-hydraulic phenomena and the resulting thermo-mechanical loadings for the surrounding structures (e. g. the reactor pressure vessel) can be achieved using a finite-element-code like ANSYS/Multiphysics®. The finite volume code CFX® has been proven to be an appropriate tool for modelling of 3D single phase fluid phenomena.
The governing mechanism in passive components for decay heat removal is natural convection and heat transfer with internal heating. To assess the capability of both codes describing a mixed convection flow, post test calculations of an IAHR (International Association for Hydraulic Research) benchmark exercise were performed (Kamide et. al, 1991). The working fluid was water in the temperature range of 15 to 50°C at atmospheric pressures.
This paper presents a discussion of the problems and capabilities of each code to calculate complex flow regimes and temperature fields.