INSTITUTE OF PHYSICS PUBLISHING and INTERNATIONAL ATOMIC ENERGY AGENCY NUCLEAR FUSION

Nucl. Fusion 43 (2003) 1540–1554 PII: S0029-5515(03)68825-3

Overview of JET resultsJ. Pamela, Emilia R. Solano, and JET EFDA Contributors1

EFDA Close Support Unit, Culham Science Centre, Abingdon, Oxon OX14 3EA, UK

E-mail: Emilia.Solano@jet.efda.org

Received 4 December 2002, accepted for publication 5 September 2003Published 1 December 2003Online at stacks.iop.org/NF/43/1540

AbstractScientific and technical activities on JET focus on the issues likely to affect the ITER design and operation. Ourunderstanding of the ITER reference mode of operation, the ELMy H-mode, has progressed significantly. Theextrapolation of ELM size to ITER has been re-evaluated. Neoclassical tearing modes have been shown to bemeta-stable in JET, and their beta limits can be raised by destabilization (modification) of sawteeth by ion cyclotronradio frequency heating (ICRH). Alpha simulation experiments with ICRH accelerated injected 4He beam ionsprovide a new tool for fast particle and magnetohydrodynamic studies, with up to 80–90% of plasma heating byfast 4He ions. With or without impurity seeding, a quasi-steady-state high confinement (H98 = 1), high density(ne/nGW = 0.9–1) and high β (βN = 2) ELMy H-mode has been achieved by operating near the ITER triangularity(δ ∼ 0.40–0.5) and safety factor (q95 ∼ 3), at Zeff ∼ 1.5–2. In advanced tokamak (AT) scenarios, internal transportbarriers (ITBs) are now characterized in real time with a new criterion, ρ∗

T. Tailoring of the current profile withlower hybrid current drive provides reliable access to a variety of q profiles, lowering access power for barrierformation. Rational q surfaces appear to be associated with ITB formation. Alfven cascades were observed inreversed shear plasmas, providing identification of q profile evolution. Plasmas with ‘current holes’ were observedand modelled. Transient high confinement AT regimes with H89 = 3.3, βN = 2.4 and ITER-relevant q < 5 wereachieved with reversed magnetic shear. Quasi-stationary ITBs are developed with full non-inductive current drive,including ∼ 50% bootstrap current. A record duration of ITBs was achieved, up to 11 s, approaching the resistivetime. For the first time, pressure and current profiles of AT regimes are controlled by a real-time feedback system, inseparate experiments. Erosion and co-deposition studies with a quartz micro-balance show reduced co-deposition.Measured divertor thermal loads during disruptions in JET could modify ITER assumptions.

PACS numbers: 52.55.s

1. Introduction

With its divertor configuration, plasma size, heating, currentdrive and diagnostic systems, tritium, beryllium and remotehandling capabilities, the JET device can access a wide rangeof operating regimes in experimental conditions very close tothose of a burning plasma experiment. Prior to and during thedesign phase of ITER, JET produced major contributions to theITER physics basis [1] used to extrapolate plasma performanceto ITER. In the last two years, the JET research programme hasfocused on issues critical to finalizing the details of the ITERdesign and preparing its operation.

The shaping capability of JET has been extended towardsa plasma shaping very close to that of ITER, in particular withhigh triangularity, δ ∼ 0.5, achieved at up to 2.5 MA. Relevanthigher current experiments require higher power, expected tobe available in future experimental campaigns. The shapingcapability has been further extended by removal of the domeand septum from the MarkII-gas box (GB) divertor. The new

1 See complete list of contributors in the Appendix.

configuration is called the MarkII-septum replacement plate(SRP) divertor. Operating at or near the foreseen ITER plasmashape has proved to be essential due to the strong effect ofparameters like triangularity on edge localized modes (ELMs),ELMy H-mode performance at high density and advancedmodes. ITER operational scenarios can be optimized in ahighly relevant and consistent range of parameters achievedsimultaneously (δ, κ, q95, as well as ρ∗, ν∗, etc).

In addition, diagnostic and analysis capabilities have beenimproved. New or upgraded diagnostics commissioned inthe last two years include correlation reflectometry, electroncyclotron emission (ECE) profile system, quartz micro-balance (QMB), improved motional Stark effect (MSE),shape controller upgrade, halo current diagnostics and pelletspectrometers. A real-time control system has been developedand proved to be successful in demonstrating the capability tocontrol pressure or current profiles of plasmas with internaltransport barriers (ITBs).

This overview paper presents only a selection of recentJET results, first concentrating on advances in the two mainoperating regimes foreseen on ITER, the ELMy H-mode and

0029-5515/03/121540+15$30.00 © 2003 IAEA, Vienna Printed in the UK 1540

Overview of JET results

advanced tokamak (AT) scenarios, and then describing newresults in various operating regimes. More details, as well asother results, can be found in [2–19].

2. ELMy H-mode

One of the key scientific programmes conducted on JETis dedicated to further consolidating the ITER referencescenario, the ELMy H-mode [20, 21]. ELM studies arepresented in section 2.1, followed in section 2.2 by keycore magnetohydrodynamic (MHD) and fast particle issues,such as neoclassical tearing modes (NTMs) and α-simulationinvestigated in ELMy H-modes. Section 2.3 summarizesperformance achievements: high confinement at high densityand high β.

2.1. Edge localized modes

Dedicated studies have investigated ELM size scaling [2],power loads at the divertor target [3], techniques for ELMamelioration and mitigation [4], modelling and MHD stability[5,6] in the usual ELMy H-mode as well as in AT scenarios [7].The new expectations for Type I ELMs in ELMy H-mode inITER now span the tolerable range.

2.1.1. Scaling of ELMs with pedestal characteristics, ITERpredictions, mitigation. Characterization of the pedestalregion in JET ELMy H-mode with Type I ELMs shows thattypically the plasma pressure collapses in about the outer15–20% of the outer plasma minor radius in a ∼200 µstimescale. Increasing the ne,ped value (the electron densityat the top of the pedestal) decreases the relative drop in Te,ped

(electron temperature at the top of the pedestal) but not in ne,ped

(convective ELM losses). For some configurations, Type IELMs with purely convective energy losses are identified [22].

Figure 1. Loss power balance of a gas scan. At the highest pedestaldensity, average ELM losses are reduced, indicating an additionalloss mechanism between ELMs. Input power, Pin, radiation lossesPrad,bulk and plasma stored energy, Wth, remain relatively constant(also, see section 2.3.1).

New results at high δ in JET experiments have showna break in the relation between Type I ELM frequency andsize with increasing pedestal density, as illustrated in figure 1.At high ne,ped the Type I ELM energy loss decreases, dueto a reduced ELM frequency (fELM) and a slight decreaseof �WELM (the change in diamagnetic energy at the ELM).This anomalous behaviour of the Type I ELM frequency isassociated with an enhanced level of fluctuations in betweenType I ELMs with characteristics similar to those seen inType II ELMs [23]. The anomalous ELM behaviour has beeninstrumental in demonstrating that the ELM energy loss isdetermined by the values of the pedestal plasma parameters,while the ELM frequency is a consequence of the in-betweenELM transport. Defining the pedestal energy as Wped =1.5ne,ped(Te,ped + Ti,ped)V , with V = plasma volume, anempirical correlation was found between the normalized Type IELM energy loss, �WELM/Wped, and the plasma collisionalityat the pedestal top, ν∗

ped, for a large range of plasma conditions,as shown in figure 2. Alternatively, the �WELM/Wped data arealso well ordered by the ion transit time along the connectionlength, τ||ion, calculated with ion pedestal values. This couldbe related to the impedance to heat flux through the divertortarget sheath during the ELM and the connection of the pedestalplasma with the divertor target along ‘broken’ field lines [24].

Estimates of expected divertor lifetime in ITER havebeen reviewed and modified [25, 3] as data from JET, otherdevices and material tests provided a more refined basisfor modelling of acceptable ELM power load on targetcomponents. The heights of the vertical arrows in figure 2 markthe present estimates of the acceptable ELM energy in ITER,assuming either 60% (green arrows) or 100% (blue arrows) ofplasma energy loss at the target (50–80% observed at JET).The upper and lower values of each arrow originate fromdifferences in the assumptions of ELM power deposition timeand ITER divertor target options [25].

A different approach to controlling the impact of ELMson plasma facing components (PFCs) is ELM mitigation by

0.20

0.15

0.10

0.05

0

0.25

0.10 1.000.01

∆WE

LM /

Wpe

d

ν∗ ped (neo)

JG02

.448

-1c

ν∗ ITER

∆WdivELM ~ 60% ∆WELM

∆WdivELM = 100% ∆WELM

δ = 0.33 1.9MA/2.0Tδ = 0.41 2.5MA/2.47Tδ = 0.50 2.5MA/2.7Tδ = 0.45 2.5MA/2.4T Ar

δ = 0.25 3.5MA/3.0Tδ = 0.39 2.5MA/2.7Tδ = 0.33 2.5MA/2.7TBefore Pelletδ = 0.33 2.5MA/2.7TAt Pelletδ = 0.22 2.5MA/2.4T Ar

Figure 2. Scaling of Type I ELM size with pedestal collisionality,calculated at the top of the pedestal. Pellet discharges arerepresented by two points: ‘before pellet’ displays the plasmacondition into which the pellet is injected, ‘at pellet’ reflectsconditions for the pellet-triggered ELM.

1541

J. Pamela et al

impurity seeding, which has been demonstrated in JET [9, 4].With Ar, in Type I ELMy H-modes, the increase in edgeradiated power leads to a reduced target peak heat load,measured by IR thermography of the divertor target plates.Ar seeding leads to a decrease in the ELM frequency and areduction in �WELM/Wped, correlated with a change in thepedestal parameters (increase of ν∗

ped), as can be seen in the Arseeded data presented in figure 2.

2.1.2. MHD stability of the H-mode edge pedestal. Thestability of H-mode edge pedestals has been analysedconsidering medium to large n kink/ballooning modes[26, 23] and including stabilizing diamagnetic effects [27]. Inthe second stable regime for infinite-n ballooning modes, themedium-n ballooning modes set the actual limit on the pressuregradient: higher than first stability but not second stable.Integrated transport/MHD modelling is being used to study theeffect of edge density on plasma collisionality and therefore onbootstrap current and stability. The JETTO predictive transportcode has been connected with the MISHKA and IDBALLMHD codes, so ELM models can be explored [5]. Study ofELM precursors and post-cursors and inter-ELM activity isongoing, as described in [28, 6].

Controlling ELMs is important in AT scenarios as largeType I ELMs can destroy the ITB [29]. In typical AT dischargesin JET it has been observed that Type III ELMs can bemaintained at heating powers exceeding the usual thresholdfor the Type III to Type I ELM transition in H-mode by afactor of 2 [30]. Nevertheless, at a high triangularity, Type IELMs are particularly difficult to avoid, even in AT scenarios.Experiments with current ramps were carried out to study theeffect of changes in the edge current density on the ELMbehaviour of high triangularity plasmas with a previouslyestablished ITB. With that technique, it was found that Type IELMs can be transiently suppressed in AT scenarios [7].

2.2. Core MHD issues in ELMy H-mode

In ELMy H-mode discharges, the core MHD, which candepend on kinetic effects, can have an impact on confinement.

2.2.1. Sawteeth and NTMs. In JET, it has long beenknown that MHD activity triggered by large sawteeth canreduce confinement. Notably, during the JET DT series,sawtooth control raised fusion power by 30% from a saturatedvalue of 10 MW. This was obtained by delaying the sawtoothcrash (fine-tuning the gas puff rate) and therefore avoidingconfinement degradation. Recent modelling shows that fastparticle stabilization leads to larger sawteeth, triggering NTMsand reducing confinement in these fusion-relevant plasmas[31]. This lends credibility to α-simulation, discussed laterin this section, where equivalent evolution is observed.

Concern about the effect of NTMs on confinement inITER prompted a series of studies in JET on NTM stability,in ELMy H-mode plasmas [8, 32]. In neutral beam powerramp experiments, it is found that the onset β-limit for NTMsdepends very much on the seeding process. Most commonly,the NTM seed perturbation is a sawtooth, suggesting theopportunity for NTM control via control of the sawtooth seed.Detailed scans have shown that heating and current drive from

second harmonic ion cyclotron radio frequency (ICRF), whenlocated near the sawtooth inversion radius, cause small andfrequent sawteeth. This sawtooth control has been exploitedto raise the 3

2 NTM onset β-limit in JET discharges up toβN = 3.6 [33]. Conversely it is found that long period sawteethare found to result in a low onset βN for the 3

2 NTM, as shownin figure 3 (partly derived from [34]). It remains unclear whysawtooth periods, rather than measurements based on magneticprecursor amplitude or soft x-ray crash size, are best correlatedwith NTM thresholds. By generating long period sawteeth,generally by ICRH (ICRF heating) fast particle stabilization,NTMs can be accessed at a high toroidal field in JET; allowingthe properties of NTMs to be studied at ρ∗ values to withina factor of ∼1.5 of the ITER Q = 10 value (albeit at lowerβN values, where the confinement degradation effects of theNTMs are modest). Direct power-law type extrapolations forthe onset βN are of dubious value, given the non-linearitiesintroduced by the seeding process. Indeed βN and ρ∗ scalingslargely reflect the natural discharge evolution, and we shouldnote that even with a correction for collisionality dependence infigure 3 (based on fits in [34]), there remains an offset in trendlines for NTM onset between the smaller devices and JET.However studies of the marginal-β (below which the NTMsare unconditionally stable), as opposed to the onset βN, arenot complicated by the effects of the seeding process. Suchstudies reveal that typical H-mode conditions in JET are meta-stable to 3

2 NTMs (which is why long period sawteeth cangenerate NTMs at low βN), and initial scalings further indicatethat the same will be true in ITER [35], where NTMs maythen be generated by long period sawteeth occurring due toα-particle stabilization effects. Overall, these results indicatethe importance of developing NTM control strategies for ITERand of further studying the physics of the seeding process topredict onset-βN limits for ITER; JET is well placed to makesuch studies in ITER-relevant regimes with access to low ρ∗

and large sawteeth.

Normalized

Figure 3. Onset βN for 32 NTM, as a function of ρ∗

i (toroidal Larmorradius at q = 3

2 divided by minor radius). The DIII-D andASDEX-U data are corrected for collisionality dependence (seefigure 6 of [34]), whereas the JET data shows no significantcollisionality dependence.

1542

Overview of JET results

2.2.2. α-simulation. On-axis ion cyclotron resonanceheating of neutral beam injected 4He ions (third harmonic)produced a high energy population of 4He ions, which simulate3.5 MeV fusion-born alpha particles [36]. The strongest tailswere created with the highest energy 4He seed beams, asexpected. In these stationary plasmas, fast 4He provided upto 80–90% of plasma heating. The successful acceleration of4He beam ions to the MeV energy range was confirmed bymeasurements of γ -ray emission (see section 4.1.1) [37] andexcitation of Alfven eigenmodes [38, 10] and was consistentwith the observed heating of the background electrons andsawtooth stabilization. This experiment provided yet anotherexample of the interconnection between various physicalmechanisms: fast particles stabilize sawteeth, which acquirea large amplitude and trigger NTMs, and confinement isdegraded (see [8]). The scheme can now be used inforthcoming JET campaigns with 4He plasmas for dedicatedalpha particle simulation studies.

2.3. Integrated ELMy H-mode scenarios

High performance ELMy H-mode integrated scenarios havebeen achieved with various methods [20, 9], as displayed intable 1 and described in sections 2.3.1–2.3.4. Scaling resultsare presented in section 2.3.5. At reduced performance, asimulation of fusion burn dynamics was undertaken, describedin section 2.3.6.

2.3.1. High confinement at high density, achieved withhigh triangularity, high power. Experiments studying thebeneficial effect of high triangularity [23] and high heatingpower [39] on confinement have confirmed and extendedearlier results demonstrating a high H factor at high density, asshown in table 1. These discharges have a lower single null,with ion grad B drift towards the X-point, standard q profile(monotonically increasing from axis to edge), NBI (neutralbeam injection) heating and gas fuelling. The separatrixgeometry was similar to that envisaged for the standardQ = 10operation in ITER, with average triangularity δ ∼ 0.45–0.5 andelongation κ ∼ 1.75. Figure 4 illustrates the beneficial effectof high triangularity on confinement at high density, up to theGreenwald density (nGW).

In those high δ, high density plasmas, a decrease in Type IELM frequency with density is observed, associated witha high pedestal density: ne,ped � 0.7nGW. The reductionof fELM is associated with high frequency fluctuations in

Table 1. Performance achievements in ELMy H-mode in JET.

High δ, high power; High δ, Ar seeding;#53299 (Type I–II) #53550 (Type I) ITER

H98(y,2) 0.91 0.96 1.0βN,th 2.0 2.0 1.81n/nGW 1.1 1.0 0.85Zeff 1.5 2.2 1.7Prad/Ptot 0.4 0.7 0.58κ, δ 1.74, 0.48 1.7, 0.4 1.84, 0.5q95 3.2 3.1 3.0τpulse/τ

aE 15 10 110

a Limited by technical constraints only.

the Dα emission (and enhanced Dα baseline level) in theinter-ELM period, accompanied by the appearance of acharacteristic broadband MHD turbulence and by an increasein density fluctuations in the pedestal region. This inter-ELMactivity is also associated with a high pedestal pressure andincreased plasma power losses (see section 2.1.1 and figure 1).A comparison with results from Asdex-U [40] suggests thatthese high frequency Dα oscillations could be identified asType II ELMs. Plasmas where Type I and Type II ELMs coexistare also observed in both Asdex-U and JT-60U [41], when thecritical parameters for complete suppression of Type I ELMsare approached.

2.3.2. Impurity seeding. Stationary high performance phases(up to 12τE) have been obtained with impurity seeding,simultaneously satisfying or exceeding the ITER Q = 10ELMy H-mode requirements for H98(y,2), ne/nGW and βN [9],as shown in table 1. In low δ plasmas, with the X-pointon the divertor dome, there is a pronounced peaking of theelectron density profile. In high δ plasmas, the formation of aradiating belt inside the separatrix has been clearly establishedwith high radiative power fractions up to 70%. Centralimpurity accumulation was avoided by application of ICRH,maintaining the safety factor on-axis below 1, so sawteethexpelled impurities [42].

2.3.3. High confinement at high density, obtained by pelletinjection. An optimized pellet sequence has been developed,consisting of a first phase where the density is increased (whileallowing a small confinement loss), followed by a second phaseminimizing the particle flux in order to maintain the densityand to recover confinement. In this way, simultaneouslyH98(y,2) ∼ 0.8 with βN > 1.8 and ne ∼ nGW have beenreached, albeit not yet under stationary conditions. The plasmadensity profile shows a large peaking with ne,axis/ne,ped = 2and a decrease in Zeff to about 1.7 is obtained at the highestdensities reached [43].

Figure 4. H factor versus Greenwald factor, as a function oftriangularity™. Marked GB are data points from the previous GBconfiguration of the divertor (closed), and HT3 is a new hightriangularity plasma shape in the new divertor.

1543

J. Pamela et al

2.3.4. Density peaking. Long timescale density peaking,without pellet injection, has been observed in JET plasmas,leading to densities exceeding the Greenwald value [44].These neutral beam heated discharges are characterized byType I ELMs and good energy confinement. The centraldensity is limited by NTMs or by termination of the H-mode,preceded by loss of sawteeth. When these limiting factorsare avoided (i.e. at an intermediate power level and withoptimization of gas puffing), quasi-stationary high densityplasmas with peaked density profiles are obtained. Whetherthe peaking is associated with an anomalous pinch is an issuestill being investigated.

Analysis shows that this density peaking is consistent withthe Ware pinch if the ratio of particle and heat diffusivityis D/χeff ∼ 0.25. Assuming a strong anomalous pinch, oforder V = 10 × VWare, would result in a particle diffusioncoefficient of order D/χeff ∼ 1. The modelling of alarge number of density profiles in H-mode did not allowdiscrimination between these two hypotheses. On the otherhand, an anomalous particle pinch must be invoked to explainthe evolution of the density profile in L-mode, in particularwith pellet injection [45]. In H-mode, simulations of iontemperature gradient (ITG) and trapped electron mode (TEM)driven turbulence exhibit a clear anomalous pinch. Resultsin L- and H-modes may be reconciled by noting that a coreH-mode plasma is less turbulent than one in L-mode. Theactual pinch velocity is the sum of neoclassical and turbulentcontributions, whereas the diffusion coefficient is essentiallythe turbulent one. The Ware pinch is therefore dominantin quiescent plasmas. This hypotheses is currently underinvestigation at JET.

2.3.5. Scaling laws. The influence of peaking, triangularityand proximity to the Greenwald limit on energy confinementscaling has been investigated and has a minor effect onconfinement predictions for ITER (<10% in τE) [46]. A setof experiments in helium [47] (with purity CHe/CD = 85%,where C indicates concentration) added to our knowledge ofthe scaling of confinement with Z: τE ∝ Z−0.75±0.08. TheL–H threshold power in He plasmas shows the same Bt andne dependence as in D plasmas but is 42% (±10%) higher inabsolute value.

A two term scaling law has been developed in the frameof the International Tokamak Physics Activity (ITPA) H-modeDatabase Working Group [48]. The best fit for an ELMy-Hmode is compatible with an L-mode core and a pedestalconfinement that depends sensitively on β. This findingis not compatible with strongly stiff profiles since in thiscase the central temperature is proportional to the pedestalheight. The question of stiffness has been investigated fromboth the experimental [11, 17] and theoretical sides. ICRHmodulation experiments with mode conversion in L-modeshow the existence of a threshold for electrons [12]. Forions, evidence has been obtained from steady-state profilesin H-mode [49]. In general, a transport model based on acritical gradient gives a satisfactory prediction of confinement,but with a moderate stiffness factor. In such a case, theglobal confinement time obtained with the values of theinstability threshold and medium stiffness factor found in JETis compatible with the ITPA two term scaling law.

2.3.6. Burn control simulation. ICRH, via fast ions, mainlyheats the electrons and is centrally deposited. These arecharacteristics of plasma self-heating from alpha particles. Inone experiment a component of the ICRH heating, Pα,sim, wasmade proportional to a scaled fusion reaction rate, accordingto measured plasma parameters. Two scenarios were used:(a) Pα,sim was made proportional to the change in the measuredDD reaction rate when ICRH was superposed onto 2 MW NBI(found to vary approximately as T 2

e,axis) and (b) Pα,sim was madeproportional to n2 and to a parametrized function, f (Te,axis,volume average Te), reflecting the temperature dependence ofthe DT reaction rate. A separate component of either ICRHor NBI was used in the role of auxiliary power, Paux. Thedischarge trajectories were programmed to simulate a reactorsituation approaching ignition: magnetic field, power level anddensity were all chosen to lead to an L–H transition towards theend of the Paux ramp, when the sum Paux + Pα,sim reaches theL–H transition power threshold. The real-time control systemacted on the Paux heating component to control the simulatedburn, precluding a ‘thermal runaway’, as shown in figure 5.More details on these experiments are presented in [11,50,51].

2.4. Future plans, in ELMy H-mode regimes

The plasma characteristics shown in table 1 demonstrate netprogress towards the ITER targets, together with encouragingdevelopments in ELM scaling. Extensions of ELMy H-modestudies to plasmas with higher currents, fields and heatingpowers and longer flat-top duration are planned. Plasmas withhigher absolute densities and nearer to ρ∗

ITER can be developed,testing the limits of high performance regimes, enabling studiesof Type I ELM scaling at low, ITER-like collisionality, and ofNTM scaling. New plasma configurations with ITER valuesof κ and δ have already been developed and tested with currentup to 2.5 MA. The estimated disruptive force implies that suchplasmas can be explored, up to 4 MA/4 T, when more auxiliarypower becomes available. NTM control techniques will betested. Work on Type II ELM regimes will continue.

3. AT research on JET

The ultimate goal of AT research is to provide steady-state operational regimes with possibly improved fusionperformance. These modes are candidates for the steady-state operation of ITER [1]. Presently, AT research on JET ismainly focused on plasmas with ITBs [52, 7]. In section 3.1,experimental and modelling progress in ITB physics isreported. In section 3.2 the achievement of high performanceand long pulse ITB plasmas and the demonstration of real-time control of pressure and current profiles are presented.Comments on future plans are given in section 3.3.

3.1. ITB physics

3.1.1. A novel criterion for characterization of ITBs. Barrierformation and strength are operationally characterized in JETby ρ∗

T = ρs/LT (ρs: ion Larmor radius, at the ion sound speed;LT: local Te gradient length). Analysis of the JET ITB databaseand of a set of turbulence simulations led to the ITB onsetcriterion ρ∗

T > 0.014 [13, 53]. ρ∗T can be computed in real

1544

Overview of JET results

0

0.5

0

1.0

10

40

0

2

4

8

6

18 20 22 24 26 28Time (s)

JG01

.137

-5c

Pulse No: 52608 2.5MA/2.5T

(MW

)(a

.u.)

Qeff

Pα,sim

Dα

Paux

Control reference

Figure 5. ICRH in the roles of Pα,sim and Paux in a simulation of aburning plasma. Pα,sim depends upon measured plasma parametersaccording to scenario (a) defined in the text. Pα,sim is stabilizedusing feedback control of the Paux component, and a step-change inthe Pα,sim demand level is illustrated.

Te

Te

Figure 6. Plot of ITB criterion versus time and plasma radius.When qmin crosses 2, the barrier is broadened (radius increased by∼30 cm) and two barriers follow the evolution of the q = 2 surfaces.The time when qmin crosses 2 is confirmed from the excitation ofAlfven cascades. At either side of qmin, the magnetic shear, s,changes sign as indicated.

time, characterizing ITB formation and evolution (see figure 6)and providing an input for real-time control (section 3.2).Obviously this criterion does not capture the special role ofrational q surfaces and magnetic shear, on which we commentbelow (section 3.1.4).

3.1.2. Lower hybrid current drive aids current profile shaping.Coupling of the lower hybrid (LH) wave in all types of plasmaswas greatly improved by puffing CD4 near the launcher andmatching the plasma shape [54–56]. Further optimization ledto a dramatic increase in the availability of the LH system, withpower levels in the 3–4 MW range routinely coupled to highpower 15–20 MW plasmas. The lower hybrid current drive(LHCD) became an efficient tool to access ITBs and controlthem through off-axis current drive. Off-axis co-LHCD inthe prelude phase (current ramp-up) of JET plasmas easilyand reliably provides magnetic shear reversal, providing targetplasmas with appropriate current profiles, often exhibiting anearly ‘electron barrier’. The current profile is then ‘frozen-in’by application of heating (NBI and ICRH). The applied heatprevents fast evolution of the current profile by raising theplasma temperature and reducing resistivity, and electron andion ITBs develop.

3.1.3. Current holes. MSE measurements of the poloidalmagnetic field [57] showed plasmas with essentially no toroidalcurrent density in the core of JET ITB discharges (a regionof up to 20 cm diameter). These were explained as an effectof the non-inductive, off-axis, increasing co-LHCD. The timederivative in the LH driven co-current induces a negativevoltage in the plasma core, therefore reducing core currentdensity. Similar ‘current holes’ were observed in JT-60Ubut driven solely by NBCD (neutral beam current drive) andbootstrap [58,59]. The core current density does not appear tobecome negative (error bars on plasma core current are ±10–20 kA), although current diffusion calculations indicate thatthere is sufficient core negative voltage to cause current reversal[60]. This apparent clamping of the core current density nearzero is consistent withn = 0 reconnection events redistributingthe core current soon after it goes negative, as shown in reducedresistive MHD simulations in a cylindrical geometry [61] andin two-dimensional non-linear resistive MHD simulations in atoroidal geometry [62].

3.1.4. Rational q surfaces and transport barrier trigger.Analysis of both low shear and reversed shear ITBsdemonstrates the key role played by the q profile itself and, inparticular, by the presence of low order rational q values, suchas q = 1, 2 or 3 surfaces in the plasma core, near the position ofthe minimum in q, qmin [14]. The formation of low shear ITBscan also be well correlated with the occurrence of rational edgeq values, qedge = 4–6. A model of coupled magnetic islandsdoes show that a significant amount of rotational shear can begenerated around qmin = 2 by an edge magnetic perturbationwith rational qedge, in agreement with measurements [63].With reversed magnetic shear, experimental evidence alsosuggests a link between the appearance/development of anITB and rational q surfaces (e.g. qmin = 2) entering the plasmacore [64], as shown in figure 6. Note that the effect of integer q

values as an ITB trigger is only evident when marginal heatingpower is applied: with sufficient power, barrier formation canbe independent of the rational values in the q profile. Possibleexplanations for the influence of q on ITB formation areput forward: coupling of MHD modes located at rational q

surfaces can generate a stabilizing velocity shear, and gaps inthe density of rational surfaces near rational q surfaces can have

1545

J. Pamela et al

a stabilizing effect on low wave number ITG/TEM turbulence[13, 14]. Simulations of ITG/TEM turbulence support thissecond explanation, showing that a barrier is more easilytriggered when qmin is close to a low order rational surface,provided the magnetic shear is low enough. The coexistenceof two barriers located on a double q = 2 surface is alsodemonstrated.

Further, ITG/TEM simulations of ITB plasmas success-fully reproduce the amplification of a cold pulse at the outeredge of the barrier, as observed in JET when launching ashallow pellet or injecting impurities [12]. Cold pulses aredamped inside the ITB region, as expected in a region withlow diffusivity.

3.1.5. Alfven cascades. In JET reversed shear ITB plasmas,cascades of Alfven modes sweeping upwards in frequencyhave been observed [65], similar to observations in JT-60U[66]. Each cascade consists of many modes, driven by ICRF-accelerated ions. These Alfven cascades have been interpretedin terms of a novel type of energetic particle mode [67–69]localized at the point where q(r) has a minimum. Thefrequency of the eigenmode tracks the evolution of qmin, whichopens the possibility of Alfven spectroscopy measurement ofq(r, t) [70]. Indeed Alfven cascades are now used as anauxiliary rational q measurement on JET, as well as identifiersof shear reversal. ‘Grand-cascades’, sets of various toroidalAlfven eigenmodes (TAEs) with (m, n) values correspondingto integer qmin, can be used to diagnose the qmin value as afunction of time. The intrinsic interest of TAEs and cascadesin potential steady-state burning plasmas is being explored byvarious groups [10, 38, 71].

3.2. High performance and steady-state ITBs

AT scenarios have been extended in two directions: improvedconfinement and steady state. The ultimate objective is tocombine both, together with appropriate density.

3.2.1. High performance ITBs. In transient states, reversedshear high performance plasmas with wide ITBs (r/a ∼ 0.6)have been achieved. Given sufficient heating power, thesestronger ITBs (with larger value of ρ∗

T) are triggered whenqmin reaches 2. Within experimental uncertainties, ITBs aredetected simultaneously and at the same location in electronand ion temperatures (Te and Ti), density and plasma rotationprofiles.

With reversed magnetic shear, the access power is low, andβN and the fusion yield for a given input power are high. Theparameters of one of the highest performance ITB plasmas areshown in table 2 [72]. Notably, βN = 2.4, and q95 = 4.9, with2.5 MA of plasma current. The current profile is broad, and themaximum available heating (4 MW ICRH and 17 MW NBI)was used. Density peaking leads to impurity accumulation[15]. In this case, the transient high performance phase endswith a large ELM, followed by a disruption (global pressuren = 1 kink mode instability).

3.2.2. Record long pulse ITBs. By tailoring the internalplasma profiles to keep confinement and stability under control,the duration of ion ITBs could be prolonged up to 7.5 s ofplasma flat-top, corresponding to 27×τE [73, 74, 7]. Thedischarge parameters are presented in table 2. This is a recordduration for an ion ITB in JET. In this case, an electron ITB wasformed during the plasma current ramp-up and had an overallduration of 37 × τE . In this long duration ITB discharge,the fusion performance is lower than in transient cases (seetable 2), while the non-inductive current fraction was muchlarger. Radiative or MHD core collapses of the ITB areobserved: plasma pressure, toroidal rotation and electric fieldscollapse on a fast timescale. Surprisingly, the ITB is reformedimmediately after the collapses. The implication, supported byturbulence simulation studies, is that as long as the q profileis not affected by the collapses, it remains favourable for ITBformation, and the ITB will re-form. At a slightly lower plasmacurrent of 1.8 MA, the full non-inductive current drive wasrealized, as shown in figure 7.

3.2.3. Real-time control of pressure and current profiles.Achieving steady-state real-time control of ITBs requiresdiagnostics allowing reliable characterization of the keyparameters and appropriate actuators. The new ITB criterion,ρ∗

T, is used to characterize ITBs in real time; interfero-polarimetry provides a q profile measurement and the neutronproduction rate characterizes ITB plasma performance. Allthese real-time measurements are used as input values for thenew real-time control system. The main actuators used arethe three heating systems, LHCD, NB and ICRH, all havingdifferent (but coupled) effects on ITB behaviour.

The control tools have been used in a first series ofreal-time feedback demonstrations, based on the largely non-inductive long pulse discharges described above. Pressureprofile and barrier strength control are illustrated in figure 7.Having established an appropriate target current profile,reliable and reproducible ITB operation is achieved bycontrolling the neutron yield by NBI heating, while ICRH isused to act on the maximum electron temperature gradientdirectly [75]. The LHCD is maintained in a pre-programmedway to provide a broad q profile, ensuring core confinementand stability. During the control phase, the plasma parametersof the discharge are fairly stationary with mild and continuousELM activity. Thanks to the real-time control of the ITBcharacteristics, the improved confinement state is maintainedaround its reference in a reproducible and stationary manner,e.g. avoiding the occurrence of core collapses. The real-time control of the ITB strength has been achieved withinthe technical limits of the heating and current drive systems.In separate experiments, it has also been shown that theLHCD feedback can be used in conjunction with real-timemeasurement of q to maintain a desired q profile fixed forseveral seconds [76].

Here we should comment that the applicability of thesecontrol strategies to ITER and to a fusion reactor is notnecessarily direct: pressure and bootstrap profiles would belargely determined by fusion heating. The external controlsystems would then be used to steer the plasma towardsdesirable initial conditions, such that the plasma could developa self-sustained AT regime, and to control the plasma around

1546

Overview of JET results

Table 2. AT scenario development on JET.

High performance Long pulse ITER targetJET ITBs (pulse 51976, (pulse 53521, as derived fromand t = 6.8 s, Ip = 2.5 MA, t = 10 s, Ip = 2.0 MA, ITER designITER target Bt = 3.45 T) Bt = 3.45 T) basis [1]

HH(IPB98y-2)(thermal) 1.9 1 >1.5HITER89L-P 3.3 2.1βN 2.4 1.7 >3.0INon-inductive/Ip (%) 55 90 100Iboot/Ip (%) 40 50 >60q95 4.9 5.5 4.5Prad/Ptot (%) 15 23 75nio/neo 0.7 = (3.85/5.5)1019 0.65 = (4/6.2)1019

τ/τE, τ/τresistivea 2, ∼0.1 27, ∼1

neo/nGW, nel/nlGW 0.65, 0.35 0.8, 0.5 nel/nGW > 0.6Edge Type III ELMs Type III ELMs Mild ELMs

a τ = duration of ITB regime; τresistive is calculated at the ITB foot-point.

Te

Figure 7. Time evolution of the main parameters of a dischargewith combined ρ∗

Te and RNT (total neutron yield) feedback controlusing ICRH and NBI powers (#53697, Bt = 3.4 T, Ip = 1.8 MA).

a desired reference state. Nevertheless, the JET real-timecontrol tools can contribute to development of steady-state,fully controlled, AT scenarios, hopefully resulting in a deeperunderstanding of the situations to be expected and desired in asteady-state reactor regime.

3.3. Summary and the future: high performance in steadystate

The plasmas shown in table 2 demonstrate net progress towardsthe ITER target of steady-state plasmas. In order to achievesteady-state high fusion performance, the current and pressureprofile must be controlled simultaneously, the density mustapproach the Greenwald value, impurity accumulation must be

avoided or controlled, bootstrap must be high (high pressureand high power) and edge radiation must be increased, or theSOL broadened. Operation in an ITER-like triangularity, madepossible by the recent divertor modification, contributes tomake the results more directly applicable to ITER prediction.Understanding of the physics behind ITB formation and loss,and development of integrated, steady-state, high performanceAT scenarios are clear goals of JET AT experiments.

4. Physics and technology research in variousoperating regimes

We present here some important physics studies that do notnecessarily focus on ELMy H-mode or ITB plasmas, as wellas studies on plasma–wall interactions and tritium technologyissues.

4.1. Diagnostic developments

4.1.1. Gamma rays. Nuclear reaction γ -ray diagnosis hasproved its value in studying fast ions in JET [37]. Intenseγ -ray emission is produced when fast ions (ICRF-drivenions, fusion products, NBI-injected ions) react either withfuel ions or with plasma impurities such as carbon andberyllium. Analysis of γ -ray spectra led to the identification,in different experiments, of several nuclear reactions involvingfast particles accelerated by ICRF and plasma ions. Thereaction 9Be(α, nγ )12C, proposed as a fusion γ -particlediagnostic, led to the identification of the tail α-particle withenergies in excess of 2 MeV, as described in section 2.2.2.Identification of RF induced orbit pinch was possible with thisdiagnostic (see section 4.2.2).

4.1.2. Non-Maxwellian bulk electron distributionmeasurements. Evidence of non-Maxwellian bulk electrondistribution functions was found on FTU [77] in low shearplasmas with central ECRH (∼0.8 MW) and in JET [78, 79]in regimes with high levels of ICRF and NBI power (hot ionH-mode). An analysis of the various harmonics of ECE spec-tra has been carried out, based on extensive modelling. Thedistorted distribution function obtained from the ECE data canexplain the discrepancy between the Thomson scattering (TS)

1547

J. Pamela et al

and ECE measurements of electron temperature (Te). Such dis-crepancies between ECE and TS temperatures have also beenobserved in regimes with alpha heating on both TFTR [80] andJET [81], making the understanding of this phenomenon andof its implications an important fusion issue.

4.2. ICRF studies

As well as supporting a variety of other experiments, the ICRHsystem in JET was used for some specific experiments brieflydescribed here.

4.2.1. Rotation. In future fusion devices, externalmomentum input will be much lower than in most present-day tokamaks. Rotation is known to increase the stabilizingeffect of a resistive wall; rotation shear plays a role in improvedconfinement regimes. For extrapolation to next step devices,it is important to study these issues in plasmas with lowermomentum input. The toroidal rotation of ICRF-only heatedplasmas has been investigated in the JET tokamak. Whenthe minority cyclotron resonance layer is far off-axis, thetoroidal rotation profiles are hollow, and dominantly co-current[82–84] (ions are accelerated in the direction of the toroidalplasma current). In contrast to the predictions of theoriesthat rely mainly on effects arising from ICRF-driven fast ionsto account for ICRF-induced plasma rotation, there is only aslight difference between the high-field side and low-field sideposition of this resonance layer. A more central depositionof the ICRF power in L-mode and operation in H-mode bothlead to more centrally peaked rotation profiles, both in theco-direction.

4.2.2. Pinch of fast particles. Theory predicts that ICRFwaves with a directed spectrum will influence the radialtransport of the minority particles with which they interact.Waves with a co-directed spectrum (i.e. propagating in thedirection of the plasma current) lead to an inward pinch ofthe energetic particles, and waves with a counter-directedspectrum to an outward pinch of these particles. Recentadvances [37] in the tomographic reconstruction of themeasured γ -ray emission data provide direct experimentalevidence of this effect. This result [85], together withconcurrent experimental evidence from Alfven eigenmodes,sawtooth periods, electron temperatures and fast ion energies,shows that the ICRF-induced pinch provides a tool to affectthe radial fast ion profile and the plasma heating profile.

4.3. Transient heat transport studies, cold pulses

Heat pulse propagation studies were carried out in JET, usingICRF mode conversion heating (3He in D and 4He) [86] as alocalized source of electron heating. Results of the transportanalysis [12] are consistent with the existence of a threshold ofthe inverse critical electron temperature gradient length and a‘medium’ level of stiffness in the electron channel, as discussedin section 2.3.5.

On the other hand, cold pulse propagation experiments, inL-mode and ITB plasmas, show fast propagation from edge tocore, which cannot be explained by stiffness alone: non-localmodels must be considered [12].

4.4. Disruption studies

4.4.1. Thermal loads. The current ITER assumption fordisruptions is that the energy from the thermal quench isdeposited in the divertor. Recent analysis shows that only a fewper cent of the thermal and thermal plus magnetic energy canbe accounted for in the JET divertor [87] and it is spread ratheruniformly. If the same applies in ITER, then the energy densitywithin the divertor would be below the level at which carbonablation and tungsten melt layer formation are expected [3].

4.4.2. Plasma position stability and control. Modelling ofvertical displacement events (VDEs) in JET has improved,with a better representation of eddy currents in the vacuumvessel now incorporated in the simulation codes. Thereis good agreement between theoretical predictions of VDEevolution and experimental results: growth rates have beenestimated within an accuracy of 5% for plasmas with a growthtime longer than 2 ms [88, 89]. The neutral point has beenidentified in dedicated experiments on JET [90], confirmedwith simulations done with the linearized equilibrium codeCREATE-L [88]. It has been observed that a plasma speciallydesigned to be set at different vertical equilibrium positionswithout altering its shape moves upwards (downwards) whenthe disruption is triggered by argon puffing with the plasmabelow (above) the neutral point.

4.4.3. Runaways. Regions of parameter space have beenidentified in which disruptions lead to runaway generationin JET [91]. They appear when q95 and Bt exceedthreshold values (i.e. Bt above ∼2 T), in good agreement withobservations on JT-60U [92]. The delay in runaway generationfollowing the temperature collapse is found to be caused bythe very high density generated by the disruption. It can beconcluded that avalanching is probably the dominant processin runaway electron formation in JET and therefore runawaycontrol techniques can be tested for ITER. Initial tests in thisregime have shown that it is possible to quench the plasma withlarge helium puffs without producing runaways.

4.4.4. Halo currents. The JET halo current diagnostics havebeen refurbished [93] with measurements of the poloidal halocurrent now available in three locations, 90˚ apart, at the top ofthe vessel. The old system had at most two locations 180˚ apartavailable at the top of the vessel. The toroidal peaking factor(ratio of the largest to the toroidally averaged halo current) cannow be determined with improved certainty.

The data collected in the past year of operation, whichinclude dedicated disruption experiments, indicate that themaximum halo fraction is <30% and the maximum toroidalpeaking factor is <2. This is in reasonable agreement withwhat was observed using the old set of probes and maintainsthe JET data below the envelope used for the design of ITER[1]. Thus, the improved JET data continue to validate the ITERdesign assumptions.

4.5. Erosion and co-deposition

Erosion and deposition constitute one of the key topics inpreparation of ITER operation, as they influence the lifetime

1548

Overview of JET results

of divertor tiles and other PFCs, and the in-vessel tritiuminventory. JET experiments provide a database against whichmodels can be tested [16].

Tiles from the MkII-GB divertor were removed during the2001 shutdown. Analysis of tile surfaces [94, 95] confirmedprevious observations: the inner divertor is subject to netdeposition of plasma impurities (and retention of D and T).No comparable net erosion is observed in the outer divertor,implying that the primary source of eroded impurities is themain chamber PFCs. This analysis shows that depositedfilms are enriched in Be and other metals, while carbon ischemically eroded and transported to shadowed regions of theinner divertor [96].

Measurements made during the 2001 He seriesdifferentiate physical and chemical sputtering. The mainsource of C in L-mode plasmas appears to be chemicalsputtering at the inner divertor and main chamber wall. InELMy H-mode, the situation is less clear, but the data point toa strong physically sputtered C-source from the main chamberand outer divertor [97], while chemical erosion dominates inthe inner divertor in all cases.

In Ohmic plasmas, atoms of 13C were injected into thetop of the JET vessel, at the end of the GB operation (divertortemperature had been lowered to ∼100˚C). About one-half ofthe injected 13C is found on the inner divertor tiles intersectedby the SOL, and a negligible amount is found at the outerdivertor [94], consistent with previous findings. No 13C wasfound on shadowed regions of the tiles, where otherwise thickcarbon layers have built up in the past [96]. This may indicatea reduced long range carbon transport due to reduced chemicalerosion at lower operating temperatures [95], but it mightalso be connected with the special (Ohmic) plasma operatingconditions [16].

First measurements of pulse resolved growth of carbonlayers near the inner louvre situated inside the divertorpumping throat have been made with a newly developed QMB.This diagnostic is capable of measuring deposited layers at themono-atomic layer level. The average growth rate (depositedcarbon atom on louvre divided by total ion flux into the innerdivertor) was found to be of the order of 2 × 10−4 C perion [98], 200 times smaller than previously estimated fromthe DT campaign. A thorough investigation is underway toexplain these differences.

4.6. Tritium technology R&D

Special detritiation techniques are developed for componentsreplaced during operations to allow waste disposal undereconomically sound and environmentally safe conditions[99–101]. Water detritiation techniques are also being

developed: laboratory size plants with liquid phase catalyticexchange (LPCE) columns have been built to test theperformance of various packing materials and catalysts andto determine mass transfer for D and T between water andgas [102, 103]. Design studies of a water detritiation plantat JET are ongoing [104]. During the 2001 shutdown, flakesfrom the sub-divertor region of the tokamak were collectedremotely to help reconcile the tritium balance at JET [105,106],as described in section 5.

5. Future plans for the JET facilities

The present plan of exploitation of the JET facilities for theyears 2003 and 2004 foresees experimental series in 2003and until February 2004. In 2003, the following experimentsare planned: development of ELMy H-mode and ITB modes,trace tritium experiments (TTEs) in deuterium plasmas andreversed magnetic field experiments in deuterium. The lastseries before the 2004 shutdown is reserved for hydrogen andhelium plasmas: it will provide data in preparation of ITER’snon-activation phase. From March 2004 a long shutdownperiod is planned for installation of enhancements.

The TTEs will take full advantage of the unique tritiumhandling capability of JET. The TTEs are planned with ∼1.5%tritium in deuterium plasmas. Transport of tritium particleswill be investigated with neutron and γ emission diagnostics.A study of particle (fuel ion, helium and impurities) transportis planned in ELMy H-modes and AT scenarios, continuingand greatly extending the work initiated in the previousTTE [107]. In particular, a study of fuel ion transport hasnever been done in AT scenarios and would be important inpreparation of future DT AT operation on JET or/and ITER.The use of trace tritium also offers an opportunity to studyaspects of ICRH schemes and fast particle physics (fast ionsin current hole scenarios). The TTE has the added benefitof exercising and maintaining technical expertise on tritiumhandling in a tokamak environment. It finally permits relevanttesting of new systems foreseen to find application in ITER,such as detritiation catalysts and an ITER-design cryo-panel[108, 109].

In the 2004 shutdown, it is planned to install a new in-portICRF antenna. The aims are to validate key elements of theITER ICRF design (such as ELM tolerant operation at a highpower density) and to deliver an additional 7 MW of heatingpower, allowing higher performance operation closer toITER plasma parameters. Diagnostic upgrades planned bythen include CXRS core and edge, magnetics, bolometry,neutron spectrometry, microwave waveguides and divertordiagnostics. New diagnostics to be installed include a highresolution TS system, TAE active antenna system, wideangle infra-red view, halo sensors, lost α detectors and acomprehensive system of erosion/re-deposition diagnostics.

6. Conclusions

JET continues to provide vital contributions to ourunderstanding of tokamak physics, scaling predictions forITER, further development of ITER operating scenarios andtesting of ITER relevant systems such as heating and currentdrive, diagnostics and tritium technologies.

Acknowledgments

This work was performed under EFDA by the JET EFDAcontributors listed (Appendix).. Special contributions to thispaper were made by M. Becoulet, D. Brennan, F. Crisanti,S. Ciattaglia, P. Coad, X. Garbet, T.T.C. Jones, E. Joffrin,N. Hawkes, T. Hender, G. Huysmans, V. Krivenski,X. Litaudon, P.J. Lomas, A. Loarte, D. McDonald, P. Mantica,M. Mantsinen, G. Matthews, P. Monier-Garbet, D. Moreau,

1549

J. Pamela et al

F. Nave, J.-M. Noterdaeme, J. Ongena, S. Pinches, R. Pitts,V. Philipps, J. Rapp, V. Riccardo, G. Saibene, J. Sanchez,R. Sartori, O. Sauter, D. Stork, B. Stratton, D. Testa,A. Tuccillo, M. Valovic, S. Sharapov, H. Weisen andK.-D. Zastrow. The dedication and hard work of the UKAEAcolleagues responsible for JET operation are gratefullyacknowledged. We dedicate this paper to the memory ofDerek C. Robinson, FRS, who contributed greatly to fusiondevelopment for four decades.

Appendix. JET EFDA contributors

J.M. Adams15, G. Agarici15, M. Agarici2, H. Akhter15,R. Albanese5c, S. Alberti12, S. Allfrey12, B. Alper15,D. Alves9, J. Amarante17, F.V. Amerongen6, P. Andrew15,Y. Andrew15, J.M. Ane2, C. Angioni12, C. Antonucci1,5g,G. Ambrosino5c, G. Apruzzese5b, M. Ariola5c, G. Artaserse5c,J.F. Artaud2, E. Ascasibar3, E. Asp10, M. Axton15, A. Baciero3,M. Badarelli5b, W. Baity29, R. Balbin3, S. Balme2, O. Barana5a,Y. Baranov15, E. Barbato5b, R. Barnsley15, V. Basiuk2,G. Bateman27, S. Baumel8, P. Bayetti2, L. Baylor29,B. Beaumont2, P. Beaumont15, A. Becoulet2, M. Becoulet2,M. Bekris14, M. Beldishevski15, A.C. Bell15, P. Bennet15,G. Berger-By2, H.L. Berk25, S. Bernabei30, L. Bertalot5b,B. Bertrand2, M. Beurskens15, M. Beurskens6, P. Bibet2,A. Bickley15, M. Bigi15, R. Bilato8, T. Blackman15,P. Blanchard12, J. Blum21, T. Bolzonella5a, A. Bondeson10,W. Bongers6, G. Bonheure17, X. Bonnin8, K. Borass8,D. Borba1,9, K. Bosak21, P. Bosia2, H. Boyer15, G. Bracco5b,G.C. Braithwaite15, B.N. Breizman25, S. Bremond2,P.D. Brennan15, J. Bresslau30, S. Brezinsek7, B. Brichero17,F. Briscoe15, M. Brix7, G. Brolatti5b, D.P.D. Brown15,A. Bruggeman17, A. Bruschi5d, S. Bryan15, J. Brzozowski10,J. Bucalossi2, G. Buceti1,5, M.A. Buckley15, T. Budd15,R. Budny30, P. Buratti5b, P. Butcher15, R.J. Buttery15,G. Calabro5c, C.J. Caldwel Nichols14, J. Callen36,D. Campbell16, D.C. Campling15, B. Cannas5e, A.J. Capel15,P.J. Card15, T. Carlstrom24, C. Castaldo5b, R. Causey31,F.E. Cecil23, R. Cesario5b, C. Challis15, V. Chan24,Ph. Chappuis2, M. Charlet15, J.M. Charreau2, C. Cheron2,D. Child15, G. Chitarin5a, S. Ciattaglia1,5, S. Cirant5d,D. Ciric15, R. Clarke15, J.P. Coad15, P. Coates15, V. Coccorese5c,V. Cocilovo5b, S. Coda12, R. Coelho9, I. Coffey15, A. Coletti5b,S. Collins15, J. Conboy15, S. Conroy10, G. Conway8,S.R. Cooper15, G. Cordey15, Y. Corre10, G. Corrigan15,S. Cortes9, D. Coster8, G.F. Counsell15, M. Cox15, S.J. Cox15,S. Cramp15, C. Crescenzi5b, F. Crisanti5b, I. Cristescu20,I.R. Cristescu14, B. Crowley4a, N. Cruz9, L. Cupido9,R. Cusack15, P. Da Silva Aresta Belo9, S. Dailey15, E. Daly15,C. Damiani1,5f , D. Darrow30, D. Darrow15, O. David2,N. Davies15, C. Day14, R. De Angelis5b, M. de Baar15,O. De Barbieri16, M. De Benedetti5b, J. De Grassie24,E. De la Luna3, P. De Vries6, F. Degli Agostini5a,M. Dentan1,2, A. Dimits34, A. Dines15, J.A. Dobbing15,L. Doceul2, J. Doncel1,3, A. Donne6, T. Donne6, W. Dorland22,S.E. Dorling15, P. Doyle15, V. Drozdov15, O. Dumbrajs13a,P. Dumortier17, A. Durocher2, F. Durodie17, B. Duval12,R. Dux8, T. Edlington15, A.M. Edwards15, D.C. Edwards15,D.T. Edwards15, P. Edwards15, T. Eich8, A. Ekedahl2,D. Elbeze2, A.R. Ellingboe4a, R. Ellis30, C.G. Elsmore15,

B.S.Q. Elzendoorn6, S.K. Erents15, G. Ericsson10, L. Eriksson2,B. Esposito5b, H.G. Esser7, T. Estrada3, M. Evrard17,C. Ewart15, J. Fanthome15, J.W. Farthing15, D. Fasel12,A. Fasoli12, R. Felton15, M. Fenstermacher26, C. Fenzi2,H. Fernandes9, A. Fernandez3, J. Ferreira9, J.A. Fessey15,A. Figueiredo9, P. Finburg15, J. Fink8, K.H. Finken7,P. Fiorentin5a, U. Fischer14, C. Fleming15, R. Forrest15,C. Fourment2, B. Franel2, P. Franzen8, T. Fredian28,J.P. Fricconneeu2, D. Frigione5b, G. Fu30, J.C. Fuchs8,K. Fullard15, W. Fundamenski15, F. Gabriel2, J. Gaffert8,E. Galutschek11, T. Gans4a, X. Garbet2, I. Garcia Cortes3,L. Garzotti5a, E. Gauthier2, J. Gedney15, A. Geier8,C. Gentile30, S. Gerasimov15, A. Geraud2, P. Ghendrih2,R. Giannella2, R.D. Gill15, C. Gimblett15, E. Giovannozzi5b,C. Giroud6, J. Goff15, P. Gohil24, A. Gondhalekar15,M. Goniche2, A. Goodyear15, N. Gorelenkov30, G. Gorini5d,R. Goulding29, C.W. Gowers15, M.E. Graham15, L. Grando5a,G. Granucci5d, M. Graswinckel6, N. Green15, C. Greenfield24,N. Greenough30, M. Greenwald28, D. Gregoratto10,F.S. Griph15, C. Grisolia2, A. Grosman2, G. Grosso5d,S. Gruenhagen14, A. Gude8, S. Guenter8, K. Guenther15,C. Guerin2, M. Gugla14, A. Guigon1,2, R. Guirlet2, J. Gunn2,G. Haas8, L. Hackett15, S. Hacquin9, T.S. Hahm30, B. Haist15,S. Hallworth Cook15, D. Hamilton15, G. Hammett30,R. Handley15, J.D.W. Harling15, J. Harrison15, D. Hartmann8,T. Haupt15, N.C. Hawkes15, J. Hay15, I. Hayward15,J. Hedin10, P. Heesterman15, J. Heikkinen13b, B. Heinemann8,P. Helander15, P. Hellingman6, T. Hellsten1,10,O.N. Hemming15, R. Hemsworth2, T.C. Hender15,M. Henderson12, P. Hennequin2, A. Henshaw15, A. Herrmann8,C. Hidalgo3, J. Hill15, D. Hillis29, M. Hitchin15, T. Hoang2,J. Hobirk8, F. Hoekzema7, J.A. Hoekzema7, F. Hofmann12,C. Hogan29, C. Hogben15, D. Hogeweij6, D. Homfray15,A. Horton15, L.D. Horton8, J. Hosea30, A.J. Hoskins15,S. Hotchin15, M.R. Hough15, W. Houlberg29, J. How2,D. Howell15, M. Hron18, A. Hubbard28, A. Huber7,Z. Hudson15, H. Hume15, D. Humphries24, F. Hurd16,I. Hutchinson28, T. Hutter2, S. Huygen17, G. Huysmans2,F. Imbeaux2, C. Ingesson6, P. Innocente5a, S. Jachmich17,G. Jackson24, S. Jardin30, A. Jarmen10, O.N. Jarvis15,R. Jaspers6, A. Jaun10, I. Jenkins15, H.S. Jensen14,E. Joffrin2, M.F. Johnson15, R. Johnson15, E.M. Jones15,G. Jones15, H.D. Jones15, T. Jones15, T. Jonsson10,C. Jupen10, A. Kallenbach8, J. Kallne10, S. Karttunen13b,W. Kasparek14, A. Kaye15, D. Keeling15, D. Kelliher4b,N. Kemp15, R.F. King15, J. Kinsey27, V. Kiptily15, K. Kirov8,A. Kirschner7, T. Kiviniemi13a, P. Knight15, S. Knipe15,R. Koch17, W. Kooijman6, A. Korotkov15, H.R. Koslowski7,G. Kramer30, W. Kraus8, A. Kritz27, O.G. Kruijt6,T. Kurki-Suonio13a, R.J. La Haye24, B. Labombard28,R. Laesser14, N. Lam15, Ph. Lamalle1,23, G. Land6,P. Lang8, L. Lao24, J. Last15, M. Laux8, C. Laviron2,K.D. Lawson15, M. Laxaback10, E. Lazzaro5d, H. Leggate15,M. Lehnen7, M. Leigheb5b, M. Lennholm2, A. Leonard24,C. Lescure15, F. Leuterer8, K. Likin3, J. Likonen13b,K. Lingier6, J. Linke7, B. Lipshultz28, J. Lister12, X. Litaudon2,Y. Liu10, G. Lloyd15, T. Loarer2, A. Loarte16, D. Loesser30,P.J. Lomas15, F. Long15, J. Lonnroth13a, A. Lorenz1,8,P. Lotte2, F. Louche17, M. Loughlin15, A. Loving15, T. Luce24,R.M.A. Lucock15, A. Lyssoivan17, J. Maagdenberg6,

1550

Overview of JET results

A. Maas2, J. MacGregor15, P. Macheta15, G. Maddaluno5b,G.P. Maddison15, P. Maget2, R. Magne2, A. Mahdavi24,J. Mailloux15, D. Maisonnier16, J. Manickam30, G. Mank7,M.E. Manso9, P. Mantica5d, M. Mantsinen13a, M. Maraschek8,D. Marcuzzi5b, M. Marinucci5b, Ph. Marmillod12, D. Martin15,M. Martin15, Y. Martin12, K.F. Mast8, M. Mattei5c,G.F. Matthews15, M. Mayer8, M.L. Mayoral15, D. Mazon2,G. Mazzone5b, E. Mazzucato30, E. McCarron15,K. McClements15, K. McCormick8, P.A. McCullen15,D. McCune30, D. McDonald15, M. Mead15, F. Medina3,A. Meigs15, W. Melissen6, L. Meneses9, F. Meo8,Ph. Mertens7, V. Mertens8, A. Messiaen17, F. Milani15,A. Miller15, S. Mills15, J. Milnes15, J. Mlynar12, I. Monakhov15,P. Monier-Garbet2, R. Monk8, R. Mooney15, D. Moreau2,Ph. Moreau2, P.D. Morgan15, A.W. Morris15, J. Morris15,J.L. Mort15, D. Mossessian28, A. Muck8, G. Mueller14,M. Murakami29, A. Murari5a, D. Murdock16, Y.-S. Na8,F. Nabais9, M.F.F. Nave9, R. Nazikian30, C. Negus15,G.F. Neil15, J.D. Neilson15, B. Nelson30, R. Neu8, W. Nevins26,G.J. Newbert15, F. Nguyen2, K. Nichols15, A. Nicolai7,L. Nicolas2, P. Nielsen5a, M. Nightingale15, H. Nordman10,J.-M. Noterdaeme8, S. Nowak5d, M. O’Mullane15, J. Ongena17,T. Onjun27, J. Orchard15, F. Orsitto1,5, T. Osborne24,J. Palmer16, J. Pamela1,2, L. Panaccione5b, V. Parail15,A. Parkin15, B. Parsons15, R. Pascualotto1,5, B. Patel15,I. Pavlenko17, A.T. Peacock16, R. Pearce15, A.L. Pecquet2,A. Pedrosa Luna3, A. Peeters8, R.D. Penzhorn14, M. PeresAlonso9, G. Pereverzev8, A. Perevezentsev15, C. Perezvon Thun7, V. Pericoli5b, S. Peruzzo5a, G. Petravich19,V. Petrizilka18, Y. Petrov32, C. Petty24, V. Phillips7,F. Piccolo15, M. Pick16, S.D. Pinches8, T. Pinna5b,B. Piosczyk14, A. Pironti5c, C.S. Pitcher28, R. Pitts12,V. Plysnin9, A. Pochelon12, S. Podda5b, P. Polinari5b,N. Pomaro5a, S. Popovichev15, C. Portafaix2, G. Porter26,A. Pospieszczyk7, G. Preece15, R. Prentice15, R. Prins6,M. Proschek11, R. Pugno8, M.E. Puiatti5a, K. Purahoo15,E. Rachlew10, M. Rainford15, D. Raisback15, K. Rantamaki13b,J. Rapp1,7, D. Rasmussen29, D. Reiser7, D. Reiter7,G. Rewoldt30, T.M. Ribeiro9, V. Riccardo15, E. Righi16,F.G. Rimini2, M. Riva5b, D.S. Robinson15, S.A. Robinson15,D.W. Robson15, M. Roccella5b, L. Rodriguez Rodrigo3,G. Rogero2, A. Rogister7, T. Roguemore30, A. Rolfe15,D. Ronden6, S. Rosanvallon2, D. Ross15, M. Rubel10,F. Ryter8, S. Saarelma13a, F. Sabathier2, R. Sabot2,G. Saibene16, J.F. Salavy2, I. Sall15, J. Sanchez3, S. Sanders15,S.G. Sanders15, D. Sands15, Y. Sarazin2, F. Sartori15,R. Sartori16, F. Sattin5a, O. Sauter12, C. Sborchia16,F. Scaffidi-Argentina1,14, A. Scarabosio12, G. Schilling30,D. Schissel24, J. Schlosser2, G. Schmidt30, W. Schmidt5a,C. Schuller6, B. Schweer7, J. Schweinzer8, J.L. Segui2,L. Semeraro5b, G. Sergienko7, F. Serra9, S.E. Sharapov15,S.R. Shaw15, M. Siegrist1,12, C.A. Silva9, D. Simpson15,S. Sipila13a, A.C.C. Sips8, H. Sjostrand10, C. Skinner30,P.G. Smith15, J. Snipes28, E.R. Solano1,3, P. Sonato5a,J. Sousa9, C. Sozzi5d, J. Spence15, E. Speth8, A. Staebler8,R. Stafford Allen15, R. Stagg15, J. Stakenborg6, M.F. Stamp15,P. Stangeby35, D. Starkey15, S.A. Stauton-Lambert15,A. Stephen15, A. Sterk6, A.B. Sterk6, A. Stevens15,J. Stillerman24, J. Stober8, R. Stokes15, D. Stork15,J. Strachan30, B. Stratton30, K. Stratton30, P. Stubberfield15,

H.P. Summers15, E. Surrey15, D. Sutton15, W. Suttrop8,J. Svenson8, I. Symonds15, F. Tabares3, J. Tait15, T. Tala13b,C. Talarico5b, A.R. Talbot15, C. Taliercio5a, U. Tam14,C. Tame15, G. Tardini8, M. Tardocchi10, N. Tartoni5b,J. Tavernier2, G. Telesca17, A.O. Terrington15, D. Testa12,P. Testoni5e, J.M. Theis2, J. Thomas15, P. Thomas2,H. Thomsen8, K. Thomsen16, M. Thumm14, C. Thyagaraja15,J. Tichler6, P. Tigwell15, T. Tiscornia15, J.M. Todd15,M.Z. Tokar7, G. De Tommasi5c, Q.M. Tran12, J.M. Travere2,G. Tresset2, V. Tribaldos3, E. Tsitrone2, A.A. Tuccillo5b,O. Tudisco5b, E. Turker15, M.M. Turner4a, V. Udintsev6,B. Unterberg7, M.P. Valetta2, M. Valisa5a, M. Valovic15,F. Van Amerongen6, M. Van de Poll6, T. Van derGrift6, D. Van Eester17, J. Van Gorkorm6, P. Varela9,D. Vender4a, G. Veres19, A. Verhoeven6, L. Villard12,E. Villedieu15, F. Villone5a, J.E. Vince15, M. Von Hellermann6,K. Vulliez2, M. Wade24, D. Wagner8, A. Walden15,M. Walker15, B. Walton15, M.L. Watkins1,15, M.J. Watson15,J. Weiland10, H. Weisen12, A. Werner8, F. Wesner8,P. West24, E. Westerhof6, B. Weyssow17, M.R. Wheatley15,A. Whitehead15, A. Whitehurst15, D. Whyte33, S.J. Wicks15,L. Widdershoven6, P. Wienhold7, A. Wilson15, C. Wilson15,D. Wilson15, D.J. Wilson15, D.W. Wilson15, R. Wilson30,M. Wischmeier12, R. Wolf7, P. Wouters17, J.S. Yorkshades15,C. Young15, D. Young15, I.D. Young15, K. Young30, L. Zabeo2,F. Zacek18, L. Zakharov30, P. Zanca5a, K.D. Zastrow15,W. Zeidner8, M. Zerbini5b, H. Zohm8, S. Zoletnik19, F. Zonca5b

and W. Zwingman2.

1 EFDA Close Support Unit, Culham, UK.2 Association Euratom-CEA, CEA-Cadarache, 13108 St Paul-

lez-Durance, France.3 Asociacion EURATOM-CIEMAT para Fusion, Avenida

Complutense, 22, Madrid, E-28040, Spain.4 Association Euratom DCU, Ireland.

a Dublin City University, Glasnevin, Dublin 9;b Department of Physics, University College

Cork, Cork.5 Associazione Euratom ENEA sulla Fusione, Italy.

a Consorzio RFX, 4-35127 Padova;b ENEA CR Frascati, Via Enrico Fermi 27, 00044

Frascati, Roma;c Association EURATOM/ENEA/CREATE, Via Clau-

dio 21, 1-80125 Napoli;d Instituto di Fisica del Plasma CNR-EURATOM, Via

Cozzi 53, 20125 Milano;e Electrical and Electronics Engineering Dept–Univ.

Cagliari, Piazza d’Armi-09123, Cagliari, Italy;f ENEA, C R Brasimone, CP1, 1-40032 Camugnano

(Bo);g ENEA, Via Martiri di Montesole 4, 1-40129 Bologna.

6 FOM-Rijnhuizen, Association Euratom-FOM, TEC,PO Box 1207, Nieuwegein, 3430 BE The Netherlands.

7 Institut fur Plasmaphysik, Forschungszentrum Julich,Trilateral Euregio Cluster, D-52425 Julich, Germany.

8 Max-Planck-Institut fur Plasmaphysik, EURATOM-Assoziation, Garching, D-85748 Germany.

9 Associacao EURATOM/IST, Centro de Fusao Nuclear,1049-001 Lisbon, Portugal.

1551

J. Pamela et al

10 EURATOM-VR Association, Swedish Research Council,SE 10378, Stockholm, Sweden.

11 Association Euratom-OAW, Kegelgasse 27/3 Wien, A-1030,Austria.

12 Centre de Recherches en Physique des Plasmas, AssociationEURATOM-Confederation Suisse, EPFL, 1015 Lausanne,Switzerland.

13 Association Euratom-TEKES, Finlanda Helsinki U. of Technology, Association Euratom-

TEKES, PO Box 2200, FIN-02015 HUT, Helsinki,Finland.

b VTT Processes, Otakaari 3A, Espoo, PO Box 1608,FIN-02044 VT.

14 Association Euratom-FZK, Forschungszentrum Karlsruhe(FZK), Postfach 3640, Karlsruhe, D-76021 Germany.

15 Euratom/UKAEA Fusion Association, Culham ScienceCentre, Abingdon, OX14 3DB, UK.

16 EFDA Close Support Unit, MPI, Garching, D-85748,Germany.

17 Association Euratom-LPP-ERM/KMS, TEC, RoyalMilitary Academy, Brussels, B-1000, Belgium.

18 Association Euratom-IPP, Institute of Plasma Physics,PO Box 17, Za Slovankou 3, 182 21, Prague 8, CzechRepublic.

19 KFKI-Research Institute for Particle and Nuclear Physics,Hungarian Academy of Sciences (HAS), PO Box 49,H-1525, Budapest, Hungary.

20 Ministry of Education and Research (MEC), Romania.21 Universite de Nice-Sophia-Antipolis, Nice, France.22 Imperial College London, South Kensington campus,

London SW7 2AZ, UK.23 Department of Physics, Colorado School of Mines, Golden,

CO 80401, USA.24 General Atomics, San Diego, California, USA.25 Institute for Fusion Studies, University of Texas, Austin,

TX 78712, USA.26 Lawrence Livermore National Laboratory, Livermore, CA

94550-9234, USA.27 Lehigh University, Bethlehem, PA 18015, USA.28 Plasma Science and Fusion Center, Cambridge, MA 02139,

USA.29 Oak Ridge National Laboratory, Oak Ridge, TN 37831-

8070, USA.30 Princeton Plasma Physics Laboratory, Princeton, NJ 08543-

0451, USA.31 Sandia National Laboratory, Albuquerque, NM 87185,

USA.32 Prairie View A&M University, Prairie View, TX 77446-

0519, USA.33 University of California San Diego, La Jolla, CA 92093-

0417, USA.34 University of Maryland, College Park, MD 20742-4111,

USA.35 University of Toronto, Institute for Aerospace Studies,

Toronto M3H 5T6, Canada.36 University of Wisconsin, Madison, WI 53706-1687, USA.

References

[1] ITER Physics Basis 1999 Nucl. Fusion 39 2137–638[2] Loarte A. 2002 Type I ELM energy and particle losses in JET

ELMy H-modes and implications for ITER Proc. 19th Int.Conf. on Fusion Energy 2002 (Lyon, 2002) (Vienna:IAEA) CD-ROM file and http://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm,paper EX/P1-08

[3] Matthews G.F. 2002 Steady state and transient power loads inJET Proc. 19th Int. Conf. on Fusion Energy 2002 (Lyon,2002) (Vienna: IAEA) CD-ROM file andhttp://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper EX/D1-1

[4] Rapp J. 2002 Reduction of divertor heat load in JET ELMyH-modes using impurity seeding techniques Proc. 19thInt. Conf. on Fusion Energy 2002 (Lyon, 2002) (Vienna:IAEA) CD-ROM file and http://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm,paper EX/P1-09

[5] Parail V. 2002 Integrated modelling of JET H-mode plasmaswith type-I ELMs Proc. 19th Int. Conf. on Fusion Energy2002 (Lyon, 2002) (Vienna: IAEA) CD-ROM file andhttp://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper TH/P3-08

[6] Koslowski H.R. 2002 Observation of pre- and postcursormodes of Types I ELMs on JET Proc. 19th Int. Conf. onFusion Energy 2002 (Lyon, 2002) (Vienna: IAEA)CD-ROM file and http://www.iaea.org/programmes/ripc/physics/fec2002/ html/fec2002.htm, paperEX/P1-14

[7] Litaudon X. 2002 Progress towards steady state operationand real time control of internal transport barriers in JETProc. 19th Int. Conf. on Fusion Energy 2002 (Lyon, 2002)(Vienna: IAEA) CD-ROM file and http://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm,paper EX/C3-4

[8] Hender T. 2002 Sawtooth, neo-classical tearing mode anderror field studies in JET Proc. 19th Int. Conf. on FusionEnergy 2002 (Lyon, 2002) (Vienna: IAEA) CD-ROM fileand http://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper EX/S1-2

[9] Ongena J. 2002 Towards the realization on JET of anintegrated H-Mode scenario for ITER Proc. 19th Int. Conf.on Fusion Energy 2002 (Lyon, 2002) (Vienna: IAEA)CD-ROM file and http://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper EX/C2-4

[10] Testa D. 2002 Experimental study of the stability of Alfveneigenmodes on JET Proc. 19th Int. Conf. on Fusion Energy2002 (Lyon, 2002) (Vienna: IAEA) CD-ROM file andhttp://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper EX/P1-17

[11] Noterdaeme J.-M. 2002 Heating, current drive and energeticparticles studies on JET in preparation of ITER operationProc. 19th Int. Conf. on Fusion Energy 2002 (Lyon, 2002)(Vienna: IAEA) CD-ROM file and http://www.iaea.org/programmes/ripc/physics/fec2002/ html/fec2002.htm,paper EX/W-1

[12] Mantica P. 2002 Transient heat transport studies in JETconventional and advanced tokamak plasmas Proc. 19thInt. Conf. on Fusion Energy 2002 (Lyon, 2002) (Vienna:IAEA) CD-ROM file and http://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm,paper EX/P1-04

[13] Garbet X. 2002 Micro-stability and transport studies ofinternal barriers in JET Proc. 19th Int. Conf. on FusionEnergy 2002 (Lyon, 2002) (Vienna: IAEA) CD-ROM fileand http://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper TH/2-1

[14] Joffrin E. 2002 Internal transport barrier triggering byrational magnetic flux surfaces Proc. 19th Int. Conf. onFusion Energy 2002 (Lyon, 2002) (Vienna: IAEA)

1552

Overview of JET results

CD-ROM file and http://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper EX/P1-13

[15] Dux R. 2002 Impurity transport in internal barrier dischargeson JET Proc. 19th Int. Conf. on Fusion Energy 2002(Lyon, 2002) (Vienna: IAEA) CD-ROM file andhttp://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper EX/C3-5Ra

[16] Philipps V. 2002 Recent results on long term fuel retention inJET and TEXTOR and predictions for ITER Proc. 19thInt. Conf. on Fusion Energy 2002 (Lyon, 2002) (Vienna:IAEA) CD-ROM file and http://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm,paper EX/P5-08

[17] Suttrop W. 2002 Parameter similarity studies in JET, ASDEXUpgrade and ALCATOR C-Mod Proc. 19th Int. Conf. onFusion Energy 2002 (Lyon, 2002) (Vienna: IAEA)CD-ROM file and http://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper EX/P5-07

[18] Hidalgo C. 2002 Experimental evidence of fluctuations andflows near marginal stability in the plasma boundary regionin fusion plasmas Proc. 19th Int. Conf. on Fusion Energy2002 (Lyon, 2002) (Vienna: IAEA) CD-ROM file andhttp://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper EX/C4-4

[19] Castaldo C. et al 2002 Improved plasma confinement by ionbernstein waves (IBWs) interacting with ions in JET (JointEuropean Torus) Proc. 19th Int. Conf. on Fusion Energy2002 (Lyon, 2002) (Vienna: IAEA) CD-ROM file andhttp://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper PD/P-01

[20] Special issue on ELMy H-Mode research on the JET facilityunder EFDA during the 2000–2001 experimentalcampaign Plasma Phys. Control. Fusion 44

[21] Suttrop W. 2002 Phys. Plasmas 9 2103[22] Loarte A. et al 2002 Phys. Plasma 44 1815–44[23] Saibene G. 2002 Phys. Plasma 44 1769–99[24] Janeschitz G. et al 2001 Plasma–wall interaction issues in

ITER J. Nucl. Mater. 290–293 1–11[25] Federici G. et al 2003 Key ITER plasma edge and

plasma–material interaction issues J. Nucl. Mater.313–316 11–22

Federici G. et al 2002 Proc. 15th Int. Conf. on PlasmaSurface Interactions in Controlled Fusion Devices (Gifu,Japan, 27–31 May 2002)

[26] Overview of MHD stability in edge transport barriers 9thEuropean Fusion Physics Workshop 2001 (Saariselka,Finland)

[27] Huysmans G.T.A. et al 2001 Phys. Plasmas 8 4292[28] Perez C. et al Type-I ELM precursor modes in JET Nucl.

Fusion submitted[29] Soldner F.X. et al 1997 Plasma Phys. Control. Fusion 39

B353[30] Becoulet M. et al 2002 Plasma Phys. Control. Fusion 44

A103[31] Nave M.F.F. et al 2002 Nucl. Fusion 42 281–9[32] Guenter S. et al 2001 Phys. Rev. Lett. 87 275001[33] Sauter O. et al 2002 Phys. Rev. Lett. 88 105001[34] Lahaye R. J. 2000 Phys. Plasmas 7 3349[35] Sauter O. 2002 Plasma Phys. Control. Fusion 44 1999[36] Mantsinen M. et al 2002 Phys. Rev. Lett. 88 105002-1[37] Kiptily V.G. et al 2002 Nucl. Fusion 42 999[38] Jaun A. et al 2002 Collective modes and fast particle

confinement in ITER Proc. 19th Int. Conf. on FusionEnergy 2002 (Lyon, 2002) (Vienna: IAEA) CD-ROM fileand http://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper CT/P-06

[39] Sartori et al 2002 Plasma Phys. Control. Fusion 44 1801[40] Stober J. et al 2001 Nucl. Fusion 41 1123[41] Kamada Y. and the JT-60U team 2001 Nucl. Fusion

41 1311[42] Nave F. et al 2003 Nucl. Fusion 43 1204–13[43] Lang P.T. et al 2002 Nucl. Fusion 42 388

[44] Valovic M. et al 2002 Plasma Phys. Control. Fusion 44 1911[45] Garzotti L. et al 2002 Particle transport and density profile

analysis of different JET plasmas 29th EPS Conf. onPlasma Phys. Control. Fusion (Montreux, 17–21 June2002) vol 26B (ECA) P-1.035

[46] Cordey G. 2002 Plasma Phys. Control. Fusion 44 1929[47] McDonald D. C. et al Plasma Phys. Control. Fusion to be

submitted[48] Cordey G. et al 2002 A two term model of confinement in

ELMy H-modes using the global confinement and pedestaldatabases Proc. 19th Int. Conf. on Fusion Energy 2002(Lyon, 2002) (Vienna: IAEA) CD-ROM file andhttp://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper CT/P-02 Nucl. Fusion submitted

[49] Wolf R.C. et al 2003 Characterisation of ion heat conductionin JET and ASDEX-Upgrade plasmas with and withoutinternal transport barriers Plasma Phys. Control. Fusion45 R1–91

[50] Jones T.T.C. et al 2001 Simulation of alpha particle plasmaself-heating using ICRH under real-time control 28th EPSConf. on Plasma Phys. Control. Fusion (Madeira, 2001)vol 25A (ECA) 1197–1200

[51] Tuccillo A.A. et al 2001 Proc. 14th Topical Conf. onRadiofrequency Power in Plasmas (Oxnard, 2001)ed T.K. Mau and J. deGrassie AIP Conf. Proc. 595 209–16

[52] Special issue on Advanced tokamak research in EFDA-JETduring the 2000–2001 experimental campaigns 2002Plasma Phys. Control. Fusion 44 1031

[53] Tresset G. et al 2002 Nucl. Fusion 42 520[54] Mailloux J. et al 2002 Phys. Plasmas 9 2156[55] Pericoli Ridolfini V. et al 2001 LHCD coupling during

H-mode and ITB in JET plasmas Proc. 14th Topical Conf.on Radiofrequency Power in Plasmas (Oxnard, 2001)ed T.K. Mau and J. deGrassie AIP Conf. Proc. 595 245–8

[56] Pericoli Ridolfini V. et al 2002 Study and optimisation oflower hybrid waves coupling in advanced scenariosplasmas in JET Plasma Phys. Control. Fusion submitted

[57] Hawkes N.C. et al 2001 Phys. Rev. Lett. 87 115001-1[58] Fujita T. 2001 Phys. Rev. Lett. 87 245001[59] Miura Y. 2002 Stable existence of current hole in the JT-60U

tokamak/equilibria and stability of JET current holedischarges Proc. 19th Int. Conf. on Fusion Energy 2002(Lyon, 2002) (Vienna: IAEA) CD-ROM file andhttp://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper EX/C3-1Ra

[60] Tala T.J.J. et al 2001 Plasma Phys. Control. Fusion 43 507[61] Huysmans G.T.A. et al 2001 Phys. Rev. Lett. 87 245002-1[62] Stratton B.C. et al 2002 Plasma Phys. Control. Fusion 44

1127[63] Joffrin E. et al 2002 Nucl. Fusion 42 235[64] Joffrin E. et al 2002 Plasma Phys. Control. Fusion 44 1739[65] Sharapov S.E. et al 2001 Phys. Lett. A 289 127[66] Kimura H. et al 1998 Nucl. Fusion 38 1303[67] Sharapov S.E. et al 2002 Phys. Plasmas 9 2027[68] Zonca F. et al 2002 Phys. Plasmas 9 4939[69] Berk H.L. et al 2001 Phys. Rev. Lett. 87 185002[70] Fasoli A. et al 2002 MHD spectroscopy Plasma Phys.

Control. Fusion 44 159–72[71] Breizman B.N. et al 2002 Alfven eigenmodes in shear

reversed plasmas Proc. 19th Int. Conf. on Fusion Energy2002 (Lyon, 2002) (Vienna: IAEA) CD-ROM file andhttp://www.iaea.org/programmes/ripc/physics/fec2002/html/fec2002.htm, paper TH/4-3

[72] Challis C.D. et al 2002 Plasma Phys. Control. Fusion 441031

[73] Crisanti F. et al 2002 Phys. Rev. Lett. 88 145004-1[74] Litaudon X. et al 2002 Plasma Phys. Control. Fusion 44 1057[75] Mazon D. et al 2002 Real-time control of internal transport

barriers in JET Plasma Phys. Control. Fusion 44 1087[76] Mazon D. et al 2003 Proc. Plasma Phys. Control. Fusion

45 L47[77] Krivenski V. Phys. Rev. Lett. submitted

1553

J. Pamela et al

[78] Krivenski V. et al 2002 Evidence of non-Maxwellian electronbulk distributions on JET 29th EPS Conf. on Plasma Phys.Control. Fusion (Montreux, 2002) vol 26B (ECA) O-1.03

[79] De La Luna E. et al 2003 Impact of bulk non-Maxwellianelectrons on electron temperature measurements Rev. Sci.Instrum. 74 1414

De La Luna E. et al 2002 Proc. 14th Topical Conf. on HighTemperature Plasma Diagnostics (Madison)

[80] Taylor G. et al 1996 Phys. Rev. Lett.[81] Thomas P.R. et al 1998 Phys. Rev. Lett. 80 5548[82] Noterdaeme J.-M. et al 2001 Spatially resolved plasma

rotation profiles with ICRF on JET 14th Topical Conf. onRadio Frequency Power in Plasmas (Oxnard, CA, 2001)vol 595 (New York: AIP) pp 98–101

[83] Noterdaeme J.-M. et al 2001 Toroidal plasma rotation withICRF on JET 28th EPS Conf. on Controlled Fusion andPlasma Physics (Funchal (Madeira) Portugal, 2001)vol 25A (ECA) (EPS) pp 777–80

[84] Noterdaeme J.-M. et al 2003 Spatially resolved plasmarotation profiles with ICRF on JET Nucl. Fusion 43 274

[85] Mantsinen M. et al 2002 Controlling the profile ofion-cyclotron-resonant ions in JET with the wave-inducedpinch effect Phys. Rev. Lett. 89 115004-1–115004-4

[86] Mantsinen M. et al 2001 ICRF mode conversion experimentson JET 28th EPS Conf. on Controlled Fusion and PlasmaPhysics (Funchal (Madeira) Portugal, 2001) vol 25A(ECA) (EPS) pp 1745–8

[87] Riccardo V. et al 2002 Plasma Phys. Control. Fusion 44905–29

[88] Albanese R. et al 1998 Nucl. Fusion 38 723[89] Albanese R. et al 2002 Plasma response models for current,

shape and position control in JET Proc. 22nd Symp. onFusion Technology (SOFT) (Helsinki, September 2002)Fusion Eng. Des. submitted

[90] Villone F. et al 2002 Neutral point detection in JET Proc.22nd Symp. on Fusion Technology (SOFT) (Helsinki,2002) Fusion Eng. Des. submitted

[91] Gill R. D. et al 2002 Behaviour of disruption generatedrunaways in JET Nucl. Fusion submitted

[92] Yoshino R. et al 1999 Nucl. Fusion 39 151[93] Riccardo V. et al 2002 Refurbishment of the JET halo current

diagnostics Proc. 22nd Symp. on Fusion Technology(SOFT) (Helsinki, 2002) Fusion Eng. Des. submitted

[94] Likkonen J. et al 2003 Fusion Eng. Des. 66–68 219–24[95] Coad J.P. et al 2002 Erosion/deposition in JET during the

period 1999–2001 J. Nucl. Mater. 313–316 419–23Coad J.P. et al 2002 Proc. Plasma–Surface Interactions in

Control. Fusion Devices (Gifu, Japan, 26–31 May 2002)vol 15

[96] Coad J.P. et al 2001 J. Nucl. Mater. 290–293 224[97] Pitts R. et al 2002 Comparing scrape-off layer and divertor

physics in JET pure He and D discharges J. Nucl. Mater.313–316 777–86

Pitts R. et al 2002 Proc. Plasma–Surface Interactions inControl. Fusion Devices (Gifu, 2002) vol 15

[98] Esser H.G. et al 2003 Fusion Eng. Des. 66–68 860–6[99] Bell A.C. et al 2002 Fusion Sci. Technol. 41 626

[100] Rosanvallon S. et al 2002 Fusion Sci. Technol. 41 695[101] Perevezentsev A. et al 2002 Fusion Sci. Technol. 41 706[102] Cristescu I. et al 2003 Fusion Eng. Des. 69 109–113[103] Perevezentsev A. et al 2002 Fusion Sci. Technol. 41 1107[104] Perevezentsev A. et al 2002 Fusion Eng. Des. 1–5[105] Perevezentsev A. et al 2002 Fusion Sci. Technol. 41 821[106] Coad J.P. et al 2002 J. Nucl. Mater. 290–293 224[107] JET Team (presented by Zastrow K.-D.) Nucl. Fusion 39

1891[108] Brennan D.P. 2002 Fusion Sci. Technol. 41 578[109] Day C.H.R. et al 2003 Fusion Eng. Des. 69 97–102

1554