Publications Repository - Helmholtz-Zentrum Dresden-Rossendorf

The density of amorphous SiC layers formed by 2 MeV Si⁺ implantation into single crystalline 6H-SiC was measured by X-ray reflectometry and compared with the resultts of step height measurements. Reactive ion
etching was used to investigate the density as function of depth. The density of the as-amorphized SiC is about 12 % less than that of the crystalline material. Within the experimental accuracy the density reduction is homogeneous
over the whole layer thickness. Low temperature annealing leads to the formation of relaxed amorphous SiC with a density about 7 % below the crystalline one. These large density changes are in contrast to results in amorphous
Si. They can be explained by the high atomic density of SiC and the chemical disorder in the amorphous state of SiC.

A model which explains the characteristic distortions of the spectral lineshape of a Si detector during the recording of heavy ions is proposed. The appearence of a group of pulses in the right-hand (high-energy) region of the amplitude spectrum is associated with nonequilibrium-charge recombination. The recombination occurs in the skin layer of the detector structure, where charge transfer occurs by diffusion. The amplitude of the signal fluctuates because of the participation of Si recoil atoms, which generate electron-hole pairs at specific cross sections of the "input-window" of the detector. Heavy particles directly produce ionization uniformly over depth, which results in definite losses of charge but is not accompanied by appreciable fluctuations.

Ein Ultraschall-Laufzeit-Beugungsverfahren wurde zur Messung des Rißfortschritts im quasistatischen Dreipunkt-Biegeversuch an Proben unterschiedlicher Werkstoffe eingesetzt. Der mit diesem Verfahren bestimmte Gesamtrißfortschritt steht in guter Übereinstimmung mit dem jeweiligen fraktografisch bestimmten Wert.
Durch eine geeignete Darstellungsweise der Ultraschallsignale konnte die Aussagekraft der Messungen deutlich erhöht werden. Insbesondere für Gußeisen zeigt sich ein ausgeprägter Übergang von Rißspitzenabstumpfung zu Rißausbreitung. Durch Ausnutzen der Bildinformationen kann der Rißfortschritt auch dann gemessen werden, wenn das Rißspitzenecho durch ein anderes Echo überlagert wird.
Es wurde gezeigt, daß Rißfortschrittsmessungen prinzipiell auch auf der Grundlage von rißspitzengebeugten Longitudinalwellen möglich ist.

Abstract:

The influence of the nuclear Coulomb field on transverse spectra of λ ⁺ and λ^- measured in Pb + Pb reactions at 158 A GeV has been investigated. Pion trajectories are calculated in the field of an expanding fireball. The observed enhancement of the λ^-/λ⁺ ratio at small momenta depends on the temperature and transverse expansion velocity of the source, the rapidity distribution of the net positive charge, and mainly the time of the freeze-out.

The project contributes to the improved evaluation of the mechanical integrity of the soviet-type VVER-440 reactors especially, to a sensitive early failure detection and to the localization of mechanical damages of reactor components by means of vibration monitoring. For that purpose the mechanical vibration of all primary circuit
components was modelled by finite elements. Modeling was built on the finite element code ANSYS. The interaction between the coolant flowing in the downcomer and the vibrating components has been considered by a fluid-structure element, which describes additional mode selective damping and intertia due to the coolant displacement when the downcomer geometry changes. The calculation model was adjusted using results from experimental vibration investigations. To some extent data from earlier measurements were available. But additionally dedicated experiments had to be performed at original VVERs. Now, the model can be regarded to be widely verified. Mainly it was applied to clarify how hypothetical damages of reactor internals influence the vibration signature of the primary circuit. Such kind of damage simulation is an appropriate means to find sensitive measuring positiones for on-line monitoring and to define physically based threshold values. In principle, the model is even suited to estimate the loads of reactor components which might be imposed by external events (explosion, earthquake).

Abstract
The distortion of the regular motion in a quantum system by its coupling to the continuum of decay channels is investigated. The regular motion is described by means of a Poissonian ensemble. We focus on the case of only few channels K < 10. The coupling to the continuum induces two main effects, due to which the distorted system differs from a chaotic system (described by a Gaussian ensemble): 1. The width distribution for large coupling becomes broader than the corresponding Χ² _K distribution in the GOE case. 2. Due to the coupling to the continuum, correlations are induced not only between the positions of the resonances but also between positions and widths. These correlations remain even in the strong coupling limit. In order to explain these results, an asymptotic expression for the width distribution is derived for the one channel case. It relates the width of a trapped resonance state to the distance between its two neighboring levels.

Abstract.

Small-angle correlations of Proton pairs produced in central Ni+Ni collisions at a beam energy of 1.93 AGeV are investigated with the FOPI detector System at GSI Darmstadt. A well-defined emission source is selected by triggering on central events which comprise about 8% of the total cross section. Simultaneous comparison of longitudinal and transverse correlation functions with the predictions of the Koonin model allows to unravel the space-time ambiguity of the emission process. Taking into account the strong collactive expansion of the participant zone, which introduces a reduction of the extracted source radius of more than 30% r.m.s. radius and emission time parameters of R_rms = ( 4.2 ± 1.2) fm and t_rms = (11⁺⁷ _-5) fm/c are extracted, respectively. In contrast, the analysis of the angle-integrated correlation function gives an upper limit R_rms = (7.0 ± 1.4) fm of the source radius.

The wear of silicon nitride ceramics is reduced by titanium
implantation albeit the friction coefficient is slightly increased. X-ray
excited photoelectron spectroscopy indicates that silicon oxide is formed
in the wear track while possible lubricating titanium oxides are not found.
Therefore we attribute the observed wear reduction to the amorphization
of the surface as proved by transmission electron microscopy.

Excited states of the nuclei ^{122, 126, 130}Te were populated
via the (gamma, gamma') reaction at endpoint energies of the bremsstrahlung
between 4.5 and 5.5 MeV. Gamma rays were detected with a EUROBALL CLUSTER
detector and a single HPGe detector. In all investigated nuclei two or
three prominent dipole transitions were identified at E_gammaca.
3 MeV. The corresponding low-lying J=1 states are interpreted as two-phonon
excitations. Quasiparticle-phonon-model calculations predict at about 3
MeV on 1^- state arising from coupling of the first quadrupole
and the first octupole phonon, and one 1⁺ state arising from
the coupling of the first and the second quadrupole phonon, where the latter
has isovector character. Such an excitation mode can be considered as an
analogue of the scissors mode in vibrational nuclei. The calculated transition
strengths are compatible with experimental ones within a factor of about
1.5.

Ein berührungsloses Dichtemeßverfahren für Medien geringer Dichte wird vorgestellt, das auf der Abhängigkeit der Reichweite von Positronenstrahlung von der Dichte des Meßmediums beruht. Die Positronen werden von einem Radionuklid in einer festen Quelle erzeugt und in das zu untersuchende Volumen emittiert. Seitlich angeordnete Detektoren registrieren die bei der Annihilation der Positronen entstehenden Gammaquanten. Das Verfahren wurde in einem Dichtebereich von 15 kg/m³ bis 38 kg/m³ getestet, eine Erweiterung des Meßbereichs von 5 kg/m³ bis 300 kg/m³ ist möglich.

Einsatzfelder sind mehrphasige Strömungen mit hohem Massenstromanteil der gasförmigen Phase (Schäume und Dämpfe) sowie unter Druck stehende Gase. Der Vorteil des Verfahrens besteht in einer Kombination der hohen Dichtesensitivität der Betastrahlung mit der geringen Absorption der bei der Annihilation entstehenden Gammastrahlung in Rohr- bzw. Behälterwänden.

Wenn ein zylindrischer Lagertank durch ein externes Feuer von der Seite intensiver Wärmeeinwirkung ausgesetzt ist, so setzt der Massenverlust infolge von Dampffreisetzung viel eher ein, als im Fall einer Wärmezuführung vom Behälterboden. Mittels eines vereinfachten Modells konnten die Details des Erwärmungsprozesses und die Ursachen für das unterschiedliche Verhalten in Abhängigkeit von der Art und Weise der Wärmezufuhr erklärt werden. Die ZUfuhr vom Boden führt zu einer irregulären thermischen Konvektion und damit einer guten Durchmischung des Fluids, so daß die Sättigungstemperatur an allen Punkten im Behälter nahezu gleichzeitig erreicht wird. Die maximale Enthalpie im Behälter verbleibt während der gesamten Zeit nahe dem Mittelwert. Wird der Behälter von der Seite beheizt, so bildet sich eine stabile horizontale Temperaturschichtung heraus. Die Bildung von Dampf erfolgt bereits lange bevor die mittlere Temperatur den Siedepunkt erreicht hat. Passive konstruktive Maßnahmen, wie der Einbau von Prallplatten, können die Schichtung stören, das verstärkte Aufheizen an der Oberfläche vermeiden und die Freisetzung von Dampf verzögern. Eine durchgreifende Vermischung des flüssigen Inhalts konnte jedoch nicht erreicht werden. Die genaue Gestalt und die Wirkung dieser Prallplatten hängt von der konkreten Behältergeometrie ab und läßt sich weiter optimieren.

Auf der Basis der Ergebnisse eines speziellen Windmeßprogramms und unter Nutzung des Windatlasprogramms WASP wurde das sächsische Windenergiepotential zu 6000 GWh/Jahr ermittelt. Das Erwartungspotential liegt bei 1700 GWh/Jahr (entsprechend 10% der derzeitigen Elektroenergiebedarfes). Bei einem jährlichen Ertrag einer Windenergieanlage von 2000 Vollbenutzungsstunden kann diese unter den derzeitigen Rahmenbedingungen wirtschaftlich betrieben werden. Die bisher vorliegenden Betriebsergebnisse unterstreichen die Notwendigkeit qualifizierter Standortgutachten.

Electromagnetic body forces, i.e. Lorentz forces, have been used to modify the boundary layer around a circular cylinder in cross flow. Depending on the polarity of the applied electric field different effects on the flow can be obtained. Lorentz forces directed with the mean flow are able to prevent the boundary layer from separation. Therefore flow separation as well as the von Karman vortex trail can be suppressed. When the momentum gain produced by the Lorentz forces in the boundary layer is high enough, thrust is produced, what results in a jet flow originated from the cylinders back side. If the Lorentz forces are directed opposite to the mean flow, the separation points are shifted towards the front stagnation point, the recirculation region broadens and the von Karman vortex trail is modified. The described technique gives a variety of opportunities to control the flow around the cylinder and the flow structure of the cylinder wake. Results from flow visualizations and numerical calculations are presented.

The exclusive deuteron break-up reaction is analyzed within a covariant approach based on the Bethe-Salpeter equation with realistic meson-exchange interaction. Relativistic effects in the cross section, tensor analyzing power and polarization transfer are investigated in explicit form. Results of numerical calculations are presented for kinematical conditions in forthcoming p + D reactions at COSY.

A rate equation approach is applied for the description of the self-organization (layering) phenomenon predicted in recent computer experiments. This layering was observed in finite precipitate systems during annealing and is caused by Ostwald ripening. We demonstrate that the layering is triggered by local inhomogenities in the spatial distribution of precipitate sizes and should not be treated in terms of the frequently considered "spinodal
decomposition" approach. The change of the spatial profile of precipitate size starts from the boundary of the system and occurs within a "reaction shell" having a thickness of the order of the diffusional screening
length. During annealing this "reaction shell" shifts progressively into the system, leaving behind layers of precipitates. The layering is shown to occur only in sufficiently large systems with characteristic dimensions of at least several diffusional screening lengths. The reason for the weak sensitivity of interlayer distance to variations of system parameters is elucidated.

One of the most critical problems in practical applications of spectrum adjustment in pressure vessel dosimetry is the insufficient knowledge of the input spectrum covariance matrix. Usually very crude approximations for this covariance matrix are used in adjustment procedures and, moreover, it is not clear to which extent these approximations influence the final results. Therefore it appeared useful to calculate this matrix as carefully as possible for a typical system and in positions for which a sufficient amount of activation measurements and precise transport calculations was available. The paper presents and discusses results of such calculations for a VVER-1000 type reactor. They base on the one hand on sensitivity coefficients estimated in transport calculations by means of the ANISN code and on the other hand on cross section covariance data taken from the Obninsk ABBN-93 data library. The results of spectrum adjustments based on the estimated covariance matrix are compared with results based on formerly used matrices. The obtained differences are discussed.

In 1995 at the integral test facility ISB-VVER in Elektrogorsk near Moscow natural circulation experiments were performed, which were scientifically accompanied by the Forschungszentrum Rossendorf. These experiments were the first of this kind at a test facility, which models VVER-1000 thermalhydraulics. Using the code ATHLET which is being developed by "Gesellschaft für Anlagen- und Reaktorsicherheit", pre- and posttest calculations were done to determine the thermalhydraulic events to be expected and to define and tune the boundary conditions of the test. The conditions found for natural circulation instabilities and cold leg loop seal clearing could be confirmed by the tests. Besides the thermalhydraulic standard measuring system, the facility was equipped with needle shaped conductivity probes for measuring the local void fractions.

Three dimensional Monte Carlo calculations considering all known geometrical and composition details were used for the determination not only of fluence integrals but also of fluence spectra with good statistical accuracy. This was possible by the application of own specially developed codes (TRAMO) and successfully modifying variance reduction on the basis of the weight window method (TRAWEI). Modern multigroup data sets were applied. For the real representation of fission sources the burn up and power history also for the fuel rods within the fuel assemblies was considered.
The investigation was accomplished for the irradiation period seven with a special low leakage loading of the reactor. During this period Russian and German detectors were irradiated in the ex-vessel cavity and analysed by advanced gamma spectrometric methods.
The reaction rates obtained with the calculated fluence spectra were compared with activation reaction rates measured in the ex-vessel cavity. Furthermore the calculated spectra were adjusted to the experimental reaction rates taking into account the variances and correlations of calculated input spectra, detector sensitivities and measurements. Using the adjustment code COSA2 based on the maximum likelihood method best estimate fluence spectra and fluence integrals could be provided.

Plasma immersion ion implantation was used to implant
oxygen ions into silicon with applied voltage pulses of -40 kV and 2.5
ms lenght. Positive ions, O2+ and O+, with a continuous energy distribution
between 0 and 40 keV were implanted. Between 3x104 and 3x105 pulses, corresponding
to nominal doses from 2x1016 to 2x1017/cm2 were used. The resulting oxygen
depth profiles were measured with elastic recoil detection analysis (ERDA)
using 13.4 MeV a particles. Rutherford backscattering was used to determine
possible co-implanted contaminations. The obtained depth profiles were
simulated using a linear supersposition of calculated single-energy profiles.
The results obtained for the energy distribution of the incident ions are
compared with calculations obtained from a theoretical model, and the agreement
is very good. The incident flux is found to be composed of 34(5)% O2+ and
66(5)% O+ ions with an Fe contamination of ca. 0.5%.

In this final design report the general layout and the optics of the beamline ROBL is described. The beamline is buildt by the Research Center Rossendorf at the bending magnet source BM20 of the European Synchrotron Radiation Facility (ESRF), Grenoble. The design criteria for the beamline optics and the adopted solutions are reported in detail. Some performance data of the beamline as resolution, focussing and intensity are modelled by the ray-tracing code SHADOW. Furthermore tha layout of the vacuum system is described.

We studied high dose implantation of tungsten into 6H-silicon carbide in order to synthesize an electrically conductive layer. Implantation was performed at 200 keV with a dose of 1.2×10(17) W(+)cm(-2) at temperatures between 200°C and 400°C. The influence of implantation temperature on the distribution of W in SiC was investigated and compared to results obtained earlier from room temperature (RT) and 500°C implants. Rutherford backscattering spectrometry (RBS) was employed to study the structure and composition of the implanted layers. Implantation at temperatures between RT and 300°C did not influence the depth distribution of C, Si and W. The W depth profile shows a conventional Gaussian shape. Implanting at higher temperatures led to a more confined W rich layer in the SiC. This confinement is explained by Ostwald ripening which is enabled during implantation at temperatures above 300°C. The depth of the implantation induced damage decreases slightly with increasing implantation temperature, except for 400°C implantation. The amount of damage, however, is significantly reduced only for implantation at 500°C.

The semi-magic nucleus ⁸⁹Y has been investigated in a (gamma,gamma')
experiment at an endpoint energy of the bremsstrahlung of E₀=
7 MeV. The scattered photons have been detected with a EUROBALL Cluster
detector. The observed excitations are discussed in the framework of the
shell model and a coupling of the unpaired proton to milti-phonon structures
in the doubly even neighbours. Transitions above 4.7 MeV are considered
to have E1 character. Around 6.3 MeV an unusually large concentration of
E1 strength is found. Its origin is likely to correspond to similar structures
in the ⁸⁸Sr and ⁹⁰Zr isotones which can be interpreted
to result from the constructive interference of strong two-phonon amplitudes
with weak admixtures from the low-energy tail of the giant dipole resonance.

The report gives an overview on the scientific work of the Institute of Safety Research in 1995.

The accelerator mass spectrometry (AMS) facility at the
3 MV Tandetron in Rossendorf has been applied for quantitative depth profiling
of deuterium and tritium in samples cut from the vessels walls of the fusion
experiment ASDEX-Upgrade in Garching. The tritium originates from D(d,p)T
fusion reaction in the plasma and it is implanted in the vessel walls together
with deuterium atoms and ion from the plasma. The high deuterium concentrations
in the samples were measured by a conventional SIMS mode. The concentrations
in the surface layers down to the analysing depth of about 25 mm are for
deuterium in the range of up to 3x1018 to 8x1021 atoms/cm3, corresponding
to a deuterium retention of 2.5x1016 to 2.5x1018 atoms/cm2 in a surface
layer £ 25 mm, and for tritium in the range form 1011 to 5x1015 atoms/cm3,
corresponding to a tritium retention of 3.0x1010 to 3.5x1012 atoms/cm2
in a surface layer £25 mm. The estimated total amount of tritium
in the vessel walls is in agreement with the total number of neutrons produced
in D(d,n)3He reaction.

Acoustic waves induced by an intensity-modulated focused
ion beam (FIB) have been measured. The experiments were performed with
Ga+ ions of 35 keV at a current of 3 nA and variable chopping frequency
up to 10 MHz. The acoustic signals were detected by means of a piezoelectric
sensor with integrated pre-amplifier. A dependence on the sample material
was found. The results suggest that the ion-acoustic effect can be utilized
for imaging and material characterization of surface and subsurface structures
in FIB systems.

Acoustic waves induced by an intensity modulated focused ion beam (FIB) have been measured. The experiments were performed with Ga+ ions of 35 keV at a current of 3 nA and variable chopping frequency up to 10 MHz. The acoustic signals were detected by means of a piezoelectric sensor with integrated pre-amplifier. Frequency and position sensitivity of the sensor has been measured by laser excitation. Ion acoustic measurements have been carried out at the resonance frequencies of the sensor. A dependence on the sample material was found. The results show that the ion-acoustic effect can be utilized for imaging and material analysis in FIB systems.

Fe-Al show interesting physical properties and offer some
special industrial applications. There are phase transitions combined with
changes in the magnetic behaviour. Another interesting fact is the excellent
oxidation and corrosion resistance of iron aluminides even at high temperatures.
Thin Fe-Al layers can be produced in different ways. Ion beam methods enable
to produce surface layers on bulk material modifying the initial properties
completely. The properties depend strongly on the phase structure induced
by the preparation process. 57-Fe Mössbauer spectroscopy, and in the
case of surface layers conversion electron Mössbauer spectroscopy
is very suitable to identify this phase structure. In the present work,
it is described for Al implanted Fe layers. Depending on implantation dose
and energy both magnetic and non-magnetic phases can be produced. Due to
the inhomogeneous distribution of Al in the Fe target a layer structure
of different phases can be created. Moreover, due to double implantation
an Fe-Si-Al alloy can be prepared

Abstract
The mixing of the doorway components of a giant resonance due to the interaction via the common decay channels influences significantly the distribution of the multipole strength and the energy spectrum of the decay products of the giant resonance. The photoemission turns out to be most sensitive to the overlapping of the doorway states. At high excitation energies, the interference between the doorway states leads to a restructuring towards lower energies and apparent quenching of the dipole strength.

Abstract
Using a phenomenological schematic model of multipole giant resonances we consider the effects of overlapping of their doorway components. The conccpt of the partial widths of a giant resonance becomes ambiguous when the escape widths get comparable with the spacings between the components. In such a case, the partial widths determined in terms of the K- and S-matrices differ from each othcr. The mixing of the doorway components due to the interaction via the common decay channels influences significantly their multipole strengths, widths and positions in energy.

The Tilted Axis Crankiug theory is applied to the model of two particles coupled to a triaxial rotor. Cornparirig with the exact quantal solutions, the interpretation and quality of the mean field approximations studied. Conditions are discussed when the axis of rotation lies inside or outside of the principal planes of the triaxial density distribution. The planar solutions represent ∆ I = 1 bands, whereas the aplanar solutions represent pairs of identical ∆ I = 1 bands with the same parity. The two bands differ by the chirality of the principal axes with respect to the angular momentum vector. The transition from planar to chiral solutions is evident in both the quantal and the mean field calculations. Its physical origin is discussed.

The tribological and microstructural properties of gas-pressure
sintered Si3N4 ceramics after chromium implantation have been investigated
in dependence on implantation energy. The fluence was kept constant at
1017 ions/cm2. The wear of the modified material is reduced for a load
of 2 N independent of ion energy. This behaviour is accompanied by a slight
incerase of the friction coefficient. At higher loads only high energy
implantations result in an improved wear behaviour. The microstructural
investigations show that no new phase are formed due to implantation. For
all energies a part of the surface becomes amorphous. For smaller energies
this amorphous layer reaches up to the surface whereas at higher energies
it is covered by still crystalline but damaged material. The amorphization
of the surface is accompanied by a large surface swelling which incerases
with implantation energy. We discuss the relationship between the microstructural
features and the observed wear behaviour.

Limitations of the pulse-shape discrimination (PSD) technique - a promising method to identify the charged particIes stopped in planar Si-detectors - have been investigated. The particle resolution turned out to be basically determined by resistivity fluctuations in the buk silicon which cause the charge-collection time to depend on the point of impact. Detector maps showing these fluctuations have been measured and are discussed. Furthermore we present a simple method to test the performance of detectors with respect to PSD. Another limitation of the PSD technique is the finite energy threshold for particle identification. This threshold is caused by an unexpected decrease of the total charge-collection time for ions with a short range, in spite of the fact that the particle tracks are located in a region of very low electric field.

Iron and magnesium are insoluble elements with each other
and there is no phase diagram. However, it is possible to produce artificial
alloys by ion implantation, in this study by iron implantation into magnesium.
Samples are investigated by conversion electron Mössbauer spectroscopy,
Auger electron spectroscopy and x-ray diffraction. While at low doses gaussian
shaped iron profiles and paramagnetic doublets as Mössbauer spectra
are obtained, the iron concentration reaches at the highest dose 90 at.-%
in maximum and the Mössbauer spectrum shows a dominant ferromagnetic
fraction. The x-ray diffraction pattern let conclude that a dilated a-iron
lattice is formed. Microhardness of all samples is clearly increased due
to the implantation.

Iron implanted and subsequently annealed n-type Si(111) was
studied by conversion electron Mössbauer spectroscopy for phase analysis
and Auger electron spectroscopy for sputter depth profiling and element mapping.
During implantation (200 keV, 3 x 1017 cm-2, 350 °C) a mixture of ß-
and a-FeSi2 and after the subsequent annealing (900 °C for 18 h and
1150 °C for 1 h) a complete transition to the ß- and a-phase can
be detected. The as-implanted profile has Gaussian shape which becomes broad
during annealing at 900 °C to a plateau-like profile and shows only
a slight broadening and depth depending fluctuations of the iron concentration
after the 1150 °C annealing. With scanning Auger electron spectroscopy
the lateral iron and silicon distribution were investigated show for the
sample annealed at 900 °C large separated ß-FeSi2 precipitates
which grow due to the process of Ostwald ripening. At 1150 °C additionally
coalescence of the precipitates occur and a wide extended penetration a-FeSi2
network structure is formed.