[en] In the first part of this thesis it was demonstrated how on a semiclassical base a RPA theory is developed and applied to electron scattering. It was shown in which fields of nuclear physics this semiclassical theory can be applied and how it is to be understood. In this connection we dedicated an extensive discussion to the Fermi gas model. From the free response function we calculated the RPA response with a finite-range residual interaction which we completely antisymmetrize. In the second part of this thesis we studied with our theory (e,e') data for the separated response functions. (orig./HSI)

[en] Empirical scalings of energy confinement are used to predict ignition parameters in a Helias reactor. These scaling laws are: Lackner-Gottardi scaling (LGS), international stellarator scaling (ISS95) and international stellarator scaling with W-7 data only (ISS95_W7). For comparison, tokamak scaling laws (ITER89 and ELMYH92y) are also taken into account. The results show that ISS95 yields a confinement time which is too small forignition. LGS and ISS95_W7, however, are sufficient for ignition. An isotope factor or any improvement by H-mode confinement are not taken intoaccount. The Helias reactor (HSR, R = 22 m, a = 1.8 m, B = 4.75 T. left angle β right angle = 4.4%, P_fus = 3040 MW) requires a confinement time τ_E = 1.7 s. (orig.)

[en] The collaboration between the six tokamak teams that led to a joint global confinement database has been continued. In the paper some features of the second version of this database (ITERH.DB2) are discussed, and an analysis of the thermal confinement time scaling based on these data is provided, which may be interesting from a scientific point of view (the database and its analysis can serve as an empirical benchmark for various plasma physical theories and can also be useful for predicting the performance of future machines). In addition, a threshold database (ITERTH.DB1) has been assembled by the same working group, of which some preliminary analysis providing information complementary to the global confinement scalings is presented. (author). 22 refs, 4 figs, 3 tabs

[en] In order to predict the global energy confinement time in the next generation of large tokamaks it is essential to have data from machines of different sizes and operating parameter regimes. This data can also be used to construct dimensionless scalings and thereby attempt to differentiate between Bohm and gyro-Bohm based transport models. Previously, at the request of the ITER project, H-mode global confinement data was assembled from six machines ASDEX, DIII-D, JET, JFT-2M, PBX-M, and PDX into a single database. This collaboration has continued with the initial database being expanded by extending the plasma parameter space as well as by improving the precision of some of the relevant calculated plasma parameters. This paper summarizes work that has been performed on the newest version (ITERH.DB2) of the confinement database

[en] An analysis of the local ion energy transport is performed for more than one hundred well documented ohmic ASDEX discharges. These are characterized by three different confinement regimes: the linear ohmic confinement (LOC), the saturated ohmic confinement (SOC) and the improved ohmic confinement (IOC). All three are covered by this study. To identify the most important local transport mechanism of the ion heat, the ion power balance equation is analyzed. Two methods are used: straightforward calculation with experimental data only, and a comparison of measured and calculated profiles of the ion temperature and the ion heat conductivity, respectively. A discussion of the power balance shows that conductive losses dominate the ion energy transport in all ohmic discharges of ASDEX. Only inside the q=1-surface losses due to sawtooth activity play a role, while at the edge convective fluxes and CX-losses influence the ion energy transport. Both methods lead to the result that both the ion temperature and the ion heat conductivity are consistent with predictions of the neoclassical theory. Enhanced heat losses as suggested by theories eg. on the basis of η_i modes can be excluded. (orig.)

[en] Spontaneous transitions from low-confinement (L-mode) into high-confinement (H-mode) regimes are supposed to be triggered by turbulence generated poloidal ExB shear flows, so called zonal flows (for a review, see Ref. [1]). According to theoretical models, ExB shear flows act on turbulence and related radial turbulent transport through the shear decorrelation mechanism [2]. In particular, the reduction in transport could be due to the reduction of the radial correlation lengths [2] leading to a reduced radial step size of diffusive transport processes associated with turbulent structures [3] or even due to a deformation or tilt of these structures through changes in the phase relation between density fluctuations (en) and velocity fluctuations (evr = e E^mu⁼B) [4], where B and e E^mu^are the magnetic field strength and the fluctuations in the poloidal electric field, respectively. In this work, the microscopic mechanism is investigated, which leads to the decor- relation of turbulent structures and the reduction of radial turbulent transport due to shear flows. In TJ-K, indications for the presence of zonal flows have not been found, yet. Hence, external plasma biasing is applied in order to generate sheared poloidal E x B flows. Poloidal plasma rotation is driven by currents imposed by a biased electrode [5, 6]. Transitions into regimes with H-mode-like properties have been achieved [7] with biasing. Results of biasing experiments in TJ-K are reported. As a main result it turned out that the cross-phase between E x B velocity and density fluctuations plays the key-role in transport reduction. Spectral energy has been shifted from low frequencies to a coherent high-frequency mode where the cross-phase indicates inward transport. (Author)

[en] A reduction of turbulent transport is an important task on the way of an economic fusion power production. The knowledge gathered on the driving forces and the dynamics of plasma turbulence helps to identify different techniques, which have been successfully applied in fusion experiments to improve confinement. The techniques can be divided into two classes, those related to sheared plasma flows and the others related to an optimisation of the magnetic configuration. The paper gives an overview of the different approaches and aims at an intuitive understanding of the mechanisms leading to reduced transport. (authors)

[en] The confinement behaviour of an L-mode plasma in ASDEX is studied under transient conditions, when the total plasma current is changed on a time scale comparable to the energy confinement time. The investigation reveals a decoupling of the conductivity - from the current density profile and shows that the scaling laws derived for steady state conditions can not be applied. The analysis of the total energy content of the plasma constituents reveals similar temporal behaviour. While the current density diffuses on a time scale determined by the plasma resistivity, the loss of thermal energy is delayed until the central confinement region is affected within the q=1 flux surface; this is correlated with a sudden change in the sawtooth frequency and the shape of the electron temperature profiles. When the current is ramped down, the total energy on the other hand begins to drop slightly earlier; it is also remarkable that the electron energy exchange across the plasma radius due to sawtooth action occurs almost simultaneously on a time scale much shorter than skin time or energy confinement time. The observed delay of the electron energy change indicated by the temperature profiles depends on whether the current is ramped down or up. In the latter case the response of the central electron temperature occurs much faster although the central q-value has not altered yet. The current ramp experiments exhibit an interesting tool for clarifying the different properties and confinement physics of the plasma boundary as compared to the hot core. (author) 3 refs., 5 figs

[en] We calculate in a semiclassical RPA approach the nuclear charge response of ¹²C, ⁴⁰Ca and ⁵⁶Fe in the quasi-elastic peak region for moderate momentum transfers (q=1.0 - 2.0 fm^-1). Using the Gogny force and taking full account of antisymmetrisation effects we find with no free parameters good agreement with the (e,e^') data for ¹²C. However for ⁴⁰Ca and ⁵⁶Fe the missing charge problem persists. Arguments that this may be due to strong influence of 2p-2h states in the isovector channel are advanced

[en] In the transverse spin-isovector channel a ρ-exchange force is compared to a realistic G-matrix interaction. An energy and momentum dependent g' is extracted that can be used together with a ρ-exchange potential in ring approximation yielding for the collective response function in the quasi-elastic peak the same result as a fully antisymmetrized G-matrix interaction. (orig.)