24th IAEA Fusion Energy Conference - IAEA CN-197

Oct 7, 2012, 9:30 AM → Oct 13, 2012, 7:00 PM US/Pacific

With a number of next-step fusion devices currently being implemented — such as the International Thermonuclear Experimental Reactor (ITER) in Cadarache, France, and the National Ignition Facility (NIF) in Livermore, USA — and in view of the concomitant need to demonstrate the technological feasibility of fusion power plants as well as the economical viability of this method of energy production, the fusion community is now facing new challenges. The resolution of these challenges will dictate research orientations in the present and coming decades.

The scientific scope of the 24th IAEA Fusion Energy Conference (FEC 2012) is, therefore, intended to reflect the priorities of this new era in fusion energy research. The conference aims to be a platform for sharing the results of research and development efforts in both national and international fusion experiments that have been shaped by these new priorities, and thereby help in pinpointing worldwide advances in fusion theory, experiments, technology, engineering, safety and socio-economics. Furthermore, the conference will also set these results against the backdrop of the requirements for a net energy producing fusion device and a fusion power plant in general, and will thus help in defining the way forward.

Book of Abstracts and Programme abstract_and_programme_version_oct_9.pdf

Conference Announcement Poster cn197_Poster.pdf

Sun, October 7

IFRC Meeting - Upon Invitation Only

12:30 PM Lunch

IFRC Meeting - Upon Invitation Only

Energy Forum

Conference Registration

Registration will be open on Sunday October 7, 2012 from 16:30 to 19:30 as well as on Monday October 8 to Thursday October 11, 2012 from 07:30 to 18:30.

Mon, October 8

7:30 AM Conference Registration

Registration will be open on Sunday October 7, 2012 from 16:30 to 19:30 as well as on Monday October 8 to Thursday October 11, 2012 from 07:30 to 18:30.

Opening and Keynote Presentation

Convener: Mr Tony Taylor (USA)

1 O/1: Opening Address

Speaker: IAEA Representative

2 O/2: Welcome Address

Speaker: Host Government Representatives

3 O/3: Nuclear Fusion Prize Award

Speaker: IAEA Representative

4 O/4: Keynote Presentation: Fusion Energy: For Now, and Forever

Fusion is, of course, mankind’s energy source forever, starting as soon as 2050. For now, our main problem is to get the truth about fusion out to the general public. Without grass-roots support for fusion, governments’ commitments to ITER may be eroded, and budget increases to cover rising costs may be impossible. This is a plea for fusion scientists to help spread the word, countering the extensive press coverage of wind and solar power. Arguments in favor of fusion will be given, as well as news about climate change and the state of renewable energies. The talk will start with reminiscences of the 1958 Geneva and the 1968 Novosibirsk conferences, with photographs.

Speaker: Mr Francis F. Chen (USA)

Slides chen.pdf IAEA_plenary.ppt

10:15 AM Coffee Break

Overview: Inertial & Magnetic Fusion: OV/1

Convener: Mr Anatoly Krasilnikov (Russian Federation)

5 OV/1-1: DIII-D Overview - Research Toward Resolving Key Issues for ITER and Steady-State Tokamaks

The DIII-D Research Program has made significant advances in the physics understanding of key ITER issues and operating regimes important for ITER and future steady-state fusion tokamaks. Edge localized mode (ELM) suppression with resonant magnetic perturbations (RMP) has been now been demonstrated in the ITER baseline scenario at q_95=3.1 by controlling the poloidal mode spectrum of n=3 RMP. Temporal modulation of the n=2 and n=3 RMP toroidal phase reveals a complex plasma response that includes an island-like modulation in T_e consistent with recent theory that predicts such island formation can inhibit the pedestal expansion. Pellet pacing experiments with injection geometry similar to that planned for ITER produced a ten-fold increase in the ELM frequency and a strong reduction in ELM divertor energy deposition. Disruption experiments producing reproducible runaway electron beams (I_RE~300 kA with 300 ms lifetimes) reveal RE dissipation rates ~2x faster than expected and demonstrate the possibility of full RE ramp down with feedback control. Long-duration ELM-free QH-mode discharges have been produced with co-current NBI by using n=3 coils to generate sufficient counter-I_P torque. With electron cyclotron heating, ITER baseline discharges at beta_N=2 and scaled neutral beam injection torque have been maintained in stationary conditions for more than 4 resistive times. Successful modification of a neutral beam line to provide 5 MW of adjustable off-axis injection has enabled sustained operation at beta_N~3 with minimum safety factors well above 2 accompanied by broader current and pressure profiles than previously observed. With qmin above 1.5, stationary discharges with beta_N=3.5 have been extended to 2 tau_R, limited only by available beam energy (power and pulse length). This work was supported by the US Department of Energy under DE-AC52-07NA27344 and DE-FC02-04ER54698.

Speaker: Mr David N Hill (USA)

Slides OV1-1_Hill.pdf Slides from Talk with embedded movie

6 OV/1-2: The Status of the ITER Project

Over the last 2 years, ITER has made the transition from building up the infrastructure (ITER Organisation (IO) staff, Domestic Agencies (DAs)) and completing the design to become a real project. A new management structure was introduced in 2010 and several new communication channels were established with the DAs. To date 69 Procurement Arrangements have been signed with the DAs, representing 75% of the total value to be procured in kind. Thus, a large percentage of the work is now performed in the DAs and in the industry of the ITER members. The work performed by the IO is changing during this transition to oversight of the work performed in the DAs and industry. A Strategic Management Plan (SMP) is used to status the project performance and to flag issues and delays. With the SMP and the matching Detailed Work Schedule, ITER is now in the position to develop recovery plans and to tackle delays at an early stage. Construction is progressing rapidly in Cadarache and factories in Members have begun manufacturing the components: Over 4 years, 300 tons of advanced Nb3Sn for the TF coils have been produced; manufacturing facilities for conductors and magnets have been set up; and full-scale TF radial plate, case and sub-scale winding mock-ups have been manufactured, together with relevant mock-ups of the vacuum vessel, divertor and blanket. The base mat and seismic pads on which the tokamak building will rest are under construction; the electrical yard is close to be finished and the ITER HQ building too. A decision has been taken to start operation with an all W divertor. The main technical reason is to learn to deal with the issues of melt layers and potential flaking in the non-nuclear phase of the operation. Elimination of the CFC/W divertor results in a considerable cost saving which is used to pay for deferred items. ITER is presently well into the construction phase and is facing the problems to be expected in such phases. However, it has a more complex organization and procurement scheme than other large science projects and so the solution of problems tends to be more complicated. The achievements over the last 2 years have shown that an international cooperation such as ITER can work successfully and that it is able to deal with the usual problems which arise in large and technically ambitious construction projects.

Speaker: Mr Osamu Motojima (ITER)

Slides OV1-2_Motojima.pptx

7 OV/1-3: Overview of the JET Results with the ITER-like Wall

The JET programme is strongly focussed on the consolidation of the ITER design choices and the preparation of ITER operation. To this aim, during the last two years the materials of the plasma facing components (PFCs) have been replaced with the same combination foreseen in ITER, namely a combination of Be for the main wall and W for the divertor. The installation of the ILW required more than 3000 tiles to be fitted by remote handling manipulators and interfaces. In addition to the new wall, JET has installed several upgrades to active protection systems and diagnostics, vertical stability control and on heating capability with an increase of the routinely-available neutral beam power from 20 MW up to 30 MW. The JET programme in the first set of campaigns after the shutdown was devoted to the development of operational scenarios with the new plasma facing materials and the investigation of the retention properties. In particular, validation of the predictions for ITER in retention levels, breakdown at low voltage, H-mode threshold, confinement level and erosion rates are part of the on-going studies and results obtained so far demonstrate agreement with initial expectations.

Speaker: Mr Francesco Romanelli (EU)

Slides Overview_JET_new.pdf

8 OV/1-4: The National Ignition Campaign: Status and Progress

Since the completion of the National Ignition Facility (NIF) construction project in March 2009, a wide variety of experiments have been completed in support of NIF’s mission areas: national security, fundamental science, and fusion energy. NIF capabilities and infrastructure are in place with over 50 X-ray, optical and nuclear diagnostic systems and the ability to shoot cryogenic DT layered capsules. NIF, a Nd:Glass laser facility, has operated routinely at 1.45-1.6 MJ of 3ω light with very high reliability since September 2011 and is on track to reach its design goal of 1.8 MJ and 500 TW of ultraviolet light in 2012. The National Ignition Campaign (NIC), an international effort with the goal of demonstrating thermonuclear burn in the laboratory, is making steady progress toward achieving ignition. Experiments in 2011 demonstrated the ability to achieve densities over 500 g/cm2 utilizing precision pulse shaping, along with neutron yields within a factor of 5-6 of those required for entering the regime of strong alpha particle heating. Experiments in 2012 are being carried to further optimize this performance. Other experiments have been completed in support of high-energy-density science, materials equation of state, and materials strength. In all cases, records of extreme temperatures and pressures, highest neutron yield and highest energy densities have been achieved. This paper will provide status update of the unprecedented experimental capabilities of the NIF and describe the progress achieved so far on the path toward ignition.

Speaker: Mr Edward Moses (USA)

Slides OV-1-4_Moses_E..pdf

12:30 PM Lunch

Overview: Magentic Fusion: OV/2

Convener: Mr Yong Liu (China)

9 OV/2-1: Extension of Operation Regimes and Investigation of Three-dimensional Current-less Plasmas in the Large Helical Device

Progress in parameter improvement as well as physical understanding of three-dimensional net-current free plasmas in LHD is overviewed. Efficient ion heating by the upgraded low energy NBI and improvement in ion transport have led to the central ion temperature of 7 keV at the density of 1.5x10^19 m^-3. The pre-wall conditioning by ICRF is effective to achieve high ion temperature. The volume averaged beta value has reached 5.1 % and the high beta state over 4.5% has been maintained for longer than 100 energy confinement times. A new dipole antenna has been installed for ICRF heating, which can control toroidal phasing and excite the fast wave with large wavenumbers. It shows better heating efficiency at higher density than the conventional monopole antenna as expected. Two of ten sections of inboard side helical divertor were modified to a baffled structure as a pilot, and neutral compression of more than 10 times was demonstrated, which agrees with the prediction of 3-D Monte Carlo simulation. A resonant magnetic perturbation (RMP) with m/n = 1/1, which has resonance at the plasma periphery, has been applied to the experiments of divertor detachment, ELM mitigation and penetration of perturbed field. The RMP has a stabilizing effect on detached plasmas by changing the radiation pattern in the edge. A radiating plasma with reduced divertor heat load by a factor of 3-10 can be sustained stably. The RMP has reduced the ELM amplitude and increased the ELM frequency. It is noted that ELMs of LHD are thought to be induced by interchange modes destabilized at the edge while peeling ballooning modes cause ELMs in tokamaks. Two types of intrinsic toroidal rotation have been identified in LHD. One is intrinsic rotation in the counter direction driven by the positive radial electric field near the plasma periphery, and the other is that in the co-direction driven by the ion temperature gradient at half of the plasma minor radius. The existence of non-linearity in the latter rotation strongly suggests that the driving mechanism is Reynolds stress due to turbulence, not neoclassical viscosity. The dynamic response of micro-turbulence to ECH modulation has been studied in terms of the long distance radial correlation of turbulence, which is expected to be the most possible candidate for causing non-local transport.

Speaker: Mr Osamu Kaneko (Japan)

Slides FEC2012_LHD_OV_rev2.ppt

10 OV/2-2: Overview of ASDEX Upgrade Results

The medium size divertor tokamak ASDEX Upgrade possesses flexible shaping capability and versatile heating and current drive systems. Recently the technical capabilities were extended by increasing the ECRH power and by installing 2x8 internal magnetic perturbation coils. Using these coils, reliable suppression of large type I ELMs could be demonstrated in a wide operational window, which opens up above a critical plasma density. The edge plasma parameters are little affected. Nevertheless, the ELMs are replaced by repetitive small scale events, which cause lower energy losses but are sufficient to keep the tungsten concentration in the core plasma at a low level. The temperature in the outer divertor rises moderately during ELM mitigation but the inner divertor remains detached at all times. The pellet fuelling efficiency was observed to increase which opened a path to H-mode discharges with peaked density profiles at line densities clearly exceeding the empirical Greenwald limit. With a maximum total heating of 23 MW, high P/R H-modes with moderate divertor peak power loads below 5 MW/m2 were achieved by feedback-controlled radiation cooling. Owing to the increased ECRH power to 4 MW, H-mode discharges could be studied in regimes with dominant electron heating and low plasma rotation velocities, i.e. under conditions particularly relevant for ITER. At low densities the roles of electron and ion temperatures in the L-H transition could be disentangled. It is found that transitions appear at a critical value of the ion pressure gradient, pointing to the neoclassical radial electric field and the related flow shear as important parameters for the transition. Reynolds stress as additional source of flow shear was investigated by means of Doppler reflectometry. ECRH was used to shape the electron temperature profile and to switch between ITG and TEM dominated turbulence regimes. The relation of the dominant instability with particle, momentum, ion and electron heat transport was assessed in detail. Using probes in the SOL, for the first time electron temperature fluctuations were measured and excellent agreement of the turbulent statistics with GEMR simulations is found. Using a retarding field analyser, fluctuating and average ion temperatures were measured in the far SOL which allows assessing the power flux on the first wall.

Speaker: Mr Ulrich Stroth (Germany)

Slides IAEA12_05.pdf

11 OV/2-3: Overview of Experimental Results and Code Validation Activities at Alcator C-Mod

Recent research on the Alcator C-Mod tokamak has focused on a broad range of scientific issues with particular emphasis on ITER needs and on detailed comparisons between experimental measurements and predictive models. It was possible for the first time to demonstrate quantitative, simultaneous agreement between nonlinear gyrokinetic calculations (GYRO) of impurity transport fluxes and experimental observations for impurity diffusion, convection and ion heat flux. Experiments into self-generated rotation in torque-free C-Mod discharges, found an unexpected connection between momentum and energy transport in which spontaneous flow reversal was observed at a critical density, identical to the transition density between LOC and SOC regimes. It was shown for the first time that ITBs could be created with the aid of self-generated ExB shear flow. Research on ICRF heating focused on understanding and mitigating production of metallic impurities. 3D finite-element antenna modeling predicts significantly reduced parallel electric fields if the antenna were aligned to the total magnetic field. A novel field-aligned antenna has been recently installed and has demonstrated sharply reduced impurity generation. LHCD experiments have shown efficient current drive and creation of electron internal transports barrier via modification of magnetic shear. At higher densities, experiments and modeling with ray tracing and full-wave codes suggest that excessive wave interaction in outer regions of the plasma combined with low single-pass absorption are responsible for markedly lower efficiency. Experiments with I-mode have increased the operating window for this promising ELM-free regime. Extrapolation to ITER suggests that Q~10 could be possible in ITER. H-mode studies have measured pedestal widths consistent with KBM-like instabilities, while the pedestal heights quantitatively match the EPED predictions. Investigations into the physics and scaling of the heat-flux footprint showed a scaling proportional to Te3/2. The width was found to be independent of conducted power, BT or q95 and insensitive to the SOL connection length. As at the midplane, IP is the dominant control parameter. At the same time, heat flux can be reduced to about 10% of the total input power via impurity seeding, while maintaining good H-mode energy confinement even with power close to threshold.

Speaker: Mr Martin Greenwald (MIT)

Slides OV-2-3_2012_IAEA_talk_V2.pptx

12 OV/2-4: Overview of KSTAR Results

After the first H-mode discharge in 2010, the H-mode has been sustained longer and the operational regime of plasma parameters has been significantly extended. Due to the proper tuning of equilibrium configuration and plasma control, values of βN = 1.9 and Wtot = 340 kJ have been achieved with energy confinement time τE = 171 ms. Typical H-mode discharges were operated with the plasma current of 600 kA at the toroidal magnetic field BT = 2 T. L-H transition was obtained with 0.8-1.5 MW of PNBI in double null (DN) configuration and the H-mode lasted up to ~5 sec. The measured power threshold as a function of density shows a roll-over with the minimum value of ~0.8 MW at n ̅_e~2x1019 m-3. Based on the achievement of high beta H-mode, various methods of ELM control has been implemented including RMP(Resonant Magnetic Perturbation), SMBI(Supersonic Molecular Beam Injection), vertical jogging and ECCD injection on the pedestal. We observed various ELM responses, i.e., suppression, mitigation and mode locking depending on the relative phases of the IVC coils. During ELMs suppression by 90 degree phase n=1 RMP, ELMs were completely suppressed. ELM pace making by fast vertical jogging of the plasma column has been also demonstrated. A newly installed SMBI system was also utilized for ELM control and a state of mitigated ELMs for a few tens of ELM periods has been sustained by the optimized repetitive SMBI pulses. A simple cellular automata (Sand-pile) model predicts that shallow deposition near the pedestal foot induced small-sized high-frequency ELMs mitigating large natural ELMs. In addition to the ELM control experiments mentioned above, various physics topics were explored focusing on ITER related physics issues such as the intrinsic rotation driven by RF power injection and sawtooth physics in H-mode. In 2012, goal of the machine performance is the long pulse H-mode over 10 s which would be the first step for validating the physics issues at the normal conductor tokamak.

Speaker: Mr Jong-Gu Kwak (Republic of Korea)

Slides FEC2012-KSTAR-v4.pptx

13 OV/2-5: Progress of Long Pulse and H-mode Experiments on EAST

Significant experimental progress has been made since last IAEA FEC towards improved confinement and high plasma performance regimes under long pulse conditions. In particular, the following key results have been achieved with a combination of RF and LHCD under a very low recycling wall condition: 1MA plasma current, fully steady long pulse diverted plasma entirely driven by LHCD over 100 s, and a new type of stationary H-mode discharges lasted much longer than several ten times the energy confinement time. Experiments have been performed to identify the key role that zonal flows play in mediating the L-H transition and sustaining H-mode confinement. This leads to many new advances in H-mode physics, including the observations of a new limit-cycle oscillation preceding the L-H transition at marginal input power and the kinetic energy transfer between shear flows and ambient turbulences, as well as modulations of a high-frequency broad-band turbulence by oscillating zonal flows in a new small-ELM regime, etc. In addition, clear current filaments have been observed only in the far SOL where the voltage difference of plasma sheath between two divertor plates connected by a filament is large. Furthermore, an initial acceleration of the central toroidal rotation has been observed before the L-H transition, with rotation breaking over a longer time scale, accompanied by momentum loss, at the edge during every type III ELM event. Detailed analysis shows that the magnitude of the edge NTV torque that is needed for rotation breaking is roughly in accord with the experimental observations. This paper presents these recent advances and characteristics of and means to control different H-modes. New advances and issues with enhanced capabilities in present campaign that may arise will also be presented.

Speaker: Mr Baonian Wan (China)

Slides OV-EAST.pdf OV-EAST.ppt

Overview Posters: OV/P

These posters remain in display throughout the conference

14 OV/1-1: DIII-D Overview - Research Toward Resolving Key Issues for ITER and Steady-State Tokamaks

The DIII-D Research Program has made significant advances in the physics understanding of key ITER issues and operating regimes important for ITER and future steady-state fusion tokamaks. Edge localized mode (ELM) suppression with resonant magnetic perturbations (RMP) has been now been demonstrated in the ITER baseline scenario at q_95=3.1 by controlling the poloidal mode spectrum of n=3 RMP. Temporal modulation of the n=2 and n=3 RMP toroidal phase reveals a complex plasma response that includes an island-like modulation in T_e consistent with recent theory that predicts such island formation can inhibit the pedestal expansion. Pellet pacing experiments with injection geometry similar to that planned for ITER produced a ten-fold increase in the ELM frequency and a strong reduction in ELM divertor energy deposition. Disruption experiments producing reproducible runaway electron beams (I_RE~300 kA with 300 ms lifetimes) reveal RE dissipation rates ~2x faster than expected and demonstrate the possibility of full RE ramp down with feedback control. Long-duration ELM-free QH-mode discharges have been produced with co-current NBI by using n=3 coils to generate sufficient counter-I_P torque. With electron cyclotron heating, ITER baseline discharges at beta_N=2 and scaled neutral beam injection torque have been maintained in stationary conditions for more than 4 resistive times. Successful modification of a neutral beam line to provide 5 MW of adjustable off-axis injection has enabled sustained operation at beta_N~3 with minimum safety factors well above 2 accompanied by broader current and pressure profiles than previously observed. With qmin above 1.5, stationary discharges with beta_N=3.5 have been extended to 2 tau_R, limited only by available beam energy (power and pulse length). This work was supported by the US Department of Energy under DE-AC52-07NA27344 and DE-FC02-04ER54698.

Speaker: Mr David Hill (USA)

15 OV/1-2: The Status of the ITER Project

Over the last 2 years, ITER has made the transition from building up the infrastructure (ITER Organisation (IO) staff, Domestic Agencies (DAs)) and completing the design to become a real project. A new management structure was introduced in 2010 and several new communication channels were established with the DAs. To date 69 Procurement Arrangements have been signed with the DAs, representing 75% of the total value to be procured in kind. Thus, a large percentage of the work is now performed in the DAs and in the industry of the ITER members. The work performed by the IO is changing during this transition to oversight of the work performed in the DAs and industry. A Strategic Management Plan (SMP) is used to status the project performance and to flag issues and delays. With the SMP and the matching Detailed Work Schedule, ITER is now in the position to develop recovery plans and to tackle delays at an early stage. Construction is progressing rapidly in Cadarache and factories in Members have begun manufacturing the components: Over 4 years, 300 tons of advanced Nb3Sn for the TF coils have been produced; manufacturing facilities for conductors and magnets have been set up; and full-scale TF radial plate, case and sub-scale winding mock-ups have been manufactured, together with relevant mock-ups of the vacuum vessel, divertor and blanket. The base mat and seismic pads on which the tokamak building will rest are under construction; the electrical yard is close to be finished and the ITER HQ building too. A decision has been taken to start operation with an all W divertor. The main technical reason is to learn to deal with the issues of melt layers and potential flaking in the non-nuclear phase of the operation. Elimination of the CFC/W divertor results in a considerable cost saving which is used to pay for deferred items. ITER is presently well into the construction phase and is facing the problems to be expected in such phases. However, it has a more complex organization and procurement scheme than other large science projects and so the solution of problems tends to be more complicated. The achievements over the last 2 years have shown that an international cooperation such as ITER can work successfully and that it is able to deal with the usual problems which arise in large and technically ambitious construction projects.

Speaker: Mr Osamu Motojima (ITER)

16 OV/1-3: Overview of the JET Results with the ITER-like Wall

The JET programme is strongly focussed on the consolidation of the ITER design choices and the preparation of ITER operation. To this aim, during the last two years the materials of the plasma facing components (PFCs) have been replaced with the same combination foreseen in ITER, namely a combination of Be for the main wall and W for the divertor. The installation of the ILW required more than 3000 tiles to be fitted by remote handling manipulators and interfaces. In addition to the new wall, JET has installed several upgrades to active protection systems and diagnostics, vertical stability control and on heating capability with an increase of the routinely-available neutral beam power from 20 MW up to 30 MW. The JET programme in the first set of campaigns after the shutdown was devoted to the development of operational scenarios with the new plasma facing materials and the investigation of the retention properties. In particular, validation of the predictions for ITER in retention levels, breakdown at low voltage, H-mode threshold, confinement level and erosion rates are part of the on-going studies and results obtained so far demonstrate agreement with initial expectations.

Speaker: Mr Francesco Romanelli (EU)

17 OV/1-4: The National Ignition Campaign: Status and Progress

Since the completion of the National Ignition Facility (NIF) construction project in March 2009, a wide variety of experiments have been completed in support of NIF’s mission areas: national security, fundamental science, and fusion energy. NIF capabilities and infrastructure are in place with over 50 X-ray, optical and nuclear diagnostic systems and the ability to shoot cryogenic DT layered capsules. NIF, a Nd:Glass laser facility, has operated routinely at 1.45-1.6 MJ of 3ω light with very high reliability since September 2011 and is on track to reach its design goal of 1.8 MJ and 500 TW of ultraviolet light in 2012. The National Ignition Campaign (NIC), an international effort with the goal of demonstrating thermonuclear burn in the laboratory, is making steady progress toward achieving ignition. Experiments in 2011 demonstrated the ability to achieve densities over 500 g/cm2 utilizing precision pulse shaping, along with neutron yields within a factor of 5-6 of those required for entering the regime of strong alpha particle heating. Experiments in 2012 are being carried to further optimize this performance. Other experiments have been completed in support of high-energy-density science, materials equation of state, and materials strength. In all cases, records of extreme temperatures and pressures, highest neutron yield and highest energy densities have been achieved. This paper will provide status update of the unprecedented experimental capabilities of the NIF and describe the progress achieved so far on the path toward ignition.

Speaker: Mr Edward Moses (USA)

18 OV/2-1: Extension of Operation Regimes and Investigation of Three-dimensional Current-less Plasmas in the Large Helical Device

Progress in parameter improvement as well as physical understanding of three-dimensional net-current free plasmas in LHD is overviewed. Efficient ion heating by the upgraded low energy NBI and improvement in ion transport have led to the central ion temperature of 7 keV at the density of 1.5x10^19 m^-3. The pre-wall conditioning by ICRF is effective to achieve high ion temperature. The volume averaged beta value has reached 5.1 % and the high beta state over 4.5% has been maintained for longer than 100 energy confinement times. A new dipole antenna has been installed for ICRF heating, which can control toroidal phasing and excite the fast wave with large wavenumbers. It shows better heating efficiency at higher density than the conventional monopole antenna as expected. Two of ten sections of inboard side helical divertor were modified to a baffled structure as a pilot, and neutral compression of more than 10 times was demonstrated, which agrees with the prediction of 3-D Monte Carlo simulation. A resonant magnetic perturbation (RMP) with m/n = 1/1, which has resonance at the plasma periphery, has been applied to the experiments of divertor detachment, ELM mitigation and penetration of perturbed field. The RMP has a stabilizing effect on detached plasmas by changing the radiation pattern in the edge. A radiating plasma with reduced divertor heat load by a factor of 3-10 can be sustained stably. The RMP has reduced the ELM amplitude and increased the ELM frequency. It is noted that ELMs of LHD are thought to be induced by interchange modes destabilized at the edge while peeling ballooning modes cause ELMs in tokamaks. Two types of intrinsic toroidal rotation have been identified in LHD. One is intrinsic rotation in the counter direction driven by the positive radial electric field near the plasma periphery, and the other is that in the co-direction driven by the ion temperature gradient at half of the plasma minor radius. The existence of non-linearity in the latter rotation strongly suggests that the driving mechanism is Reynolds stress due to turbulence, not neoclassical viscosity. The dynamic response of micro-turbulence to ECH modulation has been studied in terms of the long distance radial correlation of turbulence, which is expected to be the most possible candidate for causing non-local transport.

Speaker: Mr Osamu Kaneko (Japan)

19 OV/2-2: Overview of ASDEX Upgrade Results

The medium size divertor tokamak ASDEX Upgrade possesses flexible shaping capability and versatile heating and current drive systems. Recently the technical capabilities were extended by increasing the ECRH power and by installing 2x8 internal magnetic perturbation coils. Using these coils, reliable suppression of large type I ELMs could be demonstrated in a wide operational window, which opens up above a critical plasma density. The edge plasma parameters are little affected. Nevertheless, the ELMs are replaced by repetitive small scale events, which cause lower energy losses but are sufficient to keep the tungsten concentration in the core plasma at a low level. The temperature in the outer divertor rises moderately during ELM mitigation but the inner divertor remains detached at all times. The pellet fuelling efficiency was observed to increase which opened a path to H-mode discharges with peaked density profiles at line densities clearly exceeding the empirical Greenwald limit. With a maximum total heating of 23 MW, high P/R H-modes with moderate divertor peak power loads below 5 MW/m2 were achieved by feedback-controlled radiation cooling. Owing to the increased ECRH power to 4 MW, H-mode discharges could be studied in regimes with dominant electron heating and low plasma rotation velocities, i.e. under conditions particularly relevant for ITER. At low densities the roles of electron and ion temperatures in the L-H transition could be disentangled. It is found that transitions appear at a critical value of the ion pressure gradient, pointing to the neoclassical radial electric field and the related flow shear as important parameters for the transition. Reynolds stress as additional source of flow shear was investigated by means of Doppler reflectometry. ECRH was used to shape the electron temperature profile and to switch between ITG and TEM dominated turbulence regimes. The relation of the dominant instability with particle, momentum, ion and electron heat transport was assessed in detail. Using probes in the SOL, for the first time electron temperature fluctuations were measured and excellent agreement of the turbulent statistics with GEMR simulations is found. Using a retarding field analyser, fluctuating and average ion temperatures were measured in the far SOL which allows assessing the power flux on the first wall.

Speaker: Mr Ulrich Stroth (Germany)

20 OV/2-3: Overview of Experimental Results and Code Validation Activities at Alcator C-Mod

Recent research on the Alcator C-Mod tokamak has focused on a broad range of scientific issues with particular emphasis on ITER needs and on detailed comparisons between experimental measurements and predictive models. It was possible for the first time to demonstrate quantitative, simultaneous agreement between nonlinear gyrokinetic calculations (GYRO) of impurity transport fluxes and experimental observations for impurity diffusion, convection and ion heat flux. Experiments into self-generated rotation in torque-free C-Mod discharges, found an unexpected connection between momentum and energy transport in which spontaneous flow reversal was observed at a critical density, identical to the transition density between LOC and SOC regimes. It was shown for the first time that ITBs could be created with the aid of self-generated ExB shear flow. Research on ICRF heating focused on understanding and mitigating production of metallic impurities. 3D finite-element antenna modeling predicts significantly reduced parallel electric fields if the antenna were aligned to the total magnetic field. A novel field-aligned antenna has been recently installed and has demonstrated sharply reduced impurity generation. LHCD experiments have shown efficient current drive and creation of electron internal transports barrier via modification of magnetic shear. At higher densities, experiments and modeling with ray tracing and full-wave codes suggest that excessive wave interaction in outer regions of the plasma combined with low single-pass absorption are responsible for markedly lower efficiency. Experiments with I-mode have increased the operating window for this promising ELM-free regime. Extrapolation to ITER suggests that Q~10 could be possible in ITER. H-mode studies have measured pedestal widths consistent with KBM-like instabilities, while the pedestal heights quantitatively match the EPED predictions. Investigations into the physics and scaling of the heat-flux footprint showed a scaling proportional to Te3/2. The width was found to be independent of conducted power, BT or q95 and insensitive to the SOL connection length. As at the midplane, IP is the dominant control parameter. At the same time, heat flux can be reduced to about 10% of the total input power via impurity seeding, while maintaining good H-mode energy confinement even with power close to threshold.

Speaker: Mr Martin Greenwald (MIT)

21 OV/2-4: Overview of KSTAR Results

After the first H-mode discharge in 2010, the H-mode has been sustained longer and the operational regime of plasma parameters has been significantly extended. Due to the proper tuning of equilibrium configuration and plasma control, values of βN = 1.9 and Wtot = 340 kJ have been achieved with energy confinement time τE = 171 ms. Typical H-mode discharges were operated with the plasma current of 600 kA at the toroidal magnetic field BT = 2 T. L-H transition was obtained with 0.8-1.5 MW of PNBI in double null (DN) configuration and the H-mode lasted up to ~5 sec. The measured power threshold as a function of density shows a roll-over with the minimum value of ~0.8 MW at n ̅_e~2x1019 m-3. Based on the achievement of high beta H-mode, various methods of ELM control has been implemented including RMP(Resonant Magnetic Perturbation), SMBI(Supersonic Molecular Beam Injection), vertical jogging and ECCD injection on the pedestal. We observed various ELM responses, i.e., suppression, mitigation and mode locking depending on the relative phases of the IVC coils. During ELMs suppression by 90 degree phase n=1 RMP, ELMs were completely suppressed. ELM pace making by fast vertical jogging of the plasma column has been also demonstrated. A newly installed SMBI system was also utilized for ELM control and a state of mitigated ELMs for a few tens of ELM periods has been sustained by the optimized repetitive SMBI pulses. A simple cellular automata (Sand-pile) model predicts that shallow deposition near the pedestal foot induced small-sized high-frequency ELMs mitigating large natural ELMs. In addition to the ELM control experiments mentioned above, various physics topics were explored focusing on ITER related physics issues such as the intrinsic rotation driven by RF power injection and sawtooth physics in H-mode. In 2012, goal of the machine performance is the long pulse H-mode over 10 s which would be the first step for validating the physics issues at the normal conductor tokamak.

Speaker: Mr Jong-Gu Kwak (Republic of Korea)

22 OV/2-5: Progress of Long Pulse and H-mode Experiments on EAST

Significant experimental progress has been made since last IAEA FEC towards improved confinement and high plasma performance regimes under long pulse conditions. In particular, the following key results have been achieved with a combination of RF and LHCD under a very low recycling wall condition: 1MA plasma current, fully steady long pulse diverted plasma entirely driven by LHCD over 100 s, and a new type of stationary H-mode discharges lasted much longer than several ten times the energy confinement time. Experiments have been performed to identify the key role that zonal flows play in mediating the L-H transition and sustaining H-mode confinement. This leads to many new advances in H-mode physics, including the observations of a new limit-cycle oscillation preceding the L-H transition at marginal input power and the kinetic energy transfer between shear flows and ambient turbulences, as well as modulations of a high-frequency broad-band turbulence by oscillating zonal flows in a new small-ELM regime, etc. In addition, clear current filaments have been observed only in the far SOL where the voltage difference of plasma sheath between two divertor plates connected by a filament is large. Furthermore, an initial acceleration of the central toroidal rotation has been observed before the L-H transition, with rotation breaking over a longer time scale, accompanied by momentum loss, at the edge during every type III ELM event. Detailed analysis shows that the magnitude of the edge NTV torque that is needed for rotation breaking is roughly in accord with the experimental observations. This paper presents these recent advances and characteristics of and means to control different H-modes. New advances and issues with enhanced capabilities in present campaign that may arise will also be presented.

Speaker: Mr Baonian Wan (China)

23 OV/3-1: Overview of Physics Results from the National Spherical Torus Experiment

Research on the National Spherical Torus Experiment, NSTX, targets physics understanding needed for extrapolation to a steady-state ST Fusion Nuclear Science Facility, pilot plant, or DEMO. The unique ST operational space is leveraged to test physics theories for next-step tokamak operation, including ITER. Present research also examines implications for the coming device upgrade, NSTX-U. Energy confinement increases as collisionality is reduced by lithium (Li) wall conditioning. Nonlinear microtearing simulations match experimental electron diffusivity quantitatively and predict reduced electron heat transport at lower collisionality. Measured high-k turbulence is reduced in H-mode. Beam-emission spectroscopy measurements indicate that the poloidal correlation length of pedestal turbulence increases at higher electron density and gradient, and decreases at higher Ti. Plasma characteristics change nearly continuously with increasing Li evaporation, and ELMs stabilize due to edge density gradient alteration. Global mode stability studies show stabilizing resonant kinetic effects are enhanced at lower collisionality. Combined radial and poloidal field sensor feedback controlled n = 1 perturbations and improved stability. The disruption probability due to unstable RWMs is reduced at high betaN/li > 11. Greater instability seen at lower betaN is consistent with decreased kinetic RWM stabilization. A model-based RWM state-space controller produced long-pulse discharges exceeding betaN = 6.4 and betaN/li = 13. Precursor analysis shows 99% of disruptions can be predicted with 10ms warning and a false positive rate of only 4%. Disruption halo currents rotate toroidally and can have significant toroidal asymmetry. Global kinks cause measured fast ion redistribution. Full-orbit calculations show redistribution from the core outward and toward Vpar/V = 1. The snowflake divertor configuration enhanced by radiative detachment shows large reductions in steady-state and ELM heat fluxes (peak values down from 7 MW/m2 to less than 1 MW/m2). Non-inductive current fraction (NICF) up to 65% is reached experimentally. NSTX-U scenario development calculations project 100% NICF at Ip = 1MA. Coaxial helicity injection has reduced the inductive startup flux, with L-mode plasmas ramped to 1MA requiring 35% less inductive flux. *Supported in part by US DOE Contract DE-AC02-09CH11466.

Speaker: Mr Steven Sabbagh (USA)

Poster IAEA_FEC12_OV3-1-Sabbagh-v23p.pdf

24 OV/3-2: Overview of Physics Results from MAST towards ITER/DEMO and the Upgrade

New diagnostic, modelling and plant capability on MAST have delivered important results in key areas for ITER/DEMO and the upgrade. Linear gyro-kinetic calculations highlight the interplay of kinetic ballooning and micro tearing modes in setting the pedestal width. Using the upgraded 18 coil ELM control array ELM mitigation by resonant magnetic perturbations (RMP) has now been demonstrated with n=3, 4 and 6 proving the importance of the alignment of the perturbation to the magnetic field lines with the unique phasing capability. Already a 2D equilibrium deformation consistent with lobe structures at the X-point, only observed when the RMPs affect the plasma, destabilises the ELM despite the lower pressure gradient. The interplay of edge flow shear and turbulence at the L-H transition measured by a new 50 kHz Doppler spectroscopy diagnostic shows limit–cycle like dynamics during ~4-5 kHz H-mode dithers. Further insight into dn/dr, j and v may be provided by a novel 3D electron Bernstein emission imaging system. For the first time ELM resolved scrape-off-layer T_i profiles have been characterised simultaneously in the divertor and mid-plane with retarding field analysers. Divertor design is aided by ELM resolved data on the n_e and I_p dependence of target heat flux profiles. Under conditions foreseen for the upgrade (off-axis NBI, q_min>1.3, I_p> 0.8 MA) measurements of profiles of the neutron and fast ion D_alpha emission indicate a substantial reduction of the fast-ion redistribution compared to on-axis injection. To model the fast-ion redistribution due to the wave particle interaction driven by the super Alfvénic beam ions accurately dynamic friction is now included into the modelling recovering the observed non-linear mode behaviour well. Low-k turbulence, measured using a new 2D beam emission spectroscopy diagnostic, shows L-mode turbulence at r/a>0.7 propagating in the ion diamagnetic direction with respect to the local ExB flow and reduced turbulence levels in H-mode. Using in detail event triggering on NTMs the T_e perturbation due to the 2/1 island has been measured allowing detailed assessment of the NTM stability. Further studies of disruption mitigation with massive gas injection compared different gas mixtures and injection locations including a newly developed controllable high-field-side valve. Work supported by UK EPSRC and EURATOM.

Speaker: Mr Hendrik Meyer (UK)

Poster IAEA_2012_MAST_OV3-2_poster.pdf

25 OV/3-3: Overview of HL-2A Recent Experiments

Since the last FEC, the experiments on HL-2A tokamak have been focused on the investigations on H-mode related physics including ELM mitigation, energetic particle physics, transport including nonlocality and edge impurities, MHD instabilities, turbulence and zonal flow physics, etc. In particular, it’s demonstrated for the first time that the supersonic molecular beam injection (SMBI) and cluster jet injection (CJI) can convert large ELMs into a series of small ELMs in HL-2A. After SMBI or CJI, the ELM frequency rises by a factor of 2 to 3 on average and its amplitude decreases by about 50%. With high power ECRH, the energetic particle induced modes have been observed in different frequency ranges. The high frequency (200–350 kHz) modes with relatively small amplitude are close to the gap frequency of toroidacity induced Alfvén eigenmode (TAE). The coexistent multi-mode magnetic structures in the high temperature and low collision plasma will affect the plasma transport greatly. The L-I-H transition has been studied in pure NBI-heating deuterium plasmas. For the first time, the absolute rate of nonlinear energy transfer between turbulence and zonal flows was measured and the secondary mode competition between low frequency zonal flows (LFZFs) and geodesic acoustic modes (GAMs) was identified in experiments, which demonstrated that the energy transfer is large enough to affect the turbulence saturation level and its dynamics and that zonal flows play an important role in the low to high (L-H) plasma confinement transition. The increasing turbulent energy at 30 - 60 kHz, the spontaneous EB flow shear are identified to be responsible for the generation of LSCSs, which is in agreement with the theoretical prediction and provides unambiguous experimental evidences for LSCS generation mechanism in tokamak edge plasmas. New meso-scale electric potential fluctuations (MSEFs) at frequency f~10.5 kHz with two components of n = 0 and m/n = 6/2 are identified in the edge plasmas for the first time. The MSEFs coexist and interact with the magnetic islands of m/n = 6/2, turbulence and LFZFs. These results benefitted from the substantial improvement of the hardware such as 3 MW ECRH and 1.5 MW NBI as well as diagnostics, and have significantly contributed to the understanding of the underlying physics.

Speaker: Mr Xuru Duan (China)

26 OV/3-4: Towards an Emerging Understanding of Nonlocal Transport

In this overview, recent progress on the experimental analysis and theoretical models for non-local transport (non-Fickian fluxes in real space) are overviewed. The non-locality in the heat and momentum transport observed in the plasma, the departures from linear flux-gradient proportionality and the spontaneously and externally triggered non-local transport phenomena will be described in both L-mode and improved-mode plasmas in various devices (LHD, JT-60U, HL-2A, Alcator C-mod, KSTAR, etc.). Non-locality of transport has been observed in the response to perturbations, such as a core temperature rise associated with the cooling at edge by pellet injection, sustainment of the core temperature increase for repetitive perturbations by supersonic molecular beam injection (SMBI), strong coupling of transport at different radii as seen in the curvature transition of ITB, and spatial propagation of ITB regions. The probability distribution function (PDF) analysis for differences in temperature gradient from steady state values, studies of the fluctuation response during ‘non-locality events’, and micro-meso scale turbulence coupling studies are discussed as new approaches to investigate the mechanism of non-local transport. The experimental observation of meso-scale fluctuation with long range correlation during non-local phenomena is reviewed as a possible agent causing the non-locality of transport. The turbulence with long correlation, which is one of the strong candidates for causing the non-locality of the transport, was confirmed experimentally and the coupling between the different turbulence scales has also been identified in various devices (LHD, HL-2A, TJ-II, TJ-K, PANTA, etc.). Theoretical models of non-locality fall into two categories, namely models which are intrinsically local but which support fast front propagation (e.g. for turbulence spreading, barrier propagation etc.) and those which are intrinsically non-local (i.e. relate heat flux to temperature gradient by a non-local kernel). The intrinsically non-local models link the Kernal scale to the interval between steps in the ‘ staircase’ zonal flow pattern. Radial propagation of turbulence and barrier fronts models have had some success in explaining phenomena such as fast response to cold-pulses and profile rigidity in response to off-axis heating perturbations.

Speaker: Mr Katsumi Ida (Japan)

27 OV/4-1: Progress of the JT-60SA Project

The shared procurement and construction of the JT-60SA device by Japan and EU is progressing well, including preparation of the plan for key research and development. The JT-60SA device has been designed in order to complement ITER in all areas of fusion plasma development necessary to decide DEMO construction. Detailed studies to predict plasma performance have confirmed these capabilities. The tokamak construction will start in Dec.2012.JT-60SA enables explorations in ITER- and DEMO-relevant plasma regimes in terms of the non-dimensional parameters (beta, the normalized poloidal gyro radius, the normalized collisionality, the fast ion beta etc.) under ITER- and DEMO-relevant heating conditions (such as dominant electron heating and low central fuelling, and low external torque input). Detailed studies of plasma performance prediction support these capabilities. Under these conditions, heat/particle/momentum transport, L-H transition, ELM/RMP/Grassy-ELM characteristics, the pedestal structure, high energy ion behaviors and the divertor plasma controllability can be quantified. By integrating these studies, the project provides ‘simultaneous & steady-state sustainment of the key performance characteristics required for DEMO’ with integrated control scenario development. Assuming HH=1.3-1.4, the expected Ip for high beta-N (=4.3), high bootstrap fraction (=70-80%) full non-inductive current drive is 2.1-2.3MA at the Greenwald density ratio (=1). The central reference of DEMO for JT-60SA is a compact steady-state reactor. However, the JT-60SA research project has to treat the ‘DEMO regime’ as a spectrum spreading around the reference design, and has to assess reliable DEMO design targets.

Speaker: Mr Yutaka Kamada (Japan)

28 OV/4-2: Present Status of Fast Ignition Realization EXperiment (FIREX) and Inertial Fusion Energy Development

Controlled thermonuclear ignition and subsequent burn will be demonstrated in a couple of years on the central ignition scheme. Fast ignition has the high potential to ignite a fuel using only about one tenth of laser energy necessary to the central ignition. This compactness may largely accelerate inertial fusion energy development. One of the most advanced fast ignition programs is the Fast Ignition Realization Experiment (FIREX). The goal of its first phase is to demonstrate ignition temperature of 5 keV, followed by the second phase to demonstrate ignition-and-burn. The second series experiment of FIREX-I from late 2010 to early 2011 has demonstrated a high (≈20%) coupling efficiency from laser to thermal energy of the compressed core, suggesting that one can achieve the ignition temperature at the laser energy below 10 kJ. Given the demonstrations of the ignition temperature at FIREX-I and the ignition-and-burn at the National Ignition Facility, the inertial fusion research would then shift from the plasma physics era to power generation era.

Speaker: Mr Hiroshi AZECHI (Japan)

29 OV/4-3: Energetic Particle Instabilities in Fusion Plasmas

A remarkable progress was made in diagnosing energetic particle instabilities on present-day machines and in establishing a theoretical framework for describing them. This overview presents a point-by-point comparison between the much improved diagnostics of Alfvén Eigenmodes (AEs) and modelling tools developed world-wide, and outlines progress in interpreting the observed phenomena. A multi-machine comparison is presented giving a fair idea on the performance of both diagnostics and modelling tools for different plasma conditions. On JET equipped with 2D gamma-ray camera and interferometry, core-localised TAEs were detected causing redistribution of fast ions from inside the q=1 radius to outer plasma region followed by monster sawtooth crashes. TAE modelling using the HAGIS code showed a good agreement with the measured re-distribution and its effect on sawteeth. On DIII-D and AUG, ECE-imaging provides detailed measurements of amplitude and structure of AEs. A successful modelling was performed using the ORBIT code for reproducing the anomalously flat beam profiles on DIII-D. In AUG, a monitoring of the fast-ion redistribution and losses with an array of scintillators and fast-ion D-alpha spectroscopy has shown how a radial chain of overlapping AEs enables the transport of fast-ions from the core to the fast-ion loss detector. On NSTX, beam-driven AEs were observed in the form of “avalanches” consisting of coupled modes with strong frequency chirp. These modes caused ~10% drops in the neutron rate explained by effects of decrease in the beam energy and beam losses resulting from the interaction with TAE. A nonlinear model for near-threshold beam-driven instabilities has successfully encompassed many of the temporal characteristics of AEs seen in experiments. A steady-state nonlinear mode saturation and bursts of mode activity were found to be associated with both the strength and type of relaxation process in the phase-space region surrounding the resonance of the distribution function. An extension of the model to modes with a frequency sweep comparable to the starting frequency opened the opportunity for understanding the chirping modes in DIII-D, MAST, NSTX, START, and LHD. Finally, this presentation will outline expectations for ITER based on our present knowledge. This work was funded by the RCUK Energy programme and EURATOM.

Speaker: Mr Sergei Sharapov (UK)

30 OV/4-4: Overview of Recent and Current Research on the TCV Tokamak

Through a diverse research program, TCV addresses physics issues and develops tools for ITER and for the longer-term goals of nuclear fusion, relying especially on its extreme plasma shaping and ECRH launching flexibility and preparing for an ECRH and NBI power upgrade. Localized edge heating was unexpectedly found to decrease the period and relative energy loss of ELMs. Successful ELM pacing has been demonstrated by following individual ELM detection with an ECRH power cut before turning the power back on to trigger the next ELM, the duration of the cut determining the ELM frequency. In a parallel study, negative triangularity was also seen to reduce the ELM energy release. Both stabilizing and destabilizing agents (ECCD on or inside the q=1 surface, respectively) were used in a similar scheme to pace sawtooth oscillations, permitting precise control of their period. Locking of the sawtooth period to a pre-defined ECRH modulation period has also been demonstrated. In parallel with fundamental investigations of NTM seed island formation by sawtooth crashes, sawtooth control has permitted nearly failsafe NTM prevention when combined with backup NTM stabilization by ECRH. Additional work has addressed the destabilization of NTMs in the absence of a sawtooth trigger, and particularly its relation to plasma rotation. Further real-time control developments include the demonstration of joint current and internal inductance control using the Ohmic transformer and the validation of an ECRH power absorption observer based on transmitted stray radiation, for eventual polarization control. A new profile control methodology was also introduced, relying on real-time modelling to supplement diagnostic information; the RAPTOR current transport code in particular has been employed for joint control of the internal inductance and temperature profile. H-mode studies have focused on the L-H threshold dependence on the main ion species and on the divertor leg length. In L-mode, a systematic scan of the auxiliary power deposition profile, with no effect on confinement, has ruled it out as the cause of confinement degradation. Both L- and H-modes have been explored in the snowflake regime with emphasis on edge measurements, revealing that the heat flux to the strike points on the secondary separatrix increases as the X-points approach each other, well before they coalesce.

Speaker: Mr Stefano Coda (Switzerland)

31 OV/4-5: Science and Technology Research & Development in Support to ITER and the Broader Approach

Magnetic Fusion Energy has now entered its development era that steers the activities of traditional fusion laboratories. Recent achievements in fusion science and technology in support to both the ITER and the Broader Approach (BA) projects are reported here. On top of the direct contribution to ITER and JT-60SA procurement packages, many scientific activities, aiming at reducing risks in operation of ITER and BA, have been carried out using the CEA dedicated in-house facilities (Tore Supra tokamak, ICRH test facility for ITER, remote operated diagnostics, actively cooled PFC qualification, cryogenics test facilities from strand to sub size superconducting conductors characterization, etc). The paper reviews the research and development actions taken in the past two years by CEA in this context, in order to ensure an ITER safe operation (quench detection, disruption mitigation, surface monitoring of plasma facing components), qualify the long pulse RF Heating and Current Drive systems, and progress in MHD, turbulence and transport first principle simulations. A fully documented project, turning Tore Supra into a long pulse actively cooled diverted plasma test facility, is now being proposed to the ITER partners. This evolution allows the R&D and commissioning tests of actual ITER actively cooled tungsten divertor elements under ITER-relevant edge plasma conditions, during the ITER procurement phase, and targets its risk reduction. In parallel, the contribution to the Broader Approach projects is shown to be complemented by an ambitious programme on integrated modeling of the main scenarios and an assessment of EC power needed for NTM stabilization.

Speaker: Mr Alain BECOULET (France)

32 OV/5-1: Dynamics of Flows and Confinement in the TJ-II Stellarator

Operation with Li coated wall is the basis for a significant improvement in the performance of TJ-II, and lies behind the findings in this overview, related to the role of flows in confinement improvement. Specific progress has been performed in the use of Li as alternative to solid plasma facing materials for future fusion devices. Recently a liquid lithium limiter (LLL) based on the Capillary Porous System (CPS) has been installed in TJ-II and first results will be reported. Although TJ-II presents a strong damping, the simulations predict that the presence of an ambipolar radial electric field as well as turbulence driven flows provide driving mechanisms for mean and fluctuating flows that will cause long range toroidal correlation whose typical frequencies are in agreement with the experiment. The transitions to improved confinement are accompanied by an amplification of long-range correlation in the plasma potential, which is a footprint of zonal flows. The amplitude of these structures have been seen to modulate the particle transport into the SOL for the first time. We also investigate the relation between the zonal flows and the turbulent flux of particles and momentum via the Reynolds and Maxwell stresses as well as suprathermal particles. Suprathermal ion can contribute with significant energy content, with poloidal rotation up to 2–5 higher than the thermal component even the ECRH regime. Taking advantage of the flexibility of TJ-II, low order rationals are introduced in the plasma, which helps to transit from L to H mode. The influence of the so developed MHD modes on transport is investigated showing a bursty behaviour and evidence of radially propagating events. During L-H transitions, an oscillating low frequency non-damped sheared flow appears in the edge prior to the change to H mode, which presents a predator-pray relation with the turbulence. The spatial evolution of this turbulence-flow shows both radial outward and inward propagations. These results show the need to study L-H transition within a 1-D spatio-temporal framework. The dynamical coupling between density gradients and particle transport has been investigated and compared in the plasma boundary of different tokamaks (JET, ISTTOK) and stellarator (TJ-II), showing that the size of turbulent events is minimum in the proximity of the most probable density gradient.

Speaker: Mr Enrique Ascasibar (Spain)

33 OV/5-2Ra: Overview of Results from the MST Reversed Field Pinch Experiment

This overview of results from the MST program summarizes physics important for the advancement of the RFP as well as for improved understanding of toroidal magnetic confinement in general. Evidence for the classical confinement of ions in the RFP is provided by analysis of impurity ion transport. With inductive current profile control, the test-particle diffusivity for ions in a stochastic magnetic field is reduced below the classical transport level. (The neoclassical enhancement of radial transport is negligible in the RFP.) Carbon impurity measured by CHERS reveals a hollow profile and outward particle convection. Modeling of classical transport agrees with the profile evolution, and temperature screening explains the hollow profile. Classical confinement is also observed for energetic ions created by 1 MW NBI. The energetic ion confinement is consistent with classical slowing-down and ion loss by charge-exchange. The first appearance of Alfven eigenmodes and energetic particle modes by NBI in a RFP plasma are obtained. MST plasmas robustly access the quasi-single-helicity state that has commonalities to the stellarator and “snake” formation in tokamaks. The dominant mode grows to 8% of the axisymmetric field strength, while the remaining modes are reduced. Energy confinement is improved as a result. Predictive capability for tearing mode behavior has been improved through nonlinear, 3D, resistive MHD computation using the measured resistivity profile and Lundquist number, which reproduces the sawtooth cycle dynamics. New two-fluid analysis that includes Hall physics and gyro-viscosity has established a new basis for understanding physics beyond a single-fluid model. Nonlinear two-fluid (NIMROD) computation reveals coupling of parallel momentum transport and current profile relaxation. Large Reynolds and Maxwell stresses, plus separately measured kinetic stress, indicate an intricate momentum balance and possible origin for MST’s intrinsic plasma rotation. Microturbulence from drift-wave-like instabilities might be important in the RFP when magnetic fluctuations are reduced. New gyrokinetic analysis indicates that micro-tearing modes can be unstable at high beta, with a critical gradient for the electron temperature that is larger than for tokamak plasmas by roughly the aspect ratio. Supported by US DoE and NSF.

Speaker: Mr John Sarff (USA)

Slides Sarff_OV5_2R.pdf

34 OV/5-2Rb: Overview of the RFX Fusion Science Program

With a program well-balanced among the goal of exploring the fusion potential of the reversed field pinch (RFP) and that of contributing to the solution of key science and technology problems in the roadmap to ITER, the European RFX-mod device has produced a set of high-quality results since the last 2010 Fusion Energy Conference. RFX-mod is a 2 MA RFP, which can also be operated as a tokamak and where advanced confinement states have 3D features studied with stellarator tools. Self-organized equilibria with a single helical axis and improved confinement (SHAx) have been deeply investigated and a more profound understanding of their physics has been achieved. First wall conditioning with Lithium provides a tool to operate RFX at higher density than before, and application of helical magnetic boundary conditions favour stationary SHAx states. The correlation between the quality of helical states and the reduction of magnetic field errors acting as seed of magnetic chaos has been robustly proven. Helical states provide a unique test-bed for numerical codes conceived to deal with 3D effects in all magnetic configurations. In particular the stellarator equilibrium codes VMEC and V3FIT have been successfully adapted to reconstruct RFX-mod equilibria with diagnostic input. The border of knowledge has been significantly expanded also in the area of feedback control of MHD stability. Non-linear dynamics of tearing modes and their control has been modelled, allowing for optimization of feedback models. An integrated dynamic model of the RWM control system has been developed integrating the plasma response to multiple RWMs with active and passive conducting structures (CarMa model) and with a complete representation of the control system. RFX has been operated as a tokamak with safety factor kept below 2, with complete active stabilization of the (2,1) Resistive Wall Mode (RWM). This opens the exploration of a broad and interesting operational range otherwise excluded to standard tokamaks. Control experiments and modelling led to the design of a significant upgrade of the RFX-mod feedback control system to dramatically enhance computing power and reduce system latency. The possibility of producing D-shaped plasmas is being explored.

Speaker: Mr Piero MARTIN (Italy)

35 OV/5-3: Theory of Ignition, Burn and Hydro-equivalency for Inertial Confinement Fusion Implosions

Recent advances in the theory of ignition and burn for inertial confinement fusion are presented and related to the experimental observables of the current indirect-drive ignition campaign on the National Ignition Facility (NIF) and the direct-drive implosion campaign on the OMEGA laser. The performance parameter currently used for the ignition campaign (the Experimental Ignition Threshold Factor or ITFX) is related to the well-known Lawson criterion. Hydro-equivalent curves are derived and used to extrapolate current results from OMEGA to future direct-drive ignition experiments on the NIF. The impact of laser-plasma instabilities, hot electron and radiation preheat on the hydrodynamic scaling is discussed. Remedies to mitigate the detrimental effects of laser-plasma and hydrodynamic instabilities are presented. It is also shown that ignition through a late shock launched at the end of the laser pulse (shock ignition) may be possible on the NIF at sub-megajoule energies.

Speaker: Mr Riccardo Betti (USA)

36 OV/5-4: Multimodal Options for Materials Research to Advance the Basis for Fusion Energy in the ITER Era

Sustained worldwide efforts on fusion energy research have led to substantial improvements in understanding of plasma physics and fusion technology issues. Several options with varying degrees of technological risk are being contemplated for the next major fusion energy device that will be constructed after ITER begins operation. These options include a variety of plasma confinement configurations, coolants, tritium breeding materials, power conversion systems, and operating temperatures. In many cases, variations in the aggressiveness of the design parameters for next-step devices are associated with uncertainties in the performance of the materials systems to be used in the divertor, first wall and breeding blanket, and tritium extraction and power conversion systems. In order to reduce some of these uncertainties and to assist in the selection of the most appropriate design concept(s) that meet national needs, well-coordinated international fusion materials research on multiple fundamental feasibility issues can serve an important role during the next ten years. There are two inter-related overarching objectives of fusion materials research to be performed in the next decade: 1) understanding materials science phenomena in the demanding DT fusion energy environment, and 2) Using this improved understanding to develop and qualify materials to provide the basis for next-step facility construction authorization by funding agencies and public safety licensing authorities. There are several important fundamental materials questions that should be resolved soon due to their potential major impact on next-step fusion reactor designs, including radiation effects on mechanical properties, structural stability and tritium permeation and trapping. An overview will be given of the current state-of-the-art of major materials systems that are candidates for next-step fusion reactors, including a summary of existing knowledge regarding operating temperature and neutron irradiation fluence limits due to high temperature strength and radiation damage considerations, coolant compatibility information, and current industrial manufacturing capabilities. The critical issues and prospects for development of high performance fusion materials will be discussed along with recent research results and planned activities of the international materials research community.

Speaker: Mr Steven Zinkle (USA)

37 OV/P-01: Overview of FTU Results

Since the 2010 IAEA-FEC Conference, FTU has exploited improvements in cleaning procedures and in the density control system to complete a systematic exploration of access to high-density conditions in a wide range of plasma currents and magnetic fields. The line-averaged densities at the disruptive limit increased more than linearly with the toroidal field, while no dependence on plasma current was found, in fact the maximum density of 4.3e20 was reached at B=8T even at the minimum current of 0.5MA, corresponding to twice the Greenwald limit. The lack of plasma current dependence is due to the increase of density peaking with the safety factor. Experiments with the 140 GHz ECRH system were focused on the sawtooth period control and on the commissioning of the new launcher with real-time-steering capability that will act as the front-end actuator of a real time system for sawtooth period control and tearing modes stabilization. Various ECRH and ECCD modulation schemes have been used; with the fastest one, the sawtooth period synchronized with the 8 ms modulation period. The observed period variations were simulated using the JETTO code with a critical shear model for the crash trigger. The new launcher is of the plug-in type, allowing quick insertion and connection to the transmission line. Both beam characteristics and steering speed were in line with design expectation. Experimental results on the connection between improved coupling of lower hybrid waves in high-density plasmas and the reduction of spectral broadening of the injected wave have been compared with the results of fully kinetic non-linear model calculations. The effect of wall conditioning by lithium on MHD activities has been studied by comparing discharges with and without lithium conditioning at low-q and at the density limit. In both cases lithium conditioning has the same effect of reducing MHD modes associated with edge cooling by light impurities as a careful wall preparation. Experiments with the liquid lithium limiter inserted in the SOL, which have shown the formation of a radiative belt that acts as a virtual toroidal limiter, have been interpreted by the edge code TECXY as an effect of strong radiation from Li+ ions in non-coronal equilibrium.

Speaker: Mr Paolo Buratti (Italy)

38 OV/P-02: New Developments, Plasma Physics Regimes and Issues for the Ignitor Experiment

The IGNIR collaboration between Italy and Russia is centered on the construction of the core of the Ignitor machine in Italy and its installation and operation within the Triniti site (Troitsk). The scientific goal of the experiment is to approach for the first time, the ignition conditions of a magnetically confined D-T plasma. A parallel initiative has developed that integrates this program, involving the study of plasmas in which a high energy population is present, with ongoing research in high-energy astrophysics and with a theory effort involving the National Institute for High Mathematics and the Inter University Space Physics Consortium, CIFS. Innovations in the adopted diagnostic systems are expected from the collaboration with INAF, the National Institute for Astrophysics (X-ray diagnostics), with INFN (nuclear diagnostics), and with the University of Pisa. The Ignitor core construction has been fully funded by the Italian government and the management role for it has been assigned to INFN. Meanwhile, considerable attention has been devoted toward identifying the industrial groups with the facilities necessary to build the main machine components. An important step for the Ignitor program is the adoption of the superconducting material MgB2 for the largest poloidal field coils (P14) that is compatible with the He gas cooling system designed for the entire machine. The progress mode for the construction of these coils is described. The main physics issues that the Ignitor experiment is expected to face are analyzed considering the most recent developments in both experimental observations and theory of weakly collisional magnetically confined plasmas. Of special interest is the I-Regime that has been investigated in depth only recently and combines advanced confinement properties with a high degree of plasma purity. This is a promising alternative to the high density L-Regime that had been observed by the Alcator program and had motivated the Ignitor project. The provisions that are incorporated in the machine design in order to prevent the development of macroscopic instabilities with deleterious amplitudes are presented.

Speaker: Mr Bruno Coppi (Italy)

39 OV/P-03: On the Physics of Intrinsic Torque in Toroidal Plasmas

Intrinsic rotation is a critical physics issue for ITER, both for resistive wall mode mitigation and for confinement optimization. Rotation control requires predictive understanding of intrinsic rotation, and so this goal has stimulated a great deal of research on intrinsic torque, driven by the non-diffusive fluctuation-induced residual stress[1]. In this OV, we discuss recent theoretical, simulation and experimental progress which elucidates the important physics of intrinsic torque. In contrast to recent OV’s on rotation, here we focus on intrinsic torque rather than on the momentum pinch[2]. We present important new results on the critical role of boundary stresses in intrinsic torque. The heat flux-driven character of intrinsic torque is emphasized throughout. Specifically, we present a novel, unifying theory of intrinsic rotation in terms of fluctuation entropy balance. We then discuss various symmetry breaking mechanisms and their underlying microphysical structure. Applications of the theory to MFE phenomenology are discussed in detail. Novel results on symmetry breaking effects at the boundary, which reveal boundary-specific intrinsic torques, are discussed. We also consider RMP effects on intrinsic rotation. [1] P.H. Diamond, et al., NF (2009). [2] A. Peeters, et al., FEC (2010), C. Angioni, et al., H-Mode Workshop (2011).

Speaker: Mr Patrick H. Diamond (Republic of Korea)

40 OV/P-05: Overview of IFERC Project in Broader Approach Activities

In order to contribute to ITER and to an early realization of the DEMO reactor, International Fusion Energy Research Centre (IFERC) started the activity from 2007/7/1 with 10 years period under the Broader Approach (BA) framework and now implements the three sub-projects; DEMO Design and R&D Coordination Centre (DDA & R&D), Computational Simulation Centre (CSC), and ITER Remote Experimentation Centre (REC). DDA & R&D: The DEMO Design Activity (DDA) in 2011 covered the studies on system codes, divertor, in Vessel Components, operation modes and maintenance. After benchmark test of the system codes by EU and JA, the codes were used to investigate an R=8.5m steady-state device and an R=10m pulsed device. In addition, comparison between steady-state and pulsed was done from engineering points of view. Maintenance study was another important issue and the various schemes were compared to review the features. These works will provide the databases toward conceptual designs planned in the period from 2015. As for R&D, preparation of facilities, installation of equipment, and preliminary R&Ds have been performed in 5 tasks (T1-T5). R&D activities are now being upgraded both in EU and JA for T1 (lifetime and off-axial mechanical properties and electrical properties of SiC_f/SiC composites), T2 (tritium accountancy etc. in Tritium technology), T3 (production and characterization of reduced activation ferritic/martensitic steel in Materials engineering), and T4/ T5 (mass production technology of Be-Ti intermetallic compounds for neutron multiplier as well as Li_2SiO_4 and Li_2TiO_3 for tritium breeders). CSC: The mission of CSC is to exploit high-performance and large-scale magnetically confined fusion (MCF) simulations. The Light House Project was carried out by using codes with a good scalability, in order to show MCF simulations could exploit a new research field or a frontier research. For example, particle-fluid hybrid simulations on energetic particle-driven instability and alpha particle transport have been performed in an ITER steady state scenario plasma, and it is found out that saturation level of multiple TAE modes with medium toroidal mode numbers is so low that the confinement of alpha particles is not degraded. REC: The mission of REC is to prepare ITER remote experiments and verify the functions. The overall plan of REC will be created in 2012.

Speaker: Mr Noriyoshi Nakajima (Japan)

41 OV/P-06: The Recent Research Work on the J-TEXT Tokamak

The main results from the J-TEXT tokamak in the last two years, which emphasized the observation and analysis of MHD activity, are summarized and presented in this meeting. Static resonant magnetic perturbations generated by saddle coil currents are applied to J-TEXT Ohmic plasmas in order to study their influence on MHD instabilities. With sufficiently large RMPs, the m/n=2/1 (m and n are the poloidal and toroidal mode numbers) mode locking is easily obtained. The analysis of the mode locking thresholds varied by scanning of the spatial phase of RMPs shows that the m/n=2/1 component of intrinsic error field of the J-TEXT tokamak is about 0.4Gs. In addition to normal mode locking events, the (partial) stabilization of the m/n=2/1 tearing mode by moderate magnetic perturbation amplitude is observed experimentally. With experimental parameters as input, both the mode locking and mode stabilization by RMPs are also obtained from nonlinear numerical modeling based on reduced MHD equations. It is found that the suppression of the tearing mode by RMPs of moderate amplitude is possible for a sufficiently high plasma rotation frequency and low Alfvén velocity. Gas puffing is also used to affect the MHD activity in J-TEXT. For example, neon gas injection can cause inverse sawtooth-like activity that spreads from the q=1 surface to the axis; in particular, small amplitude m/n=1/1 mode oscillations superimposed on the inverse sawtooth waveform around the q=1 surface are observed after the impurity injection. Nevertheless, other impurities such as helium and argon impurities can’t trigger such events. In addition, gas puffing can also be applied to mitigate disruptions, especially on the current quench phase. It is found that no runaway current generation occurs in intentionally provoked disruptions when the toroidal magnetic field is lower than 2.2 T. The runaway currents can be suppressed by the intensive gas puffing of H2. To meet the requirement of charactering the MHD activity, a far-infrared polarimeter-interferometer has been developed to measure the current density profile, while a tangential X-ray imaging crystal spectrometer (XICS) provides ion temperature and toroidal rotation velocity measurements. First results are obtained with observation of perturbations associated with sawtooth and MHD activities. The details will be given in this meeting.

Speaker: Mr Ge Zhuang (China)

42 OV/P-07: Fusion Prospects of Axisymmetric Magnetic Mirror Systems

Studies of the magnetic mirrors have been started in 50s – 60s of the last century. Very soon it was found out that axially symmetric mirrors suffer from the curvature-driven instabilities which results in unacceptable plasma losses. Introduction of non-axisymmetric (quadrupole) configurations with min-B magnetic field improved plasma confinement significantly. Then, plasma losses associated with less destructive kinetic instabilities become dominative. Later on, the methods of suppression of the kinetic instabilities were successfully developed and confinement was further improved. However, it was recognized that even the axial plasma losses determined by classical binary collisions appear to be too high for practical application of the mirrors. In the 70s, some ideas emerged how to reduce these losses. These ideas are briefly reviewed in [1] and in the references therein. Another advance in magnetic mirror studies was invention of axisymmetric mirrors with improved longitudinal and transverse plasma confinement. Several techniques for achieving MHD stabilization of the axisymmetric mirrors are considered in [2]. These versions of the mirror systems are of particular interest as neutron sources, fusion-fission hybrids, and pure fusion reactors. At present, two versions of the advanced magnetic mirrors, namely a gas dynamic trap (GDT) and multi mirror confinement system are studied in the Budker Institute. References [1] A.V. Burdakov, A.A. Ivanov, E.P. Kruglyakov, Plasma Phys. Cont. Fus., 52 124026 (2010). [2] D.D. Ryutov, H.L. Berk, B.I. Cohen, A.W. Molvik, and T.C. Simonen, Phys. Plasmas 18, 092301 (2011).

Speaker: Mr Edward Kruglyakov (Russian Federation)

43 OV/P-08: 20 Years of ISTTOK Tokamak Scientific Activity

The ISTTOK tokamak commissioning began in January 1990, after the establishment operating coincidently with the signing of the EURATOM association agreement established with Instituto Superior in 1990 on the field of controlled nuclear fusion. In 1991 the first article have been issue in the Portuguese journal “Gazeta de Física” and the firsts scientific reports were mentioned to the international community on the 17th Symposium on Fusion Technology, Rome 1992. This communication aims to present an overview of ISTTOK scientific activity since its commissioning in the main áreas of its activity, namely (i) Plasma diagnostics an in particular the Heavy Ion Beam (HIBD), (ii) AC current operations, (iii) plasma control, (iv) liquid metal limiters, (v) plasma fluctuation studies and (vi) study of fusion relevant materials. The ISTTOK achievements demonstrate that small tokamaks can play an important role in the fusion plasma physics community as a result of their flexibility, high availability and good opportunity for the development of sophisticated diagnostics and technology tools.

Speaker: Mr Horácio João Fernandes (Portugal)

4:10 PM Coffee Break

ITER Physics, Scenarios and Heating & Current Drive Technology: ITR/1 & FTP/1

Convener: Mr Masahiro Mori (Japan)

44 ITR/1-1: Scaling of the Tokamak near Scrape-off Layer H-mode Power Width and Implications for ITER

The presence of a steep edge pedestal gradient in H-mode divertor plasmas implies that strong gradients should also exist across the separatrix, forcing most of the PSOL~ 100 MW of power arriving in the SOL at QDT = 10 in ITER to flow inside a narrow channel on open field lines connecting to the divertor target plates. Recent results (coordinated in part through ITPA DivSOL group) indicate that the ITER assumed value is too large. Scaling from the new database provides a clear dependence on the poloidal magnetic field, little variation with other key variables found in previous scalings and suggests λSOL ≅ 1 mm for ITER. Measurements from DIII-D, C-Mod and NSTX indicate a systematic narrowing of the inter-ELM divertor heat flux width with plasma current in H-mode plasmas. For the near SOL power width, the data indicate λSOL ∝ (q95/Btor)~1 ~ a/Ip ∝ 1/Bpol, with little or no dependence on PSOL or R (major radius). Analysis of data obtained in the same way from JET and ASDEX Upgrade yields λSOL (mm) =0.73·Bt^(−0.78)·qcyl^1.2·PSOL^0.1·R^0, again with no dependence on R. Data are consistent in absolute magnitude with a recent heuristic drift-based theory. These new findings are based on IR analysis of strongly attached H-mode discharges. Key improvements here have been the avoidance of ELM effects, accounting for changes in the deposition profile due to heat diffusion across the divertor legs into the private flux region. Experimentally, essentially the full operational range of plasma current and toroidal field in each device was scanned. The value of λSOL ≅1 mm obtained jointly from these scalings for ITER at 15 MA in the Baseline Inductive Scenario, is about a factor 3 shorter than the lowest values predicted on the basis of earlier studies. Such narrow power channels are a concern for ITER, though preliminary SOLPS simulations indicate that they could be tolerated, since volumetric power dissipation in the divertor can still be sufficient to maintain heat flux densities at acceptable levels provided the outer divertor leg is partially detached. Simple estimates show that for ITER, if λSOL ≤ 2 mm the implied upstream pedestal pressure gradient would exceed ideal ballooning stability by some margin, assuming that the SOL pressure width is a measure of that in the pedestal. This issue will be examined in the context of the current database.

Speaker: Mr Thomas Eich (Germany)

Slides ITR-1-1_Eich_T..pdf

45 ITR/1-2: Progress on the Application of ELM Control Schemes to ITER

High fusion performance DT operation in ITER is based on the achievement of the H-mode confinement regime with H98 ≥ 1 and an edge transport barrier that is expected to lead to the quasi-periodic triggering of ELMs. Operation of ITER with H-mode plasmas is also foreseen during the non-active (H & He) and DD operation allowing the development of ELM control schemes before DT operation. The non-linear MHD evolution of the plasma during ELMs in ITER has been modelled with the JOREK code which shows that the non-linear MHD growth during the ELM causes a temporary ergodisation of the plasma edge leading to the appearance of striations in the ELM power flux at the divertor target. On the contrary, power fluxes to the first wall are expected to be dominated by the convection of energy by the radial propagation of plasma filaments produced during the ELM crash. JOREK has also being applied to investigate the capabilities of the ITER pellet injection system to meet the requirements for ELM control following its validation with DIII-D experimental results. The application in-vessel coils to create edge magnetic field perturbations for ELM control and the associated power/particle fluxes to PFCs have been studied for ITER. Evaluations of the edge magnetic field perturbation by in-vessel coils show that the toroidal symmetry of the applied currents in the coils (n=3 or n=4 symmetry) can have a significant impact on the level of current required (up to 50% current level reduction) to achieve a given level of edge ergodisation. Optimization of the relative toroidal phasing of the currents applied to the 3 rows of coils shows that there is an appropriate margin (factor of ~ 1.5 - 2) in coil current magnitude required to achieve the design criterion (vacuum approximation) for the 15 MA QDT = 10 scenario. Power and particle fluxes in the perturbed edge magnetic field have been evaluated in the vacuum field approximation and including plasma response. The application of edge magnetic field perturbations leads to the appearance of non-toroidally symmetric divertor power/particle fluxes extending up to ~ 50 cm from the separatrix but also to a reduction of the peak heat flux by a factor of 2-3. The inclusion of plasma response decreases transport of energy/particles from the main plasma and the detrimental effects on plasma energy/particle confinement.

Speaker: Mr Alberto Loarte (ITER)

Slides Loarte_IAEA_2012_v3.0_conference_final.pptx

46 FTP/1-1: Evaluation of Optimized ICRF and LHRF Antennas in Alcator C-Mod

Ion cyclotron range of frequency heating (ICRF) and lower hybrid range of frequency current drive (LHCD) are expected to be key heating and current drive actuators for future fusion reactors and devices. However, impurity contamination associated with ICRF antenna operation remains a major challenge, particularly in devices with metallic plasma facing components. For LHCD, maximizing coupled power to the plasma remains a challenge, particularly to maintain low reflection coefficient over range of plasma conditions. Here, we report on an experimental investigation to test whether a field aligned (FA) ICRF antenna can reduce the impurity contamination and SOL modification associated with antenna operation. We also report on results from a new limiter for the LH coupler designed to reduce reflection coefficients across a wider range of plasma conditions. The unique feature of the so-called FA-antenna is that the current straps and antenna box structure are perpendicular to the total magnetic field. This alignment allows integrated E|| (electric field along a magnetic field line) to be minimized through symmetry. Using finite element method and a cold plasma model, the FA-antenna has been found to have lower integrated E|| relative to the previous antenna geometry. Initial results indicate that the impurity contamination associated with the FA-antenna is lower relative to our standard ICRF antennas. Configured as a 2-strap antenna, the antenna has lower core impurity contamination and lower impurity source at the antenna at high power density (~15 MW/m2). An array of core and boundary plasma diagnostics are presently being used to characterize the impurity behavior and impact on the SOL transport and SOL density profiles; the latest results will be presented. For LHCD, reflection coefficients are very sensitive to the local density and its profile in front of the LHCD coupler. Previously the local LH coupler protection limiter was fixed to the outer wall of the vacuum vessel. The new limiter is mounted on the coupler and protrudes 0.25 mm beyond the coupler for plasma heat flux protection. The protection tiles allow the LH launcher to be moved closer to the plasma than previously possible. Initial high power (Pnet ~ 700 kW) results show lower reflection coefficients were achieved (Γ2 ~ 0.1) as compared to the old configuration (Γ2 ~ 0.2).

Speaker: Mr Stephen Wukitch (MIT PSFC)

Slides FTP/1-1-wukitch

47 ITR/1-3: Design of the MITICA Neutral Beam Injector: From Physics Analysis to Engineering Design

For ITER heating and current drive, two neutral beam injectors (NBIs) are planned, delivering a total of 33 MW in stationary conditions up to one hour; each injector will accelerate a 40 A negative deuterium ion current up to 1 MV. Such requirements have never been achieved simultaneously. Hence the PRIMA (Padova Research on ITER Megavolt Accelerator) facility is under construction at Consorzio RFX in Padua, Italy. PRIMA will include a test bed named MITICA (Megavolt ITer Injector and Concept Advancement), with the aim of meeting the ITER beam requirements in terms of negative ion yield, beam uniformity, high voltage holding, operation of beam line components and power supplies, overall reliability of the NBI. The present contribution describes the current status of numerical simulations, devoted to the optimisation of MITICA, providing the main inputs for the design of accelerator, beam line components, diagnostics and power supplies. Physics and engineering aspects include: beam optics, dumping of co-extracted and stripped electrons, thermo-mechanical behaviour of grids and beam line components during long pulse operation, voltage holding capabilities. The optimised geometry of the accelerator is characterised by equal acceleration gaps (increased voltage holding capability) and a combination of horizontal and vertical magnetic fields in the accelerator (reducing heat loads and electrons exiting the accelerator); the gas pressure profile is also simulated in the accelerator and in the injector. The design of the accelerator power supplies has been supported by simulations of static and dynamic performances, including the investigation of overvoltages by a sophisticated fast transient model and the modelling of matching network and RF systems. Moreover the signals expected from the diagnostic systems have been simulated, with realistic beam features, providing prescriptions for the design of diagnostics, like beam emission spectroscopy, beam tomography and neutron diagnostic. Most of the design of MITICA plants and components are well developed and close to finalisation.

Speaker: Mr Piergiorgio Sonato (Italy)

Slides ITR_1-3_FEC24_MITICA_sonato_2.pptx

48 FTP/1-2: Acceleration of 1 MeV H- Ion Beams at ITER NB-relevant High Current Density

ITER neutral beam (NB) system requires deuterium negative ion beams of 1 MeV, 40 A at the current density of 200 A/m^2 from a single large negative ion source and an accelerator. This paper summarizes progress in R&D with a reduced size accelerator, so-called “the MeV accelerator” at Japan Atomic Energy Agency (JAEA). In the last Fusion Energy Conference, we reported achievement of 1 MV voltage holding in vacuum for more than one hour. Physics of beamlet deflections due to their own space charges and magnetic field was also reported utilizing a sophisticated three dimensional beam trajectory analyses. The improved voltage holding and a trajectory compensation technique have been applied to the MeV accelerator. Many discharge burn marks have been observed inside the accelerator after long pulse operation reported in the last conference. It was turned out that such discharge marks were observed at positions facing to high local electric field, such as edges, corners, and steps between grid and its support.　In the present MeV accelerator, such positions have been modified, for example, by increasing radii of corners around grid supports, and increasing gap length between grids to lower the local electric concentrations to about 3 ~ 4 kV/mm. For compensation of magnetic deflection, aperture offset was applied at the bottom of the EXG. Magnetic field is formed by small permanent magnets embedded in EXG between aperture lines. Since the polarities are arranged so as to be alternative in each line between apertures, aperture offset of 0.8 mm was defined in the direction against the magnetic deflection. To counteract the beamlet deflection by space charge repulsion, a field shaping plate, a metal plate to deform electric field, were installed around the aperture area for deflection of outermost beamlet inward. Position and thickness of the plate was designed by the analyses. It should be highlighted that reduction of beam direct interception at grids has brought substantial improvement in voltage holding during beam acceleration at around 1 MV. By the improved voltage holding even under beam acceleration, H^- ion beams of 185 A/m^2 (430 mA in total) have been successfully accelerated up to 0.98 MeV. This is a world first demonstration of negative ion beams at high current, high current density and high energy close to the ITER requirements.

Speaker: Mr Takashi Inoue (Japan)

Poster FTP_1-2_talk_poster.pdf.pptx

Slides IAEA_FEC_FTP1-2_INOUEslide121005.ppt

49 ITR/1-4Ra & FTP/1-3Rb: Development in Russia of Megawatt Power Gyrotrons for Fusion; Progress on the Development of High Power Long Pulse Gyrotron and Related Technologies

ITR/1-4Ra: Development in Russia of Megawatt Power Gyrotrons for Fusion During last years several new gyrotrons were designed and tested in Russia. Main efforts were spent for development 170GHz/1MW/50%/CW gyrotron for ITER and multifrequency gyrotrons. Additionally other new gyrotrons were shipped and installed at running plasma installations. The industrial production prototypes of the ITER gyrotron were tested at power 1.0 MW in 400…500 second pulses and 0.8-0.9 MW in 1000 second pulses. For 1 MW power regime the gyrotron efficiency is 55%. The last gyrotron versions operate in LHe-free magnet. It is important that two last gyrotrons (V-10 and V-11) demonstrate very similar output parameters. Time traces for the main gyrotron parameters are stable and confirm possibility of the gyrotron operation even in longer pulses. Detail analysis of the test results showed that a slightly modified ITER gyrotron prototype is capable to operate at power 1.2 MW. First tests of the modified tube are rather encouraging: microwave power 1.2 MW at MOU output was demonstrated in 100 second pulses with efficiency of 53%. Additionally two gyrotron models with TE28.12 operating mode were tested in short-pulse experiments. The use of step-tunable gyrotrons can greatly enhance performance of ECRH/ECCD systems due to larger accessible radial range, possible replacement of steerable antennas, higher CD efficiency for NTM stabilization. The main problems in development of multifrequency gyrotrons are to provide: efficient gyrotron operation at different modes, efficient conversion of the modes into a Gaussian beam, reliable operation of broadband or tuneable window. Considering this three key problems one can say that first two of them are solved, but realization of a CVD diamond window for a megawatt power level multi-frequency gyrotron met real difficulties. Now a new tunable window concept is under consideration. FTP/1-3Rb: Progress on the Development of High Power Long Pulse Gyrotron and Related Technologies In the development of a higher power dual-frequency gyrotron, a high order mode gyrotron, which permits to select the oscillation at 170GHz or 137GHz, has been fabricated and tested. Short pulse experiments (0.5ms) were performed with 1.3MW power output at more than 30% of the oscillation efficiency for both frequencies. In long pulse experiments, 760 kW/46%/60 s at 170GHz and 540 kW/42%/20 s at 137 GHz are achieved. It is the first time long pulse experiments with the dual-frequency gyrotron/triode electron gun. Since the RF beam direction from the output window is designed to be almost the same for both frequencies, good power couplings to the transmission line, which are 96% for 170GHz and 94% for 137GHz, are obtained by using a pair of identical phase correcting mirrors. Pulse extension is underway aiming for >1MW at CW operation. A 5kHz full power modulation experiment was performed using the 170 GHz gyrotron of TE31,8 mode oscillation. The 5kHz full power modulation was achieved with the full beam modulation by employing a fast voltage switching between the anode and cathode of the triode type electron gun. This satisfies the requirement of ITER. For further improvement, an advanced anode power supply system is proposed to reduce the oscillation period of adjacent mode at the start-up phase of each pulse.

Speaker: Mr Alexander Litvak (Russian Federation)

Slides Litvak-ITR-1_4a_V3.ppt

Welcome Reception

Tue, October 9

Overview: Magnetic Fusion: OV/3

Convener: Mr Predhiman K. Kaw (India)

50 OV/3-1: Overview of Physics Results from the National Spherical Torus Experiment

Research on the National Spherical Torus Experiment, NSTX, targets physics understanding needed for extrapolation to a steady-state ST Fusion Nuclear Science Facility, pilot plant, or DEMO. The unique ST operational space is leveraged to test physics theories for next-step tokamak operation, including ITER. Present research also examines implications for the coming device upgrade, NSTX-U. Energy confinement increases as collisionality is reduced by lithium (Li) wall conditioning. Nonlinear microtearing simulations match experimental electron diffusivity quantitatively and predict reduced electron heat transport at lower collisionality. Measured high-k turbulence is reduced in H-mode. Beam-emission spectroscopy measurements indicate that the poloidal correlation length of pedestal turbulence increases at higher electron density and gradient, and decreases at higher Ti. Plasma characteristics change nearly continuously with increasing Li evaporation, and ELMs stabilize due to edge density gradient alteration. Global mode stability studies show stabilizing resonant kinetic effects are enhanced at lower collisionality. Combined radial and poloidal field sensor feedback controlled n = 1 perturbations and improved stability. The disruption probability due to unstable RWMs is reduced at high betaN/li > 11. Greater instability seen at lower betaN is consistent with decreased kinetic RWM stabilization. A model-based RWM state-space controller produced long-pulse discharges exceeding betaN = 6.4 and betaN/li = 13. Precursor analysis shows 99% of disruptions can be predicted with 10ms warning and a false positive rate of only 4%. Disruption halo currents rotate toroidally and can have significant toroidal asymmetry. Global kinks cause measured fast ion redistribution. Full-orbit calculations show redistribution from the core outward and toward Vpar/V = 1. The snowflake divertor configuration enhanced by radiative detachment shows large reductions in steady-state and ELM heat fluxes (peak values down from 7 MW/m2 to less than 1 MW/m2). Non-inductive current fraction (NICF) up to 65% is reached experimentally. NSTX-U scenario development calculations project 100% NICF at Ip = 1MA. Coaxial helicity injection has reduced the inductive startup flux, with L-mode plasmas ramped to 1MA requiring 35% less inductive flux. *Supported in part by US DOE Contract DE-AC02-09CH11466.

Speaker: Mr Steven A. Sabbagh (USA)

Poster IAEA_FEC12_OV3-1-Sabbagh-v23p.pdf

Slides IAEA_FEC12_OV3-1-Sabbagh-v24t.pdf

51 OV/3-2: Overview of Physics Results from MAST towards ITER/DEMO and the Upgrade

New diagnostic, modelling and plant capability on MAST have delivered important results in key areas for ITER/DEMO and the upgrade. Linear gyro-kinetic calculations highlight the interplay of kinetic ballooning and micro tearing modes in setting the pedestal width. Using the upgraded 18 coil ELM control array ELM mitigation by resonant magnetic perturbations (RMP) has now been demonstrated with n=3, 4 and 6 proving the importance of the alignment of the perturbation to the magnetic field lines with the unique phasing capability. Already a 2D equilibrium deformation consistent with lobe structures at the X-point, only observed when the RMPs affect the plasma, destabilises the ELM despite the lower pressure gradient. The interplay of edge flow shear and turbulence at the L-H transition measured by a new 50 kHz Doppler spectroscopy diagnostic shows limit–cycle like dynamics during ~4-5 kHz H-mode dithers. Further insight into dn/dr, j and v may be provided by a novel 3D electron Bernstein emission imaging system. For the first time ELM resolved scrape-off-layer T_i profiles have been characterised simultaneously in the divertor and mid-plane with retarding field analysers. Divertor design is aided by ELM resolved data on the n_e and I_p dependence of target heat flux profiles. Under conditions foreseen for the upgrade (off-axis NBI, q_min>1.3, I_p> 0.8 MA) measurements of profiles of the neutron and fast ion D_alpha emission indicate a substantial reduction of the fast-ion redistribution compared to on-axis injection. To model the fast-ion redistribution due to the wave particle interaction driven by the super Alfvénic beam ions accurately dynamic friction is now included into the modelling recovering the observed non-linear mode behaviour well. Low-k turbulence, measured using a new 2D beam emission spectroscopy diagnostic, shows L-mode turbulence at r/a>0.7 propagating in the ion diamagnetic direction with respect to the local ExB flow and reduced turbulence levels in H-mode. Using in detail event triggering on NTMs the T_e perturbation due to the 2/1 island has been measured allowing detailed assessment of the NTM stability. Further studies of disruption mitigation with massive gas injection compared different gas mixtures and injection locations including a newly developed controllable high-field-side valve. Work supported by UK EPSRC and EURATOM.

Speaker: Mr Hendrik Meyer (UK)

Poster IAEA_2012_MAST_OV3-2_poster.pdf (Protected)

Slides IAEA_2012_MAST_OV3-2_v3.1.pdf (Protected)

52 OV/3-3: Overview of HL-2A Recent Experiments

Since the last FEC, the experiments on HL-2A tokamak have been focused on the investigations on H-mode related physics including ELM mitigation, energetic particle physics, transport including nonlocality and edge impurities, MHD instabilities, turbulence and zonal flow physics, etc. In particular, it’s demonstrated for the first time that the supersonic molecular beam injection (SMBI) and cluster jet injection (CJI) can convert large ELMs into a series of small ELMs in HL-2A. After SMBI or CJI, the ELM frequency rises by a factor of 2 to 3 on average and its amplitude decreases by about 50%. With high power ECRH, the energetic particle induced modes have been observed in different frequency ranges. The high frequency (200–350 kHz) modes with relatively small amplitude are close to the gap frequency of toroidacity induced Alfvén eigenmode (TAE). The coexistent multi-mode magnetic structures in the high temperature and low collision plasma will affect the plasma transport greatly. The L-I-H transition has been studied in pure NBI-heating deuterium plasmas. For the first time, the absolute rate of nonlinear energy transfer between turbulence and zonal flows was measured and the secondary mode competition between low frequency zonal flows (LFZFs) and geodesic acoustic modes (GAMs) was identified in experiments, which demonstrated that the energy transfer is large enough to affect the turbulence saturation level and its dynamics and that zonal flows play an important role in the low to high (L-H) plasma confinement transition. The increasing turbulent energy at 30 - 60 kHz, the spontaneous EB flow shear are identified to be responsible for the generation of LSCSs, which is in agreement with the theoretical prediction and provides unambiguous experimental evidences for LSCS generation mechanism in tokamak edge plasmas. New meso-scale electric potential fluctuations (MSEFs) at frequency f~10.5 kHz with two components of n = 0 and m/n = 6/2 are identified in the edge plasmas for the first time. The MSEFs coexist and interact with the magnetic islands of m/n = 6/2, turbulence and LFZFs. These results benefitted from the substantial improvement of the hardware such as 3 MW ECRH and 1.5 MW NBI as well as diagnostics, and have significantly contributed to the understanding of the underlying physics.

Speaker: Mr Xuru Duan (China)

Slides OV3-3__DUAN_XR_overview_HL-2A.pdf

53 OV/3-4: Towards an Emerging Understanding of Nonlocal Transport

In this overview, recent progress on the experimental analysis and theoretical models for non-local transport (non-Fickian fluxes in real space) are overviewed. The non-locality in the heat and momentum transport observed in the plasma, the departures from linear flux-gradient proportionality and the spontaneously and externally triggered non-local transport phenomena will be described in both L-mode and improved-mode plasmas in various devices (LHD, JT-60U, HL-2A, Alcator C-mod, KSTAR, etc.). Non-locality of transport has been observed in the response to perturbations, such as a core temperature rise associated with the cooling at edge by pellet injection, sustainment of the core temperature increase for repetitive perturbations by supersonic molecular beam injection (SMBI), strong coupling of transport at different radii as seen in the curvature transition of ITB, and spatial propagation of ITB regions. The probability distribution function (PDF) analysis for differences in temperature gradient from steady state values, studies of the fluctuation response during ‘non-locality events’, and micro-meso scale turbulence coupling studies are discussed as new approaches to investigate the mechanism of non-local transport. The experimental observation of meso-scale fluctuation with long range correlation during non-local phenomena is reviewed as a possible agent causing the non-locality of transport. The turbulence with long correlation, which is one of the strong candidates for causing the non-locality of the transport, was confirmed experimentally and the coupling between the different turbulence scales has also been identified in various devices (LHD, HL-2A, TJ-II, TJ-K, PANTA, etc.). Theoretical models of non-locality fall into two categories, namely models which are intrinsically local but which support fast front propagation (e.g. for turbulence spreading, barrier propagation etc.) and those which are intrinsically non-local (i.e. relate heat flux to temperature gradient by a non-local kernel). The intrinsically non-local models link the Kernal scale to the interval between steps in the ‘ staircase’ zonal flow pattern. Radial propagation of turbulence and barrier fronts models have had some success in explaining phenomena such as fast response to cold-pulses and profile rigidity in response to off-axis heating perturbations.

Speaker: Mr Katsumi Ida (Japan)

Slides OV_paper_for_IAEA_2012_v8.pptx

Poster: P1

54 FTP/1-1: Evaluation of Optimized ICRF and LHRF Antennas in Alcator C-Mod

Ion cyclotron range of frequency heating (ICRF) and lower hybrid range of frequency current drive (LHCD) are expected to be key heating and current drive actuators for future fusion reactors and devices. However, impurity contamination associated with ICRF antenna operation remains a major challenge, particularly in devices with metallic plasma facing components. For LHCD, maximizing coupled power to the plasma remains a challenge, particularly to maintain low reflection coefficient over range of plasma conditions. Here, we report on an experimental investigation to test whether a field aligned (FA) ICRF antenna can reduce the impurity contamination and SOL modification associated with antenna operation. We also report on results from a new limiter for the LH coupler designed to reduce reflection coefficients across a wider range of plasma conditions. The unique feature of the so-called FA-antenna is that the current straps and antenna box structure are perpendicular to the total magnetic field. This alignment allows integrated E|| (electric field along a magnetic field line) to be minimized through symmetry. Using finite element method and a cold plasma model, the FA-antenna has been found to have lower integrated E|| relative to the previous antenna geometry. Initial results indicate that the impurity contamination associated with the FA-antenna is lower relative to our standard ICRF antennas. Configured as a 2-strap antenna, the antenna has lower core impurity contamination and lower impurity source at the antenna at high power density (~15 MW/m2). An array of core and boundary plasma diagnostics are presently being used to characterize the impurity behavior and impact on the SOL transport and SOL density profiles; the latest results will be presented. For LHCD, reflection coefficients are very sensitive to the local density and its profile in front of the LHCD coupler. Previously the local LH coupler protection limiter was fixed to the outer wall of the vacuum vessel. The new limiter is mounted on the coupler and protrudes 0.25 mm beyond the coupler for plasma heat flux protection. The protection tiles allow the LH launcher to be moved closer to the plasma than previously possible. Initial high power (Pnet ~ 700 kW) results show lower reflection coefficients were achieved (Γ2 ~ 0.1) as compared to the old configuration (Γ2 ~ 0.2).

Speaker: Mr Stephen Wukitch (MIT PSFC)

55 FTP/1-2: Acceleration of 1 MeV H- Ion Beams at ITER NB-relevant High Current Density

ITER neutral beam (NB) system requires deuterium negative ion beams of 1 MeV, 40 A at the current density of 200 A/m^2 from a single large negative ion source and an accelerator. This paper summarizes progress in R&D with a reduced size accelerator, so-called “the MeV accelerator” at Japan Atomic Energy Agency (JAEA). In the last Fusion Energy Conference, we reported achievement of 1 MV voltage holding in vacuum for more than one hour. Physics of beamlet deflections due to their own space charges and magnetic field was also reported utilizing a sophisticated three dimensional beam trajectory analyses. The improved voltage holding and a trajectory compensation technique have been applied to the MeV accelerator. Many discharge burn marks have been observed inside the accelerator after long pulse operation reported in the last conference. It was turned out that such discharge marks were observed at positions facing to high local electric field, such as edges, corners, and steps between grid and its support.　In the present MeV accelerator, such positions have been modified, for example, by increasing radii of corners around grid supports, and increasing gap length between grids to lower the local electric concentrations to about 3 ~ 4 kV/mm. For compensation of magnetic deflection, aperture offset was applied at the bottom of the EXG. Magnetic field is formed by small permanent magnets embedded in EXG between aperture lines. Since the polarities are arranged so as to be alternative in each line between apertures, aperture offset of 0.8 mm was defined in the direction against the magnetic deflection. To counteract the beamlet deflection by space charge repulsion, a field shaping plate, a metal plate to deform electric field, were installed around the aperture area for deflection of outermost beamlet inward. Position and thickness of the plate was designed by the analyses. It should be highlighted that reduction of beam direct interception at grids has brought substantial improvement in voltage holding during beam acceleration at around 1 MV. By the improved voltage holding even under beam acceleration, H^- ion beams of 185 A/m^2 (430 mA in total) have been successfully accelerated up to 0.98 MeV. This is a world first demonstration of negative ion beams at high current, high current density and high energy close to the ITER requirements.

Speaker: Mr Takashi Inoue (Japan)

Poster FTP_1-2_talk_poster.pdf.pptx

56 FTP/1-3Rb: Progress on the Development of High Power Long Pulse Gyrotron and Related Technologies

In the development of a higher power dual-frequency gyrotron, a high order mode gyrotron, which permits to select the oscillation at 170GHz or 137GHz, has been fabricated and tested. Short pulse experiments (0.5ms) were performed with 1.3MW power output at more than 30% of the oscillation efficiency for both frequencies. In long pulse experiments, 760 kW/46%/60 s at 170GHz and 540 kW/42%/20 s at 137 GHz are achieved. It is the first time long pulse experiments with the dual-frequency gyrotron/triode electron gun. Since the RF beam direction from the output window is designed to be almost the same for both frequencies, good power couplings to the transmission line, which are 96% for 170GHz and 94% for 137GHz, are obtained by using a pair of identical phase correcting mirrors. Pulse extension is underway aiming for >1MW at CW operation. A 5kHz full power modulation experiment was performed using the 170 GHz gyrotron of TE31,8 mode oscillation. The 5kHz full power modulation was achieved with the full beam modulation by employing a fast voltage switching between the anode and cathode of the triode type electron gun. This satisfies the requirement of ITER. For further improvement, an advanced anode power supply system is proposed to reduce the oscillation period of adjacent mode at the start-up phase of each pulse.

Speaker: Mr Ken Kajiwara (Japan)

57 FTP/P1-01: Investigation and Testing of KTM Divertor Model on Basis of Lithium CPS

Lithium capillary porous system (CPS) as power divertor receivers tiles are proposed to study at the tokamak KTM. Trial start-up of KTM tokamak was done in September, 2010.The main goal of work is creation and testing the model of Li divertor (MLD) at KTM condition. As a result of project realization the Li-technology will be developed and Na-K cooling module of lithium divertor for tokamak KTM will be designed and tested. The construction of MLD, development of process for preparation, protection, clearing and rehabilitation of lithium CPS surfaces in tokamak conditions was completed. At the first stage the test MLD external systems, and test of MLD analog without temperature stabilization system are planned to be done during this year. Second stage will include the adjustment tests of demonstration models of KTM tokamak divertor using reduce plasma parameters. Research of lithium influence on the parameters of plasma discharge and specific power load on the plasma facing components; optimum operating modes definition of the lithium divertor will be done after KTM physical start-up. In order to justify the use of lithium divertor module the experiments were carried out to study sorption characteristics of lithium CPS against hydrogen isotopes. Goal of present work is to assess parameters of hydrogen isotope interaction with lithium CPS under conditions modeling the operation of the tokamak KTM modes to regulate pre-start modes of working gases input. It is necessary to know also the influence of neutron irradiation on parameters of hydrogen isotopes interaction with Li CPS as well as tritium generation, accumulation and release by Li CPS under neutron irradiation for future Li technology application on fusion power reactor. Therefore, we carried out investigations of hydrogen isotopes interaction with Li/Li CPS under reactor conditions using reactor IVG-1M of NNC RK. The gas absorption technique was used to study hydrogen/helium isotope interaction with the samples of lithium CPS. This work is carried out jointly Kazakhstani- Russian organizations in framework of the ISTC project K-1561 with ENEA collaboration.

Speaker: Ms Irina Tazhibayeva (Kazakhstan)

58 FTP/P1-02: Heat Flux and Design Calculations for the W7-X Divertor Scraper Element

The W7-X stellarator is a high-performance optimized stellarator currently under construction in Greifswald, Germany. W7-X will operate under near steady-state conditions (~30 minute pulses), with high input power (15-20MW, 8-10MW in first operational phase). The power and particle exhaust will be handled using an island divertor, in which last closed flux surface is defined by an island chain. New divertor components are being designed to protect the edges of the primary targets during the bootstrap current evolution in scenarios that deviate from the ‘minimum bootstrap current’ configurations. These new components will have peak heat fluxes ~10-12 MW/m2, and will be constructed using CFC (carbon fiber composite) monoblocks of the same type that has been qualified for ITER. The heat flux distribution to the plasma facing components is calculated from field line following in a 3D magnetic field that includes the plasma contribution. The magnetic field is determined from the VMEC [1] (3D equilibrium) and Extender [2] (fields outside the last closed flux surface) codes. The heat flux and strike patterns in the 9 reference W7-X operating configurations will be presented for various values of the total bootstrap current during its evolution. The calculated heat fluxes to the scraper element are used as an input to heat transfer calculations. Several quantities which guide the scraper design are calculated, e.g., CFC surface temperature, fluid temperature rise, and fluid pressure drop. The results of both the heat flux calculations and the heat transfer model are used in a coupled optimization procedure to develop the geometry of the scraper element. The latest divertor geometry will be presented, along with results from the optimization and analysis procedures. *This research was supported by the US Department of Energy, Contracts DE-AC05- 00OR22725. [1] S.P. Hirshman, W.I. van Rij, and P. Merkel, Comp. Phys. Commun. 43, 143 (1986). [2] M. Drevlak, D. Monticello, and A. Reiman, Nucl. Fusion 45, 731 (2005).

Speaker: Mr Jeremy Lore (ORNL)

59 FTP/P1-03: Temporal and Spatial Evolution of In-vessel Dust Characteristics in KSTAR and Dust Removal Experiments in TReD

Visible CCD images from 2010 and 2011 campaigns were analyzed by using image analysis technique. It is found that limiter machine like Tore Supra (TS) has main localized origins as well as many random dust creation events while divertor machines like ASDEX Upgrade (AUG) and KSTAR have origins localized mainly at divertor. The number of dust creation event per second (DCEs; dust creation frequency) is large in both machine configuration just after the machine restart, and decreases as a function of time during the machine conditioning. In TS, DCEs were in a range between 3-6/s while that in divertor machines AUG and KSTAR are between 0.5-4. Dust velocity distribution in 2010 and 2011 campaign was evaluated by using a dedicated software. Only "well defined dust trajectories" at in-board side are considered. Dust velocity is in the range of 10-400m/s with peak velocities of 30 and 50m/s respectively. It is observed that metal dusts created by runaway electron impact have much faster velocity, probably hypervelocity. Short term (daily) and long term (campaign integrated) dust samples were collected and analyzed. Statistics on short term dust samples indicates that the dusts are getting more and more smaller and rounded as a function of time. Most of dusts collected by short term based method are carbon-based materials. Average particle flux is ~1.2×10^4 part/cm2/s and it decreases slightly as a function of plasma operation time. Most of dusts have effective radius in the range between 0.075-3µm, peak@ 0.115µm. Total mass of dusts (extrapolated with area of mid-plane) during the analyzed period is ~122mg (average ~7mg/day, ~0.072µg/cm2). Dusts from long term campaign integrated samples are large size broken graphite pieces, stainless steel, copper, etc. Various shapes and flake-like dusts are observed. Areal distribution of dusts is from 0.43-1701µm^2 (peak at ~2µm^2), and average flux of ~3.48 part/cm^2/s is obtained (~10 part/cm^2/s in AUG). The mass difference of silicon wafer between before and after the dust collection was 0.1±0.05mg on 6.78 cm^2(14.8µg/cm^2). Extrapolate the amount of dusts using the area under the divertor(~3.38m2), 497 mg would be present(~9.6mg/day). The transportation efficiency of Al2O3 dusts in He glow discharge in TReD machine was ~10-15% (injected vs collected). Currently, dust removal rate is obtained as ~10mg/h.

Speaker: Mr Suk-Ho Hong (Republic of Korea)

60 FTP/P1-04: Tungsten Divertor Target Technology and Test Facilities Development

Tungsten divertor target technology development is in progress at IPR for water-cooled divertors of ITER-like tokamak. Test mock-ups are fabricated using tungsten materials in macro-brush as well as mono-block fashion. Vacuum brazing technique is used for macro-brush fabrication whereas high pressure high temperature diffusion bonding technique is used for mono-block fabrication. Experimental facilities are also being set-up at IPR for Non-destructive testing and high heat flux testing of divertor targets. Present paper describes recent results on high heat flux testing of the test mock-ups and briefly mention about some of the experimental test facilities being set-up at IPR.

Speaker: Mr Samir Khirwadkar (India)

61 FTP/P1-05: Comparative Study of Chemical Methods for Fuel Removal

To extend the availability of ITER, tritium stored in the vessel has to be removed on a regular basis. The research on the fuel removal at Forschungszentrum Jülich has been concentrated in recent years on chemical methods including thermo-chemical erosion (TCE) also known as baking in reactive gases, glow-discharge conditioning (GDC) and ion-cyclotron wall conditioning (ICWC). The studies were conducted in the tokamak TEXTOR and in laboratory devices using pre-characterized samples with deuterated carbon layers. GDC, in contrast to TCE and ICWC, is not applicable in the presence of the nominal magnetic field. Our investigations showed that GDC can be operated at a magnetic field of up to 10 mT and is therefore compatible with the ferritic inserts foreseen in ITER. The TCE using oxygen as the removal gas can effectively be employed at elevated temperatures of at least 300C. Plasma-based GDC and ICWC can also be applied at lower wall temperatures. TCE is equally efficient in cleaning from the wall surface as from the remote areas such as gaps of castellations. GDC is homogeneous along the wall surface except for small recessed areas like gaps. ICWC is typically inhomogeneous along the poloidal circumference. Applying the radial magnetic field, we were able to control the poloidal position of the main IC plasma production. Thus, some wall regions, e.g. the divertor, can selectively be exposed to ICWC. Owing to higher ion fluxes to the wall, ICWC activates a larger amount of neutrals than GDC, which then penetrate in gaps and clean gap walls efficiently. Removal rates with oxygen were typically by a factor of 3 - 10 higher than with hydrogen and ammonia and 10 - 30 than with nitrogen. The estimates using the highest removal rate for ICWC show that about 2 hours are needed to remove the layer deposited within one ITER pulse. The application of ammonia in TCE led to the pealing-off of layers, which is a potential dust production mechanism. However, it appears to be suitable for the non-oxidizing cleaning of metallic mirrors envisaged for optical diagnostics in ITER.

Speaker: Mr Arkadi Kreter (Germany)

62 FTP/P1-06: Analysis of Establishment and MHD Stability of a Free Curve-Surface Flow for Liquid Metal PFCs

An innovation concept of three layer s guidable liquid metal free curve-surface flow is addressed and its establishment and MHD stability are also analyzed on theoretically Layer I is a basic conduction layer, layer II is a key adjust layer, layer III is the surface layer. To adjust layer I and II in suitable flowing conditions, the MHD effect stability surface layer III can be obtained. In meantime, the layer I and II can be as the heat sink and the coolant flow (it is also suitable to a flat surface flow to avoid rivulet flow). According to Newton's laws and fluid mechanical principles, the analysis results show that an MHD effect stability free curve surface flow can be established under a given curve surface in a gradient magnetic field.

Speaker: Mr Xuru Duan (China)

63 FTP/P1-07: “Snow Flakes” Divertor and 10 MA Scenarios in FAST

The overarching FAST goals lead to a more flexible design and to a research plan based on three DEMO and ITER priorities: A) exploring plasma wall interaction in reactor relevant conditions, B) testing tools and scenarios for safe and reliable tokamak operation up to the border of stability, C) addressing fusion plasmas with a significant population of fast particles, plus being complementary to the JT60-SA missions. Unique of the FAST approach is the capability of addressing all of them simultaneously in a single, fully integrated scenario with dimensionless physics parameters very close to DEMO and ITER. FAST has the possibility to tackle the power exhaust problem in regimes relevant to DEMO with an actively cooled W-divertor capable to sustain loads up to 20 MW/mq with P/R~22 MW/m. A “Snow Flakes” divertor can be implemented in FAST with the present poloidal coils up to the reference scenario with Ip=6.5 MA. According to the 2-D plasma edge code TECXY the peak power flow along the SOL field lines can be reduced by a factor ≈3.0. A new FAST scenario has then been designed at Ip=10 MA, BT=8.5T, q(95)≈2.3. Transport simulations by using the code JETTO and the first principle transport model GLF23 indicate that, under these conditions, FAST could achieve an equivalent Q≈3.5. FAST will be equipped with a set of feedback controlled active coils located between the first wall and the vacuum vessel (≈25 cm far from the plasma edge) and accessible for maintenance with the remote handling system. Preliminary studies indicate that these coils can carry currents up to 20 kA (≈4 MA/mq) with AC frequency up to few kHz. The coil set will produce magnetic perturbation with toroidal number n=1 or n=2. MHD analysis performed with the linear code MARS (both assuming the presence of a perfect conductive wall at r/a=1.3 and using the exact 3D resistive wall structure) shows the possibility of the FAST conductive structures to stabilize n=1 and n=2 ideal modes. This leaves therefore room for active mitigation of the resistive mode (down to a characteristic time of 1 ms) for safety purposes. The main target of this experiment is the preparation of a complete and reliable low q(95) scenario, ready to be transferred to a possible ITER scenario very close to its operational limits.

Speaker: Mr Flavio Crisanti (Italy)

64 FTP/P1-08: Preliminary Safety Analysis of the Indian Lead Lithium Cooled Ceramic Breeder Test Blanket Module System in ITER

Safety analysis has been carried out for the safety licensing of Lead Lithium cooled Ceramic Breeder (LLCB) Test Blanket Module (TBM) system; INDIA’s proposed prototype of DEMO blanket concept for testing in International Thermonuclear Experimental Reactor (ITER). A set of four reference accidents is identified for LLCB TBM System. Each accidental sequence begins with a Postulated Initiating Event (PIE) identified through Failure Modes and Effects Analysis (FMEA) at component level. The analysis address specific reactor safety concerns, such as passive removal of decay heat, pressurization of confinement buildings, vacuum vessel pressurization, release of activated products and tritium during these accidental events and hydrogen production from chemical reactions between lead-lithium liquid metal and beryllium with water. An in-house customized computer code is developed and through these deterministic safety analyses the prescribed safety limits are shown to be well within limits for Indian LLCB-TBM design and it also meets overall safety goal for ITER. This paper reports transient analysis results of the safety assessment.

Speaker: Mr Vilas Chaudhari (India)

65 FTP/P1-09: Technology Gaps for the Fuel Cycle of a Fusion Power Plant

Control and management of the fuel and fusion product streams is one of the most difficult issues for fusion power plant development. This function is provided by the fusion fuel cycle and addresses the vacuum pumping systems, the fuelling systems, the tritium plant systems and the in-vessel components, especially the divertor and the breeding blankets with their associated periphery. The design of these systems decides the accumulated tritium inventories and the processing times of the unburnt fuel, which both must be absolutely minimised. As main tool to match this requirement, a simplified fuel cycle architecture has been developed based on the concept of Direct Internal Recycling (DIR) of the unburnt fuel close to the divertor. Advantages and compromises of this concept will be discussed. In all areas where ITER does not serve as a convincing basis for technology scale-up, new technologies have to be developed. This paper is reporting results of ongoing activities in EU to assess the fuel cycle systems of DEMO. It will go through the current state of technology, address the technical readiness, identify the areas which are considered to require essential supporting R&D towards a functional system for a reactor, and propose potential solutions together with experimental information – if available - or outline R&D paths for their development.

Speaker: Mr Christian Day (Germany)

Poster Day_IAEA-FECPoster_2012_A0.pdf

66 FTP/P1-10: Simulation Experiments of ELM-like Transient Heat and Particle Loads using a Magnetized Coaxial Plasma Gun

A magnetized coaxial plasma gun (MCPG) device has been developed for simulation experiments of transient heat and particle loads during type I edge localized modes (ELMs) predicted in ITER. The MCPG has been recently upgraded to increase surface absorbed energy density up to ~ 2 MJ/m^2 that makes it possible to investigate of tungsten (W) melting behaviors. In the experiment, mono-block W samples, to be used for the ITER divertor, were exposed to repetitive pulsed hydrogen plasmas with duration of ~ 0.2 ms, incident ion energy of ~ 50 eV, and surface absorbed energy density of ~ 0.7, 1.4, and 2 MJ/m^2. No melting occurred on the mono-block W surface at energy density of ~ 0.7 MJ/m^2, while major cracks were formed. Cracking and melting of the mono-block W surface were clearly observed at energy density of ~ 1.4 and 2 MJ/m^2. Micro-sized cracks were identified for energy density above the melting threshold. It is considered that the micro-sized cracks were formed due to surface melting and resolidification in each plasma pulse. The mono-block W samples with pulsed plasma irradiation will be exposed to cyclic heat loads of ~ 20 MW/m^2 in an electron beam facility JEBIS at JAEA in order to investigate damage of ITER divertor materials under a combination of steady-state and transient heat loads. Moreover, we introduce a new experiment using two MCPG devices to understand vapor shielding effects of a W surface under ELM-like pulsed plasma bombardment. The second plasmoid is applied with a variable delay time after the first plasmoid. A vapor cloud layer in front of the W surface produced by the first plasmoid irradiation could shield the second pulsed plasma load on the W surface. In this upgrade, weight loss measurements of W samples after pulsed plasma exposures became possible, which is a grate advantage for quantitative evaluation of vapor shielding effects on erosion of W.

Speaker: Mr Yusuke Kikuchi (Japan)

67 FTP/P1-11: Plasma Characteristics of the End-cell of the GAMMA 10 Tandem Mirror for the Divertor Simulation Experiment

In this paper, detailed characteristics and controllability of plasmas emitted from the end-cell of the GAMMA 10 tandem mirror are described from the viewpoint of divertor simulation studies. In the case of only ICRF plasma, the heat flux of 0.8 MW/m^2 has already been achieved and proportionally increased with the ICRF power for ion heating. The energy analysis of ion flux by using end-loss ion energy analyzer (ELIEA) proved that the obtained high ion temperature (100 - 400 eV) was comparable to SOL plasma parameters in toroidal devices and was controlled by changing the ICRF power. Parallel ion temperature Ti// determined from the probe and calorimeter shows a linear relationship with the stored energy in the central-cell and agrees with the results of ELIEA. Recently additional plasma heating experiment using ICRF in the anchor-cell (RF3) was carried out in order to improve the performance. The time behavior of the plasma line-density and end-loss ion flux is shown in Fig. 4. A significant enhancement of the line-density is observed and the resultant ion flux becomes two times higher than that without RF3. The particle flux is estimated to be 6.5×10^22 particles/s•m^2, which indicate an effectiveness of additional heating with ICRF wave in the neighboring cells toward the improved E-divertor experiments for achieving the targeted parameters of this project (PHEAT ~20 MW/m^2, Γi = 10^23-24/ m^2 sec). We have started various experiments such as radiator gas injection onto the tungsten target and visible measurement of plasma-gas-material interactions with a fast camera. Numerical simulation studies have also started in the end-cell for understanding the behavior of plasmas in divertor simulation experiments. In this spring a large-sized divertor experimental module will be installed and radiative cooling experiments of the end-cell plasma are planned by using gas injection into the module for realizing the detached plasma condition. [1] Y. Nakashima, et al., Fusion Eng. Design volume 85 issue 6 (2010) 956-962. [2] Y. Nakashima, et al., Trans. Fusion Sci. Technol. 59 No.1T (2011) 61-66. This work is supported by the bidirectional collaboration research program with National Institute for Fusion Science.

Speaker: Mr Yousuke Nakashima (Japan)

68 FTP/P1-12: Progress of High Heat Flux Component Manufacture and Heat Load Experiments in China

High heat flux components for first wall and divertor are the key subassembly of the present fusion experiment apparatus and fusion reactors in the future. It is requested the metallurgical bonding among the plasma facing materials (PFMs), heat sink and support materials. As to PFMs, ITER grade vacuum hot pressed beryllium CN-G01 was developed in China and has been accepted as the reference material of ITER first wall. Additionally pure tungsten and tungsten alloys, as well as chemical vapor deposition (CVD) W coating are being developed for the aims of ITER divertor application and the demand of domestic fusion devices, and significant progress has been achieved. For plasma facing components (PFCs), high heat flux components used for divertor chamber are being studied according to the development program of the fusion experiment reactor of China. Two reference joining techniques of W/Cu mockups for ITER divertor chamber are being developed, one is mono-block structure by pure copper casting of tungsten surface following by hot iso-static press (HIP), and another is flat structure by brazing. The critical acceptance criteria of high heat flux components are their high heat load performance. A 60 kW Electron-beam Material testing Scenario (EMS-60) has been constructed at Southwestern Institute of Physics (SWIP),which adopts an electron beam welding gun with maximum energy of 150 keV and 150×150 mm2 scanning area by maximum frame rate of 30 kHz. Furthermore, an Engineering Mockup testing Scenario (EMS-400) facility with 400 kW electron-beam melting gun is under construction and will be available by the end of this year. After that, China will have the comprehensive capability of high heat load evaluation from PFMs and small-scale mockups to engineering full scale PFCs. A brazed W/CuCrZr mockup with 25×25×40 mm3 in dimension was tested at EMS-60. The heating and cooling time are 10 seconds and 15 seconds, respectively. The experiment procedure is 3 MW/m2 by 200 cycles and then 6 MW/m2 by 1000 cycles, following by 8.5 MW/m2 for 200 cycles and 11 MW/m2 for 100 cycles. No off-normal surface temperature change and cracks were observed. The similar screening tests of small-scale mono-block W/CuCrZr mockups will be tested soon. Next large size brazed W/CuCrZr components will be manufactured and evaluated.

Speaker: Mr Xiang Liu (China)

69 FTP/P1-14: Recent Progress in the NSTX/NSTX-U Lithium Program and Prospects for Reactor-relevant Liquid-lithium Based Divertor Development

Developing a reactor compatible divertor has been identified as a particularly challenging technology problem for magnetic confinement fusion. While tungsten has been identified as the most attractive solid divertor material, the NSTX/NSTX-U lithium (li) program is investigating the viability of liquid lithium (LL) as a potential reactor compatible divertor plasma facing component (PFC). In the near term, operation in NSTX-U is projected to provide reactor-like divertor heat loads ≤ 40 MW/m^2 for 5 s. During the most recent NSTX campaign, ~ 0.85 kg (1.5 liter) of li was evaporated onto the NSTX PFCs where a ~50% reduction in heat load on the LLD was observed, attributable to enhanced divertor bolometric radiation signal. This reduced divertor heat flux through radiation observed in the NSTX LLD experiment is consistent with the results from other li experiments and calculations. These results motivate an LL-based closed radiative divertor concept proposed here for NSTX-U and fusion reactors. With an LL coating, the li is evaporated from the divertor strike point surface due to the intense heat. The evaporated li is readily ionized by the plasma due to its low ionization energies, and the ionized li ions can radiate strongly, resulting in a significant reduction in the divertor heat flux. Due to the rapid plasma transport in divertor plasma, the radiation values can be significantly enhanced up to ~ 11 MJ/cc of LL. This radiative process has the desired function of spreading the focused divertor heat load to the entire divertor chamber, which facilitates the divertor heat removal. The LL divertor surface can also provide a “sacrificial” surface to protect the substrate solid material from transient high heat flux such as the ones caused by the ELMs. The closed radiative LLD concept has the advantages of providing some degree of partition in terms of plasma disruption forces on the LL, li particle divertor retention, and strong divertor pumping action from the li-coated divertor chamber wall. By operating at a lower temperature than the first wall, the LLD can serve to purify the entire reactor chamber, as impurities generally migrate toward lower temperature li-condensed surfaces. To maintain the LL purity, a closed LL loop system with a modest capacity (e.g., ~ 1 liter/sec for ~ 1% level “impurities”) is envisioned for a steady-state reactor.

Speaker: Mr Masayuki Ono (USA)

70 FTP/P1-15: Effects of the Lithium Concentration on Tritium Release Behaviors from Advanced Tritium Breeding Material Li_2+xTiO_3

Lithium-enriched Li_2TiO_3, such as Li_2.2TiO_3 and Li_2.4TiO_3 (Li_2+xTiO_3), is considered as one of advanced tritium breeding materials in fusion reactors. Densities of irradiation defects in Li_2+xTiO_3 will increase with increasing lithium concentration. It is expected that tritium is trapped by irradiation defects and its release behavior will be affected by the density of the defects. Therefore, elucidation of enhancement effects of the irradiation defects on tritium release behavior in Li_2+xTiO_3 is an important issue from a viewpoint of tritium recovery Thermal neutron irradiation was performed for Li_2.0TiO_3, Li_2.2TiO_3 and Li_2.4TiO_3 at the Research Reactor Institute, Kyoto University. The thermal neutron flux was 5.5 x 10^12 n cm^-2 s^-1 and fluence was 3.3 x 10^15 n cm^-2. Thermal Desorption Spectroscopy (TDS) measurements were carried out to investigate the release behaviors of tritium generated in Li_2+xTiO_3. Electron Spin Resonance (ESR) measurements were also carried out to estimated the densities and the annihilation behaviors of irradiation defects introduced by thermal neutron irradiation. The densities of the irradiation defects in Li_2+xTiO_3 evaluated by the peak areas of the ESR spectra were increased with increasing lithium concentrations. An X-ray diffraction (XRD) showed that the Li_4TiO_4 structure was formed in the lithium enriched samples. The Li_4TiO_4 structure may affect the enhancement to the density of defects for lithium enriched Li_2TiO_3. It was found that the defects for Li_2.0TiO_3 observed by ESR were annihilated around 600 - 800 K, although those for Li_2.2TiO_3 and Li_2.4TiO_3 were annihilated around 400 - 600 K. In addition, two tritium desorption stages were observed at 450 K and 600 K by the TDS measurement. The tritium release in lower temperatures was enhanced with increasing lithium concentration, corresponding to the annihilation temperature region of the defects in Li_2.2TiO_3 and Li_2.4TiO_3. It was concluded that tritium release at the lower temperatures was initiated by the existence of Li_4TiO_4 structure formed in the lithium-enriched Li_2TiO_3. These results indicate that use of lithium-enriched Li_2TiO_3 enables the recovery of tritium about lower temperature compared to Li_2.0TiO_3 and would be an advantage for the fuel recovery from the tritium breeding materials in the fusion reactors.

Speaker: Mr Hiromichi Uchimura (Japan)

Poster FTP_P1-15_Hiromichi_Uchimura.pdf

71 FTP/P1-16: Progress in the Development of the ECRF System for JT-60SA

The electron cyclotron range of frequency (ECRF) system for JT-60SA is composed of 9 gyrotrons with the total injection power of 7 MW and the pulse duration of 100 s, transmission line with the total length of ~80 m, and linear-motion launchers. This paper comprehensively presents recent progress in the development of the ECRF system. Major results are (1) the extension of gyrotron output energy (60 MJ) by a factor of ~2 compared with the results presented in the last IAEA FEC through the installation of a new 60.3 mm diameter transmission line, (2) successful tests on optical and mechanical characteristics of a linear-motion launcher, which enables >1 MW, 100 s injection with a wide coverage of toroidal (typically –15 deg. to +15 deg.) and poloidal (–40 deg. to +20 deg. with respect to the horizontal plane) directions, (3) the development of a dual-frequency gyrotron which can output 110 GHz and 138 GHz ECRF at >1 MW for 100 s for heating (ECH) and current drive (ECCD) typically around the half minor radius at the full toroidal magnetic field of 2.3 T in JT-60SA.

Speaker: Mr Akihiko Isayama (Japan)

72 FTP/P1-17: Feasibility and R&D Needs of a Negative Ion Based Neutral Beam System for DEMO

The R&D requirements of a heating and current drive (H&CD) system for a demonstration fusion power plant (DEMO) are presently assessed within the EFDA 3PPT activities. The requirements of the H&CD system will strongly depend on the DEMO scenario; the most demanding requirements are defined by a steady state tokamak. For such a cw CD system with several 100 MW power, the key issues are the achievement of adequate plug-in efficiency, availability and reliability. The neutral beam injection (NBI) system, at present a key system H&CD of magnetic fusion devices including ITER, is also a candidate for DEMO, due to its high current drive efficiency. A DEMO cw CD NBI system will be based like the ITER NBI system on production, acceleration and neutralisation of negative deuterium ions. For sufficient current drive neutral particle energies of several 100 kV at minimum are required. The neutralisation efficiency of negative ions is still about 60% at these high energies whereas the efficiency of positive ions is below a few percent. The paper concentrates on two issues which are presently addressed by IPP R&D activities: (i) the enhancement of the overall wall plug efficiency from approx. 25% for the ITER system to more than 50%; and (ii) the reliability and maintainability of the negative deuterium ion source. At present, the only promising technology to enhance the plug-in efficiency to the required values above 50% is a laser neutraliser system with neutralisation efficiency of almost 100%. Among other benefits, a laser neutralizer does not require high heat flux components, i.e. a residual ion dump, in the beamline, so that the negative ion source and the accelerator will be the most crucial parts for the reliability of the system. Both parts are exposed to local power densities of several 10 MW/m2 in the present ITER design. The operation at reduced source filling pressure (below 0.3 Pa, the ITER requirement) and a Cs-free source are identified to be very beneficial by reducing the local power loads and removing the need for a regular maintenance. On the other hand, NBI systems with energies above 1 MeV for enhanced current drive efficiency seem to be unfeasible with the present technology. Details and concepts will be discussed in more detail in the paper, highlighting the feasibility and R&D needs from the present available ITER technology.

Speaker: Ms Ursel Fantz (Germany)

73 FTP/P1-18: Progress in the Development of Long Pulse Neutral Beam Injectors for JT-60SA

To realize a 100 s injection of neutral beams for JT-60 Super Advanced (JT-60SA), the feasibility of the long-pulse production of the ion beams is investigated. Using the JT-60 positive ion source, the long-pulse production of the positive ion beams is confirmed to be feasible for JT-60SA by analyzing the results obtained in the productions of full power D+ ion beams of 85 keV, 27.5 A for 30 s and a half power H+ ion beams of 60 keV, 18A for 200 s. This shows that the JT-60 positive ion source is expected to be reused in JT-60SA without modifications. To realize a stable acceleration of the high current D- ion beams of 500 keV for 100 s without interruptions due to breakdowns, vacuum voltage holding capability of the multi-aperture grid designed for JT-60 SA is firstly clarified in the world. Based on results, the gap length of the accelerator was tuned for the grid area and number of the apertures for JT-60SA. As the result, high voltage holding during 100 s without breakdowns was attained at 200 kV for one acceleration gap at an optimum gap length for sufficient suppressions of the beam losses. This suggests that stable long-pulse acceleration to 500 keV could be expected for three acceleration stages in JT-60SA.

Speaker: Mr Masaya Hanada (Japan)

74 FTP/P1-19: Local Current Injector System for Nonsolenoidal Startup in a Low Aspect Ratio Tokamak

The Pegasus experiment is an ultralow aspect ratio spherical tokamak that is developing nonsolenoidal startup and current growth techniques. Helicity injection from localized current sources in the plasma periphery have produced plasma currents up to 0.15 MA with less than 4 kA injected, and the resulting plasmas provide stable target plasmas for further current drive. This localized helicity injection startup technique requires the development of robust, high current density sources (~ 1 kA/cm^2) that can exist in the plasma scrapeoff region during plasma initiation, growth, and possibly sustainment. An integrated assembly of active arc plasma sources and a passive electrode emitter is under development for this application to MA-class spherical tokamak applications. Compact arc plasma current sources are used for initial current injection along vacuum field lines to produce a tokamak-like plasma through null formation and Taylor relaxation. Further current growth is realized through helicity injection from these arc sources or passive electrodes in the plasma edge region. Use of passive metallic electrodes can greatly simplify the design and allow for higher injected currents to optimize the resulting plasma current. The compact, active arc sources provide an extracted current stream that appears to be governed by a double layer sheath at the arc exit region. At voltages greater than eV/kT ~ 10 and high currents, the extracted current scales as V^1/2, presumably due to sheath expansion or the Alfven-Lawson cuurent limit for electrons. Control of the arc plasma density through active gas feed control and detailed design of the arc chamber should provide active control of the effective loop voltage applied to the tokamak plasma. The arc source and electrode structures are isolated from the edge plasma by a local BN limiter and nearby scraper limiter assembly. This mitigates interactions between the injector assembly and the plasma, and resulting impurities in the plasma are negligible. High-current power supplies using IGCT solid-state switches regulate the injector current according to a pre-programmed waveform. This startup approach places minimal demands on machine design, appears scalable to MA levels in large facilities, and offers the possibility of hardware that can be withdrawn before a fusion plasma enters the nuclear burn phase.

Speaker: Mr Raymond Fonck (USA)

Poster Fonck_FTP-P1-19_poster.pdf

75 FTP/P1-20: Development of MW Gyrotrons for Fusion Devices by University of Tsukuba

Over-1 MW power gyrotrons for ECH have been developed in the joint program of NIFS and University of Tsukuba. The obtained maximum outputs are 1.9 MW for 0.1 s on the 77 GHz LHD tube and 1.0 MW for 1 ms on the 28 GHz one, which are the new records in these frequency ranges. In long pulse operation, 300 kW for 40 min at 77 GHz and 540 kW for 2 s at 28 GHz were achieved. Three 77 GHz gyrotrons have already been installed and operated stably in LHD. More than 3.4 MW has been injected into LHD plasma contributing to producing the electron temperature Te of 20 keV. A new program of 154 GHz 1 MW development has started for high density plasma heating in LHD and the first tube has been fabricated. A 28 GHz gyrotron is also required at QUEST in Kyushu University, where 0.4 MW continuous wave (CW) one is needed. A few sec. with multi-MW gyrotron is useful for novel Electron Bernstein Wave (EBW) heating system of NSTX in PPPL. Based on the first 28 GHz 1 MW tube results, the design of the new 28 GHz tube has progressed, which satisfies both requirements. These lower frequency tubes like 77 GHz or 28 GHz one are also important for advanced magnetic fusion devices, which use the EBW heating / current drive.

Speaker: Mr Ryutaro MINAMI (Japan)

Poster IAEA2012poster_v2_10_5_2012.pdf

76 FTP/P1-21: Status and Plan of the Key Actuators for KSTAR Operation

After 1st plasma in 2008, the Korea Superconducting Tokamak Advanced Research (KSTAR) has been intensively upgraded to supply key actuators such as in-vessel components and heating system. The in-vessel components satisfactorily provided an essential condition for experiments on plasma shaping, H-mode, and edge localized mode (ELM) suppression in 2010 and 2011. There was also a big progress in upgrade of heating system. The neutral beam injection system demonstrated 1.5 MW of NB power with 95 keV of beam energy. The 170 GHz ECCD system provided an additional tool for ECH pre-ionization experiment with 0.6 MW/2 s of launched power. Upgrade of the key actuators in 2012 is mainly focused on upgrade of NBI and Lower Hybrid Current Drive (LHCD) system. After upgrade of the NBI system, available total NB power is expected to be more than 3.0 MW in 2012 campaign. LHCD system using 5 GHz is being developed with 8 arrays of 4-way splitter waveguide launcher. After minor upgrade, there will be a major change in hating system and in the diverter concept in 2nd stage of KSTAR operation from 2012 to 2016. In this period, the heating and current drive system will be extended in the deliverable power to 14 MW with 8 MW of NB and 2 MW of LHCD and ECCD.

Speaker: Mr Hyung Yeol Yang (Republic of Korea)

77 FTP/P1-22: Advances in Lower Hybrid Current Drive Technology on Alcator C-Mod

Lower Hybrid Current Drive is an attractive option for non-inductive tokamak operation due to its high current drive efficiency and ability to drive current off axis. The parameters of the Alcator C-Mod LHCD system (f = 4.6 GHz, B ≈ 5.5 T, n_e ≈ 10^20 m^-3) are similar to the proposed LHCD system on ITER. The 0.5 s pulses achieved in previous operation are sufficiently long as compared to the current relaxation timescale (tau_R ~ 0.2 s) for quasi-steady state non-inductive operation. Longer pulses are necessary for other plasma parameters to reach equilibrium once the current profile has been modified. Modeling of LH wave propagation indicates that the loss of LHCD efficiency at high n_e can be mitigated by enhancing the single pass power absorption. This paper will describe improvements in LHCD technology on C-Mod designed to extend pulse length (to ~ 3 s), increase power delivered to the plasma through reducing reflection coefficients (to < 10 %), and increase single-pass absorption at high n_e. Total net LH power with the additional antenna will be ~2 MW. An off mid-plane launcher has been designed combining the 4-way poloidal splitting concept of the current LH launcher on C-Mod with a toroidal bi-junction. The new antenna was optimized to decrease reflected power and increase directivity over a broad range of plasma conditions and launched n|| values. The four rows of the launcher are located above the mid-plane in order to exploit the poloidal upshift of n|| as rays propagate from the antenna into the plasma. The n|| upshift results in better wave penetration to the plasma core at high n_e ( > 10^20 m^-3) and stronger single-pass absorption of the LH waves. The maximum LHCD pulse length on C-Mod is limited by heating in the collector of the klystrons. Modeling shows that the klystron can operate for 5 seconds without boiling the coolant at full RF power, but the coolant will boil after 1.2 s of beam-on time with no RF power. The maximum pulse length was restricted to 0.5 s to prevent boiling. Increasing the LH pulse length to ~3 s will allow the LH system to remain on for ~15 x tau_R and extend the I_p flattop. The Transmitter Protection System was redesigned to model the coolant temperature in real time. The electron beam is shut off if the TPS determines that the coolant boils. The TPS upgrade has been installed and operated on C-Mod.

Speaker: Mr Gregory Wallace (USA)

78 FTP/P1-23: Preparation of Steady State Operation of the Wendelstein 7-X Stellarator

Wendelstein 7-X has been designed to demonstrate the steady state capability of the stellarator concept. At 10 MW of heating power a pulse duration of 30 minutes is envisaged. Short pulses of additional heating power are foreseen to access beta and equilibrium limits and study fast ion confinement and fast ion driven instabilities. The large variety of time scales is strongly affecting the design of plasma diagnostics, heating and fuelling systems, data acquisition and device control. For steady state heating ten 1 MW continuous wave gyrotrons are foreseen, operating at 140 GHz second harmonic heating at 2.5 T. Using a system of mirrors, relaying the micro waves through air to Wendelstein 7-X, a very high transmission efficiency has been achieved. Front steering mirrors, one for each gyrotron, individually change the poloidal and toroidal launch angles, thus controlling the radial deposition and current drive. Recent modifications to the gyrotron design include an improved power handling in the collector using a rotating transverse magnetic field. The main heating scenarios are 2nd harmonic X-mode (X2) heating below the cut-off density of 1.2x10^20 m^-3 and 2nd harmonic O-mode (O2) heating at higher densities. Owing to non-absorbed power, significant levels of stray radiation are expected for O2-heating, during the transition from X2- to O2-heating, and also during plasma start-up with electron cyclotron resonance waves. Therefore all in-vessel components have to be qualified and if necessary protected to withstand up to 50 kW/m^2 of continuous micro-wave power flux. Many diagnostic techniques require a specific adaptation or even new developments to cope with steady state operation. Besides the measurement of fast events, also the long times scales have to be covered. As a consequence not only data rates increase, but the total amount of data. This requires special efforts for real time plasma control, and for continuous data acquisition and data archiving, and makes new, automated concepts for data processing and analysis indispensable. The paper summarizes the main technologies required for steady state operation of Wendelstein 7-X, including concepts for integrated data analysis and interpretation, and discusses the possible relevance for other experiments aiming at long pulse operation.

Speaker: Mr Robert Wolf (Germany)

Poster Poster-v1.pdf

79 FTP/P1-25: ECH-assisted Startup using ITER Prototype of 170 GHz Gyrotron in KSTAR

The newly installed electron cyclotron heating and current drive (EC H&CD) system with a frequency of 170 GHz was successfully commissioned and used for the second-harmonic ECH-assisted startup in 2011 operational campaign of the KSTAR. As a RF power source, ITER pre-prototype of 170 GHz, 1 MW continuous-wave gyrotron was loaned from the Japan Atomic Energy Agency (JAEA). The Gaussian beam output from the gyrotron passes through an edge-cooled diamond window and is coupled to an HE11 corrugated waveguide via two phase correcting mirrors in a matching optics unit (MOU). The power coupled to the HE11 corrugated waveguide is delivered to the launcher by the transmission total length of 70 meters. For the first 1 MW EC H&CD system, 1-beam based 1 MW equatorial launcher is installed in the KSTAR Bay E-m. The launcher has been designed and fabricated in collaboration with both Princeton Plasma Physics Laboratory (PPPL) and Pohang University of Science and Technology (POSTECH). During the KSTAR 2011 campaign, 10-s pulse at 0.6 MW EC beam was reliably injected into the plasma. Also, 170 GHz second harmonic ECH-assisted start-up was successful leading to reduce the flux consumption at toroidal magnetic field of 3 T. In this experiment, the flux consumption until the plasma current flat-top was reduced from 4.13 Wb for pure Ohmic to 3.62 Wb (12 % reduction) for the perpendicular injection. When the EC beam is launched with toroidal angle of 20 deg with respect to the outward radial direction at the steering mirror, more reduced magnetic flux consumption was obtained with 3.14 Wb (24 % reduction) compared with pure OH plasmas. After the 2011 campaign, the gyrotron has been fully commissioned with the output power of 1 MW at the diamond window and the frequency of 170 GHz by precise alignment of the magnet to the gyrotron axis.

Speaker: Mr Jin Hyun Jeong (Republic of Korea)

80 FTP/P1-26: The Influence on System Design of the Application of Neutral Beam Injection to a Demonstration Fusion Power Plant

Steady state fusion power plants require significant non-inductive current drive possibly provided by neutral beam injection (NBI); in addition, NBI can be used for q-profile control for plasma stability. Economic considerations impose limitations on the necessary current drive and electrical efficiencies of the NBI system, with a figure of merit defined by the product of these quantities being greater than 0.25Am-2W-1. The impact of the plasma density profile on non-inductive current drive is assessed using the PENCIL code and the results expressed in terms of a spatial map of current drive efficiency. The effect of beam shinethrough on the far first wall is used to determine limiting tangency radius and elevation for beam injection and thus define an average current drive efficiency. This is used in conjunction with the figure of merit to define the lower limit of electrical efficiency, for a given plasma profile, that will provide an economic power plant. A NBI system code has been developed to allow the electrical efficiency of a combination of different technologies to be calculated. Gas, plasma and photon neutralizers and energy recovery systems can be considered along with the effect of beam divergence, background gas density and DC electrical efficiency. The code is used to investigate system sensitivities and to identify strategy for reducing technical risk. The primary determinant of electrical efficiency is neutralization efficiency, followed by beam stripping and transmission losses. Despite the gas neutralizer being limited to 58% neutralization efficiency, it is shown that the figure of merit can be achieved by a combination of energy recovery and reduced stripping. For the photoneutralizer system, where the neutralization efficiency can be varied up to 95%, the range of options is greater. Adding energy recovery to the system reduces the sensitivity of electrical efficiency to the neutralization, offering reduced risk and hence improved reliability options. However, the range of options available depends on the current drive obtainable for a particular plasma density profile, thus the plasma itself influences the choice of NBI technology. This work was funded by the RCUK under the contract of Association between EURATOM and CCFE. This work was carried out within the framework of the European Fusion Development Agreement.

Speaker: Ms Elizabeth Surrey (UK)

81 FTP/P1-27: Fusion Material Irradiation Test Facility at SNS

Computational modeling and experimental studies provide compelling evidence that displacement damage formation induced by fission neutrons and the 14.1 MeV neutrons representative of D-T fusion are quite similar. However, helium and hydrogen production levels with a D-T neutron energy spectrum are much higher. The impact of these gaseous transmutation products is a critical unresolved issue which is being addressed by combining numerical models and specialized ion and neutron irradiation experiments. Because of the uncertainties associated with both modeling and ion irradiation experiments, there is a clear need for an accessible irradiation facility that can provide near prototypic levels of helium and hydrogen. A modest range of He/dpa ratios is desirable to help calibrate and verify the modeling studies. The scientific understanding obtained would also enable more effective use of a future large-volume fusion engineering irradiation facility (such as IFMIF) when it becomes available. A conceptual level design for a fusion materials irradiation test station (FMITS) for installation at the Spallation Neutron Source (SNS) has been completed. Samples would be located within two horizontal tubes in front of the mercury target. For these specimen locations, the back-scattering neutron flux spectra should be close to the ITER fusion spectrum. The PKA spectra at the FMITS samples were also compared to those for ITER, and the results show good agreement. Material damage rates would be 1.6–5.5 dpa/yr for steel, and 1.8–3.4 dpa/yr for SiC. The test station would be water-cooled with a variable inert-gas blanket for temperature control. Thermal analysis shows that the sample temperatures can be maintained as high as 600°C even if average beam power varies by 50%. The FMITS assembly is designed to be installed over a target module and can be reused with multiple targets. The paper describes the design concepts for the FMITS gas system with additional hardware and controls, the mechanical layout of the FMITS assembly including an example experiment, revisions to the existing target shroud, target carriage modifications for FMITS utilities, and remote handling procedures and logistics. Safety and reliability impacts were also evaluated and appear to be acceptable.

Speaker: Mr Mark Wendel (Oak Ridge National Laboratory)

82 FTP/P1-28: Potential for Improvement in High Heat Flux HyperVapotron Element Performance Using Nanofluids

HyperVapotron (HV) elements have been used extensively in high heat flux neutral beam stopping devices in nuclear fusion research facilities such as JET and MAST. These water-cooled heat exchangers use a cyclic boiling heat transfer mechanism to effectively handle power densities of the order of 10-20 MW/m2, but are inherently limited by their critical heat flux. The use of a nanofluid as the coolant, instead of water, promises to enhance the heat transfer performance of the HV and increase the critical heat flux by a factor of 2 or 3. Such enhancement would produce a step-change improvement in the power handling capability of future high heat flux devices such as the divertor and heating and current drive systems. This paper describes the potential to improve HV performance using nanofluids. A molecular dynamics simulation (MDS) code has been used to determine the existence of heat transfer mechanisms that depart from classical thermodynamics and that might explain the augmented heat transfer performance of nanofluids. A basic nanofluid is synthesised and the path of the nanoparticle along a temperature gradient is tracked and compared with a simulation of the base fluid alone. The results indicate a new type of heat transfer mechanism in which the nanoparticle exhibits greater thermal diffusion, with an optimum particle size manifesting in the data. Experiments have been conducted in which the flow field in a full-scale HV model is visualised and measured using particle image velocimetry (PIV). The experimental rig design is supported by computational fluid dynamics (CFD) analysis. Relevant past studies have yielded qualitative experimental results, but the PIV results reported here provide quantitative data to aid the understanding of the preliminary flow field inside the HV (i.e., before a heat flux is applied) and to assist CFD validation. Further, using both water and Al2O3 nanofluid as the flow medium, the PIV observed flow structures are compared to allow an assessment of the effect on heat transfer performance. Thus, these PIV measurements offer the first evidence for enhanced heat transfer in HV devices when nanofluids are used as the coolant. The improved understanding of the cooling advantage of nanofluids and their effect on the HV flow regime facilitates the design of advanced high heat flux systems of future fusion machines.

Speaker: Mr Antonis Sergis (UK)

83 FTP/P1-29: Overview on CEA Contributions to the Broader Approach Projects

France is participating to the joint Europe-Japan so-called “Broader Approach Activities” in support of ITER and DEMO activities, consisting in 3 projects: The Engineering Design and Validation of a 14 MeV neutrons irradiation facility (IFMIF/EVEDA), the building of an International Fusion Research Center (IFERC) and the ITER Satellite Tokamak Programme (STP – JT-60SA). For IFMIF/EVEDA activities, CEA is in charge of the high intensity Injector, the first part of the Superconducting RF Linac, the Cryoplant, the Accelerator Control System, the delivery of specific Beam Diagnostics and RF power tubes. This activity covers the fabrication of the systems, test of most critical components, installation and commissioning at the Japanese site. For IFERC, in 2011, CEA contracted with Bull the procurement of the “Helios” computer which reached a performance of 1.5 Petaflop/s peak, making it, with its 70000 compute cores, the largest supercomputer dedicated to a single scientific community. It is complemented by large high bandwidth storage, pre/post processing, visualization systems and services for 5 years. The work started in Sept. 2007. In Jan. 2012, first users started to work. Helios was inaugurated in March 2012. For JT-60SA tokamak, CEA is responsible for providing the Cryoplant System (CS), 9 of the 18 Toroidal Field Coils and supporting structures, the TF Coils Test Facility (CTF) and cold tests, and 5 Magnet Power Supplies (MPS). CEA is also participating to the JT-60SA Research Plan update. TF coil activities started in 2007; a manufacturing contract was placed to Alstom in July 2011. The last coil will be delivered to Japan in late 2016. Prior shipment, coils will be tested in the CTF at CEA-Saclay. The CS was optimized to smooth the transient heat loads during plasma operation. For CS and MPS contract are expected to be placed in June and Dec. 2012 respectively. This paper gives an overview of present status of these projects since 2007.

Speaker: Mr Pascal Bayetti (CEA, IRFM, F-13108 St-Paul-Lez-Durance, France)

84 FTP/P1-30: Fusion Technology Facility – Key Attributes and Interfaces to Technology and Materials

On the way to a Demonstration Fusion Power Plant (DEMO), a number of fusion technology issues will need to be resolved including the long burn or steady state DT operation, net tritium breeding ratio of >1 and the application of the Fusion Technology Facility (FTF) as a material and component testing vehicle. This paper focuses on four interface areas between physics and technology that will have significant impacts on the design of FTF. For the interface area of divertor peak heat flux, both water and helium-cooled divertor designs are projected to be able to handle a maximum heat flux of 10 MW/m^2. When extended to the FTF, both mantle and divertor radiation will be needed including an innovative snowflake or super-X divertor concept. For a robust divertor design, based on results from edge localized mode (ELM) and disruption simulation experiments, both high power ELMs and disruptions will have to be avoided; otherwise the surface material will suffer significant damage. For the interface area of uniform chamber wall surface heat flux, 1-D estimates were performed and, due to the minimum and maximum temperature limits of >350°C and <550°C for the selected RAFM steel structural material, the maximum heat flux that the design can handle is <1 MW/m^2. For the area of robust chamber wall surface material, presently W is the favored surface material for the chamber wall and divertor. Recent vertical displacement event exposures to Si-W samples in DIII-D indicated the formation of the lower melting point eutectic tungsten silicide, which forms when the surface temperature reached 1400°C. The intent was for silicon to protect the tungsten by the vapor shielding effect. For the area of low activation structural material, recent boron-doped RAFM steel results indicate the possible increase of the minimum operating temperature of RAFM steel to higher than 350°C. This could significantly narrow the operating temperature window of the RAFM steel at higher neutron fluence, leading to the need for development of ODS and nano-ferritic-alloys. These areas of plasma edge physics, chamber surface and divertor materials and advanced structural material interactions and the associated necessary research directions for both tokamak and stellerator approaches are discussed in this paper. This work was supported by the US Department of Energy under DE-FG02-09ER54513.

Speaker: Mr Clement P.C. Wong (USA)

85 FTP/P1-31: Plasma Jets for Runaway Electron Beam Suppression

Multi-MA relativistic (~10-20 MeV) runaway electrons (REs), likely to be produced during disruptions in ITER, are a major threat. Dissipation of the REs energy through collisions requires reaching the Rosenbluth critical electron density (4.2×10^22 m-3) by impurity injection, which is extremely challenging. Sufficient impurity mass, very short reaction and delivery time, deep penetration to RE beam location, and efficient ablation and assimilation into the post-current quench plasma are key parameters, whose values are very difficult to be achieved simultaneously on the required time scale for ITER safe and fast shutdown (~1.2 ms). Complex plasma jets with nanoparticles are an attractive candidate for a REs suppression technique with real-time capability because: (a) they carry a large mass as compared to common gases; (b) can be accelerated to several km/s as a plasma slug in a plasma gun; (c) the resulting plasma jet has a sufficiently high ram pressure to overcome the magnetic field pressure; (d) ablation and assimilation are much enhanced due to their very large surface-to-volume ratio; (e) expanding plasma jet facilitates achieving toroidal uniformity of electron density. FAR-TECH has been developing a coaxial plasma gun prototype with a solid state TiH_2/C_60 pulsed power injector, capable of producing a hyper-velocity (>4 km/s), high-density (>10^17 cm-3), C_60¬ nanoparticle plasma jet in ~0.5 ms, with an overall reaction-to-delivery time of ~1-2 ms. We present a comprehensive characterization of the TiH_2/C_60 cartridge, which produced ~180 mg of C_60 molecular gas by explosive sublimation of C_60 powder, and the first results for a coaxial plasma gun producing a hyper-velocity C_60 plasma jet. In the next step, the prototype system is proposed for a small scale proof-of-principle REs suppression experiment on DIII-D, which has carbon tiles, can produce and control RE beams, and has a broad range of diagnostics. As injection time is ~1 ms, the nanoparticle plasma jet can be used during the thermal quench to remove RE ‘seeds’, the current quench to stop the REs ‘avalanche’, or the RE current plateau, to dissipate the REs energy. This work was supported by the US Department of Energy under DE-FG02-08ER85196 SBIR grant.

Speaker: Mr Ioan-Niculae Bogatu (USA)

Poster Bogatu.I-FTP-P1-31-FEC2012-Poster.pdf

86 FTP/P1-32: Advances in the Electrical , Control Systems, General Analysis of the Coils Design in the Mexican Tokamak Experimental Facility

The Fusion Research Group of the Autonomous University of Nuevo Leon (UANL, Spanish acronyms) presents its advances into the electrical, control systems and its coils design developed toward the Tokamak Experimental Facility[1]. This Research and Development Project (R+D) was approved from the Mexican Education Ministry (UANL-EXB-156). The present electrical and control systems studies are mainly oriented to establish our Magnetic Confinement Facility into our University Campus, with a D-shaped tokamak design with the next main characteristics: major radius 41cm(R), minor radius 18.5cm(a), aspect ratio 2.2162(A), safety factor 1.9552(q), plasma current 277kA (Ip), toroidal field 1.3T (Bt), electronic plasma density 2 - 3x10^13 cm^-3(ne). The present study at this time is an effective electrical engineering proposal to our University involving studies over the electrical power quality provided by Federal Electricity Commission. We define our parameters in voltage, current, frequency, to implement the correct strategies of electrical supplies in order to protect our facility. The analysis was performed measuring in the five domestic circuits of the University Campus: phase imbalances (current and voltage), harmonic distortion total and individual (voltages and currents of 1-50), transient capture, presentation of the power factor, registration of electrical interruptions and reclosing, measuring and recording quality power systems, crest factors (voltage and current), accurate RMS measurements of voltage and current, presentation of phasor diagrams. Our tokamak design contains a proposal coils arrangement capable for generate 1.6T, with a coil current range (10,000–30,000 A), short circuit times from 0.3s to 1s. The entirely systems uses Cu like first analysis material. The coils are designed with 3D CAD modeling and after, we apply finite element analysis through the software COMSOL Multiphysics. Our numerical calculus programs run under our SGI Altix XE250 GNU/Linux Platform. Our computational resources can give us an absolute develop systems involved in our magnetic confinement fusion research line, focused in the stronger application of engineering, technology and science concepts, developing systems and devices into this new form of energy generation. [1] Mexican Design of a Tokamak Experimental Facility, FTP/P6-36, IAEA FEC 2010.

Speaker: Mr Max Salvador (Mexico)

87 FTP/P1-33: An Advanced Plasma-material Test Station for R&D on Materials in a Fusion Environment

A new era of fusion research has started with ITER under construction and DEMO for power demonstration on the horizon. However, the related fusion reactor material science requires further development before DEMO can be designed. One of the most crucial and most complex outstanding science issues to be solved is the plasma surface interaction (PSI) in the hostile environment of a nuclear reactor. Not only are materials exposed to unprecedented steady-state and transient power fluxes, but they are also exposed to unprecedented neutron fluxes. Both the ion fluxes and the neutron fluxes will change the properties and the micro-structure of the plasma-facing materials (PFM) significantly, even to the extent that their structural integrity is compromised. A new PMTS (Plasma Material Test Station) is proposed to address these challenges, utilizing a new high-intensity plasma source concept. This device will be well suited to test toxic, as well as irradiated material samples. The advanced plasma source is based on an RF based plasma production and heating system. The source is electrode-less, so that impurity generation in the source region that could invalidate the interpretation of PSI processes will be minimized. This is especially important for high fluence experiments, accelerated lifetime studies and reduced maintenance in a radiological managed environment. B2-Eirene simulations demonstrate that ion fluxes in excess of 10^24 m^-2s^-1 should be achievable at the target delivering power fluxes of > 30 MW/m^2. Upstream temperatures at the exit of the source system should be high enough (T_e > 30 eV) to study also radiative dissipation of heat fluxes in this device. The RF source system consists of a helicon antenna for plasma production and additional electron as well as ion heating to increase electron and ion temperature separately. A pre-prototype helicon antenna has been tested at moderate magnetic fields. A maximum electron density of n_e = 4.0 x 10^19 m^-3 has been achieved in deuterium discharges. Electron heating has been investigated in a separate experiment. Electron temperatures in excess of 10 eV have been measured in the device. The roadmap for the source development to provide the prototype plasma source for PMTS will be presented.

Speaker: Mr Juergen Rapp (USA)

88 FTP/P1-34: Overview and Status of the Linear IFMIF Prototype Accelerator

Under the framework of the Broader Approach (BA) Agreement, the Linear IFMIF Prototype Accelerator (LIPAc) has been launched with the objective to validate the low energy part (9 MeV) of the two IFMIF linacs (40 MeV, 125 mA of D+ beam in continuous wave). Starting in mid-2007, the project is managed by two Home Teams (JA-HT and EU-HT) and coordinated by the Project Team at the BA site in Rokkasho with the aim to complete the validation activity with the commissioning and operating the whole LIPAc by June 2017. This paper describes the activities underway with a view to the arrival of the first components in Rokkasho at the beginning of 2013, following prior testing in Europe.

Speaker: Mr Philippe CARA (EU)

89 FTP/P1-35: IFMIF: Overview of the Validation Activities

The International Fusion Materials Irradiation Facility (IFMIF) Engineering Design and Engineering Validation Activities (EVEDA), which started in 2007 under the framework of the Broader Approach (BA) Agreement between EU and Japan, are coming to the final stage with the exception of the Accelerator Prototype Validation subproject (running till 2017). By June 2013, the engineering design of IFMIF will be completed by delivering an Interim IFMIF Engineering Design Report (IIEDR) together with the final reporting for the majority of the validation activities. IFMIF/EVEDA consists of the following sub projects: • The engineering design of IFMIF providing all the necessary information to take decisions for its construction, operation and decommissioning. • The validation of the main challenging technologies of IFMIF through the design, construction and tests of: - an Accelerator Prototype, fully representative of the IFMIF low energy (9 MeV) accelerator (125 mA of D+ beam in continuous wave) to be completed in June 2017; - a Lithium Test Loop, integrating all elements of the IFMIF lithium target facility, already commissioned in February 2011; - the High Flux Test Module (different designs) and its internals to be irradiated in a fission reactor and tested in the helium loop HELOKA-LP. An overview of the engineering design will be reported together with the outcome of the validation activities already achieved and still expected. In particular, the present status of the Accelerator Prototype (LIPAc) Validation activities ahead of the start of the installation of sub-subsystems in Rokkasho BA Site in early 2013 will be a highlight of these validation activities.

Speaker: Mr Roland Heidinger (EU)

90 ICC/P1-01: Magnetic System for the Upgraded Spherical Tokamak Globus-M2

The necessity of toroidal magnetic field increase for further gain in plasma parameters is evident from experiments conducted on the spherical tokamaks. Modernization of the machines is planned for NSTX (US), MAST (UK) and Globus-M (Russia) and aimed at toroidal magnetic field magnification. For the upgraded spherical tokamak Globus-M2 it means toroidal magnetic field (TF) increase from the present value of 0.4 T up to 1 T as well as the plasma current rise up to 0.5 MA. The vacuum vessel stays unchanged in order to reduce project costs. Present parts of the magnetic system also will be used as far as possible. The key point of the design is the novel central stack with the inductor winded above. In the current report conception of tokamak upgrade is discussed and mechanical and thermal stress analysis results for the magnets under increased field and plasma current are presented.

Speaker: Mr Vladimir Minaev (Russian Federation)

91 ITR/1-1: Scaling of the Tokamak near Scrape-off Layer H-mode Power Width and Implications for ITER

The presence of a steep edge pedestal gradient in H-mode divertor plasmas implies that strong gradients should also exist across the separatrix, forcing most of the PSOL~ 100 MW of power arriving in the SOL at QDT = 10 in ITER to flow inside a narrow channel on open field lines connecting to the divertor target plates. Recent results (coordinated in part through ITPA DivSOL group) indicate that the ITER assumed value is too large. Scaling from the new database provides a clear dependence on the poloidal magnetic field, little variation with other key variables found in previous scalings and suggests λSOL ≅ 1 mm for ITER. Measurements from DIII-D, C-Mod and NSTX indicate a systematic narrowing of the inter-ELM divertor heat flux width with plasma current in H-mode plasmas. For the near SOL power width, the data indicate λSOL ∝ (q95/Btor)~1 ~ a/Ip ∝ 1/Bpol, with little or no dependence on PSOL or R (major radius). Analysis of data obtained in the same way from JET and ASDEX Upgrade yields λSOL (mm) =0.73·Bt^(−0.78)·qcyl^1.2·PSOL^0.1·R^0, again with no dependence on R. Data are consistent in absolute magnitude with a recent heuristic drift-based theory. These new findings are based on IR analysis of strongly attached H-mode discharges. Key improvements here have been the avoidance of ELM effects, accounting for changes in the deposition profile due to heat diffusion across the divertor legs into the private flux region. Experimentally, essentially the full operational range of plasma current and toroidal field in each device was scanned. The value of λSOL ≅1 mm obtained jointly from these scalings for ITER at 15 MA in the Baseline Inductive Scenario, is about a factor 3 shorter than the lowest values predicted on the basis of earlier studies. Such narrow power channels are a concern for ITER, though preliminary SOLPS simulations indicate that they could be tolerated, since volumetric power dissipation in the divertor can still be sufficient to maintain heat flux densities at acceptable levels provided the outer divertor leg is partially detached. Simple estimates show that for ITER, if λSOL ≤ 2 mm the implied upstream pedestal pressure gradient would exceed ideal ballooning stability by some margin, assuming that the SOL pressure width is a measure of that in the pedestal. This issue will be examined in the context of the current database.

Speaker: Mr Thomas Eich (Max-Planck-Institut f. Plasmaphysik)

92 ITR/1-2: Progress on the Application of ELM Control Schemes to ITER

High fusion performance DT operation in ITER is based on the achievement of the H-mode confinement regime with H98 ≥ 1 and an edge transport barrier that is expected to lead to the quasi-periodic triggering of ELMs. Operation of ITER with H-mode plasmas is also foreseen during the non-active (H & He) and DD operation allowing the development of ELM control schemes before DT operation. The non-linear MHD evolution of the plasma during ELMs in ITER has been modelled with the JOREK code which shows that the non-linear MHD growth during the ELM causes a temporary ergodisation of the plasma edge leading to the appearance of striations in the ELM power flux at the divertor target. On the contrary, power fluxes to the first wall are expected to be dominated by the convection of energy by the radial propagation of plasma filaments produced during the ELM crash. JOREK has also being applied to investigate the capabilities of the ITER pellet injection system to meet the requirements for ELM control following its validation with DIII-D experimental results. The application in-vessel coils to create edge magnetic field perturbations for ELM control and the associated power/particle fluxes to PFCs have been studied for ITER. Evaluations of the edge magnetic field perturbation by in-vessel coils show that the toroidal symmetry of the applied currents in the coils (n=3 or n=4 symmetry) can have a significant impact on the level of current required (up to 50% current level reduction) to achieve a given level of edge ergodisation. Optimization of the relative toroidal phasing of the currents applied to the 3 rows of coils shows that there is an appropriate margin (factor of ~ 1.5 - 2) in coil current magnitude required to achieve the design criterion (vacuum approximation) for the 15 MA QDT = 10 scenario. Power and particle fluxes in the perturbed edge magnetic field have been evaluated in the vacuum field approximation and including plasma response. The application of edge magnetic field perturbations leads to the appearance of non-toroidally symmetric divertor power/particle fluxes extending up to ~ 50 cm from the separatrix but also to a reduction of the peak heat flux by a factor of 2-3. The inclusion of plasma response decreases transport of energy/particles from the main plasma and the detrimental effects on plasma energy/particle confinement.

Speaker: Mr Alberto Loarte (ITER)

93 ITR/1-3: Design of the MITICA Neutral Beam Injector: From Physics Analysis to Engineering Design

For ITER heating and current drive, two neutral beam injectors (NBIs) are planned, delivering a total of 33 MW in stationary conditions up to one hour; each injector will accelerate a 40 A negative deuterium ion current up to 1 MV. Such requirements have never been achieved simultaneously. Hence the PRIMA (Padova Research on ITER Megavolt Accelerator) facility is under construction at Consorzio RFX in Padua, Italy. PRIMA will include a test bed named MITICA (Megavolt ITer Injector and Concept Advancement), with the aim of meeting the ITER beam requirements in terms of negative ion yield, beam uniformity, high voltage holding, operation of beam line components and power supplies, overall reliability of the NBI. The present contribution describes the current status of numerical simulations, devoted to the optimisation of MITICA, providing the main inputs for the design of accelerator, beam line components, diagnostics and power supplies. Physics and engineering aspects include: beam optics, dumping of co-extracted and stripped electrons, thermo-mechanical behaviour of grids and beam line components during long pulse operation, voltage holding capabilities. The optimised geometry of the accelerator is characterised by equal acceleration gaps (increased voltage holding capability) and a combination of horizontal and vertical magnetic fields in the accelerator (reducing heat loads and electrons exiting the accelerator); the gas pressure profile is also simulated in the accelerator and in the injector. The design of the accelerator power supplies has been supported by simulations of static and dynamic performances, including the investigation of overvoltages by a sophisticated fast transient model and the modelling of matching network and RF systems. Moreover the signals expected from the diagnostic systems have been simulated, with realistic beam features, providing prescriptions for the design of diagnostics, like beam emission spectroscopy, beam tomography and neutron diagnostic. Most of the design of MITICA plants and components are well developed and close to finalisation.

Speaker: Mr Piergiorgio Sonato (Italy)

94 ITR/1-4Ra: Development in Russia of Megawatt Power Gyrotrons for Fusion

During last years several new gyrotrons were designed and tested in Russia. Main efforts were spent for development 170GHz/1MW/50%/CW gyrotron for ITER and multifrequency gyrotrons. Additionally other new gyrotrons were shipped and installed at running plasma installations. The industrial production prototypes of the ITER gyrotron were tested at power 1.0 MW in 400…500 second pulses and 0.8-0.9 MW in 1000 second pulses. For 1 MW power regime the gyrotron efficiency is 55%. The last gyrotron versions operate in LHe-free magnet. It is important that two last gyrotrons (V-10 and V-11) demonstrate very similar output parameters. Time traces for the main gyrotron parameters are stable and confirm possibility of the gyrotron operation even in longer pulses. Detail analysis of the test results showed that a slightly modified ITER gyrotron prototype is capable to operate at power 1.2 MW. First tests of the modified tube are rather encouraging: microwave power 1.2 MW at MOU output was demonstrated in 100 second pulses with efficiency of 53%. Additionally two gyrotron models with TE28.12 operating mode were tested in short-pulse experiments. The use of step-tunable gyrotrons can greatly enhance performance of ECRH/ECCD systems due to larger accessible radial range, possible replacement of steerable antennas, higher CD efficiency for NTM stabilization. The main problems in development of multifrequency gyrotrons are to provide: efficient gyrotron operation at different modes, efficient conversion of the modes into a Gaussian beam, reliable operation of broadband or tuneable window. Considering this three key problems one can say that first two of them are solved, but realization of a CVD diamond window for a megawatt power level multi-frequency gyrotron met real difficulties. Now a new tunable window concept is under consideration.

Speaker: Mr Alexander Litvak (Russian Federation)

Slides Litvak-ITR-1_4a_V3.ppt

95 ITR/P1-01: Commissioning and First Results of the ITER-Relevant Negative Ion Beam Test Facility ELISE

For heating and current drive the ITER NBI system requires a negative hydrogen ion source capable of delivering above 40 A of D¯ ions for up to one hour pulses with an accelerated current density of 200 A/m^2. In order to limit the power loads and ion losses in the accelerator, the source must be operated at a pressure of 0.3 Pa at maximum and the amount of co-extracted electrons must not exceed the amount of extracted negative ions. As presently these parameters have not yet been achieved simultaneously, also due to a lack of adequate test facilities, the European ITER domestic agency F4E has defined an R&D roadmap for the construction of the neutral beam heating systems. An important step herein is the new test facility ELISE (Extraction from a Large Ion Source Experiment) for a large-scale extraction from a half-size ITER RF source which was constructed in the last 2 years at IPP Garching. The early experience of the operation of such a large RF driven source (1x1 m^2 with an extraction area of 0.1 m^2) will give an important input for the design of the Neutral Beam Test Facility PRIMA in Padova and the ITER NBI systems and for their commissioning and operating phases. PRIMA consists of the 1 MeV full power test facility MITICA, operational 2017, and the 100 kV ion source test facility SPIDER, operational 2015. The aim of the design of the ELISE source and extraction system was to be as close as possible to the ITER design; it has however some modifications allowing a better diagnostic access as well as more flexibility for exploring open questions. The extraction system is designed for the acceleration of 20 A of negative hydrogen ions of up to 60 kV. Plasma operation of up to one hour is foreseen; but due to the limits of the IPP HV system, pulsed extraction only is possible. ELISE went into operation in spring 2012 with first plasma and beam pulses. The paper discusses critical issues of the manufacturing and describes the commissioning phases of the different subsystems with a special emphasis on the HV conditioning of the large grids. First results of the dependence of the plasma homogeneity on the magnetic filter field, measured by optical emission spectroscopy, are shown and compared with beam homogeneity measurements by calorimetry and beam emission spectroscopy.

Speaker: Mr Peter Franzen (Germany)

Poster Poster-IAEA-FEC2012-Franzen.pdf

96 ITR/P1-02: Nuclear Analyses For ITER NB System

Detailed nuclear analyses for the latest ITER NB system are required to ensure that NB design conforms to the nuclear regulations and licensing. A variety of nuclear analyses was conducted for the NB system including a tokamak building and outside the building by using Monte Carlo code MCNP5.14, activation code ACT-4 and Fusion Evaluated Nuclear Data Library FENDL-2.1. A special “Direct 1-step Monte Carlo” method is adopted for the shutdown dose rate calculation. The NB system and the tokamak building are very complicated, and it is practically impossible to make geometry input data manually. We used the automatic converter code GEOMIT from CAD data to MCNP geometry input data. GEOMIT was improved for these analyses, and the conversion performance was drastically enhanced. Void cells in MCNP input data were generated by subtracting solid cells data from simple rectangular void cells. The CAD data were successfully converted to MCNP geometry input data, and void data were also adequately produced with GEOMIT. The effective dose rates at external zones (non-controlled areas) should be less than 80 microSv/month according to French regulations. Shielding structures are under analysis to reduce the radiation streaming through the openings. We are confirming that the criterion is satisfied for the NB system. The effective dose rate data in the NB cell after shutdown are necessary to check the dose rate during possible rad-works for maintenance. Dose rates for workers must be maintained as low as reasonably achievable, and at locations where hands-on maintenance is performed should be below a target of 100 microSv/h at 12 days after shutdown. We are specifying the adequate zoning and area where hands-on maintenance can be allowed, based on the analysis results. The cask design for transport activated NB components is an important issue, and we are calculating the effective dose rates. The target of the effective dose rate from the activated NB components is less than 25 microSv/h at 30 cm from the outer surface of the iron cask.

Speaker: Mr Satoshi SATO (Japan)

97 ITR/P1-03: Status of the Negative Ion Based Diagnostic Neutral Beam for ITER

In ITER a dedicated 100keV Diagnostic Neutral Beam (DNB) based on negative ion technology will be injecting 18-20A of hydrogen to provide helium ash measurements via Charge Exchange Recombination Spectroscopy (CXRS). The CXRS diagnostics will also provide measurements of ion temperatures and other essential plasma parameters and the DNB will also be used for Beam Emission Measurements (BES). Recently on-axis Motional Stark Effect (MSE) has been proposed as an additional complementary diagnostics. Reliable DNB performance predictions are essential input parameters for the design of the diagnostic systems which will pass the conceptual design stage this year. They also assure the correct characterization of all ITER interfaces to allow safe operation of the ITER plant. The expected performance parameters of the system will be critically re-assessed taking into account the recently increased neutral edge density predictions in the plasma chamber. The DNB duty cycle will be discussed in detail, indicating all operational limitations identified in the design phase. An overview of the current status of the DNB system and the interface constraints will be given. The consolidated design, including the recently updated magnetic field reduction system, will be presented and the most challenging design features highlighted. While the performance of the RF ion source – identical to the one used in the Heating Neutral Beams but operated at much more challenging current densities - will be demonstrated in the dedicated Ion Source Test Bed SPIDER in Padua, the full DNB Beamline will be demonstrated in the Indian Test Facility (INTF). The INTF will specifically allow far-field measurements to confirm the beam divergence and allow reliable predictions of the DNB power delivered to the plasma. This parameter is essential for obtaining dependable signal to noise estimates for the design of the diagnostic systems. Procurement of the DNB Beamsource, part of the DNB power supplies and the proto-type DNB beamline components will start in 2012, while the INTF building premises are already available. The availability of the DNB as a diagnostic tool in the hydrogen/helium phase will be discussed shortly looking both at the shine through limitations and the commissioning planning to identify the earliest date the DNB will be available for CXRS measurements.

Speaker: Ms Beatrix Schunke (ITER)

98 ITR/P1-04: EU Development of the ITER Neutral Beam Injector and Test Facility

The activities towards the establishment of the NB Test Facility (NBTF) in Padova Italy and those related to the procurement of the heating neutral beams for ITER have recently reached a good level of progress thanks to the finalization of the agreements on the NBTF between F4E (the EU Domestic Agency for ITER), Consorzio RFX (the host of the NB test facility) and the ITER organization. This paper presents the status of the design of the various components within the EU scope of procurement, with a focus on the modifications implemented in the last years as a result of intense R&D activity undertaken in EU.

Speaker: Mr Antonio Masiello (EU)

99 ITR/P1-05: Development of ITER Equatorial EC Launcher

The present day EC launcher typically injects a few MW power and the pulse length is 10~20s. On the other hand, the ITER equatorial EC launcher is making an advanced technology to injecting ≥20M and CW operation. The ITER equatorial EC launcher consists of an unique blanket shield structure and a port plug installing millimeter (mm) wave components, neutron shields, cooling water lines and so on. The design of the blanket shield structure that tolerates thermal and electromagnetic load is attained. The mm-wave design that enables to guide the wave power of 20MW into plasma with toroidal steering capability of 20º~40º and efficiency of 98.4~99% assuming HE11 fundamental wave mode + TEM000 gaussian mode are described. Reduction of the heat load to 2.1MW/m2 on the steering mirror and the optimization of beam radius at plasma, 16~22cm that satisfies the requirement, are attained. The mock-up of the mm-wave launching system and the subcomponents are fabricated to investigate the design availability. High power (0.5MW) experiment of the mock-up confirmed the expected wave beam propagation and steering capability.

Speaker: Mr Koji Takahashi (Japan)

Poster Poster_ITR-P1-05.pdf

100 ITR/P1-06: Optimization of the EC Heating and Current Drive Capabilities

A 24MW CW Electron Cyclotron Heating and Current Drive (EC H&CD) system operating at 170GHz is to be installed for the ITER tokamak. The ITER EC system will represent a large step forward in the use of microwave systems for plasma heating and current drive applications. Present day systems are operating in relatively short pulses (≤10s) and installed power levels of ≤4.5MW, while the ITER EC system parameters are CW operation and 20MW injected power. The technical challenge facing the development and installation of the EC system is further complicated due to the harsh ITER in-vessel environment and complicated procurement strategy. The ITER EC international community has confronted these challenges, aiming at integrating the modifications proposed from the 2007 ITER design review and further enhancing the EC system capabilities. These changes have not only simplified the technical design, but have also simplified the procurement interfaces and increased the functional capabilities for plasma heating and current drive applications. The functional improvements include increasing the access of the EC power from ~50% to nearly ~90% of the plasma cross section. In particular the UL has been modified to allow power deposition over the range of ~0.3≤ρT≤~0.9 compared to previous access of ~0.55≤ρT≤~0.85 (where ρT is the square root of the normalized toroidal flux). This allows the UL to be applicable for a broader access for stabilization of neoclassical tearing modes (NTMs) and sawtooth instability. The EC heating is functionally limited in magnetic field region depending on the resonant harmonic. The heating access in ITER was assumed to be applicable over ~33% of the range from 2.3T to 5.3T, regions of fundamental and second harmonics. Recent analysis associated with the EL has demonstrated that the EC system is applicable over a much broader range: ~75% for central heating (ρT≤~0.5) and ~90% for L to H-mode assist (ρT≤~0.85). Further improvements to the EC system are now being considered, which include adapting the PS to accommodate future higher power gyrotrons (≥1.2MW) and modifying the sweeping direction of the EL from a toroidal to poloidal steering more than doubling the driven current in the region of 0.4≤ρT≤0.6. These and other improvements under study will be reviewed in this paper.

Speaker: Mr Mark Henderson (ITER)

101 ITR/P1-07: Validation of the RF Properties and Control of the ITER ICRF Antenna

One ITER ICRF antenna consists of a close-packed array of 24 straps arranged in a 6 poloidal by 4 toroidal array. Three poloidally adjacent straps (a “triplet” of straps) are fed in parallel from one single feeding line through a 4-port junction. A shunt service stub is inserted on the feeding line inside the antenna. It has been optimized to provide a broad-band RF frequency response of the array. Load tolerance is achieved by feeding each pair of two poloidal triplets through a 3dB hybrid coupler. The array has to radiate 20MW of RF power (with a 45kV limit on the system) in quasi-CW operation for frequencies ranging from 40MHz to 55MHz and different toroidal phasings to provide a wave spectrum appropriate for both heating and current drive. Two identical ICRF antennas are foreseen on ITER. In order to gain confidence in the design options the RF properties of the ITER ICRH antenna array are experimentally validated on reduced-scale mock-ups of part or of the whole antenna array. Experimental measurements allow to check the optimization of the antenna front-end geometry, to confirm the RF frequency response of the antenna array and hence the expected performance of the antenna for the different proposed phasings, to verify the impact of critical parameters as the vertical septum recess on the coupling properties of the antenna as well as on the mutual coupling terms and finally to test and validate the proposed grounding options to the vessel. An overview of the experimental measurement results on the different mock-ups is given. Comparisons with simulations performed by various codes (Topica, CST MWS and ANTITER II) are given together with the expected performance on plasma deduced from measurements with dielectric load. The effect of limitations (voltage, current) on the maximum total radiated power is discussed. The importance of a good decoupling network and of grounding is emphasized. Finally the control of the antenna wave spectrum and of the matching is performed by implementing a feedback control of the matching-decoupling system close to the one foreseen for ITER on a low-power reduced-scale mock-up of the whole ITER ICRH antenna array. It uses 23 feedback loops simultaneously operated. Control algorithms are developed and tested together with the stability of the system. This work was performed under the F4E grant F4E-2009-GRT-026.

Speaker: Mr Pierre DUMORTIER (Belgium)

Poster IAEA2012_PDR_Poster_lr.jpg

102 ITR/P1-08: RF Optimization of the Port Plug Layout and Performance Assessment of the ITER ICRF Antenna

ITER ICRF antenna’s capability to couple power to plasma is determined by the plasma Scrape-Off Layer (SOL) profiles, shaping of the front strap array, organized as a 6 poloidal by 4 toroidal array of short straps, overall layout of the feed network and detailed design of its RF components. The first two factors are taken into account in the TOPICA [1] calculated strap array 24x24 S/Z-matrices. This data is coupled to a RF circuit model of the strap feeding circuit of which the components are S/Z matrices calculated with CST Microwave Studio (MWS) [2], such as the 4 Port Junction (4PJ) which combines 3 poloidaly adjacent straps so as to obtain a 2 poloidal by 4 toroidal array of triplets, or simple Transmission Line (TL) sections for the rest of the Removable Vacuum Transmission Line (RVTL) which include the Service Stub tee (SST) and Vacuum Ceramic Windows (VCW). The RF feeding circuit inside the port plug is optimized in order to maximize the power coupled to the plasma for the various phasings considered for operation. This takes into account geometrical constraints, assembly requirements and RF quantities (E-field less than 2kV/mm along the magnetic field in the torus vacuum areas and less than 3 kV/mm perpendicular to the magnetic field and private vacuum areas, voltages less than 45 kV and currents through RF contacts less than 2 kA). The ITER-Like Antenna on JET results obtained in 2008-9 [3] support the proposed design by having validated the TOPICA coupling estimations as well as demonstrated that there were no unforeseen difficulties in operating at 42 kV and power densities in the range required by ITER in terms of reliability and possible ICRF impurity production. The effect of the RF grounding of the Front Housing Module, carrying 2 toroidaly adjacent triplets, as well as the grounding of the whole antenna port plug to the vessel is analyzed as well in conjunction with the Blanket Shielding Modules surrounding the antenna. [1] “TOPICA: an Accurate and Efficient Numerical Tool for Analysis and Design of ICRF Antennas”, V. Lancellott et al., Nuclear Fusion, Vol. 46, pp. S476-S499, 2006. [2] CST GmbH, CST Microwave Studio®, User Manual (2011) [3] “Latest Achievements of the JET ICRF Systems in View of ITER”, F. Durodié et al., 23rd IAEA FEC, Daejeon, Republic of Korea, 2010.

Speaker: Mr Frederic Durodie (Belgium)

103 ITR/P1-09: On the Use of Lower Hybrid Waves at ITER Relevant Density

Collisional Absorption (CA), Parametric Instabilities (PI) and Scattering from Density Fluctuations (SDF) can preclude the penetration of Lower Hybrid (LH) waves, dissipating the power in the plasma periphery. A multi-machine assessment started at the end of 2009 under the coordination of ITPA-IOS group. It aims at understanding the complex physics underlying the phenomenon and increasing confidence in LH modeling for ITER’s advanced scenarios. Results from C Mod, FTU, HT-7, JET and Tore Supra will be reported, while results from EAST are expected in the near future. The reported experiments have relevant plasma and waves parameters that encompass ITER’s. Limiter and divertor operations are included, as well as operation with metallic and carbon walls, with plasma edge conditioned by wall boronisation and, on FTU, by using liquid lithium. Experiments on FTU show that LH waves at 8 GHz does penetrate bulk plasmas, with density profiles even higher than those expected in ITER, only if operating with the increased edge electron temperature provided by operating with lithized walls. The reduced broadening of the launched spectrum, as detected by loop antennas, suggests a PI influence. C-Mod detects an improved penetration of LH waves at 4.6 GHz, at relatively high density, when edge electron temperature is higher and absorption condition is closer to single-pass. This trend is qualitatively reproduced in both ray tracing and full wave model as reduced effect of CA due to increased single pass absorption. A gradual shift of 3.7 GHz LH power deposition to the periphery, is observed in JET with increasing density. Reduced density and increased pedestal temperature lead to broader but still very peripheral absorption. Decrease of LH effects, at 3.7 GHz in Tore Supra, correlates with increasing edge fluctuations, suggesting a role of SDF. LH experiments at the lowest frequency of 2.45 GHz on HT-7 did not find anomalies, at moderate density, in discharges with Li and B coatings. From the available information a preliminary conclusion on encouraging trend can be inferred for ITER. There will be a high edge temperature that helps avoid PI and SDF while LH waves will clearly be in single pass regime, minimising CA. A strong modelling effort is nevertheless progressing to reproduce the experimental evidence with physics models to strengthen the prediction for ITER.

Speaker: Mr Angelo A. Tuccillo (Italy)

104 ITR/P1-10: Self-consistent Simulation of Plasma Scenarios for ITER Using a Combination of 1.5D Transport Codes and Free Boundary Equilibrium Codes

Self-consistent transport simulation of ITER scenarios is an important tool for the exploration of the operational space and for scenario optimisation. It also provides an assessment of the compatibility of developed scenarios (which include fast transient events) with machine constraints, in particular with the poloidal field (PF) coil system, heating and current drive (H&CD), fuelling and particle and energy exhaust systems. Credible prediction of the plasma and plasmas systems behaviour can only be achieved when the best combination of high quality transport codes, using the most advanced theory-based transport models, are combined with state of the art free boundary equilibrium codes. This paper summarises results of predictive modelling of all reference ITER scenarios with two EU suites of transport and free boundary codes. Modelling of 15MA baseline DT scenario with Q=10 and its variants was mostly based on GLF23 for the H-mode part of scenario, combined with the explicit modelling of edge barrier and type-I ELMs. The L-mode phase was simulated with Bohm/gyroBohm model. One of the novel elements was predictive modelling of fast transient phenomena, such as L-H and H-L transitions as well as predictive modelling of D and T densities and He ash accumulation. Self-consistent simulations of fast transients revealed potential difficulty for ITER PF position control system to maintain the plasma-inner wall distance during fast uncontrolled H-L transition, due to voltage saturation in the CS. Since Hybrid and Steady State (SS) scenarios have less established theoretical and experimental basis, their predictive simulation rely more on ad-hoc assumptions about heat and particle transport inside the edge barrier. It was assumed that hybrid scenario has heat transport which ensures energy confinement time with H98y=1.3 during the flat top burn. The main emphasis of the simulation was on the selection of the heating and current drive scheme to ensure that qmin stays above 1 for at least 1000s. Also, the ability of ITER PF system to sustain fast transient phenomena (such as sudden loss of the internal transport barrier in SS scenario) as well as the control of MHD stability was studied.

Speaker: Mr Vassili Parail (UK)

105 ITR/P1-11: Demonstrating the ITER Baseline Operation at q95=3

ITER requires robust operation of various plasma scenarios within the hardware constraints of the device. Operation in H-mode at 15MA and q95=3 is planned to achieve Q=10 in deuterium-tritium mixtures. The Integrated Operation Scenario Topical Group of the ITPA has coordinated experiments in C-Mod, ASDEX Upgrade, DIII-D and JET to obtain optimum data assessing H-mode scenarios at q95~3. Previous results for the plasma formation at low loop voltage and the ramp-up phase were reported. Recent joint studies on the flat top and ramp down phase show that entering H-mode is generally observed at Ptot/PL-H~1. Regular ELMing H-modes achieving H98~1 require Ptot/PL-H=1.3-2 at JET, for Ip up to 4.5MA, in DIII-D regular ELMing H-modes are only achieved at higher plasma beta (beta_N=2.0), while in C-Mod Ptot/PL-H~1 is only achieved in stationary H-modes at high input power and higher radiation fraction using seeding. JET data show no significant difference in plasma performance or temperature and density profile shapes when using ion cyclotron heating compared to neutral beam heating, despite the rotation profile changing dramatically when the co-neutral beam is reduced to zero. For H-modes at high plasma current, some experiments only reach ne/nGW~0.65-0.7 using gas fuelling. At DIII-D the stability of long pulse operation at q95=3, shows susceptibility to n=1 tearing modes. The current ramp down requires H-mode combined with a reduction in plasma elongation to control the plasma inductance excursion without additional flux consumption, for ohmic or L-mode discharges a stronger reduction in elongation would be required. The experimental data presented provide important input for benchmarking integrated code simulations using sophisticated models for transport and heating or current drive systems; moreover they give additional confidence and insight into the possible operation domain of the baseline plasma scenario in ITER.

Speaker: Mr Adrianus Sips (EU)

106 ITR/P1-12: Modelling of ITER Plasma Shutdown with Runaway Mitigation Using TSC

Fast shutdown of an ITER plasma discharge without generating large runaway current has been an area of active research over the past several years. In ITER, during the thermal quench preceding the current disruption, the toroidal electric field can resistively grow to values about 50 times the critical electric field for runaway current generation, which can give rise to avalanche generation of runaway electrons with energies up to 15MeV and currents of unprecedented magnitude of more than 10MA, causing potential damage to the ITER first wall. Such a scenario can be avoided by rapidly increasing the plasma electron density by about 50-60 times, close to Rosenbluth density values, by which the toroidal electric field remains lower than the critical electric field. In this case, the runaway electrons are slowed down through collisional processes, suppressing the runaway current. In this paper we have carried out simulations for ITER plasma shutdown using both impurity doped deuterium pellets using the TSC code. The pellet model of TSC is also extended to approximately model massive gas injection into ITER plasmas. Using pellet initial radius of 5mm and initial pellet velocity of 500m/s and repetition frequency of around 100 Hz launched from the outboard mid-plane, we have explored possibilities of safe plasma shutdown without large runaway current generation. The MGI is modeled in TSC by a train of 3-5 large pellets of pure neon with total number of atoms equal to that in a gas jet cartridge is injected with velocity of 500m/s. The major difference between a pellet and a gas jet in the low ablation rate of the pellets, especially at low target Te, is overcome by artificially increasing the ablation rate to such an extent that the entire pellet atoms are ablated by the time it crosses the q=2 surface. We are carrying out detailed parametric study of the ITER plasma shutdown in the wide parameter space of pellet impurity content, number of injected pellets, radius of the pellets etc. The relative merits and issues with the different options will be discussed.

Speaker: Mr Indranil Bandyopadhyay (India)

107 ITR/P1-13: CORSICA Modelling of ITER Hybrid Mode Operation Scenarios

The hybrid mode operation observed in several tokamaks is characterized by further enhancement over the high plasma confinement (H-mode) associated with reduced MHD instabilities linked to a stationary flat safety factor (q) profile in the core region. The proposed ITER hybrid mode is currently aiming at operating for a long burn duration (> 1000s) with a moderate fusion power multiplication factor, Q, of at least 5. This paper presents candidate ITER hybrid mode operation scenarios developed using a free-boundary transport modelling code, CORSICA, taking relevant physics and engineering constraints into account. First, we have developed a 12.5MA ITER hybrid mode operation scenario by tailoring the 15MA ITER inductive H-mode scenario. Second, we have studied accessible operation conditions and achievable range of plasma parameters. ITER operation capability for avoiding the poloidal field (PF) coil current, field and force limits are examined by applying different current ramp rates, flat-top plasma currents and densities, and pre-magnetization of the PF coils. Various combinations of heating and current drive (H&CD) schemes have been applied to investigate several physics issues, such as the plasma current density profile tailoring, enhancement of the plasma energy confinement and fusion power generation. A parameterized edge pedestal model based on EPED1 was recently added to the CORSICA code and applied to hybrid scenarios. Finally, fully self-consistent free-boundary transport simulations have been performed to provide information on the PF coil voltage demands and to study the controllability with the ITER controllers.

Speaker: Mr Sun Hee KIM (ITER)

108 ITR/P1-14: Disruption Impacts and Their Mitigation Target Values

Major disruptions (MD) and vertical displacement events (VDE) in ITER will be the cause of a variety of deleterious impacts due to the high stored thermal and magnetic energy. Extrapolation from the disruption database obtained on current tokamaks and the results of numerical simulations, demonstrate that the thermal loads which will be produced during the thermal quench (TQ) of MD and VDEs in ITER, will lead to large scale macroscopic melting of metallic plasma-facing components (PFC). During the current quench (CQ) phase, runaway electrons (RE) are expected to deposit severe thermal loads on PFCs. The electromagnetic (EM) loads due to halo and eddy currents during the CQ are a critically important factor in the mechanical design of the vacuum vessel (VV) and in-vessel components (IVC). For reliable operation and machine protection in ITER, mitigation of these heat and EM loads and REs during MDs and VDEs is mandatory. It is essential to realize high performance in the following three elements simultaneously for reliable mitigation, (i) high mitigation performance factor (high reduction of impact), (ii) high prediction performance (high success rate simultaneously with low false rate) and (iii) low disruptivity (by passive and active disruption avoidance). Proper target values of each element are of primary importance to achieve the overall mitigation performance required in ITER and to promote the physics R&D for the development of mitigation, prediction and avoidance systems and algorithms. In this paper, we will quantify proper target values for each element of mitigation based on the assessment of the impacts by heat and EM loads and REs using physics databases and modeling for the present ITER design. Key components like VV, IVC and PFC will be carefully examined to quantify these requirements.

Speaker: Mr Masayoshi Sugihara (ITER)

109 ITR/P1-15: Development of ITER Scenarios for Pre-DT Operations

In preparation for the full deuterium-tritium (DT) operation in ITER, a significant period of experimentation will be dedicated to plasma operations that generate no or minimal activation products. This operation would utilize plasmas with helium (He) or hydrogen (H) gas species since these generate no fusion reactions producing tritium or neutrons that result in materials activation. Present planning also includes consideration for deuterium (D) operation with possibly short pulse duration to limit the production of tritium and activation from DD fusion reaction neutrons. Low-activation operation is needed to qualify all major mechanical and electrical subsystems before reliance on remote handling capability. Access to and sustainment of the high-confinement mode (H-mode) must be demonstrated with sufficient control of edge localized modes to limit the power flow to the divertor. Results from time-dependent 2-dimensional equilibrium with 1-dimensional transport predictive simulations explore possible operating scenarios. Simulations in current flat top evaluate steady performance to determine the parameter operating space at full plasma current IP=15MA and magnetic field BT=5.3T and for reduced performance at 7.5MA and 2.65T. In addition to baseline neutral beam injection and electron cyclotron heating, techniques for application of ion cyclotron heating under steady conditions are presented. Full duration time-dependent simulations with start-up limited on the inside wall, IP ramp up to full current, flat top burn, and IP ramp down to develop controllable operating scenarios in H and D are presented. With present understanding of H-mode threshold scaling, the proposed auxiliary heating power level of 63MW (nominal) should allow access to H-mode operation in helium at 7.5 MA/ 2.65 T. Access to H-mode in pure H-plasmas would be, at best, marginal. Time-dependent free-boundary equilibrium simulations using controllers for the baseline operation indicate there is sufficient capability in the coil system to produce and control short pulse flat top plasmas suitable to validate physics and engineering systems before DT operation. These simulations are completed under a joint modeling effort in the Integrated Operations Scenarios (IOS) group of the International Tokamak Physics Activity (ITPA).

Speaker: Mr Thomas Casper (ITER Organization)

110 ITR/P1-16: ITER Plasma Position Control System and Scenario Optimization

The ITER machine is reaching a stage in which the design is in large part frozen. Nevertheless design changes are necessary in the procurement phase due to additional constraint linked to manufacturing techniques and/or cost containment. In this framework, the reference ITER scenario and the control system strategy are in continuous evolution. The aim is preserving the final goal of a 15 MA Q=10 burning plasma in ITER, which will require a careful optimization of the scenario in order to fully exploit the machine capabilities within the engineering limits which define and restrict the operational space available. This paper presents a summary of the activities carried out within the EU-DA on the engineering optimization of the ITER plasma scenarios and of the magnetic plasma position control system strategy.

Speaker: Mr Mario Cavinato (EU)

111 ITR/P1-17: Integrated Modelling of ITER Hybrid Scenarios Including Momentum Transport, NTMs, and ELMs in Preparation for Active Control

Hybrid scenario is an operational regime designed to achieve a long pulse operation with a combination of inductive and non-inductive current drive. It was suggested for the operation of ITER to allow high fusion power in long pulse operations over 1000 s at a plasma current lower than the inductive reference scenario. Engineering tests of reactor-relevant components, such as breeding blankets are planned to perform in this scenario . Here, we report integrated simulation results of ITER hybrid scenario including momentum transport, neoclassical tearing mode (NTM), and edge localised mode (ELM). In this work, ASTRA is used for integrated simulations of plasma equilibrium, transport, heating and current drive, and magnetic island evolution, self-consistently. Firstly, the effect of toroidal rotation to confinement is addressed by solving the momentum transport equation including inward pinch, turbulent transport and residual stress. Secondly, the ELM activities are simulated and the pedestal height of ITER hybrid scenario is predicted. Lastly, the NTM activities are simulated and the capability of the ECH upper launcher is evaluated. The methodology of simulations presented can be applied to design feedback controllers for ELMs and NTMs in ITER.

Speaker: Mr Yong-Su Na (Republic of Korea)

112 ITR/P1-18: Challenges in Burning Plasma Physics: the ITER Research Plan

Following First Plasma, currently scheduled for late 2020, the ITER project aims to develop the capability for DT operation as rapidly as possible in order to address the key mission goal of demonstrating long pulse operation at Q ≥ 10 with approximately 500 MW of fusion power. The ITER Research Plan (IRP) has been developed to analyze the experimental programme necessary to develop ITER’s operational capability from First Plasma to the achievement of the Q ≥ 10 mission goal. It integrates the experimental activities required to develop a robust capability for high current (15 MA) H-mode operation using DT fuel and, incorporating the planned schedule for the installation and commissioning of ITER auxiliary and plant systems, develops a schedule to allow full DT operation in late 2027 and the exploration of high fusion gain DT plasmas in 2028. The experimental programme is foreseen to develop through 3 phases: H/ He (non-active), D and DT (nuclear). During the first phase, all systems necessary for operation at full technical performance (15 MA/ 5.3 T) will be commissioned and integrated into plasma operation to establish the plasma operating regimes and the plasma control capability required to provide a robust basis for the transition to DT Operation. A second, relatively short, phase in deuterium completes the plasma commissioning activities, allows H-mode operation to be extended to high current and DT relevant parameters, and initiates the transition to full DT operation via a series of “trace tritium” experiments. The experimental programme on high fusion power DT scenarios to be explored in the third phase of operations must address several challenges in burning plasma physics to achieve and sustain the necessary level of fusion performance to satisfy the Q ≥ 10 mission goal. The paper will discuss the key physics issues to be resolved, the elements of the experimental programme foreseen to address them and the opportunities for burning plasma research which the experimental programme will provide.

Speaker: Mr David Campbell (ITER)

113 ITR/P1-19: Tokamak Experiments to Study the Parametric Dependences of Momentum Transport

Momentum transport and plasma rotation have been studied extensively on many tokamaks in recent years. Numerous experimental results have been reported on individual devices – yet no dedicated multi-machine momentum transport experiments have been performed. This paper reports dedicated scans to study momentum transport that have been carried out on JET, DIII-D, AUG, NSTX and C-Mod within the ITPA framework. NBI modulation technique to create a periodic rotation perturbation has been exploited on JET, DIII-D and AUG from which the convective velocity vpinch and diffusion coefficient χ_phi profiles are determined. On NSTX, 50 ms pulses of n=3 non-resonant magnetic fields were applied to extract vpinch and χ_phi profiles. On C-Mod, either ICRH modulation or septum sweeping between the lower and upper null configurations was applied to create the rotation perturbation. A 3-point collisionality υ* scan by varying collisionality by a factor of 4–5 while keeping the other dimensionless quantities, such as ρ*, βN, q, R/Ln and Ti/Te, as constant as possible (10–20% variation), has been performed independently both on JET and DIII-D in L-mode plasmas. Neither the pinch nor the Prandtl number depends on collisionality on JET or on DIII-D. On NSTX, no clear trend between the pinch number and υ* was found either. The dependence of the Prandtl and pinch numbers on R/Ln obtained from JET, DIII-D and AUG NBI modulation shots indicates that Pr does not depend on R/Ln although JET plasmas tend to have higher values in general. On the contrary, the pinch number shows a clear dependence on R/Ln in each device separately and also as a joint database. This increase in the pinch number with increasing R/Ln is qualitatively consistent with theory and gyro-kinetic simulations. Based on the results from these multi-tokamak parametric scans, one can conclude that the inward pinch will have a significant impact on the rotation profile in ITER provided that the density profile is at least moderately peaked (R/Ln≥1) and some rotation source is available at the edge of the plasma. In all these analyses, the possible intrinsic torque has not been taken into account.

Speaker: Mr Tuomas Tala (Finland)

114 ITR/P1-20: Integrated Magnetic and Kinetic Control of Advanced Tokamak Scenarios Based on Data-Driven Models

The first real-time profile control experiments integrating magnetic and kinetic variables in tokamaks are described. Parameters such as the current, toroidal rotation and pressure profiles play a crucial role in governing plasma confinement and stability and their control is important for extrapolating advanced tokamak scenarios to future tokamaks. The integrated model-based approach presented here is being developed under the framework of the International Tokamak Physics Activity for Integrated Operation Scenarios and was initially explored on JET for current profile control. A generic method to identify device-specific, control-oriented models from experimental data was validated on JET, JT-60U and DIII-D. Such data-driven models were used to synthesize integrated magnetic and kinetic profile controllers with different levels of model integration. Closed-loop experiments were performed on DIII-D for the regulation of (a) the poloidal flux profile, psi(x), (b) the inverse of the safety factor profile, iota(x)=1/q(x), and (c) either the poloidal flux profile or the inverse of the safety factor profile together with the normalized pressure parameter, betaN. The neutral beam injection (NBI) and electron cyclotron current drive (ECCD) systems provided the heating and current drive sources for these experiments. Available beamlines and gyrotrons were grouped to form, together with the plasma surface loop voltage, Vext, or current, Ip, five independent heating and current drive actuators: (a) co-current NBI power, PCO, (b) counter-current NBI power, PCNT, (c) balanced NBI power, PBAL, (d) total ECCD power from all gyrotrons in an off-axis current drive configuration, PEC, and (e) Vext or Ip. Control of iota(x) or psi(x) and simultaneous control of iota(x) or psi(x) together with betaN were performed through a mixed-sensitivity robust control algorithm and a near-optimal proportional-plus-integral control algorithm, respectively. With the same approach, closed-loop control simulations have been performed for ITER. The nonlinear burning plasma evolution and closed-loop response to the specific ITER actuators under the controller action are modelled. Results for various control configurations and targets are discussed. Work supported by the US DOE under DE-FG02-09ER55064, DE-FG02-92ER54141, and DE-FC02-04ER54698, and by the European Fusion Development Agreement.

Speaker: Mr Didier Moreau (France)

115 ITR/P1-21: Stability and Performance of ITER Steady State Scenarios with ITBs

Steady state scenarios envisaged for ITER aim at optimizing the bootstrap current drive, while maintaining sufficient confinement and stability to provide the necessary fusion yield. Non-inductive steady state scenarios on ITER will need to operate with Internal Transport Barriers (ITBs) in order to reach adequate fusion gain at typical currents of 9 MA. We have analyzed the ideal MHD stability of steady state scenarios that use 8-33 MW of Neutral Beam with addition of different mixtures of external heating, namely 20-40 MW of EC, 20-40 MW of LH and 5-20 MW of IC. Target plasmas have H98~1.6, which is necessary to access 100% non-inductive current in the range of 7-10 MA and fusion yield of Q~3-6 [1]. These configurations are established as relaxed flattop states with time-dependent transport simulations, which have been run with TSC [2]. The sensitivity of n=1,2 kink (with and without a wall) and n=∞ ballooning instabilities has been analyzed against variations of the pressure profile peaking between 2.3 and 2.9, and of the Greenwald fraction up to 1.2. It is found that LH heating is desirable to maintain the safety factor profile above 1.5. By operating with broad pressure profiles and ITBs at 2/3 of the minor radius, plasmas with 33MW of NB and 40 MW of LH have q_min>2, are ideal MHD stable and can achieve beta_N >4l_i with an ideal wall. The EC provides localized heating and current drive profiles, but 40MW of EC raise the self-inductance and drive q_min down below 1.5, driving n=1 and n=2 kinks instabilities. The stability of these configuration can be improved for deposition outside ρ(q_min)=0.35. The trade-off of current drive efficiency vs ideal MHD stability with rho(q_min) will be discussed, with focus on the effectiveness of the upper launcher in controlling the current profile. [1] Kessel CE et al, Development of ITER Advanced Steady State and Hybrid Scenarios, IAEA Fusion Energy Conference 2010. This work was supported by the US Department of Energy under DE-AC02-CH0911466

Speaker: Ms Francesca Poli (USA)

116 ITR/P1-22: Non-linear MHD Modelling of ELM Triggering by Pellet Injection in DIII-D and Implications for ITER

ITER operation in its high fusion performance DT scenarios (inductive, hybrid and steady-state) relies on the achievement of the H-mode confinement regime, which is expected to lead to the quasi-periodic triggering of ELMs. Extrapolation of measurements of ELM energy fluxes to plasma facing components (PFCs) in present devices to ITER indicates that, for naturally occurring ELMs (or “uncontrolled”), these will produce an unacceptably low PFC lifetime because of excessive erosion and/or superficial surface damage. Controlled triggering of ELMs by the injection of small pellets has been demonstrated in present experiments as a viable technique to reduce ELM energy fluxes. The application of this technique to ITER requires frequencies exceeding (by typically more than one order of magnitude) those of uncontrolled ELMs. Thus, significant uncertainties remain for its practical application to ITER regarding both the optimization of the pellets for efficient triggering as well as of the associated fuel throughput required by this technique. In order to provide a firmer physics basis to the triggering of ELMs by pellet injection, and to reduce the uncertainties with regards to its application in ITER, non-linear MHD modelling of ELM triggering by pellet injection in DIII-D experiments has been carried out with the JOREK code. The modelling results show that the triggering of ELMs is associated with the pressure (and its gradient) in the pedestal region of the plasma reaching a critical value during the pellet ablation process. As a consequence, a minimum pellet size for ELM triggering (for given pellet velocity and injection geometry) has been identified. Modelling of the effect of pellet injection location (outer midplane, X-point region, high field side) and of the details of the ablation profile on ELM triggering by pellets at DIII-D, as well as of the power fluxes to PFCs during pellet triggered ELMs, are in progress and will be reported in the paper. Comparison between simulations for DIII-D and experimental measurements will be used to validate and refine modelling assumptions. Predictions of the pellet requirements for optimum ELM triggering in ITER will be carried out with the validated model and implications for the requirements and implications for pellet pacing as a technique for ELM control in ITER will be drawn.

Speaker: Mr Guido Huijsmans (ITER)

117 ITR/P1-23: Non-linear MHD Simulation of ELM Energy Deposition

Measurements of the power deposition profile on the divertor target due to ELMs show that the wetted area during the ELMs increases significantly with the amplitude of the ELM energy losses. Taking the broadening into account in the estimates for the allowable ELM size in ITER leads to a larger operating window in plasma current (up to ~8 MA) where natural ELMs can be tolerated. The allowable ELM sizes for the high performance DT scenario at 15 MA are very small and thus no significant broadening is expected for tolerable ELMs in this scenario. Non-linear MHD simulations of ELMs have previously shown good qualitative agreement on features like the formation of filaments and their propagation speed and the fine structure in the power deposition profile during ELMs. The next step towards a quantitative validation of MHD simulations of ELMs is to compare observed trends in the experimental data with results from nonlinear MHD simulations. Here, the non-linear MHD code JOREK is used to study the origin of the observed broadening of the wetted area as a function of the ELM size, and to provide a physics basis for the expected ELM power losses in ITER. MHD simulations of ELMs for JET-scale plasmas show a significant broadening of the power deposition profiles at the divertor targets during ELMs. The broadening is a function of the ELM amplitude, the dependence being in reasonable agreement with experimental observations. The broadening is due to the ergodisation of the magnetic field due to the magnetic perturbation of the unstable ballooning mode. The ergodised field shows the characteristic homoclinic tangles which lead to filamentary structures on the divertor target with multiple strike points. The broadening of the power deposition varies during the temporal evolution of the ELM itself, which is correlated with the amplitude of the magnetic perturbation. Simulations of ELMs in the ITER 15 MA, Q=10 reference scenario, including the ITER divertor and first wall geometry, show significant broadening due to the ELM magnetic perturbation on the outer vertical target, less on the inner target. The paper will describe in detail the influence of the various MHD and transport processes occurring during an ELM on the modelled ELM heat flux pattern.

Speaker: Mr Guido Huijsmans (ITER)

118 ITR/P1-24: Three-dimensional Fluid Modeling of Plasma Edge Transport and Divertor Fluxes during RMP ELM Control at ITER

Application of resonant magnetic perturbation (RMP) fields is one option for control of edge-localized modes (ELMs) at ITER. During RMP ELM suppression at DIII-D the measured heat and particle fluxes are rearranged into a three-dimensional (3D) pattern. In this contribution, the consequences of this 3D boundary formation on the divertor heat and particle loads during RMP application at ITER are studied. We use EMC3-Eirene, a 3D fluid plasma and kinetic neutral transport code. An n=3 RMP field at 90kAt maximum coil current was applied in (a) the vacuum limit and (b) including plasma response from non-linear MHD. A low recycling divertor regime is addressed at an input power of PH=100MW without impurity radiation losses. This represents an extreme case to test the upper bound for the 3D heat fluxes. A clear reduction of the electron temperature Te field is seen with vacuum RMP field. The outer boundary of the temperature field is set by the perturbed separatrix, which forms a finger like 3D boundary structure. These fingers reach out to the divertor target and the divertor fluxes are not axisymmetric anymore but deposited in a helical pattern reaching out as far as 70 cm from the unperturbed strike line. This yields spreading of the heat flux with a reduction of the peak heat flux from qpeak=28 MW/m2 without RMP field to qpeak=9 MW/m2 with RMP where the integral heat flux is reduced by 30%. This is induced by an increase of the divertor recycling flux by a factor of 10, which is necessary to maintain the density with RMPs applied. The increasing divertor density and enhanced ionization power losses reduce the divertor heat load. A comparison of these ionization losses with impurity radiation shows that they determine the actual divertor heat fluxes. The plasma response to the n=3 field applied was investigated based on a non-linear MHD plasma response modeling. For the case studied, all modes inside normalized flux of ΨN<0.96 are shielded out while in the edge for ΨN>0.96 a vacuum like penetration is predicted. This has a strong impact on the modeling result. The electron temperature is restored to the no-RMP value and the extension of the finger like structure at the separatrix is reduced. The maximum reach out of the helical heat fluxes are reduced to 30 cm and the heat flux deposition is more localized with increasing peak heat flux to q_peak=17 MW/m2.

Speaker: Mr Oliver Schmitz (Germany)

119 ITR/P1-25: 3D Vacuum Magnetic Field Modeling of the ITER ELM Control Coils during Standard Operating Scenarios

ITER edge localized mode (ELM) coil current optimization and failure studies have been completed for nine standard operating scenarios using an automated program that calculates the vacuum island overlap width (VIOW) and compares the results to a criterion that has been correlated with ELM suppression in DIII-D. The analysis was done, using n=3 and 4 perturbation fields, by varying the spatial phase of a cosine waveform approximated by the currents in the 9 coils making up each of the three rows of ELM coils. Results from the cases studied show that the minimum ELM coil current needed to satisfy the DIII-D correlation criterion varies from 20 kAt to 50 kAt depending on the operating scenario and that the available phase angle operating space increases rapidly with coil current above the minimum. It is also found that the DIII-D correlation criterion can be satisfied in the most demanding ITER scenario with n=3 perturbation fields and with failures in up to 8 of the full 27 coil set although the available phase angle operating space is reduced from 79% with no failures to 27% with eight failures using a maximum ELM coil operating current of 90 kAt. Details of these results will be discussed along with plans to extend the analysis to include the plasma response to the perturbation field. This work was supported in part by the US Department of Energy under DE-FG02-05ER54809, DE-FG02-07ER54917, and ITER Task Agreement C19TD42FU.

Speaker: Mr Todd E. Evans (USA)

120 ITR/P1-26: Analysis of Tungsten Dust Generation under Powerful Plasma Impacts Simulating ITER ELMs and Disruprions

In this paper, experimental simulations of ITER transient events with relevant surface heat load parameters (energy density 0.45-1.1 MJ/m2 and the pulse duration of 0.25 ms) as well as particle loads (varied in wide range of 1023-1027 ion/m2 s) were carried out with a quasi-stationary plasma accelerator QSPA Kh-50. Particular attention is paid to the material erosion due to particles ejection from the tungsten surfaces both in the form of droplets and solid dust. The erosion products flying from the tungsten target have been registered using high-speed 10 bit CMOS digital camera. Traces of particles ejected from the tungsten surface allow calculation of the particles velocity and the time moment when it started from the target surface. Additionally the mass loss of the target was measured after several shots. Erosion products ejected in the form of droplets and solid dust were also collected and examined with microscopy. Several mechanisms of dust generation under the transient energy loads to the tungsten surfaces have been identified. Dust particles with sizes up to tens microns are ejected from the surface due to the cracking development and major cracks bifurcation. Fatigue cracks after a large number of transient impacts to the preheated W surface became a source of smaller dust. Surface modification of tungsten material after the repetitive plasma pulses with development of ordered submicron cellular structures contributes significantly to the nm-dust generation. It is shown that majority of generated dust nano-particles, generated due to cells evolutions, are deposited back to the surface by a plasma pressure, in contrast to micrometer-size dust. For plasma exposures with energy loads above the melting threshold both droplets splashing and solid dust ejection is observed. Analytical estimations of dust production rate have been performed, and the results are verified with experiments in plasma guns having typical for ITER ELMs.

Speaker: Mr Boris Bazylev (Germany)

121 ITR/P1-27: Narrow Heat Flux Widths and Tungsten: SOLPS Studies of the Possible Impact on ITER Divertor Operation

Recent experimental observations of a very narrow SOL for energy flow in the inter-ELM H-mode have raised a concern for the consequences for target heat loading if it were to occur in ITER. Simulations using SOLPS4.3 for a carbon divertor have shown that because of the energy dissipation by impurity radiation in the divertor, the peak divertor power load could be maintained at an acceptable level even if the SOL width were reduced down to 1 mm instead of the 5 mm resulting from current transport assumptions. In other words, partial detachment of the divertor can counteract the effect of narrowing the SOL in ITER. However, the reduced radial transport in the narrow SOL case leads to relatively high separatrix density. Whereas this should not be a problem during the burn phase in ITER, low line average densities will almost certainly be necessary to obtain an acceptably low power threshold for the L-H transition, and this may be incompatible with high separatrix density. Investigations using the integrated core-pedestal-SOL model are underway to explore the expected plasma response around the L-H transition. In this study, we initially consider a carbon divertor with two sets of boundary conditions corresponding to the wide (L-mode) and narrow (H-mode) SOL and switch between them when the core plasma matches the usual criterion for the L-H transition (the transport reduction in the pedestal area is also applied at this point). An initial simulation study is also underway to investigate divertor performance during the burn phase in the case of W targets, using seeded Ne impurity to replace the naturally produced C radiation in the C divertor case. Since the transport and radiation properties of the seeded impurity differ from those of C, this can affect the controllability of the discharge and the divertor plasma conditions, as well as the distribution of radiative power loading on divertor components. As an example, initial modelling runs already available indicate that the radiation contribution to the peak power load on targets is lower for the W divertor than for C, since Ne radiates further upstream than C. A beneficial effect is a factor 2 reduction of the maximum radiation power loads in the gaps between the cassettes, where the heat removal can be critical.

Speaker: Mr Horst D. Pacher (ITER)

122 ITR/P1-28: Multi-machine Comparisons of Divertor Heat Flux Mitigation by Radiative Cooling

Due to the absence of carbon as intrinsic low-Z radiator, and tight limits for the acceptable power load on the divertor target, ITER will rely on impurity seeding for radiative power dissipation. Partial detachment of the outer divertor needs to be achieved and integrated with an ELM mitigation technique. This contribution reports about cross-machine studies of impurity seeded scenarios initiated by the ITPA group for integrated operational scenarios (IOS-1.2). In current devices, the plasma response to impurity seeding reveals a quite broad phenomenology. In Alcator C-Mod, energy confinement degrades when the power flux through the separatrix is reduced towards the L-H threshold power, PLH, under EDA H-mode conditions. In AUG, improvement of confinement is seen during nitrogen seeding at higher values of Psep/PLH, correlating with increasing Zeff due to N seeding. In JET, degradation of normalized confinement was usually obtained during seeding. Different responses of the pedestal on the increased impurity level is supposed to be the main origin of the observed behaviour. To allow the comparison of different experiments with regard to the heat flux reduction, an analytical scaling relation for the normalized divertor power load has been developed, based on the main plasma Zeff and the divertor neutral pressure. Higher divertor neutral pressures and higher heating powers in terms of P/R will be required to approach ITER divertor conditions in present day devices. No issues are foreseen for the combination of impurity seeding with ELM mitigation by pellet ELM pace-making. Regarding ELM mitigation by magnetic perturbations, the quite different responses of the plasma in AUG, DIII-D and JET do not allow a clear forecast for ITER. So far, no negative effects of magnetic perturbations on the radiative efficiency or core impurity penetration have been observed in DIII-D and AUG. On the technical side, possible sensors and combined control schemes for the facilitation of partial detachment in ITER will be discussed based on recent tokamak experience.

Speaker: Mr Arne Kallenbach (Germany)

123 ITR/P1-29: PTRANSP Tests of TGLF and Predictions for ITER

Time-dependent integrated predictive modeling is important for helping ITER achieve the physics goals of studying reactor-relevant burning plasmas. The PTRANSP code is being used to generate time-dependent integrated predictions. These are self-consistent in that the heating, current-drive, torques and equilibria are calculated using predicted plasma profiles, and vice versa. Predictions for ITER have incorporated physics-based models such as GLF23. An improved Trapped gyro-Landau Fluid model TGLF contains physics not included in GLF23 such as realistic shaped finite aspect ratio flux geometry, and collisionality. TGLF achieves more accurate predictions of temperatures measured in L-mode, H-mode and hybrid discharges than does GLF23. This paper describes a major upgrade to PTRANSP which implements TGLF. The upgrade uses a new robust solver for stiff transport models. Both GLF23 and TGLF are incorporated. The solver has both standalone and PTRANSP-coupled modes. The implementation of TGLF is verified by comparing results derived using the XPTOR code, and is tested using H-mode ITER-like plasmas. Predictions for ITER plasmas are given and compared with predictions using GLF23.

Speaker: Mr Robert Budny (USA)

124 ITR/P1-30: ITER Implications of the Beta Scaling of Energy Confinement

There is emerging evidence that the variation in the measured beta dependence of confinement in H-mode plasmas is due in part to different turbulent modes being dominant, with ITG modes being important in weak beta scaling cases and micro-tearing modes being potential candidates explaining strong beta degradation. Additionally, the normalized H-mode pedestal height may not be constant over a beta scan, which affects core transport and global confinement. Determining the beta scaling of transport helps to differentiate between various proposed theories of turbulent transport that are primarily electrostatic or primarily electromagnetic. Initial experiments on JET, DIII-D and NSTX found a weak dependence of confinement on beta; however, this picture of primarily electrostatic transport was brought into question by experiments on JT-60U and ASDEX Upgrade that observed a strong unfavorable beta scaling. The ITPA topical group on Transport and Confinement has coordinated experimental and modeling activity to better understand the origin of these different beta scalings. An important factor that can impact scaling results is experimental imperfections in the beta scans. For some experiments the normalized H-mode pedestal height decreases with higher beta, which can result in an unfavorable beta scaling even if core transport is primarily electrostatic. A DIII-D experiment with joint participation by the ASDEX Upgrade team found no beta dependence in the local thermal diffusivities outside of the ≈15% error bars, and the magnitude and trend with beta of density fluctuations were in reasonable agreement with GYRO simulations for electrostatic ITG-mode turbulence. In contrast, turbulence modeling of ASDEX Upgrade experiments using GS2 found that micro-tearing modes are unstable in the high beta cases but their contribution to the beta degradation remains to be assessed quantitatively. Micro-tearing modes should be important for high collisionality and flat density profiles, which were the conditions for ASDEX Upgrade and JT-60U. Therefore, the disparate beta scalings may be explained by either different dominant turbulence modes or experimental imperfections such as changes in the H-mode pedestal height during the beta scan. Supported in part by the US DOE under DE-FC02-04ER54698, DE-AC02-09CH11466, DE-FG02-07ER54917, DE-FG02-89ER53296, and DE-FG02-08ER54999.

Speaker: Mr C. Craig Petty (USA)

125 ITR/P1-31: Assessing the Power Requirements for Sawtooth Control in ITER through Modelling and Joint Experiments

Recent advances in theoretical understanding and numerical modelling of sawtooth oscillations have allowed the invention and application of experimental control techniques. This enhanced understanding, coupled with demonstration of control techniques in ITER-relevant plasmas and using real-time feedback, has facilitated prediction of control actuator requirements for ITER. The control of sawteeth is important for baseline scenario operation of burning plasmas, since plasmas with long sawtooth periods are empirically more susceptible to neoclassical tearing modes, which result in substantial confinement degradation. The stabilising effects of alpha particles are likely to exacerbate this, so recent experiments have identified methods for amelioration. Sawtooth control using electron cyclotron current drive has been demonstrated in ITER-like plasmas with a large fast ion fraction, wide q=1 radius and long uncontrolled sawtooth period in DIII-D and ASDEX Upgrade. Operation at beta(n)=3 without NTMs has been achieved in ITER demonstration plasmas in DIII-D using only modest ECCD power for sawtooth control. Further, real-time ECCD control techniques have been developed in TCV and Tore Supra. Numerical modelling suggests that the achieved driven current changes the local magnetic shear sufficiently to compensate for the stabilising influence of the fast particles. Extrapolating this to ITER, transport modelling coupled to ray-tracing predictions and using the linear stability thresholds for sawtooth onset suggests that 13MW of ECCD could be sufficient to reduce the sawtooth period by 30%, and this being the case, dropping it below the NTM triggering threshold. However, since the ECCD control scheme is solely predicated upon changing the local magnetic shear, it is prudent to plan for 10MW off-axis ICRH using 3He minority as a complementary scheme which directly damps the internal kink potential energy drive responsible for trapped fast ion stabilisation. Experimental evidence from JET plasmas heated with toroidally propagating ICRH using a He3 minority exhibited sawtooth control avoiding NTMs in H-mode, as predicted by drift-kinetic modelling. Such modelling suggests that 10MW of ICRH in ITER will negate the stabilising effect of alphas. *See Appendix of F. Romanelli et al., Proc 23rd IAEA FEC 2010, Daejeon, Korea Work funded by RCUK Energy Programme and EURATOM

Speaker: Mr Ian Chapman (UK)

126 ITR/P1-32: Observation of Localized Fast-Ion Induced Heat Loads in Test Blanket Module Simulation Experiments on DIII-D

Heat loads on the first wall of ITER can potentially be very high in localized regions such as the divertor or in regions on the first wall where magnetic field perturbations can channel energetic ions to create localized hot spots. One area where hot spots can be created in ITER is on the Test Blanket Modules (TBMs) because of the ferritic steel required for these components and their effect on the distortion of the poloidal and toroidal magnetic field near the modules. Simulating the level of those heat loads is important for assessing their effects on the ITER first wall. It is therefore essential that the codes that perform such assessments are validated against experimental results using configurations similar to those expected in ITER. The development of the mock-up ITER-like TBM on DIII-D allows just such a validation to be carried out on DIII-D for the case of neutral beam ions. An important new capability in the last run period was the direct infra-red imaging of the front surface of the protective TBM tiles and the calibration of the images to infer heat loads induced by the localized deposition of deuterium beam ions. A key result of the experiments is that the detailed simulations using a variety of particle following codes reproduce well the heat loads observed using the infra-red camera. This work was supported supported in part by the US DOE under DE-AC02-09CH11466, DE-AC52-07NA27344, DE-FC02-04ER54698, SC-G903402, and DE-AC05-00OR22725.

Speaker: Mr Gerrit J. Kramer (USA)

127 ITR/P1-33: Fast Ion Power Loads on ITER First Wall Structures in the Presence of ELM-mitigation Coils and MHD Modes

The new physics introduced by ITER operation, of which there is very little prior experience, is related to the large number of fast ions: fusion alphas, NNBI deuterons and multi-MeV minority ions from ICRH. These particles present a potential hazard to the surrounding material structures. Assuming axisymmetry and neoclassical transport only, the fast ion wall power loads are found tolerable in all scenarios and for all particle species. However, in ITER the axisymmetry is destroyed by several mechanisms: Finite number of TF coils causes toroidal ripple, the field is further perturbed by the test blanket modules, and the proposed ELM control coils (ECC) cause a field modulation at their own periodicity. All the deviations can cause significant fast ion leakage, leading to localized power loads on the walls. Furthermore, it is highly unlikely that ITER plasmas will be MHD quiescent: the massive fast ion population can drive energetic particle modes that act back to the fast ion population. ITER is also prone to NTM islands. All these MHD phenomena can increase fast ion population at the edge, where transport due to the field aberrations can lead to unacceptably high peak power loads on some first wall components. In this contribution we address these issues: the effect non-axisymmetry on NBI power loss, and the effect of NTM islands and Alfvénic modes (AEs) on fast ion distribution. Non-axisymmetry: we use the 5D Monte Carlo orbit-following code ASCOT to simulate fast ions in the presence of all relevant mechanisms perturbing the edge magnetic field. The fast ion power loads are found tolerable as long as the edge magnetic field does not become stochastic. The very high NBI power losses, reported earlier, can also be reproduced, but are found to correspond to stochastic edge field that does not support the NBI source profile assumed for the simulations. MHD effects: we have developed a model applicable for both stationary (NTMs) and rotating (AEs) MHD modes. The island structures are included in the equations of motion, expressed in vector form. Thus the allows arbitrary coordinate system and simulations all the way to the wall, and permits the use of 3D fields. We use the model for ITER to 1) determine the critical NTM island size not be exceeded from the fast ion confinement point-of-view, and 2) estimate the effect of AEs on the fast ion losses.

Speaker: Ms Taina Kurki-Suonio (Finland)

128 ITR/P1-34: Benchmark of Gyrokinetic, Kinetic MHD and Gyrofluid Codes for the Linear Calculation of Fast Particle Driven TAE Dynamics

Fast particles in ITER may originate from the fusion process itself or from external heating, as Neutral Beam Injection (NBI). It is well known that those non-thermal populations of fast particles may interact with otherwise stable Alfven waves in the bulk plasma driving them unstable. This process takes place as a resonance phenomenon that requires a kinetic treatment of the fast particles but not necessarily a kinetic treatment of the bulk plasma. The rising amplitude of the oscillating electro-magnetic field in the plasma may lead to a redistribution or a loss of supra-thermal particles. In consequence, a degradation of the heating efficiency or even hazards for in-vessel components of the machine are possible. In the last decades, much effort has been invested in the development of theory and codes that can be used to describe and explain the related phenomena. However, up to now, there is no well-understood standard case that these models have been tested against quantitatively. Providing the first quantitative code comparison, the ITPA Energetic particle Topical Group is contributing to the design activity on the ITER operation scenario. The benchmark of the codes for the energetic ion driven modes is necessary for the accurate prediction of plasma behavior in ITER. The international benchmarking effort between a variety of codes shall ensure scientific quality and reliability when predictions for ITER will be made. Comparisons will be made between the different codes at different levels of approximation for growth rates, frequencies and eigenfunctions. While the limit of zero orbit width has been well met by all codes, the spread of the results in the zero Larmor radius case illustrates the necessity of a quantitative comparison. Furthermore, the importance of finite Larmor radius (FLR) effects is shown which lower the growth rates considerably. The benchmark will provide codes capable of predicting Alfven eigenmode physics in ITER reliably.

Speaker: Mr Axel Könies (Germany)

129 ITR/P1-35: Effects of ELM Control Coil on Fast Ion Confinement in ITER H-mode Scenarios

This paper reports the effects of the ELM control coil on fast ion confinement in ITER H-mode plasmas. The effects of the ELM coil on the loss of NBI-produced fast ions and fusion-produced alpha particles have been investigated using an orbit following Monte-Carlo code for an ITER 9MA non-inductive plasma, where the simulations have been performed for the vacuum fields produced by the ELM coils and then magnetic screening effects could influence the numerical conclusions. The effect of the ELM coil field dominates the loss of NBI-produced fast ions over the effect of magnetic field ripple by the toroidal field coils and the test blanket modules, leading to a significant loss of fast ions on the order of 16-17%. A significant transit particle loss occurs in the cases of the toroidal mode number n=4 in which magnetic surfaces are ergodic near the plasma periphery. Concerning the resonance of fast-ion trajectories, the anti-resonant surfaces of the main mode n=4 are very close to the resonant surfaces of the complementary mode nc=(9-4)=5 and vice versa. Since the effect of resonance on fast-ion trajectories dominates that of anti-resonance, a synergy effect of the main and complementary modes enlarges the resonant regions. The simulations also shows that the optimization of current phase of the ELM coils is not effective for the loss of fast ions for the n=4 case in this plasma. The two-dimensional heat load on the first wall due to the NBI ion loss was evaluated. With a stationary magnetic field pattern the peak heat load near the upper ELM coils due to the NBI ion loss is as high as 1.0-1.5 MW/m^2, which exceeds the allowable level in ITER. The peak heat load can be reduced to 0.2-0.25 MW/m^2 by rotation of the ELM coil field pattern, a feature foreseen in the design of the ELM control system. Most loss particles hit the inner side of the torus of the dome in the ITER divertor, and the peak heat load averaged over the 9 toroidal sectors of ELM coils is in a range of 0.3-0.5 MW/m^2, which is in the acceptable level again. Simulations have been done also for 3.5 MeV alpha particles. The loss of alpha particles also increases due to the ELM coil field. However, the loss is still acceptably low at less than 1.0 %.

Speaker: Mr Toshihiro Oikawa (ITER)

130 ITR/P1-36: Assessment of the H-mode Power Threshold Requirements for ITER

This paper contains a comprehensive multi-machine assessment on accessing and maintaining H-mode plasmas in ITER. The results from these joint experiments address L-H transition power threshold issues, which are not adequately included in the scaling from the ITPA H-mode power threshold database. Consequently, these results affect the ability to make accurate predictions for the H-mode threshold in ITER using the presently available H-mode scaling relationships and can be used to improve and reduce the uncertainty in these predictions. For the non-nuclear operational phase in ITER with H and/or He plasmas, experiments have been performed in ASDEX Upgrade (AUG), C-Mod, DIII-D, JET, NSTX, and MAST. The ratio of P_TH(H)/P_TH(D) appears to be relatively consistent at about a value of 2. However, for helium there is a large variation in P_TH(He)/P_TH(D) from 1.0-2.0. On detailed examination of the results from the many devices, the ratio of P_TH(He)/P_TH(D) decreases towards unity with increasing L-mode (or target) electron density, which is a favorable trend for ITER operational scenarios at relatively higher target densities. The application of resonant magnetic perturbation (RMP) fields can lead to significant increases in the H-mode power threshold as has been determined in AUG, DIII-D, MAST and NSTX. Multi-machine results will also be presented on the H-mode confinement in He and also on modeling of ITER scenarios, based on the above results, and the predictions and implications for accessing H-mode plasmas in ITER. This work was supported in part by the US Department of Energy under DE-FC02-04ER54698, DE-AC52-07NA27344, DE-FC02-99ER54512, and DE-AC02-09CH11466.

Speaker: Mr Punit Gohil (USA)

131 ITR/P1-37: A Model for the Power Required to Access the H-mode in Tokamaks and Projections for ITER

We describe a new model for the L-H transition which, in common to other models, determines the physics requirements to stabilize turbulent transport by ExB shear but also derives the required edge power flow to achieve these physics requirements. Plasma transport in the L-mode plasma edge (typically from Psi = 0.95 to 1.0 is assumed to be dominated by resistive ballooning turbulence. Because of the access of an H-mode transport barrier, the scale length of the edge plasma density (Ln) is expected to be set by ionization balance and to be smaller than that of the temperature (LT). The power out flux from the core plasma is transported through the edge by conduction, convection and part of it is also radiated by impurities in this edge region before reaching the separatrix. Under the assumptions of dominant resistive ballooning transport the edge power balance is derived. Suppression of resistive ballooning transport is assumed to occur when the ExB rotational shear exceeds the typical autocorrelation time of the turbulence. This condition is used to evaluate the H-mode power threshold. It is shown that the H-mode power threshold for the existing tokamaks found to be in reasonable agreement with the measurements and predictions for ITER is shown. Power threshold can also be used to identify the physics processes leading to the increase of the H-mode threshold at low density: a) the scale length of density profile becomes longer than that of the temperature and b) the ratio of ion to electron temperature decreases for electron heated plasmas because of the inefficiency of equipartition at these densities as found in experiments.

Speaker: Mr Raghvendra Singh (India)

132 ITR/P1-38: Studying the Capabilities of Be Pellet Injection to Mitigate ITER Disruptions

Principal goal of the ITER disruption mitigation system (DMS) is retention of the heat and electro-mechanical loads on the machine components during disruption within the tolerable limits. This includes heat loads on divertor and plasma facing components during thermal quench (TQ), electromagnetic forces on vacuum vessel and other constructive elements during current quench (CQ) and heat loads due to the loss of runaway electron (RE) beam to the wall. Capabilities of the Be pellet injection to meet the goals of the ITER DMS are studied in the present report using the set of numerical codes providing integrated simulation of ITER disruptions with RE generation. Simulations of the TQ with ASTRA code demonstrated a principle possibility of re-radiation of more than 80% of plasma thermal energy with injection of about 2*1023 Be atoms. Subsequent simulation of the CQ with DINA code showed that the duration of CQ is close to, but exceeds the engineering limit of 36ms, especially in the case of most probable Upward vertical displacement events (VDE). The RE generation in the plasma with Be impurity was found to be strongly suppressed. However, at the CQ stage of simulations it was found that even very small RE seed current of about 100A would be sufficient to convert up to the half of the pre-disruption plasma current into RE one. Maximum electromagnetic forces on the VV of order of 100MN were found in DINA simulations of the Downward VDE with RE. Possibilities of the mitigation of the RE current and energy by means of the Be pellet injection at the CQ stage are discussed.

Speaker: Mr Sergey Konovalov (Russian Federation)

133 ITR/P1-39: Modelling of Material Damage and High Energy Impacts on Tokamak PFCs during Transient Loads

Tungsten (divertor) and beryllium (first wall) will be the plasma-facing components used on ITER. In reactor-scale tokamaks using metallic PFCs, transient events such as ELMs, VDEs and disruptions will produce strong vaporization and surface melting. Likewise, intense heat loads due to the impact of runaway electrons (RE) generated during the current quench phase of disruptions become a major issue in devices operating at high plasma current. Even if the thermal quench energy of major disruptions is expected to be successfully dissipated by mitigation using massive gas injection (MGI), the resulting photonic radiation loads on the ITER Be wall can be very intense. Unfortunately, no existing tokamak or laboratory device can simultaneously match all the conditions of ITER transients and so estimates of expected damage to ITER PFCs can only be provided by numerical simulations, supported by benchmarking on existing experiments. This paper describes a series of applications of the codes MEMOS, ENDEP and TOKES, developed at the Karlsruhe Institute of Technology, to specific ITER transient loading on both W and Be surfaces in the case of W divertor PFC melting due to disruptions (MEMOS), RE impact on Be first wall panels (MEMOS and ENDEP) and estimates of MGI driven photon radiation flash first wall heating (TOKES). An account is also given of benchmarking studies in which these codes have been compared with results obtained on the JET and TEXTOR tokamaks.

Speaker: Mr Boris Bazylev (Germany)

134 ITR/P1-40: Control of Major Disruptions in ITER

It is argued that major disruptions in ITER can be avoided by the feedback control of the causative MHD precursors. The sensors will be 2D-arrays of ECE detectors and the suppressors will be modulated ECH beams injected radially to produce non-thermal radial pressures to counter the radial dynamics of MHD modes. The appropriate amplitude and phase of this signal can stabilize the relevant MHD modes and prevent their evolution to a major disruption. For multimode MHD precursors, an optimal feedback scheme with a Kalman filter is discussed.

Speaker: Mr Amiya Sen (USA)

10:15 AM Coffee Break

Plasma Scenarios: EX/1

Convener: Mr Nikolay Ivanov (Russian Federation)

135 EX/1-1: Scenarios Development at JET with the New ITER-like Wall

In the recent JET experimental campaigns with the new ITER-like-Wall(ILW), major progress has been achieved in the characterisation of the H-mode regime: i) plasma breakdown and L-mode operation have been recovered in a few days of operation, ii) stable type I ELMy H-modes with H98y2 close to 1 and BetaN~1.6 have been achieved in high triangularity ITER–like shape plasmas on the bulk divertor tungsten tile, iii) it has been shown that the ELM frequency is a determining factor for the control of the core radiation level from metallic impurity and iv) in comparison to carbon equivalent discharges, total radiation is similar but the edge radiation is lower and the plasma core radiation higher. The maximisation of confinement, the control of metallic impurity sources and heat loads are the main challenges facing the development of the ITER scenarios in JET in the ILW environment at higher current and toroidal which will get closer to dimensionless ITER parameters in terms of rho* and nu*. This paper reviews the major physics and operational achievements and challenges that JET has to face to produce stable plasma scenarios (baseline and hybrid scenarios) with maximised performances with the ILW in support of ITER future operation.

Speaker: Mr Emmanuel Joffrin (France)

Slides EX-1-1_Joffrin_E..ppt

136 EX/1-2: Real time ELM, NTM and Sawtooth Control on TCV

TCV’s real-time (RT) control system uses a range of diagnostic signals to detect plasma events and react with programmable, controllable actuators such as orientable ECH power and Tokamak control actuators. Three examples of MHD phenomena control are presented with RT recognition of temporal and/or spatial extents of events controlled by synchronous modulation of the power or heating position displacement. ELM energy release changed with ECH in the plasma edge. Although these ELMs satisfy the Type I designation, their frequency increases, the relative energy loss per ELM decreases, although the coupled power decreases. ELM pacing was achieved by cutting the heating power at the ELM event for a given time before turning the power back on to trigger the arrival of the next ELM. The delay before the ELM appearance is found consistent with type-I ELM frequency vs average power scaling. At sufficiently high beta, ST crashes of sufficient amplitude generate NTM seed islands that grow. NTMs can be avoided by destabilising ST (the foreseen ITER strategy), or be stabilised altering the current distribution inside an island i.e. “healed” by localised ECH or current drive (ECCD). NTMs may be mitigated with continuous pre-emptive ECCD in the expected region, predictive ECCD if the generating ST crash time is known, and/or reactive ECCD when a NTM is observed. Stabilising individual STs was demonstrated on TCV by applying ECCD near the q=1 surface for a set duration after a ST crash. A new ST event is triggered a short, and relatively constant, time after the ECCD is then removed. Using the RT system, the ST period was changed on a crash to crash basis by varying the stabilising ECCD injection period. Pacing through destabilisation was also achieved by applying ECCD inside the q=1 surface with opposite phasing. Longer ST, that trigger NTMs, can now be generated. A second gyrotron providing ECCD at the q=3/2 surface just before the ST demonstrated efficient pre-emptive NTM stabilisation. An integrated control system simultaneously controls the ST period, preempts NTM formation and suppresses any NTMs that appear. The combination of RT identification and multi-actuator reaction has been demonstrated on TCV using a combination of targeted ECH and ECCD to efficiently pact, pre-empt and/or “heal” MHD modes.

Speaker: Mr Basil Duval (Switzerland)

Slides EX-1-2_Duval_B..pdf

137 EX/1-3: Progress in Performance and Understanding of Steady ELM-free I-modes on Alcator C-Mod

The I-mode regime of operation has been extended in recent Alcator C-Mod campaigns in duration and robustness, over a wide range of parameters. This attractive regime features an edge thermal barrier, and H-mode like energy transport, in combination with L-mode like density profiles and particle transport. This prevents accumulation of impurities, and means that ELMs are not needed to expel them. I-modes are now routinely maintained in stationary conditions for over 10 Tau_E. They are usually ELM free, a key advantage given the concern over divertor heat pulses on ITER. Instead, a continuous pedestal fluctuation appears to enhance selectively particle over thermal transport [2]. High performance I-modes are usually obtained with unfavourable ion drift direction. They have been produced in both upper and lower null plasmas, with q_95= 2.5-5.3 and extending to low nu*. Tau_E is in the range of H-mode, with H98,y2 up to 1.2, and exhibits less degradation with power (W~P^0.7). Power thresholds for I-mode are somewhat higher than typical L-H scalings, and increase with Ip as well as with density. The widest power range for I-mode, nearly a factor of two above the L-I threshold, has been obtained in reversed field, lower null discharges at moderate ne. Detailed measurements have been made of profiles and turbulence in the edge pedestal region, aiming to understand the separation of particle and energy transport. At the L-I transition, broadband turbulence in the 50-150 kHz range decreases. A pedestal-localized weakly coherent mode at ~200-250 kHz is observed on density, magnetic and Te diagnostics [3]. Stability analysis using ELITE shows that the pedestal is deeply stable to peeling-ballooning modes, consistent with the lack of ELMs. Initial assessments of the potential application of the I-mode regime on ITER, extrapolating from C-Mod results, indicate that an attractive operating scenario is possible, if issues of operation in the unfavourable drift configuration can be addressed. The L-I transition should be accessible at low density, and Q=10 is projected at n¬e_95 =5x10^19 m-3. This exercise also highlights some of the key issues remaining to be addressed, on C-Mod and in joint experiments. [1] Whyte D G et al 2010 Nucl. Fusion 50 105005. [2] Hubbard A E et al 2011 Phys. Plasmas 18 056115. [3] White A E et al Nucl. Fusion 51 (2011) 113005.

Speaker: Ms Amanda Hubbard (USA)

Slides Hubbard_IAEA12_EX1-3_I-mode_Talk.pdf

138 EX/1-4: Dominant ECR Heating of H-mode Plasmas on ASDEX Upgrade Using the Upgraded ECRH System and Comparison to Dominat NBI or ICR Heating

In contrast to ITER and future fusion reactors, plasmas used today for preparational or fundamental studies are often heated dominantly via the ion channel and with strong concurrent momentum input. This is due to the dominance of NBI heating systems, which are widely used because of their reliability and universal applicability. The potential danger of this approach is an over-estimation of the scaled fusion performace, which depends on T_i and not T_e, but the ratio of these temperatures may depend on the heating mix. Even worse, an increasing value of T_e/T_i is expected to increase the ITG dominated turbulent transport in the ion channel reducing T_i further. Additionally, the significantly reduced torque of future NBI-systems may increase transport due to a reduction of rotational shear. To clarify above mentioned uncertainties it is important to study the effect of dominant electron heating with a minimum of momentum input. At ASDEX Upgrade such studies have been started using the upgraded ECRH system which delivers 4 MW of ECRH to the plasma, exceeding the minimum H-mode power threshold for typical high I_p, B_t conditions by a factor of two. Additionally, 6 MW of ICRH and 20 MW of NBI are installed at AUG. This contribution reports on the upgrade of the ECRH system and its application in studying H-mode plasmas in which the heating mix between the three available systems is varied while keeping the total heating power constant. This was done for several different levels of total heating powers. Kinetic profiles (n,T,v,E) are measured using the recently upgraded suite of diagnostics in the core and, with high spatial resolution, in the pedestal region. Transport analysis and comparisons to predictions based on stability of drift modes (GS2) are presented. The major findings are: no effect of the heating mix on pedestal pressure although plasma rotation varies significantly, no direct correlation of the rotational shear with the shape of density or temperature profiles, and a significant increase of T_e/T_i as the fraction of electron heating exceeds a certain threshold. The cases analysed so far were all at high collisionality. These studies will be continued towards higher heating powers and lower collisionality.

Speaker: Mr Joerg Stober (Germany)

Slides Stober_IAEA12.ppt

139 EX/1-5: Fully Noninductive Scenario Development in DIII-D Using New Off-Axis Neutral Beam Injection

New off-axis neutral beam injection (NBI) capability on DIII-D has expanded the range of achievable and sustainable current and pressure profiles of interest for developing the physics basis of steady-state scenarios in future tokamaks. Fully noninductive (NI) scenarios are envisioned to have broad current and pressure profiles with elevated minimum safety factors (q_min), high normalized beta (beta_N), and a large fraction of the plasma current I_P sustained by the bootstrap current. Using off-axis NBI, plasmas have been produced with q_min between ~1.3 and ~2.5 to evaluate the suitability for steady-state operation (f_NI≡I_NI/I_P=1). Nearly stationary plasmas were sustained for two current profile relaxation timescales (3 s), with q_min=1.5, beta_N=3.5, f_NI=70%, and performance that projects to Q=5 in an ITER-size machine. The duration of the high beta_N phase is limited only by the available NBI energy. Low-order tearing modes are absent and the predicted ideal-wall n=1 kink beta_N limit is >4. To achieve higher f_NI, higher beta_N is needed to increase the bootstrap current, and higher q_min will decrease the required external current drive near the axis. Experiments to produce plasmas with q_min>2 showed that the use of off-axis NBI results in higher sustained q_min, with q_min at a larger radius (i.e. a broader current profile), and a broader pressure profile. These changes increased the predicted ideal-wall n=1 kink mode beta_N limit from below to above beta_N =4. These plasmas achieved a maximum beta_N=3.2 limited by the available NBI power and reduced confinement (H_98~1) relative to similar plasmas with lower q_min. beta_N=4 with q_min>1.5 was transiently obtained albeit with only 2 out of 5 MW of off-axis NBI available. Off-axis fishbones and low-order tearing modes were observed in the course of the q-profile scan. These studies indicate that obtaining a sustained, high performance, f_NI=1 scenario involves a number of trade-offs related to the choice of q-profile. This work was supported by the US Department of Energy under DE-AC52-07NA27348, DE-FC02-04ER54698, DE-FG02-04ER54761, DE-AC02-09CH11466, DE-FG02-08ER85195, DE-FG02-08ER549874, DE-AC05-00OR22725, and SC0G804302.

Speaker: Mr Christopher T. Holcomb (USA)

Slides Holcomb_v8.pdf

12:30 PM Lunch

Overview: Inertial & Magnetic Fusion: OV/4

Convener: Mr Raymond Fonck (USA)

140 OV/4-1: Progress of the JT-60SA Project

The shared procurement and construction of the JT-60SA device by Japan and EU is progressing well, including preparation of the plan for key research and development. The JT-60SA device has been designed in order to complement ITER in all areas of fusion plasma development necessary to decide DEMO construction. Detailed studies to predict plasma performance have confirmed these capabilities. The tokamak construction will start in Dec.2012.JT-60SA enables explorations in ITER- and DEMO-relevant plasma regimes in terms of the non-dimensional parameters (beta, the normalized poloidal gyro radius, the normalized collisionality, the fast ion beta etc.) under ITER- and DEMO-relevant heating conditions (such as dominant electron heating and low central fuelling, and low external torque input). Detailed studies of plasma performance prediction support these capabilities. Under these conditions, heat/particle/momentum transport, L-H transition, ELM/RMP/Grassy-ELM characteristics, the pedestal structure, high energy ion behaviors and the divertor plasma controllability can be quantified. By integrating these studies, the project provides ‘simultaneous & steady-state sustainment of the key performance characteristics required for DEMO’ with integrated control scenario development. Assuming HH=1.3-1.4, the expected Ip for high beta-N (=4.3), high bootstrap fraction (=70-80%) full non-inductive current drive is 2.1-2.3MA at the Greenwald density ratio (=1). The central reference of DEMO for JT-60SA is a compact steady-state reactor. However, the JT-60SA research project has to treat the ‘DEMO regime’ as a spectrum spreading around the reference design, and has to assess reliable DEMO design targets.

Speaker: Mr Yutaka Kamada (Japan)

Slides OV-4-1_Kamada_Y..pptx

141 OV/4-2: Present Status of Fast Ignition Realization EXperiment and Inertial Fusion Energy Development

Controlled thermonuclear ignition and subsequent burn will be demonstrated in a couple of years on the central ignition scheme. Fast ignition has the high potential to ignite a fuel using only about one tenth of laser energy necessary to the central ignition. This compactness may largely accelerate inertial fusion energy development. One of the most advanced fast ignition programs is the Fast Ignition Realization Experiment (FIREX). The goal of its first phase is to demonstrate ignition temperature of 5 keV, followed by the second phase to demonstrate ignition-and-burn. The second series experiment of FIREX-I from late 2010 to early 2011 has demonstrated a high (≈20%) coupling efficiency from laser to thermal energy of the compressed core, suggesting that one can achieve the ignition temperature at the laser energy below 10 kJ. Given the demonstrations of the ignition temperature at FIREX-I and the ignition-and-burn at the National Ignition Facility, the inertial fusion research would then shift from the plasma physics era to power generation era.

Speaker: Mr Hiroshi AZECHI (Japan)

Slides OV-4-2_Azechi_H..pdf

142 OV/4-3: Energetic Particle Instabilities in Fusion Plasmas

A remarkable progress was made in diagnosing energetic particle instabilities on present-day machines and in establishing a theoretical framework for describing them. This overview presents a point-by-point comparison between the much improved diagnostics of Alfvén Eigenmodes (AEs) and modelling tools developed world-wide, and outlines progress in interpreting the observed phenomena. A multi-machine comparison is presented giving a fair idea on the performance of both diagnostics and modelling tools for different plasma conditions. On JET equipped with 2D gamma-ray camera and interferometry, core-localised TAEs were detected causing redistribution of fast ions from inside the q=1 radius to outer plasma region followed by monster sawtooth crashes. TAE modelling using the HAGIS code showed a good agreement with the measured re-distribution and its effect on sawteeth. On DIII-D and AUG, ECE-imaging provides detailed measurements of amplitude and structure of AEs. A successful modelling was performed using the ORBIT code for reproducing the anomalously flat beam profiles on DIII-D. In AUG, a monitoring of the fast-ion redistribution and losses with an array of scintillators and fast-ion D-alpha spectroscopy has shown how a radial chain of overlapping AEs enables the transport of fast-ions from the core to the fast-ion loss detector. On NSTX, beam-driven AEs were observed in the form of “avalanches” consisting of coupled modes with strong frequency chirp. These modes caused ~10% drops in the neutron rate explained by effects of decrease in the beam energy and beam losses resulting from the interaction with TAE. A nonlinear model for near-threshold beam-driven instabilities has successfully encompassed many of the temporal characteristics of AEs seen in experiments. A steady-state nonlinear mode saturation and bursts of mode activity were found to be associated with both the strength and type of relaxation process in the phase-space region surrounding the resonance of the distribution function. An extension of the model to modes with a frequency sweep comparable to the starting frequency opened the opportunity for understanding the chirping modes in DIII-D, MAST, NSTX, START, and LHD. Finally, this presentation will outline expectations for ITER based on our present knowledge. This work was funded by the RCUK Energy programme and EURATOM.

Speaker: Mr Sergei Sharapov (UK)

Slides OV-4-3_Sharapov_S..pdf

143 OV/4-4: Overview of Recent and Current Research on the TCV Tokamak

Through a diverse research program, TCV addresses physics issues and develops tools for ITER and for the longer-term goals of nuclear fusion, relying especially on its extreme plasma shaping and ECRH launching flexibility and preparing for an ECRH and NBI power upgrade. Localized edge heating was unexpectedly found to decrease the period and relative energy loss of ELMs. Successful ELM pacing has been demonstrated by following individual ELM detection with an ECRH power cut before turning the power back on to trigger the next ELM, the duration of the cut determining the ELM frequency. In a parallel study, negative triangularity was also seen to reduce the ELM energy release. Both stabilizing and destabilizing agents (ECCD on or inside the q=1 surface, respectively) were used in a similar scheme to pace sawtooth oscillations, permitting precise control of their period. Locking of the sawtooth period to a pre-defined ECRH modulation period has also been demonstrated. In parallel with fundamental investigations of NTM seed island formation by sawtooth crashes, sawtooth control has permitted nearly failsafe NTM prevention when combined with backup NTM stabilization by ECRH. Additional work has addressed the destabilization of NTMs in the absence of a sawtooth trigger, and particularly its relation to plasma rotation. Further real-time control developments include the demonstration of joint current and internal inductance control using the Ohmic transformer and the validation of an ECRH power absorption observer based on transmitted stray radiation, for eventual polarization control. A new profile control methodology was also introduced, relying on real-time modelling to supplement diagnostic information; the RAPTOR current transport code in particular has been employed for joint control of the internal inductance and temperature profile. H-mode studies have focused on the L-H threshold dependence on the main ion species and on the divertor leg length. In L-mode, a systematic scan of the auxiliary power deposition profile, with no effect on confinement, has ruled it out as the cause of confinement degradation. Both L- and H-modes have been explored in the snowflake regime with emphasis on edge measurements, revealing that the heat flux to the strike points on the secondary separatrix increases as the X-points approach each other, well before they coalesce.

Speaker: Mr Stefano Coda (Switzerland)

Slides iaea_12.ppt

144 OV/4-5: Science and Technology Research & Development in Support to ITER and the Broader Approach

Magnetic Fusion Energy has now entered its development era that steers the activities of traditional fusion laboratories. Recent achievements in fusion science and technology in support to both the ITER and the Broader Approach (BA) projects are reported here. On top of the direct contribution to ITER and JT-60SA procurement packages, many scientific activities, aiming at reducing risks in operation of ITER and BA, have been carried out using the CEA dedicated in-house facilities (Tore Supra tokamak, ICRH test facility for ITER, remote operated diagnostics, actively cooled PFC qualification, cryogenics test facilities from strand to sub size superconducting conductors characterization, etc). The paper reviews the research and development actions taken in the past two years by CEA in this context, in order to ensure an ITER safe operation (quench detection, disruption mitigation, surface monitoring of plasma facing components), qualify the long pulse RF Heating and Current Drive systems, and progress in MHD, turbulence and transport first principle simulations. A fully documented project, turning Tore Supra into a long pulse actively cooled diverted plasma test facility, is now being proposed to the ITER partners. This evolution allows the R&D and commissioning tests of actual ITER actively cooled tungsten divertor elements under ITER-relevant edge plasma conditions, during the ITER procurement phase, and targets its risk reduction. In parallel, the contribution to the Broader Approach projects is shown to be complemented by an ambitious programme on integrated modeling of the main scenarios and an assessment of EC power needed for NTM stabilization.

Speaker: Mr Alain BECOULET (France)

Transparents IAEA_FEC_2012_IRFM_live.pptx

Poster: P2

145 EX/1-1: Scenarios Development at JET with the New ITER-like Wall

In the recent JET experimental campaigns with the new ITER-like-Wall(ILW), major progress has been achieved in the characterisation of the H-mode regime: i) plasma breakdown and L-mode operation have been recovered in a few days of operation, ii) stable type I ELMy H-modes with H98y2 close to 1 and BetaN~1.6 have been achieved in high triangularity ITER–like shape plasmas on the bulk divertor tungsten tile, iii) it has been shown that the ELM frequency is a determining factor for the control of the core radiation level from metallic impurity and iv) in comparison to carbon equivalent discharges, total radiation is similar but the edge radiation is lower and the plasma core radiation higher. The maximisation of confinement, the control of metallic impurity sources and heat loads are the main challenges facing the development of the ITER scenarios in JET in the ILW environment at higher current and toroidal which will get closer to dimensionless ITER parameters in terms of rho* and nu*. This paper reviews the major physics and operational achievements and challenges that JET has to face to produce stable plasma scenarios (baseline and hybrid scenarios) with maximised performances with the ILW in support of ITER future operation.

Speaker: Mr Emmanuel Joffrin (France)

146 EX/1-2: Real time ELM, NTM and Sawtooth Control on TCV

TCV’s real-time (RT) control system uses a range of diagnostic signals to detect plasma events and react with programmable, controllable actuators such as orientable ECH power and Tokamak control actuators. Three examples of MHD phenomena control are presented with RT recognition of temporal and/or spatial extents of events controlled by synchronous modulation of the power or heating position displacement. ELM energy release changed with ECH in the plasma edge. Although these ELMs satisfy the Type I designation, their frequency increases, the relative energy loss per ELM decreases, although the coupled power decreases. ELM pacing was achieved by cutting the heating power at the ELM event for a given time before turning the power back on to trigger the arrival of the next ELM. The delay before the ELM appearance is found consistent with type-I ELM frequency vs average power scaling. At sufficiently high beta, ST crashes of sufficient amplitude generate NTM seed islands that grow. NTMs can be avoided by destabilising ST (the foreseen ITER strategy), or be stabilised altering the current distribution inside an island i.e. “healed” by localised ECH or current drive (ECCD). NTMs may be mitigated with continuous pre-emptive ECCD in the expected region, predictive ECCD if the generating ST crash time is known, and/or reactive ECCD when a NTM is observed. Stabilising individual STs was demonstrated on TCV by applying ECCD near the q=1 surface for a set duration after a ST crash. A new ST event is triggered a short, and relatively constant, time after the ECCD is then removed. Using the RT system, the ST period was changed on a crash to crash basis by varying the stabilising ECCD injection period. Pacing through destabilisation was also achieved by applying ECCD inside the q=1 surface with opposite phasing. Longer ST, that trigger NTMs, can now be generated. A second gyrotron providing ECCD at the q=3/2 surface just before the ST demonstrated efficient pre-emptive NTM stabilisation. An integrated control system simultaneously controls the ST period, preempts NTM formation and suppresses any NTMs that appear. The combination of RT identification and multi-actuator reaction has been demonstrated on TCV using a combination of targeted ECH and ECCD to efficiently pact, pre-empt and/or “heal” MHD modes.

Speaker: Mr Basil Duval (Switzerland)

147 EX/1-3: Progress in Performance and Understanding of Steady ELM-free I-modes on Alcator C-Mod

The I-mode regime of operation has been extended in recent Alcator C-Mod campaigns in duration and robustness, over a wide range of parameters. This attractive regime features an edge thermal barrier, and H-mode like energy transport, in combination with L-mode like density profiles and particle transport. This prevents accumulation of impurities, and means that ELMs are not needed to expel them. I-modes are now routinely maintained in stationary conditions for over 10 Tau_E. They are usually ELM free, a key advantage given the concern over divertor heat pulses on ITER. Instead, a continuous pedestal fluctuation appears to enhance selectively particle over thermal transport [2]. High performance I-modes are usually obtained with unfavourable ion drift direction. They have been produced in both upper and lower null plasmas, with q_95= 2.5-5.3 and extending to low nu*. Tau_E is in the range of H-mode, with H98,y2 up to 1.2, and exhibits less degradation with power (W~P^0.7). Power thresholds for I-mode are somewhat higher than typical L-H scalings, and increase with Ip as well as with density. The widest power range for I-mode, nearly a factor of two above the L-I threshold, has been obtained in reversed field, lower null discharges at moderate ne. Detailed measurements have been made of profiles and turbulence in the edge pedestal region, aiming to understand the separation of particle and energy transport. At the L-I transition, broadband turbulence in the 50-150 kHz range decreases. A pedestal-localized weakly coherent mode at ~200-250 kHz is observed on density, magnetic and Te diagnostics [3]. Stability analysis using ELITE shows that the pedestal is deeply stable to peeling-ballooning modes, consistent with the lack of ELMs. Initial assessments of the potential application of the I-mode regime on ITER, extrapolating from C-Mod results, indicate that an attractive operating scenario is possible, if issues of operation in the unfavourable drift configuration can be addressed. The L-I transition should be accessible at low density, and Q=10 is projected at n¬e_95 =5x10^19 m-3. This exercise also highlights some of the key issues remaining to be addressed, on C-Mod and in joint experiments. [1] Whyte D G et al 2010 Nucl. Fusion 50 105005 [2] Hubbard A E et al 2011 Phys. Plasmas 18 056115 [3] White A E et al Nucl. Fusion 51 (2011) 113005

Speaker: Ms Amanda Hubbard (USA)

148 EX/1-4: Dominant ECR Heating of H-mode Plasmas on ASDEX Upgrade Using the Upgraded ECRH System and Comparison to Dominat NBI or ICR Heating

In contrast to ITER and future fusion reactors, plasmas used today for preparational or fundamental studies are often heated dominantly via the ion channel and with strong concurrent momentum input. This is due to the dominance of NBI heating systems, which are widely used because of their reliability and universal applicability. The potential danger of this approach is an over-estimation of the scaled fusion performace, which depends on T_i and not T_e, but the ratio of these temperatures may depend on the heating mix. Even worse, an increasing value of T_e/T_i is expected to increase the ITG dominated turbulent transport in the ion channel reducing T_i further. Additionally, the significantly reduced torque of future NBI-systems may increase transport due to a reduction of rotational shear. To clarify above mentioned uncertainties it is important to study the effect of dominant electron heating with a minimum of momentum input. At ASDEX Upgrade such studies have been started using the upgraded ECRH system which delivers 4 MW of ECRH to the plasma, exceeding the minimum H-mode power threshold for typical high I_p, B_t conditions by a factor of two. Additionally, 6 MW of ICRH and 20 MW of NBI are installed at AUG. This contribution reports on the upgrade of the ECRH system and its application in studying H-mode plasmas in which the heating mix between the three available systems is varied while keeping the total heating power constant. This was done for several different levels of total heating powers. Kinetic profiles (n,T,v,E) are measured using the recently upgraded suite of diagnostics in the core and, with high spatial resolution, in the pedestal region. Transport analysis and comparisons to predictions based on stability of drift modes (GS2) are presented. The major findings are: no effect of the heating mix on pedestal pressure although plasma rotation varies significantly, no direct correlation of the rotational shear with the shape of density or temperature profiles, and a significant increase of T_e/T_i as the fraction of electron heating exceeds a certain threshold. The cases analysed so far were all at high collisionality. These studies will be continued towards higher heating powers and lower collisionality.

Speaker: Mr Joerg Stober (Germany)

149 EX/1-5: Fully Noninductive Scenario Development in DIII-D Using New Off-Axis Neutral Beam Injection

New off-axis neutral beam injection (NBI) capability on DIII-D has expanded the range of achievable and sustainable current and pressure profiles of interest for developing the physics basis of steady-state scenarios in future tokamaks. Fully noninductive (NI) scenarios are envisioned to have broad current and pressure profiles with elevated minimum safety factors (q_min), high normalized beta (beta_N), and a large fraction of the plasma current I_P sustained by the bootstrap current. Using off-axis NBI, plasmas have been produced with q_min between ~1.3 and ~2.5 to evaluate the suitability for steady-state operation (f_NI≡I_NI/I_P=1). Nearly stationary plasmas were sustained for two current profile relaxation timescales (3 s), with q_min=1.5, beta_N=3.5, f_NI=70%, and performance that projects to Q=5 in an ITER-size machine. The duration of the high beta_N phase is limited only by the available NBI energy. Low-order tearing modes are absent and the predicted ideal-wall n=1 kink beta_N limit is >4. To achieve higher f_NI, higher beta_N is needed to increase the bootstrap current, and higher q_min will decrease the required external current drive near the axis. Experiments to produce plasmas with q_min>2 showed that the use of off-axis NBI results in higher sustained q_min, with q_min at a larger radius (i.e. a broader current profile), and a broader pressure profile. These changes increased the predicted ideal-wall n=1 kink mode beta_N limit from below to above beta_N =4. These plasmas achieved a maximum beta_N=3.2 limited by the available NBI power and reduced confinement (H_98~1) relative to similar plasmas with lower q_min. beta_N=4 with q_min>1.5 was transiently obtained albeit with only 2 out of 5 MW of off-axis NBI available. Off-axis fishbones and low-order tearing modes were observed in the course of the q-profile scan. These studies indicate that obtaining a sustained, high performance, f_NI=1 scenario involves a number of trade-offs related to the choice of q-profile. This work was supported by the US Department of Energy under DE-AC52-07NA27348, DE-FC02-04ER54698, DE-FG02-04ER54761, DE-AC02-09CH11466, DE-FG02-08ER85195, DE-FG02-08ER549874, DE-AC05-00OR22725, and SC0G804302.

Speaker: Mr Christopher T. Holcomb (USA)

150 EX/P2-01: Access and Sustained High Performance in Advanced Inductive Discharges with ITER-Relevant Low Torque

Recent experiments on DIII-D have demonstrated that advanced inductive discharges with high normalized fusion gain approaching levels consistent with ITER Q=10 operation can be accessed and sustained with very low amounts (~1 Nm) of externally driven torque. This level of torque is anticipated to drive a similar amount of rotation as the beams on ITER, via simple consideration of the scaling of the moment of inertia and confinement time. These discharges have achieved beta_N~3.1 with H_98~1 at q_95~4, and have been sustained for the maximum duration of the counter neutral beams (NBs). In addition, plasmas using zero net neutral beam torque from the startup all the way through to the high beta phase have also been created. Advanced inductive discharges are found to become increasingly susceptible to 2/1 neoclassical tearing modes as the torque is decreased, which if left unmitigated, generally slow and lock, terminating the high performance phase of the discharge. Access is not notably different whether one ramps the torque down at high beta_N, or ramps the beta_N up at low torque. The use of electron cyclotron heating (ECH) has proven to be an effective method of avoiding such modes, allowing stable operation at high beta and low torque, a portion of phase space that has otherwise been inaccessible. In many cases, the ECH has been aimed to drive current near the q=2 surface, although this does not appear to be a critical element in order to gain the benefits of the ECH. Indeed, high beta_N~3 discharges at low torque have been sustained using ECH without current drive, and deposited significantly inside of the q=2 surface. The insensitivity to the deposition position, together with the lack of need for current drive, suggests that the EC assists stability in a different way than the simply replacing the bootstrap current caused by the flattening of the pressure profile in the island. These advanced inductive discharges are measured to have significant levels of intrinsic torque at the edge, consistent with a previously determined empirical scaling considering the role of the turbulent Reynolds stress and thermal ion orbit loss. This work was supported by the US Department of Energy under DE-AC02-09CH11466, DE-FC02-04ER54698, DE-FG02-04ER54761, DE-FG02-08ER85195, and DE-AC52-07NA27344.

Speaker: Mr Wayne M. Solomon (USA)

151 EX/P2-02: ITER Demonstration Discharges on Alcator C-Mod in Support of ITER

Alcator C-Mod is providing discharges that match several simultaneous parameters expected in ITER in the rampup, flattop and rampdown phases, and simulations of these discharges with time-dependent transport evolution codes. Discharges have been produced at both B = 5.4 T and 2.7 T, utilizing H-minority heating fundamental and 2nd harmonic, respectively. Discharges with rampup durations appropriate to ITER’s show that ICRF heating obtains V-s savings with only weak effects on the current profile, in spite of strong modifications of the central electron temperature. Significant V-s savings in ohmic rampup by utilizing lower densities appears only to be effective at very low density, n/nGr ~ 0.08. Injection of lower hybrid during the rampup is effective for saving V-s, again only at similarly low densities. Simulations of C-Mod rampup discharges have been performed with the Tokamak Simulation Code (TSC), utilizing TORIC full wave calculation of the ICRF deposition, and using the Current Diffusive Ballooning (CDBM), Bohm-gyro Bohm, Coppi-Tang, and modified GLF23 (enhanced thermal diffusivity near plasma edge) transport models showing that they are not reproducing the temperature profile evolution, and consequently do not reproduce the experimental internal self-inductance or the V-s. Discharges which obtained EDA H-modes during flattop, with B = 2.7 T and Ip = 0.65 MA, obtain parameter values between (bN, n/nGr, H98) = (1.9, 0.60, 1.0) to (1.5, 0.8, 0.67). The lower n/nGr values are associated with the higher H98 and bN. Discharges showed a degradation of the energy confinement as the higher densities were approached, but also an increasing H98 with net power to the plasma (Pnet = PICRF + POH – dW/dt – Prad). For these discharges up to 3 MW was injected, while intrinsic impurities (B, Mo) provided radiated power fractions (Prad/Pin) of 25-37%, typical of those required in ITER. Experiments at 5.4 T have demonstrated the plasma can remain in H-mode as the rampdown phase is entered with at least 0.75 MW of ICRF injection, the back transition occurring when the net power reaches 1 MW, and the density will decrease at the same rate as Ip when in H-mode, maintaining the flattop n/nGr value. Maintaining the H-mode longer into the rampdown and ramping the plasma current down faster can mitigate the OH coil over-current associated with the back transition.

Speaker: Mr Charles E. Kessel (USA)

152 EX/P2-03: Overview of ASDEX Upgrade ‘Improved H-mode’ Scenario Developments

‘Improved H-mode’ discharges in ASDEX Upgrade (AUG) are characterized by enhanced confinement factors, high beta and a q-profile with almost zero shear in the core of the plasma at q(0) around 1. First attempts to reach high performance in the all-tungsten AUG relied on the observation that energy confinement was most improved when nitrogen (N) was puffed. In such plasmas the achieved confinement factors H_98 turned out to be as good as, or even better than, the values obtained in the carbon dominated AUG when discharges at the same values of F_GW are compared. The improved confinement of N seeded discharges is an effect of higher pedestal temperatures which extend to the plasma core via profile stiffness. Further developments of the N seeded scenario aim to extend the duration of the high performance phase to a few seconds and also to increase the triangulatity. The latter step helps to extend the operational window towards higher F_GW values. By introducing the so called current over-shoot technique significantly improved confinement with H_98 around 1.3 was reached in the carbon walled era at JET. This technique has now been applied to the all-W AUG where ‘improved H-modes’ with beta_Ν of 4 and H-factors of 2.5 have been produced transiently. The scenario at AUG uses a fast plasma current ramp up to 1.2 MA with NBI heating starting with the X-point formation and ending at the time of maximum I_p. Then an ohmic phase with a current ramp down to 1 MA modifies the edge part of the q-profile. In the consecutive heating power ramp to about 10 MW many parameters, e.g. stored energy and electron density, but also the radiation, increase strongly. The most stable pulses so far utilise also on-axis ECRH for impurity control and off-axis ECCD to stabilise an occurring (2,1) mode in the periphery of the plasma. With small modifications (gas puff rate, timing of pre-heat in I_p ramp-up) and in particular with less aggressive heating the stability and duration of the high performance phase has been extended. This paper will deal with the present status of AUG plasma operation of ‘improved H-Mode’ scenarios at optimized performance reporting on improvements of the nitrogen seeded as well as the current overshoot scenario. For the latter experimental results will be accompanied by TGLF calculations.

Speaker: Mr Josef Schweinzer (Germany)

153 EX/P2-04: Study of H-mode Access in the Alcator C-Mod Tokamak: Density, Toroidal Field and Divertor Geometry Dependence

Knowing the conditions for H-mode access is important for ITER high performance plasma operation. The experimental study carried out on Alcator C-Mod in support of this research primarily focused on (1) determining optimum global and local plasma conditions for promoting H-mode access, (2) characterizing plasma behaviors before L-H transition at low density, and (3) demonstrating a strong (>50%) reduction in H-mode threshold power (P_th)with modified divertor geometry. It is known that “hidden” variables other than those indicated by the multi-machine scaling law can also have a great impact on P_th. In C-Mod, we observed a strong reduction in P_th at medium and high densities with slot divertor operation. A minimum P_th of 0.7MW appears at 1.5x10^20m^-3 in this configuration, which is only 40% of the scaling law prediction. Interestingly, the edge T_e and n_e profiles prior to L-H transition are not apparently affected by divertor geometry. This result is promising and of particular interest for H-mode access at reduced power. H-mode access at low density is a potential concern for ITER H-mode operation. This issue has been studied in dedicated C-Mod experiments operated at two different B_T (5.4T and 3.5T). At 5.4T, both P_th and T_e,95 (T_e at psi=0.95) preceding L-H transition rise considerably for density below 1.0x10^20m^-3. The ion and electron temperature near the pedestal top remain well equilibrated in the low-density regime, which contrasts the AUG result. The few plasmas with very low density (<0.8x10^20m^-3) show an edge T_e pedestal formed well before L-H transition, however, no clear edge n_e or T_i pedestal emerged until after L-H transition. Another significant finding is that the low density limit for H-mode access moves to lower values of density when B_T is reduced. Scaling of P_th and local plasma edge conditions for H-mode access was examined over a wide range of plasma parameters in C-Mod divertor plasmas with ion grad-B drift in the favorable direction for H-mode access. The obtained local conditions were employed to test the L-H transition models based on the suppression of resistive-ballooning mode and drift-Alfven wave turbulence. A new model developed recently to predict H-mode access power was also tested

Speaker: Mr Yunxing Ma (MIT PSFC)

Poster Ma.Y_IAEA_poster_2012.pdf

154 EX/P2-07: Investigation of Plasma Rotation Alteration and MHD Stability in the Expanded H-mode Operation of KSTAR

Initial H-mode operation of the KSTAR has been expanded to higher normalized beta, betaN, and lower plasma internal inductance, li, moving toward design target operation. As a key supporting device for ITER, an important goal for KSTAR is to produce physics understanding of MHD instabilities at long pulse with steady-state profiles, at high betaN and over a wide range of plasma rotation profiles. Instability characteristics in the present expanded H-mode operational space, the influence of varied plasma rotation, and methods to access ITER-relevant low plasma rotation are presently investigated. Equilibria have reached new maximum values in key parameters betaN = 1.9, stored energy of 340 kJ with an energy confinement time of 171 ms in 2011. These results mark substantial progress toward the n = 1 ideal no-wall stability limit, most closely positioned at betaN = 2.5, li = 0.7. Rotating MHD modes are observed with perturbations having tearing parity as determined by Mirnov and ECE measurements. Modes with m/n = 3/2 are triggered during the H-mode phase at 0.5 < betaN < 1 but and do not substantially reduce plasma stored energy. In contrast, 2/1 modes to date are only observed when both the confinement and plasma rotation profiles are lowered after H-L back-transition, and mode locking creates a repetitive crash of betaN by more than 50%. A correlation is found between the 2/1 amplitude and local rotation shear from an X-ray imaging crystal spectrometer, and additionally, 2/1 modes appear to onset only below ~1,200 km/s/m of local rotation shear. Plasma rotation alteration by applied n = 1, 2 fields and the associated neoclassical toroidal viscosity (NTV) induced torque can be used as a rotation profile alteration tool, but also to study the collisionality dependence and steady-state behavior of NTV. Initial success in non-resonant alteration of the H-mode rotation profile in KSTAR were made by using an n = 2 applied field. Analysis is pending to determine if the observed stronger rotation damping by n = 1 braking is resonant or non-resonant. The implications of kinetic RWM stability using the measured KSTAR rotation profiles and their variations are presently being evaluated using the MISK code to further examine the initial theoretical guidance that showed unfavorable RWM stability for projected device target plasmas using projected rotation profiles.

Speaker: Mr Young-Seok Park (USA)

155 EX/P2-08: Long-pulse Stability Limits of ITER Baseline Scenario Plasmas in DIII-D

Long duration plasmas, stable to m/n=2/1 tearing modes (TM), with an ITER similar shape and an ITER similar value of I_p/aB_T have been demonstrated in DIII-D, evolving to stationary conditions with the most stable operating point at beta_N approximately 2. Lower beta_N, corresponding to an ITER baseline scenario 2 value of 1.8, led to a higher probability of m/n=2/1 tearing modes modes, which is the opposite of predictions from neoclassical tearing mode theory. These plasmas (delta t_duration less equal 7.5 s and less equal 11tau_R), without electron cyclo¬tron current drive (ECCD) for TM mitigation, have extended shorter pulse experiments in which the internal inductance was continually evolving [1] often until rotating m/n=2/1 TMs or locked modes occurred, which are a concern for ITER operation. Although long-pulse plasmas have reached sta¬tionary conditions, in some cases with similar programming m/n=2/1 TMs and locked modes limited the duration, indicating operation near stability limits. In addition to the plasmas described above, the use of ECCD, broadly deposited near q=3/2, allowed stable operation in plasmas with reduced torque which were otherwise found to be 2/1 TM unstable. We note that direct stabilization of 2/1 TMs was not attempted in these experiments. With one toroidal row of the DIII-D internal coil set (n=3 configuration) and broad ECCD for 2/1 TM mitigation, edge localized mode suppression with periods up to 1 s was observed (q_95=3.15) in plasmas with an ITER similar shape. This work was supported by the US Department of Energy under DE-FC02-04ER54698, DE-AC02-09CH11466, DE-FG02-04ER54761, and DE-FG02-08ER54984. [1] F. Turco and T.C. Luce, Nucl. Fusion 50 (2010) 095010.

Speaker: Mr Gary L. Jackson (USA)

156 EX/P2-09: First-Principles Model-based Closed-loop Control of the Current Profile Dynamic Evolution on DIII-D

Recent DIII-D experiments represent the first successful use of first-principles, model-based, full magnetic profile control in a tokamak. For ITER to be capable of operating in advanced tokamak operating regimes, characterized by a high fusion gain, good plasma confinement, magnetohydrodynamic stability, and a noninductively driven plasma current, for extended periods of time, several challenging control problems still need to be solved. For instance, setting up a suitable toroidal current density profile is key for one possible advanced operating scenario characterized by noninductive sustainment of the plasma current and steady-state operation. The control approach at the DIII-D tokamak is to create the desired current profile during the ramp-up and early flat-top phases of the plasma discharge and then actively maintain this target profile for the remainder of the discharge. The evolution in time of the current profile in tokamaks is related to the evolution of the poloidal magnetic flux profile, which is modeled in normalized cylindrical coordinates using a nonlinear dynamic partial differential equation referred to as the magnetic diffusion equation. This first-principles control-oriented model of the current density profile evolution in response to auxiliary heating and current drive systems [Neutral Beam Injection (NBI)], line-averaged density, and electric field due to induction, was developed and used to synthesize a combined feedforward + feedback control scheme to drive the current profile to a desired target profile. The model combines the magnetic diffusion equation with empirical correlations obtained at DIII-D for the temperature and noninductive current. Static and dynamic plasma response models were integrated into the design of the feedback controllers by employing robust, optimal, and backstepping control theories. A general framework for real-time feedforward + feedback control of magnetic and kinetic plasma profiles was implemented in the DIII-D Plasma Control System. Experimental results are presented to demonstrate the ability of the first-principles model-based feedback controllers to control the toroidal current density profile. This work was supported by the NSF CAREER award program (ECCS-0645086) and the US DOE under DE-FG02-09ER55064, DE-FG02-92ER54141 and DE-FC02-04ER54698.

Speaker: Mr Justin E. Barton (USA)

157 EX/P2-10:Non-inductive Plasma Start-up in NSTX Using Transient CHI

Transient Coaxial Helicity Injection (CHI) in the National Spherical Torus Experiment (NSTX) has generated toroidal current on closed flux surfaces without the use of the central solenoid. When induction from the solenoid was added, CHI initiated discharges in NSTX achieved 1 MA of plasma current using 65% of the solenoid flux of standard induction-only discharges. In addition, they have lower density, which is difficult to achieve by other means in NSTX, and a low normalized internal plasma inductance of 0.35, which is consistent with broad current profiles expected for high performance NSTX-U discharges. The Tokamak Simulation Code (TSC) has now been used to understand the scaling of CHI generated toroidal current with variations in the external toroidal field and injector flux. These simulations show favorable scaling of the CHI start-up process with increasing machine size. Tokamaks and spherical tokamaks have generally relied on a central solenoid to generate the initial plasma current. The inclusion of a central solenoid in a steady-state tokamak to provide plasma startup limits the minimum aspect ratio and increases the device complexity. For reactors based on the ST concept, elimination of the central solenoid is essential, making alternate methods for plasma start-up necessary. CHI is implemented in NSTX by driving current from an external source along field lines that connect the inner and outer lower divertor plates. The NSTX is now undergoing a major upgrade (NSTX-U) to increase the capabilities of its toroidal and poloidal field coils and to add a second neutral beam line. Analysis of the NSTX results shows that the amount of closed-flux current generated by CHI is closely related to the initially applied injector flux. On NSTX-U the available injector flux is about 340 mWb, considerably exceeding the 80 mWb in NSTX. Simulations with the TSC projects that it should be possible to generate 500 kA of closed-flux current with CHI in NSTX-U. At this current, the second more tangentially injecting neutral beam should be capable of providing sufficient current drive to ramp-up the plasma current. The results from TSC simulations show that CHI could be an important tool for non-inductive start-up in next-step STs. This work was supported by U.S. DOE contracts DE-FG02-99ER54519 and DE-AC02-09CH11466.

Speaker: Mr Roger Raman (USA)

158 EX/P2-11: Non-inductive Plasma Initiation and Plasma Current Ramp-up on the TST-2 Spherical Tokamak

To realize a compact spherical tokamak (ST) reactor, operation without the central solenoid must be demonstrated. In particular plasma current ramp-up from zero to a level required for fusion burn is crucial. Plasma initiation and current ramp-up in ST by waves in the lower-hybrid (LH) frequency range were demonstrated for the first time on TST-2. A combline antenna was used to inject RF power of ~ 100 kW at 200 MHz. Formation of a low current (~ 1kA, mainly driven by pressure gradient) ST configuration can be accomplished by waves in a broad frequency range (21 MHz to 8.2 GHz in TST-2). However, further current ramp-up (to ~ 10 kA, mainly driven by RF) is most efficient with a uni-directional traveling wave in the LH frequency range. Sufficient RF power must be supplied and the vertical field must be ramped up to maintain equilibrium. Plasma current ramp-up to 15 kA was achieved with 60 kW of net RF power. Soft X-ray emission in the direction of electron acceleration by RF wave was enhanced more strongly in the co-drive case (acceleration in the direction to increase the plasma current) compared to the counter-drive case. Hard X-ray spectral measurements showed that the photon flux is an order of magnitude higher and the photon temperature is higher in the co-current-drive direction than in the counter-current-drive direction (60 keV vs. 40 keV). These observations are consistent with acceleration of electrons by a uni-directional RF wave. The combline antenna excites vertical electric fields which match the polarization of the fast wave. There is evidence that the LH wave is excited nonlinearly, based on the frequency spectra measured by magnetic probes in the plasma edge region. While the pump wave at 200 MHz has a stronger toroidal component (fast wave polarization), the lower sideband has a stronger poloidal component (LH wave polarization). The time evolution indicates the tendency of the pump wave toroidal component to weaken when both sideband poloidal and toroidal components intensify. It is expected that the effectiveness of current drive would improve if the LH wave could be excited directly by the antenna. Two types of traveling-wave LH wave antennas will be tested on TST-2, a dielectric-loaded waveguide array, and an array of mutually coupled structure with the electric field polarized in the toroidal direction.

Speaker: Mr Yuichi Takase (Japan)

Poster IAEA_poster_takase.pdf

159 EX/P2-12: Steady State Operation Using Improved ICH and ECH for High Performance Plasma in LHD

The steady state operation (SSO) of high-performance plasma in LHD [1] has progressed since the last IAEA conference by means of a newly installed ICH antenna (HAS antenna [2], HAnd Shake type) and an improved ECH system. HAS antenna could control the launched parallel wave number and heated a core plasma efficiently. The heating power of steady state ICH and ECH exceeded 1 MW and 500 kW, respectively, and the higher-density helium plasma with minority hydrogen ions was sustained. Plasma performance improved; e.g., an electron temperature of more than 2 keV at a density of more than 2 x “10^19” “ m^-3” became possible for more than 1 min. Dipole phasing operation of the HAS antenna is better than that of monopole operation, and the monopole operation gives almost the same performance with the poloidal array antenna [1]. Three 77-GHz high-power gyrotrons were also installed for high-power ECH in LHD [3]. The frequency of 77-GHz is selected to heat the plasma core region for the wider plasma operation condition, and to increase sustainable plasma density to mitigate the high energetic ion population produced by ICRF wave. The injected power to plasma is finally absorbed by divertor plates, antenna side protectors and the chamber wall. The ratios of heat flow through various channels are estimated and about half of the heat flow goes to the divertor plates, and around 10% is goes to the ICRF antenna protectors. The non-uniform heat flow to the chamber wall decreased from 30% to 15% as the density increased. The particle balance during SSO was also analyzed. The ratio of the total supplied particles (helium and hydrogen) to the externally pumped particles is around 20, which indicates that wall pumping is a dominant particle sink during the SSO of 320 sec. The vacuum chamber works as a large particle sink in LHD. In the case of 54-min plasma operation in 2006 [1], the LHD chamber wall also worked as a particle sink even after the very long operation time. These experiences of steady state operation give us useful information for ITER and future fusion devices. [1] T. Mutoh, R. Kumazawa, T. Seki, et al., Nuclear Fusion47 (2007) pp1250-1257 [2] H. Kasahara,　K. Saito, et al., Plasma and Fusion Research, Vol.5, S2090-1-5 (2010) [3] T. Shimozuma, et al., Fusion Science and Technology, Vol.58, No.12, 2010, pp530-538

Speaker: Mr Takashi Mutoh (Japan)

Poster Mutoh_EX_P2-12.pdf

160 EX/P2-13: Validation of Off-axis Neutral Beam Current Drive Physics in the DIII-D Tokamak

DIII-D experiments on neutral beam current drive (NBCD) using the new tilted beamline have clearly demonstrated off-axis NBCD in agreement with modeling. Two of the eight neutral beam sources have been modified for downward vertical steering to provide significant off-axis current drive for AT scenario development. For validation of off-axis NBCD physics, the local NBCD profile was measured in H-mode plasma and compared with modeling under a range of beam injection and discharge conditions. The full radial profile of NBCD measured by the magnetic pitch angles from the motional Stark effect (MSE) diagnostic shows a clear hollow NBCD with the peak NBCD location at rho~0.45, which is in good agreement with the classical model calculation using the Monte-Carlo beam ion slowing down code, NUBEAM. Time evolution of the MSE signals is consistent with transport simulation with realistic current drive sources. The beam-stored energy estimated by equilibrium reconstruction and neutron emission data do not show any noticeable anomalous losses of NBCD and fast ions. The measured magnitude of off-axis NBCD is very sensitive with the toroidal magnetic field direction that modifies the alignment of the off-axis beam injection to the local helical pitch of the magnetic field lines. If the signs of the toroidal magnetic field and the plasma current yield the proper helicity, both measurement and calculation indicate that the efficiency is as good as on-axis NBCD because the increased fraction of trapped electrons reduces the electron shielding of the injected ion current. This dependency of the off-axis NBCD efficiency on the toroidal field direction is crucial to optimum use of the off-axis beams not only for DIII-D but also for ITER. A detailed NB and Electron Cyclotron Heating (ECH) power scan with variation of the ratio of beam injection energy to electron temperature (E_b/T_e) at fixed beta and with variation of beta at fixed E_b/T_e, around the anticipated ITER parameters, implies that ITER is not likely to suffer from the loss of NBCD efficiency due to additional transport from microturbulence. This work supported in part by U.S. Department of Energy under DE-AC05-00OR22725, DE-FC02-04ER54698, SC-G903402, DE-AC05-06OR23100 and DE-AC52-07NA27344.

Speaker: Mr Jin Myung Park (USA)

161 EX/P2-14: Non-inductive Current Start-up and Plasma Equilibrium with an Inboard Poloidal Field Null by Means of Electron Cyclotron Waves in QUEST

ECW Start-up scenarios in QUEST: Non-inductive current start-up using ECW (f=8.2 GHz, P_{rf}\ll 140 kW, O-mode, N_\parallel 0<0.4) has been investigated from a view point of multiple ECR interaction, large up-shift and auto resonance condition (N_\parallel~1) in QUEST. Due to large up-shift of N_\parallel at R_{1ce}, ECCD with mildly relativistic electrons moving in the wave momentum direction can be expected. The start-up rate dI_p/dt of 0.3-0.5 MA/sec was achieved within 0.2 s after rf injection. During this phase rapid build up of the energy spectrum of mildly relativistic electrons is observed. This fact strongly supports the ECR current drive scenario based on the relativistic resonace interaction. Equilibrium with an inboard poloidal field null: MHD equilibrium state with an inboard poloidal field null is characterized by high beta-p. Parameters are a=0.27 m, R0=0.79 m, \delta/a=0.4 and epsilon*beta-p=1.5, respectively. Here delta denotes the Shafranov shift. The Rs locates at ~ 0.5 m. During the flattop phase, Ip reached to 25 kA with a positive dependence of Bz and could be sustaiend for ~ 10 s. Loop voltage was less than 10 mV ,the line density 0.5E18 m^3 (< the cutoff density) and Te was measured < 100 eV by Thomson scattering. The value of Shafranov lamnda was kept ~5. Using observed HX temperature of ~ 50 keV and assumed density of the energetic electrons of ~ 10 % of density beta-p (=phot/poloidal mag. pressure) due to energetic electrons is consistent with that in MHD equilibrium. Relaxation oscillations of this equilibrium: High beta-p equilibrium oscillates at the frequency of ~ 20 Hz under some conditions for 10 – 20 kA. The slow rise and sharp drop in Ip are correlated to changes in density, Rs, shift and the HX energy spectrum emitted from the current carrying electrons. The out-of-phase relation in two energy windows(20-100 keV and 200-400 keV) of HXs suggests that electrons are preferably accelerated against the induced return electric field (~20 mV/m) during the Ip rise. Thus, confinement of the energetic electrons affects the stability of this equilibrium.

Speaker: Mr Hideki Zushi (Riam Kyushu University)

162 EX/P2-15: ECRH Pre-ionization and Assisted Startup in HL-2A Tokamaks

ECRH pre-ionization and assisted startup is foreseen in ITER, because the electric field applied for ionization and ramp-up of plasma current is limited to a very low value about 0.3 V/m. Many data for ECRH pre-ionization and assisted startup have been obtained from C-Mod, ASDEX Upgrade, DIII-D, JT-60U, JET and KSTAR. This paper presents the experiment results in HL-2A Tokamak from 2010 to 2012, with emphasis on the following two new and better results. 1) The minimum loop voltage for successful plasma establishment is reduced up to 1V, corresponding toroidal electric field of 0.1V/m, which is 3 times smaller than the ITER value of 0.3V/m, and smaller than the best value of 0.15V/m ever obtained from DIII-D. 2) Plasma can be established successfully with ECRH second harmonic X mode (X2) switching on before or after application of the reduced loop voltage, which presents important experiment data to the answer of the open issue addressed in ITPA joint experiment IOS-2.3. IOS-2.3 focuses on the breakdown-assist experiment results with X2 mode ECRH when ECRH power is launched after application of the reduced loop voltage, which is the situation operating in ITER during its commissioning phase, and so far never performed before. In HL-2A experiments, ECWwith the fundamental O mode (O1) or X2 is launched, and the toroidal injection angle can be changed from 00 to 200 in the equatorial plane. During 2010-2011 HL-2A experiment campaigns, the loop voltage can be reduced to 3V by O1 and X2 mode ECW, with minimum power 200 kW and 600kW, respectively. The effects of toroidal inclination, prefill pressure, wall conditioning and poloidal field null structure on X2 mode pre-ionization and assisted startup have been studied. During 2012 HL-2A experiment campaigns, the breakdown condition can be greatly improved with siliconization and lithiumization, and good results have been obtained. The loop voltage for successful plasma establishment is reduced up to 1V, and plasma can be established readily with X2 mode ECRH switching on 30 ms before or 30 ms after application the reduced loop voltage. The minimum ECRH power is about 500kW. The minimum loop voltage for successful pure ohmic breakdown is also presented, which is 3.4V.

Speaker: Mr Xianming SONG (China)

Poster songxm_IAEA_ECRH.pptx

163 EX/P2-16: Progress toward Steady-state Regimes in Alcator C-Mod

Over 1 MW of RF power in the lower hybrid (LH) frequency range (4.6 GHz) has been injected into C-Mod plasmas. Fully non-inductive discharges were sustained for 2-3 current relaxation time with n_{e} = 0.5x10^{20} m^{-3}, I_p = 0.5 MA and B_T = 5.4 T. Sawteeth are completely suppressed and modestly reversed shear plasmas, with q_0 ~ 2 and q_{min} ~ 1.5, were obtained. CD efficiency is in the range of 2.0-2.5x10^{19} A/Wm^2, consistent with what is assumed in ITER. Also, spontaneous formation of an ITB was observed during these discharges after the current profile relaxed. Accessing ITER-relevant steady-state regimes with f_{BS} ~ 50% in C-Mod requires increasing the density to ~ 1.5x10^{20} m^{-3} with T_{e0} ~ 5 keV [1]. Target plasmas with these parameters have been produced in C-Mod both by mode-converted ICRF heating as well as in I-Mode [2]. However, in low temperature (T_{e0} ~ 2 keV) Ohmically heated plasmas, LHCD efficiency drops precipitously as the density is increased above ~ 5x10^{19} m^{-3} even though this density is well below conventional limits set by either accessibility or parametric decay [3,4]. This falloff with density has been explored with two newly developed and independent simulations, ray-tracing and fullwave codes coupled with 3D Fokker-planck codes [5,6]. Both simulations indicate that wave interactions in edge plasmas including SOL can lead to the observed loss in efficiency, either by direct collisional absorption and/or by causing an up-shift in the parallel refractive index. The simulations also indicate that the loss in efficiency can be recovered by increasing the first-pass absorption. Experiments aimed at verifying this result have been carried out in mode-converted ICRF-heated He discharges with high T_{e0} (up to 5.3 keV). Good agreement with the simulations is found, verifying that low parasitic absorption occurs with stronger absorption. Base on this result, additional launcher was design to aim maximizing the velocity space synergy between two launchers. Implications for realizing steady-state regimes in C-Mod will be discussed. [1] P. T. Bonoli et al, NF, 40 1251 (2000) [2] D. G. Whyte et al, NF 50 105005 (2010) [3] G. M. Wallace et al, PoP 17 082508 (2010) [4] G. M. Wallace et al, NF 51 083032 (2011) [5] O. Meneghini et al, PoP 16, 090701 (2009) [6] S. Shiraiwa et al, PoP 18, 080705 (2011)

Speaker: Mr Syun'ichi SHIRAIWA (USA)

164 EX/P2-17: Resonant and Non-resonant type Pre-ionization and Current Ramp-up Experiments on Tokamak ADITYA in the Ion Cyclotron Frequency Range

Here we report the pre-ionization experiments carried out in ICRF range using poloidal type fast wave antenna, and 200 kW RF system at 24.8 MHz frequency which corresponds to the second harmonic resonance layer at the center of the vacuum vessel of tokamak ADITYA at 0.825 T. The experiments are carried out in different phases like only RF plasma, RF plasma in presence of toroidal magnetic field, RF pre-ionization with higher loop voltages and then at lower loop voltages by decreasing current and then at lower loop voltages with constant volts-seconds. In last phase the current ramp-up is carried out at lower loop voltages as well as with slow rise time to simulate the requirements of steady state tokamaks like SST-1. The toroidal magnetic field is varied from 0.825 T to 0.075 T, pressure is varied from 3 x 10^-5 torr to 8.0 x 10^-4 torr and RF power is varied from 20 kW to 120 kW. The diagnostics used are Langmuir probes, visible camera, spectroscopy, soft and hard X-ray detection techniques, diamagnetic loop and microwave diagnostics like interferometer and reflectometer. After plasma production at different magnetic fields, the pre-ionization experiments are carried out at different loop voltages to ramp up the current and we could ramp up current at all available loop voltages starting from 22 volts to 8 volts of the ohmic loop voltage to get normal plasma discharge of 90 kA and 90 ms duration. There exists a minimum value of the toroidal magnetic field (0.09T) below which no plasma spread is observed and also below 20 kW RF power we did not observe pre-ionization. It is observed that at lower loop voltages the current ramp-up is possible only in presence of pre-ionization.

Speaker: Mr Sanjay Kulkarni (India)

165 TH/P2-01: The European Transport Solver: an Integrated Approach for Transport Simulations in the Plasma Core

The “European Transport Solver” (ETS) is the new modular package for scenario simulations developed within the EFDA Integrated Tokamak Modelling (ITM) Task Force. It solves 1-D transport equations to which the geometry (2-D equilibrium), the transport coefficients and the sources are provided by stand alone modules coupled in a self consistent way to the transport solver through generalized data structures. It uses the KEPLER collaborative software environment to compose and manage the scientific workflows, where physics modules are built into the ETS workflows as precompiled "actors". The high level of modularity of KEPLER allows one to have several complex workflows solving similar problems (for instance, those can either solve for the electron density or the ion density). This paper presents the status of the ETS developments, the results on verification and validation of the package and its first physics applications.

Speaker: Mr Denis Kalupin (EU)

166 TH/P2-02: Progress in the Plasma Science and Innovation Center

Highlights of recent progress in the Plasma Science and Innovation (PSI) Center include adding reacting neutrals in the MHD model, providing capability for CAD description to grid generation to MHD simulation, incorporating energetic particles in extended MHD modeling, simulating 3D physics with Hall MHD, such as rotating magnetic field (RMF) current drive and inductive asymmetric current drive. The PSI Center is a collaborative effort to refine existing computational tools with the goal of improving computational predictability. The Center collaborates with experimental research groups to test the codes and to support the experiments. The Center refines primarily NIMROD and HiFi to have sufficient physics, boundary conditions, and geometry to be calibrated with experiments to achieve predictive capabilities. This paper describes some of the recent code advances, applications to experimental devices, and comparison to experimental data. The HIT-SI bow tie spheromak uses geometrically asymmetric injectors to inject helicity. From experimental data and Hall-MHD NIMROD simulations, the spheromak is formed and sustained by a combination of reconnection and quiescent dynamo drive. The HiFi code uses a 3D finite element spatial discretization that uses a multi-block grid, imported from a CAD description from the experimental design. The grid quality is determined using an a priori error estimator that identifies regions that need improved grid resolution. The TCSU field reversed configuration (FRC) investigates RMF current drive to generate and sustain an FRC. Three-dimensional Hall-MHD NIMROD simulations using experimental parameters show the generation of an FRC with toroidal magnetic field and size that compare well with experimental results. A reacting plasma model, which has a singly ionized plasma and a dynamic, neutral gas, that undergo ionization, recombination, and charge exchange reactions, has been implemented in HiFi and used to simulate the ELF experiment of an FRC plasma drifting through a background of initially static neutral gas. The FRC leading edge ionizes and imparts momentum to the neutral gas. By collaborating with a variety of experiments, the PSI Center is able to focus efforts on adding appropriate physics capabilities to existing fluid codes and thereby provide computational support and eventually predictive capability for experiments.

Speaker: Mr Uri Shumlak (USA)

167 TH/P2-03: Model Validation and Integrated Modelling Simulations for the JT-60SA Tokamak

JT-60SA will be at the forefront of the international fusion programme for many years, both before and during the D-T phase of the ITER operation. The preparation of its scientific programme is now progressing in the framework of a Japan-EU collaboration and integrate advances coming both from experiments on other tokamaks and theoretical developments. As for ITER and DEMO, integrated modelling of full discharges will be the main ingredient to perform this preparation effectively and on a rational and coherent basis. To this end, a coordinated Japan-EU modelling activity has started with the ambitious goal of providing predictive simulations of the main JT-60SA scenarios. The first milestone of this activity is the critical comparison and benchmark of Japanese and EU models and codes used for integrated tokamak modelling. The benchmark of the H&CD codes, in particular of NBI codes for the complex injector configuration of the JT-60SA machine will be discussed. The second milestone is the validation of the main models and simulation framework used in both Japanese and EU integrated modelling suites of codes. These include, e.g., energy and particle transport models, pedestal models, rotation sources and transport, synthetic diagnostics. It is assumed that simulations of JT-60SA scenarios should be based at least on experimental results of the two machines that are the most similar, for size and configuration: JT-60U and JET. On this basis, a validation exercise has been undertaken, involving: data exchange of reference JT-60U and JET shots, representing the main scenarios (H-mode, hybrid, advanced); predictive simulations of the reference shots with both Japanese and EU codes and models, with the aim of finding a unified modelling framework that works for both machines: this should give the maximum confidence for prediction of JT-60SA. The first results of this comparison will be presented and discussed, with particular emphasis on the transport model comparison. The third milestone is the predictive modelling of JT-60SA scenario, logically to be carried out after the previous steps are completed. Nevertheless, preparation of this activity has been done by simpler models, both 0-D and 0.5-D. These results and the methods used to obtain them will be presented and discussed.

Speaker: Mr Gerardo Giruzzi (France)

168 TH/P2-04: Turbulent Optimization in Stellarators & Tokamaks via Shaping

A method[1,2] has recently been developed for evolving toroidal configurations to ones with reduced turbulent transport, using the STELLOPT optimization code and the GENE gyrokinetic code. The potential for this method is now being explored and extended. The growing body of results has found that the effectiveness of the current “proxy” figure of merit Qprox used by STELLOPT to estimate transport levels depends on the class of toroidal device considered. The present proxy works well for quasi-axisymmetric stellarators and tokamaks, and modestly for quasi-helically symmetric designs, but not for the W7X quasi-omnigenous/quasi-isodynamic design. We are exploring the origin of this variation, and improving the dependence of the proxy on key geometric factors, extending it to apply to transport channels other than the ITG turbulence it was originally developed for. We are also examining incorporating GENE directly into STELLOPT to improve the turbulent Qprox , and the relative effectiveness of different search algorithms. To aid in these efforts, we have adapted STELLOPT to provide a new capability for mapping the topography of the cost function in the search space. [1] H.E. Mynick, N. Pomphrey, P. Xanthopoulos, Phys. Rev. Letters, 105, 095004 (2010). [2] H.E. Mynick, N. Pomphrey, P. Xanthopoulos, Phys. Plasmas, 18, 056101 (2011).

Speaker: Mr Harry Mynick (USA)

169 TH/P2-05: Modelling of Hybrid Scenario: from Present-day Experiments toward ITER

An attractive operating scenario for ITER has recently emerged that combines long plasma duration similar to the steady-state scenario, together with the reliability of the reference H-mode regime: the so-called ’hybrid’ scenario. Worldwide a significant experimental effort has been devoted to explore the operating space in present day tokamaks. This paper is an overview of the recent European modelling effort carried out within the Integrated Scenario Modelling group which aims at (i) understanding the underlying physics of the hybrid regime in ASDEX-Upgrade and JET, and, (ii) extrapolating them toward ITER. Six JET and two ASDEX-Upgrade hybrid scenarios performed under different experimental conditions have been simulated in an interpretative and predictive way in order to address the current profile dynamics and its link with core confinement, the relative importance of magnetic shear, s, and ExB flow shear on the core turbulence, pedestal stability and H-L transition. The correlation of the improved confinement with an increased s/q at outer radii observed in JET and ASDEX-Upgrade discharges is consistent with the predictions based on the GLF23 model applied in the simulations of the ion and electron kinetic profiles. Projections to ITER hybrid scenarios have been carried out focusing on optimization of the heating/current drive schemes to reach and ultimately control the desired q-profile with the ITER actuators and constraints. Firstly, access condition to the hybrid-like q-profiles during the current ramp-up phase has been investigated. Secondly, from the interpreted role of the s/q ratio, ITER hybrid scenario flat-top performance has been optimized through tailoring the q-profile shape and pedestal conditions. EPED predictions of pedestal pressure and width have been used as constraints in the interpretative modelling while the core heat transport is predicted by GLF23. Finally, model based approach for real-time control of advanced tokamak scenarios has been applied to ITER hybrid regime for simultaneous magnetic and kinetic profile control.

Speaker: Mr Xavier LITAUDON (France)

170 TH/P2-06: The Role of Convective Structures in the Poloidal and Toroidal Rotation in Tokamak

Mixed regimes consisting of coexistence large scale flows (H-mode, Internal Transport Barriers) and turbulence are expected to be the current state in ITER. The rotation, either spontaneous or induced, will play a major role in the quality of the confinment. We investigate the influence of the poloidal rotation on the toroidal flow. The efficiency of the sheared poloidal rotation to control the instabilities is considerably higher than that of the toroidal rotation but it is usually assumed that the poloidal rotation should be at the neoclassical level due to the damping induced by magnetic pumping. This is true if the drive of the poloidal rotation relies on the Reynold stress of a poloidally cuasi-symmetric turbulence. Much higher drive of the poloidal rotation is however provided by flows associated with the convective structures that can be generated in the plasma cross-section beyond a threshold in the plasma pressure gradient. This drive overcomes the damping due to the magnetic pumping and the poloidal rotation is sustained. Cells of convection consisting of closed, large scale flows can be spontaneously generated, triggered by streamers sustained by the baroclinic term able to generate vorticity. Similar to the Reyleigh-Benard first bifurcation (from purely conductive to convective regime), the onset is very fast and the drive exerted on the poloidal rotation leads to a fast time variation of the polarization radial electric field. This is sufficient to create a distinction between the phases (first half, second half) of bounce on a banana of trapped ions and, implicitely, leads to acceleration in the toroidal direction (F. Spineanu and M. Vlad, arXiv.org/pdf/1202.4426) The structure consisting of cells of convection, breaking the azimuthal symmetry, is not a stationary state, mainly because they will induce magnetic perturbations. The reconfiguration of the flow due to the magnetic reconnections will reinstall the conductive state even if the gradients are still favorable to a convective response. The intermittent generation of convective structures must then be treated as a stochastic process consisting of a random sequence of events. The effect on the (toroidal) angular momentum transport will be discussed on the basis of the fluctuating drive of the Reynolds stress (a doubly random process).

Speaker: Mr Florin Spineanu (Romania)

171 TH/P2-08: Fusion Power Production in Baseline ITER H-Mode Discharges

Previous studies of ITER steady state and hybrid scenarios [1,2] are extended to studies of ITER ELMy H-mode 15 MA scenarios. Results are obtained using the PTRANSP predictive integrated modeling code with time evolved boundary conditions and the plasma shape provided by the Tokamak Simulation Code (TSC). Current is ramped to 15 MA during first 100 sec and the central density is ramped to values in the range of 0.85x10^20 to 1.05x10^20 m-3. Impurities are 2% Be, Argon in the range 0.12% to 0.3% and 2% He3. Variation in density and in level of Argon impurity provides insight to the sensitivity of fusion power production to variation in these parameters. The dependence of confinement and the associated fusion power production on the level and mixture of beam and RF heating as well as on the choice of RF heating mixes are examined. Components of the ITER H-mode plasma current density (ohmic, bootstrap, beam and radio frequency) are shown as functions of plasma radius along with the integrated current contained within a given plasma radius. Also shown, for various plasma conditions, are the neutron density production as a function of radius and the rate at which neutrons are produced within a given radius. Various discharge conditions are studied in which the input power is reduced once alpha power heating is underway. Pedestal pressure is varied in the range of EPED1 [3] predictions to examine the sensitivity of fusion power production to the pedestal pressure. The PTRANSP code is used to compute temperature, magnetic q and toroidal rotation profiles using either the MMM7.1 or the GLF23 anomalous transport model, combined with neoclassical transport. Verification studies are carried out repeating PTRANSP simulations using the ASTRA code. The effects of low, reverse and strong magnetic shear on internal transport barriers are explored. In the PTRANSP simulation scans the same rotation velocity boundary condition will be employed for each discharge. Simulations are carried out turning off flow shear in order to separate the flow shear and magnetic shear effects on internal transport barriers. This is an important issue for ITER since it is expected that the toroidal rotation will be small. [1] A.H. Kritz, et al., Nucl. Fusion 51, 123009 (2011) [2] T. Rafiq, et al., Phys. Plasmas, 18, 112508 (2011) [3] P.B. Snyder, et al., Phys. Plasmas 16, 056118 (2009)

Speaker: Mr Arnold Kritz (USA)

172 TH/P2-09: New Results in the Theory of Non-Diffusive Heat Transport and Anomalous Electron-Ion Energy Coupling

We present new results in the theory of non-diffusive heat transport, with special emphasis on electron thermal transport. Two foci of this paper are a) the theory of the convective energy velocity (i.e. heat pinch) for both electrons and ions [1], and b) the theory of collisionless electron-ion energy coupling [2]. Both these topics are important for ITER, since in a burning plasma, alpha particles slow down through Coulomb collisions with the electrons. We emphasize that the heat pinch and collisionless transfer are physically distinct and independent processes, which can, in principle, co-exist in low collisionality regimes. For both ions and trapped electrons, an inward heat pinch is predicted for flat density profiles, while outward energy convection is predicted for steep density profiles. An analysis of various energy dissipation channels shows that collisionless energy transfer and the consequent ion heating will occur predominantly by quasilinear processes and through zonal flow dissipation. This implies that any putative ITER transport model must include the effect of turbulent heating and inter-species coupling, in addition to predicting the electron heat flux. [1] L. Wang and P. H. Diamond, Nucl. Fusion 51, 083006 (2011). [2] Lei Zhao and P. H. Diamond, U.S. Transport Task Force Workshop, San Diego, California, April 6-9, 2011.

Speaker: Ms Lu Wang (Republic of Korea)

173 TH/P2-10: Microtearing Mode Fluctuations in Reversed Field Pinch Plasmas

Improved confinement scenarios in RFP plasmas that reduce global tearing modes are expected to lead to plasmas where confinement is limited by microturbu­lence driven by gradients of pressure, density, and temperature. Because enhanced confine­ment regimes in MST yield temperature profiles with core gradients near the critical thresh­old for temperature-gradient driven instability, a linear analysis of temperature-gradient driven micro-instabilities in MST-like RFP equilibria is undertaken using toroidal gyrokinet­ics for beta values ranging from 0 to 10%. These simulations show that when the ratio of minor radius to temperature gradient scale length is greater than 3 - 4, MST plasmas are unstable to ITG at low beta and unstable to microtearing at high beta ~10%. The beta at which microtearing dominates ITG is 5%, with ITG becoming completely stable just above 10%. Theory shows that the higher critical beta for ITG stabilization, relative to tokamaks, is associated with the shorter scale lengths for magnetic curvature. At the MST-relevant beta of 10% the micro­tearing mode growth rate peaks at a poloidal wavenumber of = 1.4 inverse gyroradii. However, instability is strong even for low wavenumbers, where there is a growth rate 2-3 times that of ITG at its maximal wavenumber for zero beta. The growth rate remains large even for very low collisionality, with indications that different microtearing branches are associated with low and moderate collisionalities. With these growth rate values significant transport is expected. MST has several diagnostics that will access microturbulence spatial scales, including FIR interferometry/scattering, fast Thomson scattering, heavy ion beam probe, and material probes. Work is underway to prepare these diagnostics for electrostatic and magnetic turbulence measurements for model validation in high-performance plasmas.

Speaker: Mr Daniel Carmody (USA)

174 TH/P2-11: Tokamak Discharges with Electron Thermal Conductivity Closed to the Neoclassical One

At the development of plasma transport models it is very important to have the asymptotic transitions to the cases when transport coefficients are well described by some first principle based theories. For the ion component of tokamak plasma this is a transition to results of the neoclassical theory (in Ohmically heated high density discharges). Another stable opinion has been formed in characterizing the electron transport, as being principally anomalous. Conclusions of L.A. Artsimovich [1], who analyzed the first tokamak experiments, are contributed to this. He called the electron thermal conductivity of the minimal level as “pseudo-classical” due to “classical” behavior of its dependences on plasma parameters despite of the several times higher amplitude. This opinion was supported later by B.B. Kadomtsev [2] and by many other authors. However, these conclusions have been done as a result of simplified consideration of transport processes using only diagonal terms of transport matrix. In this work it is shown that, taking into account the off-diagonal terms of neoclassical transport coefficient matrix, one can reduce imaginary anomality of electron thermal conductivity coefficients in tokamaks. Presented results of predictive modeling of several representative tokamak discharges with the ASTRA transport code [3] show that profiles of electron and ion temperatures measured in these experiments can be well described in the frame of the neoclassical transport theory. Contributions of different channels of energy losses from the ion and electron plasma components are investigated. So, results of the modeling of transport properties of high density discharges for some tokamaks show that electron and ion heat transport in the plasma core in these regimes can be described using the complete matrix of the neoclassical coefficients. Therefore minimal coefficients of electron heat transport in tokamaks are not pseudo-classical” but the neoclassical ones. [1] Artsimovitch L.A., Nucl. Fus., 12 (1972), 215 [2] Kadomtsev B.B., Plasma Phys. Reports (Rus), V.9, N5 (1983), 938. [3] Pereversev, G.V., Yushmanov, P.N., Preprint IPP 5/98 2002, Garching. Germany.

Speaker: Mr Vladimir Leonov (Russian Federation)

Poster TH-P2-11_Leonov_V..pdf

175 TH/P2-12: Indications of Nonlocality of Plasma Turbulence

Nonlocality, which may manifest itself as a breaking of the favorable gyro-Bohm scaling of transport via phenomena such as avalanches or turbulence spreading and is an important threat to ITER operation has been considered and observed to various extents in simulations and experiments. Understanding of its details, is an essential challange for the transport community. Here, we present some indications in the form of characteristic observations from experiment, such as the recent results from the Tore Supra tokamak, using detailed high-resolution fluctuation diagnostics, that show the existence of mesoscale structures (Geodesic Acoustic Modes) and their dynamics and simulation results using the GYSELA code, developed by the IRFM-CEA, a full-f, global gyrokinetic code that simulates the ion scales but can also treat neoclassical physics, momentum transport with ripple etc. and finally discuss existing simple models that help understand, and underline the important features of non-locality of plasma turbulence. We discuss possible ways of verification of such models dealing with both turbulence and transport scales and in particular the mesoscale "interface" between these two hypothetically distinct scales. Turbulence spreading, and dynamics of the turbulence intensity field is a critical ingredient of this mixed range, which is also coupled to momentum transport via residual stress generation. It is also argued that nonlocality plays a key role in the L to H transition.

Speaker: Mr Ozgur Gurcan (LPP/Ecole Polytechnique/CNRS)

176 TH/P2-13: Physics Basis and Validation of MMM7.1 Anomalous Transport Module

The MMM7.1 anomalous transport module, recently installed in the PTRANSP code, is used to compute thermal, particle and toroidal angular momentum transport. The new MMM7.1 is documented and organized as a standalone module, which fully complies with the NTCC standards. The new transport module can be used with a standalone driver as well as within the integrated PTRANSP code and in other whole device modeling codes such as ASTRA, TSC, ONE-TWO, SWIM and FACETS. The MMM7.1 module includes a model for ITG, TE modes and MHD modes as well as a model for ETG modes and a model for drift resistive inertial ballooning modes. The model for transport driven by ITG/TE modes now includes the diffusion and radial convective pinch of toroidal angular momentum. It is found that this ITG/TE model can predict the observed intrinsic plasma rotation given a relatively small toroidal rotation at the plasma edge. The theoretical foundation of MMM7.1 is significantly advanced compared to the earlier MMM95 model. The new ITG/TE model in MMM7.1 module more accurately computes the suppression of transport at low and reverse magnetic shear. The MMM7.1 module is further improved by making better approximations to the structure of the eigenfunctions along field lines in order to include the effects of non-circular flux surfaces, finite beta, and Shafranov shift. Simulations using the MMM7.1 module compute the internal transport barrier that is experimentally observed in reversed magnetic shear discharges with sufficiently large toroidal angular rotation. The fluid approach which underlies the derivation of MMM7.1, while not as complete as a kinetic approach, allows prediction of the evolution of plasma discharges on an energy transport time scale. In this study, the theory based MMM7.1 is derived and simulations of DIII-D and JET tokamak discharges are presented to illustrate how various elements of the transport model influence the evolution of the tokamak plasma discharges. The discharges simulated include Ohmic, L-mode, H-mode plasmas and plasmas with co- and counter-rotations and plasmas with internal transport barriers. Results will be presented to understand the interaction between physical processes that influence transport in magnetically confined plasmas.

Speaker: Mr Tariq Rafiq (USA)

177 TH/P2-14: Predictive Transport Simulations Consistent with Rotation and Radial Electric Field Using TOPICS with OFMC

A toroidal momentum equation solver newly implemented in TOPICS calculates a temporal variation of a total toroidal momentum with an external torque input by OFMC. A solid scheme that can calculate parallel, poloidal and toroidal flows, and thus the radial electric field E_r is developed using the Matrix Inversion method. The coupling of the solver and the scheme with the aid of OFMC provides a means to investigate complex phenomena involving E_r and toroidal rotation. The simulations show the importance of inward pinch, the residual stress and the boundary condition for estimating toroidal rotation. Predictive capability we gained helps us seek the controllability for upcoming JT-60SA and ITER.

Speaker: Mr Mitsuru Honda (Japan)

178 TH/P2-15: Impact of Fusion Alpha Driven Current on the Magnetic Configuration of a Tokamak

The paper evaluates the influence of fusion alphas on burning plasmas in future tokamak reactors. In comparison to the relatively weak effect of charged fusion products (CFPs) in today’s devices, the substantially enhanced fusion power in burning tokamak plasmas may bring on a significant impact of CFPs on the equilibrium and bulk plasma parameters. Based on 3D Fokker-Planck modelling of DT fusion alphas we analyze here their effect on the plasma current and equilibrium in basic ITER scenarios. Particularly considering the peculiarities of the velocity and spatial distributions of confined alpha particles with energies exceeding a hundreds of keV, we calculate the poloidal profiles of the total alpha induced bootstrap current as well as of the fusion power deposition to bulk plasma electrons and ions. The present study demonstrates that fusion alphas are expected to induce an additional rotational transform of the magnetic field lines in reactor size tokamak plasmas. In reversed shear plasma scenarios the impact of the alpha driven current appears to be greater. While in the ITER steady state scenario alpha particles induce a 15% reduction of the safety factor in the core area, in the 2nd ITER Scenario with positive shear the safety factor reduction in the core is < 5% according to our calculations. Nevertheless, also such an alteration may reduce the core safety factor, which in Scenario 2 is only 1.02-1.04, to a value below 1, the crucial value for plasma stability. It is noted that, in spite of the low intensity of the total current driven by fusion alphas, the alpha driven current can play a role of a seed current for the bootstrap tokamak reactor. Evidently, the development of advanced plasma scenarios and research programs for ITER and future tokamak reactors should account for the effects of currents driven by fusion alphas.

Speaker: Mr Klaus Schoepf (Austria)

179 TH/P2-16: Nonlinear Modeling for Helical Configurations in Toroidal Pinch Systems

We present the current status of research in nonlinear modeling studies developed about the RFX-mod experiment, where several pinch configurations can be compared ranging from the self-organized helical RFP, to circular Tokamak passing through Ultra Low q ones, thus providing a flexible experiment in view of a future validation stage for several modeling tools. We focus presently on 3D nonlinear MHD modeling and nonlinear gyrokinetic tools for studies of helical configurations in the Reversed Field Pinch regime. We present new results of 3D nonlinear visco-resistive MHD simulations, which address the issue of stimulating -by suitable helical magnetic boundary condition- a helical QSH configuration with different toroidal periodicity (in particular a non resonant one) with respect to the self-organized one. On a parallel side, we present first MHD toroidal simulations with the PIXIE3D code. Preliminary indications show that QSH state is not stationary like it was in the cylindrical case. Instead, it is interrupted by crashes which may be reminiscent of the experimental phenomenology. A discussion of the magnetic topology obtained both as a result of stimulated nonresonant QSH regimes and of toroidal geometry effect will be presented, including results of code’s benchmarking of available tools like NEMATO and ORBIT. Concerning nonlinear gyrokintic studies, Ion temperature gradient (ITG) modes have been studied in the last years as a possible source of ion heat transport in the RFP. Such instabilities have revealed to be strongly stabilized compared to tokamak plasmas, due to the Landau damping acting in low-q configurations. However, for strongly outwardly peaked impurity profile – which is the case for RFX-mod plasmas – they can be more relevant. In this contribution, we address the nonlinear problem still in axisymmetric RFP geometries. A large set of 2-species gyrokinetic turbulence simulations with GS2 is presented, in order to compare the linear and nonlinear stability threshold. An up-shift in a/LTi,c is found, consistently with the picture given in tokamak plasmas (the so-called Dimits shift).

Speaker: Ms Susanna Cappello (Italy)

180 TH/P2-17: Non-diffusive Momentum Transport in JET H-mode Regimes: Modeling and Experiment

A systematic comparison of theoretical predictions for momentum transport in JET with experimental results has provided detailed insight into the physics of momentum transport, in particular non-diffusive transport. For this project, 400 representative experimental samples, selected from an extensive JET profile database with more than 1000 experimental profiles obtained over the entire baseline H-mode and hybrid operating domains, were used as input for the gyrokinetic code GKW. Linear and non-linear local calculations have allowed to quantify the contributions of diffusive transport, the Coriolis pinch and residual stresses to the overall transport and predict the expected angular velocity gradient R/L_omega for each of these samples. Direct comparisons of modeled and experimental data show that overall, the predicted Coriolis pinch account for approximately 70% of the observed non-diffusive contributions to R/L_omega. First results from linear calculations indicate that the remainder is consistent in magnitude with expectations for residual stresses. The strong contribution of the Coriolis pinch is due to the fact that most of these NBI-heated plasmas are strongly rotating, with Mach numbers in the range 0.05 to 0.35. Regressions were used to determine the overall parameter dependencies, both for the experimental and the theoretical dataset. Remarkably, the experimental scaling for non-diffusive transport matches the theoretical scaling for the Coriolis pinch for the three most relevant parameters R/L_n, q and f_t=sqrt(epsilon), the trapped particle fraction. A scaling with T_i/T_e in the experimental database and not characteristic of the Coriolis pinch, is suggestive of a contribution by residual stresses. Residual stresses are addressed in ongoing linear and non-linear simulations.

Speaker: Mr Henri Weisen (EU)

181 TH/P2-18: New Technique for the Calculation of Transport Profiles in Modulation Experiments

Transport codes provide a classical way to infer the profile of transport coefficients in fusion plasmas: assuming given functionals for the transport coefficient profiles, the free parameters are iteratively adjusted to best reproduce the measurements. This work introduces a new technique, the matrix approach (MA), which avoids any a priori constraint of the profiles, and computes them by simply inverting a 2D matrix, which also provides the uncertainty on the reconstruction for the case of modulation experiments. This is done by a controllable smoothing of the experimental data, instead of the ad hoc regularization of the profile of transport coefficients operated by transport codes. As a preliminary check, the MA was applied to already published JET data of momentum transport corresponding to three discharges that share the same initial equilibrium state. While an analysis of the data by a transport code suggests that all three cases share nearly the same transport coefficients, the MA rules this out, since the three uncertainty domains do not overlap at the various measurement positions. This suggested performing a similar analysis involving a residual stress on top of the advective and diffusive contributions to the flux. Then a single set of transport coefficients was found to be compatible with all three cases.

Speaker: Mr Dominique Escande (France)

Poster ESCST_IAEA_2012.pdf

182 TH/P2-19: Evolution of Ion Heat Diffusivity and Toroidal Momentum Diffusivity during Spontaneous ITB Development in HL-2A

Toroidal momentum torques generating V_t (toroidal flow) affect ITB evolution and decay in TFTR, JT-60U and DIII-D, suggesting that momentum inputs could offer a means for controlling barrier dynamics. An important question is whether it is possible to produce and control an ITB with inputting toroidal momentum in a discharge. To explore the role of external inputs of toroidal momentum on the development of ITBs we model the NBI heating discharges in HL-2A (R=1.64m, a=0.4m, B_t=2.8T, I_p=0.48MA) by using TRANSP. The modeled discharge: B_t=2.6T, I_p =300kA, line averaged density=2.4x10^19/m^3, single null divertor, H-mode boundary. As many experiments showed that optimized q-profile is one of the essential ingredients in establishing ITB, 0.5MW LH power in the current drive mode is injected at t=0.8s to control the current profile. Since the 2.45 GHz LH wave drives off-axis current in HL-2A, the q-profile with weak shear region extended to x~0.8, q_0>2.0 and q_a~5.3 is established after the current profile sufficiently relaxed (at t~1.1s). In order to control the toroidal momentum input, 3MW NBI (E=45keV) is injected tangentially with both co- and counter-injection during t=0.4-1.8s. The heat and momentum transport is calculated with a physics-based model GLF23. The transport model includes turbulence suppression mechanisms of EXB rotation shear. With appropriate neutral beam injection the nonlinear interplay between the transport determined gradients in V_t and T_i,e and the EXB flow shear (including q-profile) produces transport bifurcations, leading to a stepwise growing ITB. After its growth duration steady ITB is formed from t~1.35s. The ITB establishment is dependent on the toroidal momentum input. Quasi-steady ITBs can not be established unless the co-injected NBI power is in the range of 2.85MW to 2.4MW (correspondingly the counter-injected power is 0.15MW to 0.6MW respectively), which suggests that the NBI producing toroidal flow plays an important role in the ITB formation. The relationship between viscosity and ion heat transport in the ITB formation process is studied. Detailed examination of the evolution of toroidal momentum diffusivity and ion heat diffusivity in a region around the ITB development shows that the transport barrier of momentum develops more quickly and its enhanced confinement region extends further outward than that of heat.

Speaker: Mr Robert Budny (USA)

Poster Gao_24IAEA_poster2.ppt

183 TH/P2-21: Progress in the Theoretical Description and the Experimental Characterization of Impurity Transport at ASDEX Upgrade

The understanding of impurity transport from the wall to the center of the plasma and the identification of reliable methods to control central impurity accumulation are essential elements toward the achievement of practical fusion energy. A combination of theoretical and experimental research is required to identify the physical mechanisms from the theoretical standpoint, and to validate their impact on the measured impurity density profiles from the experimental side. In this contribution, advances in the theoretical description of turbulent impurity transport, particularly related to the inclusion of rotational effects, are presented. An analytical fluid model, which still captures the main elements of the physics, and linear and nonlinear numerical calculations with the gyrokinetic codes GKW and GS2 are presented and compared. In particular, GKW has the unique feature for a gyrokinetic code of including also centrifugal effects on turbulent transport. The impurity transport mechanisms produced by a radial gradient of the toroidal velocity and by centrifugal effects are singled out in the analytical calculations, and identified in the numerical results. These advances allow, in particular, the consistent prediction of the two dimensional impurity density distribution over the poloidal cross section. This more comprehensive theoretical description is also applied to the modelling of ASDEX Upgrade measurements of impurity density profiles. In neutral beam injection heated H-mode plasmas, central electron cyclotron heating is observed to increase the peaking of both the electron density and the boron density profiles. An inverse correlation is observed between the peaking of the boron density profile and the plasma toroidal rotation, as well as the boron logarithmic temperature gradient. The theoretical explanation of this phenomenology relates the boron response to the reduction of plasma rotation in the presence of central electron heating.

Speaker: Mr Clemente Angioni (Germany)

184 TH/P2-22: Transport Analysis of Oscillatory State for Plasma Dynamics in Helical Plasmas

The formation mechanism of transport barriers is important issue to realize improved confinement modes in toroidal plasmas. In helical plasmas, two kinds of the oscillation for the plasma quantities are experimentally observed. Firstly, the limit cycle phenomena in the temporal evolution of the electrostatic potential, namely the electric pulsation, have been observed in the core region. Related with the electric pulsation, the electron internal transport barrier is observed in the electron temperature profile. Therefore, the physical mechanism, which realizes the oscillatory plasma state, is critical for the study of improved confinement modes. Secondly, the density limit oscillation in the helical device was reported. The achievable limit of the density due to the radiation collapse has been studied, because the strong degradation of the confinement occurs if the radiation collapse happens. Dynamics of the radial structure for the plasma quantities are important for the study of the density limit. The temporally self-generated oscillation of the radial electric field has been shown as a simulation result in the core region of helical plasmas. The clear transport barrier in the radial profile of the temperature is obtained in the core region, which is associated with the oscillation of the radial electric field. Dynamics of the temperature gradient are shown during the self-generated oscillation to compare the experimental results. The temporal evolution of the density profile is newly included in a simulation when the radiative loss is calculated in the edge region of helical plasmas. Two kinds of the stationary states are studied. One is dominated by the transport loss and another is dominated by the radiative loss. The dependence of the achieved density limit on the heating power is derived, when the temporal evolution of the density is calculated. The dependence of the critical density on the heating power when the temporal evolution of the density profile is included alters compared with the case with the temporally fixed density profile. The multiple solutions of the radial electric field, which satisfy the ambipolar condition, are obtained in the edge region of helical plasmas. Progress in the study of the density limit oscillation in the edge region is reported.

Speaker: Mr Shinichiro Toda (Japan)

185 TH/P2-23: Turbulent Transport due to Kinetic Ballooning Modes in High-Beta Toroidal Plasmas

Turbulent transport in high-beta toroidal plasmas is investigated by means of a newly developed simulation code solving electromagnetic gyrokinetic equations combined with gyro-fluid equations for electrons. The new code allows simulations of turbulent transport at high-beta with smaller computational cost and less numerical difficulty than solving gyrokinetic equations for both electrons and ions which have disparate spatio-temporal scales. The code is applicable to a model configuration of Large Helical Device (LHD) experiments as well as tokamaks, and linear calculations show that kinetic ballooning modes (KBMs) are destabilized at high-beta. A nonlinear simulation for a tokamak plasma shows that heat transport due to KBM at high-beta (beta=2%) is significantly larger than that due to ion temperature gradient mode (ITG) driven turbulence at zero beta because of high KBM growth rate and weak zonal flow.

Speaker: Mr Akihiro Ishizawa (Japan)

186 TH/P2-24: Drift-kinetic Simulation Studies on Neoclassical Toroidal Viscosity in Tokamaks with Small Magnetic Perturbations

Effect of magnetic perturbations on neoclassical toroidal viscosity (NTV) in tokamaks is investigated by using a drift-kinetic delta-f simulation code, FORTEC-3D. The effect of magnetic perturbation on plasma transport and rotation is one of the important issues in recent tokamak experiments and forthcoming ITER project, in which the resonant magnetic perturbation (RMP) applied externally is a candidate to mitigate the edge localized modes. Finite NTV caused by RMPs damp the toroidal rotation, which might bring adverse effects on the other MHD instabilities such as resistive wall modes and locked modes. We have developed the simulation code to evaluate the NTV numerically. Compared with conventional analytic formulae which adopt many approximations, FORTEC-3D solves the drift-kinetic equation by following the exact guiding-center orbits including the finite-orbit-width effect and Coulomb collisions, and the viscosity is directly evaluated from the plasma distribution function, delta-f. In this paper, basic properties of NTV, such as the dependence of collisionality, ExB rotation speed, etc., are investigated by using the delta-f simulation. Benchmark tests of the code with analytic formulae of NTV are also reported. We have found some distinctive features which have not found in previous studies. First, we have investigated the NTV in the limit of zero ExB rotation. Single-helicity, (m,n)=(7,3) mode small perturbation field is applied to a tokamak configuration. It is found that conventional asymptotic limit formulae for the 1/nu and super-banana plateau regimes overestimate the NTV, while the combined analytic formula by J.-K. Park agrees well with FORTEC-3D in a wide range of collision frequency. Second, FORTEC-3D and Park’s formula has been benchmarked in finite-ExB rotation cases. A reference radial electric field (E_r) was given from the force balance relation, and the dependence was studied by magnifying the reference E_r amplitude. Two calculations agree well when E_r is close to the reference value. The peak of NTV at the resonant surface shrinks in both calculations as |E_r| becomes larger. However, as |E_r| increases, there appear other twin peaks of NTV on the both sides of the original resonant surface only in the delta-f simulation. It is anticipated that the finite-orbit-width effect can cause the difference from the analytic formula.

Speaker: Mr Shinsuke Satake (Japan)

Poster iaea2012-poster-satake.pdf

187 TH/P2-25: The European Integrated Tokamak Modelling Effort: Achievements and First Physics Results

The achievements and first physics results are presented of the European Integrated Tokamak Modelling Task Force (EFDA ITM-TF) effort, aiming at providing a standardized platform and an integrated modelling suite of validated numerical codes, for the simulation and prediction of a complete plasma discharge in any tokamak. The framework developed by the ITM-TF, based on a generic datastructure enclosing both simulated and experimental data, allowed for the development of sophisticated integrated simulations (workflows) for physics application. Those include the European Transport Solver (ETS), incorporating a sophisticated module for synergy effects between heating schemes, several equilibrium modules, pellets, impurities, neutrals, sawteeth and NTM modules, a variety of simple transport modules and neoclassical modules. The ETS workflows have been subject to an extensive verification and validation laying the foundations for the use of ETS for both predictive and interpretative transport simulations as well as scenario modeling on present devices and ITER. The equilibrium reconstruction and linear MHD stability simulation chain is being applied for production runs on several devices. In particular, an analysis of the edge MHD stability of ASDEX Upgrade type-I ELMy H-mode discharges and ITER hybrid scenario was performed, revealing the stabilizing effect of an increased Shafranov shift on edge modes. A successful benchmark among EC beam/ray-tracing codes (C3PO, GRAY, TORAY-FOM, TORBEAM, TRAVIS) has been performed in the ITM framework for an ITER case for different launching conditions from the Equatorial Launcher, showing good agreement of the computed absorbed power and driven current. Simulations performed within the ITM infrastructure with the turbulence code GEM for a JET hybrid discharge and the comparison of the simulated anomalous fluxes with TRANSP are presented, addressing in particular, the effect of the ExB shear on the thermal and particle confinement. Finally, recent developments on the integration and validation of synthetic diagnostics (fusion products, mse and reflectometry) on the ITM platform are shown.

Speaker: Ms Gloria Falchetto (France)

188 TH/P2-26: Advanced Confinement Regimes and Their Signatures

A unified theory [1, 2] has been developed for the modes that are excited at the edge of the plasma column and are an important signature of the advanced confinement regimes in which magnetically confined plasmas can be driven. In particular the so-called EDA H-Regime, the Elmy H-Regime and the I-Regime are considered and the modes that are identified theoretically have characteristics that are consistent with or have anticipated those of the modes observed experimentally for each of the investigated regimes. The phase velocities, the produced transport processes, the frequencies, the wavelengths and the consistency with the direction of spontaneous rotation are the factors considered for comparison with the relevant experiments. The phase velocity is in the direction of the ion diamagnetic velocity, in the plasma reference frame, for the Quasi-Coherent Mode that is present in the EDA H-Regime and is identified as a ballooning mode at Finite Larmor Radius marginal stability involving the effects of transverse ion viscosity and other dissipative effects [2]. Both in this regime and in the Elmy H-Regime impurities are driven towards the center of the plasma column. Instead, in the I-Regime the excited ``Heavy Particle'' mode [1] with a phase velocity in the electron diamagnetic velocity direction is shown to expel the impurities toward the plasma edge. The modes considered for the Elmy H-Regime are of ballooning kind, driven by the combined effects of the plasma pressure gradient and the magnetic curvature, are close to the relevant non-dissipative marginal stability, involve the effects of finite magnetic diffusivity and finite electron thermal conductivity and can have phase velocities in either directions [2]. Sponsored in part by the U.S. DOE. [1] B. Coppi and T. Zhou, Phys. Lett. A 375, 2916 (2011) and B. Coppi and T. Zhou, Phys. Plasmas, 19, 012302 (2012). [2] B. Coppi and T. Zhou, MIT(LNS) Report HEP 09/04 (2011), Cambridge, MA, to be submitted to Phys. Plasmas.

Speaker: Mr Bruno Coppi (USA)

189 TH/P2-27: Study of Neoclassical Toroidal Viscosity in Tokamaks with a delta-f Particle Code and Resonant Nature of Magnetic Braking

Non-axisymmetric magnetic perturbations can fundamentally change neoclassical transport in tokamaks by distorting particle orbits on deformed or broken flux surfaces. Understanding transport under non-axisymmetric magnetic perturbations is a critical issue for ITER and future fusion devices where non-axisymmetric perturbations are potentially important control elements to actively stabilize locked modes, edge localized modes, and resistive wall modes. Neoclassical transport with non-axisymmetry, often called Neoclassical Toroidal Viscosity (NTV) transport in tokamaks, is intrinsically non-ambipolar, and highly complex depending on parametric regimes. Thus a numerical approach is required to achieve its precise description. This paper reports the study of non-ambipolar transport and NTV torque with a new delta-f particle code, and the improved understanding of magnetic braking in perturbed tokamaks. Initial calculation of non-ambipolar particle flux clearly indicates the strong resonant nature of magnetic breaking, which is typically supposed as driven by non-resonant perturbations, while bootstrap current shows resonant or non-resonant features depending on collisionality. In addition, NTV torque is directly estimated by calculating anisotropic pressures and utilizing magnetic field spectrum method. Calculation results of NTV compared with theory and experiments will be reported, and detailed analyses on magnetic braking in tokamaks such as NSTX will be discussed. This work was supported by the US DOE Contract #DE-AC02-09CH11466.

Speaker: Mr Kimin Kim (USA)

190 TH/P2-28: Use of the 3D-MAPTOR Code in the Study of Magnetic Surfaces Break-up due to External Non-Axisymmetric Coils

We show how the outer magnetic surfaces can be broken up in a spherical tokamak, by breaking the axisymmetry using an inner tilted coil. The configuration chosen for this work is that of the MEDUSA small spherical tokamak, a small glass chamber device, which allows the introduction of such a coil. The simulation is carried out with the 3D-MAPTOR code developed by the authors. Given an initial condition for the magnetic field, it is integrated from the plasma current profile and the external currents, such as the toroidal and the vertical field. Poincaré maps along the toroidal angle and the image of the field, as seen from above can be plotted. The latter allows the identification of parameters for which the ripple effect is significant.

Speaker: Mr J. Julio E. Herrera-Velazquez (Mexico)

Poster Herrera_J.J.E._TH-p2-28_poster.pdf

191 TH/P2-29: Anisotropic Heat Transport in Integrable and Chaotic 3-D Magnetic Fields

Understanding heat transport in magnetized plasmas is a problem of fundamental interest in controlled fusion. Three issues make this problem particularly difficult: (i) The extreme anisotropy between the parallel (i.e., along the magnetic field), EMBED Equation.3 , and the perpendicular, EMBED Equation.3 conductivities; (ii) magnetic field lines chaos which in general precludes the construction of magnetic field line coordinates; and (iii) nonlocal parallel flux closures in the limit of small collisionality. Motivated by the extreme anisotropy encountered in fusion plasmas, in which the ratio EMBED Equation.3 may exceed 1010, we mainly focus on the study of purely parallel transport, i.e., EMBED Equation.3 , but also report on recent extensions of the method to incorporate perpendicular transport and sources. Me propose a Lagrangian-Green’s function (LG) that bypasses the need to discretize and invert the transport operators on a grid. The proposed method allows the integration of the parallel transport equation without perpendicular pollution, while preserving the positivity of the temperature field, and it is applicable to local and non-local parallel flux closures in integrable, weakly chaotic, and fully chaotic magnetic fields. We present applications of the method to: (i) Non-diffusive radial transport in fully chaotic magnetic fields, and fractal properties in weakly chaotic fields, (ii) parallel transport in the presence of internal transport barriers in reversed shear magnetic field configurations, and (iii) finite EMBED Equation.3 anisotropic transport. Concerning (i), we provide numerical evidence of non-diffusive effective radial transport (with both local and non-local closures) that casts doubts on the applicability of quasilinear diffusion descriptions. General conditions for the existence of non-diffusive, multivalued flux-gradient relations in the temperature evolution are derived. For (ii), we focus on the study of shearless Cantori, i.e. partial transport barriers located at the minimum of q. Finally, regarding (iii), we report on recent developments on the applications of the extended LG method to study transport in magnetic islands in the presence of sources with finite

Speaker: Mr Diego del-Castillo-Negrete (Fusion Energy Division. Oak Ridge National Laboratory)

192 TH/P2-30: Tomography of 2D Velocity-space Distributions from Combined Synthetic Fast-ion Diagnostics at ASDEX Upgrade

Collective Thomson scattering (CTS) and fast-ion D-alpha(FIDA) diagnostics measure 1D functions of the 2D fast-ion velocity distribution functions in magnetically confined plasmas. A single such 1D measurement is usually not sufficient to build accurate tomographies of 2D anisotropic fast-ion velocity distribution functions. But we can compute tomographies from several simultaneous 1D CTS and FIDA measurements. Such reconstructions contain salient features of the underlying 2D fast-ion velocity distribution functions as was shown theoretically in 2004 for two and three synthetic 1D CTS measurements. Since then FIDA measurements have been demonstrated, and several tokamaks have been equipped with multiple FIDA views. In 2012, a second CTS receiver and a second FIDA optical head have been installed on ASDEX Upgrade, so that four simultaneous 1D measurements of the 2D fast-ion distribution function are now possible if CTS and FIDA measurements are used together. We reconstruct 2D fast-ion velocity distribution functions from TRANSP/NUBEAM using combined synthetic 1D CTS and FIDA measurements. Here we develop a new reconstruction prescription that makes use of the recent idea of weight functions. In the past, reconstruction algorithms were made tractable by expansion of the 1D (synthetic) measurements as well as the 2D fast-ion velocity distribution functions into orthonormal sets of base functions. Exploiting CTS or FIDA weight functions we are lead to a simpler reconstruction prescription that is inherently tractable and obviates the use of such expansions. Our prescription is analogue to a tomography but in velocity space rather than in configuration space. Computed tomography in configuration space is widely used in medical imaging (X-ray, PET, MRI scanners) and also in nuclear fusion research. With our prescription we can build tomographies of the 2D velocity distribution function for any set of 1D fast-ion measurements obtained for any machine with multiple CTS or FIDA views or a mix of these. Applying our prescription to a set of real 1D fast-ion measurements will yield an entirely experimentally determined 2D fast-ion velocity distribution function that can be compared with simulations.

Speaker: Mr Mirko Salewski (Denmark)

4:10 PM Coffee Break

Transport: EX/2 & TH/1

Convener: Mr Myeun Kwon (Republic of Korea)

193 EX/2-1: Connections Between Intrinsic Toroidal Rotation, Density Peaking and Plasma Turbulence Regimes in ASDEX Upgrade

Recently, ASDEX Upgrade has made significant contributions to momentum transport studies thanks to the upgrade of the core charge exchange recombination spectroscopy system, which now produces much higher quality ion temperature and toroidal rotation profiles. This upgrade enabled the development of an intrinsic rotation database that contains over 200 observations. The edge rotation on AUG is always co-current, while the core rotation can be either co- or counter-current directed. The latter results in a null point in the profile at finite rotation gradient, which is clear evidence of a localized residual stress momentum flux. Moreover, the Mach number in the center of the plasma appears to be determined largely by the normalized gradient of the toroidal rotation at mid-radius, u’. This correlation holds for all of the observations regardless of plasma confinement regime or type of auxiliary heating. Further examination of the database reveals that u’ exhibits the strongest correlation with the local logarithmic electron density gradient, R/Lne: hollow rotation profiles coincide with peaked n_e profiles, while co-current rotation corresponds to low R/Lne. The known relationship between density peaking and plasma turbulence suggests a connection between the turbulence and the intrinsic rotation behavior as well. A study based on local linear gyro-kinetic calculations found good quantitative agreement between the predicted and measured values of u’ through the imposition of a finite tilting angle of -0.3 radians on the turbulent mode structure. The mechanism expected to produce such a tilting is a combination of ExB and profile shearing residual stress. These database results are also consistent with observations of residual stress in non-intrinsic rotation scenarios. Flat to hollow rotation profiles are observed concomitant with peaked electron density profiles when sufficient ECRH power is added to NBI heated H-modes causing the turbulent regime to transition from ITG to TEM. Momentum transport analyzes of these plasmas show that the observations can only be explained by the presence of a core localized, counter-current directed, residual stress induced torque of the same order of magnitude as the applied NBI. These results have important implications for torque modulation experiments, which often assume that the residual stress is negligibly small.

Speaker: Ms Rachael McDermott (Germany)

Slides EX-2-1_McDermott_R..pdf

194 EX/2-2: A Unified Explanation of Rotation Reversals, Confinement Saturation and Non-Diffusive Heat Transport in C-Mod Ohmic Plasmas

Recently, the connection among rotation reversals, energy confinement saturation (the transition between the LOC and saturated Ohmic confinement, SOC, regimes) and changes in underlying turbulence has been demonstrated. Examination of the rotation reversal results and a large body of confinement saturation observations suggests that there is a critical value of the collisionality where these effects transpire. Also occurring with the rotation reversals and the LOC/SOC transition is a saturation of the electron density profile peaking. These results may be unified with the following ansatz: at low collisionality in the LOC regime, the underlying turbulence is dominated by trapped electron modes and the rotation is directed co-current; at high collisionality in the SOC regime, ion temperature gradient modes prevail, the rotation is counter-current and the density profile peaking saturates. There are two other phenomena which appear to be related and occur at the LOC/SOC transition: a transformation from non-diffusive to diffusive heat transport and a change from symmetric up/down edge impurity density profiles to up/down asymmetric. Heat transport was investigated by means of rapid edge cooling from impurity injection by laser blow-off, and following the electron temperature profile evolution from electron cyclotron emission. In the high density in the SOC regime, there is ‘normal’ diffusive heat transport, with a drop in the core temperature lagging the edge cooling by about an energy confinement time. Also with SOC, the core rotation is counter-current, and there is a significant up/down edge impurity density asymmetry. At low density in the LOC regime, the core electron temperature increases (on a faster time scale) following the edge cooling, indicating the workings of a convective heat pinch or transient ITB. The core rotation with LOC is co-current and the edge impurity density profile is up/down symmetric. Rotation reversal, the transformation from non-diffusive to diffusive heat transport, the switch of edge impurity density profiles from up/down symmetric to asymmetric and changes in turbulence have all been observed dynamically during a single discharge with a density ramp to change the collisionality. These empirical results unify a large body of previously seemingly unrelated phenomena.

Speaker: Mr John Rice (USA)

Slides iaea12rice.pdf

195 TH/1-1: Study of Toroidal Flow Generation by the ICRF Minority Heating in the Alcator C-Mod Plasma

Important role of the plasma flow and its shear in the transport improvement is suggested by many experimental observations. The spontaneous toroidal flow has been observed during ICRF heating with no direct momentum input in many devices. Especially, in the Alcator C-Mod plasma, the spontaneous toroidal flow and ITB formation have been investigated intensively in the ICRF heating plasma and they found that the ITB plasma (ITB foot is located near r/a = 0.5.) was obtained when the ICRF resonance location was placed at well off the magnetic axis, near |r/a| = 0.5. We study the toroidal flow generation by the ICRF minority heating in the Alcator C-Mod plasma using GNET code, which can solve a linearized drift kinetic equation for minority ions including complicated orbits of accelerated energetic particles. The obtained steady state distribution of energetic minority ions is analyzed and the radial profile of the averaged toroidal flow of minority ions is evaluated. In our previous study we have found that a co-directional toroidal flow, which direction is consistent with experimental observations, is generated outside of the RF wave power absorption region and that the toroidal precession motion of energetic tail ions plays an important role in generating the averaged toroidal flow. In order to make clear the relation between the ICRF driven flow and the ITB formation we investigate the resonance location dependence of the toroidal flow profile changing the resonance location from r/a=-0.6 to +0.6 on the equatorial plane. It is found that a co-directional toroidal flow of the minority ion is generated outside of the RF wave power absorption region and that the maximum averaged velocity of the minority ion reaches about 300km/s, which is more than five times bigger than the experimentally observed bulk velocity. We consider that the energetic minority ion can drive the toroidal flow of the bulk plasma to the observed velocity level. When we shift the resonance location to the out side of |r/a| = 0.5 the opposite direction toroidal flow is enhanced near the central region and the velocity shear is increased. This suggests a role of the ICRF driven flow on the experimentally observed ITB formation during ICRF heating in the Alcator C-Mod plasma.

Speaker: Mr Sadayoshi Murakami (Japan)

Slides IAEAFEC2012_TH1-1_SMurakami_v23.pptx

196 EX/2-3: ECRH Effects on Toroidal Rotation: KSTAR Experiments and Gyrokinetic Simulations of Intrinsic Torque at ITG - TEM Transitions

Toroidal rotation is important for control of stability and transport in tokamaks. Intrinsic rotation is self-generated by ambient turbulence via the non-diffusive residual stress, which motivates the question of how macroscopic rotation profiles will evolve in response to changes in the ambient micro-turbulence. One ‘control knob’ for the micro-turbulence population is the heating mix of NBI and ECRH. The change in rotation to counter-direction by ECH in KSTAR is explained by the turbulence change from ITG to CTEM. We investigate the effect of ECRH heating on NBI-driven toroidal rotation profiles in L-mode and H-mode discharges in KSTAR tokamak. 1.3 MW of NBI is injected in the co-current direction and 350 kW of ECRH are applied. The ion temperature and toroidal rotation are measured with high resolution XICS and CES. NBI in the co-current direction drives peaked rotation profiles with (H-mode) and without (L-mode) a pedestal. Dramatic decreases in the core toroidal rotation values are observed when on-axis ECRH is added to H-mode. These increments delta(Vtor)/Vtor ~ -30% indicate the presence of on ECH-induced counter-current torque acting in the discharge core. We note that, for steady state with same external torque and boundary condition, the change of the radial gradient of plasma rotation implies the change of residual stress. Interestingly, edge and pedestal rotation velocities in H-mode are nearly unchanged. We explore the viability of the ITG--TEM transition as an explanation of the observed change in the sign of the core intrinsic torque. The global gyrokinetic code gKPSP was used for the study. We performed ITG and TEM simulations at values of eta_i = 3.1 (i.e. ITG) and eta_i = 1.0 (i.e. TEM), respectively. Note that the low value of eta_i for the CTEM case is qualitatively consistent with the reduction in core grad(Ti) observed with ECRH in KSTAR. Results show that the residual stress changes sign as ITG--TEM transition occurs, indicating a change in the direction of the net wave energy density flux. Direct simulations also reveal a mean macroscopic profile reversal at ITG--TEM transition, thus confirming the overall consistency of the argument. Also we will perform nonlinear gyrokinetic simulations to calculate the strength of intrinsic torque reversal at ITG--TEM transition and compare the simulation results with experiments.

Speaker: Mr Yuejiang SHI (Republic of Korea)

Slides EX-2-3_Shi_Y..pptx

197 EX/2-4: Dependence of Heat Transport and Confinement on Isotopic Composition in Conventional H-mode Plasmas in JT-60U

Dependence of heat transport on isotopic composition is investigated in conventional H-mode plasmas for the application to ITER. The identical profiles of electron density, electron temperature and ion temperature are obtained for hydrogen and deuterium plasmas while the required power becomes clearly larger for hydrogen, resulting in the reduction of the heat diffusivity for deuterium. The result of the identical temperature profiles in spite of different heating power suggests that the characteristics of heat conduction differs essentially between hydrogen and deuterium even at the same scale length of temperature gradient. On the other hand, the edge stability is improved by increased total poloidal beta regardless of the difference of the isotropic composition.

Speaker: Mr Hajime Urano (Japan)

Slides EX-2-4_Urano_H..pptx

198 EX/2-5: Extension of Operational Regime in High-Temperature Plasmas and the Dynamic-Transport Characteristics in the LHD

Realization of high-T_i plasmas is one of the most important issues in helical plasmas, which have an advantage for steady-state operation comparison with tokamak plasmas. Since 2010, newly installed perpendicular-NBI with the beam energy of 40 keV has been operational in the Large Helical Device (LHD) and the total-heating power of perpendicular-NBIs increased from 6 MW to 12 MW. Such low-energy NBIs are effective for ion heating and enabled us to achieve a higher T_i than that obtained previously. In the last experimental campaign, ICRF-discharge cleaning was adopted to reduce particle recycling from the wall. As a result, NBI-heating-power profile became peaked and the density-normalized ion heating power in the core region increased by 18% and the central T_i of 7 keV, which is the world’s highest value for helical devices, was successfully achieved as the new record in the LHD. In the LHD, high-T_i plasmas have been realized in combination with a carbon pellet. The kinetic-energy confinement was improved by a factor of 1.5 after the pellet injection. In the high-T_i phase, a flat or hollow profile in the electron density has been observed. This is the different characteristics from PEP mode investigated in Tokamaks. After the pellet injection, the central T_i, the gradient of T_i and that of the toroidal-flow velocity at the core region clearly increased indicating the formation of the ion-internal-transport barrier. In the high-T_i phase, reduction of the thermal diffusivity over the wide region was observed. In the core region, the time constant of the improvement of the ion-heat transport was found to be larger than that in the peripheral region. The toroidal-momentum transport was also improved accompanied with the reduction of the thermal diffusivity and the Prandtl number ignoring the intrinsic torque was close to unity. However, the confinement improvement was temporal and the gradient of T_i gradually decreased. Similarly, the toroidal-momentum transport went back to the low-confinement state in the latter phase of the discharge. Decrease of the negative radial electric field and increase of the density fluctuation were also observed in the phase.

Speaker: Mr Hiromi Takahashi (Japan)

Slides IAEA_presentation_HTakahashi_EX2-5.pdf

ITER Event

Wed, October 10

Overview: Inertial & Magnetic Fusion: OV/5

Convener: Ms Sybille Guenter (Germany)

199 OV/5-1: Dynamics of Flows and Confinement in the TJ-II Stellarator

Operation with Li coated wall is the basis for a significant improvement in the performance of TJ-II, and lies behind the findings in this overview, related to the role of flows in confinement improvement. Specific progress has been performed in the use of Li as alternative to solid plasma facing materials for future fusion devices. Recently a liquid lithium limiter (LLL) based on the Capillary Porous System (CPS) has been installed in TJ-II and first results will be reported. Although TJ-II presents a strong damping, the simulations predict that the presence of an ambipolar radial electric field as well as turbulence driven flows provide driving mechanisms for mean and fluctuating flows that will cause long range toroidal correlation whose typical frequencies are in agreement with the experiment. The transitions to improved confinement are accompanied by an amplification of long-range correlation in the plasma potential, which is a footprint of zonal flows. The amplitude of these structures have been seen to modulate the particle transport into the SOL for the first time. We also investigate the relation between the zonal flows and the turbulent flux of particles and momentum via the Reynolds and Maxwell stresses as well as suprathermal particles. Suprathermal ion can contribute with significant energy content, with poloidal rotation up to 2–5 higher than the thermal component even the ECRH regime. Taking advantage of the flexibility of TJ-II, low order rationals are introduced in the plasma, which helps to transit from L to H mode. The influence of the so developed MHD modes on transport is investigated showing a bursty behaviour and evidence of radially propagating events. During L-H transitions, an oscillating low frequency non-damped sheared flow appears in the edge prior to the change to H mode, which presents a predator-pray relation with the turbulence. The spatial evolution of this turbulence-flow shows both radial outward and inward propagations. These results show the need to study L-H transition within a 1-D spatio-temporal framework. The dynamical coupling between density gradients and particle transport has been investigated and compared in the plasma boundary of different tokamaks (JET, ISTTOK) and stellarator (TJ-II), showing that the size of turbulent events is minimum in the proximity of the most probable density gradient.

Speaker: Mr Enrique Ascasibar (Spain)

Slides OV_5_1_Sanchez.pdf

200 OV/5-2Ra & OV/5-2Rb: Overview of Results from the MST Reversed Field Pinch Experiment; Overview of the RFX Fusion Science Program

OV/5-2Ra: Overview of Results from the MST Reversed Field Pinch Experiment This overview of results from the MST program summarizes physics important for the advancement of the RFP as well as for improved understanding of toroidal magnetic confinement in general. Evidence for the classical confinement of ions in the RFP is provided by analysis of impurity ion transport. With inductive current profile control, the test-particle diffusivity for ions in a stochastic magnetic field is reduced below the classical transport level. (The neoclassical enhancement of radial transport is negligible in the RFP.) Carbon impurity measured by CHERS reveals a hollow profile and outward particle convection. Modeling of classical transport agrees with the profile evolution, and temperature screening explains the hollow profile. Classical confinement is also observed for energetic ions created by 1 MW NBI. The energetic ion confinement is consistent with classical slowing-down and ion loss by charge-exchange. The first appearance of Alfven eigenmodes and energetic particle modes by NBI in a RFP plasma are obtained. MST plasmas robustly access the quasi-single-helicity state that has commonalities to the stellarator and “snake” formation in tokamaks. The dominant mode grows to 8% of the axisymmetric field strength, while the remaining modes are reduced. Energy confinement is improved as a result. Predictive capability for tearing mode behavior has been improved through nonlinear, 3D, resistive MHD computation using the measured resistivity profile and Lundquist number, which reproduces the sawtooth cycle dynamics. New two-fluid analysis that includes Hall physics and gyro-viscosity has established a new basis for understanding physics beyond a single-fluid model. Nonlinear two-fluid (NIMROD) computation reveals coupling of parallel momentum transport and current profile relaxation. Large Reynolds and Maxwell stresses, plus separately measured kinetic stress, indicate an intricate momentum balance and possible origin for MST’s intrinsic plasma rotation. Microturbulence from drift-wave-like instabilities might be important in the RFP when magnetic fluctuations are reduced. New gyrokinetic analysis indicates that micro-tearing modes can be unstable at high beta, with a critical gradient for the electron temperature that is larger than for tokamak plasmas by roughly the aspect ratio. Supported by US DoE and NSF. OV/5-2Rb: Overview of the RFX Fusion Science Program With a program well-balanced among the goal of exploring the fusion potential of the reversed field pinch (RFP) and that of contributing to the solution of key science and technology problems in the roadmap to ITER, the European RFX-mod device has produced a set of high-quality results since the last 2010 Fusion Energy Conference. RFX-mod is a 2 MA RFP, which can also be operated as a tokamak and where advanced confinement states have 3D features studied with stellarator tools. Self-organized equilibria with a single helical axis and improved confinement (SHAx) have been deeply investigated and a more profound understanding of their physics has been achieved. First wall conditioning with Lithium provides a tool to operate RFX at higher density than before, and application of helical magnetic boundary conditions favour stationary SHAx states. The correlation between the quality of helical states and the reduction of magnetic field errors acting as seed of magnetic chaos has been robustly proven. Helical states provide a unique test-bed for numerical codes conceived to deal with 3D effects in all magnetic configurations. In particular the stellarator equilibrium codes VMEC and V3FIT have been successfully adapted to reconstruct RFX-mod equilibria with diagnostic input. The border of knowledge has been significantly expanded also in the area of feedback control of MHD stability. Non-linear dynamics of tearing modes and their control has been modelled, allowing for optimization of feedback models. An integrated dynamic model of the RWM control system has been developed integrating the plasma response to multiple RWMs with active and passive conducting structures (CarMa model) and with a complete representation of the control system. RFX has been operated as a tokamak with safety factor kept below 2, with complete active stabilization of the (2,1) Resistive Wall Mode (RWM). This opens the exploration of a broad and interesting operational range otherwise excluded to standard tokamaks. Control experiments and modelling led to the design of a significant upgrade of the RFX-mod feedback control system to dramatically enhance computing power and reduce system latency. The possibility of producing D-shaped plasmas is being explored.

Speaker: Mr John Sarff (USA)

Slides OV-5-2_Sarff_J..pdf

201 OV/5-3: Theory of Ignition, Burn and Hydro-equivalency for Inertial Confinement Fusion Implosions

Recent advances in the theory of ignition and burn for inertial confinement fusion are presented and related to the experimental observables of the current indirect-drive ignition campaign on the National Ignition Facility (NIF) and the direct-drive implosion campaign on the OMEGA laser. The performance parameter currently used for the ignition campaign (the Experimental Ignition Threshold Factor or ITFX) is related to the well-known Lawson criterion. Hydro-equivalent curves are derived and used to extrapolate current results from OMEGA to future direct-drive ignition experiments on the NIF. The impact of laser-plasma instabilities, hot electron and radiation preheat on the hydrodynamic scaling is discussed. Remedies to mitigate the detrimental effects of laser-plasma and hydrodynamic instabilities are presented. It is also shown that ignition through a late shock launched at the end of the laser pulse (shock ignition) may be possible on the NIF at sub-megajoule energies.

Speaker: Mr Riccardo Betti (USA)

Slides Betti_IAEA12.pdf

202 OV/5-4: Multimodal Options for Materials Research to Advance the Basis for Fusion Energy in the ITER Era

Sustained worldwide efforts on fusion energy research have led to substantial improvements in understanding of plasma physics and fusion technology issues. Several options with varying degrees of technological risk are being contemplated for the next major fusion energy device that will be constructed after ITER begins operation. These options include a variety of plasma confinement configurations, coolants, tritium breeding materials, power conversion systems, and operating temperatures. In many cases, variations in the aggressiveness of the design parameters for next-step devices are associated with uncertainties in the performance of the materials systems to be used in the divertor, first wall and breeding blanket, and tritium extraction and power conversion systems. In order to reduce some of these uncertainties and to assist in the selection of the most appropriate design concept(s) that meet national needs, well-coordinated international fusion materials research on multiple fundamental feasibility issues can serve an important role during the next ten years. There are two inter-related overarching objectives of fusion materials research to be performed in the next decade: 1) understanding materials science phenomena in the demanding DT fusion energy environment, and 2) Using this improved understanding to develop and qualify materials to provide the basis for next-step facility construction authorization by funding agencies and public safety licensing authorities. There are several important fundamental materials questions that should be resolved soon due to their potential major impact on next-step fusion reactor designs, including radiation effects on mechanical properties, structural stability and tritium permeation and trapping. An overview will be given of the current state-of-the-art of major materials systems that are candidates for next-step fusion reactors, including a summary of existing knowledge regarding operating temperature and neutron irradiation fluence limits due to high temperature strength and radiation damage considerations, coolant compatibility information, and current industrial manufacturing capabilities. The critical issues and prospects for development of high performance fusion materials will be discussed along with recent research results and planned activities of the international materials research community.

Speaker: Mr Steven Zinkle (USA)

Slides OV5_4_Zinkle_IAEA_FEC.pdf OV5_4_Zinkle_IAEA_FEC.pptx

Poster: P3

203 EX/2-1: Connections Between Intrinsic Toroidal Rotation, Density Peaking and Plasma Turbulence Regimes in ASDEX Upgrade

Recently, ASDEX Upgrade has made significant contributions to momentum transport studies thanks to the upgrade of the core charge exchange recombination spectroscopy system, which now produces much higher quality ion temperature and toroidal rotation profiles. This upgrade enabled the development of an intrinsic rotation database that contains over 200 observations. The edge rotation on AUG is always co-current, while the core rotation can be either co- or counter-current directed. The latter results in a null point in the profile at finite rotation gradient, which is clear evidence of a localized residual stress momentum flux. Moreover, the Mach number in the center of the plasma appears to be determined largely by the normalized gradient of the toroidal rotation at mid-radius, u’. This correlation holds for all of the observations regardless of plasma confinement regime or type of auxiliary heating. Further examination of the database reveals that u’ exhibits the strongest correlation with the local logarithmic electron density gradient, R/Lne: hollow rotation profiles coincide with peaked n_e profiles, while co-current rotation corresponds to low R/Lne. The known relationship between density peaking and plasma turbulence suggests a connection between the turbulence and the intrinsic rotation behavior as well. A study based on local linear gyro-kinetic calculations found good quantitative agreement between the predicted and measured values of u’ through the imposition of a finite tilting angle of -0.3 radians on the turbulent mode structure. The mechanism expected to produce such a tilting is a combination of ExB and profile shearing residual stress. These database results are also consistent with observations of residual stress in non-intrinsic rotation scenarios. Flat to hollow rotation profiles are observed concomitant with peaked electron density profiles when sufficient ECRH power is added to NBI heated H-modes causing the turbulent regime to transition from ITG to TEM. Momentum transport analyzes of these plasmas show that the observations can only be explained by the presence of a core localized, counter-current directed, residual stress induced torque of the same order of magnitude as the applied NBI. These results have important implications for torque modulation experiments, which often assume that the residual stress is negligibly small.

Speaker: Ms Rachael McDermott (Germany)

204 EX/2-2: A Unified Explanation of Rotation Reversals, Confinement Saturation and Non-Diffusive Heat Transport in C-Mod Ohmic Plasmas

Recently, the connection among rotation reversals, energy confinement saturation (the transition between the LOC and saturated Ohmic confinement, SOC, regimes) and changes in underlying turbulence has been demonstrated. Examination of the rotation reversal results and a large body of confinement saturation observations suggests that there is a critical value of the collisionality where these effects transpire. Also occurring with the rotation reversals and the LOC/SOC transition is a saturation of the electron density profile peaking. These results may be unified with the following ansatz: at low collisionality in the LOC regime, the underlying turbulence is dominated by trapped electron modes and the rotation is directed co-current; at high collisionality in the SOC regime, ion temperature gradient modes prevail, the rotation is counter-current and the density profile peaking saturates. There are two other phenomena which appear to be related and occur at the LOC/SOC transition: a transformation from non-diffusive to diffusive heat transport and a change from symmetric up/down edge impurity density profiles to up/down asymmetric. Heat transport was investigated by means of rapid edge cooling from impurity injection by laser blow-off, and following the electron temperature profile evolution from electron cyclotron emission. In the high density in the SOC regime, there is ‘normal’ diffusive heat transport, with a drop in the core temperature lagging the edge cooling by about an energy confinement time. Also with SOC, the core rotation is counter-current, and there is a significant up/down edge impurity density asymmetry. At low density in the LOC regime, the core electron temperature increases (on a faster time scale) following the edge cooling, indicating the workings of a convective heat pinch or transient ITB. The core rotation with LOC is co-current and the edge impurity density profile is up/down symmetric. Rotation reversal, the transformation from non-diffusive to diffusive heat transport, the switch of edge impurity density profiles from up/down symmetric to asymmetric and changes in turbulence have all been observed dynamically during a single discharge with a density ramp to change the collisionality. These empirical results unify a large body of previously seemingly unrelated phenomena.

Speaker: Mr John Rice (USA)

205 EX/2-3: ECRH Effects on Toroidal Rotation: KSTAR Experiments and Gyrokinetic Simulations of Intrinsic Torque at ITG - TEM Transitions

Toroidal rotation is important for control of stability and transport in tokamaks. Intrinsic rotation is self-generated by ambient turbulence via the non-diffusive residual stress, which motivates the question of how macroscopic rotation profiles will evolve in response to changes in the ambient micro-turbulence. One ‘control knob’ for the micro-turbulence population is the heating mix of NBI and ECRH. The change in rotation to counter-direction by ECH in KSTAR is explained by the turbulence change from ITG to CTEM. We investigate the effect of ECRH heating on NBI-driven toroidal rotation profiles in L-mode and H-mode discharges in KSTAR tokamak. 1.3 MW of NBI is injected in the co-current direction and 350 kW of ECRH are applied. The ion temperature and toroidal rotation are measured with high resolution XICS and CES. NBI in the co-current direction drives peaked rotation profiles with (H-mode) and without (L-mode) a pedestal. Dramatic decreases in the core toroidal rotation values are observed when on-axis ECRH is added to H-mode. These increments delta(Vtor)/Vtor ~ -30% indicate the presence of on ECH-induced counter-current torque acting in the discharge core. We note that, for steady state with same external torque and boundary condition, the change of the radial gradient of plasma rotation implies the change of residual stress. Interestingly, edge and pedestal rotation velocities in H-mode are nearly unchanged. We explore the viability of the ITG--TEM transition as an explanation of the observed change in the sign of the core intrinsic torque. The global gyrokinetic code gKPSP was used for the study. We performed ITG and TEM simulations at values of eta_i = 3.1 (i.e. ITG) and eta_i = 1.0 (i.e. TEM), respectively. Note that the low value of eta_i for the CTEM case is qualitatively consistent with the reduction in core grad(Ti) observed with ECRH in KSTAR. Results show that the residual stress changes sign as ITG--TEM transition occurs, indicating a change in the direction of the net wave energy density flux. Direct simulations also reveal a mean macroscopic profile reversal at ITG--TEM transition, thus confirming the overall consistency of the argument. Also we will perform nonlinear gyrokinetic simulations to calculate the strength of intrinsic torque reversal at ITG--TEM transition and compare the simulation results with experiments.

Speaker: Mr Yuejiang SHI (Republic of Korea)

206 EX/2-4: Dependence of Heat Transport and Confinement on Isotopic Composition in Conventional H-mode Plasmas in JT-60U

Dependence of heat transport on isotopic composition is investigated in conventional H-mode plasmas for the application to ITER. The identical profiles of electron density, electron temperature and ion temperature are obtained for hydrogen and deuterium plasmas while the required power becomes clearly larger for hydrogen, resulting in the reduction of the heat diffusivity for deuterium. The result of the identical temperature profiles in spite of different heating power suggests that the characteristics of heat conduction differs essentially between hydrogen and deuterium even at the same scale length of temperature gradient. On the other hand, the edge stability is improved by increased total poloidal beta regardless of the difference of the isotropic composition.

Speaker: Mr Hajime Urano (Japan)

207 EX/2-5: Extension of Operational Regime in High-Temperature Plasmas and the Dynamic-Transport Characteristics in the LHD

Realization of high-T_i plasmas is one of the most important issues in helical plasmas, which have an advantage for steady-state operation comparison with tokamak plasmas. Since 2010, newly installed perpendicular-NBI with the beam energy of 40 keV has been operational in the Large Helical Device (LHD) and the total-heating power of perpendicular-NBIs increased from 6 MW to 12 MW. Such low-energy NBIs are effective for ion heating and enabled us to achieve a higher T_i than that obtained previously. In the last experimental campaign, ICRF-discharge cleaning was adopted to reduce particle recycling from the wall. As a result, NBI-heating-power profile became peaked and the density-normalized ion heating power in the core region increased by 18% and the central T_i of 7 keV, which is the world’s highest value for helical devices, was successfully achieved as the new record in the LHD. In the LHD, high-T_i plasmas have been realized in combination with a carbon pellet. The kinetic-energy confinement was improved by a factor of 1.5 after the pellet injection. In the high-T_i phase, a flat or hollow profile in the electron density has been observed. This is the different characteristics from PEP mode investigated in Tokamaks. After the pellet injection, the central T_i, the gradient of T_i and that of the toroidal-flow velocity at the core region clearly increased indicating the formation of the ion-internal-transport barrier. In the high-T_i phase, reduction of the thermal diffusivity over the wide region was observed. In the core region, the time constant of the improvement of the ion-heat transport was found to be larger than that in the peripheral region. The toroidal-momentum transport was also improved accompanied with the reduction of the thermal diffusivity and the Prandtl number ignoring the intrinsic torque was close to unity. However, the confinement improvement was temporal and the gradient of T_i gradually decreased. Similarly, the toroidal-momentum transport went back to the low-confinement state in the latter phase of the discharge. Decrease of the negative radial electric field and increase of the density fluctuation were also observed in the phase.

Speaker: Mr Hiromi Takahashi (Japan)

208 EX/P3-01: Poloidal Variation of High-Z Impurity Density due to Hydrogen Minority ICRH on Alcator C-Mod

In the Alcator C-Mod tokamak, strong, steady-state variations of molybdenum density within a flux surface are routinely observed in plasmas using hydrogen minority ion cyclotron resonant heating. In/out asymmetries, up to a factor of 2, occur with either inboard or outboard accumulation depending on the major radius of the minority resonance layer. Quantitative comparisons between existing parallel high-Z impurity transport theories and experimental results show good agreement when the resonance layer is on the high-field side (HFS) of the tokamak but disagree substantially for low-field side (LFS) heating. Impurity accumulation on the LFS of a flux surface can be explained by the centrifugal force, and is the first observation of intrinsic rotation generating an in/out asymmetries. The accumulation of impurity density on the HFS of a flux surface is shown to be driven by a poloidal potential variation sustained by magnetically trapped non-thermal, cyclotron heated minority ions. Parallel impurity transport theory is extended to account for these fast-ion effects and shown to agree with experimentally measured impurity density asymmetries.

Speaker: Mr Matthew Reinke (USA)

209 EX/P3-02: Measurements of Core Lithium Concentration in Diverted H-Mode Plasmas of NSTX

The National Spherical Torus Experiment (NSTX) is investigating the use of lithium as a candidate plasma-facing material to handle the large power flux to the wall of fusion devices. To investigate the possible contamination of the core plasma caused by lithium influx from the plasma boundary, measurements of core lithium concentration, nLi(R), have been performed in diverted H-mode plasmas of NSTX. Various experimental scenarios, representative of the NSTX operating space, are explored from the 2010 NSTX experimental campaign, during which a total of ~1.3 kg of lithium was evaporated into the vessel. It is found that, in spite of the large amount of lithium (hundreds of milligrams) introduced in the vessel either before or during a discharge, nLi remains insignificant, typically <<0.1% of the electron density. The measured nLi is rather insensitive to variations of plasma current, toroidal field, divertor conditions and of the specific technique utilized for lithium conditioning of the vessel wall. These results enable projections to the higher field and current and longer pulse length of the NSTX Upgrade (NSTX-U), suggesting that lithium contamination will remain negligible compared to other impurities such as carbon. Work supported by U.S. DOE Contract DE-AC02-09CH11466.

Speaker: Mr Mario Podesta (USA)

210 EX/P3-03: Potential Fluctuation Study from the Core Plasma to End Region in the GAMMA 10 Tandem Mirror

Correlation between the drift type fluctuation and anomalous radial transport was observed in GAMMA 10 and these fluctuations were suppressed by electron cyclotron heating (ECH) driven radial electric field. We have developed new diagnostics to investigate for these studies, which are a simultaneous two points measuring gold neutral beam probe (GNBP) for the radial electric field and potential fluctuation and a high speed end plate potential fluctuation measurement system. The radial electric field and its fluctuation successfully obtained by using simultaneous two point measurements. The potential fluctuation phase difference between the two measuring positions in a single plasma shot was obtained for fluctuation analysis. The coherency of the drift type potential fluctuations between the core plasma by GNBP and that of the end plate measurement was clearly observed. By using end plate system with the GNBP, we can study detailed potential fluctuations in the core plasma without ECH. It is found that these potential and electric field fluctuations are clearly suppressed by the positive electric fields. We have obtained the strong tools for investigating the correlation of the radial electric field and the potential fluctuations between core and edge plasmas.

Speaker: Mr Masayuki Yoshikawa (Japan)

Poster IAEA-poster-8.pdf

211 EX/P3-04: High Current Plasmas in RFX-mod Reversed Field Pinch

High current (I_p up to 2 MA) operations in RFX-mod access the Single Helical Axis (SHAx) regime, during which the magnetic dynamics is dominated by the innermost resonant mode (m=1, n=-7): the magnetic chaos level is reduced and internal magnetic field configuration is close to a pure helix. The best plasma performances at high Ip with pure SHAx states featuring electron transport barriers have been reached with shallow values of the reversal parameter F=B_t(a)/<B_t>(-0.05<F<0). The SHAx states show back transitions to Multiple Helicity (MH) regime; at I_p>1.5 MA their total persistence is greater than 90% of the plasma current flat-top. During the SHAx state strong electron temperature (Te) gradients can show up, identifying an electron transport barrier. The dynamics of the thermal structure has been characterized with a new high time resolution T_e profile diagnostic, SXR double-filter multichord system. The high T_e gradients last up to 10-15 ms, more than the energy confinement time. The strongest barriers in the central helical region are achieved at the lowest total amplitudes of the m=1 secondary modes. Improvements of error field control capability has been recently identified in the correction of systematic errors in the edge field measurements and of the fields induced by the presence of a conductive wall with 3D structures. The T_e profile measured inside the barrier is flat; to describe the transport in this region, electrostatic turbulence and subsequent vortical drift have been taken into account in a simplified model, in alternative, the residual m=2 mode activity has been considered as the source of the magnetic chaos. In the region of the T_e gradients, the main gas diffusion coefficient is one order of magnitude lower than the MH case and the convection term is negligible, in agreement with the removal of the stochastic transport. The large plasma volume external to the barrier is crucial to improve the global confinement. The Te gradients in the region 0.7<r/a<0.95 increase at lower amplitudes of the m=0 modes, likely connected to a lower edge turbulence. Lithium wall conditioning experiments are ongoing aiming at producing higher edge temperature and temperature gradients through a reduced radiation and recycling: promising experiments with more peaked density profiles and good density control up to n/n_G = 0.5 have been produced.

Speaker: Ms Lorella Carraro (Italy)

Poster poster_IAEA_HighIp.pdf

212 EX/P3-05: Experimental Investigation of Plasma Confinement in Reactor Relevant Conditions in TCV Plasmas with Dominant Electron Heating

This paper reports on recent TCV experiments performed to investigate the confinement of electron-heated discharges simulating reactor relevant conditions with dominant electron heating. The dependence of the L-mode confinement properties on the electron heating power density profile width has been analyzed for the first time. Discharges with on-axis peaked ECR heating profiles with half-width varying between 15 and 40% of the minor radius have been performed in the range of heating power 0.5-2 MW and at line-averaged density 2*10^19 m^-3=0.15 n_Gw. A scenario with off-axis heating peaked at rho~0.7 but the same total heating power has also been studied for comparison. The following features have been found [1]: (i) the confinement is largely independent of the power deposition profile width, provided a significant fraction occurs inside the q=1 radius; (ii) the energy confinement time scales with the heating power as tau_E~ (P_tot)^-0.73, independent of the heating profile width, when the power is peaked on-axis; (iii) off-axis heating results in stronger confinement degradation (tau_E~ (P_tot)^-0.9). An investigation of the plasma confinement in high-density discharges has also been performed. A limit density close to and even exceeding the Greenwald limit n_e_lim~(0.6-1.1)*n_Gw has been achieved in ohmically heated plasmas, depending on the q_edge value. [1] N A Kirneva, K A Razumova, A Pochelon et al, Plasma Physics Control. Fusion 54 (2012) 015011

Speaker: Ms Natalia Kirneva (Russian Federation)

213 EX/P3-06: Suprathermal Ion Studies in ECRH and NBI Phases of the TJ-II Stellarator

In TJ-II we have been able to detect suprathermal ions by passive spectroscopy in the plasma interior and by means of a luminescent probe for those escaping from the confinement region. We have measured their temporal evolution in both cases and with several vision cords in the first system. The luminescent probe has very high sensitivity and it was operated in a height pulse analysis regime. Its data was processed by an ad hoc digital pulse algorithm which has made possible to determine its ion energy distribution function in ECRH and NBI regimes. We have been studying, not only with protons but also in some selected impurity ions, temperatures and population of these suprathermal ions as well as its flow and rotation velocity when the available photon statistics permit it. We have attained several conclusions: a) the typical temperatures of the suprathermal ions is a factor 4 higher than the thermal one in the case of protons; b) in many cases the ion energy content of this suprathermal component is comparable to the energy content of the thermal one with relevant implications for the TJ-II energy balance; c) the suprathermal rotation measured in a poloidal plane can be a factor between 2.5 and 5 higher than the rotation of the thermal component. All these results represent a true challenge for theories capable of explaining its generation, like in the case of ECRH phase, and its rotation: suprathermal particles constitute an excellent probe of physical mechanisms inside a high temperature plasma complementary of thermal ones. Its time behavior in transient situations has made possible to realize that the confinement time is lower as the energy of the selected population is higher, and this fact suggests that the deconfinement mechanisms of suprathermal ions cannot be explained solely by Coulomb collisions, but rather well intrinsic properties of the TJ-II magnetic configuration, like magnetic ripple, magnetic configuration, etc., must be invoked even to understand qualitatively this behavior within the stellarator field.

Speaker: Mr Bernardo Zurro (Spain)

214 EX/P3-07: Study of Fueling Control for Confinement Experiments in Heliotron J

This paper discusses the effects of fueling control on plasma performance in Heliotron J, a helical-axis heliotron device with an L/M = 1/4 helical coil (R_0 = 1.2 m, <a_p> = 0.12-0.17 m, <B_0> < 1.5 T). Here, L and M are the pole number of the helical coil and its helical pitch number, respectively. Based on recent installation/improvement of diagnostics, which give us plasma profile database for detailed transport analyses, the confinement study has been accelerated. Here fueling and recycling control is not only one of the key issues for high density and high performance plasma but also plays important roles in diagnostics. Effectiveness of supersonic molecular beam injection (SMBI) fueling has been studied based on profile data. A peaked density profile is realized by SMBI in NBI plasma, while a conventional gas puff (GP) fueling results in rather flat profile for the same heating condition. This is qualitatively consistent with the edge density profile reconstructed from an AM microwave reflectometer data. Since the amount of gas to obtain the same increment of the line-averaged density is about 30-40% higher in GP compared to SMBI, the expected difference in the neutral density outside the plasma after SMBI or GP might contribute to make the observed different density profile at ~20ms after the fueling. SMBI can also affect plasma fluctuations. Fast camera observation for filament structure in the edge turbulence has revealed that SMBI can change its rotation direction and/or speed. Similar change is observed at L-H transition in Heliotron J. In addition, recent density fluctuation measurement at different radial positions with a beam-emission spectroscopy (BES) system suggests SMBI affects the fluctuation inside the last-closed flux surface. Here, the observed fluctuation may be some MHD mode relating to high-energy ions. During about 10ms after SMBI, the fluctuation is not observed in BES data and the Mirnov-coil signal is decreased, suggesting change of excitation condition of the mode. These observations suggest more preferable control scenario of NBI deposition profile toward core heating through n_e(r) modification caused by SMBI.

Speaker: Mr Tohru MIZUUCHI (Japan)

215 EX/P3-08: Magnetic Fluctuation-Driven Intrinsic Flow in a Toroidal Plasma

Magnetic fluctuations have been long observed in various magnetic confinement configurations. These perturbations may arise naturally from plasma instabilities such as tearing modes and energetic particle driven modes, but they can also be externally imposed by error fields or external magnetic coils. It is commonly observed that large MHD modes lead to plasma locking (no rotation) due to torque produced by eddy currents on the wall , and it is predicted that stochastic field induces flow damping where the radial electric field is reduced . Flow generation is of great importance to fusion plasma research, especially low-torque devices like ITER, as it can act to improve performance. Here we describe new measurements in the MST reversed field pinch (RFP) showing that the coherent interaction of magnetic and particle density fluctuations can produce a turbulent fluctuation-induced kinetic force that acts to drive spontaneous plasma rotation. Key observations include; (1) the average kinetic force, ~ 0.5 N/m3, is comparable to the intrinsic flow acceleration, , and (2) between sawtooth crashes, the spatial distribution of the kinetic force is directed to create a sheared parallel flow profile that is consistent with the measured flow profile, suggesting the kinetic force could be responsible for intrinsic plasma rotation.

Speaker: Mr Weixing Ding (UCLA)

216 EX/P3-09: Understanding the Dynamics of Cold Pulse Nonlocality Phenomena

Resolving the long standing question of whether tokamak transport is local or non-local is crucial for successful predictive modeling of ITER. Though most theoretical and modeling schemes are based on the local Fickian formulation, there are many conspicuous experimental results suggesting otherwise. Most notable of these is the cold pulse nonlocality in which edge cooling produces a central temperature rise on a time scale drastically shorter than the bulk confinement time. In this paper, we present experimental and theoretical results which illuminate the enigma of the cold pulse and support the hypothesis that the associated central temperature rise is due to the formation of a transient thermal barrier structure. We present fluctuation measurements which support this conclusion. Experiments show that Supersonic Molecular Beam Injection (SMBI) into a LOC plasma results in fast edge cooling and the subsequent central heating. Sequential SMBI sustains the elevated core temperature if the interval between SMBI pulses is shorter than one confinement time. Microwave reflectometry studies of the response of core turbulent density fluctuations to SMBI show that small scale, high frequency fluctuations are reduced in this state with sustained increase of Te(0). Theoretical studies have focused on the elucidation of the origin of the rapid profile response to the peripheral cooling. Studies of model simulations of cold pulse experiments recover all of the elements of non-locality – i.e. the inversion, a transient thermal barrier due to the self-consistent shearing and a fast response. Further studies reveal that the key to the inversion and transient thermal barrier is diamagnetic shearing, self-consistently produced in response to dynamic profile perturbations. Thus, we propose that the long standing mystery of turbulent nonlocality phenomena is ultimately rooted in a simple but strongly nonlinear feedback loops in the transport dynamics, and that dynamic shearing can produce the transient barrier structures required to explain the cold pulse phenomenology.

Speaker: Mr Hongjuan Sun (Republic of Korea)

217 EX/P3-10: Collisionality Dependence of Confinement in T-10 L-Mode Plasmas

Investigation aimed at the understanding of a general origin of the thermal and particle transport has been carried out in regimes with the dominant electron heating in the T-10 tokamak. ECR heating with the power of 0.25-3 MW has been used. Two scans have been summarized for the analysis: density scan at the constant EC heating power value, (P_EC=0.9 MW, P_tot=1 MW), and recently obtained EC heating power scan at the fixed density n_e=1.8*10^19 m^-3=0.25 n_Gw. The value of the effective collisionality was changed in the range nu_eff~0.1-10. For the first time it was shown that the main regularities of the energy and particle confinement do not depend of the method of collisionality modification. These features are the following: i) energy confinement time increase with collisionality and saturates at nu_eff~1-2; ii) the density peaking increases with collisionality and goes to higher level at the same nu_eff value, nu_eff~1-2; iii) the density profile flattening becomes stronger with collisionality increase in the ECR heated discharges in comparison with the ohmically heated discharges taken at the same collisionality.

Speaker: Ms Natalia Kirneva (Russian Federation)

218 EX/P3-11: The Value of Flexibility: the Contribution of RFX to the International TOKAMAK and STELLARATOR Programme

RFX-mod embodies the characteristics of flexibility of an experiment where cross-configuration studies can be carried out. Such condition has aroused lively interest and has led to important collaborations with laboratories worldwide (JT60-SA, DIII-D, AUG, PPFL, PPPL, Auburn University, ORNL). As a Reversed Field Pinch (RFP), RFX-mod addresses many basic physics issues that are common to both Tokamak and Stellarators, just in a different region of the parameters space of a hot plasma, characterized by low magnetic fields. In addition RFX-mod can be run directly as a low current, ohmic, circular Tokamak and apply to it its state-of-the-art system for active MHD feedback control and investigate, for instance, Resistive Wall Modes and Resonant Field Amplification processes. The control of the (2,1) mode has allowed exploring equilibria down to q(a)=1.6, showing the importance of correctly treating the aliasing of the sidebands generated by the correction coils. A high triangularity plasma with double x-point has also been produced. Entering an ohmic H mode would open the way to further important studies such as ELM’s control. The bridge with the Stellarator community has been established because the RFP helical states can provide a good test-bed for numerical codes conceived to deal with 3D effects. The equilibrium codes VMEC and V3FIT developed for the Stellarator have been successfully adapted to reconstruct RFX-mod equilibria with diagnostics. Such equilibria show a good agreement with the results of the RFP equilibrium reconstruction code SHEq, providing the additional information on the role of pressure. The resulting q profiles show a non monotonic radial shape and the presence of a maximum where usually a strong thermal barrier develops. The contribution of coherent structures to the transport of particles and energy at the plasma edge has been studied on RFX-mod with direct observations of current density filaments in the edge region both in the RFP configuration - where drift kinetic Alfvén structures have been identified - and in the Tokamak configuration, where small scales current filaments have been found. These results are compared with the findings in the TORPEX experiment, those obtained in the AUG experiments during type one ELM’s, and with those in the TJ-II Stellarator device where similar investigations are in progress.

Speaker: Mr Marco Valisa (Italy)

219 EX/P3-12: Real-time Model-based Reconstruction and Control of Tokamak Plasma Profiles

A new paradigm for real-time plasma profile reconstruction is demonstrated in the TCV tokamak. Predictions based on physics models are merged with available real-time diagnostic data to construct a self-consistent profile state estimate compatible with a time-dependent model of transport processes in the plasma. This is enabled by a new RApid Plasma Transport simulatOR (RAPTOR), implemented in the new TCV real-time control system. RAPTOR simulates the radial current diffusion including the ohmic coil transformer voltage and non-inductive sources in real-time, while the plasma physically evolves in the tokamak. This makes available an extensive set of quantities which are normally not known in real-time such as the bootstrap current fraction, safety factor, magnetic shear and loop voltage profiles. This approach represents a generalization of existing approaches for real-time equilibrium reconstruction with measurement-constrained current density profile, as transport physics knowledge is now included in the reconstruction. The same rapid transport code is also used in predictive mode, including a model of the electron temperature evolution, for off-line studies of optimal actuator trajectories during plasma ramp-up scenarios. Constraints are included in the optimization to reflect realistic operational limits. These studies show that a plasma current overshoot combined with appropriately timed heating are beneficial for rapidly reaching a stationary q profile with flat central shear. The demonstration of this new paradigm paves the way for further integration of real-time tokamak plasma simulations for prediction, scenario monitoring, disruption avoidance and feedback control. This work was supported in part by the Swiss National Science Foundation

Speaker: Mr Federico Felici (Netherlands)

220 EX/P3-13: Turbulent Transport and Gyrokinetic Analysis in Alcator C-­Mod Ohmic Plasmas

Transport in ohmically heated plasmas in Alcator C-Mod was studied in both the linear (LOC) and saturated (SOC) confinement regimes and the importance of turbulent transport was established with gyrokinetic analysis. The presence of turbulence was measured with an absolutely calibrated phase contrast imaging (PCI) method and was compared with theoretical predictions. While in the SOC regime the measured electron an ion transport coefficients were comparable in magnitude and in agreement with GYRO predictions due to the dominance of ITG turbulence, in the LOC regime the measured transport coefficients disagreed with predictions [1]. Importantly, in the experiments electron transport was dominant, whereas GYRO found dominant ion transport due to prevailing ITG turbulence. After extensive analysis with TGLF and GYRO, in the present work it is found that in the LOC regime using an effective impurity ion species with Zi ≤ 8, and moderately high Zeff (2.0-5.6, in agreement with experimental measurements as the density was decreased) electron transport became dominant due to excitation of dominant TEM/ETG turbulence [2,3]. The key ingredient in the present results is the observation that dilution of the main ion species (deuterium) by impurity ion species of moderate charge state (Zi ≈ 8) results in the onset of TEM/ETG dominated turbulence. The turbulent spectrum measured with the Phase Contrast Imaging (PCI) diagnostic is in good agreement with predictions of a synthetic PCI diagnostic installed in global GYRO as long as the Doppler shift due to the measured ErxB is included. However, the measured flow shear is too weak to impact the stability of TEM or ITG modes in a significant way. Experiments are underway where low-Zi impurities are injected into C-Mod plasmas to test the impact of ion dilution. Work supported by the US DOE. [1] L. Lin, et al, Plasma Phys. Contr. Fusion 51, 065006 (2009) [2] M. Porkolab, et al, 38th EPS Conf. on Plasma Phys. Strasbourg (2011) [3] M. Porkolab et al, Bull. Am. Phys. Soc. 56, no 12, 139 (2011) [4] M. Porkolab et al, Bull. Am. Phys. Soc. 55, no 15, 312 (2010)

Speaker: Mr Miklos Porkolab (USA)

221 EX/P3-14: Inter-Machine Validation Study of Neoclassical Transport Modelling in Medium- to High-Density Stellarator-Heliotron Plasmas

Stellarator-Heliotrons (S-H) offer an alternative route to steady-state fusion reactors and one mission of the largest S-H devices is to provide a physics basis for burning S-H plasmas. The S-H devices closest to reactor conditions play key-roles: the Large Helical Device and Wendelstein 7-X (under construction). In contrast to tokamaks, 3D magnetic fields in S-H lead even in the plasma core to localized, trapped particles significantly enhancing the radial neoclassical (NC) transport for reactor-relevant conditions (long-mean-free-path (lmfp), Te~Ti, high nT\tau). In order to test a recently concluded benchmarking of calculations of NC transport coefficients, this study compares experimental findings with NC predictions for medium- to high-density S-H plasmas. The focus of this study is put on recent experiments conducted in LHD and TJ-II also involving findings from Wendelstein 7-AS. This experimental inter-machine study was to test NC transport models in the transition from medium- to high-density, lmfp S-H (3D) plasmas at high heating power. In 3D magnetic configurations, radial electric fields (E_r) must arise to satisfy the ambipolarity condition which is not intrinsically satisfied as in axisymmetric tokamaks. In the plasma core, E_r complies with predictions for LHD and W7-AS, differences are found for TJ-II operating at lower densities. Steady-state energy balance analyses were performed using the integrated transport code, TASK3D for LHD discharges, and by ASTRA in TJ-II. For W7-AS, the experimentally determined particle and energy fluxes were compared to NC fluxes with transport coefficients from DKES and found to be consistent with NC theory up to 2/3 of the minor radius under lmfp, high nT\tau conditions. Cases with similar findings can be reported from LHD, but depend sensitively on the shift of the major plasma axis. The inter-machine dataset obtained a variety of magnetic configurations will be comprehensively analyzed to assess the ranges of validity of NC transport predictions. This study is providing quantitative tests of the NC theory employed in the optimization of W7-X and to allow assessment of NC transport in the reactor-relevant regime of S-H devices. Moreover, the coupling of NC energy and particle transport is highly relevant for the discharge scenario development for large S-H devices in view of reactor operation scenarios.

Speaker: Mr Andreas Dinklage (Germany)

222 EX/P3-15: Tungsten Screening and Impurity Control in JET

For the ITER-like wall at JET, the screening of the divertor W source is investigated along with possibilities to influence the central metal transport. From visible spectroscopy the erosion fluxes of W are determined at the outer strike line, which intersects the horizontal, solid tungsten target tile. The W-fluxes as determined by visible spectroscopy are related to the W-content in the main plasma, which is derived from VUV spectroscopy, in order to obtain an effective W confinement time tau_W. The investigations have been performed in low- and high-triangularity H-mode discharges at 2.0MA and 2.5MA, while a deuterium fuelling gas puff has been varied from shot to shot. Both the W-erosion in the divertor and the W-screening behave beneficial for increasing gas puff, i.e. the erosion decreases and tau_W decreases. Thus, an increasing gas puff leads to a strong reduction of the W-content in all investigated H-mode plasmas. Important players for this behaviour are the divertor electron temperature, the SOL and plasma edge transport and the ELM frequency. While the absolute numbers for the W-concentrations are very low (a few 1E-6), impurity accumulation is observed for cases without any gas puff. This behaviour is discussed also considering the influence of other metal impurities. Independent of the radiating species, it is explored to what extent central wave heating (ICRH and LH) and impurity avoidance strategies, i.e. gas puffing and ELM-pace-making, help to avoid impurity accumulation.

Speaker: Mr Thomas Pütterich (Germany)

223 EX/P3-16: Classical Confinement of Impurity Ions and NBI-born Fast Ions in the Reversed Field Pinch

Classical behavior of two types of ions (impurity and NBI-born fast bulk) has recently been observed in the MST RFP plasma. Both have positive implications, as NBI-born fast ions (with normalized Larmor radius similar to that of fusion alphas in a reactor-sized plasma) are well confined and transfer their energy to the background plasma. Classical transport of impurities in this specific collisionality regime leads to a decrease in core impurity density thereby reducing bremsstrahlung losses in the dense core plasma. The confinement time and radial transport properties of carbon impurity ions are determined by classical theory during periods of suppressed magnetic turbulence in MST. The measured density of fully stripped carbon rapidly evolves to a hollow profile due to outward convection, consistent with the temperature screening mechanism in classical transport modeling. A confinement time is deduced from the decay of core carbon ions sourced by methane pellet injection and agrees with classical modeling. Classical behavior of NBI-born fast ions is also observed. A new 1 MW injector sources 25 kV hydrogen (and roughly 3% deuterium) atoms in the core of MST. The measured fast H distribution and time decay of beam-target neutron flux both indicate classical slowing without enhanced radial transport, even in a stochastic magnetic field. This leads to a substantial population of fast ions and has several effects on the bulk plasma including enhanced rotation, electron heating, and stabilization of the core resonant tearing mode. Beam driven instabilities in the RFP are observed for the first time, as both continuum energetic particle modes and discrete toroidal Alfven eigenmodes are excited by inverse Landau damping. Work supported by USDOE

Speaker: Mr Jay K. Anderson (USA)

224 EX/P3-17: Measurement and Simulation of Electron Thermal Transport in the MST Reversed-Field Pinch

Comparison of measurements made in the MST Reversed-Field Pinch (RFP) to the results from extensive single-fluid nonlinear resistive MHD simulations provides two key observations. First, thermal diffusion from parallel streaming in a stochastic magnetic field is reduced by particle trapping in the magnetic mirror associated with the toroidal equilibrium. Second, the structure and evolution of long-wavelength temperature fluctuations measured in MST shows remarkable qualitative similarity to fluctuations appearing in a finite-pressure nonlinear MHD simulation. New high-time-resolution measurements of the evolution of the electron temperature profile [Te(r,t)] through a sawtooth event in high-current RFP discharges have been made using the recently enhanced capabilities of the multi-point, multi-pulse Thomson scattering diagnostic on MST. Thermal diffusion is calculated by performing a low resolution fit of the χe profile to the electron temperature data via the energy conservation equation, assuming Fourier’s law qe = -ne χe ∇Te. These measurements are then compared directly to simulations using the nonlinear, single-fluid MHD code DEBS, run at parameters matching the RFP discharges in MST. These simulations display MHD activity and sawtooth behavior similar to that seen in MST. In a zero beta simulation, the measured χe is compared to the thermal diffusion due to parallel losses along diffusing magnetic field lines, vpar Dmag, where Dmag is determined from the simulation by tracing magnetic field lines. Agreement within uncertainties is only found if the reduction in thermal diffusion due to electron trapping is taken into account. In a second simulation, the pressure field was evolved self consistently assuming Ohmic heating and anisotropic thermal conduction. Although these pressure-evolved simulation results need further confirmation, the fluctuations in the simulated temperature are very similar in character and time evolution to temperature fluctuations measured in MST.

Speaker: Mr Daniel Den Hartog (USA)

Poster Den_Hartog.D-EX_P3-17-poster.pdf

225 EX/P3-18: Experimental Tests of Stiffness in the Electron and Ion Energy Transport in the DIII-D Tokamak

Drift wave theories (ion or electron temperature gradient modes) have an onset threshold in gradient beyond which the flux transported is predicted to increase very rapidly. For fixed boundary condition, this type of behavior would manifest itselfg as a strong resistance to change in the temperature profiles or “stiffness”. A new series of experiments exploiting the unique tools available in the DIII-D tokamak have explored this concept of stiffness in the electron channel in L-mode plasmas and in the ion channel in H-mode plasmas, specifically as a function of applied torque. In L mode, the electron temperature scale length in a narrow region was varied by a factor of 4 by changing the deposition location of the electron cyclotron heating (ECH) and changing the electron heat flux by a factor of 10. The response of the temperature profile is not dependent on the applied torque, as seen by self-similar response of the profile with balanced, co-current, and counter-current injection of neutral beams (NBI). One ECH source was also modulated to probe the flux/gradient relationship dynamically. The response is consistent with a threshold in gradient or scale length at quite low values, which is exceeded for virtually all cases with auxiliary heating in DIII-D. A similar tool for variation of the ion heat deposition is not available in DIII-D. However, the majority of the NBI heating power is deposited in the ions, so the response of the ion temperature profile to a power scan in H mode with low and high torque was obtained. The ion temperature scale length increases significantly with heating power at a normalized radius of 0.4 for both low and high torque. The change in the ion temperature scale length decreases strongly at increasing radius. The scale length is virtually constant at low torque input at a normalized radius of 0.7. This dependence on applied torque is in contrast to the electron profile results in L mode, which show little correlation with the applied torque. Work supported by the US DOE under DE-FC02-04ER54698, DE-FG02-04ER54762, DE-FG02-08ER54984, DE-FG02-07ER54917, DE-FG02-89ER53296, DE-FG02-08ER54999, DE-AC02-09CH11466, and DE-FC02-99ER54512.

Speaker: Mr Timothy C. Luce (USA)

226 EX/P3-19: Measurement of Deuterium Ion Toroidal Rotation and Comparison to Neoclassical Theory in the DIII-D Tokamak

Recent experimental comparisons of the bulk deuterium ion toroidal rotation to neoclassical theory have revealed a significant discrepancy with neoclassically predicted bulk ion toroidal rotation. Performance of ITER plasmas will depend strongly on the level of main-ion toroidal rotation achieved due to the beneficial effects of rotation for stabilization of MHD as well as the toroidal rotation contribution to the radial electric field and associated ExB shear stabilization of turbulence. Recent measurements of the main-ion toroidal rotation in deuterium plasmas have been made through new spectroscopic capabilities and integrated modeling. Neoclassically the main-ion species is generally predicted to rotate faster in the co-current direction than impurity ions. However, recent measurements of carbon and deuterium ion toroidal rotation in ECH dominated H-mode conditions have revealed than the main-ions rotate slower than carbon in the co-current direction, opposite to the neoclassical predictions, and similar to previous measurements in helium plasmas. The discrepancy lies in the neoclassical prediction of the main-ion poloidal rotation. We compute the main-ion poloidal rotation from our direct measurements and find that the deuterium ion poloidal flow velocity is significantly larger than neoclassical theory from NCLASS predicts. In low toroidal rotation ITER scenarios the performance will depend on the E_r shear stabilization of turbulence through toroidal rotation, poloidal rotation and pressure gradient contributions to the total radial electric field. The modeling of ITER performance displays a strong dependence on the predicted levels of toroidal rotation obtained by fixing the ratio of chi_psi/chi_i, however this ratio is poorly understood, and our current experimental databases of this scaling are based on measurements of impurity ions. As toroidal rotation approaches zero, E_r will be dominated by the pressure and poloidal rotation contributions; hence an accurate determination of the poloidal flow is required in plasmas with low toroidal rotation. This work supported in part by the U.S. Department of Energy under DE-AC02-09CH11466, DE-FC02-04ER54698 and SC-G903402.

Speaker: Mr Brian A. Grierson (USA)

227 EX/P3-20: Observation of ELM-Free H-Mode in the HL-2A Tokamak

For the first time, a high performance ELM-free H-mode was observed in the HL-2A tokamak.It was realized with Electron Cyclotron Resonance Heating (ECRH) and co-current Neutral Beam Injection (NBI) heating. The H98y2 factor increased by about 40-50% compared with the one before the transition. This regime is triggered by a minor disruption via the edge current change, where the temperature profile undergoes a sudden contraction. The mechanism leading to this regime is investigated. During the ELM-free phase, the edge particle transport barrier is formed with peaked central density, and the edge turbulence is significantly reduced. An EHO mode has been identified in the ELM-free H-mode plasmas. It propagates poloidally in the direction of the electron diamagnetic drift velocity and toroidally in the direction of the plasma current and NBI. The edge particle transport is enhanced with EHO which locates near the q95 surface. Interesting, the EHO/ELM transition during the plasma current ramp phase has been observed with a transient phase where the EHO and grassy ELM coexist. And each burst of the grassy ELM tends to stabilize the EHO. Thus the amplitude of the EHO becomes more fluctuant in the case of mixture of EHO and grassy ELM. During this phase, the EHO amplitude is decreasing and then it is completely disappeared, and ELMs appear with increasing density. It is likely that these transitions are governed by the edge current profile and the pedestal gradient.

Speaker: Mr Wulu Zhong (China)

228 EX/P3-21: Enhancement of Edge Impurity Transport with ECRH in HL-2A Tokamak

In next-generation fusion devices reduction of impurity concentration is of great importance for mitigating the radiation losses and the fusion fuel dilution to achieve a high confinement and high density plasma with a high radiation loss fraction at the edge for divertor compatibility. In HL-2A tokamak electron cyclotron resonance heating (ECRH) has been extensively carried out for the particle and impurity transport studies. The flattening of impurity density profile is frequently observed during the ECRH phase, while the interaction of plasma facing components with edge plasma and nonthermal electrons usually becomes stronger during ECRH phase. Radial profiles of impurity ions have been observed with a space-resolved vacuum ultra-violet (VUV) spectrometer recently developed in HL-2A, of which the intensity is absolutely calibrated using bremsstrahlung continuum. The impurity transport has been studied with this spectrometer in both edge and core plasmas for the ohmic discharges with ECRH using carbon emissions of CIII to CV. A quick decrease against the electron density in the ratio of CV to CIV with ECRH is observed, but a gradual decrease of the ratio is obtained in Ohmic plasmas. The dependence of CV/CIV on ne in Ohmic plasmas can be explained by the change in the edge density where the CIV is located. However, the quick decrease of CV/CIV with ECRH is caused by a change of the transport of C4+ ions in the core plasma. The effect of electron temperature on CV emission is small. Based on the analysis of the radial profile of CV with a 1D impurity transport code an outward flux to carbon ions in the ECRH plasma and an inward flux in the Ohmic plasmas have been obtained in the core plasma. In the SOL region, the C2+ ions are moved upstream because the increased ion temperature gradient along magnetic field transfers impurity ions upstream, while the density gradient along the magnetic field transfers them downstream. The carbon transport in the SOL is enhanced for both the Ohmic and ECRH plasmas and radiation loss from carbon ions is increased in the SOL. The effect of impurity screening is reduced. The results indicate that ECRH puts out the impurity from the core plasma and enhances the radiation in the edge plasma, suggesting a favorable condition to next-generation fusion devices.

Speaker: Mr Xuru Duan (China)

229 EX/P3-22: Dynamical Coupling Between Gradients and Transport in Tokamaks and Stellarators

Understanding the relation between free energy sources and transport is a fundamental issue in systems far from thermal equilibrium that has been debated for years. Instabilities governed by a gradient will typically produce transport events at all scales connecting different regions of plasma. In the case of a critical gradient mechanism, the functional dependence between the transport flux and the gradient is expected to show a sharp increase as the system crosses the instability threshold and finite background transport below the threshold, implying a non-linear relation between gradients and turbulent transport [1]. It is well known that edge turbulent transport is strongly bursty and that a significant part is caused by few large transport events [2]; this is possibly reflecting the fact that systems out of thermal equilibrium are dynamically exploring different accessible states. In this paper, the dynamical coupling between gradients and transport has been investigated in the plasma boundary of different tokamak (JET, ISTTOK) and stellarator (TJ-II) devices, showing that the size of turbulent events is minimum in the proximity of the most probable gradient. The local system relaxes to the most probable state in a time comparable to the auto-correlation time of turbulence. Experimental results were found to be consistent with results from two very different models [3, 4] of plasma turbulence and transport, where non-local effects play an important role. These non-local effects are resulting from a series of feedback mechanisms at different radial locations where at a given point in the plasma the local gradients drive the turbulence and turbulence controls the transport. These observations [5] provide a guideline for further developments in plasma diagnostics, transport modelling and data processing to characterize transport and gradients in terms of joint probability distribution functions. [1] F Ryter et al., Phys. Rev. Lett. 86 (2001) 2325 [2] M. Endler et al., Nuclear Fusion 35 (1995) 1307 [3] L. Garcia et al., Phys. Plasmas 9 (2002) 841 [4] L. Garcia et al., Phys. Plasmas 8 (2001) 4111. [5] C. Hidalgo et al., Phys. Rev. Lett 108 (2012) 065001.

Speaker: Mr Carlos Hidalgo (Spain)

230 EX/P3-23: Quantitative Comparison of Experimental and Gyrokinetic Simulated ICRH and I_p Dependent Impurity Transport

For the first time, quantitative comparison of nonlinear gyrokinetic simulation and experiment is found to demonstrate simultaneous agreement in the ion heat and impurity transport channels. Linear and nonlinear simulation was used to interpret changes in measured transport as changes in turbulence drive and suppression terms. Extensive sensitivity analysis of the GYRO predicted impurity transport to uncertainty in experimental measurement was performed to assess quantitative agreement or disagreement between simulation and experiment. The modification of measured and gyrokinetic simulated impurity transport in response to changes in plasma current (I_p) and Ion Cyclotron Resonance Heating (ICRH) were also studied for the first time in the core of Alcator C-Mod. Utilization of a novel multi-pulse laser blow-off system coupled with the unique measurements provided by a high resolution x-ray crystal spectrometer allows for precise characterization of the spatial and temporal behavior of the full, time-evolving profile of the He-like calcium charge state. Changes in the experimental impurity transport coefficients have been determined during scans of ICRH (1.0 -3.3 MW) and I_p (0.6-1.2 MA) using a synthetic diagnostic developed around the impurity transport code STRAHL. At fixed values of a/L_{T_e}, a/L_n, and s ̂, increasing ICRH input power is observed to reduce the drive for Ion Temperature Gradient (ITG) turbulence. Linear stability analysis performed using the gyrokinetic code GYRO suggests that the character of the core plasma turbulence transitions from ITG to Trapped Electron Mode (TEM) dominated during the power scan. Measured changes in the experimental transport coefficients with input power have been compared with qualitative predictions of quasi-linear theory. During the I_p scan, significant modification of the measured impurity confinement time and the parameters a/L_n and s ̂ is observed. Analysis using high fidelity, global (0.29 < r/a < 0.62), nonlinear GYRO simulations predict a decrease in the inward impurity pinch and diffusion with increasing plasma current which is both quantitatively and qualitatively consistent with experimental observations. Physical interpretation of I_p and ICRH driven changes in turbulent impurity transport as well as critical comparison of code predictions with measured transport will be presented.

Speaker: Mr Nathan Howard (MIT - Plasma Science and Fusion Center)

231 EX/P3-24: Observation of Electron Energy Pinch in HT-7 ICRF Heated Plasmas

Inward energy pinch in electron channel is observed in HT-7 superconducting tokamak using off-axis ion cyclotron resonance frequency (ICRF) heating. The experimental results and power balance transport analysis by TRANSP code are presented in this article. With the aids of GLF23 transport model, which predicts energy diffusivity in experimental condition, the estimation of electron pinch velocity is obtained by experimental data and is reasonably similar to the results in previous study, such as Song in Tore Supra. The parametric dependence of pinch velocity and the benchmarks between HT-7 experiment and existent theories will be performed soon.

Speaker: Mr Siye Ding (China)

232 EX/P3-25: Survey of Density Modulation Experiments on the HT-7 Tokamak

The particle diffusion coefficient and the convection velocity have been studied by means of the density modulation using pulsed deuterium gas puffing on the HT-7 tokamak. It was observed in AC plasmas that the particle transport coefficient and confinement time of the positive current plasma is different from that of the negative current plasma (Gao X et al 2008 Nucl. Fusion 48 035009) on HT-7 tokamak. New experimental result improved our understanding in AC plasma operation on HT-7 tokamak. It was found that the particle confinement time becomes much higher when the directions of plasma current and toroidal field are uniform. Recently, the density modulation experiments are carried out with advanced liquid lithium limiter on HT-7 tokamak. The interesting results are compared and discussed in detail with previous results under the graphite limiter on HT-7 tokamak.

Speaker: Mr Xiang Gao (China)

Poster Poster of EX/P3-25

233 EX/P3-26: Particle Transport Results from Collisionality Scans and Perturbative Experiments on DIII-D

Recent GYRO simulations predict that particle flux, as a function purely of collisionality, should show a strong increase at low collisionality, but with little change above a critical value, nu^*~0.01. In an L-mode experiment in which collisionality was varied from nu^*~0.01 to 0.05, little change was observed in the density profile, profile peaking, measured D, V, (obtained using perturbative techniques), or measured fluctuation levels, in agreement with the GYRO predictions made prior to the experiment. This experiment was performed using similarity techniques to vary the collisionality; i.e. the magnetic field B_T was changed from 2.1 to 1.65 T, and heating power was also varied, while matching key dimensionless parameters such as relative gyroradius, beta and safety factor. TGLF and GYRO simulations of the actual experimental discharges are underway. In a second set of experiments, perturbative transport techniques using oscillating gas puffs were utilized to measure D and V in multiple operating regimes, including conventional ELMing H mode, ELM-suppressed operation obtained using resonant magnetic perturbations (RMPs), QH-mode, and L-mode. These experiments provide the first direct measurement confirming an increase in D and decrease in V with RMP application. One important feature of the measurements is that the changes in D and V with RMP application extend deep into the plasma core, past the edge region where the applied RMP fields are expected to directly impact the magnetic field topology. In the plasma core, clear increases in plasma turbulence levels are observed, consistent with TGLF modeling, while ExB shear decreases to a level below the linear growth rate. This work was supported in part by the US Department of Energy under DE-FG02-08ER54984, DE-FC02-04ER54698, DE-FG02-89ER53296, DE-FG02-08ER54999, DE-FG02-05ER54809 and DE-FG02-07ER54917.

Speaker: Mr E. J. Doyle (USA)

234 EX/P3-27: Transition of Poloidal Viscosity by Electrode Biasing in the Large Helical Device

The transitions to the improved confinement mode, which were accompanied with bifurcation phenomena characterized by a negative resistance, were clearly observed in various magnetic configurations on the Large Helical Device (LHD) by the electrode biasing. The configuration dependence of the transition condition and the radial resistivity qualitatively agreed with neoclassical theories. The effects of the ion viscosity maximum on the transition to an improved confinement mode were experimentally investigated by the externally controlled J x B driving force for a poloidal rotation using the hot cathode biasing in Tohoku University Heliac (TU-Heliac) [1, 2]. Here, J and B are a biasing electrode current and a magnetic field. In steady state the J x B driving force balances with the ion viscous damping force and the friction to neutral particles. Then the local maximum in ion viscosity can be evaluated experimentally from the external driving force at the transition. The remaining problems in the biasing experiments on TU-Heliac were (1) to spread a ripple component in magnetic configuration, and (2) to improve a target plasma to a high-performance region. In LHD the effective helical ripple and viscosity maxima have wider region than those in TU-Heliac. The relation between a poloidal Mach number and ion viscosities in LHD predicted by the neoclassical theory [3] shows that viscosity maxima change drastically depending on the position of the magnetic axis that mainly changes the effective helical ripple. LHD can produce higher temperature plasmas at higher field than the plasma in TU-Heliac. We tried the biasing experiment in LHD, and reported the first observation of transition phenomena [4]. In this paper we report the difference of the transition condition in three configurations by the electrode biasing experiments in LHD and the radial electric field and the viscosity estimated from the neoclassical transport code FORTEC-3D [5] for a non-axisymmetric system. [1] H. Takahashi et al., Plasma Phys. Control. Fusion, 48, 39 (2006). [2] S. Kitajima et al., Nucl. Fusion, 46, 200 (2006). [3] K. C. Shaing: Phys Rev. Lett. 76, 4364 (1996). [4] S. Kitajima et al., Nucl. Fusion, 51, 083029 (2011). [5] S. Satake et al., PPCF, 53, 054018 (2011).

Speaker: Mr Sumio Kitajima (Japan)

235 EX/P3-28: Production of Internal Transport Barriers by Intrinsic Flow Drive in Alcator C-Mod

New results suggest that changes observed in the intrinsic toroidal rotation influence the internal transport barrier (ITB) formation in the Alcator C-Mod tokamak. Detailed plasma rotation and ion temperature profile measurements are combined with linear and non-linear gyrokinetic simulation to examine the effects of the self-generated rotational shear on the transport changes that occur in C-Mod ITB plasmas. These arise when the resonance for ICRF minority heating is positioned off-axis at or outside of the plasma half-radius. These ITBs form in a reactor relevant regime, without particle or momentum injection, with T_i≈T_e, and with monotonic q profiles (q_min < 1). C-Mod H-mode plasmas exhibit strong intrinsic co-current rotation that increases with increasing stored energy without external drive. When the resonance position is moved off-axis, the rotation decreases in the center of the plasma resulting in a radial toroidal rotation profile with a central well which deepens and moves farther off-axis when the ICRF resonance location reaches the plasma half-radius. This profile results in strong E×B shear (>1.5x10^5 Rad/sec) in the region where the ITB foot is observed. Gyrokinetic analyses indicate that this spontaneous shearing rate is comparable to the linear ion temperature gradient (ITG) growth rate at the ITB location and is sufficient to reduce the turbulent particle and energy transport. The newly available detailed measurement of the ion temperature demonstrates that the radial profile flattens as the ICRF resonance position moves off axis, decreasing the drive for ITG the instability as well. These results are the first evidence that intrinsic rotation can affect confinement in ITB plasmas, and suggest that this regime could be achievable in ITER and in future reactor experiments.

Speaker: Ms Catherine Fiore (USA)

Poster IAEAPoster_Fiore.pdf

236 EX/P3-29: Experiments on GOL-3 Multiple Mirror Trap for Fusion Program

The GOL-3 Multiple Mirror Trap is an 11-m-long solenoid with axially-periodical (corrugated) magnetic field. In the basic operation regime the solenoid consists of 52 magnetic corrugation cells with Bmax/Bmin=4.8/3.2 T. Deuterium plasma of 10^20÷10^22 m^-3 density is heated up to ~ 2 keV ion temperatures (at ~10^21 m^-3 density and confinement time ~1 ms) by a high power relativistic electron beam. Main conclusion from data is that plasma heating and confinement in the multiple mirror traps are of essentially turbulent nature. In general, achieved plasma parameters support our vision of a multiple mirror trap as the alternative path to a fusion reactor with beta~1 and 10^21÷10^22 m^-3 plasma density. Project of a new linear trap with multiple mirror plugs is in progress in Novosibirsk BINP. Several new experiments in support of the fusion program based on linear machines are presented. . An intense electron beam source of a new type, based on a gaseous arc plasma emitter, was developed and first experiments with this beam are carried out; the new data on plasma rotation and electromagnetic radiation in the GOL-3 will be presented.

Speaker: Mr Aleksandr Burdakov (Russian Federation)

237 EX/P3-30: Comparison of Plasma Flows and Currents in HSX to Neoclassical Theory

The Helically Symmetric Experiment (HSX) was designed to have an axis of symmetry in the helical direction, reduced neoclassical transport and small equilibrium currents due to the high effective transform. Unlike other stellarators in which |B| varies in all directions on a flux surface, plasmas in HSX are free to rotate in the direction of quasihelical symmetry. In this paper we will present measurements with Charge Exchange Recombination Spectroscopy (CXRS) that demonstrate for the first time that intrinsic plasma flows with a velocity up to 20 km/s are predominantly in the direction of symmetry. Whereas previous neoclassical calculations did not conserve momentum, we show that the experimental results agree better with recent modifications to neoclassical theory that do conserve momentum. Also, we present for the first time a 3-D equilibrium reconstruction of the plasma pressure and current profile based on a set of magnetic flux loops. Early in time, the magnetic signals indicate that, because of the absence of toroidal curvature, a helical Pfirsch-Schlüter current develops. Later in time, the bootstrap current evolves over a time scale longer than the plasma discharge and is modeled using a 3-D suseptance matrix method. The reconstructed pressure profile agrees well with the experimental measurements. The reconstructed current profile agrees well with the neoclassical calculations of the bootstrap current including momentum conservation. However, a wide range of current profiles also show reasonable agreement with the data, indicating that the measured edge magnetic signals are not that sensitive to the small bo