TFTR Discharges Research Articles

Z eff behavior following Li and C pellet injection into TFTR

Injection of both Li and C pellets into TFTR discharges has provided a variety of almost instantaneous changes to the normal equilibrium profiles. The profile measurements of visible bremsstrahlung intensity, electron density, and temperature were used to deduce Zeff profiles versus time. The evolution Zeff is examined in L mode and supershot cases. A comparison is made to the soft x-ray profile behavior for these same conditions.

Measurement of internal magnetic field pitch using Li pellet injection on TFTR (invited)

A diagnostic technique which measures the direction of the internal magnetic field pitch angle has been used successfully on TFTR. The technique requires the injection of high-speed Li pellets. The magnetic field direction is measured by observing the polarization direction of the intense visible line emission from Li+ (λ≊5485 Å, 1s2p 3P0,1,2→1s2s 3S0) in the pellet ablation cloud. The presence of the large (primarily toroidal) magnetic field causes the line to be split due to the Zeeman effect, and the unshifted π component is polarized with its polarization direction parallel to the local magnetic field. In devices with sufficiently strong fields (B≳4.5 T), the Zeeman splitting of the line is large enough, relative to the linewidth of each Zeeman component, that enough residual polarization remains. Because the pellet moves about 1 cm before the Li+ is ionized (τionization≲10 μs), the time history of the polarization direction (as the pellet penetrates from the outside toward the plasma center) yields the local magnetic field direction. In the TFTR experiment, spatial resolution of the measurement is typically ∼7 cm, limited by the requirement that a large number of photons must be collected in order to make the measurement of the polarization angle. Typically, the pitch of the field is measured with an accuracy of ±0.01 rad, limited by the photon statistics. The measurements of the internal field pitch angle, combined with external magnetic measurements, have been used in a code which finds the solution of the Grad–Shafranov equation, yielding the equilibrium which is the best fit to the measured inputs. The q profile constructed from this equilibrium is believed to be accurate to ∼±10% over the region where there are internal magnetic measurements. Internal field measurements and equilibrium reconstructions have been performed for a variety of TFTR discharges, including 1.6 MA ohmic plasmas where the internal field is measured at the beginning of the current flat top (before the onset of sawteeth) and 2 s into the flat top (with sawteeth), and in extremely high βp(Ip=0.3 MA, βp≊4.5) discharges.

Self-consistency of the principle of profile consistency results for sawtoothing tokamak discharges

The principle of profile consistency states that for fixed limiter safety factor qa there are unique natural equilibrium profile shapes for the current density j(r) (and consequently q(r)) and the electron temperature Te(r) for any tokamak plasma, independent of the shapes of the heating power deposition profiles. The mathematical statements of the three basic consequences of this principle for sawtoothing discharges (i.e. discharges with q(0) ≤ 1) are: (1) r1/a = F1(l/qa) (≈ 1/qa, empirically); (2) ⟨Te⟩/Te0 = F2(l/qa); and (3) a unique scaling law for the central electron temperature,where q(r1) = 1, ⟨Te⟩ is the volume average electron temperature, and F3(qa) = qa/q0. Since Ohm's law relates j(r) to Te(r), the principle of profile consistency dictates that this unique set of functions F1, F2 and F3 remain the same for all sawtoothing discharges in any tokamak, regardless of its size (i.e. a and R), Ip, VL, BT, etc. The paper presents a rather complete and detailed analysis of this self-consistency of the measured values of Te(r), F1, F2 and F3 for sawtoothing TFTR discharges. In particular, the analytical predictions of the profiles of Coppi's Gaussian, exponential, modified exponential and trapezoidal models, as well as the model profiles of Kadomtsev and Campbell et al. are compared with TFTR and TFR data. Some of the principal results are: (1) The empirical profile consistency relation r1/a = 1/qa is an acceptable solution of q(r1) = 1 for all qa dependent profiles. (2) A comparison between experiment and the present analytic results yields [⟨Te⟩/Te0]EXP ≈ [⟨Te⟩/Te0]AN + 0.05 for the profiles of Coppi's Gaussian model, and for the profiles of the models of Kadomtsev and Campbell et al. (3) For all qa independent profiles, F3(qa) = qa/q0 = constant, and, consequently,for all qa dependent profiles, F3(qa) = qa/q0 ∼ qa when r1/a ≈ 1/qa, and, consequently,.(4) Coppi's and Ohkawa's forms of xe(r) yield,while the INTOR value of xe(r) yields.(5) For r1/a ≈ 1/qa, the profiles of Coppi's Gaussian model and those of the models of Kadomtsev and Campbell et al. all predict that the normalized sawtooth amplitude ΔTe/Te ∼ 1/qa, in agreement with the experimental observations. (6) For qa dependent models, universality of profiles exists in suitably reduced co-ordinates when r1/a ≈ 1/qa.

Charge exchange recombination spectroscopy measurements in the extreme ultraviolet region of central carbon concentrations during high power neutral beam heating in TFTR

The carbon concentration in the central region of TFTR discharges with high power neutral beam heating has been measured by charge exchange recombination spectroscopy (CXRS) of the C5+ n = 3–4 transition in the extreme ultraviolet region. The carbon concentrations were deduced from absolute measurements of the line brightness using a calculation of the beam attenuation and the appropriate cascade corrected line excitation rates. As a result of the high ion temperatures (20–30 keV) in most of the discharges, the contribution of beam halo neutrals to the line brightness was significant and therefore had to be included in the modelling of the data. Carbon concentrations have been measured in discharges with plasma currents Ip in the range 1.0-1.6 MA and beam power in the range 2.6–30 MW, including a number of supershots. The results are in good agreement with carbon concentrations deduced from the visible bremsstrahlung Zeff and with metallic impurity concentrations measured by X-ray pulse height analysis, demonstrating the reliability of the atomic rates used in the beam attenuation and line excitation calculations. Carbon is the dominant impurity species in these discharges; the oxygen concentration measured via CXRS in a high beam power case was 0.0006 of ne, compared to 0.04 for carbon. Trends with plasma current and beam power in the carbon concentration and the inferred deuteron concentration are presented. The carbon concentration is independent of plasma current and decreases from 0.13 at 2.6 MW beam power to 0.04 at 30 MW, while the deuteron concentration increases from 0.25 to 0.75 over the same range of beam power. These changes are primarily the result of beam particle fuelling, as the carbon density did not vary significantly with beam power. The time evolutions of the carbon and deuteron concentrations during two high power beam pulses, one which exhibited a carbon bloom (a sudden influx of carbon due to local heating of the limiter) and one which did not, are compared. In both types of discharge, the carbon concentration decreases early in the beam pulse as a result of beam particle fuelling, and the carbon density rises slowly during the beam pulse until the start of the bloom. The electron density rise during the bloom is primarily due to the increase in the carbon density.

Characterization of deposition and erosion of the TFTR bumper limiter and wall

To understand material transport by the plasma in the TFTR tokamak, graphite bumper limiter tiles and metal surfaces have been studied. Detailed measurements of the TFTR inner bumper limiter POCO ™AXF-5Q graphite tiles indicate areas of net erosion and areas of net deposition. These areas are poloidally asymmetric and on the scale of a bay (l/20th of the torus) repeat regularly toroidally. Finer scale measurements indicate that there are subtle variations in the interaction of the plasma with the wall. Furthermore, a study of the correlation of limiter deposits with the type of discharges in TFTR indicates that the material composition depth distribution was determined by the tokamak operational history. In particular, the relative amounts of carbon, hydrogen, oxygen, and metal in the deposits changed over time, reflecting plasma impurity levels. The outer wall of the vessel was not exposed to direct plasma flux and does not show evidence of erosion.

Collector probe measurements of ohmic conditioning discharges in TFTR

Special limiter conditioning techniques using low density deuterium or helium discharges have produced enhanced plasma confinement in TFTR. Measurements with a rotatable collector probe have been made to increase our understanding of the boundary layer during these conditioning sequences. A set of silicon films behind slits facing in the ion and electron drift directions was exposed to four different D + and He 2+ discharge sequences. The amounts of deuterium and impurities trapped in the surface regions of the samples have been measured by different analytical techniques, including nuclear reaction analysis for retained deuterium, Rutherford backscattering spectroscopy for carbon and metals, and Auger electron spectroscopy for carbon, oxygen and metals. Up to 1.9 × 10 −17 cm −2 of deuterium was detected in codeposited carbon layers with D/C generally in the range of the bulk saturation limit. Radial profiles and ion drift/electron drift asymmetries are discussed.

Measurements of radial profiles of the ion temperature and the plasma rotation velocity with the TFTR vertical x-ray crystal spectrometer

The TFTR vertical x-ray crystal spectrometer has now been operating with three crystals and position-sensitive detectors according to the original design specifications. The observed spectra of heliumlike Ti xxi, Cr xxiii, Fe xxv, and Ni xxvii have permitted a detailed comparison with the predictions from atomic theories, and they have provided data on the radial profiles of the ion temperature and toroidal rotation velocity, as well as the radial ion charge-state distribution in TFTR discharges. Central ion temperatures of 20 keV and central plasma rotation velocities of 5×105 m/s have been recorded from plasmas with auxiliary neutral beam heating. These experimental results are presented. Also discussed are further instrumental improvements, such as the installation of two additional crystals and detectors and the installation of γ and neutron shielding, which will make it possible to measure under full DD and DT operation with 27 MW of neutral beam injection where neutron production rates of 1019 neutrons/s are expected.

Measurements of radial profiles of the ion temperature and the plasma rotation velocity with the TFTR vertical x-ray crystal spectrometer (abstract)

The TFTR vertical x-ray crystal spectrometer has now been operating with three crystals and position-sensitive detectors according to the original design specifications. The observed spectra of heliumlike Ti xxi, Cr xxiii, Fe xxv, and Ni xxvii have permitted a detailed comparison with the predictions from atomic theories, and they have provided data on the radial profiles of the ion temperature and toroidal rotation velocity, as well as the radial ion charge-state distribution in TFTR discharges. Central ion temperatures of 20 keV and central plasma rotation velocities of 5×105 m/s have been recorded from plasmas with auxiliary neutral beam heating. These experimental results are presented. Also discussed are further instrumental improvements, such as the installation of two additional crystals and detectors and the installation of γ and neutron shielding, which will make it possible to measure under full DD and DT operation with 27 MW of neutral beam injection where neutron production rates of 1019 neutrons/s are expected.

Conditioning of the graphite bumper limiter for enhanced confinement discharges in TFTR

A strong pumping effect was observed with plasma operation on the toroidal graphite bumper limiter in TFTR. The pumping effect was induced by conditioning the limiter with a short series (10-20) of low density deuterium or helium initiated discharges. The decay constantfor gas fuelled Ohmic discharges was reduced frombefore conditioning to a minimum ofafter conditioning, which corresponds to a reduction in the global recycling coefficient from about 100% to less than 50%. Coincident with the low recycling conditions, low current neutral beam fuelled discharges showed global energy confinement times which were enhanced by a factor of two over the values obtained with an unconditioned limiter. Two models are proposed for the observed pumping effects: (1) a depletion model, based on pumping of hydrogenic species in the near-surface region of the limiter after depletion of the normally saturated surface layer by (carbon and helium) ion induced desorption; and (2) a co-deposition model, based on adsorption of hydrogenic species in carbon films formed by material sputtered from the limiter during conditioning.

Spectroscopic study of impurity behaviour in neutral beam heated and ohmically heated TFTR discharges

Quantitative spectroscopic measurements of Zeff, impurity densities and radiated power losses have been made for ohmically heated and neutral beam heated TFTR discharges at a plasma current of 2.2 MA and a toroidal field of 4.7 T. Variations in these quantities with line average plasma density (n̄e) and beam power up to 5.6 MW are presented for discharges on a movable graphite limiter. A detailed discussion of the use of an impurity transport model to infer absolute impurity densities and radiative losses from line intensity and visible continuum measurements is given. These discharges were dominated by low-Z impurities, with carbon having a considerably higher density than oxygen, except in high n̄e Ohmic discharges where the densities of carbon and oxygen were comparable. Metallic impurity concentrations and radiative losses were small, resulting in hollow radiated power profiles and fractions of the input power radiated being 30–50% for Ohmic heating and 30% or less for beam heating. Spectroscopic estimates of the radiated power were in good agreement with bolometrically measured values. Because of an increase in the carbon density, Zeff rose from 2.0 to 2.8 as the beam power increased from 0 to 5.6 MW, pointing to a potentially serious dilution of the neutron producing plasma ions with increasing beam power. Both the low-Z and the metallic impurity concentrations were approximately constant with minor radius, indicating no central impurity accumulation in these discharges.

Impurity transport in ohmic and neutral-beam-heated TFTR discharges

Impurity transport has been studied in ohmic and neutral-beam-heated TFTR discharges by observation and numerical modeling of time evolutions of emissions from germanium injected using the laser-blowoff method. VUV spectral lines, the total radiated power, and soft X-ray intensities are compared with code predictions in which the impurity flux is assumed to have diffusive and convective terms. In plasmas with I p = 1.4 MA and n e ≈ 2.5 × 10 13 cm −3 , no difference in transport between ohmic and beam-heated discharges is observed. At I p = 0.7 MA and n e ≈ 1.0 × 10 13 cm −3 , a three-fold increase in the diffusion coefficient is found during beam heating.

Evolution of TFTR scrape-off plasmas with neutral beam injection

Langmuir probe radial scans were achieved for several types of neutral beam heated 0.8 MA discharges in TFTR. The I( V) characteristics were analysed to derive n e and T e profiles and extrapolated to infer the global particle confinement time τ p ( t). A one-dimensional edge model was used to infer transport coefficients before and during neutral beam injection. The model indicates that ionization in the scrape-off increases the ion flux to the limiter by a factor ∼ 2 and alters the transport coefficients.

Global particle recycling with ohmic and neutral beam heating in TFTR

High I p , high n e TFTR discharges exhibit a strong density increase throughout most of the period of 0.5 s neutral beam pulses. This increase is consistent with beam fueling. On the other hand, low I p , low n e discharges show n e saturation shortly after the beam turn-on. The latter phenomenon has been observed in smaller devices operated at lower I p (“density-clamping”). The H α emission near the limiter shows a strong density and plasma current scaling [ ( n e/I p) 3 ] but shows little change with beam power toroidal field, and plasma species, or with a change in wall condition (with and without Cr gettering and ZrAl gettering).

Confinement of high-density pellet-fueled discharges in TFTR

TFTR pellet injection results reported by Schmidt (1985) have been extended to higher density and n tau in plasmas limited by a graphite inner-wall belt limiter. Increased pellet penetration and larger density increases were achieved by operation at reduced plasma current (1.6 MA) and minor and major radius (70 cm and 235 cm). Under these conditions, beam heating results have been extended to 7 MW.

Confinement studies of neutral beam heated discharges in TFTR

The TFTR tokamak has reached its original machine design specifications (Ip=2.5 MA and BT=5.2 T). Recently, the D degrees neutral beam heating power has been increased to 6.3 MW. By operating at low plasma current (Ip approximately=0.8 MA) and low density (ne approximately=1*1019 m-3), high ion temperatures (9+or-2 keV) and rotation speeds (7*105 m/s) have been achieved during injection. At the opposite extreme, pellet injection into high current plasmas has been used to increase the line-average density to 8*1019 m-3 and the central density to 1.6*1020 m-3. This wide range of operating conditions has enabled the authors to conduct scaling studies of the global energy confinement time in both ohmically and beam heated discharges as well as more detailed transport studies of the profile dependence.

Surface analysis of TFTR limiter tiles

The plasma contacting surfaces of the TFTR movable limiter consist of an array of titanium carbide coated graphite tiles. The surface characteristics of the graphite tiles were analyzed prior to installation to establish the initial limiter conditions, and after exposure to 800 Ohmically heated discharges in TFTR to provide data on the limiter conditioning process. Some tile surfaces showed nearly complete removal of the TiC surface as a result of plasma exposure. 3D and 2D elemental mapping of the tile near surface regions show deposition of vacuum vessel metals (primarily Ti and Ni) in the form of oxides.