Publications Repository - Helmholtz-Zentrum Dresden-Rossendorf

Bacterial diversity in two uranium waste piles in Germany was studied. For this study, total DNA was recovered by two alternative procedures for direct lysis from a large number of soil samples drawn from different sites and depths of the piles. Significant differences in the bacterial composition of the samples investigated were revealed by the use of rep-APD, RISA, and 16S ARDREA. The 16S rDNA retrieval allowed also to demonstrate that both uranium wastes were dominated by Thiobacillus ferrooxidans and by several Pseudomonas species of g-subdivision of the Proteobacteria. Three kinds of T. ferrooxidans 16S and IGS rDNA specific fragments corresponding to the recently described three phylogenetic groups of this species (Flemming at al., 1999) were amplified in different ratios from distinct soil samples which were polluted with heavy metals to different extent. This microdiversity probably reflects the genetic adaptation of the uranium waste strains to the different heavy metal concentrations in their natural surrounding.

Nuklidvektor in verbrauchtem Kernbrennstoff;
Globale Plutoniumbilanz;
Recycling oder Endlagerung von Plutonium;
Accelerator-Driven-Systems (ADS) zur Aktiniden- und Spaltprodukttransformation

There is evidence that intrauterine growth restriction (IUGR) is associated with altered dopaminergic function in the immature brain. However, the relevant enzyme activities have not been measured in the living neonatal brain together with brain oxidative metabolism. Therefore ¹⁸F-labelled 6-fluoro-L-3,4-dihydroxyphenylalanine (FDOPA) was used together with positron-emission-tomography to estimate the activity of the aromatic amino acid decarboxylase in the brain of 10 newborn IUGR piglets (2 to 5 days old; body weight 908±109g) and in 10 normal-weight (3 to 5 d old; body weight 2142±373g) newborn piglets. The regional transport of FDOPA to the brain and the clearance rate of labeled metabolites from brain tissue were broadly similar in the two groups. However, the regional rate constant for back flux from the brain was markedly increased in IUGR piglets for striatum (72%) and frontal cortex (83%) (p < 0.05). Furthermore, the rate constant for conversion of FDOPA to fluordopamine was markedly increased (between 48 % in cerebellum and 91 % in mesencephalon, p < 0.05) in all brain regions of IUGR piglets studied.
Thus, it is suggested that IUGR induces an up-regulation of amino acid decarboxylase activity that is not related to alterations in brain oxidative metabolism.

A novel surface layer (S-layer) protein was found on Bacillus sphaericus isolate recovered from a uranium mining waste pile. Formation of highly ordered platinum and palladium nanoclusters on this crystalline protein template was studied. The average diameter of the metal clusters is 1.9±0.6 nm. They are located in the nano-sized pores and gaps of the crystalline S-layer. The atomic structure of the S-layer-metal nano-clusters was analysed by the use of EXAFS spectroscopy.
In addition, it was demonstrated that the B. sphaericus isolate is able to accumulate selectively large amounts of U, Pb, Cd, and Al from drain waters of the uranium mining waste pile. For this reason, this strain was chosen for construction of so called bio-ceramic "Biocere" which may be used for bioremediation of heavy metal contaminated liquid wastes. Our first analyses of the properties of this nano-scalic bio-ceramic will be presented.

Tritium depth profiling measurements by accelerator mass spectrometry have been performed at the facility installed at the Rossendorf 3 MV Tandetron. In order to achieve an uniform erosion at the target surface inside of a commercial Cs ion sputtering source and to avoid edge effects, the samples were mechanically scanned inside of a commercial Cs sputter ion source. The sputtered negative ions were mass analysed by the injection magnet of the Tandetron. Interesting ions except tritium are measured with the Faraday cup between the injection magnet and the accelerator, while the tritium is counted after the accelerator with semiconductor detectors. Depth profiles have been measured for carbon samples which had been exposed to the plasma at the first wall of the Garching fusion experiment ASDEX-Upgrade and from the European fusion experiment JET, Culham/UK. Tritium contents in the JET samples were up to six orders higher than in samples from ASDEX-Upgrade. Tritium beam currents from samples with high tritium content can be measured partly in the Faraday cup before the accelerator. A dedicated tritium AMS facility with an air-insulated 100 kV tandem accelerator is under construction. First results of test operation are presented.

For focused ion beam applications of Erbium as well as of several metal ions which are of interest for optical and semiconductor investigations a liquid metal ion source operating with an Er-stainless steel alloy was developed and investigated. This alloy consists of a mixture of Er70Fe22Cr5Ni3 and has a melting point of about 860°C. The wetted needle type tungsten emitter showed a stable emission behaviour down to 1µA emission current. The I-V characteristics, the temperature dependence of the extraction voltage, the mass spectrum, and the energy spread of the main beam components depending on the emission current were investigated. While in the case of singly charged metal ions the beam is not defined due to isotope interference in the case of doubly charged ions the species well separated for focused ion beam applications are available.

After 16alpha-[¹⁸F]fluoroestradiol ([¹⁸F]FES) has been successfully prepared in an automated module, the synthesis of 16alpha-[¹⁸F]fluoroestradiol-3,17beta-disulphamate ([¹⁸F]FESDS) is described as a module-assisted one-pot procedure which can provide 10 GBq [¹⁸F]FESDS with a radiochemical purity better than 99 %. The procedure is reliable and reproducible and requires a time of about 90 min. Because of its high sulphatase-inhibitory effect [¹⁸F]FESDS is thought to be a new PET tracer to image sites of high sulphatase activity.

The synthesis of 16alpha-[¹⁸F]fluorestradiol-3,17beta-disulphamate ([¹⁸F]FESDS) is described. 16alpha-[¹⁸F]fluorestradiol ([¹⁸F]FES) is converted using excess sulphamoyl chloride in absolute acetonitrile in the presence of Kryptofix 2.2.2 and potassium carbonate using an automatically operating module. The required time for the synthesis related to end of bombardment is 3h, the maximum yield is 6%, and the maximum decay-corrected yield is 20%. The radiochemical purity of [¹⁸F]FESDS is > 99%. The specific radioactivity of [¹⁸F]FESDS is found to be between 150 and 200 GBq/µmol.

Starting from the lead structure of ketanserin, a prototypic serotonin (5-HT) antagonist, a series of oxotechnetium(V) complexes were synthesized which are able to compete with [³H]ketanserin in receptor binding assays.
In order to imitate organic 5-HT₂ receptor ligands, fragments of ketanserin were combined with chelate moieties. Lipophilic complexes, consisting of a small S₄ or S₃N thiolate/thioether chelate unit, protonable nitrogen containing spacer and benzyl moiety significantly inhibited the specific binding of [³H]ketanserin with IC₅₀ values between 10 and 50 nM.
Low brain uptake is a generally accepted problem in developing ^99mTc brain receptor imaging agents. For some representatives of 5-HT_2A receptor binding agents it was tried to improve the original low brain uptake of 0.4 % ID in rats 5 min p.i. by modifying the lipophilic properties of the molecules. Because of the presence of a protonable nitrogen, which according to the pK value leads to ionization of the molecule at blood pH, the pK value was considered to be the parameter most suitable for adjustment of lipophilicity. Insertion of ether-oxygen in the molecule of 5 candidates lowers the apparent pK value from 10.0 to 8.3 and dramatically increases the brain uptake up to 1.3 % ID at 5 min.
Finally, the synthesis and in vitro autoradiography of a novel Tc-99m ligand with subnanomolar affinity to the 5-HT_2A receptor is reported. The complex combines the 4-(4-fluoro)-benzoyl piperidine portion derived from ketanserin with a neutral oxotechnetium(V) chelate in form of a mixed ligand "3+1" unit containing the SNS/S donor set. The analogous ⁹⁹Tc compound has been synthesized as a surrogate for the Tc-99m complex for use in receptor binding assays and for complete structural characterization.
In competition experiments the Tc-99 complex as well as its Re analogue display subnanomolar affinity towards the 5-HT_2A receptor (K_i 0.44 nM for Tc, 0.25 nM for Re).
In vitro autoradiographic studies clearly indicate the accumulation of the Tc-99m compound in 5-HT_2A receptor rich areas of the brain. This enrichment can be blocked by 5-HT_2A receptor antagonists such as mianserin and ketanserin and is therefore specific.

The effect of copper, phosphorus and nickel on the radiation embrittlement was investigated by using an irradiation experiment at surveillance positions in two Russian VVER-type reactors with 8 iron-base testing alloys. Their chemical composition varied between 0.015 to 0.42 % Cu, 0.002 to 0.039 % P, 0.01 to 1.98 % Ni, 0.09 to 0.37 % Si, and 0.35 to 0.49 % Mn. Charpy-V impact tests and tensile tests were performed with specimens machined from these alloys. The specimen were tested in the as-received state, in the irradiated state (fluence: 1×10 19 and 8×10 19 cm -2 [En > 0.5 MeV]) and in the post-irradiation annealed state.
Irradiation produces strong hardening and embrittlement. The effect increases with the Cu and P content. Ni causes an additional embrittlement. It does not depend on the Ni content within the range of 1.1 to 2 % Ni. The irradiation effect can be eliminated by annealing at 475 °C / 100 h but not absolutely in the case of high Cu or P content.
Small angle neutron scattering experiment at these alloys show that not only the volume fraction but also the type of the radiation defects vary in dependence on the composition of the alloys.

Dendrimer chemistry is a rapidly espanding field for both fundamental studies and applications. Their unique structural features and properties make them ideally suited for use in diagnostic imaging and therapy, transfection, and drug delivery. A short overview about some fields of application in medicine was given. It was reported the binding and transport behaviour of dendrimers having urea, azobenzene, methyl orange and diamidopyridine moieties at the periphery towards pertechnetate, and the nucleotides AMP, ADP and ATP.

Sequence analysis of the 16S rRNA genes of several reference strains and uranium mining waste pile isolates of Thiobacillus ferrooxidans revealed specific signatures which distinguish three types within the species. This allowed to develop a technique for analysis of the distribution of the T. ferrooxidans eco-types in the soil samples of a uranium mining waste pile.
The technique is based on amplification of 16S rDNA fragments in total soil DNA by the use of two T. ferrooxidans species specific primers 16S458F and 16S1473R (De Wulf-Durand et al., 1997). The resulting amplicons were then digested with a frequently cutting enzyme RsaI which produced three different type-specific profiles (Selenska-Pobell et al., 2000). Using this direct approach we have demonstrated that one of the T. ferrooxidans types (type I) was predominant in the soil samples studied and was found in more polluted sites, whereas the type II was found in less contaminated samples. The type III was found mostly to coexist with the type II.
The complexation of uranium by the three eco-types of T. ferrooxidans was investigated. The uranium sorption studies showed that the strains from the three different types possess different capability to accumulate uranium. The process of uranium accumulation is affected by the initial uranium concentration and pH. Using sodium carbonate as a desorbing agent, 97% of the biosorbed uranium was recovered from the cells of T. ferrooxidans type III, and 88.33 and 88.50% from the cells of the type I and II, respectivley. Extended x-ray absorption fine structure (EXAFS) and infrared (IR) spectroscopy were performed to obtain information about the exact chemical structure of the uranium complexes formed on the surfaces of the three eco-types of T. ferrooxidans.

Gram-positive spore-forming bacterial isolates belonging to the species Bacillus cereus, B. sphaericus and B. megaterium and also Gram-negative chemolithoautotrophic isolates of the species Thiobacillus ferrooxidans and Leptospirillum ferrooxidans were cultured from soil and sediment samples of several uranium polluted waste piles in Saxony, Germany. The interactions of these natural bacterial isolates with uranium (in particular, sorption - desorption, accumulation, reduction, and tolerance) were studied.
We have demonstrated that the natural Bacillus isolates are able to accumulate selectively and reversible a variety of toxic metals from drain waters of a uranium mining waste pile (Selenska-Pobell et al., 1999). By the use of extended X-ray absorption fine structure (EXAFS) analysis and time resolved laser fluorescence spectroscopy (TRLFS) it was demonstrated that the main chemical group involved in the interaction of the mentioned Bacillus strains with U(VI) is the phosphate group (Hennig et al., 1999; Panak et al., 2000).
Our studies on the interactions of the recently described in our laboratory three ecological types (eco-types) of T. ferrooxidans (Selenska-Pobell et al., 1999) and two eco-types of L. ferrooxidans (Tzvetkova et al., 1999) with uranium (VI) by the use of EXAFS, TRLFS, and infrared spectroscopy (IRS) indicate that the nature and the strength of the uranium complexes built by these Gram-negative acidophilic chemolithoauthotrops differ significantly from those built on the cell surfaces of the Gram-positive Bacilli. The tolerance and the interactions of the natural T. ferrooxidans isolates with U were type-specific.

The complexation of uranium by the cells of the three recently described eco-types of T. ferrooxidans (Flemming et al., 1999; Selenska-Pobell et al., 1999) was studied. The uranium sorption studies showed that the strains from the three different types possess different capability to accumulate uranium. We have found that the representatives of type II accumulate significantly higher amounts of uranium in comparison to the other T. ferrooxidans strains studied. The amount of accumulated uranium gradually increased with increasing metal concentration at pH 1.5 and 4. The uranium tolerance investigations demonstrated that the types I and III are resistant to 8 and 9 mM of uranium respectively, whereas the type II does not tolerate more than 2 mM of uranium. Sodium carbonate was the most efficient desorbing agent tested, removing 97% of the uranium sorbed from the cells of T. ferrooxidans type III, and 88.33 and 88.50% from the cells of the types I and II, respectivley. Sodium citrate was the second best metal desorbing agent, whereas EDTA could remove only between 10 and 30% of the accumulated uranium. By the use of extended x-ray absorption fine structure (EXAFS) and infrared (IR) information was obtained about the exact chemical structure of the uranium complexes formed on the surface by the three eco-types of T. ferrooxidans.

The tilted axis cranking model is used in combination with the random phase approximation and particle number projection to analyze the influence of dynamical pairing correlations in the high-K bands of 178-W and their effect on relative energy and angular momentum. The calculations show the importance of dynamical pairing correlations to describe the experiment as well as advantages and problems with the different models in the superfluid and normal state regions.

A description of the fuel rod behaviour and heat transfer model used in the code DYN3D for nuclear reactor core dynamic simulations is given. Besides the solution of heat conduction equations in fuel and cladding, the model comprises detailed description of heat transfer in the gas gap by conduction, radiation and fuel-cladding contact. The gas gap behaviour is modeled in a mechanistic way taking into account transient changes of the gas gap parameters based on given conditions for the initial state. Thermal, elastic and plastic deformations of fuel and cladding are taken into account within 1D approximation. A creeping law for time-dependent estimation of plastic deformations is implemented. Metal-water reaction of the cladding material in the high temperature region is considered. The cladding - coolant heat transfer regime map covers the region from one-phase liquid convection to dispersed flow with superheated steam. Special emphasis is put on taking into account the impact of thermodynamic non-equlibrium conditions on heat transfer.

For the validation of the model, experiments on fuel rod behaviour during RIAs carried out in Russian and Japanese pulsed research reactors with shortened probes of fresh fuel rods are calculated. Comparisons between calculated and measured results are shown and discussed. It is shown, that the fuel rod behaviour is significantly influenced by plastic deformation of the cladding, post crisis heat transfer with subcooled liquid conditions and heat release from metal water reaction.

Numerical studies concerning the fuel rod behaviour under RIA conditions in power reactors are reported about. It is demonstrated, that fuel rod behaviour at high pressures and flow rates in power reactors is different from the behaviour under atmospheric pressure and stagnant flow conditions in the experiments. The mechanisms of fuel rod failure for fresh and burned fuel reported from the literature can be qualitatively reproduced by the DYN3D model. However, the model must be extended and improved for proper description of burned fuel behaviour.

The realistic simulation of the fuel rod behaviour is important not only under RIA conditions, but also for the analysis of operational transients. This has been shown in a calculation of an operational transient with power decrease after switching-off one from the two working feed water pumps in the NPP Balakovo (VVER-1000).

The reactivity of labile 3+1 mixed-ligand ^99mTc complexes of the type [^99mTcO(SES)(RS)] with SES being a tridentate dithiol ligand and glutathione or dimethylcysteamine as monodentate ligands RSH towards proteins was investigated in vitro and in vivo. It was found that the complexes undergo reversible transchelation reactions with SH group-containing components of blood such as albumin or hemoglobin. High labelling yields were obtained when 3+1 complexes with the tridentate SSS ligand were used. The biodistribution of blood proteins labelled by ligand-exchange reaction with the [^99mTcO(SSS)] or [^99mTcO(SNMeS)] core was studied and compared with the in vivo distribution of the labile 3+1 complexes containing glutathione as monodentate ligand.

He⁺ ions were implanted at 40 keV into Si <100> random direction at room temperature to fluences between 5×10¹⁵ He⁺/cm² and 3.5×10¹⁶ He⁺/cm². The samples were than contaminated with Cu in order to study the gettering of Cu atoms at the defective layer. A subsequent annealing at 800ºC was performed in order to recover the implantation damage and redistribute the Cu into the wafer. The samples were analyzed by Rutherford backscattering/channeling (RBS/C) and transmission electron microscopy (TEM) techniques. The Cu distribution was measured by secondary ion mass spectrometry (SIMS). The SIMS results show that the 5×10¹⁵ He⁺/cm² implant induces gettering at the He⁺ projected range (Rp) region, while the 8×10¹⁵ He⁺/cm² and 3.5×10¹⁶ He⁺/cm² implants have resulted in gettering at both the Rp and Rp/2 depths. The TEM micrographs show no visible defects at Rp/2 region. This corresponds to the so called Rp/2 gettering effect. It is important to note that in this study the Rp/2 related effect is observed after an implantation of a light ion at low energy.

The TEM study of titanium contained ZSM-5 zeolite before and after hydrothermal treatment was performed. The use of different TEM techniques - such as conventional TEM, HRTEM and EDX line scans provides important information about the zeolite catalyst consisting from several phases. The hydrothermal treatment of zeolite powder leads to strong changes in the morphology of the constituting particles. They are characterised by a homogeneous structure before hydrothermal treatment while the occurrence of holes after thermal treatment was observed. These changes lead to the enrichment of zeolite with titanium that enhance its catalytic activity. Some of the titanium surplus precipitates as TiO2 anatase nanoparticles within the holes.

Strong gettering of Cu atoms beyond the projected ion range RP has been found in single-crystal Si implanted with P+ and As+ ions at MeV energies. We call this phenomenon the "trans-RP effect". The formation of a separate Cu gettering band below RP, as detected by secondary ion mass spectrometry, indicates the presence of a significant amount of defects therein. These defects are not detectable by transmission electron microscopy and we suggest that they are small interstitial clusters. The amount of Cu atoms gettered beyond RP is, particularly for the P implants, much greater than that in the implanted gettering layer, indicating that the gettering ability of the defects beyond RP is higher than that of the extended defects at RP. A mechanism responsible for their formation and clustering in the trans-RP region is proposed, and an explanation is given of the differences in the results for the P and As implants.

There are two major reasons to study the effects of high dose implantation in SiC. First, the production of low-resistivity, p-type layers requires high acceptor concentrations which are far above the equilibrium solubility. It is essential to know the critical acceptor concentration for thermally stable, homogeneously doped SiC. On the other hand, there is a lack of information about the precipitation of new phases in SiC, their properties and influence on the SiC matrix. Such information is needed for the assessment whether insulating or conductive layers with device quality can be formed in SiC by ion beam synthesis. Two examples of high dose implantation in 6H-SiC are presented.

In the first example high doses of 350 keV Al+ ions were implanted in 6H-SiC single crystals at temperatures between 400oC and 800oC in order to preserve the crystallinity of the SiC matrix. The phase formation was studied by TEM, SIMS and AES. A critical Al concentration of about 10 at% was found below that the 6H-SiC structure remains stable during implantation at 500°C. The Al atoms occupy preferentially Si sites in the SiC lattice. At higher Al concentrations the SiC matrix is decomposed and precipitates of Si and Al4C3 are formed. It was found that the Al4C3 precipitates have a perfect epitaxial orientation to the SiC matrix. The phase transformation is accompanied by atomic redistribution and strong volume swelling. The observed atomic depth profiles can be accounted for by a a simple chemical reaction model. Implantation at 800°C leads to the formation of Si and Al precipitates accompanied by a dramatic volume swelling which results in surface cracking and material outflow. Post-implantation annealing above 1500°C severely deteriorate the SiC matrix by local melting of Al and Si precipitates.

In the second example the evolution of the binary system C -> SiC was studied as function of implantation temperature. 1 1018 cm-2, 60 keV carbon ions were implanted into single crystalline 6H silicon carbide (SiC) at elevated temperatures. The formation of carbon phases in the crystalline SiC lattice was investigated by cross sectional transmission electron microscopy. An amorphous, carbon rich phase was produced at 300oC. Precipitates of graphite were obtained at 600oC, whereas at 900oC small diamond grains were produced. These grains are in perfect epitaxial relation with the surrounding SiC lattice. The mechanism of the diamond formation by IBS in SiC is not quite clear. It can be assumed that the tetrahedrally coordinated SiC lattice, which is preserved during the high temperature implantation, acts as a template for the growth of diamond. In addition, local nonequilibrium conditions in the ion cascades (thermal spikes, shock waves) or temporary stress on interstitially incorporated carbon atoms by the surrounding SiC lattice could contribute to the diamond nucleation.

Results of a joint German/Russian irradiation programme performed on the prototype pressurized water reactor VVER-2 of the Rheinsberg nuclear power plant Germany) are summarized. The experiment comprises Charpy V-notch (CVN), precracked Charpy size (SENB) and compact tension (CT) specimens made of different heats of Russian VVER type reactor pressure vessel (RPV) base and weld metals. Reference temperatures, T0, were evaluated according to the Master Curve (MC) concept using the
multi temperature method. Neutron irradiation induced ductile-to-brittle transition temper ature (DBTT) shifts determined on the basis of CVN and SENB tests are compared. On the base of the DBTT the neutron embrittlement sensitivity and the annealing behaviour of tested RPV steels are evaluated. Different heats of the same VVER-RPV steel exhibit different neutron induced embrittlement and annealing behaviour. The determined CVN transition temperatures correlates to the T0 temperatures evaluated by the MC concept.

The sequence of phase transitions during the hot history of the universe is followed within a phenomenological framework.
Particular emphasis is put on the QCD confinement transition, which is at reach under earth laboratory conditions.
A tepid inflationary scenario on the GUT scale with bubble grow at moderate supercooling is discussed.

The overall process of uptake of heavy elements (actinides) into plant tissues from contaminated soils has been quantified using the soil-to-plant transfer factor (TF). Considering the radionuclide transfer, it is quantified as TF(RN) = plant specific activity (Bq/kg-1) /soil specific activity (Bq/kg-1). In practice the predominant pathway of plant contamination is direct contamination by deposition and resuspension. The comparable uptake via soil-plant transfer is about 10%. But this ratio strongly depends on the real accident scenario (distribution of contaminants via aerosol in air, as soluble or colloidal species in aquifer or surface water).Up to now, the chemistry of the various actinides has not been considered along the pathway soil-plant-animal-humans. The transfer factor has to be unfolded into the basic parameters like changes in oxidation state and complexation, differences in bonding of the elements like adsorption and ion exchange processes, formation of colloids e.g.
It was our goal to obtain first results on the change of uranium speciation during the soil-to-plant transfer. X- ray absorption spectroscopy (XAS) with synchrotron radiation and the time-resolved laser-induced fluorescence spectroscopy (TRLFS) were used as tools for the determination of uranium speciation in soil, water, and different parts of the plants. The reported investigations about the distribution and speciation of uranium were focused on lupins.

Radionuklidkontaminationen können über den Belastungspfad Boden-Pflanze-Nahrungkette deutlich zur Strahlenbelastung des Menschen beitragen. Die Bestimmung des Transfers von Radionukliden aus kontaminierten Böden ist daher für Risikoabschätzungen unbedingt erforderlich. Bisher stand die Bestimmung von Transferfaktoren im Mittelpunkt der Untersuchungen. Um zuverlässigere Abschätzungen über eine Radionuklidmigration im System Boden-Pflanze machen zu können, ist die Charakterisierung des chemischen Zustands der einzelnen Radionuklide notwendig. Bislang ist eine direkte Bestimmung der chemischen Speziation von Uran und anderen Actiniden in Pflanzen aus der Literatur nicht bekannt. Für die Bestimmung der chemischen Speziation von Uran in Pflanzen wurden insbesondere die Lupine und der Löwenzahn auf natürlich und künstlich mit Uran kontaminierten Böden und als Hydrokulturen in uranhaltiger Lösung gezüchtet. Frische und trockene Pflanzenproben wurden vorrangig mit Hilfe der zeitaufgelösten Laserfluoreszenzspektroskopie und der Röntgenabsorptionsspektroskopie als äußerst leistungsstarke und empfindliche Speziationsmethoden untersucht. Über die bisherigen Ergebnisse soll hier berichtet werden

The paper presents a method for the prediction of the radial gas profile in vertical pipe flow for a given bubble size distribution. The method assumes equilibrium of the forces acting on a bubble perpendicular to the flow direction. On basis of the radial profile of the liquid velocity, radial distributions of the volume fraction are calculated separately for all bubble classes according to a given bubble size distribution. The sum of these distributions is the radial profile of the gas fraction. The model is the basis of a fast running one-dimensional steady state computer code. The results of the model are compared with experimental data for a number of gas and liquid volume flow rates.

The hydrate is the simplest chemical form of plutonium in aqueous solution. The knowledge of the structural parameters of the hydration sphere is important for the interpretation of EXAFS results on complicated aqueous plutonium complexes where the water molecules are partly or fully replaced by other ligands. Plutonium(VI) hydrate was prepared by dissolution of PuO2 (Pu-242, AEA Technology, QSA GmbH) and electrochemical oxidation. Part of this solution was reduced to Pu(III) in an electrochemical cell. The Pu(III) and Pu(VI) hydrates were in perchloric and nitric media (1 M acidic solution), respectively. The final Pu concentration was 50 mol/L. The Pu oxidation states were confirmed by UV/Vis spectroscopy. For the measurements 4.7 mL of solution (7.5 MBq)
was filled and sealed in polyethylene cuvettes. The samples were measured at the Rossendorf Beamline ROBL at the European Synchrotron Radiation Facility (ESRF), Grenoble, France within 48 hours after their preparation.

Multiple scans of the Pu LIII-edge EXAFS were collected in transmission mode at room temperature using the Si(111) double-crystal monochromator in fixed-exit mode /1/. The energy scale was calibrated using the first inflection point of the absorption spectrum of a Zr foil (17998 eV). The scattering phases and amplitudes where calculated for hypothetical clusters of PuO8, and PuO2O5 using FEFF6.

Results
As one can see from the XANES spectra given in Fig. 1, the LIII absorption edge of Pu(VI) is shifted by 4 eV toward higher energy as compared to that of Pu(III). The energy shift and the distinct XANES features of these two Pu hydrates can be used for the determination of the Pu oxidation states /2/.

The different electronic and molecular structures of Pu(III) and Pu(VI) hydrates are also reflected in the EXAFS shown in Fig. 2. The coordination sphere of Pu(III) hydrate can be written as Pu(H2O)83+ with an average Pu-O bond distance of 2.48 Å. The Fourier transform corresponding to the EXAFS of Pu(VI) hydrate shows two coordination shells. The Pu(VI) forms a plutonyl ion PuO2(H2O)4-52+. The axial and equatorial Pu-O bond distances are 1.74 and 2.42 Å, respectively. The structural parameters of Pu(VI) hydrate are nearly identical to those of U(VI) and Np(VI) hydrates (see Tab. 1), which were measured recently at ROBL /3/.

Fig. 2: Raw Pu LIII-edge k3-weighted EXAFS spectra (left) and corresponding Fourier transforms (right) of 50 mMol/L Pu hydrates. Solid line - theoretical fit; dots - experiment.

Tab. 1: EXAFS structural parameters for 50 mMol/L Pu hydrates and comparison with 50 mMol/L Np(VI) and U(VI) hydrates.
Sample Shell R(Å) N s2 a)
Pu(III) Pu-O 2.48 7.6(2) 1.02
Pu(VI) Pu -Oax 1.74 1.9 0.12
Pu -Oeq 2.42 4.4(2) 0.50
Np(VI) Np-Oax 1.75 2.0 0.15
Np-Oeq 2.42 4.6(2) 0.56
U(VI) U-Oax 1.76 1.9 0.12
U-Oeq 2.41 4.9(2) 0.61
a) s2 in units of 10-2 Å2

Two principal methods of neutron spectrum adjustment have been compared. Their advantages and problems were discussed. Thereby, the possible use of spectrum adjustment at the LR-0 facility Rez was considered. Recent improvements of the Rossendorf adjustment procedure COSA3 were described. A wider use of spectrum adjustment has been justified by its close relation to uncertainty analysis always required in serious investigations.

The dependence of ion flux, ion energy distribution, and neutral density of a planar radio frequency (RF) driven inductively coupled plasma source on pressure and power is analyzed using an energy selective mass spectrometer and a retarding field analyzer. The ion flux is about 7 mA/cm^2 at 5 Pa and 300 W and increases as RF power and argon pressure increase. The ion energy distribution consists of a single peak with a full width at half maximum of 3 eV for a discharge power in the range from 50 to 300 W and for a pressure in the range from 0.5 to 5 Pa. This indicates that inductive coupling mainly drives the discharge while capacitive coupling between coil and plasma is weak. A significant decrease in Ar neutral density is observed when the plasma is ignited. The Ar depletion increases with increasing RF power and increasing Ar base pressure and reaches 30 % at 5 Pa and 300 W. The decrease in neutral density cannot be explained by the ionization of Ar atoms only but is also attributed to heating of Ar atoms by collisions with energetic particles. A simple model is proposed to understand and to estimate the effect of these collisions on the neutral temperature.

The dependence of the iron nitride formation and the phase transformations in the nitrogen implanted layer on the thickness of the implanted material is investigated by Conversion Electron Mössbauer Spectroscopy (CEMS). Phase transformations in nitrogen-implanted iron layers of various thickness ranging from 60 nm to 870 nm deposited on SiO2 substrate by rf sputtering technique have been studied. Such prepared samples were implanted at room temperature with nitrogen ions with the energy of 50 keV with ion doses ranging from 1x1017 at. N/cm2 to 5x1017 at. N/cm2. After each implantation step the samples were characterized by CEMS using the gas flow electron counter with He-6%CH4 gas. The implanted samples were also investigated by the Grazing Angle X-ray Diffraction (GXRD). The results revealed that the nitride formation is strongly enhanced in thin films as compared to thicker layers or bulk iron samples. In thin iron films a given nitride phases (‘/”, -Fe3N, -Fe2N) are formed at lower nitrogen doses and the transformation of the original iron layer into iron nitrides is more complete than in the bulk -Fe. In bulk iron samples about 20 % of Fe atoms remain in the -Fe phase even for the highest implantation doses (5x1017 at. N/cm2) whereas the complete transformation of thin 60 - 230 nm iron layer into Fe nitrides has been observed.
The observed effects are not fully understood yet. The GXRD experiments performed for the same samples revealed that the stress in the interfacial region of Fe layer and the substrate is significantly larger than in the top (free) surface of the layer. The higher stress may favor the formation of iron nitrides. On the other hand the presence of the interface between Fe film and the substrate may block the in-bulk diffusion of N atoms leading to the higher nitrogen concentration in thin films as compared to the bulk.
The strong dependence of the nitride formation due to N-implantation on the thickness of the implanted material is important for implantation metallurgy, especially for implantation of thin coatings often used for the extension of lifetime of precise tools and details. The phase transformations in thin films may occur in significantly different way as compared to the bulk materials leading to the formation of new phases characterized by extremely high impurity concentration.

Considering the hypothetical core melt down scenario for a Light Water Reactor the behaviour of the Reactor Pressure Vessel during the late phase of the accident has to be investigated. The FOREVER-experiments, currently underway, are simulating the lower head under the load of a melt pool with internal heat sources. Due to the multi-axial creep deformation of the vessel with a non-uniform temperature field these experiments are an excellent possibility to validate numerical creep models. Therefore a Finite Element model is developed on the basis of a commercial multi-purpose code. Using the CFD module the temperature field within the vessel wall is evaluated. The transient structural mechanical calculations are performed using a new numerical approach which avoids the use of a single creep law with constants evaluated for a limited stress and temperature range. Instead of this a three-dimensional array is developed where the creep strain rate is evaluated according to the actual total strain, temperature and equivalent stress. Performing post-test calculations it was found that the assessment of the experimental data and of the numerical results has to be done very carefully. Considering the experiment FOREVER-C2, for example, the recorded creep process appears to be tertiary, if a constant temperature field is assumed. But slight temperature changes during the creep deformation stage could explain the observed creep behaviour, too. Taking into account both - experimental and numerical results - gives a good opportunity to improve the simulation and understanding of real accident scenarios.

In dem Vortrag werden Ergebnisse von Bestrahlungsexperimenten vorgestellt, die im Druckwasserreaktor WWER-2 des KKW Rheinsberg durchgeführt worden sind. Im WWER-2 wurden ISO-V und Bruchmechanikproben aus unterschiedlichen Chargen
von WWER-Grundwerkstoff und Schweißgut bestrahlt. Der Schwerpunkt liegt auf Ergebnissen eines Bestrahlungsprogrammes (WTZBRh), das im Rahmen der wissenschaftlich-technischen Zusammenarbeit zwischen Deutschland und Russland mit den Zielstellungen:der Untersuchung des Einflusses von Begleit- (P, Cu) und Legierungselementen (Ni) auf die bestrahlungsinduzierte Versprödung, der Bewertung der Gültigkeit der Vorhersageformeln des russischen Regelwerkes und der Untersuchung des Ausheilverhaltens durchgeführt worden ist. Die Prüfung der Proben des WTZBRh erfolgte im Rahmen von Förderprojekten des BMBF/BMWi durch das Forschungszentrum Jülich GmbH, die Staatliche Materialprüfungsanstalt Universität Stuttgart und das Forschungszentrum Rossendorf e.V.
Das Bestrahlungs- und Ausheilverhalten der WWER-RDB-Stähle wird auf der Basis der im ISO-V- und Bruchmechaniktests ermittelten Parameter bewertet. Als bruchmechanische Parameter wurden im unteren Spröd-Duktil Übergangsbereich
Referenztemperaturen T0 nach dem Master-Curve Konzept und im duktilen Bereich Risswiderstandskurven und Risseinleitungszähigkeiten ermittelt.

Der beschriebene Gittersensor wurde für eine schnelle Visualisierung von transienten Gasgehaltsverteilungen in einer Zweiphasenströmung entwickelt. Er basiert auf der Messung der momentanen örtlichen elektrischen Leitfähigkeit des Zweiphasengemischs. Der Sensor besteht aus zwei dicht hintereinander angeordneten Elektrodengittern mit jeweils 16 Elektroden. Die Messung erfolgt an den Kreuzungspunkten der Elektroden, d.h. an 16 x 16 Messpunkten, die gleichmäßig über die Querschnittsfläche des Strömungskanals verteilt sind. Die Zeitauflösung des Gerätes beträgt 1024 Messungen in der Sekunde.

Zur Bewertung des Energieertrages einer PV-Anlage ist das Performance Ratio der entscheidende Parameter. Die Einführung des Generatorfaktors als Verhältnis der Generator-Nennleistung und der nominalen Generatorleistung ermöglicht eine einfache Interpretation des Performance Ratio und zugleich die einfache Bestimmung der Generator-Nennleistung aus gemessenen PR-Werten.

Considering the hypothetical accident scenario of a core melt down for a Light Water Reactor the behaviour of the Reactor Pressure Vessel has to be investigated. Therefore the FOREVER-experiments (Failure Of Reactor VEssel Retention) are currently underway. These experiments are simulating the lower head of a reactor pressure vessel under the load of a melt pool with internal heat sources. The geometrical scale of the experiments is 1:10 compared to a common Light Water Reactor.
Due to the multiaxial creep deformation of the three-dimensional vessel with a non-uniform temperature field these experiments are an excellent possibility to validate numerical creep models which are developed on the basis of uniaxial creep tests.
Therefore a Finite Element model is developed on the basis of the comercial multi-purpose code ANSYS/Multiphysics®. Using the
Computational Fluid Dynamic module the temperature field within the vessel wall is evaluated. The transient structural mechanical
calculations are performed applying a creep model which is able to take into account great temperature, stress and strain variations within the model domain. The new numerical approach avoids the use of a single creep law with constants evaluated for a limited stress and temperature range. Instead of this a discrete creep data base is developed where pairs of strain and strain rate are stored for each temperature-stress combination. The creep strain increment is evaluated by linear or non-linear interpolation. Performing calculations for the FOREVER experiments gives a good opportunity to improve the simulation and understanding of real accident scenarios.

The effect of different alcohols on gas hold-up and bubble size in a semi-batch bub-ble column has been investigated. Even little amounts of alcohols added to the liquid preserve the homogeneous bubble flow up to higher gas velocities and gas hold-ups (up to 50 %). For a detailed description it is necessary to divide the flow into two flow regimes - bubble flow and foam. This work presents results from comparative meas-urements with three different methods: needle probes, gamma-densitometry and high speed video observation. The obtained information was used to check the qual-ity of existing foam models. It was found that the vertical gas hold-up profiles in wet unstable alcoholic foams deviate from the theoretical predictions closely above the bubble flow region and again at the top of the foam layer. The first is caused by the effect of inertia hindering the drainage additionally to the viscosity effects, the sec-ond is a consequence of the progressing coalescence. There is a need to improve the existing models for the application in wet unstable foams.

The synthesis is described of uniformly 14C-labelled PCB-congeners starting with [14C]benzene (1) via [14C]nitrobenzene (2), [14C]hydrazobenzene (3), [14C]benzidine dihydrochloride (4), N,N'-diacetyl-[14C]benzidine (5), 3,3'-dichloro-N,N'-diacetyl-[14C]benzidine (6), and 3,3'-dichloro-[14C]benzidine dihydrochloride (7). Elimination of the amino groups from 7 leads to [14C]PCB No. 11 (8), and their substitution by chlorine alternatively to [14C]PCB No.77 (9) /1/. Yields obtained for 8 and 9 are 20 % and 30 %, respectively. The specific activity of the products was 1.2 GBq/mmol. This approach, avoiding the isolation of free [14C]benzidine, may allow the synthesis of PCBs with higher specific activity. The described methods are also applicable to 13C-labelling. From the results of capillary GC-MS analysis /2/ of the unrefined PCBs, conclusions may be drawn about the course of chlorinating the 5: 87 % in 3,3'-position, 6 % in 3,3',5-position, 5 % in 3-position and only 2 % in 2,3'-position.
Because only negligible amounts of position isomers are produced by this method of preparation, the reaction products can be purified much easier with HPLC-techniques than the products obtained from other syntheses.

FZR-Report

The presented wire-mesh sensor produces sequences of instantaneous gas fraction distributions in a cross section with a time resolution of 1200 frames per second and a spatial resolution of about 2-3 mm. At moderate flow velocities (up to 1-2 m/s), bubble size distributions can be obtained, since each individual bubble is mapped in several successive distributions. The method was used to study the evolution of the bubble size distribution in a vertical two-phase flow. In order to assess the accuracy of the described method, a transparent flow channel with an integrated wire-mesh sensor was built. The interaction of gas bubbles with the electrode wires was visually observed. A high-speed video camera was used to record the contact of bubbles with the sensor. The video sequences are compared with the synchronously recorded sensor data. Conclusions are drawn concerning the distortion of the flow by the sensor and the accuracy of the measured gas fractions and bubble size distributions.

Crystal-GRID is a new and complementary method for studying interatomic solid state potentials in the energy range 1 eV to 1 keV. Due to a neutron capture induced photon emission an atom of the crystal starts moving with an energy of some hundreds of eV. Information on the movement of this nuclear probe can be read out by measuring the Doppler shift of a second in-flight emitted photon. The amount of information depends strongly on the nuclear level lifetime in between the two photon emissions and on the interatomic interaction. Systematic theoretical investigations will be presented showing under which requirements, e.g. for count rates or lifetimes, interatomic potentials can be investigated. Experimental results for ZnS and TiO2 will be presented. In these cases it was possible to simultaneously extract the nuclear level lifetime and the screening length of the screened Coulomb potential.

Durch das Forschungszentrum Rossendorf wurde in den Jahren 1996-1998 ein eigenes Strahlrohr fuer Experimente mit Synchrotronstrahlung an der ESRF (European Synchrotron Radioation Facility) in Grenoble/Frankreich aufgebaut. Das Strahlrohr verfuegt ueber zwei alternativ nutzbare Messplaetze fuer die Untersuchung von radioaktiven Proben mittels Roentgenabsorptionsspektroskopie und fuer Materialstrukturuntersuchungen mit Roentgendiffraktion.
Der Bericht konzentriert sich auf die Arbeiten, die fuer die Steuerung der Optik und die Nutzung der Messplaetze hinsichtlich der Elektronik, Rechentechnik und Software erforderlich waren. Nach einer Beschreibung der Randbedingungen und einer Kurzcharakteristik der geraetetechnischen Basis werden wichtige Hardwarekomponenten fuer die Instrumentierung der Systeme vorgestellt. Die rechentechnische Basis wird anschliessend beschrieben. Die angewendeten Software-Grundprinzipien werden erlaeutert und diskutiert sowie an einigen Applikationen beispielhaft verdeutlicht. Abschliessend werden spezifische Probleme bei der Programmierung von Applikationen mit grafischer Bedienoberflaeche in Verbindung mit Geraetezugriffen behandelt. Tabellen, in denen die benutzten Hardware-Module und die Softwarekomponenten zusammengestellt sind, ermoeglichen einen Ueberblick ueber das Gesamtsystem. Das Literaturverzeichnis dient als Leitfaden fuer die Detaildokumentationen.

Irradiation effects on iron disilicide layers were induced by 3.5 keV argon ion bombardment and studied by XPS, AES and the Mössbauer spectroscopy. The use of these electron spectroscopies enables the investigation of different regions of the altered layer.

The study of the interaction processes between humic acids (HAs) and actinide ions is important for a reliable long-term risk assessment with regard to the behavior of actinides in the environment. However, due to the heterogeneous nature of HAs the description of their complex formation with metal ions is difficult and not completely understood up to now.
To further improve the understanding of the HA-metal ion-interaction we studied the influence of phenolic OH groups on the complexation behavior of HAs with uranium(VI) using chemically modified HAs with blocked phenolic OH groups (synthetic HA type M1 and natural HAs Aldrich and Kranichsee). These modified HAs were synthesized by permethylation of the original HAs with diazomethane, and subsequent alkaline saponification of the permethylated HAs. The modified and unmodified HAs were characterized regarding their functional group content. Extensive comparative 13C-CP/MAS-NMR measurements were performed to verify the modification process and to detect whether or not significant structural changes of the HAs had occurred during their modification. The complexation behavior of the modified HAs with UO22+ at pH 4 was studied by time-resolved laser-induced fluorescence spectroscopy (TRLFS) and compared with the results of the UO22+ complexation of the original, unmodified HAs.
The NMR spectra of chemically modified 13C-labeled synthetic HAs type M1 as well as of the modified HAs with carbon-13 at natural abundance level have shown that the permethylation of the HAs with diazomethane and the subsequent alkaline saponification causes the intended structural changes on carboxyl and acidic OH groups. No significant structural changes of the HAs occur during modification.
For all HAs we determined comparable complexation constants by TRLFS. However, significant lower loading capacities with UO22+ were determined for the modified HAs type M1 and Aldrich (10.5 ± 0.9 % and 9.7 ± 1.6 %, respectively) than for the corresponding original, unmodified HAs (18.0 ± 2.0 % and 17.5 ± 1.6 %, respectively). That indicates a lower amount of maximal available complexing sites of the modified HAs compared to the unmodified HAs. Hence, it can be concluded that the modification of phenolic OH groups changes the complexation behavior of HAs with UO22+. Already at pH 4, phenolic OH groups contribute to the interaction between HA and UO22+ ions.

In spite of the fact that a great deal of research has been carried out on liquid metal ion sources, surprisingly few results exist on the temperature dependence of their electric characteristics. In this article we study two liquid metal alloy ion sources (LMAISs), namely Co36Nd64 and Au77Ge14 Si9. While the results of the former alloy were as expected, the latter displayed an entirely different dependence of its electric characteristics on temperature. The unusual results of the Au77Ge14Si9 LMAIS are explained in terms of the abnormal behaviour of its surface tension coefficient with temperature.

We studied the sorption of uranium and humic acid onto phyllite and ferrihydrite both by steady-state experiments in the pH range 3.5 to 9.5 and by kinetic experiments at pH 6.5. Furthermore, the sorption of uranium onto ferrihydrite in the presence of humic acid was studied by extended X-ray absorption fine structure (EXAFS).
Phyllite was used as site-specific rock material because it is closely associated with uranium deposits of the former uranium mining areas in East Germany. Ferrihydrite is known to be formed as secondary mineral phase due to weathering of phyllite. Its sorption behavior was therefore studied and compared to that of phyllite. As humic acid, we applied both a natural humic acid (Kranichsee HA) and a 14C-labeled synthetic humic acid (14C-M1).
The steady-state experiments have shown that humic acid is strongly taken up by both solids over the entire pH range. The uranium sorption is affected both by the pH and by the presence of organic material.
The kinetic experiments have shown that sorption of uranium and humic acid onto phyllite and ferrihydrite is rapid. The initial uranium sorption rates depend on the sequence of addition of uranium and humic acid. For ferrihydrite the initial uranium sorption rate decreases with increasing amounts of uranyl humate complexes in solution. For phyllite the dependence of the uranium sorption rate on the sequence of addition of uranium and humic acid does not exhibit a clear pattern which is attributed to the complex nature of the phyllite rock. The results have shown that the amount of uranium sorbed on solids in equilibrium is determined by the total number of surface sites and by their affinity and accessibility to solutes.
The results of the steady-state and the kinetic experiments show that the sorption behavior of phyllite for uranium and humic acid is dominated by small amounts of ferrihydrite. We conclude that the uranium mobility in an environment with phyllite as a major constituent is controlled by its sorption onto amorphous iron minerals.

A 4 PI silicon ball for detection and identification of light charged particles in large multidetector -arrays as
EUROBALL is presented. The design is based on a N=42 ball with 12 pentagons and 30 hexagons as used in the
GASP array. The absorptive material for -rays is minimized to the detector thickness of 300 or 500µm and a
0.63 mm ceramic backing. The geometrical coverage is designed for about 90% of 4. A pulse shape
discrimination method with totally depleted detectors working in the reverse mount allows identifying protons
and -particles above an energy threshold of about 2MeV. The performances of the ball were tested at the
tandem - booster accelerator combination of the MPI Heidelberg in two experiments using the high-recoil
reaction of 228MeV 58Ni+46Ti and the low-recoil reaction of 95MeV 16O+58Ni. The two-dimensional spectra of
zero-crossing (ZC) versus energy confirmed an excellent discrimination of protons and -particles in all the
detectors at different angles. The energy spectra of protons and -particles measured in the experiments are
presented, too. The -spectra measured in coincidence with various combinations of emitted particles showed a
high selectivity of the ball. The reduced total efficiency for protons of 59% and 55% and -particles of 44% and
32% measured in a nuclear spectroscopy application is analyzed in a Monte-Carlo simulation (GEANT). It is
due to a combined influence of a thick target needed to stop the recoiling residual nuclei and thick absorbers
needed to protect the Si-detectors from scattered beam. The results along with the GEANT extrapolation to
optimum experimental conditions confirm that RoSiB is a highly efficient and selective device for identification
of rare reaction channels with heavy ions.

Im Rahmen einer umfassenden Untersuchung zur inhärenten Borverdünnung nach kleinen Primärkühlmittelsystemlecks für die Anlage Mülheim-Kärlich wurde als konservatives Störfallszenario mit dem größten Kondensatanfall das 20 cm2-Leck in der kalten Hauptkühlmittelleitung hinter der Hauptkühlmittelpumpe ermittelt. Der Borverdünnungsprozeß und die Verlagerung des deborierten Kondensatvolumens zum Kern durch Naturumlauf wurde in thermohydraulischen Systemanalysen detailliert berechnet. Die Mischung des Kondensats nach dem Eindringen in den Reaktordruckbehälter mit der Hochdruckeinspeisung wurde durch Experimente an einer skalierten Versuchsanlage ermittelt. Die Ergebnisse dieser Versuche wurden für eine 3-dimensionale neutronenkinetische dynamische Kernanalyse benutzt. Diese Analyse zeigte, daß der Kern bei dem konservativen Störfallszenario wieder kritisch wird und kurzzeitig Leistung erzeugt, aber eine sehr große Marge zu einzuhaltenden Schutzziel besteht. Weder die Integrität der Brennstofftabletten noch die Integrität der Hüllrohre ist zu irgendeinem Zeitpunkt gefährdet, wie eine abschließende Heißkanalanalyse zeigt, die ein minimales DNB-Verhältnis von 1,53 ausweist.

An overview of the beamline design is given. The basic experimental equipment of the materials research hutch (MRH), i.e. the special six-circle diffractometer for heavy duty, a beam-deflector for investigations of liquids with free surfaces, special sample environments (high-temperature chambers) and the various detector systems (scintillator, photo-diode, 2-dim. CCD camera) will be described.

As represantative results so far obtained at ROBL-MRH the following scientific topics of the Institute of Ion Beam Physics and Materials Research of the Forschungszentrum Rossendorf are presented in more detail:

 Stress relaxation and precipitation of SiC in Si implanted with C;
 X-ray reflectivity and diffuse scattering on Co/Cu-multilayers near the absorption edges;
 Phase transformations studied in-situ during annealing of Ti4Al6V implanted with nitrogen.

Ziel:

Bei der Präzisionstherapie von Schädelbasis-Tumoren mit Strahlen beschleunigter Kohlenstoffionen wird bei jeder Therapiefraktion eine Kontrolle der Dosisdeposition unter Ausnutzung der bei der Bestrahlung erzeugten Positronenemitter vorgenommen.
Methodik:
Durch nukleare Fragmentierungsreaktionen zwischen einem Teil der stabilen ¹²C-Ionen des Therapiestrahles und Atomen des Körpergewebes wird eine charakteristische räumliche und zeitliche Positronenemitterverteilung im Körper des Patienten erzeugt. Eine aus zwei Detektorköpfen bestehende Positronenkamera, die direkt am Bestrahlungsplatz installiert wurde, detektiert während und wenige Minuten nach der etwa zehnnminütigen Bestrahlung eines Feldes mit einer Dosis von ungefähr 0,5 Gray einige zehntausend Koinzidenzereignsse. Da die Verteilungen von Dosis und vom stabilen Strahl erzeugter Aktivität voneinander abweichen, ist eine Berechnung der erwarteten Aktivitätsverteilung für den Vergleich mit der Messung erforderlich. Wegen der sehr geringen Zahl von Ereignissen und des im eingeschränkten Winkelbereich arbeitenden Tomographen erfolgt die Bildrekonstruktion iterativ nach der Maximum-Likelihood-Methode.
Ergebnisse:
Beim deutschen Schwerionentherapie-Pilotprojekt wurden bisher ca. 50 Patienten bei der Gesellschaft für Schwerionenforschung Darmstadt behandelt. Dabei erwies sich die in-situ Kontrolle mittels PET als ein wertvolles Werkzeug für die Verifikation der Bestrahlungspläne. Darüber hinaus wurden Beiträge zu einer Verbesserung der Kalibrierung zwischen CT-Hounsfield-Einheiten und dem Schwerionen-Bremsvermögen geliefert, die Grundlage für die Bestrahlungsplanung ist.
Schlussfolgerungen:
Um dem Anspruch nach höchstmöglicher Präzision bei der Schwerionentherapie gerecht zu werden, ist ein in-situ Monitoring jeder einzelnen Therapiefraktion mit PET unverzichtbar. Bei der gegenwärtigen Planung der klinischen Einrichtung zur Krebsbehandlung mit Ionenstrahlen wird der PET zur Therapiekontrolle ein wichtiger Stellenwert zugemessen.

The condensation in horizontal tubes plays an important role e.g. for the determination of the operation mode of horizontal steam generators of VVER reactors or passive safety systems for the next generation of nuclear power plants. Two different approaches (HOTKON and KONWAR) for modeling this process have been undertaken by Forschungszentrum Jülich (FZJ) and University for Applied Sciences Zittau/Görlitz (HTWS) and implemented into the 1D-thermohydraulic code ATHLET, which is developed by the Gesellschaft für Anlagen- und Reaktorsicherheit (GRS) mbH for the analysis of anticipated and abnormal transients in light water reactors.
Although the improvements of the condensation models are developed for different applications (VVER steam generators - emergency condenser of the SWR1000) with strongly different operation conditions (e.g. the temperature difference over the tube wall in HORUS is up to 30 K and in NOKO up to 250 K, the heat flux density in HORUS is up to 40 kW/m² and in NOKO up to 1 GW/m²) both models are now compared and assessed by Forschungszentrum Rossendorf FZR e.V. Therefore post test calculations of four selected HORUS experiments were performed with ATHLET/KONWAR. It can be seen that the calculations with the extension KONWAR as well as HOTKON improve significantly the agreement between computational and experimental data.

Einleitung:
Bei Patienten mit onkologischer Fragestellung wurde die Abhängigkeit der Bildqualität von Rekonstruktionsparametern und Filtereinstellungen bei der iterativen Bildrekonstruktion (OSEM) und gefilterten Rückprojektion (FBP) an der PET-Kamera ECAT EXACT HR+ und der Software ECAT V7.1 (CTI/Siemens) untersucht. Für Patienten mit unterschiedlichem Körpergewicht bestimmten wir den Einfluß der Glättung durch verschiedene Gaußfilterbreiten auf die visuelle Bildqualität.

Durchführung:
Ausgehend von der Standardkonfiguration für die Einstellung der Rekonstruktions- und Filterparameter wurden bei einem Patienten mit 18F-FDG speichernden kleinen Metastasen die Anzahl der Iterationen und Subsets variiert. Bei der FBP wurde der Einfluß der Cut-Off-Frequenzen des HANNING-Filter untersucht. Die Wirkung der Gaußfilterbreite (Ganzkörper-Protokoll) auf die visuelle Bildqualität wurde bei einem Patienten mit abgegrenzter Metastase, bei Ganzkörperaufnahmen und bei Patienten mit unterschiedlichen Körpergewichten untersucht.

Ergebnisse:
Standardmäßig erfolgt die Rekonstruktion (OSEM) bei einer Iteration und 24 Subsets. Eine Erhöhung der Anzahl von Iterationen bei einer Verringerung der Subsetzahl brachte keine Verbesserung der Bildqualität (Untergrundrauschen, Detailerkennbarkeit), nur ein Verlängerung der Rechenzeit. Eine Variation der Subsetzahl bei einer Iteration zeigte ein Optimum bei 1:24. Die optimale Breite des Gaussfilters bei der abgegrenzten Metastase und bei Patienten mit 50 kg Körpergewicht lag etwa bei 5mm, bei 60-70 kg etwa 7mm, bei 130 kg 9 mm. Die mit veränderten Cut-Off-Frequenzen FBP rekonstruierten Bilder zeigten bei Frequenzen unter 0,3 eine verminderte Detailerkennbarkeit.

Schlussfolgerung:
Eine gute Darstellung wird bereits bei einer Iteration mit 24 Subsets erreicht. Die Gaußfilterbreite sollte in Abhängigkeit vom Patientengewicht und Umfang zwischen 4 und 9 mm eingestellt werden. Die Optimierung der Rekonstruktionsparameter reduziert den Arbeitsaufwand bei bestmöglichster Bildqualität.

Die Kerntechnische Gesellschaft hat die Notwendigkeit zu grundlegenden Reformen erkannt. In der Strategiediskussion "KTG nach 2000" sind hierzu wichtige Ziele sowie die zu Ihrer Umsetzung notwendigen Maßnahmen formuliert. In dem Strategiepapier fehlt jedoch eine klare Konzeption für die organisatorische Umsetzung. Der Vorstand der KTG unterschätzt bzw. verkennt hierbei die Möglichkeiten und Potentiale des Internets. Dies wiegt umso schwerer, da sich in den letzten Jahren der Umgang der Gesellschaft bei der Wahrnehmung und Verarbeitung von Informationen dramatisch verändert hat. Um auch zukünftig in der Gesellschaft präsent und anerkannt zu sein, nach innen sowie außen wirken zu können und speziell Interessenten und Nachwuchs für die KTG anzusprechen und zu gewinnen, ist eine angemessene Internetpräsenz überlebensnotwendig.
Das Internet könnte bereits jetzt Briefe, Rundschreiben, Aushänge, Ankündigungen in der atw und Telephonate gänzlich ersetzen. Auch heute würde ein Verzicht auf diese Kommunikationsmittel die Arbeit der FG bzw. OS nur minimal beeinträchtigen. Denn die Erfahrung zeigt, das engagierte KTG-Mitglieder von sich aus den Kontakt zur FG bzw. OS suchen und pflegen.
Erste Experimente mit Webseiten einzelner Fachgruppen - wie z. B. am Forschungszentrum Rossendorf - zeigen eine äußerst positive Resonanz. Die Seiten werden von den Mitgliedern angenommen und finden selbst bei Nicht-KTG-Mitgliedern eine Resonanz. Die KTG im Internet sollte wesentlich mehr sein als eine nette bunte Webseite, der man Hinweise zu Tagungen und Veranstaltungen entnehmen bzw. über die man mit der Geschäftsstelle, den Vorständen oder Ausschüssen kommunizieren kann. Durch den Aufbau eines effektiven Intranets ließen sich viele der in dem Strategiepapier angesprochenen Maßnahmen umsetzen und Ressourcen einsparen. Zahlreiche Anregungen und Beispiele hierzu finden sich in dem vorliegenden Beitrag. Auch für die KTG gilt derzeit leider noch der aus der Werbung bekannte Satz "Wir haben ja schon immer gewußt, daß wir das Internet brauchen, nur nicht wie dringend".

Due to their excellent mechanical and thermal properties C-based materials are used in High Temperature (HTR) and Fusion Reactor concepts for basic components: Fuel elements and reflectors of HTRs are made from graphite and in Fusion Reactors parts of the plasma facing components within the vacuum vessel consist of C-based materials. Because of their limited oxidation resistance at high temperatures accidents, leading to contact of oxidising gases (air or steam) with these carbon components, have to be considered in safety analyses. In addition, strength loss of carbon materials due to oxidation by impurities in cooling gas is an important aspect for HTR normal operation; accordingly, selection of carbon materials has to take into account their oxidation resistance.
This paper gives an overview on theoretical work and measurements in Jülich concerning oxidation kinetics of selected innovative C-based materials (graphite, CFC) with and without doping in oxygen (523 - 1223 K) and steam (1100 - 1523 K). Also data and theoretical models on depth of penetration of the oxidation process are presented. Data on oxidation induced release of hydrogen isotopes from redeposited a-C:H layers on first walls in fusion reactors are outlined, too.
Futheron, fields of insufficient knowledge are identified: These include data for an improved oxidation model considering both in-pore diffusion and chemical reaction and data/models on catalytic oxidation. Also, a lack of data on the Boudouard reaction, which has to be considered in course of air ingress accidents, was identified.

Studies of the turbulent dispersion of a radiotracer in an experimental setup with natural convection liquid­gaseous flow were carried out. In a series of experiments, a liquid­gaseous bubbly flow was generated in a narrow tank by injection of pressurized air into water, or by catalytic disintegration of H2O2. A small amount of a positron emitting radiotracer liquid was injected instantaneously at a certain point of the tank. The Positron Emission Tomogaphy (PET) technique was used to observe the spreading of this tracer liquid in the bubbly flow. For that case a double head gamma detector array was used for measuring the positron annihilation rate which is proportional to the tracer concentration. From the experimental data the dispersion coefficient D for the tracer liquid was calculated, assuming an isotropic spreading of the concentration profile after separation of the linear displacement of the maximum concentration point.
Calculations for this two­phase flow with spreading tracer were carried out for selected experiments with air injection into water using the computational fluid dynamics code CFX­4.3. Here, the Euler­Euler continuum approach using a homogeneous low Reynolds number K,epsilon­model was applied. The evolution of the bubble size distribution (coalescence and break-up of the bubbles) was taken into consideration. A reasonable agreement was achieved between calculated and measured values of the turbulent dispersion coefficient D. A nearly linear correlation between the gas superficial velocity vgas and D was found in agreement with other authors.

Using the FOPI facility at GSI, charged particles (1\leqZ\leq6) produced in the Au(150 A MeV)+Au reaction have been measured at laboratory angles 1.20 these central collisions, have been extracted: the intensity ratio deduced for a transverse to longitudinal emission is found to be R=1.4+0.2-0.4. Model comparisons using QMD are presented. The value of R appears to depend sensitively on the nucleon-nucleon cross section, nn. Within this model, a value of nn=25±5 mb is derived.

Charged pions are measured with the 4 pi detector FOPI at GSI using the Au on Au reaction at 1.06 AGeV bombarding energy. The pion multiplicities n(pi) increase with the number of participants A(part). The average pion multiplicities per participant are / = 0.0308 and / = 0.0182. These values are only half as large as extrapolated from the low-mass systems studied by Harris et al. The ratio n(pi-)/n(pi+) increases with A(part) and decreases with the pion kinetic energies. The pion kinetic energy spectra have concave shapes, their parametrization in terms of thermal Boltzmann distributions yields a low (T-l,T- pi) and high (T-h,T- pi) temperature which change with the cm emission angle Theta of the pions. In the angular range 45 degrees < Theta < 135 degrees the low temperature is larger than , the high temperatures , are, within experimental uncertainties, the same. The inclusive polar angular distributions of pions are anisotropic, d sigma/d Omega increases for forward and backward angles. The forward - backward enhancements are independent of the pion kinetic energies or the number of participants. In addition to the preferred forward - backward emission, also the enhanced emission into the transverse direction Theta = 90 degrees is observed for pions with high energies or for pions from near-central collisions. These observations and the shape of the rapidity spectra suggest that pions, emitted from the central rapidity region, are partly rescattered by spectator matter. The strength of the rescattering process depends only weakly on the number of participants. The experimental data are compared to the results of IQMD/GEANT calculations using momentum dependent NN interactions and a hard equation of state. The calculated pion multiplicities are approximately 50% larger than experimentally determined; the existence of secondary pion sources is reproduced by the calculation, but their predicted strengths are larger than experimentally observed.

Semi-inclusive triple differential multiplicity distributions of positively charged kaons have been measured over a wide range in rapidity and transverse mass for central collisions of 58Ni with 58Ni nuclei. The transverse mass (mt) spectra have been studied as a function of rapidity at a beam energy 1.93 A GeV. The mt distributions of K+ mesons are well described by a single Boltzmann-type function. The spectral slopes are similar to that of the protons indicating that rescattering plays a significant role in the propagation of the kaon. Multiplicity densities have been obtained as a function of rapidity by extrapolating the Boltzmann-type fits to the measured distributions over the remaining phase space. The total K+ meson yield has been determined at beam energies of1.06, 1.45, and 1.93 A GeV, and is presented in comparison to existing data. The low total yield indicates that the K+ meson can not be explained within a hadro-chemical equilibrium scenario, therefore indicating that the yield does remain sensitive to effects related to its production processes such as the equation of state of nuclear matter and/or modifications to the K+ dispersion relation.

We present new experimental results concerning the azimuthal distributions of proton-likes, light and intermediate mass fragments at midrapidity for Au(100-800 A MeV)+Au collisions measured with the FOPI phase-I detector at GSI in Darmstadt. The azimuthal distributions are investigated as a function of the collision centrality, the incident energy, the fragment charge and transverse momentum. The azimuthal anisotropy is maximum for impact parameters around 7 fm. Intermediate mass fragments present a stronger out-of-plane emission signal than light fragments and a saturation is reached for Z 4. The azimuthal anisotropy increases with the fragment transverse momentum and decreases as the incident energy increases. The azimuthal anisotropy of Z = 2 particles investigated as a function of the scaled fragment transverse momentum follows an universal curve for bombarding energies between 250-800 A MeV. A signature for a transition from in-plane to out-of-plane emission is evidenced at the lowest beam energies.

Collisions of Au on Au at incident energies of 150, 250 and 400 A·MeV were studied with the FOPI-facility at GSI Darmstadt. Nuclear charge (Z15) and velocity of the products were detected with full azimuthal acceptance at laboratory angles 1lab30. Isotope separated light charged particles were measured with movable multiple telescopes in an angular range of 6-90. Central collisions representing about 1% of the reaction cross section were selected by requiring high total transverse energy, but vanishing side flow. The velocity space distributions and yields of the emitted fragments are reported. The data are analysed in terms of a thermal model including radial flow. A comparison with predictions of the quantum molecular model is presented.

Using the FOPI detector at GSI Darmstadt, excitation functions of collective flow components were measured for
the Au+Au system, in the reaction plane and out of this plane, at seven incident energies ranging from 100A MeV
to 800A MeV. The threshold energies, corresponding to the onset of sideward-flow (balance energy) and
squeeze-out effect (transition energy), are extracted from extrapolations of these excitation functions toward
lower beam energies for charged products with Z>= 2. The transition energy is found to be larger than the balance
energy. The impact parameter dependence of both balance and transition energies, when extrapolated to central
collisions, suggests comparable although slightly higher values than the threshold energy for the radial flow. The
relevant parameter seems to be the energy deposited into the system in order to overcome the attractive nuclear
forces.

The hydrolysis of uranium(VI) has been the subject of extensive studies since 50 years and is relatively well-known under acidic and near-neutral conditions. A comprehensive discussion and review of the thermodynamic data is published in [1]. At total uranium(VI) concentrations above 10-4 M and in a slightly acidic (pH: 2 to 5) aqueous media, polymeric cations [e.g., (UO2)2(OH)22+, (UO2)3(OH)5+] are the major species. Structure investigations of such hydrolysis products of uranyl in solutions at environmental relevant uranium concentrations are rare [2].
We investigated the structure of dioxouranium(VI) as a function of pH at different (CH3)4N-OH concentrations with the aid of U LIII EXAFS. Polynuclear hydroxo species were identified by an U-U interaction at 3.808 Å at pH = 4.1.
The speciation of uranium(VI) in the neutral and slightly alkaline pH region is dominated by the precipitation of schoepite. The EXAFS measurements performed at the precipitate formed at pH = 7 showed a schoepite like structure.
The investigation of uranyl hydroxide complexes in alkaline solutions is challenging due to the formation of highly insoluble alkali metal mono- and polyuranate salts. Mononuclear species, UO2(OH)3- and UO2(OH)42-, have been identified mainly by solubility experiments [3] to be the major uranium(VI) complexes at pH > 12. Several attempts have been made to characterize the structure of uranyl complexes under highly alkaline conditions using potentiometric titrations, EXAFS, NMR techniques and quantum chemical methods [4-6]. In solution at high pH [0.5M (CH3)4N-OH], our EXAFS data are consistent with the formation of a monomeric four coordinated uranium(VI) hydroxide complex UO2(OH)42- of octahedral geometry. The first shell contains two O atoms with a U=O distance of 1.830 Å, and four O atoms were identified at a U-Oeq distance of 2.266 Å.
We have extended a previous study [6] by determining the structure of uranium(VI) at (CH3)4N-OH concentration below 1 M over a broad pH-range. The objective was to decide if binary polynuclear hydroxo species also occur in (CH3)4N-OH media. Additional information is provided by comparing the EXAFS results with new 17O-NMR measurements. In strong alkaline solutions [> 1 M (CH3)4N)-OH], 17O-NMR spectra indicate the presence of two species, presumably UO2(OH)42- and UO2(OH)53-, which are in rapid equilibrium with one another at 268 K in aqueous solution. New EXAFS and 17O-NMR data will be presented.

References
1. Grenthe, I., Fuger, J., Konings, R.J.M., Lemire, R.J., Muller, A.B., Nguyen-Trung C., Wanner, H., in: Wanner, H. and Forrest, I. (eds.): Chemical Thermodynamics of Uranium, NEA OECD, Issy-les-Moulineaux, France, p. 241ff (1992).
2. Dent, A.J., Ramsay, J.D.F., Swanton, S.W.: J. Colloid Interface Sci. 150, 45-60 (1992).
3. Yamamura, T., Kitamura, A., Fukui, A., Nishikawa, S., Yamamoto, T., Moriyama, H.: Radiochim. Acta 83, 139-146 (1998).
4. Palmer, D.A., Nguyen-Trung, C.: J. Solution Chem. 24, 1281-1291 (1995).
5. Clark, D.L., Conradson, S.D., Donohoe, R.J., Keogh, D.W., Morris, D.E., Palmer, P.D., Rogers, R.D., Tait, C.D.: Inorg. Chem. 38, 1456-1466 (1999).
6. Wahlgren, U., Moll, H., Grenthe, I., Schimmelpfennig, B., Maron, L., Vallet, V., Gropen, O.: J. Phys. Chem. A103, 8257-8264 (1999).

To get an improved understanding and knowledge of the processes and phenomena during the late phase of a core melt down accident the FOREVER-experiments (Failure of Reactor Vessel Retention) are currently underway. These experiments are simulating the lower head of a reactor pressure vessel under the load of a melt pool with internal heat sources. The geometrical scale of the experiments is 1:10 compared to a common Light Water Reactor. During the first series of experiments the Creep behaviour of the vessel is investigated.
Due to the multi-axial creep deformation of the three-dimensional vessel with a non-uniform temperature field these experiments are on the one hand an excellent possibility to validate numerical creep models which are developed on the basis of uniaxial creep tests. On the other hand the results of pre-test calculations can be used for an optimized experimental procedure.
Therefore a Finite Element model is developed on the basis of the multi-purpose commercial code ANSYS/Multiphysics®. Using the Computational Fluid Dynamic module the temperature field within the vessel wall is evaluated. The transient structural mechanical calculations are performed applying a creep model which is able to take into account great temperature, stress and strain variations within the model domain. The new numerical approach avoids the use of a single creep law with constants evaluated for a limited stress and temperature range. Instead of this a three-dimensional array is developed where the creep strain rate is evaluated according to the actual total strain, temperature and equivalent stress for each element.
Performing post-test calculations for the FOREVER-C2 experiment it was found that the assessment of the experimental data and of the numerical results has to be done very carefully. A slight temperature increase during the creep deformation stage of the experiment for example could explain the creep behaviour which appears to be tertiary because of the accelerating creep strain rate. Taking into account both - experimental and numerical results - gives a good opportunity to improve the simulation and understanding of real accident scenarios.

A joint German/Russian irradiation programme was performed at the prototype pressurized water reactor VVER-2 of the Rheinsberg nuclear power plant (Germany). The experiment comprises about 800 Charpy V-notch, SENB and CT specimens made from 24 different heats of Russian VVER type reactor pressure vessel base and weld metals. The testing of these specimens were performed at Forschungszentrum Rossendorf e.V. (FZR), MPA Stuttgart and the KFA Jülich. This paper presents the results of Charpy V tests performed by FZR. On the base of these results the neutron embrittlement sensitivity and the annealing behaviour are evaluated.

Der Erfindung betrifft ein Verfahren zur Implantation von Ionen gasförmiger Elemente in leitende bzw. halbleitende Werkstücke mittels Plasmaimmersionsimplantation ( PIII ) das ohne aufwendige Hochspannungspulser zu guten Implantationsergebnissen führt.
Das erfindungsgemäße Verfahren besteht darin, daß an die zu implantierenden Werkstücke eine DC-Hochspannung (Gleichspannung > 5kV ) angelegt und das Plasma gepulst wird.
Zur Erzeugung des Plasmas wird vorteilhaft eine Anordnung mit geheizten Filamenten oder eine ECR-Quelle eingesetzt. Das Plasma wird gepulst, indem die Bogenspannung an den Filamenten gepulst wird, oder aber der Mikrowellengenerator der ECR-Quelle im Pulsbetrieb betrieben wird.
Um zu verhindern, daß das das vom auf Hochspannung liegenden Werkstück ausgehende elektrische Feld bis auf die Ionenquelle durchgreift, dort den Austritt von Elektronen unterdrückt und die Zündung des Plasmas erschwert, wird die Plasmaquelle geometrisch so angeordnet bzw. gegenüber der Hochspannung abgeschirmt, daß der Durchgriff der Hochspannung auf die Plasmaquelle minimal ist. Konkret kann dies z. B. dadurch erfolgen, daß die Plasmaquelle an die Implantationskammer mit einem genügend tiefen Hals angeflanscht wird, oder daß zwischen Plasmaquelle und Werkstück in Abständen von einigen cm dünne geerdete Drähte gespannt werden.

Workshop on X-rays from electron beams with special emphasis on possible developments at ELBE

The diagnosis and staging of breast cancer could be improved by the development of imaging radiopharmaceuticals that provide a noninvasive determination of the estrogen receptor status in the tumor cells. Toward this goal, we have synthesized a number of novel Re-containing 7alpha-substituted estradiol complexes. The introduction of the 7alpha side chain involves the alkylation of tetrahydropyranyloxy-protected 6-keto estradiol. The methods used to introduce the rhenium metal involve "3+1" and "4+1" mixed ligand complexes (2a-c and 5, respectively), tricarbonyl dithioether complexes (3), and the cyclopentadienyltricarbonylmetal organometallic system (4ab, 6, 7). These complexes showed binding affinities for the estrogen receptor (as high as 45% for the "3+1" complex 2c) when compared to the native ligand estradiol. The polarity of some complexes (4ab) was modified to improve biodistribution properties by introducing (poly)ether linkages into the 7alpha side chain (6, 7). These complexes provide a further refinement of our understanding of ligand structure-binding affinity correlations for the estrogen receptor, and they furnish the synthetic groundwork for the synthesis of the analogous Tc-99m complexes for evaluation as breast tumor imaging agents.

In a former study we investigated the influence of phenolic OH groups on the complexation behavior of humic acids (HAs) with uranium(VI) using chemically modified synthetic HA (type M1) as well as modified natural HAs (Aldrich and Kranichsee) with blocked phenolic OH groups. These modified HAs were synthesized by permethylation of the original HAs with diazomethane, which converts carboxyl groups to methyl esters and acidic (phenolic) OH groups to methyl ethers and subsequent alkaline saponification of the permethylated HAs. This results in the hydrolysis of ester groups, whereas phenolic OH groups remain blocked.
In continuation, we performed in the present study comparative 13C-CP/MAS-NMR measurements of the modified and unmodified HAs type M1, Aldrich and Kranichsee to verify that there are no significant structural changes of the HAs during their modification. We also studied NMR spectra of alkaline treated HAs type M1, Aldrich and Kranichsee to determine potential structural changes of the HAs which may occur during the alkaline treatment of the permethylated HAs. In addition to the modified and unmodified HAs with carbon-13 at natural abundance we investigated 13C-labeled modified HAs type M1 (permethylated, phenolic OH groups blocked, partly methylated) that were synthesized using [13C]diazomethane as methylation reagent. These studies were performed to verify the modification process and to determine which functional groups of the HAs show the highest affinity for methylation with diazomethane.
From our studies we conclude that only carboxyl and acidic (phenolic) OH groups are methylated with diazomethane, whereby carboxyl groups show the highest affinity for methylation. The alkaline hydrolysis of diazomethylated HAs results in HAs with blocked phenolic OH groups. No significant changes of the overall HA structure occur. This method offers the possibility to synthesize selective modified HAs that can be used to study the influence of various functional groups on the complexation behavior of HAs with metal ions.

We studied the influence of sulfate on the kinetics of uranium(VI) sorption onto ferrihydrite both in the absence and presence of humic acid at pH 6.5 under aerobic conditions since seepage waters from uranium mine tailing piles in Saxony and Thuringia (Germany) contain up to 3·10-2 M sulfate. The sequence in which sulfate, uranium and humic acid were added to the ferrihydrite suspensions was varied in order to obtain more detailed information on the sorption mechanisms. The uranium(VI) and humic acid concentration was 1·10-6 M and 5 mg/L, respectively. As humic acid, we applied the 14C-labeled synthetic humic acid 14C-M1. Two different sulfate concentrations were applied (0.005 and 0.02 M Na2SO4). The ionic strength was always 0.1 M.
The influence of sulfate, present at naturally relevant concentrations (0.005 or 0.02 M), on uranium sorption was found to be different in the absence and presence of humic material.
In the absence of humic acid, both the uranium sorption rate and the total uranium sorption in equilibrium decreased with increasing sulfate concentration. The reason is that the number of strong ferrihydrite sorption sites, accessible for uranium, is reduced upon sulfate adsorption. This leads to an increased availability of uranium.
In the presence of humic acid, and especially in the presence of higher amounts of uranyl humate complexes in solution, the influence of sulfate on uranium sorption is smaller. The reducing effect of sulfate on the total uranium sorption is almost counterbalanced by humic acid.

Summary of the Scientific Activities of the Institute in 1999: Highlight Reports / Short Contributions / Statistics

Basing on data obtained by the incomplete fusion reactions 7Li (43 AMeV) + 232Th and 14N (34 AMeV) + 197Au, the energy dependence of the variance of the fragment mass in fission of highly heated nuclei has been investigated for total excitation energies ranging from 50 up to 350 MeV. This dependence shows some unexpected features when the total excitation energy exceeds a value of about 70 MeV. After this value the steady increase of the variance expected from its temperature dependence changes to some kind of plateau between 100 and 200 MeV. Further on, at energies in excess of about 250 MeV, the variance is found to increase again sharply. In order to analyze this behaviour quantitatively, a dynamical stochastic model has been developed. The model employs the one-body dissipation mechanism and describes the decay of highly excited and rotating nuclei by fission and light particle evaporation. It satisfactory explains the measured prior-to-scission neutron multiplicities and the experimental mass variances up to an energy of about 250 MeV, but the stochastic treatment does not reveal any increase at higher excitation energies in contradiction with the data.

The photoluminescence (PL) and PL excitation (PLE) properties of Sn-implanted SiO2 layers thermally grown on crystalline Si has been investigated and compared with those from Ge- and Si-implanted SiO2 layers. In detail, the violet PL of Sn-implanted SiO2 layers is approximately 2 and 20 times higher than those of Ge- and Si-implanted SiO2 layers, respectively. Based on PL, PLE and decay time measurements the violet PL is interpreted as due to a triplet-singlet transition of the neutral oxygen vacancy (NOV) typical for Si-rich SiO2 and similar Ge- and Sn-related defects in Ge- and Sn-implanted SiO2 films. The enhancement of the blue-violet PL within the isoelectronic row of Si, Ge and Sn will be explained by means of the heavy atom effect.

The microstructural, optical and electrical properties of Si-, Ge- and Sn-implanted silicon dioxide layers were investigated. It was found, that these layers exhibit strong Photoluminescence (PL) around 2.7 eV (Si) and between 3 and 3.2 eV (Ge, Sn) at room temperature (RT), which is accompanied by an UV emission around 4.3 eV. This PL is compared with that of Ar-implanted silicon dioxide and that of Si- and Ge-rich oxide made by rf magnetron sputtering. Based on PL and PL excitation (PLE) spectra we tentatively interpret the blue-violet PL as due to a T1®S0 transition of the neutral oxygen vacancy typical for Si-rich SiO2 and similar Ge- or Sn-related defects in Ge- and Sn-implanted silicon dioxide. The differences between Si, Ge and Sn will be explained by means of the heavy atom effect. For Ge-implanted silicon dioxide layers a strong electroluminescence (EL) well visible with the naked eye and with a power efficiency up to 5·10-4 was achieved. The EL spectrum correlates very well with the PL one. Whereas the EL intensity shows a linear dependence on the injection current over three orders of magnitude, the shape of the EL spectrum remains unchanged. The I-V dependence exhibiting the typical behavior of Fowler-Nordheim tunneling shows an increase of the breakdown voltage and the tunnel current in comparison to the unimplanted material. Finally, the suitability of Ge-implanted silicon dioxide layers for optoelectronic applications is briefly discussed.

A joint German/Russian irradiation programme was performed at the prototype pressurized water reactor VVER-2 of the Rheinsberg nuclear power plant (Germany). The experiment comprises about 800 Charpy V-notch, SENB and CT specimens made of 24 different heats of Russian VVER type reactor pressure vessel base and weld metals. The testing of these specimens was performed at Forschungszentrum Rossendorf e.V. (FZR), MPA Stuttgart and the KFA Jülich. This paper presents the results of Charpy V and fracture toughness tests performed by FZR. On the base of these results the neutron embrittlement sensitivity and the annealing behaviour are evaluated.

The potential of radiolabelled neurotensin ana-logues for in vivo imaging of neurotensin (NT) receptor overexpressing tumours by PET was studied.
Methods: Three newly synthesised ¹⁸F-radiolabelled NT(8-13) agonists, characterised by HPLC, mass spectrometry, and NMR were used for in vitro binding and internalisation stud-ies with HT-29 or WiDr cells. Ca²⁺ mobilisation was investigated using FLIPR measurements. Biodistribution studies were performed in nude mice bearing HT-29 or WiDr tumours without and with blocking of the NT receptors. Biokinetic studies were done in Wistar rats using PET. The specificity of the peptide binding to the tu-mour cells was studied in receptor blocking ex-periments.
Results: The K_D's of the NT analogues were in the nanomolar range. The Ca²⁺ mobilisation ef-fect of the compounds were between 99% and 129% compared to NT with EC50 between 17nM and 120nM. The tumour uptake of the peptides was moderate and was highest for the double stabi-lised peptide {4-[¹⁸F]fluoro-benzoyl-Arg-Psi-(CH2-NH)-Arg-Pro-Phe¹¹-T-Leu¹²-Leu-OH (FB-P3)} (3.5% ID/g tissue 10 min p.i.). It was specifically accumulated in vitro and also in vivo in the tumours (70% inhibition by NT(8-13)). In nude mice the tumour was clearly visible with PET if the location was distant from the excre-tory organs.
Conclusion: The ¹⁸F-labeled stabilised NT analogues are NT agonists and potential radiotrac-ers for imaging of neurotensin receptor ex-pressing tumours. However, structural optimisa-tion is still needed.
(supported by the EC, Contract No. BMH4-CT98-3198)

High-spin states in 72Br were studied with the EUROBALL III spectrometer . The negative-parity band
observed in this experiment displays a signature inversion around spin I = 16. The interpretation within cranked Nilsson-Strutinsky approach shows that this signature pattern is a signal of a substantial triaxial shape change with increasing spin where the nucleus evolves from a triaxial shape with rotation around the intermediate axis at low spin
through an axially symmetric prolate one to a triaxial shape
but with rotation around the shortest principal axis at high spin.
This is the first case in which signature inversion was clearly related to varying gamma deformation.

We report on a nuclear resonance fluorescence experiment on the semi-magic odd-mass fp-shell nucleus 51V. The
detected dipole strength distribution is discussed on the basis of modern shell-model calculations employing a
model space allowing for excitations of protons and neutrons from the 1f7/2 to the (2p1/2,2p3/2,1f5/2) shells. The
calculations indicate that the main body of transitions is of magnetic dipole type, and they are capable of
reproducing the observed distribution well. Possible electric dipole and quadrupole contributions are also
discussed.

We investigate the application of LSO and YSO scintillating crystals for Positron Emission Tomography (PET). Properties such as light output, energy resolution, detection efficiency for various energy threshold, and timing resolution are presented. These data allow us to evaluate the usage of finger-like LSO:Ce and YSO:Ce crystals coupled to photomultiplier tube, and to establish optimal operating conditions for high-resolution PET. Both crystals have advantages over BGO currently used in PET and LSO is considered as a possible replacement for BGO based systems.

Natural IIa diamond was implanted at 90 keV to 1 1015 N+/cm2 and subsequently at 150 keV to 3 1017 Si+/cm2 at a temperature of 900°C. The structure of the implanted diamond region was investigated by high resolution cross-sectional transmission electron microscopy, Raman- and infrared absorption spectrometry. A buried layer with crystalline 3C-SiC domains in perfect epitaxial relation to the diamond substrate was detected. Amorphization and graphitization was completely prevented by the elevated temperature during the implantation. Resistance measurements demonstrated low electrical resistivity in the implanted regions.

A method for the prediction of the radial gas profile for a given bubble size distribution is presented. It is based on the assumption of the equilibrium of the forces acting on a bubble perpendicularly to the flow direction. These forces strongly depend on the bubble size. For the simulation of transient flow regime effects, the modelling of several bubble classes in an 1D model and consideration of their radial profiles seems to be more promising than a detailed 3D modelling.
The radial profile of the liquid velocity is calculated by the model of Sato. On the basis of this velocity profile, radial distributions are calculated separately for all bubble classes according to the given bubble size distribution. The sum of these distributions is the radial profile of the gas fraction. It is used in an iteration process to calculate a new velocity profile. There is a strong interaction between the profiles of liquid velocity and gas volume fraction. The model is the basis of a fast running one-dimensional steady state computer code. The results are compared with experimental data obtained for a number of gas and liquid volume flow rates. There is a good agreement between experimental and calculated data. In particular the change from wall peaking to centre peaking gas fraction distribution is well predicted.

The hydrothermal treatment of zeolite powder leads to the strong changes in the morphology of zeolite particles. In the particles characterised by homogeneous structure before thermal treatment the formation of holes after thermal treatment was observed. These changes lead to the enrichment of zeolite with titanium that enhance its catalytic activity. The titanium surplus precipitates as small TiO2 anatase particles within the holes

Rotational bands in 73Br have been investigated
up to spins of 65/2 using the EUROBALL
III spectrometer. One of the
negative-parity bands displays the highest rotational
frequency 1.85 MeV reported to date in nuclei heavier than 25.
At high frequencies, the experimental dynamic moment of inertia for all bands decrease to very low values, indicating
a lose of collectivity. The bands are described in the configuration-dependent cranked
Nilsson-Strutinsky model. The calculations indicate that
one of the negative-parity bands is observed up to its terminating single-particle state at spin 63/2.
This result establishes the first band termination in the
A =70 mass region.

The stability and reactivity of mixed-ligand ^99mTc complexes of the general formula [^99mTcOL¹L², where L¹H₂ is either an N-substituted bis-(2-mercaptoethyl]amine [SNMeS] or 3-thiapentane-1,5-dithiol [SSS] and L²H is a monodentate thiol [RS], were investigated. The complexes undergo transchelation reactions with glutathione and other SH-group containing blood constituents. The reactions are reversible and can be inhibited by addition of diethylmaleate. Challenge experiments were performed with a broad set of ^99mTc mixed-ligand complexes to investigate the influence of both tridentate ligand and the monodentate ligand on the stability of this type of complex. The occurence of ligand exchange reactions with glutathione depends on the donor set of the tridentate ligands as well as the structure of the monodentate ligands.Especially the stability of complexes containing a monodentate ligand with an amine nitrogen in the side chain can be increased by lengthening the carbon chain between the sulfhydryl group and the nitrogen. Thus, it might be possible to improve their in vivo performance.

Energy filtered TEM (EFTEM) has been applied to study ion beam syn-thesized Ge nanoclusters in 30nm SiO2 thin film thermally grown on Si(100). After thermal annealing the implanted Ge, as show the RBC measurements, was redistributed in to two spatial well separated layers - around the center of SiO2 film and near the SiO2/Si interface. The Ge in the middle of SiO2 film build the well detectable with a bright field TEM the cluster band. But the conventional TEM methods (bright, dark field and EDX line scans) are not able to determine exact Ge location related to the interface the closed cluster band. The energy filtered TEM investigations show that the interface Ge completely located in the Si substrate. The study enable conclusions about influence of the protection layer on Ge distribution in SiO2/Si.

Usually, the microstructure cannot be absolutely surely reconstructed from the small angle scattering pattern produced by it. The small angle neutron scattering curves measured at four C-poor ternary iron alloys were used to demonstrate the effect of the analysis method and the applied assumption on the microstructure analysis. Apart from the mean size of the scatterer, the size distribution as well as the shape of the scatterer affect the results. As a consequence, even well-proved methods as the indirect transformation method developed by Glatter can fail. The application of additional evaluation procedures or the use of information obtained by non-small angle scattering methods increases the chance to derive plausible structure models. In the paper a fit two-parameter method, the ratio between nuclear and magnetic scattering cross section and results from atom probe field ion microscopy are used for that. In this way it is shown that one of the four alloys exhibits another type of irradiation defects. Furthermore, modelling of the mechanical behaviours can also be helpful.

Co/Cu multilayers were prepared by crossed-beam pulsed laser deposition and
characterized by high-angle X-ray diffraction as well as specular and non-specular reflection. The as-deposited structure has
a higher rms roughness than magnetron sputtered Co/Cu multilayer. In
general the roughness of the interface is described by a very large lateral correlation length and a high jaggedness.
Ion-beam mixing of the multilayers with 150 keV Cu ions and a fluence of
up to 5e15 Cu/cm^2 leads to the formation of a granular system of superparamagnetic Co clusters in Cu, as it is shown by magnetic
measurements.

Natural circulation is a basic physical principle for a lot of industrial applications. The paper describes two examples of investigations of natural circulation phenomena. The first example is drawn from the task of the safety assessment of the danger of storage tanks of hazardous fluids under the influence of an external heat source. Experiments were performed to investigate heating up processes of fluids in storage tanks under the influence of an external. As a consequence of an external fire, the heat-up of the inventory may lead to the evaporation of the liquid and to release of significant quantities of dangerous gases into the environment. Several tests were performed both with heating from the bottom and with heating from the side walls. The paper presents experimental and numerical investigations of single and two phase heating up processes of tanks with side wall heating. The measurement of the temperature and of the void fraction makes interesting phenomena evident, which could be explained by an own 2D model.
The second example concerns with investigations of the investigation of passive heat removal systems in advanced designs of nuclear power plants. The heating up processes on the large pool on the secondary side of a test facility NOKO were analysed using the CFD code CFX.

Experiments were performed to investigate heating up processes of fluids in storage tanks under the influence of an external heat source. As a consequence of an external fire, the heat-up of the inventory may lead to the evaporation of the liquid and to release of significant quantities of dangerous gases into the environment. Several tests were performed both with heating from the bottom and with heating from the side walls. In recent tests in addition to thermocouples, the tank was equipped with needle probes for measuring of the local void fraction.
The paper presents experimental and numerical investigations of single and two phase heating up processes of tanks with side wall heating. The measurement of the temperature and of the void fraction makes interesting phenomena evident, which could be explained by an own 2D model. The gained experimental results may be used for the validation of boiling models in 3D CFD codes.

The amorphous non-equilibrium state of the system Fe/Al is studied in bulk and thin-film systems. For bulk samples prepared by ion-beam mixing of laser-deposited multilayers, a bcc solid solution is found
for alloys with up to 70 at.% Al. Around 75 at.% Al a Fe2Al5-like crystalline phase is found. An amorphous state with Fe2Al5-like short-range order is found for 80 at.% Al. In thin films, e.g. the amorphous Al-rich transition layer of a laser-deposited Fe/Al multilayer, the amorphous state can exist for
up to 30 at.% Fe due to the high interfacial energy. By ion-beam mixing at low temperatures (about 140 K) an oversaturation of 40 at.% Fe could be achieved.

Since seepage and flood waters of abandoned uranium mines in Saxony and Thuringia (Germany) contain relatively high amounts of sulfate, we studied the influence of sulfate on the kinetics of uranium(VI) adsorption onto ferrihydrite both in the absence and presence of humic acid at pH 6.5 under aerobic conditions. The sequence in which sulfate, uranium and humic acid were added to the ferrihydrite containing solutions was varied in order to obtain indications to sorption mechanisms. Two different sulfate concentrations were applied (0.005 and 0.02 M Na₂SO₄). The ionic strength was held constant at 0.1 M as sum of NaClO₄ and Na₂SO₄. The uranium(VI) concentration was 1·10^-6 M and the humic acid concentration was 5 mg/L.
In the absence of humic acid, the uranium sorption rate and the total uranium sorption in equilibrium decreased with increasing sulfate concentration due to blocking of strong ferrihydrite surface sites by sorbed sulfate. This leads to an increased availability of uranium. Variations of the order of addition of uranium and sulfate primarily affected the initial uranium sorption rate but not its sorption capacity. Direct competition of uranium and sulfate for strong ferrihydrite surface sites predominates electrostatic effects due to changes in net surface charge of ferrihydrite upon sulfate sorption.
In experiments with humic acid, the competitive effect of sulfate on humic acid sorption at pH 6.5 is small. This is attributed to a higher sorption affinity of humic acid for ferrihydrite surface sites compared to that of sulfate. The initial humic acid sorption rate was not affected by sulfate, only the total humic acid sorption was somewhat reduced by sorbed sulfate. Since the total number of potential binding sites for uranyl ions is increased by sorbed humic acid, more uranium is sorbed in the presence than in absence of humic acid. Thus, the influence of sulfate on uranium sorption is smaller when humic acid is present in solution. This effect becomes even more pronounced with increasing amounts of uranyl humate complexes in solution.

We studied the kinetics of uranium(VI) and humic acid adsorption by the rock material phyllite and by the minerals ferrihydrite and muscovite at pH 6.5 under aerobic conditions. The sequence in which uranium and humic acid were added to the solid containing solutions was varied in order to obtain more detailed information on the sorption mechanisms. The results showed that sorption of uranium and humic acid onto phyllite and ferrihydrite is rapid, that onto muscovite is slower. The initial uranium sorption rates were found to depend on the sequence of addition of uranium and humic acid. For ferrihydrite the initial uranium sorption rate decreased with increasing amounts of uranyl humate complexes in solution. In contrast, addition of humic acid to muscovite resulted in a faster uranium sorption, since the amount of easily accessible surface sites for uranyl ions is enhanced largely by sorbed humic acid. Phyllite showed an overlapping of different processes which is attributed to the complex nature of the phyllite rock. The total uranium uptake decreased in the following order: ferrihydrite > phyllite > muscovite. The results have shown that the amount of uranium sorbed on minerals in equilibrium is determined by the total number of surface sites as well as their affinity and accessibility to solutes. For ferrihydrite indications to mineral dissolution in the presence of humic acid were found. The results of the kinetic experiments confirm that the sorption behavior of phyllite to uranium and humic acid is dominated by minor amounts of ferrihydrite, that is formed as secondary mineral phase due to weathering of phyllite. The contribution of minerals, naturally present in the rock material phyllite, to the total sorption behavior of phyllite could also be shown.
Furthermore, we compared the sorption behavior of two different humic acids, a natural humic acid Kranichsee HA and a 14C-labeled synthetic humic acid type M1, onto phyllite in the pH range 3.5 to 9.5 by batch experiments. Their influence on uranium(VI) sorption onto phyllite was also studied and compared to uranium sorption in the absence of humic acid. Results revealed that sorption of humic acids and their effect on uranium sorption is influenced by the total content of humic acid functional groups and its fractions that are actually involved in sorption reactions or are available for binding of uranyl ions, together with the molecular size and the amount of aromatic structural elements of humic acids. The results also confirmed that the synthetic humic acid 14C-M1 is suitable for experiments studying the kinetics and reversibility of uranium and humic acid sorption onto minerals in the pH range 3.5 to 7.5 since its sorption behavior and its influence on uranium sorption is comparable to that of natural humic acids.

Bei der Charakterisierung der Kolloidpartikel in Bergwerkswässern stehen sowohl die Partikelgröße als auch die chemische Zusammensetzung der Partikel im Zentrum des Interesses. Die Kolloidunter-suchungen an Bergwerkswässern, die den "Bulk" der fließenden Wässer darstellen (Stollenwässer, Flutungswässer), wurden fortgesetzt. Die bevorzugte Partikelgröße von 100 bis 300 nm bestätigte sich für die "Bulk"-Wässer. Bezüglich der chemischen Aspekte stand, neben dem Verhalten des As, Pb und Cu, besonders das Verhalten des U im Blickpunkt. Es tritt sehr oft "nichtkolloidal" auf, im pH-Bereich zwischen 5 und 6 kann es jedoch auch an Eisenoxyhydroxidpartikel gebunden sein und sich dann zu erheblichen Anteilen "kolloidal" verhalten.

Mit der Untersuchung von Acid Rock Drainage (ARD), deutsch "Lettenwasser", wurden die kol-loidchemischen Arbeiten auf ein andersartiges Bergwerkswasser ausgedehnt. Lettenwässer spielen eine Schlüsselrolle bei der Wassermineralisierung in Altbergwerken. Frage war, ob ein Lettenwasser aus der Himmelfahrt Fundgrube Freiberg (pH 2,7; Sulfatkonzentration 39,5 g/l) Kolloidteilchen des unteren Nanometerbereichs enthält, oder ob es nach gängiger Filtration über ein 0,45-µm-Filter parti-kelfrei ist. Wir fanden eine kleine Menge (weniger als 30 mg/l) an leichtnachweisbaren Kolloidpartikeln von 70 bis 250 nm Größe. Wichtiger war jedoch eine zweite kolloidale Komponente, welche schwerer zu detektieren ist: eine Population von ultrafeinen Partikeln einer Größe von weniger als 5 nm. Die Konzentration dieser Komponente ist ca. 1 g/l. Die ultrafeinen Partikel konnten mit Hilfe der Photonenkorrelationsspektroskopie (PCS) nach "Demaskierung" durch Filtration über Nuclepore-Filter nachgewiesen werden (die Detektion von extrem kleinen Partikeln mittels PCS wird bei Anwesenheit größerer Submicron-Partikel durch optische Maskierung verhindert, denn die Streulichtintensität hängt im Partikelgrößenbereich der Rayleigh-Streuung von der sechsten Potenz des Partikelradius ab). Die feinen Partikel wurden gründlicher mittels Ultrafiltration untersucht. Dabei ergab sich, daß mindestens 15 % des Eisens, 50 % des Arsens und 80 % des Bleis im Letten-wasser kolloidales Verhalten zeigen.

Um die röntgenamorphen, schwer untersuchbaren ultrafeinen Partikel der ARD-Lösung näher zu cha-rakterisieren, versuchten wir, mit thermodynamischen Rechnungen (EQ3/6) die übersättigten Minera-lien festzustellen (dies sind die potentiellen Kolloidbildner). Folgende Substanzen/Systeme kommen als Kolloidbildner in Frage. Für Fe: Ferrihydrit, Schwertmannit, Jarosit oder Hydroniumjarosit. Für Pb: Anglesit (PbSO4) oder Plumbojarosit. Für As: Skorodit oder adsorptive Arsenatbindung durch Oberflä-chenkomplexierung (Inner-Sphere-Komplexe). Eine weitergehende Eingrenzung der chemischen Ver-hältnisse in den Partikeln ist mit EQ3/6-Rechnungen nicht möglich.

Die Aufklärung von Bindungszuständen in den Kolloidpartikeln unterschiedlicher Bergwerkswässer (Lettenwässer, "Bulk"-Wässer) ist das Ziel von XAFS-Messungen an der Rossendorfer Beamline (ROBL) am Synchrotron der ESRF in Grenoble die im Februar 2000 angelaufen sind. Das Problem der Unterscheidung zwischen Partikeln und Mutterlösung wollen wir bei diesen in-situ-Messungen durch Vergleich der EXAFS-Spektren von Lösungen unterschiedlichen Filtrationsgrades zu lösen ver-suchen: a) unfiltrierte Lösungen, b) stark angereichertes Kolloid (Retentate der Ultrafiltration) und c) stark abgereichertes Kolloid (Fitrate der Ultrafiltration). Aussagen über die Bindungszustände in den Partikeln möchten wir dann durch Vergleich mit Spektren von definierten Referenzsystemen gewin-nen. Folgende Referenzsysteme wurden vorbereitet: Schwertmannit, Jarosit, Goethit, Skorodit, Bukowskyit, Arsenat sorbiert an Goethit und Bleisulfat.

Im Laufe von 10 Monaten schied sich aus dem untersuchten Lettenwasser eine makros-kopische Sekundärmineralphase aus. Es handelt sich im wesentlichen um das sulfatreiche Mineral H-Jarosit. Kleine Mengen an schlecht kristallinen Phasen, wie Schwertmannit, können nicht ausgeschlossen werden. Der H-Jarosit bindet signifikante Mengen an As und Blei. Die XRD ergab aber keine Hinweise auf Skorodit oder Bleiminerale.
Die ultrafeinen Partikel (<5nm) in ARD Lösungen könnten ein Zwischenprodukt im Bildungsprozeß des H-Jarosits sein. Dann könnte die Charakterisierung des Niederschlags im gealterten Lettenwasser Hinweise auf die Zusammensetzung der ultrafeinen Partikel im frischen Lettenwasser geben. EXAFS-Messungen zur Aufklärung der chemischen Bindungsverhältnisse in Kolloidpartikeln (<5nm) sind in Arbeit.