Search

The effects of drifts and associated flows and currents on the width of the parallel heat flux channel (λq) in the tokamak scrape-off layer (SOL) are analyzed using the SOLPS-ITER 2D fluid transport code. Motivation is supplied by Goldston’s heuristic drift (HD) model for λq, which yields the same approximately inverse poloidal magnetic field dependence seen in multi-machine regression. The analysis, focusing on a DIII-D H-mode discharge, reveals HD-like features, including comparable density and temperature fall-off lengths in the SOL, and up-down ion pressure asymmetry that allows net cross-separatrix ion magnetic drift flux to exceed net anomalous ion flux. In experimentally relevant high-recycling cases, scans of both toroidal and poloidal magnetic field (Btor and Bpol) are conducted, showing minimal λq dependence on either component of the field. Insensitivity to Btor is expected, and suggests that SOLPS-ITER is effectively capturing some aspects of HD physics. Absence of λq dependence on Bpol, however, is inconsistent with both the HD model and experimental results. The inconsistency is attributed to strong variation in the parallel Mach number, which violates one of the premises of the HD model. Keywords: Power scrape-off width, SOLPS-ITER, Drift effects, SOL flows

Using the new version of the SOLPS plasma boundary code package, SOLPS-ITER, this paper presents the first ever simulations of the ITER burning baseline H-mode edge plasma with drifts and currents activated. Neon (Ne) seeded discharges for divertor power dissipation are considered. The results for divertor and scrape-off layer (SOL) parameters with and without drifts are compared, both for the SOLPS-ITER simulations and against the earlier SOLPS-4.3 modelling (which did not include a drift description) constituting the bulk of the existing ITER divertor simulation database. Whereas the drift effect on the equatorial midplane (main chamber) density and temperature profiles is moderate, drifts increase the peak heat flow to the outer divertor target. This effect is more pronounced for regimes with low sub-divertor neutral pressure, when even drift-free SOLPS4.3 simulations find strong out-in target power asymmetries. An important conclusion is thus that if ITER operates as expected with partially detached divertor targets, drifts should not influence the power handling, but that in the case of divertor reattachment, they will act to worsen the target loading, increasing the need for development of reliable schemes for detachment control. Comparing SOLPS-4.3 and SOLPS-ITER results for the key peak target heat flux versus sub-divertor neutral pressure operating domain, SOLPS-ITER with drifts predicts a narrower operational window for the divertor pressure.

A numerical study is presented, using the SOLPS-ITER plasma boundary code with full drifts and currents activated, of impurity seeded JET discharges in support of high power H-mode experimental campaigns designed to compare two ITER candidate seeding species, nitrogen (N) and neon (Ne). Fluid-kinetic edge plasma simulations are first performed at lower levels of power into the scrape-off layer (SOL) and benchmarked against existing JET experimental data. Calculations are then performed for higher levels of SOL power to examine the impact of this key parameter on the efficiency of both radiators. From the code point of view, for the chosen JET parameters, Ne can be as efficient a divertor radiator as N at the same level of upstream separatrix Zeff

In the present paper the ASDEX Upgrade (AUG) experimental trend of reaching the radiating X-point with nitrogen seeding is reproduced by SOLPS-ITER code modeling. In these experiments the whole divertor region below the X-point is cooled down by the impurity radiation if the seeding rate is large enough, and the maximal radiation is registered from the X-point region, or even from the confinement zone above the X-point. It is demonstrated that for constant seeding rate SOLPS-ITER simulations of the intensively seeded AUG discharges result in that the confined plasma goes into the radiation collapse as a certain threshold in seeding rate is exceeded. This threshold value increases with increasing discharge power. No stable regimes with the electron temperature below 5 eV in the confinement zone even above the X-point are achieved in the modeling if the seeding rate is large enough, in contrast to the experiment. However, such a regime may be achieved if the fueling, seeding and pumping rates are changing in time. Since the SOLPS-ITER code can simulate only steady state, another modeling strategy is chosen. The fueling and seeding rates are artificially reduced by 3 orders of magnitude and the impurity content is set to satisfy the condition that the ratio of electrons contribution originating from fuel atoms to ones coming from impurity atoms is about unity. It is suggested that the radial width of the cooled region in the confinement zone is of the order of the scrape-off layer width λ q , since it is driven by the same physics leading the energy flux to go from mostly radial to mostly parallel. Under these conditions, the radiative spot above the X-point behaves as the energy sink similarly to the energy sink near the divertor in the conventional regime. In extreme regimes (with large seeding rate), the width of the cold region inside the separatrix may exceed λ q , and up to 90% of discharge power can be radiated from the confined region. An estimate of the poloidal length of the radiative spot is suggested as well. Flow patterns of neutrals, deuterium ions, impurities, electric current and heat flows are analyzed for the regimes with intensive X-point radiation. The formation of an electric potential peak above the X-point is observed in the simulations, and the corresponding E × B drift flux appears to give the largest contribution to the main ion and impurity fluxes. This E × B drift flux together with the large ionization source change the parallel velocity with respect to its neoclassical profile. Consequently, the radial E field deviates from the neoclassical one, which might improve the turbulence suppression.

The SOLPS-ITER transport code was released in 2015 on the basis of older versions of SOLPS and the EIRENE Monte Carlo code for neutrals. In the parallel momentum balance equations of this first release, each of the ion species was combined with the parallel momentum balance for electrons so as to eliminate the electric field term. In doing so, the expressions for the thermal and friction forces were approximated for simplification. Such approximations break down for multi-component plasma modelling. It was subsequently proposed to return to solving the original (Braginskii) ion momentum balance equations, with the electric field term included, and with more accurate treatment of friction and thermal force terms, although the treatment of these terms still assumed trace impurities. In the present work, more general expressions for the friction and thermal forces, valid for high impurity densities (z(eff) - 1 > 1 at the separatrix), are suggested. This new form for the force terms within the Braginskii equation has now been implemented in SOLPS-ITER and has been tested through the use of simulation test cases of nitrogen-seeded scenarios in ASDEX Upgrade geometry, with drifts and currents switched on. This paper discusses the effect of the new form of the terms in the Braginskii equation, emphasizing the differences with the trace impurity approach. It demonstrates that the latter already begins to fail in simulations for which z(eff) = 1.5. The influence of the transition from the old form of the momentum balance equation to the new general Braginskii formulation with corrected friction and thermal force terms is demonstrated by comparing SOLPS-ITER ASDEX Upgrade simulations employing both descriptions.

Modeling of the transition to the detachment of ASDEX Upgrade tokamak plasma with increasing density is performed using the SOLPS-ITER numerical code with a self-consistent account of drifts and currents. Their role in plasma redistribution both in the confinement region and in the scrape-off layer (SOL) is investigated. The mechanism of high field side high-density formation in the SOL in the course of detachment is suggested. In the full detachment regime, when the cold plasma region expands above the X-point and reaches closed magnetic-flux surfaces, plasma perturbation in a confined region may lead to a change in the confinement regime.

A systematical study of E → × B → drift and parallel current effects is reported based on the analysis of the simulations by B2SOLPS5.2 transport code. It is demonstrated that divertor asymmetry for connected double null (CDN) configuration is established due to poloidal E → × B → drift and parallel thermal current. For single-null configuration the existing asymmetry is amplified by account of these effects.

Radial electric fields have been simulated using the B2SOLPS5.2 transport code. Simulations were performed for three different times of one H-mode shot of ASDEX-Upgrade with three different levels of heating power. Simulation results were compared with the values of electric field measured by means of He II spectroscopy. It is demonstrated that the radial electric fields obtained in the simulations are close to the neoclassical electric fields as in the previous simulations performed both for ASDEX-Upgrade and MAST for H and L-regimes. The calculated radial electric fields are in reasonable agreement with the field profiles obtained in the experiment. Both simulations and experiment are consistent with the neoclassical character of the radial electric field in the edge barrier. The calculated toroidal rotation at the low field side equatorial midplane is of the same order as measured in the experiment and is co-current directed.

Impurity seeding into a tokamak divertor for radiative cooling is considered as a tool for achieving detached/semi-detached regimes required to meet the condition of acceptable heat loads on divertor plates. Experiments aimed at searching for an operational window with a significant reduction of poloidal heat fluxes due to the impurity radiation and without decrease of confinement are performed on many tokamaks. A critical issue in these experiments is how large a fraction of impurities is retained in the divertor region and how much is extracted upstream to the scrape-off layer. In the present paper a physical mechanism of impurity transport from a divertor towards upstream and back to the divertor is analyzed. It is demonstrated that the widespread concept that the impurity leaks if the parallel thermal force exceeds the friction due to main ions and is retained otherwise—is not correct. In this paper, we contend that the impurity leaks if it crosses the stagnation point of the impurity ion poloidal velocity profile before being ionized, and is retained if it ionizes closer to the target than the location of that stagnation point. Thus the leakage efficiency depends on the relative spatial positions of the impurity atom ionization source and the stagnation point of the impurity ion poloidal velocity profile. The impurity ion poloidal velocity is to large extent the sum of the poloidal projection of its parallel velocity and the E × B drift velocity, where the former is derived from the parallel impurity force balance equation. It is demonstrated that the solution of this equation may be approximated by the balance of friction and thermal forces in all regimes, while other terms are smaller. This allows for expressing the impurity parallel velocity through the main ion one and makes the distribution of the parallel (poloidal) fluxes of the main ions, including Pfirsch–Schluter fluxes and E × B drift fluxes, to be an important element of the impurity transport. It is shown that the impurity distribution in the edge plasma is rather sensitive to the value of the impurity ion ionization rate. This analysis is supported by simulation results obtained for the ASDEX Upgrade tokamak with various seeding rates of N and Ne with the SOLPS-ITER code. The importance of the inclusion of self-consistent drift flows is demonstrated by comparison to results of corresponding simulations with the drifts turned off.

Reported is the first modeling of ITER H-mode. Simulations were made with the code B2SOLPS5.2 with drifts and currents, which was specially developed for simulation of tokamak H-regimes. The emphasis is made on the pedestal and edge transport barrier regions. It is demonstrated that the specific feature of the ITER discharges is the density profile in the edge barrier region typical for the low particle sources in the main part of the barrier. The density, temperature and radial electric field profiles are calculated for two values of the edge barrier width (© 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

A new version of the B2SOLPS5.0 transport code, which is free from numerical problems in the barrier region, has been used to simulate H-mode shots from ASDEX-Upgrade and MAST. The radial electric field inside the edge transport barrier and in the pedestal region is close to the neoclassical prediction. The shear of poloidal E → × B → drift at the inner side of the barrier is close to the value before the transition, while inside the barrier it is significantly larger. It is demonstrated that to match the experimental density and temperature radial profiles the drop in the diffusion coefficient within the barrier should be significantly larger than the drop in the electron heat conductivity.

Modeling of the edge toroidal rotation and its dependence on the edge plasma parameters is performed by means of the B2SOPS5.0 transport code for ohmic shots both for MAST and ASDEX Upgrade configurations. The impact of plasma current, toroidal magnetic field, plasma density, and temperature is investigated. The connection between the toroidal rotation and edge radial electric field is also studied. The results of the simulation are consistent with the parametric dependence predicted analytically.

Modeling of the edge toroidal rotation and its dependence on the edge plasma parameters is performed by means of the B2SOPS5.0 transport code for Ohmic shots both for MAST and ASDEX Upgrade configurations. The impact of plasma current, toroidal magnetic field, plasma density and temperature is investigated. A connection between the toroidal rotation and edge radial electric field is also studied. The results of the simulation are consistent with the parametric dependence predicted analytically.

Simulations have been performed to investigate the impact of gas puffing position on plasma toroidal rotation. It was demonstrated that inboard puffing generates counter-current rotation in agreement with experimental observations. On the basis of the simulations a physical model was put forward, which is based on momentum transport by vertical ∇B drifts and anomalous radial transport. Toroidal rotation is generated mainly due to radial transport of the inboard/outboard particle fluxes by ∇B drifts. Its value depends on the location, radial and poloidal dimensions of the ionization source on the closed field lines, increases linearly with the ion temperature and plasma density and is inversely proportional to the plasma current.

The impurity transport in the edge plasma of a divertor tokamak is simulated by means of B2-SOLPS5.0 2D multi-fluid code where diamagnetic and E → × B → drifts for all species are taken into account. It is demonstrated that these drifts to large extent determine the poloidal distribution of impurities near the separatrix and hence penetration of impurities to the core region. The influence of the drifts on the low ionized impurity distribution in the divertor region is more modest.

Mechanisms of the impurity transfer inside a tokamak separatrix have been studied. It is shown that poloidal rotation significantly influences the poloidal distribution of impurities. An analysis of the neoclassical component of the radial impurity flux showed that this fraction is described by an expression that is more complicated than the standard neoclassical variant.

In this paper, recent progress in plasma edge modelling is presented, using the SOLPS B2–Eirene Scrape-Off Layer Plasma Simulation code. The code capabilities have been extended so that is it now possible to investigate reliably regimes dominated by strong neutral sources in the plasma edge. Two main improvements are reported. First, a fluid plasma treatment including drift and currents effects has been coupled to a Monte-Carlo treatment for the neutrals. Concurrently, a grid adaptation algorithm compatible with the coupling is demonstrated. We present simulations taking advantage of these new tools, and discuss their impact on the results. 2003 Elsevier Science B.V. All rights reserved.

New form of momentum balance equations for ion species is suggested and implemented into B2SOLPS5.2 code. The original form of Braginskii parallel momentum balance equations is used with parallel electric field on the left hand side which does not require any assumptions in contrast to the existing momentum balance equations solved in SOLPS codes. General expressions for friction and thermal force which are valid for arbitrary densities of different ions are suggested. Convergence of the simulations is the same as in existing B2SOLPS versions. It is demonstrated that results for arbitrary impurity density differs from that obtained for trace impurities due to correct account of friction and thermal forces.

Simulation of an H-mode ASDEX Upgrade shot with boron impurity was done with the B2SOLPS5.2 transport code. Simulation results were compared with the unique experimental data available for the chosen shot: radial density, electron and ion temperature profiles in the equatorial midplanes, radial electric field profile, radial profiles of the parallel velocity of impurities at the low-field side (LFS) and high-field side (HFS), radial density profiles of impurity ions at LHS and HFS. Simulation results reproduce all available experimental data simultaneously. In particular strong poloidal HFS?LFS asymmetry of B5+ ions was predicted in accordance with the experiment. The simulated HFS B5+ density inside the edge transport barrier is twice larger than that at LFS. This is consistent with the experimental observations where even larger impurity density asymmetry was observed. A similar effect was predicted in the simulation done for the MAST H-mode. Here the HFS density of He2+ is predicted to be 4 times larger than that at LHS. Such a large predicted asymmetry is connected with a larger ratio of HFS and LFS magnetic fields which is typical for spherical tokamaks. The HFS/LFS asymmetry was not measured in the experiment, however modelling qualitatively reproduces the observed change of sign of He+parallel velocity to the counter-current direction at LFS. The understanding of the asymmetry is based on neoclassical effects in plasma with strong gradients. It is demonstrated that simulation results obtained with account of sources of ionization, realistic geometry and turbulent transport are consistent with the simplified analytical approach. Difference from the standard neoclassical theory is emphasized.

Because of the complexity of the physics in the edge of divertor tokamaks, extrapolation from present-day machines to future reactor or reactor-type devices is done using two-dimensional-edge codes. Amongst the issues which are important for future machines, are those of inboard/outboard divertor power asymmetries and the compression of hydrogen and helium. The issues driving the in-out power asymmetry are explored, and a reasonable match is found to existing experiments, as well as an `adjustable knob' that could be used to further improve the match. In this work, one further effect that is known to play a role - that of E×B and diamagnetic drifts - is identified but not analysed, because the codes have not yet been able to include the drift terms in combination with the impurities and kinetic neutrals. It is also shown that the codes reproduce the fact the only small differences were seen on the ASDEX Upgrade experiment in the compression of helium when the DivII divertor was replaced by the DivIIb configuration.

The current-voltage characteristics of a biased electrode (flush mounted probe) and the potential and current distributions are analysed for three basic mechanisms of conductivity across the magnetic field (viscosity, inertia, ion-neutral collisions) in a fully ionized plasma for probe sizes larger than the ion gyro-radius. The analysis was performed both analytically (for small probe potentials) and numerically. It is shown that the slope of the transitional part of the I-V characteristics is approximately the same for all types of conductivity, provided the probe current is normalized to the ion sonic flow to the return current collecting area, while the probe voltage is measured in electron temperature units. The characteristic scale of the plasma perturbed region along the magnetic field is of the order of λmfp(mi/me)1/2 for all three mechanisms considered. The transverse scale is essentially determined by the type of transverse conductivity.

The numerical codes ASTRA and B2SOLPS5.2 are coupled to perform an integrated modeling of particle and energy transport and to obtain continuous self-consistent profiles of the main plasma parameters from the magnetic axis up to target plates. The unique distinguishing feature of the new coupling scheme is the presence of a region of overlap of the 1D and 2D computational domains, where the 1D solution coincides with the 2D one at the equatorial midplane. In the 2D transport equation system, all relevant drift flows and currents are taken into account, which allows us to calculate the poloidal variation of the density, temperatures and electrostatic potential, and obtain neoclassical radial fluxes in a self-consistent manner. Such an approach allows us to model tokamaks for which neoclassical effects give a significant contribution to the ion heat transport, and in particular, spherical tokamaks.

The I-V characteristics in a fully ionized plasma are calculated for the probes larger than the ion gyroradius. The shape of the characteristics is crucially effected by the transverse conductivity caused by the perpendicular ion viscosity.

Several simulations of the L-regimes of ASDEX Upgrade edge plasma have been performed with the B2 SOLP S5.0 2D fluid transport code for a wide range of parameters. A conclusion has been made that the radial electric field in the separatrix vicinity is close to the neoclassical value. The poloidal E × B drifts and compensating parallel fluxes in the scrape-off layer are larger in the H-mode due to steeper gradients while the qualitative pattern of the flows is similar to that of the L-mode.

The Super-X divertor edge plasma of the future MAST Upgrade tokamak (Lisgo et al 2009 36th EPS Plasma Physics Conf. 33E O4-046, Katramados et al 2011 Fusion Eng. Des. 86 1595, Morris 2012 IEEE Trans. Plasma Sci. 40 682) was simulated with the B2SOLPS5.2 transport code including, for the first time, the effects of drifts due to electric field and magnetic field gradients. The previous simulations did not contain drift effects as well as an account of the external transport barrier. The expected reduction in temperatures and heat flux densities at the low-field side divertors was obtained in the simulations. However, an account of drifts and parallel currents led to an up–down asymmetry of the power to the plates in the connected double null configuration, which was not observed in the simulations without drifts.

A systematical study of drift and parallel current effects is reported based on the analysis of the simulations by the B2SOLPS5.2 transport code. It is demonstrated that divertor asymmetry is caused or amplified by the poloidal drift and parallel thermal current.

SOLPS-ITER, a comprehensive 2D scrape-off layer modeling package, is used to examine the physical mechanisms that set the scrape-off width () for inter-ELM power exhaust. Guided by Goldston's heuristic drift (HD) model, which shows remarkable quantitative agreement with experimental data, this research examines drift effects on in a DIII-D H-mode magnetic equilibrium. As a numerical expedient, a low target recycling coefficient of 0.9 is used in the simulations, resulting in outer target plasma that is sheath limited instead of conduction limited as in the experiment. Scrape-off layer (SOL) particle diffusivity (D SOL) is scanned from 1 to 0.1 m2 s−1. Across this diffusivity range, outer divertor heat flux is dominated by a narrow (~3–4 mm when mapped to the outer midplane) electron convection channel associated with thermoelectric current through the SOL from outer to inner divertor. An order-unity up–down ion pressure asymmetry allows net ion drift flux across the separatrix, facilitated by an artificial mechanism that mimics the anomalous electron transport required for overall ambipolarity in the HD model. At m2 s−1, the density fall-off length is similar to the electron temperature fall-off length, as predicted by the HD model and as seen experimentally. This research represents a step toward a deeper understanding of the power scrape-off width, and serves as a basis for extending fluid modeling to more experimentally relevant, high-collisionality regimes.

The 2D equation system for an ionized ablatant density integrated along the magnetic field lines and for a polarization potential is solved numerically. The pellet is modelled by a plasma source of given intensity and spatial dimensions. The effects of a non-uniform tokamak magnetic field are taken into account. It has been shown that the vertical drift current inside the ionized ablatant produces the polarization that causes the ablatant drift in the direction of a tokamak major radius. The resulting E*B drifts lead to a rather complicated plasma cloud shape resembling a 'mushroom' with two clouds connected by a narrow 'stem'. The 'stem' is unstable and splits into separate striations.

Recently a scheme for the coupling of the one-dimensional core transport code ASTRA and the two-dimensional edge transport code B2SOLPS was developed, thus providing the integrated modelling of tokamak discharge. Here, this scheme is improved by taking impurities into account and by considering a real flux surface shape using the equilibrium code SPIDER. This integrated modelling is applied to discharges of the spherical tokamak Globus-M to study the dependence of the scrape-off layer (SOL) width and divertor heat loads on the discharge power and the plasma current. Since these values, together with the magnetic field, are relatively small in Globus-M, this study can test the existing scaling against data in a wider range of tokamak operational parameters. The modelling results agree reasonably with Thomson scattering and Langmuir probe measurements and allow, in principle, the determination of the physical mechanisms responsible for the SOL structure formation. It is found that the SOL width is approximately inversely proportional to the plasma current, in agreement with existing experimental scaling, while its dependence on discharge power is found to be quite weak.

The radial electric field Er at the tokamak plasma edge is simulated both with a two-dimensional (2D) fluid code solving the most complete system of transport equations and with five-dimensional (three-dimensional in configuration space and 2D in velocity space) Monte Carlo particle following code. At low to high confinement transition conditions, the E→r×B→ shearing rate is found to be high enough for turbulence suppression even though the field is essentially neoclassical. Here, B→ is the magnetic field. No bifurcation of Er is found.

The transport code B2SOLPS5.2 is used to simulate L- and H-mode discharges on MAST with and without resonant magnetic perturbations (RMP). The simulated variation of the radial electric field (less negative for RMP) and toroidal rotation (spin-up in the co-current direction for RMP) is in agreement with the experiment. The pump-out effect in the L-modes with high and medium plasma density and in the H-mode is caused by the additional neoclassical radial plasma flow in the electric field modified due to the electron loss along the stochastic field lines. The pump-out in the low-density L-mode can be reproduced only by a significant rise in the turbulent transport coefficients. The modelling suggests strong RMP screening. An analytical model for RMP screening is proposed.

The impact of resonant magnetic perturbations (RMPs) on the structure of the edge transport barrier has been studied. A model for the density pump-out mechanism during the stochastization of the plasma edge is proposed. The observed phenomena are explained as a result of the impact of the ambipolar electric field, which is modified during RMP, on the particle fluxes in the pedestal region. It is demonstrated that the rise of the particle fluxes inside the transport barrier leads to the pump-out effect on density, while the pedestal temperature increases in spite of the big electron heat conductivity in the stochastic magnetic field. The latter is not sufficient to change significantly turbulent heat conductivity in the barrier region and only compensates the rise of the pedestal temperature caused by the density drop for constant heating power. The analytical approach is supported by results of simulations with the B2SOLPS5.2 2D transport code which uses a full description of particle sources and transport phenomena in the pedestal region. Simulations are performed for ASDEX-Upgrade and MAST configurations for various values of electron stochastic conductivity. The radial electric field with RMPs is predicted to be less negative than without RMP. The density drop and temperature rise in the pedestal region are observed in accordance with the experimental results. Generation of toroidal rotation in the co-current direction is predicted. Extrapolations to ITER are discussed.

In the presence of strong ?-particle heating, active control of the formation or the properties of the edge transport barrier (ETB) might be necessary to meet the confinement and exhaust needs of burning fusion plasmas. In particular, the sensitive dependences on the magnetic configuration and divertor geometry are suitable candidates for controlling the ETB. Using both methods on the Mega Amp?re Spherical Tokamak (MAST) short L-mode periods of a few milliseconds duration have been introduced into strongly heated H-modes.In double null configuration a vertical plasma shift by ?Z < 2.5?cm opposite to the ion ?B-drift direction led to L-mode periods of 15?ms < ?tL < 20?ms length. The peak power load to the targets at the H?L transition was a factor of 2 below that measured during edge localized modes (ELMs). At the H?L transition a rapid change in toroidal He?II velocity indicating a fast collapse of the radial electric field within was observed spanning a wider radial region than the collapse during ELMs. In single null configuration the change in the connection length of the outer divertor leg by 2?m < ?Lc < 4?m led to L-mode periods of 2?ms < ?tL < 5?ms duration. In both cases fluid modelling indicates a more negative Er and consequently an increase in the E ? B shear in the L-mode just inside the last closed flux surface due to the changes to the magnetic geometry, which lead to improved H-mode access.The active control techniques demonstrated here may help to develop suitable scenarios for future devices and improve the understanding of the physics determining the ETB.