Photo: Martin Conway

  • Published on
    24-Jan-2016

  • View
    31

  • Download
    0

Embed Size (px)

DESCRIPTION

The Australian Plasma Fusion Research Facility: Overview and Upgrade Plans (is there potential to be an additional training platform in diagnostics and plasma physics for KSTAR?). - PowerPoint PPT Presentation

Transcript

<ul><li><p>The Australian Plasma Fusion Research Facility: Overview and Upgrade Plans</p><p>(is there potential to be an additional training platform in diagnostics and plasma physics for KSTAR?) B.D. Blackwell, J. Howard, D.G. Pretty, J.W. Read, H. Punzmann, J. Bertram, M.J. Hole, F. Detering, C.A. Nuhrenberg, M. McGann, R.L. Dewar, J. Bertram Australian National University, and *Max Planck IPP Greifswald, Photo: Martin ConwayAustralia-Korea Foundation mission to KSTAR, 2010</p></li><li><p>The Australian Plasma Fusion Facility: Results and Upgrade PlansIntroductionResult Overview: MHD Modes in H-1 Data miningAlfvnic ScalingOptical MeasurementsRadial StructureFacility Upgrade Aims Key areas New diagnostics for UpgradeConclusions/Future</p><p>*</p></li><li><p>H-1NF: the Australian Plasma Fusion Research FacilityOriginally a Major National Research Facility established by the Commonwealth of Australia and the Australian National UniversityMission:Detailed understanding of the basic physics of magnetically confined hot plasma in the HELIAC configurationDevelopment of advanced plasma measurement systemsFundamental studies including turbulence and transport in plasmaContribute to global research effort, maintain Australian presence in the field of plasma fusion power</p><p>MoUs for collaboration withMembers of the IEA implementing agreement on development of stellarator concept.National Institute of Fusion Science, Princeton Plasma Physics LabAustralian Nuclear Science and Technology Organisation*</p></li><li>H-1 CAD*Major radius1mMinor radius0.1-0.2mMagnetic Field1 Tesla (0.2 DC) q0.5 -1 (transform 1~2)ne1-3x1018 Te20eV (helicon) </li><li><p>Blackwell, Australian ITER Workshop, 10/2006H-1NF Photo</p></li><li><p>H-1 configuration (shape) is very flexible*</p><p>flexible heliac : helical winding, with helicity matching the plasma, 2:1 range of twist/turn</p><p>H-1NF can control 2 out of 3 of transform () magnetic well and shear (spatial rate of change) </p><p>Reversed Shear like Advanced Tokamak mode of operation Edge Centrelow shearmedium shear = 4/3 = 5/4twist per turn (transform)</p></li><li><p>Blackwell, Australian ITER Workshop, 10/2006Large Device Physics on H-1Confinement Transitions, Turbulence (Shats 1996--)H-mode 1996Zonal Flows 2001Spectral condensation of turbulence 2005</p><p>Magnetic Island Studies H-1 has flexible, controlled and verified geometryCreate islands in desired locations (shear, transform)Langmuir probes can map in detail</p><p>Alfvn Eigenmodes (next)D3D tokamakH-1</p></li><li><p>Experimental confirmation of configurationsRotating wire array64 Mo wires (200um)90 - 1440 anglesHigh accuracy (0.5mm)Moderate image quality Always available</p><p>Excellent agreement with computation</p><p>Blackwell, AIP Congress/AINSE PP 12/2006T.A. Santhosh Kumar B.D.Blackwell, J.HowardSanthosh Kumar Iota ~ 1.4 (7/5)</p></li><li><p>Effect of Islands on PlasmaLHD, JT60U results showTe profile flattened radial transport high (inside side to outside side)But this doesnt mean that internal transport is high(from the inside to the outside of the island.)Internal rotational transform is quite low and can complicate experiments10 to 100 toroidal transits to circumnavigate island?</p><p>Conditions:Blackwell, Kyoto JOB 16th March 2009Argon plasma ~10eVTe ~ 10 eV, Ti ~ 10 eV 30-80 eVelectron density 1 1018/m3nn &lt; neutral fill density 0.81 1018/m3e 0.075 mm, i 35-55 mmei 9105/sec en 1.6105/secCollision mean free path ei 2.5 m, en 8 mSanthosh Kumar</p></li><li><p>Small, core islandsDensity peaked near island axis (O-point)</p><p>Potential negatively peaked there too.</p><p>Local symmetry about local magnetic axis, but apparently not globally (axis to axis)Blackwell, Kyoto JOB 16th March 2009</p></li><li><p>Spontaneous Appearance of IslandsIota just below 3/2 sudden transition to bifurcated state</p><p>Plasma is more symmetric than in quiescent case.</p><p>Uncertainty as to current distribution (and therefore iota), but plausible that islands are generated at the axis.</p><p>If we assume nested magnetic surfaces, then we have a clear positive Er at the core similar to core electron root configuration?Many unanswered questions Symmetry? How to define Er with two axes?</p><p>Blackwell, Kyoto JOB 16th March 2009Santhosh Kumar</p></li><li><p>Unanswered QuestionsIssues Remaining:What is the correct analysis for Er?When does the plasma see/not see islandscollisionality, i, e internal transform are importantWhat is the most robust indicator of surface number? (p? varies with ne by a fraction of Te (Boltzmann relation)) </p><p>Correct analysis for Er?If we take the axis at the core core electron rootIf we assume two axes, then ion root, but field is still large (characteristic of e-root)[Conditions for usual e-root picture (1/, trapped e) not met]</p></li><li><p>MHD/Mirnov fluctuations in H-1Blackwell, ISHW Princeton 2009 *David Pretty</p></li><li><p>Identification of Alfvn Eigenmodes: neCoherent mode near iota = 1.4, 26-60kHz, Alfvnic scaling with nem number resolved by bean array of Mirnov coils to be 2 or 3.VAlfvn = B/(o) B/neScaling in ne in time (right) and over various discharges (below)</p><p>phase1/nene</p><p>Critical issue in fusion reactors:D + T He + nVAlfvn ~ fusion alpha velocity fusion driven instability!</p></li><li><p>Blackwell, KSTAR2010Identification with Alfvn eigenmodes: k||, iotaWhy is f so low? - VAlfven~ 5x106 m/s </p><p>res = k|| VA = (m/R0)( - n/m) B/(o) </p><p>k|| varies as the angle between magnetic field lines and the wave vectork|| - n/miota resonant means k||, 0 Expect Fres to scale with iota </p></li><li><p>Mode structure via synchronous 2D imagingIntensified Princeton Instruments camera synchronised with modeLight imaged for various delaysAveraging/Accumulation is performed by the camera </p><p>*Toroidal Field Coils</p><p>Helical ConductorJohn Howard, Jesse Read</p></li><li><p>Alfvn Eigenmode structure in H-1Compare cylindrical mode with optical emission measurementsTest functions for development of a Bayesian method to fit CAS3D modes to experiment.</p><p>John Howard, Jason Bertram, Matthew Hole</p></li><li><p>First Results from Gas Puff Imaging - true 2D imagingIntensified Princeton Instruments camera synchronised with modeSelect a small range of toroidal angles with a gas puff (Neon) Intensity ~ neInitial results 2D image without assumptions of rotation, mode number. </p><p>*John Howard, Jesse ReadViewlineGas Puff</p></li><li><p>Third Mirnov Array (Toroidal)New Toroidal ArrayCoils inside a SS thin-wall bellows (LP, E-static shield)</p><p>Access to otherwise inaccessible region withlargest signals and with significant variation in toroidal curvature.Shaun Haskey</p></li><li><p>Australian Plasma Fusion Research Facility Upgrade 2009 Australian Budget Papers Australian Governments Super Science Package </p><p>Boosted National Collaborative Infrastructure Program using the Educational Infrastructure Fund </p><p> $7M, over 4 years for infrastructure upgrades (no additional funding for research)</p></li><li><p>Aims of Facility UpgradeConsolidate Facility infrastructure including that required to implement the Australian ITER Forum strategy planAim to involve the full spectrum of the ITER Forum activities</p><p>More specifically:Improve plasma production/reliability/cleanlinessRF production/heating, ECH heating, baking, gettering, discharge cleaningImprove diagnosticsDedicated density interferometers and selected spectral monitors permanently in operationIncreasing opportunities for collaborationIdeas?Increasing suitability as a testbed for ITER diagnosticsAccess to Divertor like geometry, island divertor geometry</p></li><li><p>RF UpgradeRF (7MHz) will be the workhorseLow temperature, density limited by powerRequired to initiate electron cyclotron plasmaNew system doubles power: 2x100kW systems.New movable shielded antenna to complement bare antenna(water and gas cooled).Advantages:(non resonant Helicon) Very wide range of magnetic fields in Argon (ion-cyclotron resonant) New system allows magnetic field scan while keeping the resonant layer position constant. e.g. to test Alfven scaling MHD Additional ECH source (10/30kW 14/28GHz) for higher Te</p></li><li><p>Improved Impurity Control Impurities limit plasma temperature (C, O, Fe, Cu)High temperature (&gt;~100eV) desirable to excite spectral lines relevant to edge plasma and divertors in larger devices.</p><p>Strategy : Combine - Glow discharge cleaning for bulk of tankPulsed RF discharge cleaning for plasma facing components.antenna (cooled) and source (2.4GHz)Low temperature (90C) baking Gettering Titanium or Boron (o-carborane)Access Island Divertor GeometryAshley GibsonIsland Divertor</p></li><li><p>Small Linear Satellite Device Plasma Wall Interaction DiagnosticsPurpose:Testing various plasma wall interaction diagnostic conceptse.g. Doppler spectroscopy, laser interferometry coherence imaging, imaging erosion monitorFeatures:Much higher power density than H-1 H-1 cleanliness not compromised by material erosion diagnostic testsSimple geometry, good for shorter-term students, simpler projectsShares heating and magnet supplies from H-1</p><p>Magnetic Mirror/Helicon chamberMirror coils Optical Diagnosticsne ~ 1019 m-3P ~ 1MW/m2</p></li><li><p>Helicon H+ source ConceptBased on ANU, ORNL workQuartz/ceramic tubem=+1 Helicon AntennaDirectional Gas FlowHelicon Antenna is an efficient plasma source in ArHigh Density (&gt;1018 m-3) more difficult in HCombination of higher power and non-uniform magnetic field has produced ne ~ 1019 m-3in HWater cooled targetMirror Coilsne ~ 1019 m-3(Mirror coils at one end should be sufficient mainly to provide field gradient rather than full mirror effect.)</p></li><li><p>Additional Power/Plasma SourcesSheath acceleration increases power density, but if &gt;30-50V physical sputtering (not normal in fusion simulators, but may be useful to increase erosion?)</p><p>E-beam can increase dissipation, but ion bombardment damage of LaB6 cathode if pressure too high? Solid LaB6 Cathode, &gt;10A emission </p><p>Sterling Scientific washer gun H+ 1019-1020 5-15eV under the right conditions, can generate a relatively clean plasma (low W) den Hartog: Plasma Sources Sci. Technol. 6 (1997) 492498.(also useful for a simple way of obtaining first plasma)</p><p>15mm</p></li><li><p>FutureResearch New Toroidal Mirnov Array Bayesian MHD Mode AnalysisToroidal visible light imaging (Neon)Correlation of multiple visible light, Mirnov and n~ e dataSpatial and Hybrid Spatial/Temporal Coherence Imaging</p><p>Facility upgradeDevelop divertor and edge diagnosticsStudy stellarator divertors, baffles e.g. 6/5 island divertorDevelop plasma wall interaction diagnostics Linear Satellite device for materials diagnostic development multiple plasma sources, approach ITER edgeBlackwell, H1 Results &amp; Upgrade, AKF_KSTAR 2010</p></li><li><p>H-1NF is an excellent toroidal graduate student machine complements Korean facilitiesLarge device physics accessible MHD/Alfvnic modes.Island studiesVery good diagnostic access Invites imaging and multiview interferometric diagnosticsPrecise control of magnetic geometry, q (0.5 1) and shear.High repetition rate, availability0.5 Tesla :200 shots/day (hydrogen/helium/deuterium)Low field student mode :500 shots/day (argon, Langmuir probes etc.)Complementary high density hydrogen linear machine (~1019)Simpler geometryEdge plasma density and power density approaching ITER (over a small area)fast diagnostic/vacuum turnaround timeMDSPlus/JavaScope data common between KSTAR/H-1NF/MDFBlackwell, H1 Results &amp; Upgrade, AKF_KSTAR 2010</p><p>****Important: toroidal connection, relation to boundary conditions twisted field lines.**Flexible - wide range of plasma shapes (and stability)magnetic shear, the rate of change of twist with position.H1 is a medium to low shear device, can access reverse shear like mode of tokomaks**Lack of symmetry across the X point is strange, because the distance from one island to the other is that of one island to the next, so why is there symmetry across the O point (AB, a long distance ) and not the X (A or B C, a much shorter distance)*Electric field twice the magnitude of the highest observed in island case**Magnetic Probe signals around 30-50 kHz - Mode Jump in H-1 identified by datamining (see David Prettys presentation)**Critical issue for fusion reactors, as alpha velocity ~ Alfven velocityDensity scaling indicates Alfven velocity*Why is f so low? For the shear Alfven wave, k|| is the important component We can see that k|| and freq will go to zero at resonance i=n/m. Freq will increase with iota as we move away from res.*************</p></li></ul>