IN et al.

1

TEST OF THE ITER-LIKE RMP CONFIGURATIONS

FOR ELM-CRASH-SUPPRESSION ON KSTAR

Y. IN

National Fusion Research Institute

Daejeon, Republic of Korea

Ulsan National Institute of Science and Technology

Ulsan, Korea

Email: inyongkyoon@unist.ac.kr

Y.M. JEON, H.H. LEE, K. KIM, G.Y. PARK, and 3D Physics Task Force in KSTAR

National Fusion Research Institute

Daejeon, Republic of Korea

H. PARK, and M.W. KIM

National Fusion Research Institute

Daejeon, Republic of Korea

Ulsan National Institute of Science and Technology

Ulsan, Korea

J.K. PARK

Princeton Plasma Physics Laboratory

Princeton, Unites States of America

A. LOARTE

ITER organization

St. Paul Lez Durance, France

and

J.W. AHN

Oak Ridge National Laboratory

Oak Ridge, Unites States of America

Abstract

KSTAR has demonstrated divertor heat flux broadening during edge-localized-mode (ELM)-crash-suppression using

ITER-like 3-row resonant magnetic perturbation (RMP) for the first time. To address a couple of critical issues in ITER

RMP, robust ELM-crash-suppression methodology has been explored at low q95 and established in KSTAR using low-n

RMPs. Taking full advantage of the ITER-like 3-row in-vessel control coils (IVCC) in KSTAR, a set of intentionally

misaligned RMP configurations (IMC) was tested to investigate whether or not IMC could be compatible with ELM-crash-

suppression, while minimizing electromagnetic loads on RMP coils. As a result, the ITER-like 3-row IMCs were found not

only to have been compatible with the ELM-crash-suppression, but also to have broadened the heat flux in the vicinity of the

outer strike point on divertor. In comparison, the 2-row RMPs have rarely affected the near scrape-off-layer (SOL) heat flux

despite slightly broadened profile change in the far-SOL. Since the divertor heat flux broadening reflects the dispersal of the

peaked near-SOL heat flux, the experimental outcome is quite favorable to the ITER choice of 3-rows, instead of 2-rows.

Nonetheless, since the IMC-driven broadening observed in the attached plasmas in KSTAR might appear substantially

different in the partially detached plasmas in ITER, additional investigation has been conducted to see if RMP-driven, ELM-

crash-suppression could be compatible with detached plasmas. Although no detached plasmas have been identified with

ELM-crash-suppression yet, significantly reduced divertor heat flux was confirmed in high density, ELM-crash-suppressed

plasmas at q95=3.8 using n=2 RMPs. These new findings elevate the confidence level about the ITER RMP system, while the

remaining uncertainties need to be further clarified using the 3-row IVCCs in KSTAR. As long as mode-locking percussion

is minimized along with a quick recovery of wall conditions, the access to the targeted q95 (~ 3) for ITER is foreseen to be

feasible in KSTAR.

1. INTRODUCTION

The success of ITER is of paramount importance not only to enhance the scientific understanding of nuclear

fusion, but also to evaluate the technological feasibility of nuclear fusion reactor [1]. Considering that high

confinement mode (H-mode) operation in ITER is expected to be accompanied by edge-localized-modes (ELM),

EX7-1

a robust and reliable control to either suppress or mitigate

ELMs is a must, not an option, in that the ELM-driven

particle and heat fluxes could endanger the integrity of

the divertor and plasma facing components [2]. One of

the critical issues in the planned ITER resonant magnetic

perturbation (RMP) system require the divertor heat flux

to be sustained below the critical thresholds of material

safety; ~ 10MW/m2 (in steady state), except for ~20

MW/m2 (transiently)[2]. To meet the criteria, several heat

load dissipation methods can be considered, including

rotating RMP [3]. However, due to the concern about

material fatigue related to time-varying electromagnetic

(EM) loads on coils [2, 3], intentionally misaligned

configurations (IMC) have been proposed to not only

spread the divertor heat fluxes, but also alleviate the EM

loads on coils. Unlike many devices (typically equipped

with 2-row RMPs of Upper/Lower coils), the planned

ITER RMP system is composed of 3-row RMP coils. In

that regard, the KSTAR provides an ideal environment

for ITER RMP study, in that robust RMP-driven, ELM-

crash-suppression has been routinely accessible using

ITER-like 3-row in-vessel control coils (IVCC) [4].

Indeed, over the last couple of years, we have established

a robust methodology to fully suppress ELM-crashes

using low-n (i.e. n=1 or 2) RMPs. To address the ITER

relevant ELM control, a systematic exploration of various

RMP configurations at lower q95 plasmas led us to

accomplish RMP-driven, ELM suppression down to q95 =

3.3[5]. Here, n and q95 refer to the toroidal mode number

and edge safety factor at the 95 % normalized flux surface

respectively. As expected, such low q95 plasmas were

found to be much more vulnerable to mode-locking,

which might have changed the wall conditions substantially. Thus, as long as mode-lockings are avoided and a

quick recovery of the wall conditioning (e.g. cryo-pumping or divertor gas-puffing) is secured, the access to the

targeted q95 (~ 3) for ITER is foreseen to be feasible in KSTAR in the near future. Figure 1 shows one of the

nearly stationery ELM-crash-suppressions using n=1 RMP at q95 = 5, whose sustainment of ~30 sec is

comparable to the deuterium fuel wall saturation time in KSTAR [4]. While the ITER-RMP system has been

designed for n=3 or 4 under the ITER baseline scenario (IBS) at q95 ~ 3 [6], the test of ITER-like RMP

configurations has been conducted at various q95 using the low-n RMPs. Specifically, taking full advantage of 3-

row IVCC in KSTAR, a series of intentionally misaligned RMP configurations have been investigated for ELM-

crash-suppression and the relevant divertor heat flux distribution. As a result, we have not only demonstrated the

compatibility of the IMC-driven, ELM-crash-suppression, but also identified a few promising IMC

configurations that could be more effectively used to disperse the divertor heat fluxes in a wider area.

In Section 2, the ITER-like RMP configurations, along with the experimental setup, will be briefly introduced.

In Section 3, the divertor heat fluxes under various IMC-driven, ELM-crash-suppressions are summarized,

while an emphasis is given to the difference of the divertor heat flux broadening between 3-row and 2-row RMP

configurations. In Section 4, a simple linear modelling has been proposed and evaluated with field-line-tracing

calculations, which appears partly consistent with the experimental observations. In Section 5, the progress

about the compatibility study of ELM-crash-suppression with detached plasmas is reported. In Section 6, the

remaining physics tasks and prospects are discussed, along with a short summary.

2. ITER-LIKE RMP CONFIGURATIONS AND EXPERIMENTAL SETUP

To minimize the time-varying EM loads on RMP coils, 3 types of 3-row n=1 RMP configurations have been

explored, while a 2-row n=1 RMP (Upper/Lower rows) configuration without mid-RMP has been compared, as

schematically shown in Figure 2. All of the 3-row intentionally misaligned configurations (IMC) are compared

with the reference configuration of 90° phasing of n=1 RMP. Here, the phasing refers to the toroidal phase

difference between adjacent rows. In KSTAR, such 90° phasing (default) of n=1 RMP has been frequently used

and confirmed effective in suppressing ELM-crashes, as illustrated in Figure 1.

Fig. 1

IN et al.

3

Fig. 2

°

To diagnose the divertor heat fluxes during

RMP-driven, ELM-crash-suppression, a

toroidally rotating n=1 RMP has been applied,

while a high-resolution infrared (IR) camera up

to 9 kHz sampling(~100 s) is utilized [7].

According to a recent study of Type I ELMs, the

duration of ELMy burst of ~400 s (with a peak

of up to ~50 MW/m2 without RMPs) is longer

than a typical parallel connection time of ~200

s. Thus, a divertor heat flux striation pattern

associated with a slowly rotating RMP at 1 Hz

can be accurately probed by such a rotating

RMP, where its plasma response beyond

vacuum calculation needs to be separately

considered. As will be discussed in Section 4,

the measured divertor heat fluxes are compared with the field-line-tracing calculations [8].

3. DIVERTOR HEAT FLUXES DURING IMC-DRIVEN, ELM-CRASH-SUPPRESSION

First of all, all the attempted 3-row IMCs, as well as 2-row RMPs, are confirmed to be more or less compatible

with ELM-crash-suppression, although the allowable degree of misalignment from the reference phasings varies

subject to a type of 3-D configuration.

3.1. 3-row IMC-driven, ELM-crash-suppression and its divertor heat fluxes

Figure 3 shows one of the successful IMC-driven, ELM-crash-suppressions, whose toroidal phases between

rows move “toward” each other. Each phasing is sustained for 2 sec in static IMC, followed by a toroidally

rotating IMC for 1 sec, like a rigid body. As a

result, the IMC-driven, ELM-crash-suppression on

19212 has been fully sustained for both static and

rotating IMC until 10° of misalignment (up to 14

sec). Although the quality of ELM-crash-

suppression at the misalignment of 15° in static

IMC was not so perfect as those at the other

preceding misalignments, the ELM-crashes remain

suppressed in rotating IMC. Figure 4 shows the

measured divertor heat fluxes using the IR camera,

Fig. 4 Divertor heat flux measurement during the

intentionally misaligned 3-D configuration (IMC), as the

phasing between rows is moving “toward” each other by

the denoted amount in comparison with the reference

phasing of 90° (labelled as “0° ” here). Shown are

the time evolutions of the re-aligned divertor heat flux

from IR camera, whose raw signals are subject to radial

excursions within +/- a few millimeters.

Fig. 3 ELM-crash-suppression using intentionally misaligned 3-

D configuration (IMC). Shown are the time evolutions of (a)

two-color-interferometry (TCI) line-integrated average of plasma

density ne, (b) electron temperature Te,

EX7-1

whose quantitative analysis would be feasible up to 17 sec. Note that the divertor heat flux measurements have

been realigned with the border line of private flux region, whose radial position differs from that of a raw IR

camera signal within +/- a few millimeters. Since such realignment leads to the most-stringent condition for

divertor material safety, it does not compromise any physics issues related to the divertor heat flux study, while

providing a more rigorous technical constraint on divertor heat flux distribution. The other 3-row IMCs (i.e.

‘distorted’ and ‘away’ phasings) were also confirmed to be compatible with the ELM-crash-suppression, as

reported elsewhere[5].

3.2. 2-row ELM-crash-suppression and its divertor heat fluxes

While 3-row RMP-

driven, ELM-crash-

suppression is

frequently used in

KSTAR, the 2-row

RMP-driven, ELM-

crash-suppression has

been newly

established for the first

time in 2017.

Primarily, an optimal

phasing for the

Upper/Lower rows of

RMP has been

predicted at -90 °

phasing, according to

the IPEC plasma

response

calculation[9]. As

schematically shown

in Figure 5, such

helical structure

appeared nearly

orthogonal to a typical

helical configuration

associated with 3-row

90 ° phasing in

vacuum. Nonetheless,

when the plasma response is considered for this 2-row RMPs of -90° phasing, the n=1 resonant components

turned out to be dominant, consistent with the experimental outcome. Hence, a systematic ‘adjustment’ from the

reference phasing of -90° has been similarly added to 2-row RMP, which would be exactly the same as a typical

n=1 phasing scan. Interestingly enough, all the supposedly ‘non-optimal’ phasings up to 20°, as denoted in each

configuration, were also found to be compatible with ELM-crash-suppression, as shown in Figure 5. The

corresponding divertor heat fluxes during static and rotating RMPs have been similarly re-aligned.

3.3. Divertor heat flux dispersal in 3-row IMC and 2-row RMPs

As aforementioned, the divertor heat flux dispersal with minimizing EM loads on coils has been investigated

using the rotating IMC and RMPs, which probe the toroidally non-uniform striation pattern on divertor, as

shown in Figs. 4 and 5 (b). While the misalignment increases in both 3-row IMC and 2-row RMP, the peaks of

divertor heat fluxes got reduced, as shown in Figure 6 (a) and (c) respectively. However, once the near-scrape-

off-layer (SOL) peaks are normalized, the divertor heat fluxes under 3-row IMCs have been broadened but

nothing similar appears in 2-row RMPs, as shown in Figure 6 (b) and (d). This reminds us of a similarly

contrasting observation of divertor heat fluxes between ELM-crash-suppression and ELM-crash-mitigation[5],

in which the divertor heat flux dispersal was observed only during ELM-crash-suppression, while nothing

similar appears during ELM-crash-mitigation. So far, when the IMC with “away” phasing is reinforced by the

global ‘kink-resonant’ influences, such broadening has been speculated to be readily available in a non-linear

plasma simulation [5]. On the other hand, the IMC with ‘toward’ phasing is essentially distanced from strong

Fig. 5 ELM-crash-suppression using 2-row (Upper and Lower RMPs only). Shown in (a) are

the time evolutions of (i) two-color-interferometry (TCI) line-integrated average of plasma

density ne, (ii) electron temperature Te,

the time evolution of the re-aligned divertor heat flux from IR camera and the

schematic of the 2-row IMC. Note that the reference phasing of the 2-row RMP is -90°

between Upper/Lower rows, which is equivalent to 270° in a longer path.

IN et al.

5

‘kink-resonance’ effect, so another non-linear coupling, possibly associated with ‘peeling’ dominance, may have

led to a similar broadening of divertor heat flux in this particular configuration.

Arguably, it is possible for the mid-row to have played a

bigger role than the other off-midplane rows in broadening

the heat fluxes. In particular, since the magnetic flux

delivered by the mid-row is roughly doubled than by the other

row in IVCC, such conjecture appears reasonable. In that

regard, even with 2-row RMPs, a similar divertor heat flux broadening might be accessible, as long as the mid-

row RMP is utilized. A follow-up experiment is scheduled to clarify whether or not the heat flux broadening is

exclusively attributable to 3-row IMCs in KSTAR.

4. DIVERTOR HEAT FLUX MODELLING AND FIELD-LINE-TRACING CALCULATTIONS

While each IMC of 3-rows, as well as 2-row RMP, has proven the compatibility with the ELM-crash-

suppression, the fundamental physics rationale responsible for divertor heat flux broadening at the near-SOL is

being investigated. To address such a rather complicated issue in a physically sound way, a simple modeling has

been established to correlate the divertor heat flux with edge temperature and density profiles.

4.1 Divertor heat flux specification for the 3-D field impact assessment on striation pattern

To evaluate the 2-D heat flux pattern in (Rdiv, ) at the outer divertor in KSTAR under the presence of the 3-D

field, a simple model is based on the following two assumptions:

First, for field lines at the divertor which are connected to the core plasma (of the unperturbed plasma without 3-

D fields), the heat flux at the divertor is determined by the plasma density and temperature at the corresponding

flux surface. It is assumed that there is direct connection between the core plasma and the divertor and that the

heat flows mostly along the field lines (i.e. no diffusion due to ergodic diffusion of fields lines). Second, for

field lines at the divertor which are connected to the SOL plasma the heat flux has the same dependency as the

Fig. 6 . Toroidally-averaged divertor heat fluxes and their

normalized shapes in 3-row and 2-row IMCs during ELM-crash-

suppression respectively. Shown on the left are the profiles of

toroidally-averaged divertor heat flux in each IMC, in comparison

with no RMP (black) and a typical rotating RMP with n = 1, +90°

phasing (red) with 3-rows in (a) and -90° phasing with 2-rows in

(c) respectively. On the right are the corresponding normalized

shapes of divertor heat flux at each configuration. Note that the

widths of A-A’ in (b) at the near SOL with 3-rows are broader,

while those of B-B’ in (d) with 2-rows are rarely broadened. Here,

(UM, LM) in (a) and (b) refer to the phasings between rows, and

in (c) and (d) refers the phase difference between upper and lower

rows without mid-row.

Fig. 7 Comparison of divertor heat flux measurement

and field-line-tracing calculation during rotating

RMP. Shown in (a) is the time evolution of the

measured divertor heat during “distorted” IMC

during ELM-crash-suppression. On the other hand,

shown in (b) is the contour of minimum flux surface

calculated in a field-line-tracing, which suggests the

high heat flux would be equivalent to deeper RMP-

field penetration. Note that the dashed magenta box

appears quite similar to the measured heat flux,

which automatically discredits the deep penetration

field near Rdivertor ~ 1.44 m in the simulation, while

the vertical dashed line refers to a typical IR camera

location in KSTAR without the presence of rotating

RMP. The weakly striated pattern in the region A in

(b) matches well with the measurement in (a).

EX7-1

unperturbed plasma (i.e. an exponential power profile decay length of q). The edge pedestal profiles are

assumed with a typical pedestal width = 0.02 m and Tped = 500 eV, while the separatrix temperature is

typically characterized by Tsep = 100 eV. To circumvent a lack of accurate edge density profile measurement in

KSTAR, we assume a ratio of pedestal and separatrix density of nped/nsep = 3-4, based on tokamak experiences.

Here, a typical power decay length is assumed to be q = 2 mm in the KSTAR plasmas, as characterized in

Figure 6. For simplicity, both density and temperature profiles are assumed to increase linearly in the pedestal

region from the separatrix, as the radial position move inwards into the plasma towards the pedestal top:

𝑇(𝑅) = 𝑇𝑠𝑒𝑝 +𝑇𝑝𝑒𝑑 − 𝑇𝑠𝑒𝑝

∆(𝑅𝑠𝑒𝑝 − 𝑅𝑚𝑖𝑑) (1)

𝑛(𝑅) = 𝑛𝑠𝑒𝑝 +𝑛𝑝𝑒𝑑 − 𝑛𝑠𝑒𝑝

∆(𝑅𝑠𝑒𝑝 − 𝑅𝑚𝑖𝑑) (2)

where R is the major radius at the outer midplane, Rsep and Rped are the separatrix and pedestal locations at the

outer midplane respectively, along with ( = Rsep - Rped) and Rped ≤ R ≤ Rsep.

The normalized magnetic flux can be related to Rsep – R by a linear expression or mapped by unperturbed

equilibrium information:

N = 1 + C (R-Rsep) (3)

where N > 1 corresponds to the points in the SOL and N < 1 corresponds to points in the edge plasma for the

unperturbed plasma.

Assuming that all the edge profiles become available, the heat flux profile at the divertor under the 3-D fields

can be specified as (normalized to 1 at R = Rsep):

A. R ≥ Rsep

𝑞𝑑𝑖𝑣(𝑅) = 𝑒(𝑅𝑠𝑒𝑝 − 𝑅)

𝜆𝑞 (4)

B. R ≤ Rsep

In this case the heat flux depends on the plasma T(R) and n(R) from the expressions above and the transport

along the field line are assumed to be either sheath-limited (qdiv ~ nT3/2) or conduction-limited (qdiv ~ T7/2).

So, the following two expressions can be considered for R ≤ Rsep

B.1. Sheath-limited heat flux

𝑞𝑑𝑖𝑣(𝑅) = 𝑛(𝑅) 𝑇(𝑅)

32⁄

𝑛𝑠𝑒𝑝 𝑇𝑠𝑒𝑝3/2 (5.1)

B.2. Conduction-limited heat flux

𝑞𝑑𝑖𝑣(𝑅) = 𝑇(𝑅)

72⁄

𝑇𝑠𝑒𝑝7/2 (5.2)

4.2 Divertor heat flux specification for the 3-D field impact assessment on striation pattern

The field-line-tracing calculation has been successfully conducted to understand the divertor heat flux striation

pattern in KSTAR[8]. A typical comparison between experiments and simulation is the connection length

calculation. However, since the correlation between core plasma and divertor needs to be made under the 3-D

field environment, a more adequate quantity could be the minimum N, where 3-D field line may penetrate to

core plasma ergodically. Here, the field-line-tracing calculation has been performed based on the perturbed 3-D

magnetic field using the IPEC[9]. Figure 7 illustrates such an example of field line calculation of N, min, with a

divertor heat flux striation pattern in experiments under “distorted” IMC. In comparison, some striations outside

the magenta box in Figure 7 (b) do not correspond to any meaningful counterparts in the experimental results. In

fact, it is deemed necessary for us to ignore such spurious field line penetration in the rest of analysis (e.g. the

narrow spikes near Rdiv ~1.445 m, next to the near-SOL peaks in Figure 8). Based on the modeling and field-

line-tracing calculations, Figure 8 summarizes the simulation results on both 3-row IMC (with ‘toward’ phasing)

and 2-row RMPs. In general, the conduction-limited model appears more peaked than the sheath-limited model,

which may be attributable to the difference of temperature powers between two models (i.e. T7/2 vs T3/2). Despite

quite a simple modeling, the field line tracing calculation shows the reduction of the divertor heat fluxes with 3-

row IMC, reasonably consistent with the experimental observations [e.g. Comparison of Figure 6 (a) with

Figure 8 (a)]. However, once the near-SOL peak is normalized, virtually no broadening can be seen, as shown in

the conduction-limited model results in Figure 8 (b). Even if the 2nd peaks (near Rdiv ~ 1.455 m) are included in

determining the width of the peaks, it does not seem to be adequate to predict any broadening using field-line-

tracing alone. As for 2-row RMPs, the field-line-tracing does not differentiate the variation of phasing at all. In a

way, this is understandable, in that a linear simulation, based on both IPEC and field-line-tracing, may not

adequately describe the non-linear phenomena, such as RMP-driven, ELM-crash-suppression. Of course, the 3-

D field requirement necessary for RMP-driven, ELM-crash-suppression can be linearly predicted, as

successfully demonstrated in KSTAR experiments[10].

IN et al.

7

5. PROGRESS IN

THE

COMPATIBILITY

STUDY OF ELM-

CRASH-

SUPPRESSION

WITH DETACHED

PLASMAS

Since the main

focus of divertor

heat flux dispersal

would be the

redistribution of the

peaked near-SOL

heat flux, the 3-D

heat flux

broadening is quite

favourable to the

choice of 3-rows in

ITER, instead of 2-

rows. However, we

are cautiously

aware of the

expectation of

completely

different

broadening

characteristics, which

might take place in

partially detached

plasmas in ITER[11].

Thus, it has been a high-

priority research theme

in KSTAR to investigate

whether RMP-driven,

ELM-crash-suppression

would be compatible

with detached plasmas.

Although a fully

detached plasma under

RMP has not been

obtained yet, we were

able to greatly reduce

heat flux at q95=3.8 using n=2 RMPs in high density plasmas, as shown in Figure 9. THis

6. DISCUSSION AND SUMMARY

The ITER-like intentionally misaligned configuration (IMC) has been successfully demonstrated to be not only

compatible with ELM-crash-suppression, but also promising in broadening the divertor heat fluxes in a wider

area. Although the choice of 3-row ITER RMP is greatly meritorious in this aspect, the physics question about

the broadening mechanism needs to be further clarified. While the IMC with “away” phasing greatly benefited

from a global ‘kink-resonance’ that would interact with edge RMP coupling [5], it remains to be answered what

causes the IMC with “toward” phasing to be similarly accompanied by divertor heat flux broadening. Based on

Fig. 9 Comparison of high (red) and low (black) density plasmas during n=2 RMP-

driven, ELM-suppression at q95=3.4. Shown are (a) the key physics parameters and (b)

divertor heat flux profiles for each discharge.

Fig. 8 Field-line-tracing simulation results with conduction-limited (C) and sheath-limited (S)

models (as described in Section 4) with 3-row and 2-row IMCs during ELM-crash-suppression

respectively. Shown on the left in (a) and (c) are the profiles of toroidally-averaged divertor

heat flux in each IMC, in comparison with a typical rotating RMP with n = 1, +90° phasing

(red) with 3-rows in (a) and -90° phasing with 2-rows in (c) respectively. On the right are the

corresponding normalized and enlarged shapes of divertor heat flux at each configuration. Here,

6, 9, 12 and 15 secs correspond to 0° , 5° 10° , and 15° of misalignment from the reference

phasing respectively.

EX7-1

a simple edge modeling, both conduction- and sheath-limited models appear reasonably consistent with the 3-

row IMC heat flux reduction, as the degree of misalignment from the reference phasing increases. Nonetheless,

the field-line-tracing alone does not seem to be adequate to predict the divertor heat flux broadening, suggesting

the need of more comprehensive non-linear simulation. In the meantime, a more sophisticated edge pedestal

modeling beyond a simple linear modeling may be an intermediate step to test the validity of both conduction-

and sheath-limited models. One of the critically important questions goes to a physics issue of whether or not

RMP-driven, ELM-crash-suppression would be compatible with detached plasmas. Since the increase of gas

puffing is typically accompanied by not only the density increase but also collisionality, a common practice to

prefer a low collisionality plasma for ELM-crash-suppression directly conflicts with the methodology to access

the detached plasma. Considering that the wall conditions of the detached plasmas (expected in ITER) are vastly

different from those of attached plasmas (as seen in KSTAR), the experimental demonstration, as well as

theoretical assessment, is in urgent need.

In summary, KSTAR has demonstrated ITER-like 3-row RMPs could be intentionally misaligned to disperse

divertor heat fluxes in a wider area during ELM-crash-suppression, while minimizing EM loads on RMP coils.

The ITER-like 3-row IMCs were found not only to have been compatible with the ELM-crash-suppression, but

also to have broadened the near-SOL divertor heat flux, which were not seen with 2-row RMPs so far. Although

the compatibility of ELM-crash-suppression with detached plasma has not been confirmed yet, significantly

reduced divertor heat flux was measured in high density ELM-crash-suppressed plasmas. In support of the ITER

RMP system, the remaining physics uncertainties will be further clarified using the 3-row IVCCs in KSTAR.

ACKNOWLEDGEMENTS

This work was supported by the Korean Ministry of Science and ICT for the KSTAR project (NFRI-EN1801-9),

National Research Fund (NRF-2014M1A7A1A03029865), and the 2018 UNIST research fund (1.180056.01),

as well as through active collaboration with PPPL and ORNL. We acknowledge all the KSTAR Team members

that have led the successful operations throughout the 3-D physics experiments.

REFERENCES

[1] Perkins F. W., Post D. E., Uckan N. A., Azumi M., Campbell D. J., Ivanov N., Sauthoff N. R., Wakatani M.,

Nevins W. M., Shimada M., et al. 1999 Nucl. Fusion 39 2137

[2] Loarte A., Huijsmans G., Futatani S., Baylor L. R., Evans T. E., Orlov D. M., Schmitz O., Becoulet M.,

Cahyna P., Gribov Y., et al. 2014 Nucl. Fusion 54 033007

[3] Schmitz O., Becoulet M., Cahyna P., Evans T. E., Feng Y., Frerichs H., Loarte A., Pitts R. A., Reiser D.,

Fenstermacher M. E., et al. 2016 Nucl. Fusion 56 066008

[4] In Y., Park J. K., Jeon Y. M., Kim J., Park G. Y., Ahn J. W., Loarte A., Ko W. H., Lee H. H., Yoo J. W., et

al. 2017 Nucl. Fusion 57 116054

[5] In Y., Jeon, Y.M. Jeon, Park, J.-K., Loarte, A., Ahn, J.-W. Ahn, Lee, J.H., Lee, H.H., Park, G.Y., Kim, K.,

Kim, H.S., Ko, W.H., Rhee, T., Kim, J., Yoon, S.W., Park, H., and the 3D Physics Task Force in KSTAR. 2018

Tamed Stability and Transport using Controlled Non-axisymmetric Fields in KSTAR Submitted to Nuclear

Fusion

[6] Schaffer M. J., Menard J. E., Aldan M. P., Bialek J. M., Evans T. E., Moyer R. A. 2008 Nucl. Fusion 48

024004

[7] Lee H., Kang, C., Seo, D., WI, H., Frerichs, H. G, Kim, H., Kim, H. Park, J.K., Bak, J.G., Kim, K., Schmitz,

Ol., Pitts, R., 2016 H-mode divertor target heat load measurements on KSTAR. Proceedings of the 26th IAEA

Fusion Energy Conference (Kyoto, Japan)

[8] Kim Kimin, Ahn J. W., Lee H. H., Park J. K., Kang C. S., In Y., Kwak J. G., Yoon S. W., Oh O. K., Choe

Wonho. 2017 Phys. Plasmas 24 052506

[9] Park Jong-kyu, Boozer Allen H., Glasser Alan H. 2007 Phys. Plasmas 14 052110

[10] Park Jong-Kyu, Jeon YoungMu, In Yongkyoon, Ahn Joon-Wook, Nazikian Raffi, Park Gunyoung, Kim

Jaehyun, Lee HyungHo, Ko WonHa, Kim Hyun-Seok, et al. 2018 Nature Physics

[11] Pitts R. 2018 Private communications