Alcator C-Mod Quarterly Progress Report – November 2002

The primary activity at Alcator C-Mod during the fourth quarter of FY02 was operation of the tokamak. 31 ½ days of physics operation were completed out of a planned 32 days for the fiscal year.

Science Results

Core Confinement and Transport

Experiments to determine the boundary between ELM-free, EDA and ELMy regimes were carried out, exploiting the new shaping capabilities provided by the modifications to the inner divertor. The goal was to map out the regimes in the q/d plane. Most of our previous work was at d ~ 0.4 and found a threshold in q₉₅ in the range 3.5 - 4. Preliminary analysis, for low triangularity, supports our earlier work that found that EDA was harder to obtain for d < 0.35. The new work found that the q threshold for obtaining EDA at d ~ 0.3 was in the range of 4.5 - 5.5. At the higher triangularities, the quasi-coherent mode didn't disappear until q₉₅ was dropped below 3.5-4, about the same as the q threshold for our standard triangularity.

Ken Gentle (U. Texas) reported on his analysis of C-Mod sawtooth data from ECE measurements. The complete electron temperature profiles with high space and time resolution from ECE open the possibility of time-dependent transport analysis, for example of sawteeth. By averaging over many sawteeth, waveforms of sufficiently low noise can be obtained. The analysis shows the expected large outward fluxes inside the mixing radius at the crash that flatten the profile. Surprisingly, the fluxes become negative at larger radii at the crash. The negative fluxes are an indication of missing energy, in this case interpreted as the dissipation associated with reconnection. The amount is small ~ 1 kJ, consistent with a decrease in internal inductance at the crash of 1-2 percent. Excluding the brief interval about the crash, the remainder of the cycle can be analyzed for transport, presented as thermal flux versus temperature gradient. Outside the mid-radius, these curves are very steep (stiff?). Inside the mixing radius, they become almost flat. The results are preliminary and must be refined with better density and power deposition profiles and analysis of ion coupling, but they illustrate the potential of the approach.

Continuing studies of ITBs formed with off-axis ICRF heating have revealed that the ITB foot location in the electron density profile does not depend on the plasma current or on the location of the q = 3/2 radius. There is some evidence that the foot location determined from the analysis of heat pulse propagation, which is thought to be related to the location of a steep reduction in the thermal diffusivity, moves outward slightly with an increase in the plasma current. There is a strong shift in the electron density profile foot location as the toroidal magnetic field is varied, with the LFS (high magnetic field, 6.2 T, 80 MHz ICRF) ITB foot at R~0.75 m and the HFS (B_T=3.8 T, 70 MHz) ITB foot near 0.81 m. This corresponds to a movement of 1/3 of the plasma minor radius.

Experiments to extend density limit studies were undertaken using an inboard gas puff to reach the density limit without ramping down current. We were successful in gas fueling to the empirical limit, nebar ~ 6x10²⁰/m³ at I_p~ 800kA in these experiments. The physics of the limit was investigated by deep insertion of the A-side scanning probe, which was able to follow the evolution of plasma profiles and fluctuations. While there was a gradual cooling as the density is raised, a bifurcation occurred between 0.5 and 0.8 x n_G. This effect was seen as a sharp drop in Te, a somewhat smaller drop in density and a sharp increase in the floating potential. This effect is demonstrated in Fig.1. The role of the edge shear layer in stabilizing edge instabilities, and the importance of its suppression in the density limit, will be a subject of future investigations.

Figure 1

The mode converted ion cyclotron waves (ICW) in He³(H) plasmas have been observed by the PCI system and identified through TORIC modeling and solving of the dispersion equation. The ICW is on the low field side of the ion-ion hybrid layer and has a wavelength shorter than the fast wave and usually longer than the ion Bernstein wave (E. Nelson-Melby, et al, submitted to PRL). Localized direct electron heating by mode-converted ICRF waves in D(H) tokamak plasmas has been observed for the first time in Alcator C-Mod. Both on- and off-axis (high field side) mode conversion electron heating (MCEH) have been observed. The experimental MCEH profiles were compared with the predictions from a toroidal full-wave ICRF code TORIC. TORIC predictions were shown to be qualitatively in agreement with the experimental results for both on- and off-axis MC cases. From the simulations, the electron heating from mode converted ICW and IBW was examined.

The spatial and temporal characteristics of the turbulence in the outboard and inboard SOL of Alcator C-Mod have been investigated using “gas-puff-imaging” and probes. With the imaging we identify intermittently-occurring structures – blobs – as the phenomenon seen as large amplitude, non-gaussian-distributed events on probes. These intermittently-occurring structures have been called ‘striations’, ‘filaments’, blobs’, ‘IPOs’, and ‘avaloids’ by different investigators and have been shown to be responsible for large fractions of the perpendicular particle and heat transport. The ‘blob’ designation comes from the appearance of the perturbation when viewed in cross-section, along the magnetic field. However, these fluctuations have relatively long, but finite parallel wavelengths, so ‘filaments’ is probably a more descriptive term. Although the blobs are evident in the raw ‘gas-puff’ images, they are even clearer in images made by dividing out the average emission pattern. Four such “normalized” snapshots taken 17 msec apart are shown in Fig. 2. (The turbulence is completely uncorrelated between these snapshots.) Two blobs are clearly seen in the first frame. The other three frames show features mostly outside the separatrix and with significantly smaller amplitudes.

The spatial characteristics of the outboard SOL turbulence were determined for two discharge conditions with different field strengths and densities (normalized to the Greenwald density), both different by more than a factor of two. Poloidal correlation lengths were found to be somewhat different, both in magnitude and radial profile. Normalized fluctuation levels were similar. The results were compared quantitatively with 3-D non-linear simulations of the C-Mod edge for which the time-averaged profiles were fixed. The agreement between the simulations and the experimental measurements was good in some cases, e.g. in the radial profiles of poloidal correlation lengths. Good agreement was also found in the poloidal wavenumber spectra over the part of the profile, with significant differences existing in the limiter shadow. The simulations identify the resistive ballooning instability as the dominant linear instability. Inclusion of results measured from the outboard SOL of NSTX allowed comparison of turbulence characteristics over an even greater parameter range in B_T, l_ei/L_c, r_s, gradient scale length. There is a rough, but inconclusive, scaling of the turbulence scale size with the resistive ballooning mode scale size. The ballooning character of the turbulence is more clearly evidenced by the experimental comparison between the fluctuation levels in the inboard and outboard SOLs. Both the gas-puff imaging diagnostics and inboard and outboard fast-scanning probes show normalized fluctuation levels on the inboard side that are typically 5-10 times lower than on the outboard side of the same flux surface. The frequency spectra are also quite different. These observations clearly support a ballooning-like drive and do not support flute-like instability drive for which field line connection to a surface is important. Finally, in discharges close to the density limit, we observe blobs inside the LCFS, implying that their origin is tied to regions of steep gradients, rather than to the boundary between open and closed flux surfaces.

Complete details of the quarterly review may be found at:

http://www.psfc.mit.edu/cmod/sciprogram/Q_rev_Oct02/

Divertor and Edge

Data from the new inner wall scanning probe (ISP) and the scanning probe on the outer midplane (ASP) have been analyzed for a recent run where the magnetic flux balance between upper and lower X-point locations was varied (see MP#323,"Scoping Experiments for Upper Divertor Cryopump Operation in Unbalanced Double-Null H-mode Discharges"). Electron pressure profiles in the scrape-off layer (SOL) recorded by the ISP and ASP are found to agree when the flux surface mapping of the upper X-point (SSEP) was located beyond the main-chamber limiter location. As SSEP was varied towards a double-null magnetic configuration, a sharp drop-off in the density profile was seen in the ISP data. The break-point in the density profile was found to correspond to the location where magnetic flux surfaces change topology from 'common' surfaces (circulating the core plasma) to 'private' surfaces (being restricted to a region on the high field side of the torus). These observations are consistent with recent conclusions drawn from D_a camera measurements which indicated a sharp decay of the density on the high field side of the secondary separatrix flux surface. The data indicate that cross-field transport into the private, high-field magnetic flux zone is greatly reduced relative to that in the common flux zone. This result is consistent with the drive for SOL turbulence being dominated by magnetic curvature effects. In addition, normalized ion saturation current levels in the common flux zone were seen to be reduced by a factor of 5 on the high field side relative to the low field side. These data suggest that while the turbulence drive is dominated by magnetic curvature effects, the fluctuations exhibit a clear ballooning character, ruling out a simple flute-like description for the transport physics in the far SOL.

These probe observations also have important implications for experiments in C-Mod. When operating in a balanced double-null configuration, the SOL on the high field side is very thin (few mm). This should allow the plasma to be efficiently fueled by gas puffed through an existing inner wall capillary tube on the midplane. In addition, the thin inner SOL indicates that particle fluxes to the divertors predominately flow in the outer divertor legs in double-null discharges. A cryopump designed to operate in near double-null discharges should therefore be aimed at pumping the particle flux on the outer divertor leg.

Operations and Diagnostics

Since the last quarterly, there have been approximately 800 plasma discharges, with an overall startup reliability of 75%. On some days, the startup reliability was 95%, and lately there have been on average 32 shots per run day.

Initial experiments with the C-Mod Active MHD Spectroscopy antenna were performed. TAE resonances were observed on most shots showing a factor of two increase above the background at the driving frequency of 421kHz. During a series of density and toroidal field scans, the resonance was observed to be consistent with the expected scaling. Most shots were carried out at a drive power of 400W, but the resonance signature was also clearly observed down to about 20W. Results were initially obtained in inner-wall limited discharges. Two shots taken with SN diverted discharges also exhibited clear resonances at high density and field. Attempts were also made to sweep the frequency over the limited range where matching could be obtained, about 2% with the present matching circuit; resonances were observed, but further analysis is required to determine the resonance widths from these data. Initial estimates based on the TF ramps indicate a Q of about 40 or a damping rate of 2.5%, which is in the range observed in similar experiments on JET.

The plasma current has been raised successfully to Ip>1.7 MA, at a toroidal field of 7.8T. Vessel motion during disruptions at this level was found to be well within allowable values. Many of the high current shots, which had relatively low electron densities, developed locked modes which affected the performance and contributed to increased disruptivity. This effect will be the subject of additional study and analysis.

Diagnostic Neutral Beam

During the latest run campaign, all of the DNB diagnostics have begun producing physics data. The best results so far have been obtained at densities around 1.0x10²⁰/m³ and lower. CXRS has measured boron ion temperatures over the outer half of the plasma, showing good agreement with T_e profiles. Rotation velocities in ohmic plasmas are also being obtained. BES has unambiguously observed the quasi-coherent mode which is characteristic of our EDA H-modes. The large amplitude fluctuations near 100 kHz, a signature of the QC mode, were seen on channel 1 at 88.5 cm, and is demonstrated in Fig.3. The measured fluctuation level of 6.5% in the raw signals is unexpectedly high. MSE-derived profiles of the magnetic field pitch angle has shown a clear differences in central current density before and after sawtoothing starts. This change in the pitch angle between the Ohmic and RF portions of several current ramp discharges is shown in Fig. 4. Even more important, MSE has shown that intense, early ICRF does indeed impede current diffusion into the core.

Figure 3

Figure 4

During the previous maintenance period, the stainless steel cylinder was modified to allow for several off-axis tangential views of the plasma. Three point radial toroidal rotation velocity profiles are now routinely available for C-Mod discharges. Several interesting observations have been made, including the evolution of the velocity profile during ITB discharges, which shows a strong collapse of the co-current rotation inside of the ITB foot while the barrier is evolving. During normal H-mode discharges, the co-current rotation is seen to propagate in from the outside to the core, and an estimate of the momentum confinement time, directly measured for the first time without direct momentum input, is about 100 ms, about a factor of 3 larger than the energy confinement time.

ICRF

The J-port antenna has been operated successfully up to 11 MW/m² in heating phase.

(See mode conversion results above.)

Computing

The migration of the C-Mod data acquisition and analysis systems from the OpenVMS operating system to Linux is well underway. Ten Linux workstations have been configured for use as physics and engineering consoles and we will be installing perhaps an additional ten systems before the next C-Mod run campaign. Two Linux server systems have been installed for file serving and data acquisition. A new RAID subsystem providing an additional 1.5 terabytes of data storage has been installed on one of the Linux servers for C-Mod experimental data. The port of data acquisition and analysis software is near completion and in the final debugging stage.

Lower Hybrid Project

(see separate report)

Domestic Travel

Miklos Porkolab, Ron Parker, Earl Marmar and Amanda Hubbard attended the FESAC Panel meeting in Austin, Texas, Aug. 6-8.

Martin Greenwald was at ORNL Aug. 12-15 to attend a meeting of the FESAC panel on integrated simulation.

A status review of the Lower Hybrid launcher project was held at PPPL on 9/6/2002. Ron Parker, Steve Wukitch, Rui Vieira, and Bill Beck came from MIT. Joel Hosea, Stefano Bernabei, Doug Loesser, Greg Pitonak (DoE), Joe Rushinski, Gerd Schilling, Skip Schoen, and Randy Wilson attended from PPPL. Project costs and schedule were discussed, and visits were made to the shops and labs where launcher fabrication and assembly are taking place.

Miklos Porkolab attended the FESAC meeting in Gaithersburg, MD, on Sept. 12,13.

Martin Greenwald travelled to LBNL to attend a meeting of the ESnet steering committee. Topics discussed included the "Earth Simulator Rapid Response", planned network upgrades and new ESnet Services (PKI and ECS) Oct.2-4.

Amanda Hubbard participated in the meeting of the FESAC Fusion Energy Development Path Panel, at PPPL on Oct. 3.

Amanda Hubbard attended a meeting of the FESAC Development Path panel at LLNL on Oct. 30..

Most of the C-Mod Physics staff attended the APS Division of Plasma Physics Meeting in Orlando Nov.11-15. C-Mod Presentations included two Invited talks, twelve contributed oral presentations, and twenty-two contributed poster presentations.

APS presentations may be found at:

http://www.psfc.mit.edu/cmod/sciprogram/2002_aps_abstracts.htm

Amanda Hubbard attended a meeting of the FESAC panel on Fusion Development Path, held in Orlando on Nov.15 and 16.

Dave Terry, Rick Murray and Charles Schwartz of the C-Mod Engineering Group visited PPPL Nov. 14 and 15 and met with Nevell Greenough, Chris Brunkhorst and Elmer Fredd to discuss their ICRF transmitter operations and new remote controlled amplitude/phase controllers now under test. They also met briefly with Doug Lossier about the LH launcher fabrication and saw some of the work in progress.

International Travel

Martin Greenwald was in Cordoba, Spain to attend the joint EU-US TTF meeting Sept.16-20 and to present an invited talk on density limits and edge turbulence. He also took the opportunity to discuss plans for upcoming density limit experiments on JET.

Bruce Lipschultz was at JET Sept 25 - Oct 3 to participate in experiments to better understand SOL transport. Most of the time was spent becoming familiar with the data /data system and preparing for the run (Oct 1.). Dr. Lipschultz gave a presentation on two subjects - experiments done on C-Mod and DIII-D to compare transport (and the results of this analysis) and the development of a measurement of SOL ionization source rate in JET based on their wide-angle camera. The JET experiment was aimed at extending the DIII-D/C-Mod comparison to an even larger experiment. The experiment was moderately successful. Data were obtained at two currents and fields in ohmic plasmas. The H-mode data had some problems in that there was apparent interference in the Langmuir probe data.

D. Mossessian traveled to JET Oct.1-4 to participate in a joint experiment on dimensionless identity of C-Mod and JET tokamaks. The goal of the experiment is to investigate possible similarity of plasma characteristics (confinement, character of an H-mode, parameters of edge transport barrier etc.) in two different size tokamak plasmas with matched local dimensionless parameters. During this experiment the plasma was created on JET with the shape matching the shape of C-Mod plasmas and appropriately scaled values of plasma current, field, density and auxiliary heating. The H-mode was obtained and edge profiles and global plasma characteristics were measured. The results are being analysed.

A number of C-Mod physicists attended the 19th IAEA Fusion Energy Conference in Lyon Oct.14-19. Papers presented on C-Mod Results included:

Earl Marmar,"Overview of Recent Alcator C-Mod Research"

Brian Labombard, "Density Limit and Cross-Field Edge Transport Scaling

in AlcatorC-Mod"

Dmitri Mossessian, "H-Mode EdgeStability of Alcator C-Modd Plasmas"

John Rice, "Double Transport Barrier Plasmas in Alcator C-Mod"

James Terry,""Experimental/Theoretical Comparisons of the Turbulence in the

Scrape-Off-Layers of Alcator C-mod, DIII-D, and NSTX"

(presented by G. McKee of DIII-D)

Steve Wukitch,"Performance of a Compact Four-Strap Fast Wave

Antenna"

David Mikkelsen (PPPL), "Nonlinear Simulations of Drift-Wave

Turbulence in Alcator C-Mod"

IAEA presentations may be found at:

http://www.psfc.mit.edu/cmod/sciprogram/2002_iaea_abstracts.htm

In addition, Joe Snipes presented a paper on behalf of the ITPA Threshold Working Group entitled "Multi-Machine Global Confinement and H-mode Threshold Analysis".

Ron Parker, Miklos Porkolab, and Bruce Lipschultz also participated in the meeting.

At the IAEA meeting in Lyon, there was a series of meetings on U.S. bilateral collaborative agreements. Miklos Porkolab participated in the US/EU and US/Japan meetings. Earl Marmar participated in the US/Korea and US/PRC meetings; in addition, he participated in the ASDEX IEA Executive Committee Meeting.

Amanda Hubbard attended the meeting of the pedestal ITPA working group in Garching, Oct. 21-23. She also had some useful discussions on 10/24 with Lorne Horton, Wolfgang Suttrop and Gabriella Saibene regarding results of, and plans for, ASDEX-Upgrade and JET pedestal collaborations.

Joe Snipes participated in the H-mode ITPA working group meeting at Cadarache, On Oct.21-23. He then went to JET to work on q-control experiments using LHCD.

Bob Granetz participated in the MHD ITPA working group meeting in Garching Oct. 21-23.

Paul Bonoli participated in the ITPA Steady State and Energetic Particle working group meeting at Cadarache from Oct.21-23.

John Rice participated in the ITPA Internal Transport Barrier Meeting, also in Cadarache, from Oct.21-25.

Bruce Lipschultz participated in the ITPA Divertor Working Group meeting in Lausanne Oct. 21-23.

John Rice attended the IAEA Technical Workshop on Atomic and Plasma-Material Interaction Data for Fusion Science Technology at IPP Julich from Oct.28-30, and gave an invited talk entitled “Spectroscopy of High Z Elements in Alcator C-Mod”.

Near Term Plans

There is currently in progress a maintenance period at C-Mod; among the activities are improvements to the BN tile fastening technique for the ICRF antennas, improvements to the voltage handling capabilities on D and E port antennas and repair of the mirror mounts for DNB diagnostics.

Schedule

The short term schedule may be found at:

http://www.psfc.mit.edu/cmod/operations/short_term.pdf