1. This work was funded by the RCUK Energy Programme under grant EP/G003955 and the European Communities under the contract of Association between EURATOM and CCFE. The views and opinions expressed herein do not necessarily reflect those of the European Commission. VI. Edge current density L-H Transition and Pedestal Studies on MAST H Meyer 1, MFM De Bock 1,2, NJ Conway 1, SJ Freethy 1,3, K Gibson 3, J Hiratsuka 4, A Kirk 1, CA Michael 1, T Morgan 1,3, R Scannell 1, G. Naylor 1, S Saarelma 1, AN Saveliev 5, VF Shevchenko 1, W Suttrop 6, D Temple 1,7, RGL Vann 3 and the MAST and NBI Teams 1 EURATOM/CCFE Fusion Association, Culham Science Centre, Oxon, OX14 3DB 2 Eindhoven University of Technology, PO Box 513, 5600 MB Eindhoven, The Netherlands 3 University of York, Heslington, York, YO10 5DD, UK 4 The University of Tokyo, Kashiwa 277-8561, Japan 5 Ioffe Institute, Politekhnicheskaya 26, 194021 St. Petersburg, Russia 5 Max-Planck-Institute for Plasma Physics, Boltzmannstr. 2, Garching, Germany 7 Imperial College of Science, Technology and Medicine, London, UK I. Motivation  ITER success relies on H-mode operation with high confinement and tolerable ELMs  Limited theoretical understanding of H-mode access and pedestal width and stability.  H-mode access in non-activation phase (hydrogen) questionable  helium operation.  Global access conditions studied in He and for different X-point height.  Change of vertical position in single null (SN), change in elongation in double null (DN).  Evolution of local parameters at the low field side mid-plane studied during the L-H transition and the ELM cycle.  Evolution of E r, T e, n e,  E r,  T e and  n e measured on 0.2ms time scale through L-H transition.  Evolution of  p e and j  studied during ELM cycle  peeling-ballooning model seems incomplete.  Novel measurement of T i shows  T i depends on collisionality on MAST. VII. Conclusions:  On MAST P LH is about (50  30)% higher in dominant 4 He than in D plasmas.  The reduction of P LH observed with the X-point closer to the targets can hardly be explained by changes in the SOL connection length, L c.  In SN a change of P LH by a factor of two is observed with only 8% change in L c.  Changes of the mean (equilibrium) E r don’t correlate with changes in P LH.   E r seems to be a sufficient condition, but not a necessary.  No changes of E r,  E r, T e,  T e, n e,  n e prior to the L-H transition with  t < 0.2ms.  T i profile on MAST reflect the fact that entropy is conserved over the whole pedestal at low *.  Edge current density measurements indicate that the peeling-ballooning model is incomplete. II. H-mode access in helium:  Righi et.al.: P LH  A eff -1 (A eff =  a M a n a /  a n a ) [1]  P LH on ITER in hydrogen difficult.  New measurements at AUG: P LH He = P LH D [2]  High priority ITER task to improve database.  Compare P LH in D and 4 He at similar parameters.  I p = 0.7 MW, S pl = 24 m 2, =2.4  10 19 m -3, B t =0.5 T  P LH He = (2.1  0.2)MW compared to P LH D = (1.4  0.2)MW  A eff ~3.7 (85% He), detailed TRANSP analysis since D-NBI and fast-ion redistribution D FI ~1 m 2 /s.  P loss = P abs +P  -dW/dt-P rad (P abs He ~0.8 P NBI, P abs D ~P NBI )  Similar pedestal at P loss -P LH ~0.3 MW  Different atomic physics; 2 ion species, no molecules  different velocity distribution.  Fit suggests narrower in He (caution).  L-H transition dynamics differ between He and D  D: Dithering H-mode over wide power range.  He: Sharp transition after modest power increase. III. Effect of X-point height on H-mode access:  JET: P LH decreases with decreasing X-point height [3]  Only if strike point (SP) is on horizontal plates  With and without septum – even septum limited.  MAST: previously SP close/far from X-point in SN [4]  H-mode/L-mode. (2m <  L c < 4m).  New: P LH increases by factor of 2 with  Z x ~ + 10 cm (SN).  No significant change in L c (  L c ~ -1m; different direction.)  Increased   |  Z x | ~ 8 cm  better H-mode access.  No change in edge T e or n e prior to L-H transition  Critical T e or  T e favoured by many L-H transition theories  no evidence for this on slow time scale. IV. Changes in the L-mode E r  E r measured using active Doppler spectroscopy on He.  L-mode E r averaged over 20ms of the dithering phase (average over many dithers).  Little change in mean E r during power scan as P loss approaches P LH  No change in mean E r with increase of  from 1.8 to 1.9.  n=3 field  E r more positive: No H-mode  But, increased power (80%) to regain H-mode  no change in E r.  Mean E r does not correlate with P LH   E  B >  max : sufficient, but not necessary for the transition  V. Pedestal formation  Measured E r at 4 radial points with  t = 0.2ms  Synchronous bursts of Thomson scattering.   t  10ns, every 0.2ms.  Sequence of L-H, H-L and L-H transitions forced using the magnetic configuration [4]  Improved statistics, jitter of a few ms. Prior to L-H transition:  No change in E r,  E r, T e,  T e, n e or  n e. After L-H transition:  Shear layer develops at R-R sep ~ -1 cm  Increase of n e and  n e.   Er = E r /  t E r ~ 0.6 ms,  n ~ 3 ms;  L-mode filaments vanish in less than 0.1ms.  visible light acts as proxy for Note t LH is defined slightly differently in these graphs: T e, n e : top of last dither (reproducibility). E r : top of first dither. vis. light: bottom of last dither.  Ion temperature profiles are flat at low *  CX: C 6+ + D * (n=2)  C 5+ (n=8) + D +  localised gas puff; 10ms <  t < 20ms)   T i increases with increasing *.  Measurement averages over ELMs at high *.  Similar density, but different I p, B t and T e, i.  Profiles show dimensionless * scan.  T i ~ T e at high collisionality.  MAST ions are in the banana regime.  Flat profiles consistent with  Conservation of entropy [5].  T i >> T e changes total p for stability calc..  Usually p e = p i assumed  clearly wrong. t L/H transition  t = 13µs t Range shown in the 2D images of visible light 1234 5678 9101112 13141516 17181920 21222324 25262728 [1] E. Righi et.al., Nucl. Fusion 39 (1999) 309 [2] F. Ryter et.al., Nucl. Fusion 49 (2009) 062003 [3] Y. Andrew et.al., Plasma Phys. Control. Fusion 46 (2004) A87 [4] H. Meyer et.al. Plasma Phys. Control. Fusion 50 (2008) 015005 [5] G. Kagan et.al. Plasma Phys. Control. Fusion 50 (2008) 085010  Edge pitch angle,  m, measurement:  MSE (  t = 2ms,  R = 2cm)  2D analysis of EBE (  t ~ 40 ms,  R ~ 0.2 cm depends on  n e ).  MSE resolves inter ELM period, but spatial resolution is marginal.  Total edge current exceeds neoclassical prediction.  MSE: wider profile  spatial resolution?  EBE: narrow profile with high  m.  Evolution of  p and j  suggest peeling-ballooning model incomplete.  Profiles unstable with flat T i and