All opinions expressed herein are those of the authors and should not be reproduced, quoted in publications, or
used as a reference without the author’s consent.
An international journal of news from the stellarator community
Editor: James A. Rome Issue 172 February 2021
E-Mail: James.Rome@stelnews.info Phone: +1 (865) 482-5643
On the Web at https://stelnews.info
Remote participation in the
LHD experiments
International collaboration is essential for fusion plasma
research. However, due to the spreading of the COVID-19
virus, it is not easy to join collaboration experiments
directly, especially in foreign countries. To stimulate collaboration
for the Large Helical Device (LHD) experiments,
activities to support remote participation have been
started. The purpose of the effort is to extend and to
improve the present services [1] and to make remote participation
as comfortable as possible. Towards those goals,
two actions are being taken: (1) to provide data handling
capabilities similar to those available when you are in the
control room, and (2) to update the documents to make
them easier to use by newcomers. Available network
resources are listed on a Web page [2].
The data flow for LHD experiments is shown in the left
part of Fig. 1 [3]. Experimental data are acquired by the
distributed data acquisition system. Acquired binary data
are stored in the “raw data server.” Physical values are
automatically calculated using the so-called auto-ana system
[4] and stored in the “analyzed data” servers in the
“Kaiseki-data format” [5]. Two supporting libraries for
handling raw data and analyzed data are provided.
Because these libraries have interfaces to user-friendly
languages such as PV-WAVE (https://www.perforce.com/
products/pv-wave) and Python (https://www.python.org),
it is quite easy to access both raw and analzed data.
Remote access is available to users with a VPN connection.
An easy-to-use GUI data viewer for Kaiseki-data,
MyView2 [6], has been developed by NIFS and is widely
used in LHD experiments. MyView2 can be operated
remotely with a VPN connection as well. There are Windows/
Linux servers directly connected to the LHD internal
LAN. For those who need to handle large amounts of data,
using those servers can be a good solution.
In this issue . . .
Remote participation in the Large Helical
Device (LHD) experiments
Details are presented on how to join the LHD experiments
remotely. ....................................................... 1
Plasma filaments in the scrape-off layer of Wendelstein
7-X
Large, coherent, field-aligned plasma filaments are
observer at the edge of W7-X. Two-dimensional simulations
agree well with experimental measurements
and indicate that macroscopic characteristics of filament
transport in a stellarator might be independent of
the three-dimensional magnetic field. ..................... 2
Hydrogen isotope effects in confinement structure
formation LHD
In LHD, internal transport barriers are spontaneously
formed in either the ion or electron temperature profile.
We discuss the effects of operation with hydrogen
or deuterium on these barriers. ............................... 4
Fig. 1. Conceptual diagram of the data flow of the LHD
experiment and network resources for remote participants.
Stellarator News -2- February 2021
In the control room, the summary graph and camera monitor
images of the LHD plasma are shown continuously on
the main monitor screen. These images are broadcast via a
Zoom (https://zoom.us/) connection. For chatting, messaging
researchers, video meetings, and sharing documents,
services based on Microsoft Teams (https://www.microsoft.
com/en-us/microsoft-teams/group-chat-software) can
be used (Fig. 1). If you join the LHD topical experimental
group, you can be a member of a Team. Tutorial documents
(https://www-lhd.nifs.ac.jp/LHD/pdf/LHD_Guide/
Tutorial_for_beginners_v1.pdf) are also available for
detailed information.
In summary, network services for remote participants have
been significantly improved. Many international collaboration
experiments are already being performed in the
present experimental campaign. We hope many new collaborators
join the LHD experiment and enjoy experiments
remotely in a comfortable manner.
References
[1] Y. Nagayama et al., “Control, data acquisition, data
analysis and remote participation in LHD,” Fusion Eng.
Design, 83 (2008) 170–175.
[2] https://www-lhd.nifs.ac.jp/LHD/resouces.html (login
password for the collaborator is required).
[3] H. Nakanishi, “Data Acquisition and Management System
of LHD,” Fusion. Sci. Technol., 58 (2010) 445.
[4] M. Emoto et al., “Improvement of Automatic Physics
Data Analysis Environment for the LHD Experiment,”
Fusion Sci. Technol., 74 (2018) 161.
[5] M. Emoto et al., “Server for experimental data from
LHD,” Fusion. Eng. Design 81 (2006) 2019–2023.
[6] C. Moon et al., “MyView2, a new visualization software
tool for analysis of LHD data,” Fusion. Eng. Design
104 (2016) 56–60.
S. Ohdachi and Remote Participation supporting Group
National Institute for Fusion Science
Oroshi-Cho 322-6 Toki-shi, Japan
E-mail: ohdachi.satoshi@nifs.ac.jp
Plasma filaments in the
scrape-off layer of Wendelstein
7-X
Large, coherent, field-aligned plasma filaments (or blobs)
are universal phenomena at the edge of magnetic plasma
devices. They have been measured in stellarators, linear
devices, and tokamaks, where they can provide a significant
source of heat and particle transport. A recent publication
co-first-authored by C. Killer and B. Shanahan [1]
has coordinated experimental measurements and numerical
simulations to characterize filaments in the Wendelstein
7-X (W7-X) scrape off-layer (SOL).
Filaments were observed by probes that measured the ion
saturation current and floating potential. From these measurements,
the size of the density perturbation and radial
velocity of the filaments were calculated. A comparison
with target probes indicated that these filaments extend to
the plasma-facing components. Two-dimensional (driftplane)
seeded filament simulations were then performed
using a cold-ion, full-profile fluid turbulence model in the
BOUT++ framework, an open-source framework for solving
partial differential equations in curvilinear geometry.
Initial conditions characteristic of those measured by the
reciprocating probes were then used to seed the filaments.
As the simulations were two dimensional, parallel sheath
closures were implemented in order to incorporate parallel
dynamics. Furthermore, the nonaxisymmetric nature of the
W7-X SOL was ignored and an average curvature was
assumed. Despite this strong approximation, the scaling of
the experimental filament velocity with perpendicular size
was reflected in simulations with remarkable agreement
(Fig. 1).
Fig. 1. Two-dimensional numerical simulations (orange
and blue circles) recover experimental measurements
(green) of plasma filament velocity scaling as a function of
perpendicular size in the W7-X SOL [1].
Stellarator News -3- February 2021
Aside from the agreement between theory and experiment,
it was determined that the filaments in this work have a
small radial velocity due to the low ballooning drive of
W7-X and therefore do not stray far from the flux surface
at which they originate.
This work, which emphasizes collaboration between theory
and experiment, also indicates that the macroscopic
characteristics of filament transport in a stellarator might
be independent of the three-dimensional magnetic field.
[1] Carsten Killer, B. Shanahan, et al., Plasma Phys. Control.
Fusion 62 (2020) 085003.
Brendan Shanahan and the Wendelstein 7-X Group
Max-Planck-Institut für Plasmaphysik
Wendelsteinstr. 1, D-17491 Greifswald Germany
E-mail: brendan.shanahan@ipp.mpg.de
Hydrogen isotope effects in
confinement structure formation
in LHD
One of the biggest challenges for plasma physicists is to
clarify the background physics of the hydrogen isotope
effect. The hydrogen isotope effect is the phenomenon in
which the heavier the mass of the fuel gas, the better the
plasma confinement becomes. Since heavier ions are
expected to have a larger excursion length in a transport
event, intuitively it is difficult to accept the hydrogen isotope
effect. As future thermonuclear fusion plants are
planned to be operated with a tritium-deuterium fuel mixture,
whether the confinement scaling follows the simple
scaling model—the so-called gyro-Bohm scaling—is a
crucial issue for projecting reactor plasma performance.
From an academic aspect, the hydrogen isotope effect is
also an attractive theme in assessing how the basic properties
of a medium in an open non-equilibrium system determine
the eventual steady-state structure of the system.
In addition to the well-known hydrogen isotope effect
widely observed in the energy confinement time scaling,
the fuel mass dependence of the threshold power necessary
for the edge transport barrier to form across the lowto
high-confinement mode transition is experimentally
revealed. That is, the threshold power is approximately
halved in the deuterium plasma case compared to the
hydrogen plasma case [1]. Such an observation is routinely
obtained in different tokamak devices, but knowledge
concerning stellarators is scant due to the lack of
experimental cases. Since the confinement mode transition
is attractive both for fusion reactor operation and for the
study of nonlinear structure formation, the isotope effect at
the confinement mode transition event needs to be characterized
experimentally. In the Large Helical Device
(LHD), the deuterium experimental campaign was started
in 2017, and direct comparison in plasmas with different
deuterium content is being performed. This contribution is
focused on the production of internal transport barriers
(ITBs), and their relationship to the hydrogen isotope
effect. ITBs are spontaneously formed confinement structures
observed in many tokamaks and stellarators. In LHD,
they are produced typically at low density and high heating
power. Depending on the plasma species being subjected
to the intense heating, the ITB appears either in the
ion temperature profile or in the electron temperature profile.
These structures are called the ion-ITB and the electron-
ITB, respectively, and are investigated separately for
possible hydrogen isotope effects.
The study of the hydrogen isotope effect in the ion-ITB
case is begun by defining the ITB intensity. In LHD, the
typical non-ITB ion temperature profile features a domeStellarator
News -4- February 2021
shaped profile. An indicator of the ITB intensity is defined
based on how much the ion temperature surpasses the typical
L-mode profile shape, which we call the profile gain
factor [2]. Figure 1 presents the typical ion-ITB profiles in
the deuterium case and the hydrogen case. The dashed
curves show the reference L-mode profile for each case.
At first glance, a higher central ion temperature is
achieved in the deuterium case owing to the stronger ITB,
which is evident by the larger deviation of the ion temperature
from the reference L-mode profile. The hydrogen
isotope effect in the ion-ITB strength is systematically
examined by plotting the profile gain factor with respect to
the line-averaged density, as shown in Fig. 1(c). As the
line-averaged density decreases, the profile gain factor
gradually rises due to the ion-ITB formation. The ITB
intensity in deuterium plasmas is consistently higher than
that in hydrogen plasmas in a lower density range. To
explore the underlying physics of the hydrogen isotope
effect in the ion-ITB intensity, a principal component analysis
is performed in a multidimensional parameter space.
It is found that a stronger ion-ITB is formed not only
when the line-averaged density is low enough, but also if
the electron density tends to peak, [3].
The hydrogen isotope effect in the electron-ITB is
assessed based on a perturbative experiment. In a steadystate
plasma discharge, the ECH power is modulated and
the response in the electron temperature gradient is
observed. Figure 2 shows the line-averaged density dependence
of the modulation amplitude in the electron temperature
gradient. When the base electron density is
sufficiently high, the modulation amplitude in the electron
temperature gradient does not depend on the line-averaged
density. However, it monotonically rises as the line-averaged
density is decreased in the lower density range due to
the transient electron-ITB formation/deformation synchronized
with the modulating ECH. The threshold density for
the electron-ITB transition is higher in the deuterium case;
i.e., the electron-ITB is easily formed in the heavier hydrogen
isotope fuel. This isotope effect is predominantly
caused by the branch transition in the local electron thermal
diffusivity shown in Fig. 2(b). In the higher density
range, the local electron thermal diffusivity follows the
scaling, likely due to the density stabilization effect
in the global confinement scaling. Once the density
crosses the electron-ITB threshold, the local electron thermal
diffusivity drops off from the trend [4]. Unlike
the local transport property, the nonlocal contribution
scarcely has any meaningful impact on deuterium plasmas
and hydrogen plasmas [4,5] (not shown in this contribution).
To reveal the underlying physics of the isotope effects in
the ITB plasmas, multiscale turbulent fluctuation diagnostics
play a key role in examining the turbulence transport
contribution. In addition to routinely operated density
fluctuation diagnostics in LHD, such as Doppler back
scattering (DBS) and phase contrast imaging (PCI), a new
beam emission spectroscopy (BES) system is being developed
[6]. The new BES system covers the low-frequency
and low-wavenumber domain that neither the DBS nor the
PCI can access. The new BES system has a large-diameter
narrowband interference filter for selectively sensing
either the hydrogen beam emission or the deuterium beam
emission for the hydrogen isotope experiment. By combining
all of these density fluctuation diagnostics, a wide
parameter range of turbulence is obtained.
FF
Fig. 1. Radial profiles of the ion temperature with the ion-
ITB in (a) deuterium and (b) hydrogen plasmas, and (c)
line-averaged density dependence of the profile gain factor.
Fig. 2. Line-averaged density dependence of (a) the modulation
amplitude in the electron temperature gradient, and
(b) the classical diffusion coefficient in deuterium, hydrogen,
and mixed deuterium-hydrogen plasmas.
ne
–1.2
ne
–1.2
Stellarator News -5- February 2021
References
[1] C. F. Maggi et al., Plasma Phys. Control. Fusion 60,
014045 (2017).
[2] T. Kobayashi et al., Plasma Phys. Control. Fusion 61,
085005 (2019).
[3] T. Kobayashi et al., Sci. Rep. 9, 15913 (2019).
[4] T. Kobayashi et al., Nucl. Fusion 60 076015 (2020).
[5] T. Kobayashi et al., Plasma Fusion Res. 15,1402072
(2020).
[6] T. Kobayashi et al., Plasma Phys. Control. Fusion 62,
125011 (2020).
Tatsuya Kobayashi
National Institute for Fusion Science, National Institutes of Natural
Sciences,
The Graduate University for Advanced Studies, SOKENDAI
322-6 Oroshi-cho, Toki, Gifu 509-5292, Japan
E-Mail: kobayashi.tatsuya@nifs.ac.jp