Improved Confinement in JET High Plasmas with an ITER-like Wall - PowerPoint PPT Presentation

Improved Confinement in JET High  Plasmas with an ITER-like Wall Clive Challis Culham Centre for Fusion Energy Paper EX/9-3 IAEA FEC, St Petersburg 13-18 October 2014 1 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Acknowledgements Co-authors: E. Belonohy 1,2 , M. Beurskens 3 , P. Buratti 4 , P. Drewelow 2 , L. Frassinetti 5 , J. Garcia 6 , C. Giroud 3 , N. Hawkes 3 , J. Hobirk 2 , E. Joffrin 6 , D. Keeling 3 , D.B. King 3 , C. Maggi 2 , J. Mailloux 3 , C. Marchetto 7 , D. McDonald 8 , I. Nunes 9 , G. Pucella 4 , S. Saarelma 3 , S. Sharapov 3 , J. Simpson 3 and JET Contributors* 1 EFDA CSU, Culham Science Centre, Abingdon, OX14 3DB, UK 2 Max-Planck-Institut für Plasmaphysik, D-85748 Garching, Germany 3 CCFE, Culham Science Centre, Abingdon, OX14 3DB, UK 4 Unità Tecnica Fusione, C.R. ENEA Frascati, CP65, 00044 Frascati, Italy 5 VR, Fusion Plasma Physics, KTH, SE-10044 Stockholm, Sweden 6 CEA, IRFM, F-13108 Saint-Paul-lez-Durance, France 7 Instituto di Fisica del Plasma, CNR, 20125 Milano, Italy 8 EFDA CSU Garching, D-85748 Garching, Germany 9 Instituto de Plasmas e Fusão Nuclear, IST, Universidade de Lisboa, Portugal * See the Appendix of F. Romanelli et al., Proceedings of the 25th IAEA Fusion Energy Conference 2014, Saint Petersburg, Russia 2 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Outline of talk • Power scan with JET ITER-like wall at Comparison of JET power scans constant: I P , B, M eff , n e , a, R,  with different wall materials: – Carbon wall ( C-wall ) experiment – ITER-like wall with Be main chamber & W divertor ( ILW ) …and different plasma shapes: – Low triangularity ( low  ) – High triangularity ( high  ) IPB98(y,2) scaling • Identification of key factors that contribute to power scaling of confinement • Implications for prediction of plasma performance 3 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Overview of power scan experiments plasma current & field plasma conditions plasma shapes • H-mode (type I ELMs) high  (  U ~0.36) Scan I P (MA) B(T) ILW high  • q 95 ~3.9 1.4 1.7 • current ‘overshoot’ for access ILW low  1.4 1.7 to high  C-wall high  1.4 1.7 • No performance degrading C-wall low  1.7 2.0 MHD at analysis time low  time evolution • ‘engineering’ parameters (  U ~0.15) constant in each scan analysis q-profiles time 4 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Weak power degradation of confinement All scans show weak power degradation of confinement except C-wall high  • C-wall high  ILW low  C-wall low  ILW high  1.4MA/1.7T 1.4MA/1.7T 1.7MA/2.0T  E ~P -0.25  E ~P -0.35  E ~P -0.30  E ~P -0.65 IPB98(y,2) (  E ~P -0.69 ) 5 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Wall recycling affected C-wall scaling Close wall proximity of high  plasmas • • Increasing neutral deuterium population in vessel (e.g. by gas led to collinear increase in neutrals with injection) can degrade confinement power for C-wall ( Joffrin EPS 2014 ) ILW data may be overestimated due to reflections Wall recycling appears to affect confinement scaling in C-wall high  power scan • 6 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Pedestal & core play role in scaling • Strong increase in plasma thermal energy with power in pressure ILW scans due to increase in: W core Pedestal pressure (strongest for high  ) – W pedestal Core pressure peaking (strongest for low  ) – radius W core /P~P -0.12 W ped /P~P -0.48 W ped /P~P -0.61 W core /P~P -0.11 7 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Pedestal consistent with stability analysis • Increase in pedestal pressure with power consistent with edge stability modelling of peeling-ballooning stability boundary ( Maggi this conference EX/3-3 ) • Pedestal parameters before ELM • Pedestal temperature varied at fixed width to find peeling-ballooning stability boundary using ELITE 8 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Density peaking correlated with collisionality • Density peaking behaviour varies between power scans: ILW high  : weak increase with power – ILW low  : strong increase with power (compensates weak pedestal increase) – – Correlated with collisionality for C-wall and ILW power scans (as previously seen on AUG & JET – Angioni Nucl Fusion 2007 ) 9 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Core T i /T e increases with power • Electron temperature peaking constant in ILW power scans • Increase in core T i /T e with power contributes to increase in plasma stored energy 10 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Core T i /T e modelled • Quasi-linear TGLF simulations capture increase in T i /T e with power and shows importance of: – Ion heating (P i /P e larger at high power) – Electromagnetic effects & fast ion pressure (reduces ITG transport at high power) • Non-linear simulations may be needed to fully reproduce T peaking ( Citrin PRL 2013 ) ILW low  measurements w/o electromagnetic, ExB & fast ion effects full simulation P abs ~6MW P abs ~12MW 11 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Positive feedback on  leads to weak power degradation increase ITG NBI classical physics power classical fast-ion fast-ion collisions increase increase collisions fast ion temperature pressure peaking increase increase fast-ion particle increase fast-ion heating pinch beta slowing-down to ions increase time density decouple increase peaking ions & peeling- pedestal electrons ballooning temperature reduce paradigm collisionality increase classical core collisions temperature temperature ‘stiffness’ See mechanism described by Garcia this conference TH/5-2 12 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Neutrals affect confinement scaling • • Increasing neutrals by D gas injection Weak increase in pedestal pressure helps control W contamination with power at high gas not consistent with peeling-ballooning analysis • Strong power degradation of confinement at  N,th <1.5 & high D gas ( Maggi this conference EX/3-3 ) ILW low  ILW low  - high gas low gas (~0.3x10 22 /s) medium gas (~0.9x10 22 /s) high gas (~1.8x10 22 /s) See also Cesario Plasma Phys Control Fusion 2013 13 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Consistency with wider dataset • ILW database shows: – Strong power degradation of confinement at high D gas & low  with H 98 ~0.7-1.0 (‘baseline’ plasmas) – Weak power degradation of confinement at low D gas & high  with H 98 >1 (‘hybrid’ plasmas) • Behaviour reproduced in power H 98 =1 scans, suggesting that positive feedback loops on  may be fit to  N <2 data key factor in confinement improvement of JET ‘hybrid’ plasmas JET-ILW dataset from Beurskens Nucl Fusion 2014 14 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014

Conclusions Weak power degradation of confinement observed in JET high  plasmas • at low & high  , associated with: – Low neutral particle density in vacuum vessel ( more easily achieved with ILW than C-wall at high  ) Multiple positive feedback loops affecting  – • Strong power degradation seen with high neutral deuterium density in vacuum vessel ( through gas injection or wall recycling ) – Pedestal fails to reach modelled pedestal limit at high power • Prediction of plasma performance requires: Self-consistent modelling of  feedback loops – – Understanding of effect of neutrals on pedestal 15 Challis EX/9-3 IAEA FEC St Petersburg 13-18 Oct 2014