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.
Oak Ridge National Laboratory is managed by UT-Battelle, LLC, for the U.S. Department of Energy.
Published by Oak Ridge National Laboratory
Building CR 5600 P.O. Box 2008 Oak Ridge, TN 37831-6169, USA
Editor: James A. Rome Issue 145 October 2014
E-Mail: jamesrome@gmail.com Phone (865) 482-5643
On the Web at http://www.ornl.gov/sci/fed/stelnews
Plans for first plasma operation
of W7-X
Present status of the commissioning of W7-X
Wendelstein 7-X (W7-X) has now started its commissioning
in preparation for first plasma operation. A recent picture
of W7-X is shown in Fig. 1. The cryostat is closed,
and has been under vacuum for several months. It is
pumped by five turbomolecular pumps with a total pumping
speed of about 6000 L/s. Leak checking of the cryostat
was initially done with ultrasonic detection, but is now
performed with helium, nitrogen, and neon leak checking.
A number of leaks have been identified and repaired, and
the neutral pressure status as of mid-September was in the
low 10−3 mbar range, now dominated by water outgassing
from the large surface area of the multilayer insulation.
The plasma vacuum vessel remains under air, since a number
of in-vessel components are still being installed. This
also helps with leak finding from the inside (between
plasma vacuum vessel and cryostat vacuum).
The focus on the machine assembly is at the moment transitioning
to peripheral components and commissioning of
key subsystems, such as the helium refrigeration plant,
water-cooling circuits, and the main magnet systems.
A new startup plan: OP1.1, OP1.2, OP2
It was decided in 2013 to perform the first plasma operation
phase, in 2015 with a reduced set of in-vessel components.
Most importantly, the uncooled test divertor unit
(TDU) will not be installed until after this first phase;
instead a limiter configuration will be used. Because of
this, the plan for the operational phases of W7-X was
revised, as shown in Fig. 2. The plan now has three distinct
operational phases, defined primarily through the status
of the in-vessel components and the associated pulse
length and heating power restrictions.
The first plasma operation phase with the limiter is
referred to as OP1.1, the second phase with the TDU is
now referred to as OP1.2, and the steady-state-capable
phase with the high heat flux (HHF) divertor is still
referred to as OP2. OP1.1 was introduced to allow for an
accelerated, fully integrated commissioning of the main
systems on W7-X, and to gain the first physics results
from the device. This mitigates risk and stabilizes the
schedule, since many upgrades and improvements can be
Fig. 1. Recent overview shot of W7-X. (Photo by Beate
Kemnitz)
In this issue . . .
Plans for first plasma operation of W7-X
The Wendelstein 7-X commissioning phase has
started, and the first plasma will occur in 2015. The
startup plan presented here will govern the initial operations
of W7-X. ........................................................ 1
A new review paper of stellarator theory
Per Helander has authored a new review of stellarator
theory intended to be comprehensible to students
entering the field. ..................................................... 6
20th International Stellarator-Heliotron Workshop
The workshop will be held in Greifswald, Germany on
5–9 October 2015. This is intended to coincide with
first operations of Wendelstein 7-X. ........................ 7
Stellarator News -2- October 2014
done in parallel with the TDU installation in the roughly
one year experimental pause between OP1.1 and OP1.2. In
the following, we will present some of the physics elements
that can be addressed in OP1.1.
OP1.1: Limiter operation
In preparation for OP1.1, five uncooled graphite limiter
stripes have been installed symmetrically in the five beanshaped
planes on the inboard side, one of which is shown
in Fig. 3.
The limiters were designed with a three-dimensional (3D)
machined surface optimized to spread the heat loads over a
large surface area. If the five limiters can be loaded symmetrically,
plasma pulses with an integral heat input of
2 MJ should be achievable. The heating will be exclusively
electron cyclotron resonance heating (ECRH), with
heating power of at least 2 MW (up to 5 MW might be
installed).
An asymmetry of the heat loads on the limiters is expected
due to a combination of magnetic field asymmetry and
asymmetry in the actual installation locations of the limiters.
The symmetry of the magnetic field will be measured
by flux surface mapping (see next section) at the end of
OP1.1, since this is important for the future operation of
the divertor. For OP1.1, symmetry of the heat load on the
five limiter stripes will allow the goal of 2 MJ per discharge
to be reached, since each limiter can only absorb
0.4 MJ per discharge. This symmetry can be achieved by
use of the trim coils even without explicitly measuring the
cause of the asymmetric loading. Postponing the field
error measurement saves time on the path towards first
plasma.
Flux surface mapping
The first important physics results on W7-X will be from
flux-surface mapping. The flux-surface mapping will be
done with the standard electron-beam phosphorescent-rod
technique. Two movable phosphorescent rods, each with
an integrated electron gun, have been manufactured and
installed. They will work in unison, with one system used
to place the electron gun on a particular magnetic surface
and emit the electron beam, and the other performing a
Fig. 3. One of the five limiters is shown here, installed on
the heat shields, which are only partly covered with graphite
in OP1.1.
Fig. 2. A recent (tentative) plan for W7-X operation is shown here, with plasma operation phases marked with red, and
hardware installation and commissioning periods in cyan. (Of course, these time lines are always in a state of flux.)
Stellarator News -3- October 2014
sweeping motion of the phosphorescent rod through the
flux surface at a different toroidal location. Figure 4 (top)
shows one of the two systems during motion tests after
installation.
The goal of the first flux-surface mapping campaign is to
confirm that nested flux surfaces exist all the way out to
the limiter, and that the expected value of the rotational
transform has been achieved, so that no large island chains
exist near the last closed flux surface. A Poincaré plot of
the limiter configuration is shown in Fig. 4 (bottom).
The trained eye will notice the flux surface deformations
that indicate a moderately sized internal, natural n/m = 5/6
island chain in the Poincaré plot. This island chain should
be clearly detectable in the flux-surface mapping pictures,
and will therefore measure and verify the location of the
iota = 5/6 surface. This then ensures that no low order
rational values can create islands in the near scrape-off
layer (SOL) behind the limiters. Thus, parallel transport
will not cause any radial widening of the near SOL, and
the convective plasma loads will be effectively absorbed
by the limiters. Therefore, only small convective heat
loads will reach the other plasma-facing components, most
of which are bare metal surfaces in OP1.1.
The second goal of the flux-surface measurement campaigns
in OP1.1 will be to measure and eliminate resonant
low-order magnetic field errors. The so-called standard
configuration of W7-X has iota = 1 and a natural resonant
n/m = 5/5 island chain at the edge. This configuration is
sensitive to n = m resonant field components, where m and
n are poloidal and toroidal mode numbers respectively,
especially to n/m = 1/1 and, to a lesser degree, n/m = 2/2
field errors.
The 1/1 field error can be measured very accurately with
flux-surface mapping of a special magnetic configuration
designed for that purpose. The configuration has iota just
barely above 1 on the magnetic axis, and therefore the axis
itself will shift measurably for a resonant 1/1 field error as
low as B11/B0 = 2 10−5. The flux surfaces for this configuration
are shown in Fig. 5 for a field error of B11 = 1 
10−4 T on the right and no field error on the left. The direction
of the shift of the axis is directly related to the phase
of the 1/1 field error, and the magnitude of the shift allows
a direct measurement of the 1/1 field error amplitude. The
trim coils will then be used to eliminate the measured field
errors and to create well-defined error fields in order to
verify the accuracy of this method.
Fig. 4. Top: One of the installed flux surface measurement
rods during a motion and metrology measurement inside
W7-X under atmospheric pressure. The rotation direction
of the rod is indicated by the red arrow. The electron gun
will be installed at the tip of the rod, where the four-way
cross holds corner cubes in this photo. Bottom: A Poincaré
plot of the limiter configuration showing the location of the
limiter (in red), and other components in black and blue,
well in the shadow of the limiter.
Fig. 5. Flux surfaces for a configuration with iota = 1.01 on
axis are shown. Left: Without field errors. Right: With a 1/1
field error of 10−4 T.
Stellarator News -4- October 2014
SOL physics in a limiter configuration
The limiter plasmas actually provide a unique opportunity
to study SOL transport processes in a helical configuration
with comparably short connection length compared to the
island divertor geometry. Hence, comparative studies
between the limiter and the island divertor operational
phases are most informative for generic edge transport
effects such as the capability of heat flux deposition widening
due to long parallel connection lengths. In the limiter
phase (OP1.1), the parallel connection length will be
very well defined and short due to the five well-localized
limiters and the absence of edge islands. Due to shadowing
effects from one limiter onto another, each limiter will
have two distinctly loaded regions, one region with a short
connection length (about 30 m), and another with a longer
connection length (about 80 m), as shown in Fig. 6. The
limiters will be diagnosed with thermocouples, as well as
infrared (IR) cameras and embedded Langmuir probes.
While the thermocouples serve primarily to monitor the
temperature of the holding structures of the limiters, the IR
cameras and Langmuir probes will together be able to
diagnose the heat load patterns and the SOL plasma
parameters. Together they will provide the first SOL physics
measurements. Here, one outstanding challenge for
stellarators and tokamaks alike is to be able to predict and
control the width of the wetted area, q. Recent publications
[1, 2] assuming that q scales linearly with the connection
length Lc agree with observations from a number
of tokamak experiments. In a stellarator island divertor,
one has a large amount of freedom in choosing Lc and
should therefore be able to design for a desired q, assuming
that the same proportionality holds in stellarator island
divertors. The limiter configuration therefore gives valuable
data points for the SOL width at the aforementioned
relatively short Lc values, to be compared with Lc values
of 100–500 m in later island divertor operation.
Plasma startup and wall conditioning plasmas
The OP1.1 plasma operation phase will be done with
helium as the primary working gas, even though the primary
working gases for the future operation phases will be
hydrogen and deuterium. A smaller number of hydrogen
plasma discharges will be attempted in the last couple of
weeks of OP1.1, in order to gain first experience with
these plasmas. Initial plasma pulses will be short and at
moderate power. For the very first plasmas, heating pulses
at 500 kW with about 100 ms duration will be used, i.e.,
the injected energy will be on the order of 50 kJ. Even for
such short-lived plasmas, significant temperatures should
be reachable. Figure 7 shows predictions for an early discharge
with 60 kJ of absorbed energy from a transport
simulation that includes neoclassical transport and an ad
hoc model for the anomalous transport [3]. It is seen that
the electron temperature should reach several kiloelectron
volts and ion temperatures approach 1 keV. The electron
root feature in the core is also clearly seen.
Fig. 6. Left: The limiter configuration will create a SOL
which has two regions with distinctly different connection
lengths (Lc). Right: The SOL transport diffusion coefficient
D can be determined by measuring the heat load deposition
patterns on the limiters. Shown, left to right, are heat
load patterns calculated for three distinctly different values
of D: 0.5, 1.0, and 2.0 m2/s.
Fig. 7: Assumed density profile and calculated temperature
and radial electric field profiles for a plasma heated
with 500 kW for 100 ms after being fully ionized.
Stellarator News -5- October 2014
As operational experience is gained, the energy per pulse
will be increased with the aforementioned goal of reaching
2 MJ per discharge.
There are several reasons for starting with helium plasmas.
In several other stellarator/heliotrons, including WEGA
and Heliotron-J, ECRH breakdown is significantly easier
in helium. Helium can be used for wall conditioning, so
the first plasma discharges in OP1.1 will be wall (or limiter)
conditioning discharges, as well as serving to commission
the first diagnostics and the ECRH heating
system. However, it is expected that as the limiter and the
walls are conditioned, relatively clean helium plasmas can
be achieved, with significant electron temperatures (several
keV), and an interesting first plasma physics program
can be performed.
ECRH and helium beam gas inlets as feedforward
discharge control tools
Since ECRH will be the only heating method in OP1.1, the
ion temperatures will depend on our success in extending
pulse lengths and increasing density in an at least somewhat
controlled way. The fast actuators for discharge control
in OP1.1 will be the preprogrammed segment control
of the ECRH, allowing on- and off-axis heating with complex
time evolution of the heating profiles, as well as the
high-pressure fast piezo-valve gas boxes that will later be
used for the helium beam diagnostic and active divertor
gas fueling. Both of these systems have submillisecond
response times, and will be used in a feed-forward sense.
Feedback loops for density and temperature control are not
expected to be operational in OP1.1.
Heat pulse propagation studies
The fast programming of the ECRH combined with the
electron cyclotron emission (ECE) diagnostic will allow
for first studies of electron thermal transport through heat
pulse propagation studies. This is of interest, in particular
since the plasmas — owing to the ECRH heating and the
relatively low densities to be achieved — are likely to be
in the electron root with outward-pointing electric fields.
The electron root feature should help keep the plasma discharges
relatively clean from impurities. Modeling shows
that if relatively long-lived, high-density discharges are
achieved, the core plasma will change from electron root
confinement to ion root confinement later in the discharges.
Thus, a comparison between ion and electron root
confinement appears reachable, both for the plasma fluid
and for the impurities.
Diagnostics in OP1.1
The amount of physics that can be learnt in this phase will
depend on many factors. One of these is which diagnostics
are available. More than 20 diagnostics are under development
and are on track for being available for first plasma
operation. However, it is not clear if there will be enough
time and resources to bring all of them into operation for
OP1.1. The resources are being allocated according to a
priority list. The following diagnostics are deemed necessary
in order to start OP1.1:
 Flux surface measurements (verify the quality of the
magnetic field and its topology)
 Neutron counters (needed for operation allowance)
 ECE (Te measurements)
 Interferometer (line-averaged ne measurements)
 Video diagnostic (safety diagnostic, also used for flux
surface measurements)
 Limiter IR and visible observation system (heat load
patterns and symmetry, SOL studies)
 Limiter Langmuir probes (SOL studies)
 Limiter thermocouples (determine discharge duration
and dwell time)
These diagnostics receive special priority and are consequently
all far advanced.
Summary
W7-X has started its commissioning phase in preparation
for the first plasma operation campaign, OP1.1, in 2015.
The cryostat is under vacuum and is being actively leak
tested, in parallel with in-vessel and peripheral completion
work. The main goals of the first operation phase are to do
an integral commissioning of all main systems — cryostat
vacuum, cryoplant, superconducting magnets, plasma vessel
vacuum, the ECRH system, and a first basic set of
diagnostics. The high-priority physics goals of this campaign
are to obtain first results concerning magnetic field
quality, plasma breakdown behavior, confinement, and
edge transport aspects in a helical limiter plasma; dynamics
should also be achievable.
Stellarator News -6- October 2014
References
[1] T. Eich et al., Phys. Rev. Lett. 107, 215001 (2011).
[2] R. J. Goldston, Nucl. Fusion 52, 013009 (2012).
[3] Y. Turkin et al., Phys. Plasmas 18, 022505 (2011).
Thomas Sunn Pedersen, on behalf of the W7-X team and the
topical working group “OP1.1 physics”:
Tamara Andreeva, Hans-Stephan Bosch, Sergey Bozhenkov,
Andreas Dinklage, Yühe Feng, Joachim Geiger, Dirk Hartmann,
Ralf König, Hauke Hölbe, Marcin Jakubowski, Heinrich Laqua,
Matthias Otte, Thomas Sunn Pedersen, Mélanie Preynas, Torsten
Stange, Yuriy Turkin, Max Planck Institute for Plasma Physics,
Greifswald, Germany
Monika Kubkowska, Institute of Plasma Physics and Laser
Microfusion, Warsaw, Poland
Florian Effenberg, Oliver Schmitz, U. Wisconsin-Madison, WI,
USA
David A. Gates, Samuel Lazerson, Princeton Plasma Physics
Laboratory, Princeton, NJ, USA
A new review of stellarator
theory
Newcomers to the field can find it difficult to learn stellarator
theory. Only one book [1] has been published on the
subject, and although there are a number of review papers
about stellarators, these either do not go into mathematical
details or do not describe modern developments. To obtain
an overview of the field, the student must find and read
research articles scattered over several decades in the literature.
To somewhat alleviate the situation, a review paper has
recently been published in the journal Reports on Progress
in Physics [2]. This journal may not be on the radar screen
of most plasma physicists, but it has published review
papers for a general physics audience since 1934. The text
is primarily meant for students and others entering the
field, but could be of interest also to established stellarator
physicists. The emphasis is on basic theory—no experimental
results are mentioned—and the following topics
covered are:
 MHD equilibrium. Magnetic field, magnetic coordinates,
plasma current, origin of the rotational transform,
equilibrium variational principle, rational
surfaces
 Single-particle motion. Guiding-centre Lagrangian,
precession, quasisymmetry, omnigeneity and quasiisodynamicity,
maximum-J configurations
 Kinetic theory. Neoclassical transport, intrinsic ambipolarity,
plasma rotation, bootstrap current
References
[1] M. Wakatani, Stellarator and heliotron devices (Oxford
University Press, 1998).
[2] P. Helander, Theory of plasma confinement in non-axisymmetric
magnetic fields, Rep. Prog. Phys. 77,
087001 (2014).
P. Helander
Max-Planck-Institut für Plasmaphysik
Greifswald, Germany
Stellarator News -7- October 2014
20th International Stellarator-
Heliotron Workshop (ISHW)
The 20th International Stellarator-Heliotron Workshop
will take place in Greifswald, 5–9 October 2015.
The Max Planck Institute for Plasma Physics is pleased to
host this workshop, which is planned to take place against
the background of the first experiments in Wendelstein 7-
X. The workshop will be held at the Alfried Krupp Wissenschaftskolleg,
which is located in the medieval city
center of Greifswald, next to the 14th-century cathedral.
Head of the International Programme Committee: Prof.
Hiroshi Yamada, National Institute for Fusion Science,
Toki, Japan.
Head of the Local Organizing Committee: Prof. Per
Helander, IPP, Greifswald, Germany.
http://www.ipp.mpg.de/3523924/ishw_2015
Preliminary ISHW schedule:
1 May 2015 Start of conference registration
15 May 2015 Abstract submission deadline
30 June 2015 Accommodation booking deadline
30 June 2015 Conference registration deadline
5–9 October 2015 ISHW