Search

The understanding and predictive capability of transport physics and plasma confinement is reviewed from the perspective of achieving reactor-scale burning plasmas in the ITER tokamak, for both core and edge plasma regions. Very considerable progress has been made in understanding, controlling and predicting tokamak transport across a wide variety of plasma conditions and regimes since the publication of the ITER Physics Basis (IPB) document (1999 Nucl. Fusion 39 2137-2664). Major areas of progress considered here follow. (1) Substantial improvement in the physics content, capability and reliability of transport simulation and modelling codes, leading to much increased theory/experiment interaction as these codes are increasingly used to interpret and predict experiment. (2) Remarkable progress has been made in developing and understanding regimes of improved core confinement. Internal transport barriers and other forms of reduced core transport are now routinely obtained in all the leading tokamak devices worldwide. (3) The importance of controlling the H-mode edge pedestal is now generally recognized. Substantial progress has been made in extending high confinement H-mode operation to the Greenwald density, the demonstration of Type I ELM mitigation and control techniques and systematic explanation of Type I ELM stability. Theory-based predictive capability has also shown progress by integrating the plasma and neutral transport with MHD stability. (4) Transport projections to ITER are now made using three complementary approaches: empirical or global scaling, theory-based transport modelling and dimensionless parameter scaling (previously, empirical scaling was the dominant approach). For the ITER base case or the reference scenario of conventional ELMy H-mode operation, all three techniques predict that ITER will have sufficient confinement to meet its design target of Q = 10 operation, within similar uncertainties.

The predicted H-mode power threshold, PL−H, for ITER is generally estimated from the international global H-mode threshold database (IGDBTH) by ordinary least squares log–linear (OLS) regressions. Such fits assume that errors are uncorrelated and (i) errors in PL−H are much greater than those in the other parameters, (ii) errors are normally distributed and (iii) relative errors are equal for all experiments. In this paper, the validity of this statistical model for the IGDBTH is examined, by use of the more generalized maximum-likelihood method. Results indicate that all three assumptions bias the resulting scaling and so need to be relaxed. A fit relaxing all three constraints lies outside the error bars of the OLS, indicating that the choice of the statistical model makes a significant contribution to the resulting scaling. A chi-squared analysis shows that none of the studied models are entirely consistent with the data, indicating that further refinement of the physical and statistical model is required. For ITER-like parameters, a maximum-likelihood analysis shows a predicted threshold of 38.4 MW, compared with 31.1 MW for the OLS, indicating that OLS tends to under predict and that quoted confidence intervals tend to be too small. However, further studies of the sources of errors in the IGDBTH would be required before estimates based on more detailed statistical models can be given with confidence.

The condition of the latest version of the ELMy H-mode database has been re-examined. It is shown that there is bias in the ordinary least squares regression for some of the variables. To address these shortcomings three different techniques are employed: (a) principal component regression, (b) an error in variables technique and (c) the selection of a better conditioned dataset with fewer variables. Scalings in terms of the dimensionless physics variables, as well as the standard set of engineering variables, are also derived. The new scalings give a very similar performance for existing scalings for ITER at the standard beta(n) of 1.6, but a much improved performance at higher beta n.

While most of the dimensionless parameters that govern energy transport are readily apparent, there is some controversy as to whether the collisionality should be represented by the collision frequency normalized to the bounce frequency (ν*) or the density normalized to the Greenwald density limit (n/nlimit). To help resolve this question, experiments on the DIII-D and JET tokamaks have compared the normalized energy transport in ELMing H-mode plasmas with matched dimensionless parameters. When ν* was kept fixed, the normalized energy transport on JET and DIII-D was in good agreement, but when n/nlimit was kept fixed, the normalized energy confinement time on DIII-D was 20% lower than on JET (for all cases, n/nlimit < 1). Therefore, ν* appears to be the correct form of the normalized density, and scaling transport properties from present day tokamaks to ITER at fixed n/nlimit can result in incorrect predictions.

The disagreement between the weak dependence of the energy confinement time on normalized pressure, β, observed in dedicated scans and the strongly negative dependence in the confinement scaling laws used for the design of next step tokamaks and future reactors, remains an outstanding problem. As such, scans of β have been undertaken in single null, low triangularity (δ ≈ 0.2) ELMy H-mode plasmas in JET with the MarkIIGB-SRP divertor. The scans varied β by a factor of 2.8 (normalized β from 0.72 to 2.04) and covered a range of magnetic fields (1.5–2.3 T), plasma currents (1.5–2.75 MA) and safety factors (q95 = 2.8 and 3.3). A weak β dependence was observed both globally (B0τE varied less than 9% across any one scan) and locally. A scan within Type I ELMy H-modes suggests that this weaker dependence is not due to ELM regimes. A statistical analysis indicates that these results are consistent with log–linear regressions performed on a wide JET database of ELMy H-modes, if correlations in this database are considered.

The scaling of energy transport with dimensionless parameters has been measured in high-temperature plasmas with the goals of guiding theory and predicting energy confinement in future fusion devices. Validation of this approach requires demonstration of similarity in plasmas with identical dimensionless parameters but very different physical parameters. Within measurement uncertainties, the heat diffusivities and global energy confinement exhibit similarity in high-confinement regimes on the DIII-D and JET tokamaks and in low-confinement regimes on the DIII-D and Alcator C-Mod tokamaks.

Pellet injection was investigated for its fuelling capability to high densities in ELMy H mode discharges at JET. Applying the high field side launch system, optimized refuelling scenarios were developed on the basis of conventional discharge configurations withIp= 2.5 MA,Bt= 2.4 T, averaged triangularity ⟨δ⟩≈0.34 and mainly neutral beam heating at a level of approximately 17 MW. The accessible operational range was extended with respect to gas puff refuelling by the use of pellet injection. For example, H mode conditions could be maintained at densities beyond the Greenwald level. Plasma energy confinement was observed to become density independent at high densities. Deep pellet particle deposition made possible the uncoupling of edge and core density, allowing more peaked density profiles. When confinement deterioration due to pellet triggered MHD activity or parasitic pellet borne gas was avoided in appropriate pulse schedules, an enhanced particle inventory was achieved while maintaining the plasma pressure profile. In a technical assessment it was found that there is still room for further enhancement of the injection scenario by improved adaptation to high density plasma operation.

An international multi-tokamak profile database has been assembled, constituting a representative set of reference tokamak discharges for the purpose of testing local transport models against well documented data. In particular, it will allow one to measure the accuracy with which the models can reproduce experiments and draw confidence intervals for the predictions of the models outside the range covered in the database. This database is now available to the fusion community and may be accessed by anonymous ftp to iterphys.naka.go.jp; the purpose of this article is to describe the structure of the database and the discharges contributing to it so that all can take full advantage of this resource. Thus, after an introductory general discussion of the database, there is a more detailed description of its structure, with listings of variables emphasized and how to access the database. There is then a brief description of each contributing tokamak and information on the type of discharges available from that tokamak. This is followed by a more quantitative description of the data, giving the ranges of dimensional and dimensionless variables available. Some typical modelling results to illustrate the use of the database are given in the conclusion.

ELMy H-mode pulses have been obtained with different hydrogenic isotopes (H and D) but having the same profiles of the dimensionless parameters ρ*,β,ν* and q, to test whether the confinement scale invariance principle is valid in a tokamak. The fact that the confinement times, the ELM and sawtooth frequencies in the two pulses all scale as expected suggests that the invariance principle is satisfied through the plasma radial extent, in spite of the differing physical processes taking place in the plasma centre, core and edge regions. An application of this `isotope windtunnel technique' to predicting the D-T performance of the next step devices is discussed.

Because of its large size, single null divertor and flexible magnetic geometry, JET is capable of producing the most reactor relevant plasmas of any present generation tokamak. In recent DT experiments, the fusion performance of these plasmas was tested for the first time. Over 4 MW of fusion power was produced in a high power, steady state pulse of 5 s, limited by the duration of the heating power. The fusion QE, defined simply as the fusion energy produced divided by the input energy over this 5 s interval, was 0.18. These DT ELMy H mode discharges performed up to expectations based on DD preparation pulses and thus establish a firm basis for extrapolating to a next step machine. Operation at low q95 is possible in JET with no degradation in the confinement enhancement factor and provides an improved margin to ignition when extrapolated to ITER. Considerable uncertainties remain, nonetheless. In particular, access to high density, relative to the Greenwald limit, and operation in close proximity to the H mode threshold may both result in a degradation of the confinement in the next step machine.

The scaling of the energy confinement in H mode plasmas with different hydrogenic isotopes (hydrogen, deuterium, DT and tritium) is investigated in JET. For ELM-free H modes the thermal energy confinement time τth is found to decrease weakly with the isotope mass (τth ~M-0.25±0.22), whilst in ELMy H modes the energy confinement time shows practically no mass dependence (τth ~M0.03±0.1). Detailed local transport analysis of the ELMy H mode plasmas reveals that the confinement in the edge region increases strongly with the isotope mass, whereas the confinement in the core region decreases with mass (τthcore ∝ M-0.16), in approximate agreement with theoretical models of the gyro-Bohm type (τgB ~M-0.2).

Results are presented from a series of dedicated experiments carried out on JET in tritium, DT, deuterium and hydrogen plasmas to determine the dependence of the H mode power threshold on the plasma isotopic mass. The Pthr ∝ Aeff-1 scaling is established over the whole isotopic range. This result makes it possible for a fusion reactor with a 50:50 DT mixture to access the H mode regime with about 20% less power than that needed in a DD mixture. Results on the first systematic measurements of the power necessary for the transition of the plasma to the type I ELM regime, which occurs after the transition to H mode, are also in agreement with the Aeff-1 scaling. For a subset of discharges, measurements of Te and Ti at the top of the profile pedestal have been obtained, indicating a weak influence of the isotopic mass on the critical edge temperature thought to be necessary for the H mode transition.

On the basis of experiments carried out in JET in D:T mixtures varying from 100:0 to 5:95 and those carried out in hydrogen plasmas, the isotopic mass dependence of ELM parameters and the edge pedestal pressure in NBI and ICRF heated H mode plasmas is presented. The ELM frequency is found to decrease with the atomic mass number in both ICRH and NBI discharges. However, the frequency in the case of ICRH is about 8-10 times higher than that in the case of NBI. Assuming that ELMs occur at a critical edge pressure gradient, limited by the ballooning instability, the scaling of the maximum edge pressure is most consistent with the assumption that the width of the transport barrier scales as the ion poloidal Larmor radius governed by the average energy of fast ions at the edge. The critical edge pressure in NBI heated discharges increases with the isotopic mass, which is consistent with the higher deduced width of the edge transport barrier in tritium than those in deuterium and hydrogen. The critical edge pressure in ICRH discharges is smaller, presumably, due to the smaller fast ion contribution to the edge region. As a consequence of the edge pressure scaling with isotopic mass, the edge operational space in an ne-Te diagram increases with operation in tritium. If the evidence that the edge pedestal width is governed by the average energy of fast ions in the edge prevails, the pedestal in ITER would be controlled by the slowing down energy spectrum of alpha particles in the edge.

An overview of JET experimental results in DT plasmas directly relevant to ITER modes of operation is presented. Experiments in D:T mixtures varying from 100:0 to 10:90 and those carried out in hydrogen plasmas show that the H mode threshold power has an approximately inverse isotope mass dependence. Matching some of the key dimensionless parameters to the ITER values, the ITER similarity experiments with ITER shape and safety factor q show that the global energy confinement time is practically independent of isotopic mass (~A0.03±0.08), where A is the atomic mass of the hydrogenic species. Subtracting the edge pedestal energy (which scales as ~A0.57±0.2) from the total stored energy leads to a ~A-0.17±0.1 dependence of confinement in the plasma core, very similar to that expected from the gyro-Bohm transport (~ A-0.2) model. The observed scaling of the edge pedestal energy is consistent with a model in which the edge pressure gradient saturates at the ballooning limit over a region of width that scales as the ion poloidal Larmor radius governed by the average energy of the fast ions in the edge. The steady state total stored energy for a given input power in both ICRH and NBI discharges is the same despite the lower edge pedestal in the ICRH case, which is compensated for by more peaked power deposition profiles in ICRH. The ELM frequency is smaller with NBI; it decreases with isotopic mass in both NBI and ICRH discharges. A steady state, type I ELMy H mode discharge with ITER shape and q at 3.8 T/3.8 MA with an input power of 22 MW produced a Q ≈ 0.18 for 3.5 s and extrapolates well to ignition with ITER parameters. Here, Q is the ratio of fusion output power to input power. The thermal ELMy H mode confinement in both deuterium and tritium gas fuelled plasmas decreases significantly when the plasma density exceeds 0.75 of the Greenwald (nGW) limit, and the maximum density achieved is 0.85nGW. In L mode, the density limit decreases with increasing isotope mass roughly in accordance with code predictions. ITER reference ICRH scenarios have been evaluated. Second harmonic heating of tritium at the densities available in JET produces strong tails and heats electrons predominantly as expected. The 3He minority in 50:50 D:T and tritium dominated plasmas showed strong bulk ion heating leading to ion temperatures up to 13 keV with ICRH alone. Deuterium minority ion cyclotron heating in tritium plasmas at a power level of 6 MW produced steady state record values of Q ≈ 0.22 for more than 2.5 s.

In current large tokamaks, non-intrinsic seeded impurities have been used to produce divertor power loads which would be considered acceptable when extrapolated to ITER. Many devices have achieved the goals of high fractional radiated powers, small frequent ELMs and detachment which are characteristic of radiative H mode regimes. The influence of divertor geometry on these characteristics is described. It has been a matter of concern that the Zeff associated with the seeded impurities may exceed that allowable in ITER and also that the degradation in energy confinement may be unacceptable. Confidence can only be built in the prediction of these parameters in ITER if reliable scalings are available for impurity content and energy confinement which have a sound physics basis. Work is described at JET in this area whilst using multimachine data to characterize the size scaling and provide a context for the JET data. Predicted levels for the impurity content of seeded ITER plasmas appear to be of marginal acceptability. Discharges run in the JET Mark I, Mark IIA and Mark IIAP divertors are compared and indicate that increased divertor closure has brought relatively minor benefits in highly radiative discharges. The acceptability of the energy confinement of radiation for ITER remains unclear. Dimensionless parameter scaling experiments have been conducted in which β, q25, fractional radiated power and Zeff are held constant for a range of ρ*. The price paid for high edge radiation and small ELMs appears to be a 25% loss in total stored energy as a result of edge pedestal degradation. However, the underlying energy confinement scaling may still be consistent with gyro-Bohm scaling, which would give an adequate margin for ITER. This conclusion is, however, sensitive to the scaling of confinement with collisionality, which is difficult to determine due to the coupling between ρ* and ν* which is a consequence of radiation dominated regimes.

Empirical scaling laws (Goldston, Kaye-Goldston, ITER89P and ITER93H) for the confinement time τE in tokamaks depend on dimensionless global plasma parameters in a way which is difficult to associate with some general plasma physics model. It is shown that such dependences can arise from artefacts in the database used to derive the scaling expression. The artefacts involve inner relationships, for example, collinearities between variables. For the ITER L mode and H mode databases, it is shown that subsets of the scaling expressions are approximately constant. An alternative description of L mode, ELM-free, small ELM and giant ELM H mode data is given. The alternative description does not include the dependences upon plasma β, elongation and safety factor which are present in the above scaling laws; extrapolations to ITER parameters by this alternative data description are virtually the same as those by ITER89P and ITER93H. The alternative description is based on a simple electrostatic physics model (plateau scaling) in which confinement is degraded by non-collisional processes. It can describe global confinement data for all four confinement regimes; the common scaling features a multiplier C which is assigned a different value for each regime. The complex physics associated with MHD instabilities (ELMs, sawteeth, outer modes) is hidden in the multiplier C, and attempts are made to uncover such physics. In these attempts global confinement is made to scale with physics models rather than physics variables.

Plasmas with the highest deuterium fusion neutron rates yet achieved in JET have been produced by combining a hollow or flat central current profile with a high-confinement (H-mode) edge. In these discharges, lower hybrid current drive (LHCD) and ion cyclotron resonance heating (ICRH) preheating, applied early in the current ramp-up phase, `freezes in' a hollow or flat current density profile. When the combined neutral beam injection (NBI) and ICRH heating power is increased, a region of reduced transport and highly peaked profiles forms during the L-mode phase and persists into the later H-mode phase when the fusion reactivity reaches a maximum. Transport analysis shows the formation of a central region of good confinement (the internal transport barrier or ITB) which expands with radial velocity . The clearest signatures of this region are large gradients in the ion temperature and toroidal rotation profiles. Ion thermal diffusivities in the central region are of the order of the neoclassical value. The position and rate of expansion of the ITB radius correlates well with the calculated rational q = 2 surface. The confinement improvement can also be seen in electron density, and, to a lesser extent, in electron temperature. The ITB can persist in combination with the edge transport barrier characteristics of the H-mode.

An analysis of the complete JET D-D threshold database, from 1990 to 1997 inclusive is presented. A JET scaling is derived for the threshold power through the separatrix, estimated by considering also radiation losses from the bulk plasma. Such scalings are applied to obtain a first qualitative estimate of the amount of power above threshold that is needed to obtain steady-state high confinement (H89 > 1.8). The L-H and H-L transitions are also analysed in terms of local edge parameters. The first results on isotope scaling of the H-mode power threshold in D-T and T-T plasmas are also presented, confirming the scaling.

All major JET systems have been fully commissioned for D-T and the DTE1 series of experiments has started with the D-T fuel mixture and operating conditions foreseen for ITER. In the area of ITER physics, significant results have been produced in both D-D and D-T. In D-D, the LH threshold power database has been extended, the bounds on edge-electron temperature and density in ELMy H-modes have been defined and the advantages of Types I and III ELMy discharges have been compared. In D-T plasmas, the isotope effect on H-mode threshold power and transport has been determined so that a more accurate assessment can be made of the ignition margin and heating requirements for ITER. Trace tritium experiments have provided first particle transport measurements and an assessment of the ITER reference ion-cyclotron resonance-frequency heating scenarios has been started, In the area of fusion performance, record D-D neutron yields have been obtained by controlling the plasma and current profiles in hot ion ELM-free H-modes and optimized shear modes. In D-T, internal transport barriers have been readily established in optimized shear discharges and Alfven eigenmodes have been observed.

This special topic describes the contents of an L mode database that has been compiled with data from Alcator C-Mod, ASDEX, DIII, DIII-D, FTU, JET, JFT-2M, JT-60, PBX-M, PDX, T-10, TEXTOR, TFTR and Tore Supra. The database consists of a total of 2938 entries, 1881 of which are in the L phase while 922 are ohmically heated only (ohmic). Each entry contains up to 95 descriptive parameters, including global and kinetic information, machine conditioning and configuration. The special topic presents a description of the database and the variables contained therein, and it also presents global and thermal scalings along with predictions for ITER. The L mode thermal confinement time scaling, determined from a subset of 1312 entries for which the τE,th are provided, is τE,th = 0.023Ip0.96BT0.03R1.83(R/a)0.06 κ0.64ne0.40Meff0.20P-0.73 in units of seconds, megamps, teslas, metres, -, -, 10-9 m-1

The theoretical basis of the dimensionless parameter scaling technique is derived and the limitations in its application are discussed. The use of the technique is illustrated by the production on JET of a steady-state ITER similarity pulse having the same and collisionality as the ignited ITER. The key issue of the scaling of the transport with the main dimensionless parameter is discussed in detail. Finally, possible shortcomings of the technique are examined.

The ITER Threshold Database, which at present comprises data from nine divertor tokamaks, is described. The main results are presented and discussed. The properties and dependences of the power threshold in individual devices are reviewed. In particular, the analysis shows a rather general linear dependence on magnetic field, but a non-monotonic density dependence that varies from device to device. Investigation of the combined database suggests that the threshold dependence Pthres approximately=0.3neBT2.5 shows reasonable agreement with the data. This expression yields Pthres approximately=150 MW at a density of 0.5*1020 m-3 for ITER. Other expressions with weaker size dependence, and therefore lower threshold power for ITER, are also discussed. Their agreement with the present data is poorer than that of the above expression. In addition, the database is investigated by statistical discriminant analysis. The edge data included at present are described and discussed. Finally, there is a discussion of the implications of the results for ITER

The results of a series of ITER simulation experiments on JET are described. A series of H-mode threshold experiments are shown to reproduce one of the standard power threshold scaling expressions that is being used to predict the power threshold in ITER. Then, from a series of experiments in which the Larmor radius scaling of ELMy H-modes is examined, it is concluded that the scaling of the confinement is gyro-Bohm-like provided the power levels are well above the threshold. Finally, we show that at high there is a dramatic reduction in the confinement at .

The results of numerical simulation of JET L, H and VH modes are presented. It is shown that L-H transitions are accompanied by a fast and large modification of the transport coefficients not only near the plasma edge but also across a large fraction of the plasma cross-section. Both experiments and numerical simulations show that giant ELMs in JET also have a global character and their penetration length increases with the amplitude of the Dalpha signal. Transport coefficients are of the order of the L mode values during these ELMs. Analysis shows also that within experimental accuracy the transport coefficients are the same for VH and ELM-free H mode discharges. The variation in the global energy confinement is mainly due to the differences in the impurity radiation, power deposition profiles and plasma recycling and related convection

The evolution of the energy, momentum and particle transport through the L-H transition are determined in JET NBI heated discharges. Both normal and periodic L-H transitions are studied. It is found that all of the transport coefficients drop at the transition over a wide radial region and not just in the edge region as was previously thought to be the case. Indeed it is shown by two different modelling techniques that the conventional model in which the transport changes in a narrow region at the edge cannot explain the time behaviour of the electron temperature. Measurements of the fluctuation level by reflectometry also show a very fast drop over a wide radial region

A numerical study of the dynamics of the L-H transition in JET was done using models of anomalous transport which depend either on local or on global plasma parameters. Comparison with experimental results shows that the best agreement is reached when using a global model for reducing χe, χi and D in the region where shear becomes large (e.g., outside the q=1 surface). Thus the L-H transition manifests itself as a sudden drop of the transport coefficients everywhere outside the q=1 surface, not only in a narrow layer near the plasma edge. The slow improvement of plasma confinement which follows such a fast transition could then be attributed to the dependence on the local plasma parameters of the transport coefficients in the plasma core

The analysis of discharges with strong on- and off-axis ion cyclotron resonance heating shows that the transport of heat in JET is local and diffusive. Discharges with or without pellet injection are considered; in all cases, the electron temperature profile responds to the location of the heating. The total heat flow Q can be described by a local diffusive model of the form Q = -χ(qψ, |∇Te|) ne ∇Te, where χ is proportional to qψ1.75 and increases with |∇Te|. Several forms of the dependence of χ on |∇Te| are shown to give a reasonable fit to the data

Plasmas with the energy confinement time enhanced by a factor 1.4 to 2.1 over the ITER93H-P H-mode scaling expression have been obtained in 3 different H-mode regimes (excluding pellet injection) in JET. The phase of enhanced confinement has strong similarities with the VH-mode observed in DIII-D in all 3 cases. The JET data clearly show VH-mode confinement increasing with plasma current and degrading with loss power.

The change in thermal transport across the L to H transition is studied in detail for those JET high performance H-modes which have a very fast transition. It is found that in these pulses the transport changes very rapidly (

Many different models for energy transport in a tokamak have been tested against subsets of the wealth of data accumulated from tokamak experiments and claimed to describe adequately the data used in the tests. Yet no single transport model has emerged as an obvious candidate for accurate extrapolations from today's large tokamak to a reactor. The principles involved in testing a model against data are examined. It is shown how: (i) collinearities in the data, (ii) confinement degradation with power and (iii) measurement errors in the data, can make tests insensitive to particular details of transport models. This lack of sensitivity is illustrated via tests of models against global confinement data from several tokamaks (ITER L mode database) and via tests of different models against local confinement data from one single tokamak (JET database). The latter tests demonstrate that the testing of transport models must include data from two or more tokamaks of different size

This paper describes an update of the H mode confinement database that has been assembled for the ITER project. Data were collected from six machines of different sizes and shapes: ASDEX, DIII-D, JET, JFT-2M, PBX-M and PDX. The updated database contains better estimates of fast ion energy content and thermal energy confinement times, discharges with RF heating, data using boronization, beryllium and pellets, more systematic parameter scans, and other features. The list of variables in the database has been expanded, and the selection criteria for the standard dataset have been modified. We also present simple scalings of the total and thermal energy confinement time to the new dataset

The VH-mode regime of high confinement has been observed in both DIII-D and JET. VH-mode is characterized by thermal confinement twice that seen in H-mode, with the edge transport barrier penetrating deeper into the plasma. Two mechanisms have been identified as important in achieving this high level of confinement. Expansion of the E*B velocity shear turbulence suppression zone is important in allowing reductions in local transport, while access to the second ballooning stability regime in the edge allows avoidance or elimination of ELMs which impede the confinement improvement. The high performance phase of these discharges is usually terminated by an MHD event which removes energy from a large portion of the plasma cross-section, and is followed by an H-mode phase.

The properties of JET plasmas obtained under similar L mode conditions, but using different ion species, are compared. In hydrogenic limiter discharges with NBI heating, particle confinement and sawtooth activity display a clear isotopic dependence. However, the gain in global energy confinement from hydrogen to deuterium is less pronounced than that implied by popular L mode scaling relationships and the improvement appears to be only partly due to a reduction in the local conductive heat flux. No significant change in energy or particle confinement is observed when deuterium is replaced by 3He

From the first principles of plasma physics, the local thermal diffusivity can be expressed in terms of the Bohm diffusivity χB and a function, F, of dimensionless parameters. Tests of transport models, theoretical, empirical or heuristic, expressed in terms of such parameters often produce ambiguous answers, which arise from collinearities in the experimental data. In experiments on dimensionally similar discharges, the scaling of confinement with one single dimensionless parameter can be examined in a more unambiguous fashion. Experiments on JET have been carried out in which only the normalized Larmor radius rho is varied. This is achieved by varying the plasma current, the toroidal field, the density, the ion cyclotron frequency and the power level in a predetermined pattern, such that other dimensionless parameters, for example the normalized collisionality ν*, β and the safety factor qψ, are kept constant. The experiments on JET demonstrate that L mode confinement, both globally and locally, scales with ρ* according to the long wavelength or Bohm scaling. The implications for theory and future experiments are outlined

The dependence of global and local confinement upon plasma current and poloidal field is studied in an analysis of JET data from a series of current ramp experiments. In such experiments the current profile is transiently decorrelated from its steady state shape over a period lasting more than ten confinement times. Global confinement is shown to depend not only on total current, but also on some measure of the current distribution, for example the internal inductance. The origin of this dependence is found in the local analysis to be the dependence of diffusivity on the poloidal field. A weak dependence of diffusivity on shear is observed in the outer half of the plasma. Predictive transport calculations are made in which the electron-ion thermal diffusivities depend on shear; the calculations reproduce the measured temperature profiles to within the measurement errors. Simple mathematical expressions linking global confinement are used to demonstrate that the results from the global and local analysis are fully consistent

An efficient full domain method for plasma equilibrium construction in iron core tokamaks is described. Illustrative calculations and comparison with results obtained from a finite domain method using JET data are given. The results show that the two methods are in agreement with respect to global plasma parameters such as stored energy, internal inductance, volume, and edge safety factor. However, when there is an uncertainty in the magnetic flux loop measurement, the full domain method yields a more accurate separatrix location

Preliminary experiments were carried out in the Joint European Torus (JET) using 120 keV helium (He) neutral beam injection (NBI). Injected power levels up to 5 MW with 3He and 7 MW with 4He, lasting up to 3 s were reached. The 3 s helium NBI produced efficient ion heating and similar global and local energy confinements to those obtained with deuterium (D) NBI, in both limiter and X-point plasma geometries, in L-mode and H-mode plasma regimes. The elimination of beam-plasma and beam-beam fusion reactions by replacing D NBI with He NBI extended the range for measuring ion temperatures with the JET neutron profile monitor and neutron spectrometers.

It is shown that by the use of strong off-axis ICRH following pellet injection, heat transport with both positive and negative heat flux Q can be examined. It is found that a local diffusive model with Q = - χn ∇T is an adequate representation of the heat flux and that there is no need to invoke any non-diffusive flow terms such as a heat pinch

Describes the content of an H-mode confinement database that has been assembled for the ITER project. Data were collected from six machines of different sizes and shapes: ASDEX, DIII-D, JET, JFT-2M, PBX-M and PDX. A detailed description of the criteria used in the selection of the data and the definition of each of the variables is given. The authors also present an analysis of the conditions of the database, the scalings (power law and offset linear) of the data with both dimensional and dimensionless variables, and predictions of the expected confinement time for ITER

The energy confinement properties of different regimes of JET plasmas are investigated. The results of local transport analysis in both L- and H-mode regimes are discussed and the properties of H-mode plasmas produced by neutral beam injection and/or ion cyclotron resonance heating are compared. Two aspects of transport analysis are addressed. Firstly, the relative importance of the electron and ion heat conductivities and a comparison with the effective particle diffusivity since this determines the nature of the turbulence responsible for the degradation in confinement. The second issue is whether theoretical models adequately describe the radial dependence of the measured heat flux. Several theoretical models are considered: the ion temperature gradient driven turbulence model, the Rebut-Lallia critical temperature gradient model, the dissipative and collisionless trapped electron drift wave mode, the Carreras-Diamond resistive ballooning model and the neoclassical pressure gradient driven turbulence model.

A subset of the ITER L-mode and H-mode confinement data is tested against the constraints imposed by various theoretical models for thermal plasma transport. Matrix algebra is used to facilitate such tests. A new result obtained is that the fundamental constraint imposed by the high-β collisional Fokker-Planck equation (Kadomtsev) is satisfied by the data. An additional constraint on the characteristic scale length associated with thermal diffusion can also be satisfied by the data. Dimensionally correct empirical scaling laws embodying theoretical constraints can thus be derived.

The best configuration for the D-T phase of operation in JET is assessed by analyzing two of the high-performance H-mode plasmas obtained in 1989. Several different simulations are performed using the same power levels as in the original D-D case: (a) 140 kV D beams are injected into a tritium plasma, (b) 160 160 kV T beams are injected into a deuterium plasma, and (c) a mixture of beam species is injected into a plasma initially 50:50 D:T. A second set of simulations involves increasing the beam power levels to their predicted design values while retaining the same thermal conductivities as previously. The effect of using two different particle transport models to prescribe the evolution of the thermal density profiles is investigated for each scenario.

The D-D fusion reactivity from JET plasmas has been optimized by using combined neutral beam injection and radiofrequency heating. Reaction rates of up to 2.9 x 10(16) reactions per second (equivalent to neutron emission rates of 1.45 x 10(16) neutrons per second) have been achieved. The best results were obtained in the double-null magnetic configuration or with plasmas limited by the inner wall. By far the largest fraction of the neutron production originates from beam induced reactions, although, in some cases, about 25% of the production can be attributed to the acceleration of deuterium ions by the radiofrequency. The instantaneous neutron yield was limited by a strong influx of carbon impurity ions due to excessive, but local, wall heating.

Neutral beam heated H-mode DIII-D and JET expanded boundary divertor discharges were examined to study the parametric dependence of the thermal energy confinement on the plasma current, plasma size and neutral beam power. Single-null discharges in both machines were examined during the ELM-free phase (ELM stands for edge localized mode) to extract information about the intrinsic H-mode thermal energy confinement time τth. A power law dependence of ELM-free thermal energy confinement was assumed, with the result that for Bτ ≈ 2.2 T and κ = 1.8, τth = C Ip1.03±0.07 PL−0.46±0.06 L1.48±0.09. The size dependence of τth is described by the linear dimension L since the determination of the individual dependences on the minor and major radii is precluded by the similar aspect ratio of the two machines. For this representation of τth (units of s, MA, MW and m), when L is the plasma major radius, C = 0.106 ± 0.011, and when L is the plasma minor radius, C = 0.441 ± 0.044. A dimensionally correct version of the scaling, consistent with the constraints of a collisional high beta model, is τth∝ Ip1.06 PL−0.45 L1.40 ne0.07 Bτ0.06. These results indicate that, within the experimental error, the empirical scaling and the dimensionally correct scaling are the same.

The energy confinement properties of low density, high ion temperature L- and H-mode plasmas are investigated. For L-mode plasmas it is shown that, although the global confinement is independent of density, the energy confinement in the central region is significantly better at low densities than at higher densities. The improved confinement appears to be associated with the steepness of the density gradient. For the H-mode phase, although the confinement at the edge is dramatically improved, which is once again associated with the steep density gradient in the edge region, the central confinement properties are essentially the same as for the standard L-mode. The results are compared in a qualitative manner with the predictions of the ion temperature gradient instability theory and appear to be in disagreement with some aspects of this theory.

From the equations which describe local transport in a turbulent plasma the scaling of the local diffusivity with the plasma parameters can be established. It is shown that, by appropriate choices of time and length scales for the turbulence, the scaling of the global energy confinement can be cast in two limiting forms: a short and a long wavelength scaling. The scaling laws consist of a leading term and a function F whose arguments are the dimensionless normalized plasma collisionality and beta. The scaling of the leading term in the short wavelength expression for global confinement is shown to fit both L-mode and H-mode confinement data from JET. The function F which describes the precise physical mechanism of the turbulence is found to increase with collisionality, but its dependence on the plasma beta remains uncertain. Several theoretical models are found to be a reasonable fit to the JET confinement data; others, however, such as that for the dissipative trapped electron mode, can be discarded. The short wavelength confinement scaling is shown to unify a multiplicity of existing theoretical/empirical scaling laws by suitable choices of the function F. These scaling laws all exhibit a similar dependence of F upon collisionality but have different dependences upon the plasma beta. It is shown that for the JET data it is very difficult to determine the beta dependence and this explains why so many apparently different scaling laws produce reasonable fits to the data. Using the leading term of the scaling, the fusion product can be extrapolated to reactor conditions (ITER) with an uncertainty factor of four. This uncertainty can be reduced to ?30% by the use of a similarity technique in which F is determined directly from the JET data.

Auxiliary heating by neutral beam injection (NBI) or ion cyclotron resonance heating (ICRH) can broaden or steepen the plasma current density profile. A measure of changes to the current density profile J is established by a combination of current moments which yields the profile broadening parameter ζ. JET experiments are demonstrated to increase ζ (broaden J) when NBI is applied to discharges which exhibit either L-mode or H-mode confinement; pellet injection and ICRH or current decay are demonstrated to decrease ζ (peak J). The time evolution of the parameter ζ is shown to correlate with the changes to the Dα emission seen in H-mode discharges with and without edge localized modes. The broadening of J by NBI, especially for H-mode pulses, smarts in the edge region and it is identified with the non-inductive bootstrap current component as a consequence of steep density profiles. Associated with current profile broadening is an increase in the edge pressure and the total stored plasma energy W. An analysis of global confinement (variations of W with power P) is made by splitting W into three parts: the fast ion energy, Wf, the edge energy, W1, which is empirically shown to be proportional to the current profile broadening parameter ζ, and the bulk plasma thermal energy, Wp. It is shown that the scaling for Wp in the L- and H-mode confinement regimes is similar.

A similarity technique is described in which the temperature and density profiles and fusion performance of ITER and DEMO are determined from the D–T and D–D discharges completed on JET. The limitations and errors associated with this technique are also briefly described.

A possible explanation for the discrepancies observed in the scaling of confinement with respect to the dimensionless variables β, ν* and ρ* is presented. It is shown that errors in the stored energy and input power have a substantial effect on the confinement scaling, when the scans in the dimensionless variables are modest. This is mainly due to mismatches in the non scanned parameters.