H????????????????

1
Japan Atomic Energy Agency
H????????????????
?? ?
????? ????????
2
Introduction
H-mode is considered as the standard operation
mode in ITER.

• Steady state with the existence of ELM
  accompanied by pulsating expulsion of heat and
  particle.
• Favorable confinement due to edge transport
  barrier

width
Pedestal characteristics can strongly be linked
to

• Boundary condition determining the core energy
  confinement
• Characterization of ELM losses

3
Dimensionless analysis on spatial width of H-mode
pedestal
Optimizing pedestal width Dped and height is a
crucial issue to control the overall H-mode
confinement. Dimensionless analysis is urgent
issue for extrapolation to ITER.
Dped expressed by non-dimensional parameters are
not consistent among several devices.
This is because a strong co-linearity between r
and b is hard to separate out at the plasma edge.
rx bz ? F(n, q, e, k, d, )

Dped
Examine the scaling of pedestal width by
conducting the non-dimensional transport
experiment introducing hydrogen discharges.
4
Co-linearity between r and b at the plasma edge
in the H-modes
Dependence of pped on Ip yields b and r as a
function of Ip at n const.
µ
µ
I
T
n
Ip1.0-1.5
T
n
This co-linearity is hard to separate in the
H-mode plasmas only with a single species.
5
Examine dependence on r introducing hydrogen
plasmas
(H. Urano, et al., NF 48 (2008) 045008)
Examine the dependence of Dped on b or r while
keeping total bpol fixed in hydrogen and
deuterium H-modes.
(1) If DPED ? r, spatial profiles would deviate
by m0.5.
(2) If DPED ? b, spatial profiles would be
similar.
CONDITION IP1MA, BT2, DELTA0.3
Similar profiles are obtained in H/D plasmas.
This result implies r dependence is weak.
Japan Atomic Energy Agency, Naka Fusion Institute
JT-60U Tokamak p. 5
6
Experiments on r dependence of Dped
Power scan of satisfying b const. in H/D
plasmas reveals the r (? m0.5) dependence of
Dped at fixed density.
Ip 1.08 MA, BT 2.4T d 0.35
Power required to sustain the same b in H-plasma
is 2 times larger than that in D-plasma.
Higher ELM frequency is observed in H-plasma.
7

Similar profiles of ne, Te, Ti were obtained in
H/D plasmas.
Thus, mass effect clearly appeared in the r
profiles (? m0.5) while the profiles of b and n
were similar.
8
x 0.2
Despite of different r at the pedestal by mass
ratio ( 1.4), the similar edge profile are
obtained in H/D plasmas.
In a series of experiments, change of Dped is
small in the variation of rpol at fixed bpol
(Dped ? ap r0.2).
9
Experiments on bpol dependence of Dped
Power scan of satisfying bpol ? Ip4 at the
pedestal with controlling n ? Ip4 gives the
variation of bpol at fixed rpol.
Compare profiles at time slices when rpol
remains constant at the pedestal.
10

In the peripheral region, b is varied while
keeping r and n fixed.
11
0.2
0.1
0.08
Dped/ap
0.06
0.04
x 0.5
0.02
0.1
0.2
0.4
0.6
0.8
bpol
When r (? T0.5/Ip) remains constant at the
pedestal, Dped becomes wider at higher bpol
plasma.
Appropriate power scan gives a data set with the
variation of b. Then, the relation of Dped ? ap
b0.5 is obtained.
12
Scaling of pedestal width
D
µ
r

b
D
µ
r

b
0.2
0.5
a
a
ped
p
ped
p
Based on the non-dimensional transport
experiments using H/D plasmas, the scaling on
H-mode pedestal width is obtained Rho dependence
of the pedestal width is weak while it depends
strongly on beta.
D
Dped / ap
H
0.315 rpol0.2 bpol0.5
13
Summary
Conducting the H-mode experiments using hydrogen
(H) and deuterium (D) discharges, dependence of
pedestal width on non-dimensional parameters were
examined.
Similar pedestal profiles were obtained at H/D
plasmas. When b is raised, pedestal width
increased with a relation expressed as Dped ? ap
r0.2 b0.5.