2000 PROGRESS REPORT - ENEA - Fusione

E
ITALIAN AGENCY FOR NEW TECHNOLOGIES
ENERGY AND THE ENVIRONMENT
NUCLEAR FUSION DIVISION
2000 PROGRESS REPORT
Activities carried out by ENEA in the framework of the
EURATOM-ENEA Association on Fusion

These activities were carried out by ENEA in the framework of the
EURATOM-ENEA Association on Fusion, with the exception of those
indicated by asterisks.
Cover picture: Toroidal limiter sector assembly sequence
by articulated arm IVROS (In Vessel Robotic System)

1. MAGNETIC CONFINEMENT 9
1.1 INTRODUCTION 9
1.2 FTU FACILITY 9
1.3 PHYSICS RESULTS 22
1.4 PLASMA THEORY 35
1.5 NEW PROPOSALS 48
1.6 JET COLLABORATION 54
2. IGNITOR PROGRAMME(*) 63
2.1 INTRODUCTION 63
2.2 PHYSICS 63
2.3 ENGINEERING OF THE MACHINE 66
CONTENTS
3. TECHNOLOGY PROGRAMME 2000 73
3.1 INTRODUCTION 73
3.2 MAGNETS 74
3.3 VACUUM VESSEL AND SHIELD 82
3.4 FIRST WALL AND DIVERTOR 84
3.5 REMOTE HANDLING 93
3.6 BREEDING BLANKET 98
3.7 IFMIF 101
3.8 NEUTRONICS 103
3.9. FUEL CYCLE 108
3.10 SAFETY AND ENVIRONMENT 109
3.11 MATERIALS 119
3.12 LIQUID METAL AND HYDROGEN/MATERIAL
INTERACTION TECHNOLOGY 125
3.13 THERMAL FLUID-DYNAMICS 133
4. INERTIAL CONFINEMENT 147
4.1 INTRODUCTION 147
4.2 TARGET CHAMBER & DIAGNOSTIC UPGRADING 147
4.3 PREPARATION OF THE NEW EXPERIMENTAL CAMPAIGN 147
4.4 THEORY 147
4.5 DPSSL DESIGN ACTIVITY 151
5. MISCELLANEOUS 155
5.1 ADVANCED SUPER-CONDUCTING MATERIALS AND DEVICES 155
5.2 CRYOGENIC TESTING 159
5.3. NEW HYDROGEN ENERGY 160
5.4 CRYOGENICS 161
5.5 ACCELERATOR-DRIVEN SUBCRITICAL (ADS) 162
5.6 PLASMA FOCUS 165
PUBLICATIONS AND CONFERENCES 173
ORGANIZATION CHART 186
LIST OF PERSONNEL 187
ABREVIATIONS AND ACRONYMS 195
(*) Not in Associations framework

1. Magnetic Confinement
1.1 INTRODUCTION 9
1.2 FTU FACILITY 9
1.2.1 Summary of machine operation 9
1.2.2 Summary of machine maintenance 11
1.2.3 Future activities 11
1.2.4 FTU heating systems Lower Hybrid Heating and Current Drive (LHHC&CD) system 12
1.2.5 Diagnostics 13
1.3 PHYSICS RESULTS 22
1.3.1 Improved confinement at high density, high-field operation 22
1.3.2 Analysis of high electron temperature plasma in current ramp-up scenario 24
1.3.3 ECRH in the post pellet phase 27
1.3.4 LH and LH+ECRH results 28
1.3.5 Active MHD control and Tearing Mode (TM) stabilization with localized ECRH/ECCD 30
1.3.6 Energy transport and electron temperature profile stiffness with localized ECRH 32
1.4 PLASMA THEORY 35
1.4.1 Introduction 35
1.4.2 Spontaneous excitation of zonal flows by Energetic Particle Modes (EPM)
(In collaboration with University of California at Irvine) 36
1.4.3 Particle simulation studies of zonal flow excitation by NL EPM dynamics 38
1.4.4 Particle simulation applications to hierarchical distributed-shared memory parallel(*)
systems: integration of High Performance Fortran (HPF) and OpenMP (In collaboration
with Seconda Università di Napoli) 42
1.4.5 Non-linear zonal dynamics of drift and drift-Alfvén turbulences in tokamak plasmas
(In collaboration with University of California at Irvine and Princeton University Plasma
Physics Laboratory) 44
1.4.6 Transport analysis and modeling of FTU plasmas with the JETTO transport code 46
1.4.7 Poloidal rotations induced in tokamak plasmas by IBW 46
1.4.8 Complex ray-tracing method in high harmonic fast wave propagation and absorption 47
1.5 NEW PROPOSALS 48
1.5.1 FTU-D 48
1.5.2 PROTO-SPHERA (Spherical Plasma for HElicity Relaxation 50
1.6 JET COLLABORATION 54
1.6.1 High-beta plasmas in JET discharges with optimised shear 54
1.6.2 ITB dynamics in JET discharges with optimised shear 57
REFERENCES 59
(*) Not in association framework

1. Magnetic Confinement
1.1 INTRODUCTION
Compact, high magnetic field tokamaks have the advantage of producing thermonuclear grade
plasmas at high plasma density, low impurity concentration and strong electron-ion
equipartition. The Frascati Tokamak Upgrade (FTU) (a=0.3 m, R=0.93 m) can exploit these
features by working up to a magnetic field B=8 T and a plasma current I=1.6 MA. During the
year 2000 campaign, FTU has been operated up to the nominal parameters with good reliability,
avoiding light impurity contamination problems. Furthermore, the shot-by-shot use of the
titanisation system has been essential to obtain high performance discharges. Pellet injection in
these conditions allows for a substantial increase in the energy confinement time
(τ E =H×τ ITER89P , with H≈1.4-1.7), as long as deep fuelling conditions are obtained. A
maximum n eo τ E T io in the range n eo τ E T io ≈1020 m-3 has been achieved in 8T, 1.25 MA
pellet–fuelled ohmic discharges with low impurity content (Z eff ≈1.3) and strong electron-ion
equipartition.
Full exploitation of the various heating systems has been reached. The lower hybrid system
(8 GHz, t pulse =1s) is at present composed by 5 gyrotrons (1 MW each at the generator), feeding
5 grills on two FTU windows. One of the two Lower Hybrid (LH) structures showed a severe
leak problem and was dismantled in 1999, thus reducing the power available to about 1 MW at
the plasma, a value which has been routinely achieved throughtout the year 2000. The Electron
Cyclotron Resonance Heating (ECRH) system [1.1] (140 GHz, t pulse =0.5 s) has been working
at the maximum power level of about 1.1 MW at the plasma (corresponding to three gyrotrons),
making use of the launching capability of the system of injecting power at an oblique angle with
Electron Cyclotron Current Drive (ECCD) capability. The system has been employed both for
transport studies and Magneto Hydrodynamic (MHD) mode stabilisation. By using ECRH on the
current ramp, a central temperature of 14 keV has been achieved at high central plasma density.
Simultaneous pellet injection and ECRH have produced improved confinement phases.
Stabilisation of m=2 tearing modes by ECRH/ECCD have produced a substantial confinement
increase. Furthermore, the combined injection of lower hybrid and electron cyclotron waves in
B=7.2 T discharges has produced the first clear sign of synergy between the two waves, with an
increase of the order of 1 keV in the electron temperature at the injection of electron cyclotron
waves.
1.2 FTU FACILITY
1.2.1 Summary of machine operation
The machine was run during the first six months, while the second part of the year experienced
a shutdown in order to have new systems installed. The operations were characterized by a high
availability of all the systems. For the first time, full performance FTU discharges of 1.6 MA,
B t =8 T, q=2.6 were achieved. Many shots at this level were performed, and the main engineering
parameters were in agreement with the expected values.
The plasma performances of the machine are strongly influenced by light impurities, viz. by
some difficulty to obtain high density; so, the possibility of using the titanization system installed
on FTU was very important.
1564 shots were successfully completed, out of a total of 1740 performed on 76 experimental
days. The average number of successful daily pulses was 20.58. Table 1.I reports the main
parameters for evaluating the efficiency of the experimental sessions.
Figure 1.1 reports the sources of downtime in 2000. It has to be noticed that the time required to
analyse the discharges represents the largest downtime fraction with 34.42%. Throughout the
experimental period, the Tokamak power supplies, as well as the control and data acquisition
system have worked at a very high level of availability. The percentage of time lost on
experimental days because of problems due to the control system was reduced to 20.51%.
9

1. Magnetic Confinement
Table 1.I - Summary of FTU operations in 2000
Jan Feb March April May June July Sept Oct Nov Dec Total
Total pulses 103 349 310 256 316 340 66 0 0 0 0 1740
Successful pulses (sp) 84 303 289 245 294 289 60 0 0 0 0 1564
I(sp) 0,82 0,87 0,93 0,96 0,93 0,85 0.91 0,90
Potential experimental days 6 14 12 10 14 18 4 0 0 0 0 78
Actual experimental days (d) 5 14 12 10 14 17 4 0 0 0 0 76
I(ed) 0,83 1,00 1,00 1,00 1,00 0,94 1,00 0,97
Experimental minutes 1569 5612 5243 4623 5558 6274 1211 0 0 0 0 30090
Minutes of delay 1559 3412 2410 1643 3034 4015 1302 0 0 0 0 17375
I(et) 0,50 0,62 0,69 0,74 0,65 0,61 0,48 0,63
A(sp/d) 16,80 21,64 24,08 24,50 21,00 17,00 15,00 20.58
A(p/d) 20,60 24,93 25,83 25,60 22,57 20,00 16,50 22,89
Delay for system (minutes)
Jan Feb March April May June July Sept Oct Nov Dec Total %
Machine 144 389 118 9 94 379 111 0 0 0 0 1544 8,9
Power supplies 97 539 161 140 795 587 736 0 0 0 0 3055 17,6
Radiofrequency 0 248 177 119 94 42 15 0 0 0 0 695 4,0
Control system (Prometeo) 227 293 375 101 469 366 164 0 0 0 0 1995 11,5
Data Acquisition 85 337 232 15 260 37 30 0 0 0 0 996 5,7
Feedback 5 80 235 0 0 52 0 0 0 0 0 372 2,1
IBM 0 0 0 0 0 201 0 0 0 0 0 201 1,2
Diagnostic system 533 486 149 211 328 366 0 0 0 0 0 2073 11,9
Analysis 467 972 837 999 922 1538 246 0 0 0 0 5981 34,4
Others 1 68 126 49 72 147 0 0 0 0 0 463 2,7
Total 1559 3412 2410 1643 3034 4015 1302 0 0 0 0 17375 100
10

1. Magnetic Confinement
The shutdown was mainly
devoted to installing the
boronization system and the
second LH antenna.
An important outcome of this
period was the successful
utilization of a new remote
handling tool to substitute the
broken tiles of the toroidal
limiter. This new remote arm is
able to remove and/or install one
of the toroidal limiter sectors of
an adjacent port. In other words,
all the twelve toroidal limiter
sectors can be changed by merely
opening four equatorial ports.
Fig. 1.1 - Sources of downtime during the experimental campaign in 2000
The restart of the experimental activity is planned for next January.
1.2.2 Summary of machine maintenance
Maintenance of the FTU systems was carried out according to the FTU equipment maintenance
schedules.
The visual inspection of the vacuum vessel, routinely performed after venting, showed four
broken tiles in the toroidal limiter located in the sectors 2,8 and 12.
Since the disassembly procedure of the ECRH antenna located in port 12 is time consuming, the
remote handling arm was located in port 1 and used to extract the sector 12, following a fully
remote procedure. The broken tiles were then replaced in the vacuum laboratory and one of them
was sent off to the supplier for analysis.
The main activity concerning the control system consisted in the elaboration of new software
tools, namely:
• a real time monitor, to verify which resources in terms of CPUs are available for the system
in order to avoid crashes during the experimental sessions;
• a new Objects Oriented database of the FTU data archive. This work has been the subject of
a Degree Thesis;
• a data server prototype, which uses the CORBA architecture;
• the Intranet Division, which utilizes an APACHE server on COMPAQ platform.
Specific hardware and software were studied and implemented on specific requests for any
diagnostic or additional heating system.
1.2.3 Future activities
The machine will be running for most of 2001. A shutdown period is planned for the summer, in
order to install the second Ion Bernstein Wave (IBW) antenna. This is to increase the injected
power to the megawatt level.
Tests on the boronisation system are also foreseen.
Meanwhile, a new data archive will be implemented.
11

1. Magnetic Confinement
Pd (kW)
1000
800
600
400
200
Vgk = 40
Vgk = 41
Vgk = 42
Vgk = 43
Vgk = 44
Vgk = 45
Vgk = 46
Vgk = 47
Vgk = 48
Vgk = 49
Vgk = 50
Vgk = 51
Vgk = 52
Vgk = 53
Vgk = 54
0
50 60 70
80
Vk (kV)
Fig. 1.2 - FTU LH gyrotron: power vs cathode voltage
Fig. 1.02
These gyrotrons are now in operation on the LH system.
1.2.4 FTU heating systems
Lower Hybrid Heating and Current
Drive (LHH&CD) system
The last two gyrotrons of the
LLHH&CD system, delivered by
Thomson TTE, have been
successfully tested both at the factory
test bench - on water dummy load in
circular waveguide - and at ENEA’s
complete transmission line - on
matched dry loads in rectangular
waveguides.
The diagram (fig. 1.2) reports the
output power vs the cathode voltage
characteristics of one of these
gyrotrons. The linearity of these
characteristics and the extreme
versatility of the gyrotron, with power
ranging from 40 kW up to 850 kW,
have to be pointed out.
For the next experimental campaign (year 2001), the LH system will therefore operate with six
modules, that is to say with two coupling structures.
The new launchers for the LHH&CD system
Conventional Multi-junction (MJ). A grill, based on the MJ principle, has been designed by the
ENEA LH team, in the frame of a collaboration with Commissariat à L’Energie Atomique,
Cadarache France (CEA). The grill has been built by an Italian factory; it has a modular
configuration in which each MJ module is made of two consequent E-plane bi-junctions to split
by four the input Radio Frequency (RF) power. The cross section of the output waveguides is
28×4.2 mm, their geometric periodicity is 5 mm; inherent phase shifters give 90° phase pitch at
the MJ output. Twelve MJ modules are arranged in four poloidal rows and three toroidal
columns, and are fed by a single gyrotron.
This configuration represents a good compromise between the power density in the MJ output
and the peak values of the radiated spectra. The main peak is at n ||0 =1.87 with a module feeding
phase Φ e =0° while, by varying Φ e in the range ±90°, the position of the peaks moves in the
range 1.4÷2.4, as in the present FTU grill.
From the mechanical point of view, the launcher is divided into two sections: the input section,
made of copper in order to reduce the electric losses shared with the PAM (see below), holds
the first set of E-plane bi-junctions, while the output section, made of stainless steel in order to
face the harsh plasma conditions, holds the second set of bi-junctions.
The MJ is now ready for characterisation tests both at low level and full power.
Passive Active Multi-junction (PAM) launcher. The PAM antenna has been proposed in order to
inject the RF power levels (≈50 MW) required by the next generation of machines, and to
withstand the expected thermal and neutron load (respectively 10 MW/m2 and 0.5 MW/m2). The
PAM concept leads to a robust and efficient water-cooled mouth: the thick walls between the
active waveguides allow to accomodate the cooling pipes. At the launcher mouth, passive
12

1. Magnetic Confinement
waveguides (depth=0.25 λ g ) are dug in the walls to enhance the directivity of the launcher.
The launcher is also expected to show good coupling at low plasma density.
PAM electrical characteristics only will be tested on FTU, since a full-scale experiment, relevant
for ITER-like machines, will be performed at Tore-Supra, in the frame of the ENEA-CEA
collaboration.
The FTU PAM has the same modular configuration as the MJ; each module output has two active
and two passive waveguides, cross section 28×5 mm, wall thickness 0.8 mm. It has been designed
for a power density P s ≤8 kW/cm2, corresponding to a safe electric field in the waveguides.
The expected loop voltage drop: (∆V/V≈30%, for a 300 kA plasma current and an average
electron density of 4×10 19 m -3 ) is large enough to test the reliability of the concept, and allows
for direct comparison with standard FTU grills.
The radiated power spectrum has its main peak at n ||0 =2.40, with Φ e =180°; the position of the
peaks moves from 1.6 to 2.6 when Φ e varies in the range ±180°.
The PAM input section is shared with the MJ; the output section, instead, holds the tapers to
thicken the vertical walls of the waveguides, and adjusts the phase shifters to set the 270° pitch
between the active waveguides.
The input section of the launcher is ready for being tested at ENEA, while the output section
(with special attention to all the microwave tools necessary to perform the essential tests on the
two launchers) is under construction at the moment.
The previous LH system for FT (400 kW at 8 GHz) is now under revision in order to allow for
full power tests on the two structures.
ECRH system
The ECRH system [1.1] was run at half its capability for most of last year’s experimental
campaign, which is to say: 2 gyrotrons run at a nominal performance of 800 kW of total power
to the plasma, with 50 ms pulse length. Only for a small part of the experiment one more
gyrotron was available. Because of the failure of the heater filament, the two failed sources
(gyrotrons) were sent back to the manufacturer for repair.
IBW RF system
In order to double the amount of power delivered to the plasma, it was decided that the
installation of a second antenna on the tokamak would take place during the shutdown planned
for the year 2000.
More vacuum ceramic windows were ordered and the second antenna, built during the past
years, was assembled. The layout of the RF system was also completed.
A delay concerning the acquisition of the vacuum ceramic windows has so far prevented the full
system from being installed on FTU.
1.2.5 Diagnostics
γ-ray and neutron measurements
Further analyses of FTU runaway electron measurements have been carried out in collaboration
with the Universidad Carlos III (Madrid), CIEMAT (Madrid) and TRINITI (Moscow): the
13

1. Magnetic Confinement
Counts (arb. un.)
10 4
10 3
10 2
10 1
10 0
# 16362
# 15443
10 -1 0 5 10 15 20 25
Energy (MeV)
Fig. 1.3 - Comparison between Fig. pulse 1.03 height distributions of
Bremsstrahlung γ-rays due to runaway electrons (#16362) and γ-rays
from neutron capture (#15443)
Bremsstrahlung γ-ray spectra, due to
runaway electrons hitting the limiters,
have been measured by means of NaI
scintillators in both ohmic an auxiliary
heated plasma discharges [1.2]. The
neutron contribution to the signal can
be subtracted by comparison with
discharges where no runaway electrons
are present. In fig. 1.3, the pulse height
distributions of Bremsstrahlung γ-rays
due to runaway electrons (discharge
#16362) and of γ-rays from neutron
capture (discharge #15443) are shown.
A comparison has been performed
[1.3] with the predictions of a test
particle model of the runaway
electron dynamics, which includes
acceleration in the electric field,
collisions with the plasma particles
and synchrotron radiation losses. The
results indicate that the behaviour of
high energy runaway electrons (up to
20 MeV) during auxiliary heating is determined by the change in the plasma parameters induced
by the heating scheme (LH, ECRH and IBW). The energy corresponding to the end-point of the
experimental γ-ray energy distribution is in agreement with the maximum energy predicted by
the theoretical model. No evidence has been found for resonant wave-particle interaction at large
electron energy (>3 MeV), such as runaway electron interaction with LH via anomalous Doppler
broadening.
The above runaway test particle model has also been applied to the study of disruptions in JET
and ITER [1.4,1.5]. A systematic study of plasma disruptions has been started for FTU
discharges, with the creation of a database including various major parameters (e.g.: neutron
yield, toroidal magnetic field, electron density): the detailed analysis is being carried out at the
moment.
The FTU 6-channel neutron multicollimator has been re-installed and re-calibrated, after major
repair during the tokamak shutdown.
Ultrafast soft x-ray 2-D plasma imaging system based on gas electron multiplier detector
with pixel read-out
A new diagnostic device in the soft x-ray range, has been developed for magnetic fusion plasmas.
It is based on a Gas Electron Multiplier (GEM) detector and equipped with a true 2-D read-out
system [1.6]. By means of a pinhole camera configuration, the plasma has been imaged on the
detector with a very high sampling rate (tens of kHz). A read-out board with 128 pixels has been
designed for this purpose and coupled to a GEM detector with 2.5×2.5 cm active area. The
system has been set up in laboratory and then successfully tested with the FTU plasma.
GEM detector. The GEM have been developed at CERN by F. Sauli [1.7]; their principle of
operation is shown in fig. 1.4a [1.8]. The detector consists of a “drift” and a “transfer” gas
volume, separated by a composite mesh acting as an amplifier (“GEM foil”). The drift region is
defined by the cathode and by the upper face of the GEM foil, while the transfer region lies
between the GEM lower face and the read-out Printed Circuit Board (PCB). The GEM foil is a
14

1. Magnetic Confinement
thin polymer foil, metal-clad
on both sides and pierced by
narrow holes at high density
(typically 70 µm at 140 µm
pitch). Potential differences
applied between the cathode
and the read-out PCB and to
the GEM foil generate the
field structure as shown in
fig. 1.4b. Primary electrons,
produced by the photoelectron
following the absorption of
the x-ray photon in the upper
part of the chamber, drift into
the holes, where the
multiplication occurs.
Electrons are then collected on
the Lower Printed Board
(LPB). These detectors have
an important feature: namely,
the separation of the functions
of electron multiplication and
read-out. Consequently, the
read-out board can be
designed with any geometry
and optimized or adapted for
specific purposes, thus
allowing for a flexible readout,
high counting rate, large
gain range and good spatial
resolution. In our detector,
designed and built by
INFN–Pisa, the drift region is
4 mm high, the transfer region
1.3 mm and the collection
plane is a printed circuit board
with 128 square pixels (2×2
mm each). Electrons
stemming from a converted
photon are collected on the
pixel; they generate a short
current pulse (20 ns), whose
-1400 V
-3000 V
E DRIFT
-900 V
E TRANSFER
0 V
Cathode
Fig. 1.04a
Read-out PCB
total charge is proportional to the energy of the detected x-ray photon. Each pixel is connected
to a fast charge pre-amplifier (LABEN 5231), an amplifier (LABEN 5185), a low threshold
discriminator and a latched scaler.
The gas mixture of the GEM detector used in this experiment is Ar 66% and DiMethilEther
(DME) 33%. The operational voltages of the chamber are determined by the following two
requirements: a high gain to detect photons in the 3-15 keV range, and very high counting rates.
Laboratory tests and preliminary plasma measurements. The whole system has been tested in
laboratory to study the imaging properties at very high counting rates (up to 4 MHz/pixel
corresponding to an x-ray flux of 106 ph/s mm2) in the range 3-10 keV. As an example of the
a)
cathode
5 mm copper
GEM foil
50 mm kapton
Read-out PCB
Fig. 1.4 – Operation principle of the GEM detector. a) planes, electrodes and
polarizations b) cross section of the electric field structure
Fig. 1.04b
b)
GEM foil
15

1. Magnetic Confinement
mm
20
15
10
5
0 5 10 15 20
mm
Fig. 1.5 - Image of a stainless steel wrench placed
close to the detector, at Fig. very 1.05high counting rates
(about 2 MHz/pixel). In the representation, the area
of the square related to each pixel is proportional to
the counts
Counts / ms
5000
4000
3000
2000
1000
0
Lower hybrid
Electron
cyclotron
ch 1
ch 2
ch 3
ch 4
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Time (s)
Fig. 1.06
Fig. 1.6 - Signals from 4 pixels of the GEM detector aligned off
axis for a plasma heated with RF power (LH and ECRH) from
ch1 more central to ch4 at half radius from 0 to 20 cm
imaging capabilities, the shadograph of a wrench
(fig.1.5) is shown, which was obtained with an
exposure time of 50 µs.
An x-ray pinhole camera based on this detector has
been tested on FTU in a preliminary configuration.
The detector’s optical view is limited to roughly one
third of the plasma minor radius. The system has been
then tilted to investigate the intermediate radial
position (8

1. Magnetic Confinement
The camera arrived at the beginning of February, to
be mounted on FTU.
At the beginning of March, L. Delpech (from DRFC-
CEA) joined the team to mount, align, calibrate the
camera, and adapt the software in order to be able to
analyse the data.
In April the camera was operating and providing
good measurements.
Mrs. Delpech remained in Frascati for two months, in
order to ensure the camera maintenance during the
first running time.
Mr. Peysson, the responsible person of the
collaboration for the DRFC, arrived in May for a
two-week stay in order to adapt the elaboration
software and take part in an experimental campaign
with Lower Hybrid Power (LHP) and Electron
Cyclotron Heating (ECH).
Due to the shutdown of Tore-Supra, the camera will
stay in place on FTU until May 2001, to allow for
more measurements.
Technical aspects. The camera was installed on an
equatorial port: the lines of sight are on a poloidal
cross section of the plasma.
Fig. 1.7 - Camera set and aligned
The support structure, built by the
FTU team, allows the camera to
move in a suitable way for the
alignment. A laser and mirror system
has been used to align the camera,
taking the window as a reference
point.
Figure 1.7 shows the camera in situ
on FTU. Figure 1.8 shows the
geometrical lines of sight.
The complete system was calibrated
by using a 241 Am source Detectors
(CdTe, cadmium telluride semiconductors);
electronics and
threshold levels were accurately
checked with a computer system
developed by the Tore-Supra team.
Because of the geometry of the FTU
tokamak, only sixteen lines of sight
are usable.
CAEN and LeCroy modules and
suitable software have been installed
in the FTU acquisition system.
z (mm)
300
200
100
0
-100
-200
-300
700
High energy x-ray spectrometry
800 900 1000 1100 1200
R (mm)
Fig. 1.8 - Geometrical lines of sight of the HXR camera on FTU
Fig. 1.08
30
28
26
24
22
20
18
16
14
12
10
17

1. Magnetic Confinement
600
400
10 9
HXR Spectra
exp. fit
200
10 7
Counts/5 ms
0
150
100
50
0
Intensity (Counts/s)
10 5
10 3
150
100
50
0
0.5 0.6 0.7 0.8 0.9
Time (s)
Fig. 1.9 - HXR raw data (counts), Fig. shot 1.09 #18181, for some chords:
top to bottom central chord; intermediate chord; peripheric
chord
T ph (keV)
50
40
30
20
10
0
Shot 16161, chord 20 - E min = 30 (keV), E max = 170 (keV)
0.55 0.60 0.65 0.70 0.75 0.80
Time (s)
Fig. 1.11 - Time behaviour Fig. 1.11 of photon temperature
10 1
0 50 100 150 200
Energy (keV)
Fig. 1.10 - HXR spectrum Fig. 1.10for central chord
A dedicated software package has been
supplied by DRFC in order to elaborate
experimental data. So, for each energy
level, it is possible to obtain HXR spectra;
photon temperature; line integrated HXR
emission profiles; Abel inverted HXR local
emission profiles.
Some results. The FEB camera worked very
well during the 2000 FTU experimental
campaign, and turned out to be a very useful
diagnostic instrument in the analysis of
FTU experiments with LH and LH plus
ECRH additional power.
Figure 1.9 shows some raw experimental
data. Figure 1.10 shows a HXR spectrum
for central chord during LH power
injection. Figure 1.11 shows the behaviour
of photon temperature for a shot, during LH
and LH plus ECRH power. Figure 1.12
shows an example of HXR local emission
profile for a selected energy range.
The availability of this camera also for the
2001 experimental campaign would be a
great asset to the FTU experiment.
Development of active beam diagnostics
The implementation of new diagnostics, aiming to a better description of the central plasma
configuration in FTU, requires the injection of fast neutral hydrogen atoms into the plasma.
18

1. Magnetic Confinement
Radial profiles of fundamental quantities, such as poloidal magnetic field, ion temperature and
plasma rotation velocities, can be determined through the study of the Motional Stark Effect
(MSE) on the beam-emitted H α radiation and the analysis of the central line emission of the
plasma enhanced by charge exchange with the injected neutrals.
A neutral injector, developed at TdeV laboratories [1.9], has been installed in Frascati and is now
ready for testing as well as for the set up of the diagnostics.
Installation of the Neutral Beam Injector (NBI). The characteristics and scheme of the beam
injector are summarized in table 1.II and fig. 1.13.
An ion beam is produced out of a
DuoPigatron source; it is then
accelerated by a 19-hole grid into a
neutralizer tube, where the fast neutrals
are generated.
The H 2 pressure in the neutralizer has to
be maintained within a specified range in
order to maximize its neutralization
efficiency, while keeping re-ionisation to
a minimum. Due to high pressure in the
source region, a high pumping speed
(>10000 l/s) is required at the
neutralizer’s end: this is also necessary to
avoid any perturbation of the main FTU
plasma. Figure 1.14 shows the
provisional laboratory layout of the
vacuum chamber built to host the injector.
HXR normalized signal
1.0
0.8
0.6
0.4
0.2
Local inversion
40

1. Magnetic Confinement
After completion of the vacuum
tests, setup of the power supplies
and of all the electrical
connections, re-building of a new
cooling system and preliminary
formation of the plasma in the
duoPIGatron source, the system is
ready for operation.
The characterization of the beam
quality will be initially based on
optical observations of its intrinsic
emission and on infrared
observations of a target hit by the
beam itself.
Fig. 1.14 - Overall view of the vacuum chamber containing the neutral
source
Development of diagnostics. a)
MSE. The beam attenuation and
the brightness of the main Stark
component expected in a discharge
with a 1020 m-3 average
electron density are shown in
fig. 1.16. The figure also reports
the radial dependence of the
polarization angle to be measured.
Laboratory test of the Photoelastic
Modulators (PEM) [1.10]
polarimeter on a laser source have
yielded the desired accuracy of the
polarimeter angle (0.1°). The PEM
frequencies (40/46 kHz) and fast
data acquisition (5 Ms/s) will
allow for time resolution of a few
ms on the MSE measurements.
Fig. 1.15 - Internal view of the vacuum chamber showing the grids, the
neutraliser tube and the electron trap (conical cup)
A second polarimeter, based on the
polarization modulation given by a
half-wave plate rotating around its
axis, has been developed and
compared with the PEM
polarimeter. This polarimeter has
the advantage of depending on
phase only, rather than on intensity
measurements. Although comparable
in accuracy, the system
needs further development to obtain
modulation speeds as high as the
PEM ones [1.11].
Special care has been taken in the
design of the optical transport
system for the diagnostic in FTU in order to avoid refractive optics, which could be subject to
Faraday rotation perturbations by the high magnetic field of the tokamak. The optical layout is
reported in fig. 1.17.
20

1. Magnetic Confinement
10 18
n 0 (E 1 ,E 2 ,E 3 )
1.0
0.9
0.8
-H α
4
2
.2MA/5T
.5MA/5T
1MA/8T
5MA/8T
Equivalent current (s -1 )
10 17
10 16
10 15
n 3 (E 1 ,E 2 ,E 3 )
Emission per unit length (s -1 m -1 )
0.7
0.6
0.5
0.4
0.3
Polarization angle (deg)
0
-0
-4
-6
10 14
0.2
-8
0.1
10 13
0
0.5
s (m)
0
10 19 1.0 1.0
0.2 0.4 0.6 0.8
s (m)
-10
0.9 1.0 1.1 1.2 1.3
R maj (m)
Fig. 1.16 - Radial attenuation of a 40 keV/1A beam Fig. in a 1.16 n e =10 20 m -3 plasma: a) total number of atoms (upper
curves) and abundance of the 3s level for each energy component vs path length; b) H a emission per unit path
length; c) polarization angle of the σ Stark; component for some plasma configurations
b) Charge exchange. The measurements of
the ion temperature and plasma velocity will
be based on the spectroscopic observation of
the lines emitted by charge exchange ions in
the plasma centre.
b)
For the purpose of these measurements, a
new spectrometer designed by the Trinity
laboratory spectroscopic group [1.12] will be
used.
The spectrometer is characterized by high
aperture and high spectral resolution obtained
by working with an echelle grating at high
diffraction orders, as reported in table 1.III.
The vertical dimensions and aberrations of
the output image can allow for up to 20
different spatial channels on the plasma. The
optical layout and picture of the spectrometer
are reported in fig. 1.18.
Fig. 1.17 - The relay and detection optics for the MSE
measurements
a)
Fig. 1.17
21

1. Magnetic Confinement
Table 1.III - Spectrometer characteristics and typical operation of the instrument
Spectrometer characteristics
Typical operation
f number 3 λ 6614 Å 5290 Å 4686 Å
Focal length 0.48 m Diffr. orders 9 11 13
Grating 300 gr/mm Linear dispersion 3.5 3.1 2.2 Å/mm
Blaze angel 65° Band of interference filters >300 Å
Filter transmission 80% Average linear dispersion 2.8 Å
Grating reflectivity 60-70% Lines of sight

1. Magnetic Confinement
second pellet launch, central density was up to 7×10 20 m -3 and neutron yield up to 4×10 12 n/s.
Transport analysis has indicated that ion diffusivity was neo-classical. These high performance
plasmas are very sensitive to m=1 internal kink modes, which can grow to a large amplitude and
couple, in some case, to more external modes, leading to mode-locking and disruptions.
In the 2000 experimental campaign, technical operation capabilities of FTU have been extended
up to 8 T, 1.6 MA plasmas with a significant duration of the current plateau (0.6 s at 1.6 MA).
Main results
Previous enhanced confinement regimes have been extended up to B=8 T, I=1.25 MA with
multiple pellet injection, only limited by the availability of pellets and the time duration of the
plateau. Disruptions were avoided by a careful conditioning of the first wall (using titanisation)
and adjusting the target plasma density so as to allow good pellet penetration. Time intervals
between pellets was typically 100 ms, viz. of the order of one energy confinement time.
As shown in fig 1.19, the best confinement was achieved during the last sequence of pellets.
Quasi steady conditions were achieved with a thermal neutron rate up to 1.8×10 13 n/s, a line
averaged density of 4×1020m-3 (peaked density is estimated to be 7-8×1020m-3) and a central
temperature of 2 keV. The resulting values of the fusion figure of merit, nτT, are in the range of
1020m-3 keVs and are achieved in conditions where electron and ion temperatures are equal and
Z eff close to unity.
In an attempt to heat and/or stabilize m=1 modes, effective coupling of Lower Hybrid Current
Drive (LHCD) to high density, high-field plasmas, (n e ~1.2×10 20 m -3 , B T =7.9 T, I p =1.2 MA,
P LH =0.8 MW) has been achieved. A 40% increase in neutron yield, together with a 20% drop
in loop voltage and good confinement was observed with about 1 MW of LHCD power. In some
cases, m=1 modes were stabilised.
At the very high density achieved
during the multiple pellet phase,
FTU pulse #18598
LHCD with its presently available
power, is not effective enough to 4 . 10 20 Line average density (m -3 )
really affect the plasma behaviour.
Several 1.6 MA discharges were
obtained during the last week of
FTU high performance operation.
Pellet injection up to three pellets
was studied. Some of the discharges
were terminated by a disruption for
reasons under investigation, likely
linked to the lack of tuning of the
operation with pellet injection. MHD
activity, which could be associated
with high plasma current density
operation (q edge =2.5), has not
raised any significant problem.
Thanks to titanisation, disruption
recovery was achieved within two or
three standard discharges.
Steady neutron yield of 1013n/s was
achieved in the post pellet phase,
that is to say one of the highest
2 . 10 20
2 . 10 13
1 . 10 13
0
2
Neutron yield (s -1 )
1
Central electron temperature (keV)
0
0.6 0.8 1.0 1.2 1.4
Time (s)
Fig. 1.19 - Time traces of line-average density, neutron
Fig. 1.19
yield and central electron temperature for FTU pulse
#18598, with five pellets injected during the 1.2 MA
current plateau at B=8 T (edge safety factor q=3.3)
23

1. Magnetic Confinement
values achieved so far in FTU discharges. It confirms the trend, indicating that performances of
multiple pellet operation increase with increasing plasma current.
1.3.2 Analysis of high electron temperature plasma in a current ramp-up
scenario
In the current ramp-up phase, very high values of electron temperature (15 keV) and temperature
gradients are being achieved with ECRH in low-density plasmas [1.13], in conditions of low
electron-ion coupling. ECRH operates at the fundamental frequency, with perpendicular lowfield
side-launch and ordinary polarization, so that the resonant magnetic field is 5 T.
The energy transport analysis has been performed using the EVITA code, which allows for both
the interpretative and the predictive time-dependent analysis of a plasma configuration. The
current density profile is obtained through the solution of the diffusion equation for the magnetic
poloidal field (PF); consistency of the profiles obtained with the MHD behaviour of the plasma
discharges has also been checked. The global energy confinement of these discharges is close to
L-mode scaling values (ITER89-P) and the ohmic power is of the same order as the additional
power. In the plasma core, however, the ECRH power is largely the dominant input power term,
so that the local energy transport analysis can be performed in detail.
The main results are described in the following sections; they can be summarized as follows:
experiments with ECRH in the current ramp-up scenario show that low values of electron
thermal conductivity, comparable to the values measured in the core of ohmic plasma discharges,
are observed in the central plasma at much higher values of electron temperature and electron
temperature gradients. The comparison between plasmas with different shapes of current density
profiles indicates that the low thermal conductivity region could be correlated with a low or
negative value of the magnetic shear. Similar results can also be obtained in the current flat-top
phase, in cases when the current profile is far from the standard sawtoothing scenario.
Effect of the current density profile
In the current ramp-up phase, the plasma target is characterized by a variety of shapes of the
current profile, which depend on the
plasma start-up characteristics, gas
a)
10
filling and impurity content. As a
ECRH
consequence, it has been possible to
inject ECRH in plasma both with
0
0.10
0.12 0.14 hollow and peaked current profiles to
t (s)
study the consequences induced on
15
the electron energy transport. When
b) pre-ECRH temperature and current
profiles are peaked, the discharge is
characterized by a quiescient MHD
10
behaviour until the development of
the sawtooth activity, when q=1
surface enters the plasma core. For
5
hollow pre-ECRH profiles, the high
temperature phase obtained by onaxis
ECRH is quenched by a strong
OH
internal reconnection at the time
0
0.1
0.2
0.3
when the minimum value of safety
r (m)
Fig. 1.20 - Pulse #17389, a) peak T
factor q becomes lower than 2, as
e vs time; b) T e radial
profiles at t=0.098 (OH dashed), Fig. 1.20 0.108, 0.113, 0.118, 0.123 s,
shown in fig. 1.20. Here, both the
with 0.9 MW on–axis ECRH
time traces of the peak electron
T e (keV) T e (keV)
24

1. Magnetic Confinement
temperature and of the
profile evolution are
shown. Figure 1.21
compares the results of the
analysis for two plasma
discharges with different
current profiles: the values
of the electron thermal
diffusivity at the plasma
core are similar, but it must
be noticed that the q
profiles are characterized
by low magnetic shear
values (s≤ 0.5) at the
plasma core for both cases.
Off-axis heating
experiments
Off-axis heating experiments
have been performed
either by varying the
location of the resonance
layer by changing the value
of the toroidal magnetic
field B, or by tilting the
ECRH launchers. In the
latter case, the power
deposition is broader. In
figure 1.22 the results of a
radial scan of power
T e (keV)
q
deposition at a fixed B value are shown. In
three different pulses the two launchers were
set respectively as follows: 1) both launchers
at the plasma center, 2) one at the center and
one off-axis, 3) both off-axis. The shape of
the electron temperature profile follows what
is qualitatively expected from a diffusive
model for the plasma energy transport. The
interpretative transport analysis is rather
critical for off-axis experiments, because of
the uncertainty in the power deposition
profile, so a predictive approach has been
applied. The pulses were characterized by
hollow pre-ECRH temperature and current
profiles, To show the compatibility of the
results with a diffusive model, the data of the
full off-axis case have been simulated, fig. 1.23,
Fig. 1.22 - Radial scan of power deposition,
pulses #17389 (full), #17392 (dash), #17393
(dash-dot); a) T e profiles, (b,c,d) ECRH power
density
15
10
5
0
4
2
0
a)
0.1 0.2 0.3
r (m)
2.0
1.5
1.0
0.5
0
b)
-0.5
c) d)
3
2
1
0
0.1 0.2 0.3
r (m)
Fig. 1.21 - Comparison between pulse #15020 (dashed) and #17389 (full) at the
Fig. 1.21
time of maximum temperature; a) T e (r), ohmic data also shown; b) q(r); c)
magnetic shear s(r); (d) χ e (r)
T e (keV)
p ECRH (10 7 w/m 3 )
15
10
5
0
2
0
1
0
1
0
s
χ e (m 2 /s)
t = 0.116 s
1.0 1.1 1.2
r (m)
Fig. 1.22
25
a)
b)
c)
d)

1. Magnetic Confinement
8
# 17393
5
# 17386
6
4
3
T e (keV)
4
T e (keV)
2
2
OH
1
OH
0
0.1 0.2 0.3
r (m)
Fig. 1.23 - Electron temperature profiles at t=0.098 (OH),
Fig. 1.23
0.103, 0.108, 0.113, 0.118 s for off-axis 0.9 MW ECRH; full
line experimental, dashed simulation
0
0.1 0.2 0.3
r (m)
Fig. 1.24 - Electron temperature profiles at t=0.096 (OH),
Fig. 1,24
0.106, 0.11, 0.116 s for off-axis 0.9 MW ECRH; full line
experimental, dashed simulation; the shaded area indicates
the power deposition localization
T e (keV) T e (keV)
10
5
0
0.45
10
5
0
0.8
0.9
0.50
OH
1.0
r (m)
ECRH
0.55 0.60
Fig. 1.25 - Pulse #17578, a) peak Fig. T 1.25 e vs time; b) T e radial profiles
at t=0.486 (OH), 0.506, 0.511 and 0.525 s, with 0.9 MW on-axis
ECRH on current flat top
t (s)
1.1
1.2
a)
b)
by using an ad-hoc electron thermal
diffusivity profile, constant in time, with a
value in the range 0.2-0.5 m 2 /s at the
plasma core. Figure 1.24 shows the result
of an off-axis heating experiment, where
the localization of the resonance has been
changed by changing the value of B in the
case where pre-ECRH profiles are peaked.
In this case, the data can be reproduced by
using the functional form of the Bohm term
of the mixed shear Bohm Gyro-Bohm
model [1.14], multiplied by a factor 2.
Comparison with high electron
temperature on current flat-top
When the start-up phase produces a
plasma with hollow electron temperature
and current density profiles, there are
cases where these features also persist in
the current flat-top phase. This is due to
high radiation from the plasma core, so
that the local ohmic power is balanced by
radiation losses at the plasma center.
On–axis ECRH heating has been applied to such a plasma, fig. 1.25, and a peak T e value in
excess of 10 keV has been obtained at the same density, characterizing the current flat-top
scenario (=4×10 19 m -3 ), at a plasma current I p =0.4 MA. The local analysis of the ohmic
phase of this pulse shows that indeed the balance between ohmic power and radiation losses
26

1. Magnetic Confinement
terms occurs at the center, while the q profile appears to be hollow with q 0 ≈3 and q min ≈2. The
plasma is MHD-quiescent as long as the sawtooth activity does not develop. The T e profile at
the time of the maximum temperature is narrower than in the ramp-up case, fig. 1.20, possibly
because of the reduced width of the low shear (s≤ 0.5) region. Z eff is ~3 in this scenario, while
current ramp-up phase has much higher values, with Z eff =7÷10. The pulse demonstrates that
high T e values can also be obtained on the current flat-top phase, provided that the current profile
is far from the standard sawtooth regime. In this pulse comparison between the experimental
neutron yield and the estimation, made through the ion temperature, evaluated according to
Chang-Hinton ion thermal diffusivity, a degradation of the ion transport during the high electron
temperature phase is shown. Indication of density pump-out is also observed.
1.3.3 ECRH in the post pellet phase
Ohmic plasmas in the post-pellet phase are characterized in FTU by the suppression of the sawtooth
activity; peaked density profiles; reduction of the ion energy transport to the neoclassical value; and
improved global energy confinement. ECRH has been applied to this scenario to check whether
these features are maintained with additional electron heating. The cut-off density at 140 GHz
(2.4×1020 m-3) sets a strong constraint to the experiment, as the pellet injector system was designed
to inject deuterium pellets of a given
size (1-2×1020 atoms), thus allowing
the high-density limit of a high-field
tokamak to be explored. To fulfil this
constraint, a low density plasma target
( ≤ 1×1020 m-3) has been chosen
and off-axis ECRH has been applied.
The results obtained for a plasma pulse
at I p =0.6 MA, B T =5.6 T, (q a =4.8) are
shown in fig. 1.26. ECRH power (0.8
MW) is applied 50 ms after the pellet,
in a phase when the line averaged
density is very slowly decreasing,
while peak density is slightly
increasing, so that density profiles are
peaking. MHD activity is rather
quiescent as the sawtooth is
suppressed by the pellet. Plasma reheating
is helped by ECRH, electron
temperature and neutron yield
increase until a strong central m=1
mode starts. At that point, the neutron
yield increase is quenched, density
decreases and density peaking is
reduced. The global energy
confinement time, that has reached
transiently 1.5 times the value of the
ITER89-P scaling, goes back to the L-
mode value. The comparison between
the experimental neutron yield and the
estimation, produced by the solution
of the ion energy diffusion equation
using the Chang-Hinton ion thermal
diffusivity, shows that, before the m=1
mode, the neoclassical value is in
10 20 (m -3 )
10 6 (W)
10 -2 (s) 10 18 10 12 (keV)
3
1
2
0
2
1
3
1
6
2
6
2
a)
b)
c)
d)
e)
f)
I sx
P TOT
T e0
Φ DD
0.5 0.6 0.7
t (s)
Fig. 1.26 - Time traces for pulse #17839, pellet injected at t=0.55 s: a)
peak and line averaged density; b) Fig. total 1.26 and ECRH power; c) peak
electron and ion temperature; d) neutron yield: experimental (full),
neoclassical (dashed), 2 times neoclassical (dot-dashed); e) soft x
emission; f) global energy confinement time: experimental (full),
ITER89-P (dashed). temperature and ECRH power (on-axis heating)
ITER89P
n e0
P ECRH
T i0
τ E
27

1. Magnetic Confinement
n e (10 20 m -3 )
T e (keV)
0
4
3
2
1
0
1.5
1.0
0.5
OH
P ECRH
OH
0.1 0.2 0.3
r (m)
Fig. 1.27 - Pulse #17389, radial
Fig.
profiles:
1.27
a) electron density, at
t=0.595 s (OH-dashed), t=0.650 s (ECRH-full), shaded area marks
the power localizations; b) electron temperature at t=0.595s
(OH–dashed), t=0.610, 0.630, 0.650 s (full)
to 3.5×10 12 n/s, essentially because of the broader temperature profiles.
agreement with the experiment while,
during the mode, the ion diffusivity
increases by a factor 2. In fig. 1.27, the
electron density and temperature profiles
are shown, together with the radial
location of the ECRH power, before the
onset of the m=1 mode. The additional
power is deposited just below the cut-off
layer at 2.4×1020 m-3 and symptoms of
density pump-out in the deposition
region are observed. The evolution of the
temperature profiles shows the
characteristics of a diffusive behaviour,
similarly to what happens in the lowdensity
high temperature case of section
1.3.2. The current profile, as evaluated
from the diffusion equation for the
poloidal magnetic field, is very flat in the
core region, including the deposition
layer, resulting in a magnetic shear s ≤
0.5 for r/a ≤ 0.5.
The same scenario has been performed
also at a higher plasma current, viz. 0.76
MA, q a =3.7 with qualitatively similar
results as far as time behaviour of the
main quantities is concerned.
Quantitatively, the lower q a case only
shows a marginal confinement
improvement over the L–mode, and no
clear reduction of ion energy transport
can be deduced from the experiment. The
neutron yield increases by about 20% up
Some experiments have been performed in a similar scenario, displacing the resonance layer to
the plasma center. The on-axis heating induces a fast onset of the sawtooth activity, and a
stronger density pump-out has been observed. To reduce the density increase produced by the
pellet, the scenario where the pellet injection is performed on a plasma heated by off-axis ECRH
has also been foreseen. In that case, the higher electron temperature induces a more peripheral
deposition of the pellet particles, and the pellet injection is not able to suppress the persisting
sawtooth activity. In the latter plasma scenario, no improvement in the global energy
confinement nor in the ion energy transport has been observed.
1.3.4 LH and LH+ECRH results
Important achievements in the last FTU campaign have been obtained with LH power system,
both in the technological and the scientific area.
From the technological point of view:
a)
b)
• Full compatibility of the LH system with either the ECRH and the pellet injection systems has
been demonstrated.
• The total auxiliary coupled power has reached its record value of 1.95 MW in shot #17989,
28

1. Magnetic Confinement
for pulses longer than 50 ms, and the
value of 1.80 MW in shot #17983,
for pulses longer than 250 ms. The
LH coupling problems, possibly
arising from the edge density
increase induced by ECRH, have
been solved by positioning the LH
antenna inside the FTU vessel, so
that a slightly underdense plasma is
found in front of it during pre-ECRH
phases.
• Significant LH power has been
coupled in a very high-density postpellet
enhanced confinement phase
(900 kW in plasma with central
density larger than 3.6×10 20 m -3 ).
The strong edge perturbation
following the pellet injection only
prevents LH coupling from taking
place for 30 ms. Heating effects
have been observed at such high
density regimes.
From the scientific point of view:
1) Synergy between LH and ECH
waves
It has been demonstrated for the first
time in a tokamak plasma that ECH
power is efficiently absorbed by the
fast electron tails generated by the
LH waves. This occurs when no
interaction is possible between the
bulk electron distribution and the
ECH waves, because the magnetic
field is sufficiently high (=7.2 T) to
put the cold resonance well outside
the FTU vessel. The relativistic mass
increase in the fast electrons
balances that of the magnetic field.
Even though signs of this interaction
had previously been observed in
other tokamaks (as JFT–2M, TdeV,
Versator II), clear macroscopic
effects on the plasma state, such as
those observed in FTU, have never
been reported before. The enclosed
figs. 1.28 and 1.29 summarize the
FTU results. In fig 1.28, the time
evolution of the main plasma
quantities for shot #18181 is shown.
keV 10 5 W V 10 18 m -3
50
46
0.3
0.1
5
0
5
4
0.60 0.65 0.70 0.75 0.80 0.85
Time (s)
Fig. 1.28 - Time evolution of density, loop voltage and peak electron
Fig. 1.28
temperature during LH and LH+ECRH
keV
6
4
2
0
-0.3 -0.2 -0.1
0 -0.1
Fig. 1.29 - Thomson T e (r) profile Fig. at different 1.29 times: OH (red); LH 0.6
MW only (yellow); LH+ECRH 0.6+0.35 MW (green); 0.6+0.7 M (Blue)
29

1. Magnetic Confinement
Of a particular importance is the drop in the loop voltage, which indicates an increase in the CD
efficiency. Figure 1.29, instead, shows electron temperature radial profiles in the LH alone (600
kW), in the first (350 kW) and second (700 kW) ECH power step. The increase occurs within
r

1. Magnetic Confinement
be obtained with an ECRH power as
low as P ECRH ≈0.15 P OH , provided
that the mode is rotating. Figure 1.32
shows that stabilization with co-and
counter-ECCD is identical to the one
achieved with perpendicular launch,
provided that the absorption radius is
the same in all cases. Heating is
therefore the dominant stabilizing
term, CD effects being negligible in
this case. TM dynamics can be
modelled (fig. 1.33) by using a
Rutherford-type equation [1.16], in
which the change rate of the island
width is dependent on the balance
between de-stabilizing terms, related
to the current density and pressure
profiles, and stabilizing terms due to
ECRH and ECCD.
The beneficial effects of mode
suppression on energy confinement
are shown in fig. 1.34, where two
discharges are compared, which are
almost identical in the ohmic phase,
the only difference being that ECRH
absorption occurs in slightly
different positions. In shot #18004,
absorption is ≈3 cm internal to
the island O-point, while in shot
#18015 EC absorption is
correctly positioned inside the
island. In this case of a precise
alignment, stabilization occurs
on the expected resistive time
scale, and a much better thermal
and particle confinement is
observed.
Sawteeth are stabilized with
P ECRH >P OH (≈2.2 times) if
r dep >r inv , with a delay which is
dependent on r dep -r inv and
consistent with resistive diffusion
times (fig. 1.30 and 1.35). A
broadening of J(r), with a slow
decrease in central ohmic heating
Fig. 1.32 - T e at plasma centre
and at r≈a/2 (left), and Mirnov
oscillations (right) with co-ECCD
injection (top), counter injection
(bottom), and perpendicular
launch (centre)
keV keV keV
T/s
100
-100
50
-50
50
-50
20
-20
δR = -3 cm
δR = 0 cm
δR = 1 cm
δR = 2 cm
0.45 0.50 0.55 0.60 0.65 0.70
Time (s)
Fig. 1.31 - 4 Mirnov oscillations with different values of δR=R abs -
Fig. 1.31
R O–point . ECRH starts at 0.5 s. From top to bottom: δR=-3 cm, 0, +1
cm and +2 cm
4
3
2
4.0
3.5
3.0
2.5
4
3
2
0.8 0.9 1.0
Time (s)
10 2 T/s
10 2 T/s
10 2 T/s
1
0
-1
1
0
-1
1
0
-1
0.8 0.9 1.0
Time (s)
Fig. 1.32
31

1. Magnetic Confinement
w (m)
0.03
0.02
0.01
0
r s = 12.6 cm
t R = 165 ms
r s ∆' 0 = 0.12
rf ON
δ ECRH = 2 cm
δ ECCD = 0.95 cm
0.46 0.48 0.50 0.52 0.54
Time (s)
Fig. 1.33 - Island width evolution estimated by using a
Fig. 1.33
Rutherford type relation between growth rate and
stabilizing/destabilizing factors, including the term due to EC
induced helical currents (by heating and CD). Key parameters
are equal to the experimental values. Estimation is close to
experiment
10 3 J 10 18 m -3 keV keV 10 2 T/s
1
-1
3.0
2.5
2.0
1.5
40
38
4
0
r/a=0
r/a=0.5
0.48 0.50 0.52 0.54 0.56
Time (s)
Fig. 1.34 - Shot #18004 and #18015 are compared. In #18004
absorption is at r abs /a=0.46, 3 Fig. cm 1.34 and stabilization fails. Top
to bottom: Mirnov signals, T e,ECE at centre and half radius,
central line density, global energy increase. Core confinement
improves with TM stabilization
10 19 Flux 10 19 m -3 keV 10 5 W
10
6
2
3
1
4
2
0
1.0
0.5
0
0.5 1.0
1.5
Time (s)
Fig. 1.35 - Sawteeth are stabilized with a second gyrotron (r dep ≈a/2),
with a delay consistent with a resistive Fig. 1.35 diffusion time across r dep -r inv .
Particles and impurities tend to accumulate at centre, depressing core
temperatures. J(r) slowly broadens, with a further decrease in central
OH and T e
and T e , is observed after sawteeth
suppression, since reconnections no
longer provide fast stationarity. Particles
and impurities tend to accumulate,
further depressing core temperatures.
Stabilization does not enhance
confinement, and is therefore most used
for more accurate power balance
estimations and energy transport
analyses.
1.3.6 Energy transport and
electron temperature profile
stiffness with localized ECRH
Strongly localized ECRH is ideal to
bring into evidence inhomogeneities in
energy transport. As a matter of fact,
the local power balance during ECRH
is positive only where EC absorption
occurs (fig. 1.36), and most features of
thermal profiles are therefore
determined by heat transfer inside the
plasma column. A peculiar aspect of
energy confinement has been observed
32

1. Magnetic Confinement
in strongly localized off-axis
ECRH: the effective electron
thermal diffusivity χ e,eff ,
given by the ratio between
the local heat flux and the
temperature gradient, at the
switching on of the ECRH
power develops a sharp step
at the position of EC wave
absorption r dep . χ e,eff ,
strongly decreases within the
absorption radius, and
increases outside of it
(fig. 1.37). In other terms,
the temperature profile
appears to be rigid toward
strongly localized heating, in
the sense that it does not
bend at the heating point, as
expected for a medium with
a smoothly varying heat
conduction coefficient.
Profile rigidity is observed
both at low density regimes,
characterized by a negligible
e-i heat transfer, and at high
density (fig. 1.38), when
collisional ion heating is an
important element of the
power balance in the plasma
core. During off-axis ECRH
at high density, the residual
ohmic heating in the central
region may be insufficient to
provide the necessary ion
heating rate and radiation
losses. Since the electron
temperature remains rigid
and peaked also in these
conditions (fig. 1.39), some
heat transfer against
temperature gradients (heat
pinch) should be considered.
In order to characterize the
resiliency of the profiles to
modifications induced by
local heating, a
dimensionless “rigidity
parameter” ξ can be defined
as the ratio between the
relative change δ(ρ∇T) of
the temperature gradient step
-Φ heat /n∇T (m 2 /s)
0
10 6 P d,tot (W/m 3 )
2
1
0
-1
#18281
a = 0.3 m
P d,total ; ohmic
P d,total ; with ECRH
P d,ECRH (W/m 3 )
0 0.05 0.10 0.15 0.20 0.25
r (m)
Fig. 1.36 - Total electron heating power density in the ohmic (-) and
ECRH (+) phase. During ECRH
Fig.
the
1.36
local power balance is negative
everywhere except at EC wave absorption (continuous line)
1.0
0.5
#18281
a = 0.3 m
-Φ heat /n∇T; ohmic (.75÷.80 s)
-Φ heat /n∇T; ECRH (1.05÷1.10 s)
-Φ heat /n∇T; ECRH (3 ms from start)
P d,OH
P d,ECRH
-0.5 0
0 0.05 0.10 0.15 0.20 0.25
r (m)
Fig. 1.37 - Electron heat diffusivity at different times during ohmic heating (x)
Fig. 1.37
and ECRH (diamond). A step develops with ECRH where localized EC
absorption occurs
1.6
1.2
0.8
0.4
10 7 P d (W/m 3 )
33

1. Magnetic Confinement
χ e (m 2 /s)
2
1
0
-1 0
0
0.05 0.10 0.15 0.20
r (m)
Fig. 1.38 - The step in the effective Fig. heat 1.38 diffusivity at the EC absorption radius
is observed both at low (o) and high (+) density
q & nT norm
8
4
0
0
# 18290 - n e,line = .8 10 20 m -3 )
# 18015 - n e,line = .4 10 20 m -3 )
nT
q
P d,ECRH
P d,oh -P d,loss
P d,ECRH
P d,ECRH
= .8 10 20 m -3
= .4 10 20 m -3
0.1 0.2 0.3
r (m)
Fig. 1.39 - The figure shows q, Fig. nT 1.39 and heating profiles respectively at high
(continuous) and low (dashed) density. Rigidity is present in both cases, and
shapes are very similar. In the high-density case the power balance is negative
at the plasma core
2
1
0
2.0
1.5
1.0
0.5
10 7 P d (W/m 3 )
10 7 P ECRH (W/m 3 )
from inside to outside the
absorption layer, and the
relative change ∆P/P in the
power deposited in the same
volume layer (fig. 1.40). The
profile is rigid or neutral if
ξ1, a thermal
barrier may develop inside δr.
The results of several shots
with ECRH in different
conditions are summarized in
fig. 1.41. The temperature
profile appears to be really
rigid only in cases of far offaxis
deposition (ρ dep ≈0.5)
during ECRH at current flattop.
As absorption occurs at
the core (ρ dep

1. Magnetic Confinement
T e (keV)
2.5
1.5
0.5
0
0.1 0.2 0.3
r (m)
Fig. 1.40 - A local rigidity index Fig. 1.40
ξ at ρ dep is defined as the ratio
between the relative change in the δ(r∇T) jump across the absorption
layer δr and the corresponding relative change in the total power
deposited inside δr. For the ideally non-rigid case, ξ=1
∆δ (r∇T)/δ (r∇T)
30
20
10
0
T e
Φ in ÷(-r∇T) in
a/3
>a/3
O.H.
E.C.R.H.
P heat = ∫ P d,tot dV
Φ out ÷(-r∇T) out
P d,tot

1. Magnetic Confinement
transport, due to excitation of eigenmodes near marginal stability. In this case, it can only be
expected that local adjustments take place in the fast particle pressure profile, although
substantial losses may occur if a threshold in the mode amplitude is reached, yielding
stochasticity of particle orbits in phase space. Away from marginal stability, and/or in the
presence of resonant modes, transport processes are more complex and of a greater intensity, as
confirmed by experimental results in the presence of significant levels of additional power
creating fast ion populations. Due to their fundamental character, resonant modes tend to readjust
their mode structure as the fast ion source is non-linearly modified. This provides for an intuitive
reason why these modes are potentially the most dangerous ones as far as fast ion transports are
concerned, and why strictly non-perturbative investigation tools are required for a realistic
modeling of their dynamic behaviour.
All these aspects, and especially those related to the excitation of resonant modes, have been
extensively studied both by theoretical-analytic investigations of the fundamental physical
processes involved (cf. Section 1.4.2) and by self-consistent numerical simulations of the nonlinear
dynamics of collective modes, and of the associated fast ion transports (see Section 1.4.3).
The crucial importance of treating the fast ion dynamics in a strictly non-perturbative way makes
such numerical simulations very demanding on the computational side, and requires the
application of advanced numerical techniques in particle simulations to achieve significant
results (see Section 1.4.4). Meanwhile, the theoretical-analytic investigations of resonant modes
indicate that some aspects of the non-linear dynamics of collective modes, such as the radial
fragmentation of coherent eddies at saturation (see Sections 1.4.2 and 1.4.3), have a clear
analogy and possible overlaps with more general non-linear problems of dynamics, specifically
with the spontaneous excitation of zonal flows by drift-Alfvén turbulence (see Section 1.4.5).
This remark obviously opens a number of questions on the possible effect of fast ion–driven
collective modes on the transport processes of the thermal plasma, and on the analogies of fast
ion transports with those induced by electromagnetic turbulence in the thermal plasma.
Other important topics and research areas in connection with a burning plasma are those related
to the investigation of transport processes of the thermal plasma and the possibility to improve
such transports locally, e.g., by inducing and controlling a so called Internal Transport Barrier
(ITB). In this respect, the theoretical activities using the JETTO transport code to perform
interpretative analizes of FTU discharges have provided significant insights. In particular, the
transport analysis performed on the IBW experiment has shown that the effects observed, of a
simultaneous increase in plasma density and central electron temperature, could be explained in
terms of a reduced central electron thermal diffusivity (see Section 1.4.6). Such results find their
interpretation and explanation in the capability of the injected IBW to induce a poloidal sheared
flow in the plasma, which locally decorrelates some microturbulence and, eventually, limits the
transport processes which may be associated with it (see Section 1.4.7).
Finally, as a useful and routinely used investigation tool to analise the effects of wave
propagation and absorption in a burning plasma, a complex ray-tracing method has been further
refined and numerically implemented (cf. Section 1.4.8).
As a concluding remark, it is worth recalling that a relevant fraction of theoretical activities is
being pursued within established international collaborations with both National Laboratories
and Universities, while others are the result of cooperative efforts carried out with Italian
Universities.
1.4.2 Spontaneous excitation of zonal flows by Energetic Particle Modes
(EPM) (In collaboration with University of California at Irvine)
Recent results of a nonperturbative 3-D Hybrid MHD Gyrokinetic Code (HMGC) [1.18]
demonstrate evident radial fragmentation of the EPM [1.19] coherent eddies. This fragmentation
36

1. Magnetic Confinement
is being associated with a diffusive transport of fast ions, as it may be inferred from
modifications in the fast particle density, and is confirmed by analytical studies, which yield
explicit expressions for fast ion transports.
The radial fragmentation of EPM coherent eddies has a clear analogy and possible overlaps with
more general non-linear problems of dynamics, specifically with the spontaneous excitation of
zonal flows by drift-Alfvén turbulence [1.20]. Within this framework, it is demonstrated that
EPM may yield spontaneous excitation of zonal flows [1.21,1.22], since they are modulationally
unstable above a given amplitude threshold of the coherent eddies that they form above their
excitation threshold. The EPM Non-Linear (NL) evolution is dominated by fast ion nonlinearities,
which only enter the ballooning interchange term in the vorticity equation [1.21,1.22],
since they carry pressure but not inertia. Meanwhile, fast ion non-linearities play a role via NL
modifications of their non-adiabatic response, δH k ≈(δφ-ν || δA || /c) k , δH z , where k,k’ subscripts
refer to the high frequency EPMs and sidebands, generated via NL interaction with the low
frequency zonal field (subscript z). Thus, the NL fast ion response is formally equivalent to a
quasi-linear diffusion, consistently with numerical simulations. This represents a crucial
difference between fast ions and thermal particles non-linear responses, since the latter are
influenced by the decorrelation associated with the self-generated zonal flow δφ z [1.23]. The
basic reasons for these different dynamic responses are the finite orbit width of fast ions (as
compared to the inverse perpendicular wavelength) and the intrinsic resonant character of their
interaction with the EPMs, in contrast with the fluid response of thermal ions.
Considering an NL regime, dominated by fast particle dynamics such as that typical for unstable
EPMs, has two main advantages: the first and obvious one is to analyse a relatively simple
physical model; the second advantage is to explore a regime in which the zonal field δφ z never
directly enters the NL EPM equations. δφ z becomes a “passive scalar”, entirely determined by
thermal plasma non-linearities, once the NL EPM field evolution is consistently determined by
fast particles. This physical picture provides the theoretical framework which justifies the use of
a Hybrid MHD-Gyrokinetic model for the consistent simulation of NL EPM dynamics
[1.18,1.24]. The zonal field δφ z can be evaluated by a direct numerical solution of the low
frequency quasi-neutrality equation as a post-processor of the HMGC.
Detailed analytic solutions of the NL EPM equations may be obtained within the limit formally
corresponding to the most unstable linear mode, for which the energetic particle drive in the
MHD ideal region is dominated by a geodesic curvature [1.19,1.25]. With the present
assumptions and within the framework of the weak turbulence theory, the non-linear gyrokinetic
equation [1.26] may be solved for the linear and NL non-adiabatic fast ion responses, including
both turbulent broadening and real frequency shift of wave-particle resonances [1.21,1.22]. In
the analytic expressions derived, it is possible to demonstrate that turbulent broadening in the
propagators can be consistently neglected. Its formal presence in the NL wave particle
resonances, however, is what guarantees regularity of these contributions; therefore, it is
explicitly maintained.
The solution of the NL gyrokinetic equation for the zonal non-adiabatic particle response yields
the low frequency particle and energy density transport equation for fast ions, which emphasizes
the basic role played by resonant particles and formally resembling the quasi-linear diffusion
effect on the particle equilibrium distribution function. This analogy is simply formal since, here,
a single coherent EPM eddy non-linearly produces δH z via modulational instability. From the
low frequency particle and energy density transport equation for fast ions, it is easily
demonstrated that energetic particle transports are dominated by diffusive processes, after radial
fragmentation of the coherent EPM eddies sets in.
As it happens for the general case of drift-Alfvén turbulence, this work demonstrates that EPMs
are modulationally unstable. It also gives explicit expressions for the growth rate of the low
37

1. Magnetic Confinement
th
β E0
0.02
0.01
frequency zonal field perturbation, δφ z , which is associated to their radial fragmentation.
Depending on the relative ordering of the zonal field frequency as well as of the linear frequency
ω z shift of the mode ∆ z , it is demonstrated that the zonal flow growth rate scales either
proportionally to the mode amplitude A, as |ω z |∆ L |, [1.21]. Further
analyses are in progress to compare quantitatively our theoretical estimates of ω z with numerical
results.
1.4.3 Particle simulation studies of zonal flow excitation by NL EPM
dynamics
Non-linear properties of moderate-toroidal-number (n) shear-Alfvén modes in tokamaks have
been investigated by using the HMGC simulation, which solves the coupled set of MHD
equations for the electromagnetic fields and gyrocenter Vlasov equation for a population of
energetic ions. In particular, the non-linear dynamics of the EPM has been studied. The EPM
becomes destabilized above a certain threshold value of the energetic-ion pressure gradient, with
a linear growth rate fast increasing with an increasing energetic-particle pressure gradient. We
confirm previous findings that strong radial redistributions in the energetic particle source take
place whenever the EPM excitation threshold is exceeded, potentially yielding large particle
losses and, eventually, mode saturation. Such a threshold may occur at experimentally accessible
th
values of β E , e.g., as low as β E0=0.75%
(on–axis value) for n=8 EPM excitation by Maxwellian
energetic ions with ρ LE /a=0.01 (the ratio between the energetic ion Larmor radius ρ LE and the
minor radius of the torus a) and a pressure profile β E =β E0 exp(-ρ 2 /LpE), 2 with L pE /R 0 =0.075
and a/R 0 =0.1 (cf. fig. 1.42).
High-resolution simulations have shown that the coherent non-linear EPM eddies tend to be
radially fragmented by the onset of a non-linear axisymmetric perturbation of the mode structure.
After a first phase, in which the eigenmode structure forms itself (up to t=45R 0 /ν A , fig. 1.43a),
it clearly appears - from the modifications in the fast ion line density - that strong particle
redistributions take place from t=45R 0 /ν A (fig. 1.43a) up to t=75R 0 /ν A (fig. 1.43b), which are
consistent with the mode-particle pumping model (particle radial convection). However, while it
was shown that these non-linear dynamics dominate particle losses and mode saturation at low–n
above the EPM excitation threshold, fig. 1.43 indicates a new dynamic process, which becomes
important for non-linear EPM evolution already at moderate n. In Fig. 1.43b, evident radial
fragmentation of the EPM coherent eddies (k θ =k || =0, k r =0) is present, which is visible both in
the contour-plots and the radial variation of the various poloidal harmonics in which the
eigenmode is de-composed. This
fragmentation is associated to a
diffusive transport of fast ions, as
it may be inferred from
modifications in the fast particle
density profile up to t=144R 0 /ν A
(fig. 1.43c).
0 5 10 15 20
n
th
Fig. 1.42 - On-axis value of the critical β E , β E0 ù , for EPM excitation as n
th
changes. Values as low as β Fig. 1.42
E0 ù =0.75% are obtained for n=8 EPM excitation
with ρ LE /a=0.01 and a pressure profile, β E =β E0 exp(–r 2 /L 2 pE), with
L pE /R 0 =0.075 and a/R 0 =0.1
It is possible to demonstrate that,
in correspondence with the radial
fragmentation of the coherent
EPM eddies and EPM non-linear
saturation, a zonal flow is also
formed. However, the interaction
of EPM with zonal flows has little
effect on non-linear EPM
dynamics, which is essentially
regulated by resonant interactions
with fast ions. Numerical
38

1. Magnetic Confinement
10 4
10 5
10 6
10 7
10 8
10 9
energy
t = 45 r n(r) vs. r
*10 -3
20
15
10
5
t = 45
*10 -3 2
1
|ϕ m,n |(r)
t = 45
0 40 80 120 0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0
ϕ(x,y,0)
t = 45 ϕ(x,y,φ t ) t = t = 45.00 ϕ(x,y,φ t )
t = 45
dϕ-zonal/dr t = 45 |(dϕ-zonal/dr) kr | 2 t = 45 |(dϕ-zonal/dr) kr | 2 t = 45
*10 -3 10 0
10 0
20
10 -2
10 -2
10 -4
10
10 -4
10 -6
0
10 -6
10 -8
10 -10
-10
10 -8
10 -12
-20
10 -10
10 -14
10 -16
0 0.2 0.4 0.6 0.8 1.0 0 20 40 60 0 40 80 120
r
|γ E τ A | t = 45 |(γ E τ A ) kr | 2 kr
t = 45 |(γ E τ A ) kr | 2
t = 45
10 2 10 2
10 -1
10 0
10 0
10 -2
10 -2
10 -4
10 -2
10 -6
10 -4
10 -8
10 -3 10 -6
10 -10
10 -12
10 -8
10 -14
0 0.2 0.4 0.6
r
0.8 1.0 0 20 40
kr
60 0 40 80 120
Fig. 1.43 - a) Non-linear evolution of an n=8 EPM at t=45R 0 /v A . Twelve frames are visible. On the first
row, from the left to the right: the wave energy in each poloidal component, the line density rn E (r)of
energetic ions, and the radial mode structure. On the second row: the contour plot for the scalar
potential fluctuation in the laboratory frame and Fig. at, respectively, 1.43a the toroidal angles of a circulating and
of a magnetically trapped particle (white bullet). On the third row: zonal flow intensity (dϕ zonal /dr) vs
r, radial power spectra and time evolution of the strongest radial Fourier components. On the fourth
row: decorrelation rate γ E (multiplied by the Alfvén time) vs r, radial power spectra and time evolution
of the strongest radial Fourier components
39

1. Magnetic Confinement
10 4
10 5
10 6
10 7
10 8
10 9
energy
t = 75 r n(r) vs. r
*10 -3
20
15
10
5
t = 75
*10 -3
8
|ϕ m,n |(r)
t = 75
0 40 80 120 0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0
ϕ(x,y,0)
t = 75 ϕ(x,y,φ t ) t = t = 45.00 75 ϕ(x,y,φ t )
t = 75
dϕ-zonal/dr t = 75 |(dϕ-zonal/dr) kr | 2 t = 75 |(dϕ-zonal/dr) kr | 2 t = 75
*10 -3 10 0
20
10 -2
10 -2
10 -4
10
10 -4
10 -6
0
10 -6
10 -8
10 -10
-10
10 -8
10 -12
-20
10 -10
10 -14
10 -16
0 0.2 0.4 0.6 0.8 1.0 0 20 40 60 0 40 80 120
r
|γ E τ A | t = 75 |(γ E τ A ) kr | 2 kr
t = 75 |(γ E τ A ) kr | 2
t = 75
10 2 10 2
10 -1
10 0
10 0
10 -2
10 -2
10 -4
10 -2
10 -6
10 -4
10 -8
10 -3 10 -6
10 -10
10 -12
10 -8
10 -14
0 0.2 0.4 0.6
r
0.8 1.0 0 20 40
kr
60 0 40 80 120
6
4
10 0 2
Fig. 1.43b - As Fig. 1.43a but t=75R 0 /v A
Fig. 1.43b
40

1. Magnetic Confinement
10 4
10 5
10 6
10 7
10 8
10 9
energy
t = 144 r n(r) vs. r
*10 -3
20
15
10
5
t = 144
*10 -3
|ϕ m,n |(r)
t = 144
0 40 80 120 0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0
ϕ(x,y,0)
t = 45 ϕ(x,y,φ t ) t = t = 45.00 ϕ(x,y,φ t )
t = 45
dϕ-zonal/dr t = 144 |(dϕ-zonal/dr) kr | 2 t = 144 |(dϕ-zonal/dr) kr | 2 t = 144
*10 -3 10 0
20
10 -2
10 -2
10 -4
10
10 -4
10 -6
0
10 -6
10 -8
10 -10
-10
10 -8
10 -12
-20
10 -10
10 -14
10 -16
0 0.2 0.4 0.6 0.8 1.0 0 20 40 60 0 40 80 120
r
|γ E τ A | t = 144 |(γ E τ A ) kr | 2 kr
t = 144 |(γ E τ A ) kr | 2
t = 144
10 2 10 2
10 -1
10 0
10 0
10 -2
10 -2
10 -4
10 -2
10 -6
10 -4
10 -8
10 -3 10 -6
10 -10
10 -12
10 -8
10 -14
0 0.2 0.4 0.6
r
0.8 1.0 0 20 40
kr
60 0 40 80 120
3
2
10 0 1
Fig. 1.43c - As Fig. 1.43a but t=144R 0 /ν A
Fig. 1.43c
41

1. Magnetic Confinement
calculation of zonal flow intensity (dϕ zonal /dr, normalized to T E /e, mainly given by thermal
plasma non-linear response) and decorrelation rates (non-linear E×B shearing rate γ E ,
normalized to ν A /R 0 ) are also shown in figs. 1.43a-c, together with their radial power spectra.
1.4.4 Particle simulation applications to hierarchical distributed-shared
memory parallel systems: integration of High Performance Fortran (HPF) and
OpenMP (In collaboration with Seconda Università di Napoli)
Particle-In-Cell (PIC) simulation codes seem to be the most suited tool for the investigation of
turbulent plasma behaviours. Because of the large ratio between the equilibrium scale lengths
and the fluctuation ones (typically, several tens or more), a high spatial resolution is required in
such simulations (up to millions grid cells for three-dimensional PIC codes). Such a requirement,
along with the need to ensure an adequate description of the velocity-space dependence of the
particle distribution function, makes it necessary to use large numbers of simulation particles
(tens or hundreds millions). This makes the full exploitation of parallel computers unavoidable.
Hierarchical distributed-shared memory multiprocessor architectures (in which Shared Memory
Multiprocessor Systems (SMPs) are used as nodes of large-scale distributed memory
architectures) are now emerging as a flexible architectural model: in fact, it combines the two
paradigms of shared and distributed address space into one system, thus exploiting at best the
properties of the hierarchical parallelism present in most applications.
The PIC simulation consists in evolving the phase-space coordinates of a particle population in
certain fields, which are computed (in terms of particle contributions, such as, for instance,
pressure) only at the points of a discrete spatial grid, and then interpolated at each particle’s
(continuous) position. Two main strategies have been developed for workload decomposition in
distributed memory parallel environment, namely: the domain decomposition and the particle
decomposition. Standard domain decomposition techniques assign different portions of the
physical domain and the corresponding portions of the grid to different computational nodes,
together with the particles which reside on them. The distribution of all the arrays among the
computational nodes gives this method an intrinsic scalability of the maximum domain size that
can be simulated with the number of nodes. This makes the domain decomposition approach
very attractive, in principle. Two important problems with this technique, however, are
represented by the communication overhead and the need for dynamic load balancing, both
associated to particle migration from one portion of the domain to the other. While the former
problem may affect the parallelization efficiency, depending on the effective amount of particle
migration per time step, the latter can be by-passed, at the expense of a deep restructuring of the
original serial code and by adopting a message-passing approach. It is generally accepted,
however, that such an approach, based on manual partition of data, insertion of communication
library calls, handling of boundary cases, is very complicated, time-consuming and error-prone,
and affects the portability of the resulting programme. In order to avoid these features with
distributed architectures, it is worth resorting to the particle decomposition technique, which is
suited to be implemented, with relatively little effort, through the use of high-level programming
languages, such as the HPF. Particle decomposition consists of a static distribution of the particle
population among processors, while replicating the data relative to grid quantities. Since no
particle has to be transferred nor re-assigned from one computational node to the other, the
migration-associated communication and load-balancing problems are automatically overcome.
The implementation of such strategy with high-level languages is then, in principle, relatively
straightforward. On the opposite side, an overhead on memory occupancy, given by the
replication of data related to the domain, and a computation overhead related to the updating of
the fields (each node only manages the partial updating associated to its own portion of particle
population) hinders a good scalability of the maximum domain size with the number of
processors, and limits the efficiency of such a technique to cases in which both memory and
computational loads on each node are dominated by the particle-related ones.
42

1. Magnetic Confinement
When distributing the workload among the different processors of a shared memory node, the
alternative between particle and domain decomposition does not correspond to an alternative
between high and low-level languages. Indeed, even in the framework of a domain
decomposition approach, particle migration from one processor to the other does not require any
communication at all, and a high-level parallel programming language, such as OpenMP, can
still be used. The choice between the two alternatives can then be solved on the basis of different
considerations. Here, we present the results of our experience in the high-level language porting
of a specific PIC code, namely the HMGC developed in Frascati, which includes all the main
features of the PIC codes used for the investigation of magnetically confined plasmas in toroidal
devices. While fixing a particle decomposition approach (implemented in HPF) for the
distributed memory decomposition (among different nodes), both particle and domain
decomposition techniques (implemented in OpenMP) have been tested for the shared memory
(intra-node) decomposition. A trade-off between low memory requirement and high
parallelization efficiency emerges, when comparing the two approaches, which makes one
method preferable as compared to the other, depending on the practical constraints the user has
to face.
The first strategy used for intra-node workload decomposition is the particle decomposition,
which distributes the loop iterations over the particle among the different processors of the node.
Several particles, belonging to different processors, can contribute to the pressure on the same
grid point, thus rising a race condition exception. OpenMP allows such sections to be protected
by defining them as critical sections, but this practically consists of a serial execution of those
specific portions of the code. The introduction of an auxiliary (private) grid array, owned by each
processor and representing a partial and
local pressure to be summed over all the
processors at the end of the particle
loop, can overcome the memory
conflicts at the expenses of an increase
in the memory occupation of the code.
In fig. 1.44, the speed-up is shown
(namely, the ratio between the time of
execution of the serial code and that of
the parallel code) of the most
demanding section of the PIC code, viz.
the pressure-updating phase, vs the
number of processors n proc at fixed
number, n node =2, of (8-processors)
nodes for this version of the code (ν1),
which uses the particle decomposition
both in the inter-node and the intra-node
workload decomposition. Four different
values of the average number of
particles per cell (from N ppc =4 to
N ppc =256) have been considered, with
N cell =4096 cells. From fig. 1.44 it can
be observed that the speed-up values
depart from the linear scaling with
n proc only for n proc greater than a
certain value, which is higher the higher
is the average number of particles per
cell, N ppc .
The second version considered here
(ν2a) implements the coupling of the
s u
16
12
8
4
0
N ppc = 4
N ppc = 16
N ppc = 64
N ppc = 256
2 4 6 8
n proc
Fig. 1.44 - Speed-up of the pressure-updating phase vs the number
Fig. 1.44
of processors, at fixed number of (8-processors) nodes, n node =2, for
the particle decomposition version, ν1. Four different values of the
average number of particles per cell (from N ppc =4 to N ppc =256)
have been considered
ν1
43

1. Magnetic Confinement
s u
16
12
8
4
0
N ppc = 4
N ppc = 16
N ppc = 64
N ppc = 256
2 4 6 8
n proc
Fig. 1.45 - Speed-up vs the number Fig. 1.45 of processors, at fixed number
of (8-processors) nodes, n node =2, for the domain decomposition
version, ν2a
ν2a
inter-node particle decomposition and the
intra-node domain decomposition, which
do not require the introduction of the grid
auxiliary array introduced in version ν1,
but implies heavier restructuring of the
code and, possibly, addressing loadbalancing
problems. It consists of reordering
the particle population according
to the portion of domain in which each
particle resides, and assigning a different
portion to each processor. Such a
reordering gives rise, once again, to the
risk of race conditions (the particles
belonging to a certain domain portion
have to be counted within a particle loop,
and the updating of the counter is a
critical operation). Once assigned to the
processors, however, no further race
condition occurs in updating the pressure
array element, as loop iterations, which
could, in principle, concur to the updating
of the same element, are executed by the
same processor. Figure 1.45 shows the
scaling of the speed-up as compared to
n proc , at the fixed number (n node =2) of
8-processors nodes, obtained by this
version. The speed-up values, obtained
with the version ν2a, do not seem to be
very satisfactory. In fact, several operations, which are absent in the intra-node particle
decomposition version ν1, will have to be performed anyway: namely, the identification of the
domain portion into which each particle falls, a loop to balance the particles over processors, and
a re-ordering loop. However, for specific (but rather common) applications characterized by a
contained particle migration per time step from one portion of the domain to the other, a
significant efficiency improvement can be obtained by limiting the reordering phase (and then,
the critical computation) to those particles that have changed their domain portion in the last
step. Their number can be very low indeed, if it is possible to decompose the domain along a
slowly-varying coordinate. Figure 1.46 shows a comparison between the results obtained by the
version ν2a and a companion version (ν2b), which implements such a selective reordering. The
results of the particle decomposition implementation, ν1, are also shown for reference. The case
N ppc =256 is considered as an example.
It can be concluded that, at least for the specific application considered here, this mixed
“particle-domain decomposition” strategy represents an interesting compromise between the two
competing targets — namely, high speed-up and low memory requirements.
1.4.5 Non-linear zonal dynamics of drift and drift-Alfvén turbulences in
tokamak plasmas (In collaboration with University of California at Irvine
and Princeton University Plasma Physics Laboratory)
In recent years, increasing attention has been devoted to exploring non-linear dynamics of zonal
flow [1.27] associated with electrostatic drift-type turbulence [1.28-1.30]. On the other hand,
though it is well known how electrostatic drift modes couple to the electromagnetic shear Alfvén
wave as the plasma β=8πP/B2 increases [1.31-1.33], little effort has been devoted so far to
44

1. Magnetic Confinement
investigating non-linear zonal dynamics
of drift-Alfvén turbulence.
The present work focuses on the
identification of the main non-linear
physics processes, which may regulate
drift and drift-Alfvén turbulence by using
a weak turbulence approach. Within this
framework, based upon the non-linear
gyrokinetic equation [1.34] for both
electrons and ions, an analytic theory is
presented here for non-linear zonal
dynamics described in terms of two
axisymmetric potentials, δφ z and δA ||z ,
which spatially only depend on a
(magnetic) flux coordinate. Physically,
δφ z is associated with zonal flow
formation, while δA ||z , corresponds to
zonal currents δj ||z =-(c/4π)∇⊥δA 2 ||z . The
introduction of a zonal vector potential,
δA ||z , is one of the typical differences of
the electromagnetic as compared to the
electrostatic case.
Zonal potentials are characterized by time
variations on typical scales, which are
long as compared to the characteristic
ones of the drift-Alfvén instabilities. This
specific ordering of time scales - which
formally requires such a proximity to the
marginal stability, that the linear growth
s u
rate becomes smaller than the mode frequency - is exploited for explicitly manipulating formal
expressions in the theoretical analysis. In contrast to other approaches, however, which also
assume slow radial variations of the zonal fields (k z
-1 ) as compared to the typical spatial scale of
the background turbulence(k r
-1 ), (kz ≈k r is generally taken, although |∂k z /k z 2 |>>1 is still assumed
for consistency of our eikonal approach). In this respect, the present work is the generalization
of ref. [1.23], which has demonstrated that zonal flows can be spontaneously excited by
electrostatic drift turbulence and that these are characterized by k z ≈k r . The present work shows
that zonal flows in toroidal equilibria can be spontaneously excited via modulations of the radial
structure (envelope) of a single-n coherent drift-wave, with n as the toroidal mode number. In
this framework, the turbulent state and the non-linear couplings among different n’s will show
up via zonal dynamics only. Similarly to ref. [1.23], the present theory is strictly applicable to
toroidal plasma equilibria, where poloidal asymmetry forces each mode to be (at least, within the
linear limit) the superposition of many poloidal harmonics m, characterized by the same n. In
this respect, the present theoretical analysis is a systematic treatment of the radial mode structure
(envelope) of zonal fields and drift turbulence in the general electromagnetic case, including
slow time evolutions and accounting for linear (toroidal) and non-linear mode couplings on the
same footing. More specifically, it is demonstrated that zonal flows (δφ z ) are due to charge
separation effects, associated with both finite ion Larmor radius and finite ion orbit width effects
(magnetic curvature), whereas zonal currents (δA ||z ) or turbulent dynamo are due to parallel
electron pressure imbalance (see also ref. [1.35]).
Spontaneous excitation of zonal flows by electrostatic drift micro-instabilities is demonstrated
both analytically and by direct 3-D gyrokinetic simulations [1.34]. Direct comparisons indicate
16
12
8
4
0
ν1
ν2a
ν2b
2 4 6 8
n proc
Fig. 1.46 - Comparison between the speed-up obtained, at different
number of processors and fixed Fig. number 1.46 of 8-processors nodes,
n node =2, by the domain decomposition version, ν2a, and a
companion selective reordering version, ν2b. The results of the
particle decomposition implementation, ν1, are also shown for
reference. The case N ppc =256 is considered
45

1. Magnetic Confinement
good agreement between analytic expressions of the zonal flow growth rate and numerical
simulation results for Ion Temperature Gradient (ITG) modes. Analogously, it is shown that
zonal flows may be spontaneously excited by drift-Alfvén turbulence, in the form of
modulational instability of the radial envelope of the mode as well. From the analytic expression
for the growth rate of the spontaneously excited zonal flows (δφ z ) it is also shown how no-flow
generation is expected for a pure shear Alfvén wave, due to the peculiar nature of the Alfvénic
state. Meanwhile, it is also demonstrated that, generally speaking, zonal currents are also excited,
but they have a negligible effect on the turbulence itself, at least when sufficiently long
wavelengths are considered (longer than the electron collisionless skin depth). The general
results obtained within this theoretical model are also applied to Alfvénic oscillations; such as
the Kinetic Alfvén Waves (KAW), and the more recently discussed Alfvén Ion Temperature
Gradient (AITG) [1.31] driven mode.
1.4.6 Transport analysis and modeling of FTU plasmas with the JETTO
transport code
The JETTO transport code turned out to be an important tool in the interpretation and analysis
of FTU plasma discharges. It constitutes a very useful aid in predictive studies of future plasma
scenarios, both in FTU and other devices proposed such as IGNITOR, ITER, FTU-D
As a predictive tool, the JETTO code mainly uses an empirical transport model, formed by a
combination of a Bohm and a gyro-Bohm term; moreover, an appropriate dependence on the
magnetic shear s was introduced in the model in order to take into account the possible
suppression of long-scale, toroidally coupled “global” turbulence, with the formation of internal
thermal barrier for null or negative values of s.
Extensive use of JETTO was made in the interpretative analysis of FTU discharges. In particular,
the transport analysis performed, among others, on the IBW experiment is worth being
mentioned. Transport analysis was applied to IBW plasma discharges for testing whether the
observed effects of a simultaneous increase in plasma density and central electron temperature
could be explained in terms of a reduced central electron thermal diffusivity, or peaking of
plasma current density profile. The JETTO code was used to evolve ion temperature and current
density profiles, assuming both neoclassical ion transport and neoclassical electrical resistivity.
Electronic kinetic profiles, equilibrium magnetic flux surfaces, and radiation profiles were taken
from the experiment, and the IBW deposition profile expected from linear theory was used. The
experimental loop voltage behaviour was well reproduced, consistently with the measured
average Z eff .
The transport analysis shows that, during the IBW injection, the effective electron thermal
diffusivity χ e is actually reduced by about a factor of two as compared to the ohmic phase,
which had never been observed in previous IBW experiments before. Moreover, the analysis
does not predict any significant radial redistribution of the ohmic power during the IBW phase
and indicates that the global energy confinement time was not strongly affected, despite the
power added.
1.4.7 Poloidal rotations induced in tokamak plasmas by IBW
A poloidal ion sheared flow can be produced by coupling IBW to a tokamak plasma near an ion
cyclotron resonant layer, which improves the plasma confinement [1.36]. Experimental evidence
of this improved plasma confinement regime was observed in the IBW experiments on the
Princeton Beta Experiment-Modification (PBX-M) [1.37-1.40]. The theoretical interpretation of
the IBW-induced poloidal rotations has been developed in the framework of a fluid model
[1.36,1.41], which relies on the solution of the momentum balance equation, where losses are
represented by the neoclassic viscosity. Severe critics to this model, which implies plasma
46

1. Magnetic Confinement
incompressibility, have been addressed in ref. [1.42], where a fluid compressible plasma model
has been assumed, and the results were compared with the previous ones. In the same article, a
kinetic approach based on the second order solution of the Vlasov equation has been proposed
to calculate the divergence of the second order pressure tensor exactly. In general, the
fluid–based models do not clearly explain, in our opinion, how an IBW can produce a poloidal
flow, if the poloidal momentum of the wave is zero. The local balance of the poloidal
momentum, indeed, implies a poloidal momentum transfer to the plasma, as a result of the IBW
interaction with the resonant ions. This is reasonably done by the Lorentz force, because of the
toroidal magnetic field, Β φ , acting on the resonant ions when they take their radial momentum
from the wave. Following the fluid model, such a mechanism is embedded in the convective term
of the poloidal momentum balance (here the brackets denote averaging over magnetic
surfaces and over time intervals which are long as compared to the wave period, V is the fluid
velocity of the resonant ions, and V 0 indicates the velocity component in the poloidal direction).
In this work, an attempt has been made to explain the IBW-induced poloidal rotations in terms
of single particle dynamics of resonant ions. As a main purpose this could give an insight into
the mechanism of the poloidal flow driven by RF, and a guideline for a kinetic approach, which
would improve the fluid model of this process. Here, the single particle dynamics in an RF
electromagnetic field is outlined in the framework of a perturbation theory. The IBW electric
field is considered to be a perturbation of the ion helical motion in a constant magnetic field. As
a smallness parameter of the perturbed orbit calculation, the ratio between the amplitude of
RF–induced oscillations and the Larmor radius is used, multiplied by the square of the ratio
α=ω/Ω, viz. the ratio of the pump frequency to the ion-cyclotron frequency of the resonant ions.
At the second order in the perturbation, the ion dynamics involves resonant absorption when α
is half-integer, and ion poloidal flow when α is integer. This suggests that the underlying
physical mechanism of both effects is the result of a self-interaction of IBWs. In the first case,
which was widely investigated [1.43-1.45], quasi-modes at a frequency of 2ω, namely twice the
beating waves, are produced and damped on the ion-cyclotron resonance nΩ=2ω; in the second
case, the beating waves produce a d.c. radial electric field E r , which induces a poloidal drift
E r ×B φ . Here, we evaluate the mean poloidal flow induced by IBW in a Maxwellian population
of resonant ions within a plasma slab model, limiting the wave-particle interaction interval to a
slowing-down characteristic time, which depends on the relevant collisional regime. This allows
us to provide a straightforward analytic expression of the radial profile of the poloidal flow
induced. The analytical calculations were also compared with a numerical evaluation of the
poloidal flow induced by IBW in the FTU plasma (433 MHz, 7.9 T, hydrogen plasma). In fact,
a numerical code, based on the fluid model [1.36], has been implemented to evaluate the ion flow
induced by application of IBWs. The power spectrum of the waves launched from the antenna
to the absorption layer has been calculated in the framework of the ray-tracing theory, in a 3-D
toroidal geometry, taking into account the full electromagnetic dielectric tensor. Moreover, both
analytical and fluid calculations are compared with the poloidal velocity profile measured in the
recent IBW experiment on the Tokamak Fusion Test Reactor (TFTR) (76 MHz, 3.3 T) [1.46].
1.4.8 Complex ray-tracing method in high harmonic fast wave propagation
and absorption
Injection of fast waves at a moderately high harmonic number can be used to induce CD in
Tokamaks with low aspect ratio and very low external magnetic field, in order to overcome the
accessibility condition of a slow wave, such as the lower hybrid wave. Nevertheless, the use of the
fast wave is affected by the possibility that the ion cyclotron harmonic resonance, distributed along
the path of wave propagation, can absorb the RF power before the wave may reach the plasma core
and deposit its energy on the electrons. To model such an absorption mechanism, as well as the wave
propagation, the integration of the ray-tracing equation system, which comes from the Wenzel,
Kramer, Brillouin code (WKB) approximation, has been numerically performed when considering
a complex Hamiltonian in strongly nonhermitian media (complex ray-tracing method) [1.47-1.55].
47

1. Magnetic Confinement
The classical treatment of the geometric optics in non-homogeneous and dispersive media
explicitly assumes that the dielectric tensor is Hermitian or nearly-Hermitian, which means that
there is a restriction of space-time coordinates along a ray to be real values. The new formulation
removes the above constraints and generalizes the geometric optics to media with a strongly non-
Hermitian dielectric tensor, the consequence of which is that complex-value coordinates are
introduced.
The results of the ray-tracing integration for the complex phase S allow for the reconstruction of
the wave trajectory only in those zones where the wave is crossing the real space. In general, in
strongly non-Hermitian media, a complex space must be considered. The concept of “complex
space”, in the frame of the complex ray-tracing theory, is well illustrated by this sentence of
Budden&Terry: “When a ray in free space impinges obliquely on the boundary of a lossy
medium, whose refractive index is complex, it is refracted at a complex angle, and the
coordinates of points on it must take complex values. It follows that, when a ray travels from one
real point to another through media, some of which are lossy, the ray path can strictly be traced
only by using complex values of some of the space coordinates”.
In a complex space, in fact, there is an infinite number of possible ray trajectories. Therefore, to
trace a ray trajectory does not make sense, as it is illustrated by another sentence of
Budden&Terry: “.... The functional relation between z and x in the ray trajectory is not, however,
restricted to real values. It shows x as a complex function of the complex variable z, and is
represented by a Riemann surface, which would require four ordinary Euclidean dimensions.
The actual ray, here, is simply the line where this Riemann surface intersects the real space
defined by Im(x)=Im(z)=0. But, in proceeding from the origin to the end point (D,0,0), it is not
necessary to remain in real space. Any other line in this Riemann surface would satisfy the
equations (**) to (**) (ray trajectories) of the ray....”.
Beyond the possibility to trace the ray trajectory, the knowledge of the complex phase S, coming
from the solution of the ray-tracing equations, allows us to know the power damping rate for
each component of the launched spectrum. This is a crucial information in order to know the
amount of first-pass absorption of the wave power on the ions species near the harmonic resonant
layers, and on the warm electrons far from the resonance.
Near the ion harmonic resonance, the fast wave may couple with the IBW, which is a hot
electrostatic mode, characterized by a very low wavelength (very high refractive index), so that
the condition k ⊥ ρ i >>1 holds. On the IBW branch of propagation, the absorption of the RF power
by hot ions is sudden, and, if all the wave power is transmitted to the IBW branch, the wave will
be stopped on this radial location. The coupling of the fast wave to the IBW can be studied by
means of the complex ray-tracing technique.
This technique has been applied to study the wave propagation and absorption in the case of NSTX
High Harmonics Fast Wave (HHFW) experiment [1.56]. The preliminary experimental results of
NSTX (B 0 =2.5 T, A=1, f rf =30MHz) are compared to the results of the present model and discussed.
1.5 NEW PROPOSALS
1.5.1 FTU-D
The FTU-D proposal refers to a modifications of poloidal magnetic field circuit, power supplies
and vacuum wall of the present FTU device, to allow elongated plasma configurations with a
magnetic separatrix to take place. The main scientific objective of FTU-D is the study of
advanced tokamak scenarios at high beta, high aspect ratio (up to 6.3), high magnetic field and
high density, with a substantial bootstrap current fraction. This proposal has been investigated in
detail for the last three years [1.1] and, in 2000, the ENEA-Euratom Steering Committee was
48

1. Magnetic Confinement
asked by the Frascati laboratory for permission to submit FTU-D for phase I application for
preferential support by Euratom. The usual procedure for preferential support was initiated. The
Fusion Programme Committee (FPC) established an Ad Hoc Group (AHG) of European fusion
experts to examine the application and provide a report. The AHG, chaired by Dr. C. Schuller,
met the proponents in Frascati at the end of September, in a two-day session. After the
presentation of the main characteristics and objectives of the FTU-D project and an in-depth
discussion, the AHG issued their recommendations.
In their report, the AHG recognised the scientific validity of the proposal, stressing the role of
FTU-D in extending the H-mode and advanced tokamak physics to high aspect ratio and high
density. It also considered as very relevant to the European Fusion Programme the high bootstrap
current fractions expected at high aspect ratio and the divertor physics at high density, albeit in
an open divertor configuration.
The AHG also expressed some concern on the possibility of achieving a good H-mode, in view
of the additional power indicated, at the highest magnetic field proposed (5T), due to
uncertainties on LH coupling, which could be alleviated by local gas injection. It also suggested
to examine more thoroughly the role of IBW system to control poloidal flows, and thus facilitate
the access to transport barrier scenarios. It also recommended that the possibility of adding Ion
Cyclotron Resonance Heating (ICRH) (at least in a second phase) should be examined in order
to increase the ratio T i /T e , which could be important for stability and for achieving high values
of β N . It encouraged the proponents to examine the possibility of varying the aspect ratio in
elongated configurations, to assess the dependence of transport, stability and bootstrap current
fraction on this parameter. It was concerned that the choice of inserting the proposed top and
bottom toroidal limiters to act as target plates should not be related to maintaining the
compatibility with full bore (circular plasma) operation. It also encouraged the proponents to
examine scenarios at higher magnetic fields, following the positive results of FTU on the
synergetic effects of LH and ECR downshifted heating.
The AHG also recommended that some technical issues should be examined in a forthcoming
phase II, namely:
• The demonstration of sufficient flexibility in shape and position control to cope with the
envisaged scenarios, including fields significantly above 5 T.
• To examine the capability of controlling the x-point during plasma evolution.
• To demonstrate that the target plates, as envisaged, can sustain the heat loads, particularly in
view of the gaps that are needed for diagnostic accesses.
• To implement the fast sweeping of the ECRH deposition radius in order to stabilise, for
instance, neo-classical tearing modes and achieve larger beta limits, and to examine the
possibility of feedback control of LH N || spectrum.
• To examine the possibility of increasing the ECRH power and pulse length.
With these comments, in their report to FPC, the AHG recommended FTU-D for the project
priority status phase I.
The FPC discussed in November the report by the AHG. It recognised that the FTU-D project
would allow to access a parameter regime at high aspect ratios, high magnetic field and high
absolute density which hitherto was inaccessible. The exploration of this regime would be
significant to the purpose of extending the data base on advanced scenarios and in any case
would be of high value to the fusion community. However, the FPC did not recommend the
attribution of priority status phase I at that moment, recognising that many of the features of
FTU-D are strongly interlinked with phase II issues. Therefore, the FPC expected that the
necessary information and answers to the points raised by the FPC and AHG would be provided
by the proponents in a combined phase I and II application.
49

1. Magnetic Confinement
The proponents examined the comments and recommendations by both the AHG and FPC. There
is no complete agreement as to the relevance of some of the issues raised. Nevertheless, in order
to comply with all the requirements made, a first conclusion has been reached that more
resources are needed and the cost of the project will be somewhat raised, as compared to the
original proposal.
1.5.2 PROTO-SPHERA (Spherical Plasma for HElicity Relaxation
Assessment)
Chandrasekar-Kendall-Furth (CKF) configurations for magnetic confinement
A simply connected magnetic confinement scheme can be obtained by superposing two
axisymmetric homogeneous force-free fields, each with ∇ → ∧B → =µB → , both having the same
relaxation parameter value µ=µ → 0 j /B → . The first is the Chandrasekar-Kendall force-free field of
CK
order-1: ψ µ,1 . [1.57]. The second is the Furth square-toroid force-free field: ψ F µ,λ [1.58]. The
flux functions obtained by superposing the two force-free fields are written as:
CK
ψ(r,ϑ)=ψ F µ,1 +γψµ,λ . For values of the superposition constant γ≥0.402, they contain - in a simply
connected region near the origin - a toroidal current density j φ of the same sign, and are called
CKF force-free fields. The CKF fields contain a magnetic separatrix and are composed by a
“main spherical torus”, two “secondary tori” on top and bottom, and a “pinch” discharge
surrounding the three tori (see fig. 1.47). If the pinch discharge can be sustained by driving
current on its closed flux surfaces, magnetic reconnections will occur at the x-points of the
configuration, thus injecting magnetic helicity, poloidal flux and plasma current into the main
spherical torus. Also the secondary tori will be a by-product of the same magnetic reconnections.
The ideal MHD stability of the CKF force-free fields has been studied by solving the eigenvalue
problem:W → •ξ → =ω2K → •ξ → , where W → is the plasma–perturbed potential energy and K → the plasmaperturbed
kinetic energy, associated with the perturbed plasma displacement ξ → [1.59]. The
expressions for the perturbed energies become simpler if the equilibrium is analysed in nonorthogonal
periodical Boozer coordinates (ψ T -radial, θ-poloidal, φ-toroidal), with Jacobian
√g∝1/B2 [1.60]. The continuity of the rotational transform ι/(ψ T ) and of the toroidal and poloidal
plasma currents, I(ψ T ) and f(ψ T ), smoothly joins the Boozer coordinates of
these equilibria at the ST–SP interface, ψ T =ψ T
x . The energy principle for a
compressible plasma is used by Fourier-analyzing the normal ξ ψ =ξ → •∇ → ψ T ,
binormal η=ξ → •(∇ → θ−ι/∇ → φ) and parallel µ=-√g•ξ → •∇ → θ components of the
displacement away from the (up/down symmetric) equilibrium as:
ξ ψ =Σ l ξ l (ψ T )sin(m l θ-nφ), η=Σ l η l (ψ T )cos(m l θ–nφ), µ=Σ l µ l (ψ T )cos(m l θ–nφ),
in terms of a toroidal number n and of a range of poloidal numbers m l . Only
the normal displacement ξ ψ must be continuous at the separatrix between the
three tori and the surrounding pinch, where the binormal η and the tangential
µ components can make jumps [1.61]. The STABLE code uses a onedimension
Finite Hybrid Element method (in terms of ξ ψ , η, µ) and solves the
ideal MHD eigenvalue problem. The results of the STABLE code calculations
for low toroidal mode numbers (n=1,2,3), assuming fixed boundary conditions
at the edge of the plasma: ξ ψ (ψ T
x )=0, are that the CKF force-free fields are
stable in ideal MHD, when the value of the superposition parameter is greater
than γ=0.5.
Fig. 1.47 - CKF relaxed state
Although force-free fields cannot sustain anypressure gradient (∇ → p∧=0) and
are therefore unable to confine plasmas of fusion interest, a variety of unrelaxed
(∇ → µ≠0, (∇ → p≠0) MHD fixed boundary equilibria, similar in shape and topology
to the CKF force-free fields, can be calculated (see fig. 1.48). They have µ=µ 0
j→ /B
→ =constant only at the edge of the plasma ( ψ T =ψ T
max ), as a boundary
condition for the MHD equilibrium. The flux-surface averaged relaxation
50

1. Magnetic Confinement
parameter (ψ T )=µ 0 will decrease from the edge of the
plasma to the axis of the main spherical torus: if the pinch
discharge can be sustained by driving current on its closed flux
surfaces, magnetic helicity (flowing down the gradient) will
be injected into the main spherical torus, through magnetic
reconnections at the x-points. The gradient of the pressure profile
will presumably be concentrated in the same region where the
gradient of undergoes its largest variation. Unrelaxed CKF
equilibria can be stable, with this kind of (ψ T ) and p(ψ T )
profiles, with fixed boundary conditions at the edge of the plasma:
ξ ψ (ψ T
max )=0, to all low-n ideal MHD modes, up to unity value of
the beta of the main spherical torus: β ST =2µ 0 ST / ST =1.
In a reactor extrapolation, the unrelaxed CKF configurations are
contained in an almost cylindrical solenoid-external magnetic
field, and their high β opens the possibility that their internal
magnetic field can be sustained by plasma motions. Unrelaxed
CKF fusion reactors with the right helicity injection, β limit and
energy confinement, will allow for an unimpeded outflow of the
high energy-charged fusion products, easing direct energy Fig. 1.48 - Unrelaxed CKF
conversion and the use of the burner as a
space thruster. However, at present, there is
no clear idea yet about the methods for
injecting currents or torque into the
surrounding plasma, so in a preliminary
experiment the surrounding plasma will be
partially replaced by a force-free Screw
Pinch (SP), fed by electrodes. The PROTO-
SPHERA experiment proposed at
CR–ENEA Frascati, will be devoted to
demonstrating the feasibility of a Spherical
Torus (ST), (with toroidal current I p ), where
a Hydrogen force-free SP (with longitudinal
current I e fed by electrodes) replaces in part
the surrounding discharge (see fig. 1.49). The
goal of the experiment will be to compress
the ST to the lowest possible aspect ratio in a
time of about 1600 Alfvén times
(1600×τ A =1600×0.5 µs=800 µs) and to
show that efficient helicity injection can
maintain a stable configuration for at least
one resistive time (τ R =50 ms). PROTO-
SPHERA, with a Pinch current I e =60 kA,
will produce an ST of diameter 2R sph =75
cm, aspect ratio A=1.2-1.3 and I p =120-240
kA, corresponding to an edge safety factor
q 95 ≈2.5-3.
Fig. 1.49 - PROTO-SPHERA configuration
Electrodes for the PROTO-SPHERA experiment
In the PROTO-SPHERA experiment, the plasma will be magnetically shaped as a disk near each
modular annular electrode (see fig. 1.50).
When the design of the PROTO-SPHERA experiment began, the electrodes were considered the
51

1. Magnetic Confinement
Anode
module
Directly heated
cathode ring
Gas exhaust
52
Fig. 1.50
Gas flux
Water cooled
anode ring
Cathode
module
Fig. 1.50 - Electrodes for PROTO-SPHERA
a)
H 2 gas flow
Insulating nuts
Stainless steel
anodic seal plate
Pump
Pinch arc
Pump
reduction
Gate
Anode voltage
connection
Ground connection
10 cm
Anode
6 cm
Insulating plate
I e
Poloidal coil
Pyrex vessel
8 TF current
returns
Ground SS
cathodic
seal plate
V ~
Directly heated
W cathode
Fig. 1.51 - a) Scheme; b) picture of PROTO-PINCH superposed with picture of an
I e =600 A plasma arc
Fig. 1.51
most unconventional item and a major concern. It was not
clear that a feasible solution did exist, allowing for almost
steady-state (≈1 s) emission from a cathode, at a plasma
current density level of 1 MA/m2 at the plasma-cathode
interface. Neither was it clear that a working solution did
exist for an anode able to withstand 50 MW/m2 for the
same discharge duration. Other concerns arose about the
endurance to many hundred discharges, the plasma
contamination and the magnetic field perturbation, due to
the directly heated cathode. A final concern was
represented by the breakdown voltage, which could cause
insulation problems in the PROTO-SPHERA load
assembly.
In order to investigate these points, the PROTO-PINCH
benchmark of one anode and one cathode module has been
built. It is similar to PROTO-SPHERA in physical
dimensions and strength of the magnetic fields near the
electrodes (see fig. 1.51). PROTO-PINCH has produced,
within a Pyrex vacuum vessel, hydrogen and helium arcs
in the form of screw pinch discharges, stabilized by two
PF coils located outside the vacuum. Following a trial and
error procedure, about 3 anode and 10 cathode prototypes
will have been tested on PROTO-PINCH from October
1998 to March 2001.
The technical solutions (see fig. 1.50) are a tungsten-copper
hollow anode and an AC
directly heated cathode,
composed by conical
helical 2 mm filaments
double wound (zero field)
in pure tungsten. Pure
tungsten has been used
instead of W-Th since, as
soon as the arc breaks
down, the temperature
raises to about 2700 °C,
exceeding by far the
Thoria melting point. The
cathode module is
AC–heated by a total
current I cath =590 A (rms).
Arc discharges have been
obtained with B=0.1 Tesla,
I e =660 A and V e =80-100
V and sustained for 2-5 s,
limited by the heating of
PF coils, pyrex vessel and
rubber vacuum seals. The
Cu-W water cooled hollow
anode, with H 2 puffed
through it, has withstood
thousands of discharges.

1. Magnetic Confinement
The final result is that a cathode and an anode module, able to
withstand the required current and power densities, have been built
and have survived many hundred plasma shots. A further remarkable
result is that the hydrogen plasma produced has turned out to be
almost free of impurities. The final relevant result is that the
breakdown is obtained at low voltage (100 V) in the same filling
pressure range as a standard tokamak discharge (10-2-10-3 mbar).
max
ψ T = ψ T
shell
S P
X
ψ T = ψ T
I e
Ideal MHD stability of PROTO-SPHERA
The TS-3 experiment carried out at Tokyo University [1.62] has
produced and sustained a flux-core spheromak similar to PROTO-
SPHERA for tens of Alfvén times, albeit its plasma was fed by simple
axial cylindrical electrodes. This result has shown that the combined
ST+SP (spherical torus+screw pinch) configuration can be stable in
ideal MHD.
The computation of the ideal MHD stability of PROTO-SPHERA
Z
R
raises the following new problems:
Fig. 1.52 - PROTO-
• the combined configuration composed by the ST, with closed field
Fig. 1.52
SPHERA cross section
lines, and by the SP, with open field lines ending up on electrodes,
must be correctly modeled;
• a magnetic separatrix defines the interface between the ST and the SP (see fig. 1.52).
The continuity of the rotational transform ι/(ψ T ) of the toroidal and poloidal plasma currents
I(ψ T ) and f(ψ T ) smoothly joins the Boozer coordinates (ψ T ,θ,φ) [1.60] of these equilibria at the
ST-SP interface, ψ T =ψ Τ
x . Inside the force-free SP (ψΤ
x

1. Magnetic Confinement
a) I p = 120 kA
b)
I e = 60 kA
q = 1
β = 20% Stable
β = 30%
Unstable
1.6 JET COLLABORATION
During the year 2000, the exploitation of the
JET facilities has started within the new
EFDA organization. ENEA has had the
responsibility of the Task Force H on
Heating and CD, and the Task Force S2 on
Advanced Tokamak Scenarii. The ENEA
participation has been large during the
second part of the year, in particular during
FTU shutdown.
1.6.1 High-beta plasmas in JET
discharges with optimised shear
Fig. 1.53 - Arrow plot results of the PROTO-SPHERA ideal
MHD stability calculation: I e =60, I p =120 kA; a) Oscillatory
stable motion localized on Fig. resonant 1.53 surfaces at b=20%; b)
I p (MA)
Dα
(a.u.)
Power
(MW)
T e
(keV)
2.6
1.8
2
0
10
0
2
0
8
4
2
0
Argon
S N (x10 16 neutrons/s)
H 89
β N
Pulse No: 46695 B T = 2.6T
ICRH
NBI
Internal transport barriers are produced in
JET with pre-heating and main heating
during the current ramp-up phase of the
discharge; in this way, so-called optimised
shear scenarios are obtained, with flat or
hollow core q-profile The main heating
waveforms, viz. the NBI and ICRH, start as
soon as the q=2 surface is within the
plasma. The power needed to form an ITB
is several times larger than the H-mode
power threshold; for this reason, the plasma
tends to develop large ELMs, which can
destroy the ITB. In order to reduce the
ELMs amplitude, the current ramp is
continued during main heating and the
divertor pumping is maximised by locating
the separatrix strike points at the corners of
the gas box divertor. Steady ITBs are only
obtained when excessive peaking of the
pressure profile is avoided; this is achieved
by controlling the plasma edge with
impurity injection (usually argon) and by
adjusting the power waveforms to slowly
build-up the plasma pressure [1.63]. As a
consequence of edge radiation, the q=2
surface slowly broadens, and wide ITBs are
produced, allowing good confinement to be
achieved. An example of a high-beta steady
pulse is shown in fig. 1.54.
4
0
H 89 x β N
4 5
Time (s)
Fig. 1.54
6 7
Fig. 1.54 - Time traces of plasma current (I p ), D a , argon puffing,
heating waveforms, neutron rate (S N ), electron temperature (T e ),
confinement enhancement (H 89 ), normalised beta β N and quality
parameter H 89 •β N
MHD activity at high β N
Optimised shear discharges show a variety
of MHD phenomena. Plasmas with peaked
pressure profile are subject to disruptions
caused by pressure-driven kink modes at
β N

1. Magnetic Confinement
peaking, i.e. by operating at high β N
with a broader ITB radius and an
H–mode pressure pedestal. Pressure
peaking was reduced by delaying the
high power heating phase and by
adding argon puffing; in particular,
ELMs size reduction through edge
radiation was essential to achieve
broad pressure profiles. In this way, a
long high-performance phase is
obtained, in which the only significant
MHD activity consists of tearing
modes with n≥2 [1.65].
The observation of tearing modes in
high β N discharges raises the question
of the possible role played by these
modes in limiting the performance
attained so far [1.66]. Figure 1.55
shows the comparison between two
discharges with similar heating
waveforms; both discharges attain
β N >2.5, in spite of a large difference in
their mode amplitude. Moreover, the
discharge with a smaller MHD activity
has an earlier ITB collapse, and in
general no change in the modes, is
observed before ITB collapses. There is
β N
10 19
2
0
0.2
0
2
# 46695
# 46701
n=2 mode amplitude (arb. un.) b)
n-
e at R=3.75 m
0
43 44 45 46
Time (s)
Fig. 1.55 - a) Time traces of normalised Fig. 1.55 beta for two discharges
with similar heating conditions. b) Qualitative indicator for the
amplitude of modes with even n. c) Line average density at
R=3.75 m, near the plasma edge
evidence that tearing modes locally weaken the pressure gradient, but this has a small global effect,
as profiles are broad at high β N .
The signals shown in fig. 1.55b result from the superposition of all the modes with an even
toroidal number. Magnetic signals analysis, performed for the discharge with a stronger activity,
shows two main coherent modes, namely one at 42 kHz with toroidal number n=2, and another
at 62 kHz with n=6. The poloidal (m) numbers were inferred from temperature oscillations, as
measured by a multichannel Electron Cyclotron Emission (ECE) radiometer with off-equatorial
line of sight, i.e. with each channel sampling a different poloidal angle. Cross-phase analysis
gives m=3 for the n=2 mode and m=9 for the n=6 mode; both modes resonate with the same
rational q=1.5. The analysis of ECE oscillations also gives the island position, which is identified
as the radius where the phase changes by π, and with the radial displacement profile (fig. 1.56).
The island positions on the outer midplane are R=3.23 m for the m/n=3/2 mode, and R=3.58 m
for the m/n=9/6 mode. This indicates that there is a pair of q=1.5 surfaces, i.e. a reversed shear
profile. The equilibrium reconstruction gives a monotonic q-profile, but the uncertainty of such
reconstruction in the absence of motional Stark effect data for this pulse does not allow us to
exclude the presence of a central region with a negative magnetic shear.
Inspection of the displacement profile (fig. 1.56) reveals that the 3/2 mode has a very unusual,
very broad and asymmetric shape, probably due to the presence of a very low shear region. The
9/6 mode has a more usual tearing structure, with localised and antisymmetric displacement. The
island width as estimated from the displacement profile, is w≈5 cm. This island size should be
large enough to give a significant bootstrap current perturbation, but the application of simple
estimates derived through a neoclassical tearing mode theory is not straightforward, due to the
non-standard shape of the current profile in the optimised shear regime. Further progress in this
field is needed, in order to be able to predict the behaviour of tearing modes at higher β N values.
c)
a)
47 48
55

1. Magnetic Confinement
Displacement (mm)
15
10
0
0
6
4
2
0
m=3, n=2
Phase inversion
m=9, n=6
Phase inversion
3.2 3.4 3.6 3.8
R (m)
Fig. 1.56 - Radial displacement (in mm) induced by tearing modes in
Fig. 1.56
pulse #46695 as a function of major radius coordinate. As both modes
have m/n=1.5, a pair of q=1.5 surfaces is expected to be localised at
the positions marked by arrows. Data refer to the time interval
46.26÷46.39 s
β N
3.5
3.0
2.5
2.0
1.5
Data at B=2.6 T
H 89 =3
H 89 =2
1.0
12 14 16 18 20 22 24 26
Fig. 1.57 - Normalised beta Pas NBI a +P function ICRH (MW) of the heating power. Dots
represent experimental data from the B=2.6 T database; lines
correspond to fixed values of the Fig. confinement 1.57 enhancement factor
Limits of the β N value
In order to obtain an insight of the nature
of the normalised beta saturation
encountered so far (β N 23 MW, which fall
below H 89 =2.3, result from discharges
with special edge conditions (septum
avoidance, see next section). The
observation of good confinement
conditions at the maximum power
indicates that the β N values achieved
so far with broad pressure profiles are
transport-limited.
Limits of duration at high β N
The steady phase at high performance is often interrupted before the end of the main heating
phase by some event that does not correspond to any change in the MHD activity (blank circles
in fig. 1.58). The duration of the steady phase, defined as the period with β N exceeding 90% of
the top value, clearly decreases at high power and high β N . Discharges with pulse-limited
duration, i.e. with a β N plateau lasting up to the end of the main heating pulse, are only found
below 20 MW heating power; two important exceptions, marked as “septum avoidance” in
fig. 1.58, are discussed below.
56

1. Magnetic Confinement
ITB collapses are generally preceded by a
density increase at the plasma periphery
(fig. 1.55c), indicating that some edge
phenomenon is taking place. A possible
explanation can be deduced from the
inspection of magnetic surfaces in the
divertor region: with the constraint of
placing the strike points at the divertor
corners, at high β N the separatrix comes
very close to the septum part of the gas box
divertor (fig. 1.59). In order to assess the
effects on plasma associated with such an
interaction, discharges have been developed
in which the magnetic configuration has
been modified by moving the strike points
on the divertor vertical plates, once the ITB
was triggered. Energy confinement is
slightly reduced in discharges with septum
avoidance, but heating pulse-limited
duration at the maximum power levels is
obtained (fig. 1.58).
1.6.2 ITB dynamics in JET
discharges with optimised shear
The formation of internal transport barriers
has been found in agreement with theories
based on turbulence suppression by ExB
shear flow. Turbulence suppression is
expected to be effective when the ExB
shearing rate (ω ExB ) exceeds the growth
rate of ion temperature gradient driven
modes (γ ITG ). The shearing rate has to be
calculated from the radial force balance of a
single impurity, including flows and
pressure gradient. As no direct
measurements were available for the
poloidal flow velocity, this quantity was
calculated from the parallel momentum
balance equation in the framework of the
neoclassical theory [1.67]. A simple model
expression was used for the ITG growth rate
[1.67]. Both triggering and radial expansion
of the ITB were found to be in good
agreement with the condition for turbulence
suppression (ω ExB >γ ITG ). The main
contribution to the feedback mechanism
Fig. 1.59 - Magnetic surfaces Fig. in 1.59the divertor region. The
normalised beta is β N =0.84 at t=4 s and β N =2.52 at t=6.5 s,
when septum interaction occurs
leading to the ITB onset was given by toroidal rotation. Figure 1.60 shows an example of ITB
evolution during power step-down: the shearing rate takes several confinement times to drop
below the ITG growth rate, and the ITB contraction occurs on the same time scale, as shown in
fig. 1.61.
Significant progress has been done in the development of integrated scenarios with high beta,
Z (m)
Duration (s)
5
4
3
2
1
0
10 15 20 25 30
P NBI +P ICRH (MW)
Fig. 1.58
Fig. 1.58 - Duration of the high β N phase vs heating power.
Open circles represent cases with the steady phase terminated
by an ITB collapse. Filled circles represent discharges with a
β N plateau lasting up to the end of the main heating pulse
-1.2
-1.4
-1.6
Data at B=2.6 T
Pulse No: 46695 B T = 2.6 T I p = 2.6 MA
t = 4s
t = 6.5s
Pulse-limited
duration
Septum
avoidance
-1.8
2.2 2.4 2.6 2.8 3.0 3.2
R (m)
57

1. Magnetic Confinement
Fig. 1.60
Fig. 1.60 - Time evolution of total power, neutron rate,
shearing rate ω ExB and turbulence growth rate γ ITG . Both
ω ExB and γ ITG are evaluated at R=3.35 m. The ITB contracts
between t=46.8 and t=47 s
Ion temperature (keV)
0
0.4
0.2
0
20
10
0
1.0
0.5
20
15
10
P NBI +P ICRH (MW)
Neutron rate
(10 16 n/s)
ω ExB (10 6 s -1 )
γ ITG
44 45 46 47
5
Time (s)
#49651
t = 46.0 s
t = 46.4 s
t = 46.8 s
t = 47.0 s
ITB and steady conditions. The
quality factor H 89 ×β N has
reached steady values up to 7.3,
in conditions of high neutron
yield. Steady conditions have
been achieved by controlling
edge conditions and pressure
peaking. The beta values
achieved so far (β N ≤2.6) have
been limited by the heating
power available. The duration of
the steady phase at high β N was
limited by plasma interaction
with the septum of the Gas Box
divertor.
The role of turbulence
stabilisation by ExB velocity
shear has been analized in
plasmas with an ITB. Both ITB
onset and ITB radial extent
coincide with the ExB shearing
rate exceeding the linear growth
rate of ion temperature gradient
driven turbulence. Also the ITB
contraction following power
step-down is in agreement with
the same condition.
In the exploration of MHD
boundaries, disruptions or q=2
snakes at β N

References
[1.1] ENEA-Nuclear Fusion Division 1999 Progress report
[1.2] B. Esposito et al., A gamma-ray spectrometer system for fusion application, accepted for
publication in Nucl. Instrum. Methods (2000)
[1.3] J.R. Martin-Solis et al., Proc. of the 27th EPS Conf. on Controlled Fusion and Plasma
Physics (Budapest 2000), Vol 24B, 165
[1.4] J.R. Martin-Solis, R. Sanchez, B. Esposito, Phys. Plasmas 7, 3814 (2000)
[1.5] J.R. Martin-Solis, R. Sanchez, B. Esposito, Phys. Plasmas 7, 3369 (2000)
[1.6] D. Pacella et al., Ultrafast soft x ray 2D plasma imaging system based on gas electron
multiplier detector with pixel read-out, accepted for publication in Rev. Sci. Instrum.
[1.7] F. Sauli, Nucl. Instrum. Methods A386, 531 (1997)
[1.8] R. Bellazzini et al., Nucl. Instrum. Methods A419, 429 (1998)
[1.9] A. H. Sarkissian et al. Rev. Sci. Instrum. 68, 1 (1997)
[1.10] Hinds Instr. Manual, 1994
[1.11] R. De Angelis et al., The motional stark effect diagnostic in FTU, presented at the 13 th
Topical Conference on High Temperature Plasma Diagnostics, (Tucson 2000)
[1.12] S. Tugarinov et al., Rev. Sci. Instrum. 66, 603 (1995)
[1.13] G. Bracco et al., ECRH results during current ramp-up and post-pellet injection in FTU
plasma, presented at the 18 th IAEA Fusion Energy Conference (Sorrento 2000), Paper EX6(03)
[1.14] G. Vlad, et al., Nucl. Fusion 38, 557 (1998)
[1.15] S. Cirant, et al., Mode coupling trigger of tearing modes in ECW heated discharges in
FTU, presented at the 18th IAEA Fusion Energy Conference (Sorrento 2000), Paper EX3(03)
[1.16] E. Lazzaro et al., Phys. Rev. Lett. 84, 6038 (2000)
[1.17] C. Sozzi, et al., Energy confinement and sawtooth stabilization by ECRH at high electron
density in FTU tokamak, presented at the 18 th IAEA Fusion Energy Conference (Sorrento 2000),
Paper EXP5/13
[1.18] S. Briguglio et al., Phys. Plasmas 2, 3711 (1995)
[1.19] L. Chen, Phys. Plasmas 1, 1519 (1994)
[1.20] L. Chen et al., Nonlinear zonal dynamics of drift and drift Alfvén turbulences in tokamak
plasmas, presented at the 18 th IAEA Fusion Energy Conference (Sorrento 2000), paper TH4/5
[1.21] F. Zonca et al., Proc. of the Joint Varenna-Lausanne Inter. Workshop on Theory of Fusion
Plasmas (Varenna 2000), pp.17-30
[1.22] F. Zonca et al., Energetic particle mode dynamics in tokamaks, presented at the 18th
IAEA Fusion Energy Conference (Sorrento 2000), paper THP2/20
[1.23] L. Chen, Z. Lin and R.B. White, Phys. Plasmas 7, 3129 (2000)
[1.24] S. Briguglio, F. Zonca and G. Vlad, Phys. Plasmas 5, 3287 (1998)
[1.25] F. Zonca and L. Chen, Phys. Plasmas 3, 323 (1996)
[1.26] E.A. Frieman and L. Chen, Phys. Fluids 25, 502 (1982)
[1.27] A. Hasegawa, C.G. Maclennan, and Y. Kodama, Phys. Fluids 22, 2122 (1979)
[1.28] A.M. Dimits et al., Phys. Rev. Lett. 77, 71 (1996)
[1.29] Z. Lin et al., Science 281, 1835 (1998)
[1.30] M.A. Beer, Ph.D. dissertation, Princeton University (1995)
[1.31] J.Y. Kim, W. Horton, and J.Q. Dong, Phys. Fluids B5, 4030 (1993)
[1.32] F. Zonca et al.; Phys. Plasmas 6, 1917 (1999)
[1.33] B. Scott, Plasma Phys. Control. Fusion 39, 1635 (1997)
[1.34] E.A. Frieman, and L. Chen, Phys. Fluids 25, 502 (1982)
59

References
[1.35] P.H. Diamond, Private Communication (2000)
[1.36] H. Biglari et al., Proc. 9 th Topical Conf. on Radiofrequency Power in Plasmas, ed. by
D.B. Batchelor (Charleston 1991), p 376
[1.37] M. Ono et al., Proc. 15 th Int. Conf.on Plasma Physics and Controlled Nuclear Research,
ed.by IAEA (Vienna 1994) Vol. I, p. 469
[1.38] B. LeBlanc et al., Phys Plasmas 2, 741 (1995)
[1.39] S. Sesnic et al., Nucl. Fusion 38, 835 (1998)
[1.40] S. Sesnic, et al., Nucl. Fusion 38, 861 (1998)
[1.41] G.G. Craddock et al. Phys. Plasmas 1, 1944 (1994)
[1.42] L.A. Berry, E.F. Jaeger, D.B. Batchelor, Phys. Rev. Lett. 82, 1871 (1999)
[1.43] H. Abe et al., Phys. Rev. Lett. 53, 1153 (1984)
[1.44] M. Porkolab, Phys. Rev. Lett. 54, 434 (1985)
[1.45] P. Palmadesso and G. Schimdt, Phys. Fluids 14, 1411 (1971)
[1.46] B.P. LeBlanc, et al., Proc. 12th Inter. Conference on Radio Frequency Power in Plasmas,
(Savannah 1997) p. 81
[1.47] K.G. Budden, G.W. Jull, Canadian J. Phys. 42, 113 (1964)
[1.48] K.G. Budden, P.D. Terry, Proc. R. Soc. London, A.321, 275 (1971)
[1.49] K.G. Budden, Phil. Trans. R. Soc. London, A280, 111 (1975)
[1.50] P.D. Terry, Proc. R. Soc. London A.363, 425 (1978)
[1.51] L.B. Felsen, J. Opt. Soc. Am. 66, 751 (1976)
[1.52] K.A. Connor, IEEE Trans. Plasma Sci., PS8, 96 (1980)
[1.53] Z.S. Wang, Phys. Scr. 29, 482 (1984)
[1.54] A. Bravo-Ortega, PhD. Dissertation, Auburn University, (1988)
[1.55] A. Bravo-Ortega, A.H. Glasser, Phys Fluids B3, 529 (1991)
[1.56] J.R. Wilson et al., Proc. 13th Inter.Conference on Radio Frequency Power in Plasmas
(Annapolis 1999) p. 168
[1.57] S. Chandrasekar and P.C. Kendall, Astrophys. J. 126, 457 (1957)
[1.58] H.P. Furth, M.A. Levine and R.W. Waniek, Rev. Sci. Instr. 28, 949 (1957)
[1.59] I.E. Bernstein et al., Proc. R. Soc. London, Ser A 244, 17 (1958)
[1.60] A.H. Boozer, Phys. Fluids 24, 1999 (1981)
[1.61] R. Gruber and J. Rappaz, Finite element methods in linearideal magnetohydrodynamics,
(Springer-Verlag, Berlin) 1985
[1.62] N. Amemiya, A. Morita and M. Katsurai, J. Phys. Soc. Jpn. 63, 1552 (1993)
[1.63] C. Gormezano Plasma Phys. Control. Fusion 41, B 367 (1999)
[1.64] G.T.A. Huysmans et al., Nucl. Fusion 39, 1489 (1999)
[1.65] T. Hender et al, Proc. 26 th EPS Conf. on Controlled Fusion and Plasma Physics
(Maastricht 1999), Vol. 23A
[1.66] P. Buratti and the JET Team, High beta plasmas and internal barrier dynamics in JET
discharges with optimised shear, presented at the18 th IAEA Fusion Energy Conf. (Sorrento
2000), paper EX7(1)
[1.67] F. Crisanti et al., Analysis of ExB flow shearing rate in JET ITB discharges, accepted for
publication in Nucl. Fusion
60

References
[1.35] P.H. Diamond, Private Communication (2000)
[1.36] H. Biglari et al., Proc. 9th Topical Conf. on Radiofrequency Power in Plasmas, ed. by
D.B. Batchelor (Charleston 1991), p 376
[1.37] M. Ono et al., Proc. 15th Int. Conf.on Plasma Physics and Controlled Nuclear Research,
ed.by IAEA (Vienna 1994) Vol. I, p. 469
[1.38] B. LeBlanc et al., Phys Plasmas 2, 741 (1995)
[1.39] S. Sesnic et al., Nucl. Fusion 38, 835 (1998)
[1.40] S. Sesnic, et al., Nucl. Fusion 38, 861 (1998)
[1.41] G.G. Craddock et al. Phys. Plasmas 1, 1944 (1994)
[1.42] L.A. Berry, E.F. Jaeger, D.B. Batchelor, Phys. Rev. Lett. 82, 1871 (1999)
[1.43] H. Abe et al., Phys. Rev. Lett. 53, 1153 (1984)
[1.44] M. Porkolab, Phys. Rev. Lett. 54, 434 (1985)
[1.45] P. Palmadesso and G. Schimdt, Phys. Fluids 14, 1411 (1971)
[1.46] B.P. LeBlanc, et al., Proc. 12th Inter. Conference on Radio Frequency Power in Plasmas,
(Savannah 1997) p. 81
[1.47] K.G. Budden, G.W. Jull, Canadian J. Phys. 42, 113 (1964)
[1.48] K.G. Budden, P.D. Terry, Proc. R. Soc. London, A.321, 275 (1971)
[1.49] K.G. Budden, Phil. Trans. R. Soc. London, A280, 111 (1975)
[1.50] P.D. Terry, Proc. R. Soc. London A.363, 425 (1978)
[1.51] L.B. Felsen, J. Opt. Soc. Am. 66, 751 (1976)
[1.52] K.A. Connor, IEEE Trans. Plasma Sci., PS8, 96 (1980)
[1.53] Z.S. Wang, Phys. Scr. 29, 482 (1984)
[1.54] A. Bravo-Ortega, PhD. Dissertation, Auburn University, (1988)
[1.55] A. Bravo-Ortega, A.H. Glasser, Phys Fluids B3, 529 (1991)
[1.56] J.R. Wilson et al., Proc. 13th Inter.Conference on Radio Frequency Power in Plasmas
(Annapolis 1999) p. 168
[1.57] S. Chandrasekar and P.C. Kendall, Astrophys. J. 126, 457 (1957)
[1.58] H.P. Furth, M.A. Levine and R.W. Waniek, Rev. Sci. Instr. 28, 949 (1957)
[1.59] I.E. Bernstein et al., Proc. R. Soc. London, Ser A 244, 17 (1958)
[1.60] A.H. Boozer, Phys. Fluids 24, 1999 (1981)
[1.61] R. Gruber and J. Rappaz, Finite element methods in linearideal magnetohydrodynamics,
(Springer-Verlag, Berlin) 1985
[1.62] N. Amemiya, A. Morita and M. Katsurai, J. Phys. Soc. Jpn. 63, 1552 (1993)
[1.63] C. Gormezano Plasma Phys. Control. Fusion 41, B 367 (1999)
[1.64] G.T.A. Huysmans et al., Nucl. Fusion 39, 1489 (1999)
[1.65] T. Hender et al, Proc. 26th EPS Conf. on Controlled Fusion and Plasma Physics
(Maastricht 1999), Vol. 23A
[1.66] P. Buratti and the JET Team, High beta plasmas and internal barrier dynamics in JET
discharges with optimised shear, presented at the18th IAEA Fusion Energy Conf. (Sorrento
2000), paper EX7(1)
[1.67] F. Crisanti et al., Analysis of ExB flow shearing rate in JET ITB discharges, accepted for
publication in Nucl. Fusion
60

2. Ignitor
2.1 INTRODUCTION 63
2.2 PHYSICS 63
2.2.1 FTU experiments 63
2.2.2 Pellet fuelling 64
2.2.3 Consistency of the plasma pressure profile 64
2.2.2 X-point configurations for advanced scenarios 64
2.3 ENGINEERING OF THE MACHINE 66
2.3.1 Introduction 66
2.3.2 Engineering of the machine 67
REFERENCES 69

2. Ignitor
2.1 INTRODUCTION
A large part of the Physics design of Ignitor is carried out by Prof. Bruno Coppi’s Team at MIT,
USA, where this Ignitor Group directly takes part in all scientific discussions and meetings on
the strategy of the Fusion research. A significant step of this wide debate was the Snowmass
summer workshop in Colorado, in July ’99, where the fusion community found a general
consensus on the statement that “The Tokamak is technically ready for a high gain burning plasma
experiment”. Following the result of this meeting, and recognizing the widespread interest in the
burning plasma as manifested by the fusion community, the Department of Energy entrusted the
University Fusion Association Group (UFA) with the task to examine the features of Physics and
Technology to realize, in a medium term, a Burning Plasma Science eXperiment (BPSX).
The UFA sponsored a first Workshop on “Burning Plasma Science”, held at the University of
Texas, Austin on 11-13 December 2000, which focused on the scientific issues of BP Physics. It
is worth to mention the letter [2.1] addressed by Prof. M. Rosenbluth,a plasma physics famous
scientist, to the participants to the meeting, where he clearly stated: “In views of past history and
present and likely future budgetary climates here and abroad, it seems prudent to look for the last
costly experiment which has a high probability of success, both in answering most critical
Science issues and in serving to convince the world that fusion is a scientific possibility”, and
suggested Ignitor as the swiftest candidate for BPSX. Actually three experiments have been
taken into consideration so far: ITER FEAT, FIRE and Ignitor, but the last shows unique features,
with its compact design and use of known technologies, which assure low costs and contained
construction times, and make it attractive to fill the time gap between the present experiments
and the new, more ambitious, machine generation.
An international assessment has been completed during the year 2000 by the Thermonuclear
Tokamak Panel, chaired by Prof. Laval, of École Politecnique of Paris, which compared the
physics bases of ITER and Ignitor. Also this Group, issuing their final Report [2.2] defined
Ignitor as a great scientific experiment on the frontier of plasma physics.
Progress on the physics and engineering of the machine during the 2000 year are here described.
2.2 PHYSICS
2.2.1 FTU experiments
Following the involvement of the Ignitor unit in the FTU experimental campaign of the year
2000, experimental sessions of FTU have been devoted to the study of confinement properties of
plasmas at the nominal maximum field (8T), maximum current (1.6 MA), and at high density
(8×1020 m-3).
The goals of the above study were the following:
• the extension of the transport database to the parameter region in which Ignitor is designed to
operate at;
• validation of the existing scaling laws for particle and energy confinement time which have
been derived from a database dominated by low field, low current discharges;
• verification that the above scaling laws are still valid for high field, high density plasmas and
therefore valid to predict Ignitor’s confinement time.
The investigation of regimes of enhanced confinement, following the injection of pellets in
ohmic discharges. Density profile control through pellet injection and the possibility to achieve
enhanced confinement regimes is very important to Ignitor, since this would increase the
flexibility of the machine and widen the range of experiments that can be performed.
The FTU experiments described above have been successful and an analysis [2.3] of the best
63

2. Ignitor
discharges has been reported at the IAEA conference held in Sorrento (Italy). It was
demonstrated that FTU can be operated at the nominal maximum field and current and that the
discharge reacts very well to the injection of several pellets (up to 5 have been injected in the
same discharge). Although an enhancement of confinement was observed against the ITER89Lmode
scaling law (up to a factor of two), it was confirmed that ITER97P well predicts the
confinement time of high-density high field discharges.
Confinement enhancement and transport barriers have also been noticed on ALCATOR C Mod
when off-axis Ion Cyclotron Resonance Heating (ICRH) is applied as external heating. The
physics underlying this phenomenon has been investigated by Prof. Bruno Coppi and related to
the appearance of toroidal rotation of the discharge.
Similar experiments with ICRH heating have been performed on Tore-Supra (CEA) where
enhancements over the ITER97 scaling law up a factor 1:6 have been reported.
2.2.2 Pellet fuelling
A pellet injector has always been considered an integral part of the machine design in order to
have: i) fast core fuelling; ii) density profile control; iii) time-dependent burn control. The
controlled injection of tritium, to promote the formation of internal transport barriers, and
diagnostic purposes can also be envisioned.
A tentative assessment of the fuelling requirements for Ignitor has been carried out, based on
time dependant transport simulations. The particle flow required to increase the density is only
a fraction compared to that necessary to compensate for the particle losses at the edge. The
interesting results from high field side injection experiments, and the very recent ones on vertical
injection in D-III-D, suggest that the best solution for Ignitor is that of a vertically mounted,
high-speed injector, producing pellets with velocities up to 3 km/s.
2.2.3 Consistency of the plasma pressure profile
The optimal ignition conditions for Ignitor have been determined by a series of complex 1+1/2
dimensional simulations performed under a variety of assumptions regarding the plasma thermal
transport properties [2.4]. A recently observed feature is the consistency of the plasma pressure
profile at ignition. The word consistency means that, although the plasma evolution is
determined by different expressions for χ e , at ignition the pressure profile turns out to be nearly
of a unique type. The comparison concerns simulations of the 11 MA scenario using fourelectron
thermal diffusivity model. The consistency of the plasma pressure profile at ignition in
Ignitor, pointed out by previous analyses, has been confirmed by other dynamic simulations.
Such a profile turns out to depend much more on the achievement of ignition than on the
transport assumptions. It should be noted that density and temperature profile, separately, do not
exhibit the same characterization. A satisfactory analytic formula fitting the pressure profile has
been determined and compared to similar expressions considered in zero-dimensional
evaluations. The results, presented at the IAEA Fusion Conference and in other scientific
meetings, will be published in Nuclear Fusion [2.5-2.8].
2.2.4 X-point configurations for advanced scenarios
An important improvement in the design of the machine, which allows to greatly increase its
flexibility, has been studied, namely the production of x-point configurations, essential to explore
advanced scenarios. Two different plasma shapes have been examined: a double null
configuration with x-points laying just outside the first wall, and a single null configuration with
an x-point at the bottom, laying on the first wall. In both cases the plasma height must be reduced
from its reference value to allow space for the scrape off layer. The separatrix solutions are
64

0.17
0.42
2. Ignitor
constrained to have the usual design value of
q 95 ≈3, resulting in a reduced plasma current:
10 MA for the double null configuration and
≈9 MA for the single null. A common feature of
these configurations is the requirement of currents
in the PF poloidal coils close to the X-point
significantly higher than the reference values,
which might require revision of Plasma Facing
Components (PFC) design.
0.00
0.18
0.36
-0.36
In the code developed to treat single null
configurations (EQUI1X), auxiliary feedback coils
are introduced to provide a horizontal field in the
plasma region to prevent vertical displacements of
the plasma column. The relevant current is
adjusted at each iteration so as to guarantee a
preselected vertical position of the magnetic axis.
The single null configuration given by our code has
been compared to the one computed for Ignitor by
the CORSICA code [2.9]. Here the magnetic axis
is shifted to Z axis ≈0.10m. The plasma shape was
quite well matched, but the separatrix was found to
partly lay on the first wall (see fig. 2.1), thus giving
rise to a limiter configuration. By changing some
PF currents a new configuration was found which
has no contact between the separatrix and the wall
(see fig. 2.2). The feedback control introduced in
0.00
0.89
0.53
0.71
-0.18
Fig. 2.1 - Single null configuration
touching the first wall
0.00
0.00
0.21
-0.52
0.35
0.63
0.52
-0.17
0.84
0.70
-0.35
1.04
0.00
-0.21
0.00
Fig. 2.2 – Single null configuration
Fig. 2.3 - Double null configuration
65

2. Ignitor
the simulations allows to check the ‘robustness’ of the obtained configurations and to assess the
optimal position of the feedback loops. As a matter of fact, by modifying the currents of the
amount required in the feedback loops and recomputing the equilibrium, the new feedback
currents come out to be about zero. The upper and lower coils P9 were found to be the most
suitable for accommodating feedback loops. The double null configuration is shown in fig. 2.3.
The relevant thermal loads on the first wall have been evaluated by C. Ferro [2.10] and found to
be acceptable. Transport simulations relevant to these new scenarios are in progress.
2.3 ENGINEERING OF THE MACHINE
2.3.1 Introduction
Ignitor is a high-field compact machine, proposed and designed to achieve ignition in
Deuterium-Tritium (DT) plasmas. The machine has been conceived as a completely integrated
system of its major components (toroidal field system, poloidal field system and plasma
chamber) (fig. 2.4).
The toroidal field magnet is made of 12 modules (fig. 2.5), each including two Toroidal Field
Coils (TFC), which are contained in 4 C-clamp elements.
The structural performance of the machine relies on an optimised combination of“wedging” in
the TFC inboard legs and in the outboard of the C-clamp, and “buckling”, between the TFC and
the Central Solenoid (CS). The inboard legs of the TFC are pre-loaded to resist the vertical
components of the Lorenz forces. The pre-load is applied through the upper and lower parts of
the C-clamp (the C-clamp “noses”) by
means of bracing rings (passive system)
and an electromagnetic radial press
(active system).
Plasma
chamber
Toroidal field
coil
C-Clamp
elements
Fig. 2.4 - Ignitor machine cross-section
Fig. 2.5 - Ignitor magnet module
66

2. Ignitor
2.3.2 Engineering of the machine
The work was carried out by Ansaldo with ABB as a subcontractor, under supervision by ENEA.
The design activity concerning the load assembly went as far as developing the detail drawings
of its components. Design of vertical and radial
(fig. 2.6) Plasma Chamber (PC) supports were
completed, and one of the most demanding
component of the radial support was successfully
tested (fig. 2.7). Main function of the PC supports
is to react to the vertical and radial
electromagnetic loads, induced by a transient
plasma disruption; to allow for free movement
under thermal loads; and to provide for the
electrical isolation from the C-clamps and
cryostat. An overall analysis of the machine
cooling-down was performed and critical coils,
from a thermal point of view, were analysed in
detail. The complete integration among the
different major components, and the consequent
thermal contact among them, ensures cooling
down of all the components at 30 K. The thermal
interactions between C-clamp (fig. 2.8) and TF
coils, and between C-clamp and external PF
coils, have been evaluated by means of the
ANSYS code, with the same geometrical models
used for the stress analysis [2.11].
A dedicated cooling system provides for cooling
of the plasma chamber. For optimal plasma
Fig. 2.6 - Radial support
Z
C
C
CC
C
C
D
C
I
C
D CC
DD
I
H E D
D
MX
I I
I H
FGH
FGH
FG I I
FG
G
F
E
D
D
II
E
FG
H EFG
FG
H H G
DD
EE
F E
E
H G F
E
E
E FGH HH
H
G F EE
F F
F
EEFG
FG H
F
E EE
F
G H
G
EF G H
BCD E
H F
D H
H G FF
GG
G
CE FG H G
G
A
G
G
G G
G
ABC
FG
G
G G
A B
A BCDE EG
G
MN G GG
G
BG
CDEF B CD EF G G G
F GG
G
G
C
B DE
E
G
B CD
Y
CD
EF
G
G
DEF
G
D EF
G G G G G
D G
G
DEF
G G G
X
D
F
G
DEE
E F
E F
F
F F F F
E
E
F
B
E
CD
B E
E
C
D E
B
B C E
C D E
Time = 14400 (4h)
Temp
TEPC = 84.127
SMN = 28.173
SMX = 63.733
A = 30.148
B = 34.100
C = 38.051
D = 42.002
E = 45.953
F = 49.904
G = 53.855
H = 57.806
I = 61.758
G
E
E
E
E
E E
E
E
E E
E
Fig. 2.7 - Testing set up
Fig. 2.8 - Thermal map of C-clamp after 4 h from pulse
67

2. Ignitor
Fig. 2.9 - PC temperature after
4 h of pulse
Vacuum insulated cryostat
Electrical busbar of TF coils
Electrical power
supply of toroidal
field coils
Ignitor machine
Cryostat feed throw for
current and helium
(Vacuum tight end
electrically insulated)
Vacuum insulated
connection pipe
Electrofluidic lines
Auxiliary cold-box
for high pressure
line control (1 of 13)
Auxiliary cold-box
for low pressure
line control (1 of 4)
Electrical busbar of PF coils
Auxiliary cold-box
for control of
fluid distribution
Main cryogenic line
Electrical power
supply of poloidal
field coils
(1 of 14)
Refrigeration
plant
(1 of 3 groups)
Fig. 2.10 - Sketch for cooling plant
operations, a warm vessel is more desirable; a temperature of about 293 K is a good operating
point. On the outer surface of the PC, suitable ducts, fed by helium gas, are provided to cool
down the PC after baking (T=250 °C) and to remove the heat deposited on the First Wall (FW)
during each shot. Analyses are being performed using the ANSYS code (fig. 2.9). The PC is
considered to the completely thermally insulated. The outlet pressure is 13 bar; the pressure drop
is 1.7 bar. The thermo-mechanical stresses induced are maintained within the allowable values.
Finally, a preliminary sketch for cooling plant has been developed (fig. 2.10).
68

References
[2.1] M.N. Rosenbluth, From yearning to burning (Possible broad-brush guidelines for burning
plasma thinking), Letter addressed to the participants at the Workshop on Burning Plasma
Experiment Physics (Austin 2000)
[2.2] Thermonuclear tokamak panel, Final Report, (Paris 2000)
[2.3] D. Frigione et al., Steady improved confinement in FTU high field plasmas sustained by
deep pellet injection, presented at the 18th IAEA Fusion Energy Conference (Sorrento 2000)
paper P4.52(1)
[2.4] A. Airoldi, G. Cenacchi, Nucl. Fusion 37, 1117 (1997)
[2.5] A. Airoldi et al., Ignition and appropriate confinement regimes in Ignitor, presented at the
Int. Sherwood Fusion Theory Meeting (Los Angeles 2000)
[2.6] A. Airoldi, G. Cenacchi, Privileged pressure profiles at ignition, Rep. FP 00/13, Istituto di
Fisica del Plasma, Milano, (2000)
[2.7] G. Cenacchi et al., Bulletin of the American Physical Society 45, 282 (2000)
[2.8] A: Airoldi, G. Cenacchi, B. Coppi, Bulletin of the American Physical Society, 45, 283
(2000)
[2.9] R.H. Bulmer, Private Communication (2000)
[2.10] C. Ferro, Bulletin of the American Physical Society 45, 284 (2000)
[2.11] A. Bianchi, Calcolo strutturale soluzione di riferimento, Ansaldo Report
IGN.ANE.I.1009 (May 1999)
69

3. Technology Programme
3.1 INTRODUCTION 73
3.2 MAGNETS 74
3.2.1 Design of conductors and magnets for ITER European Fusion Development
Agreement (EFDA) (Task Two-T405/2) 74
3.2.2 Installation and testing of the ITER CS and TF model coils (ITER Task M20) 74
3.2.3 Survey of the TF model coil geometry 75
2.2.4 ITER TF casing manufacturing opmisation (ITER Task GB8-M45) 76
3.2.5 Numerical simulations of welds of thick steel relevant for ITER TF case 76
3.2.6 Development of new calculation codes for cable-in-conduit conductors 79
3.2.7 Development of NbTi conductors for the poloidal field coils of ITER (ITER Task M50
and EFDA Task Two-T405/1) 80
3.2.8 Stability and quench propagation on an NbTi CIC test conductor 80
3.3 VACUUM VESSEL AND SHIELD 82
3.3.1 ITER blanket modules: testing on the earth strap connections (EFDA task two-BM/STRAP) 82
3.3.2 Design of the Plasma-Facing Component (PFC) for the divertor of ITER FEAT (EFDA
Contract/00-544) and EM analyses of shielding blanket for ITER-FEAT design options, during
plasma disruptions (EFDA Contract/00-570) 83
3.4 FIRST WALL AND DIVERTOR 84
3.4.1 Runaway electrons on ITER PFCs (EFDA Contract/00-520) 84
3.4.2 Deuterium desorption measurements of W-1% La 2 O 3 (EU Task DV 7A-ITER Task EU-T438) 84
3.4.3 Design of a welded divertor cassette (NET Contract/98-488) 86
3.4.4 Neutron diffraction study of high-temperature stresses in brazed divertor mockups 87
3.4.5 Mechanical and electrical tests on the attachment keys for the divertor vertical target 88
3.4.6 Non-destructive testing of permanent components with calibrated defects (ITER task T222-14) 90
3.4.7 Thermal fatigue testing of vertical target mockups manufactured by diffusion bonding
(ITER task DV1/01) 92
3.5 REMOTE HANDLING 93
3.5.1 Overview of the ENEA contribution to the implementation of the ITER L-7 project 93
3.5.2 Divertor test platform (Tasks T308/1 and TW0/DTP01) 93
3.5.3 Divertor refurbishment platform (Tasks T308/5, T308/8, and TW0/DRP01) 94
3.5.4 Laser in-vessel viewing & ranging systems (JET order/JWO-OFT-ENEA-02)
(EFDA TWO DTP/01-6-7-11) 95
3.5.5 IVROS articulated boom 96
3.5.6 Multilink general purpose boom 97
3.6 BREEDING BLANKET 98
3.6.1 Compatibility test between EUROFER and Li 2 TiO 3 or Li 4 SiO 4 pebble bed 98
3.6.2 Exposure to lithium titanate 98
3.6.3 Exposure to lithium silicate 99
3.6.4 Li 2 TiO 3 pebbles reprocessing, recovery of 6Li as Li 2 CO 3 100
3.7 IFMIF 101
3.7.1 Activities on IFMIF optimisation and cost reduction (EFDA Contract EFDA 99/506) 101
3.7.2 Tasks of the key engineering phase 102

3.8 NEUTRONICS 103
3.8.1. Experimental validation of shutdown dose rates for ITER 103
3.8.2 Evaluation of neutron cross-sections for fusion-relevant materials (EFF project) 105
3.8.3 Neutronics benchmark experiment on SiC (EFF project) 105
3.8.4 Experimental validation of neutron cross-sections for fusion-relevant materials (EAF project) 105
3.8.5 Activation foils and real-time neutron/gamma detectors for IFMIF 107
3.8.6 Development of Chemical Vapour Deposition (CVD) diamond detectors for nuclear radiation (*) 107
3.8.7 Participation in the Astrorivelatore Gamma ad Immagini LEggero (AGILE) project - The
collimator and the coded mask of the SuperAGILE detector (*) 108
3.9. FUEL CYCLE 108
3.9.1 Development of palladium-ceramic membranes 108
3.10 SAFETY AND ENVIRONMENT 109
3.10.1 Assessment of ORE (Task SEA2) 109
3.10.2 Plant safety assessment (Task SEA4) 110
3.10.3 Validation of computer codes and models (Task SEA5) 115
3.10.4 Occupational dose and development of requirements for environmental releases (Task TRP1) 116
3.10.5 Waste management 117
3.10.6 Socio economics studies 118
3.11 MATERIALS 119
3.11.1 Manufacturing of improved Polymer Impregnation and Pyrolysis (PIP) composites 119
3.11.2 Development of high performance SiC fibres composites 120
3.11.3 Development of polymer-based joining techniques for SiC/SiC f composites 120
3.11.4 Development of sealing coating for SiC/SiC f composites 120
3.11.5 Development of a design methodology for components made of SiC/SiC f composites 121
3.11.6 Compatibility of SiC/SiC f composites with Pb17Li 121
3.11.7 Low cycle fatigue (LCF) of Reduced Activation Ferritic Martensitic (RAFM)
steel in water with additives 122
3.11.8 Mechanical properties of RAFM steel base material and joints 124
3.11.9 Microstructural investigation of the effects in RAFM steels using Small-Angle
Neutron Scattering (SANS) 124
3.11.10 Mechanical characterisation of materials with miniaturised specimens 125
3.12 LIQUID METAL AND HYDROGEN/MATERIAL INTERACTION TECHNOLOGY 125
3.12.1 Interaction between lead-lithium alloy and water in conditions relevant for DEMO 125
3.12.2 Qualification of tritium permeation in Pb17Li/gas 127
3.12.3 Transport parameters and solubility of hydrogen in Pb17Li 128
3.12.4 Feasibility study of a modified concept of WCLL DEMO blanket 129
3.12.5 Corrosion and mechanical tests on EUROFER 97 in Pb17Li 130
3.12.6 Lithium corrosion and chemistry for IFMIF target 130
3.12.7 Hydrogen permeability and embrittlement in EUROFER 97 martensitic steel 131
3.12.8 Measurements of H/D diffusivity in and solubility through tungsten and tungsten
alloys in the temperature range of 600°C to 800°C (ITER task 436) 133
3.13 THERMAL FLUID-DYNAMICS 133
3.13.1 Tests on beryllium pebble bed by Small Rectangular Test Sections (SMARTS) 133
3.13.2 Non-nuclear tests for the solid breeder blanket in the HE-FUS3 facility 134
3.13.3 Fabrication and testing of a full-scale ITER divertor outboard mockup 136
3.13.4 Fatigue tests on six mockups of the primary first wall panel prototype
(EFDA Contracts 00/529 and 00/533) 138
REFERENCES 141
(*) Not in association framework

3. Technology Program
3.1 INTRODUCTION
In the framework of the European Fusion Development Agreement (EFDA), the ENEA-Euratom
Association is also involved in the technological research program concerning the Next Step
(ITER Project), the Long Term (Breeder Blanket, Materials, IFMIF), the Power Plant Conceptual
studies, the Socio-Economics Studies and the Underlying Technologies. Technology activities
are being performed at the Frascati and Brasimone Fusion Division laboratories and take
advantage from the valuable contribution of other ENEA Units. The main fields of investigation
are the following:
• Magnets: participation in the International Thermonuclear Experimental Reactor Central
Solenoid (ITER CS) and TF (Toroidal Field) model coils test campaigns; experiments on
superconducting magnets under transient regimes; development of codes for modeling the
behaviour of cable-in-conduit conductors.
• Vacuum Vessel and Shielding: characterisation of components able to sustain high
electromagnetic loads for ITER shield blanket; electromagnetic analyses of ITER FEAT shield
blanket and vacuum vessel.
• First Wall and Divertor: assessment of the effect of electron run-away on ITER Plasma-
–Facing Components; development of new divertor cassette components; mechanical and
electrical testing on vertical target attachments; deuterium desorption measurements of W.
• Remote Handling: ENEA is responsible for the coordination and integration of the whole
ITER Large Project L-7, to demonstrate the feasibility of maintenance and refurbishment
operations of divertor components; development of radar optic viewing and metrology system
for JET and ITER and articulated booms for FTU.
• Breeding Blanket: compatibility tests between EUROFER and ceramic breeders; development
of lithium titanate reprocessing for 6Li recovery.
• International Fusion Material Irradiation Facility (IFMIF): ENEA is in charge of integration
activities and lithium target design (including remote handling) and tests.
• Neutronics: validation of shutdown dose rates for ITER, using two independent experimental
techniques; evaluation of neutron cross section for the European Fusion File (EFF) and the
European Activation File (EAF); development of gamma, x-rays, neutron detectors for IFMIF;
plasma diagnostics and space applications.
• Fuel Cycle: development of palladium-ceramics membranes for ITER fuel cycle (recovery of
tritium from tritiaded water) and industrial applications.
• Safety and Environment: Plant Safety (PS) assessment; Occupational Radiation Exposure
(ORE) assessment; validation of computer codes and models; waste management; Socio
Economics Research for Fusion (SERF2).
• Materials: manufacturing of high performance SiC fibres; development of design
methodology for SiC/SiC f composites; compatibility tests; mechanical and microstructural
characterisation of reduced activation ferritic martensitic steel.
• Liquid Metal Technology: interaction between lead-lithium alloy and water; qualification of
tritium permeation of materials in Pb17Li/gas; transport parameters and solubility of hydrogen
in Pb17Li; feasibility study of a modified concept of the Water-Cooled Lithium Lead
Demontration Reactor (WCLL-DEMO) blanket; hydrogen permeability and embrittlement in
EUROFER 97 martensitic steel; corrosion and mechanical tests on EUROFER 97 in Pb17Li;
measurement of H/D diffusivity solubility in W; lithium corrosion and chemistry for IFMIF.
• Thermal Fluid-Dynamics: tests on beryllium pebble bed and solid breeder blanket; fatigue
tests on ITER divertor cassette; set up of thermal-hydraulic tests on ITER first wall panels.
73

3. Technology Program
3.2 MAGNETS
3.2.1 Design of conductors and
magnets for ITER European
Fusion Development
Agreement (EFDA) (Task Two-
T405/2)
The poloidal coils of ITER are
expected to be superconducting.
Because of the relatively low value
of the maximum field (around 6 T, to
be compared to 12–13 T for the CS
and TF coils), the material that will
Fig. 3.1 - Von Mises stresses (Pa) of a multi-layer poloidal field NbTi be used is NbTi. As for the CS and
insert for the Toska Facility
TF coils, a long sample of a full-size
conductor (around 100 m), wound as
a single - or multi-layer “insert” coil, will be tested in conditions similar to those of ITER. ENEA
has produced a number of conceptual designs of such inserts, suitable for testing in the bore of
the European Union Large Coil Task (EU LCT) coil located at the TOroidal Spulenanlage
KArlsruhe (TOSKA) facility.
The use of this facility has been taken into account as a back-up solution, in case the more
suitable one of the Central Solenoid Model Coil (CSMC) facility at Japan Atomic Energy
Research Institute (JAERI) will not be available for testing of this insert coil.
Magnetic field and forces distributions have been calculated as well as the inductance matrix.
Finite element calculations have also been performed to evaluate the stress distribution in the
winding pack. Figure 3.1 shows the Von Mises stresses for one of the solutions examined.
3.2.2 Installation and testing of the ITER CS and TF model coils (ITER Task
M20)
In the frame of the ITER activities, testing of the CSMC and the Toroidal Field Model Coils
(TFMC) represents a very ambitious project, which aims at demonstrating the feasibility of the
reactor superconducting magnets.
The CSMC was cooled down to 4.5 K by April 2000, at the JAERI facility in Naka (Japan).
Coils were extensively tested in direct current (DC) and pulsed regimes by August 2000.
As a member of the European Union Home Team, ENEA contributed to the preparation of the
testing program, the execution of the tests, and the analysis of part of the results. It was also
charged to coordinate the activities of the other EU partners in this field. All the main goals of
the testing program were achieved and the results demonstrated that large, high-field
superconducting magnets with predictable properties can be designed and constructed [3.1-3.5].
The main outcome of the ENEA data analysis was the quantitative determination of the
continuous decrease in the conductor coupling losses during the testing campaign (fig. 3.2). The
losses are proportional to the decay time constant τ. The heat generated by this type of losses,
related to the magnetic field variation, produces a coolant temperature increase and heavily
contributes to the dimensioning of the cooling plant. The phenomenon of the loss decrease,
earlier observed during the tests of an ITER relevant coil at the ENEA laboratories [3.6], could
make for appreciable savings in cost.
As a member of the EU TFMC Operation Group, ENEA contributed to monitoring the assembly
74

3. Technology Program
of the toroidal field model coil (TFMC)
inside the TOSKA facility, at the Research
Centre Karlsruhe. The coil, completed at the
end of 2000, is a “race-track” shaped
winding, wound with a cable-in-conduit
conductor built up of 720 internal tin Nb 3 Sn
strands (0.81 mm diam.) and 360 copper
strands jacketed by a circular 1.6 mm thick
316 LN Stainless Steel (SS) tube.
To assemble the coil into the TOSKA
facility and to make it fit for the EU-LCT
coil, a heavy Inter–Coil Structure (ICS) has
been built. Within the ICS, the TFMC is
held by four wedges, shaped in a way that
the maximum stresses in coil and case can Fig. 3.2 - Coupling loss evolution for the layers 1A and 3A of the
be compared to the ones in the ITER TF CSMC. The full line is a simple power fit to the data
coils. A detailed test program has been
elaborated, which will be performed in two
phases. First, the TFMC will be tested alone and subsequently together with the LCT coil. Tests
are scheduled to take place in the second half of 2001 and during 2002.
Politecnico di Torino (POLITO) participates in these tasks in the frame of an Association
Contract. POLITO has developed the new Multi-conductor Mithrandir (M&M) code for the
simultaneous simulation of thermal-hydraulic transients in an arbitrary number of conductors,
which are thermally and hydraulically coupled together [3.7]. This novel tool has been validated
against data coming from several experiments [3.7-3.8], showing good agreement with the
experimental results, and has been applied to the analysis of the CSMC and TFMC. Concerning
the CSMC, POLITO attended the experimental campaign in April-May and June-July, with
particular reference to T cs measurements in the model coil, whose analysis through M&M is
presented in [3.9], and stability and quench tests in the CS Insert. Concerning the TFMC, a
predictive analysis of T cs measurement has been performed by using M&M, and a strategy for
helium heating has been suggested, which would lead to the initiation of the normal zone in the
conductor without quenching the joint [3.10].
3.2.3 Survey of the TF model coil geometry
The TFMC is a 36 t “race-track” shaped coil, about 4 m high and 3 m wide, scaled as compared
to the full-size ITER TF coils, and including the key technical features and manufacturing
approaches foreseen for the actual ITER TF coils. The objective of the TFMC project is to
develop the superconducting magnet technology to a level that will allow the ITER TF coils to
be built with confidence. The TFMC is going to be installed at the TOSKA facility, together with
a 27 t SS ICS interfacing the TFMC to the TOSKA facility (fig. 3.3). Due to the complexity of
the assembly operations, and in order to avoid any trial-and-error assembly process, a laser
tracking technique (fig. 3.4) has been utilized by ENEA to retrieve as-built geometry data of each
sub-assembly. The raw data have been analyzed and combined to verify the assembly procedure
and identify the corrective actions to be taken before the actual installation takes place. The
whole survey has been accomplished during three different survey campaigns, carried out at
Alstom and the Research Centre Karlsruhe.
The survey and the data analysis showed that the ICS and the coil dimensions are within the
design tolerance levels. Likewise, they showed that the positioning errors of the copper soles of
Busbars 2 in the TOSKA vessel are within 15 mm in the TOSKA coordinate system. The
resulting misalignment of the copper soles of Busbars 1 and 2 was compensated by properly
75

3. Technology Program
busbar
twisting Busbars 1 before any TFMC assembly into the
TOSKA vessel. The survey made it possible to evaluate
this misalignment and design the proper corrective
action, thus avoiding any trial-and-error process and,
subsequently, saving a considerable amount of time
and money.
TFMC
ICS
3.2.4 ITER TF casing manufacturing
optimisation (ITER Task GB8-M45)
The toroidal field coil case is a large (~18 m high and
~12 m wide) 316 SS D-shaped structure with a
rectangular section. It contains and supports the
superconductor winding pack of the toroidal field coil
for the ITER tokamak. It weighs several tons and has
to withstand heavy in- and out-of-plane magnetic loads
when energized. The superconducting cable is inserted
in the winding pack grooves and the coil case is closed
on the two faces by cover plates, that are laser-beam
welded.
Fig. 3.3 - TFMC, ICS and busbar being lowered into
TOSKA
Two non-destructive methods were developed to test
this laser welding: the eddy-current and the infrared
thermography techniques.
The eddy-current technique was successfully applied
to control the cover-plate welding of the five pancakes,
by using a computerized semi-automatic x-
y scanning system that produces an
immediate C-scan and perspective color
images (fig. 3.5a,b).
The activity on the infrared thermography
technique started with a feasibility study,
performed in the frame of the M45
deliverable #3 task. Considering the good
results obtained, the engineering design of
a suitable equipment for the control of the
laser welded structural joints of the TF
cover plate was carried out by using a
remote thermal source and developing a
computer image processing.
The image processing is performed by
analysing the thermal map generated by
Fig. 3.4 - The laser tracker and the ICS
the laser-heating of the plate. The analysis
is made in real-time, during the relative
motion of the heating source and the stainless steel plate (fig. 3.6). A numerical simulation to
validate the algorithm used in the calculation of the defect dimension and position was also
carried out.
3.2.5 Numerical simulations of welds of thick steel relevant for ITER TF case
This work, developed in the frame of the project activities "Fabrication Development and Q.A.
for TF Coil Case" for ITER machine, concerned in particular the numerical evaluation of
76

3. Technology Program
deformations and residual stresses due to the different welding phases, which are of a crucial
importance for the definition of the manufacturing requirements of ITER TF case.
This activity, being very challenging both for the difficulties in the modeling welding/processes and
the dimensions of the
TF Case (almost 240
mm thick), required
several problems to be
coped with, namely:
a) b)
• To evaluate the
possibility of
numerically simulating
welds with filler
material, through the
use of a computational
thermo-structural code,
based on Finite Element
Method (FEM) (in
particular, the
ABAQUS/S code).
• To carry out
numerical simulations,
by considering
experimental reference
models of simple
geometries and reduced
dimensions, realized “ad
hoc”.
Fig. 3.5 - Map of the two sides of the winding case in which the repaired
defected welding are red highlighted
Fig. 3.6 – Equipment scheme
77

3. Technology Program
Initial FEM Configuration
Distortion comparison between Experience
and FE Calculation:
Transverse Shrinkage
Exp = 6 mm; Calc. = 5 mm
Longitudinal Shrinkage (L=80 mm)
Exp = 0,04 mm; Calc .= 0,02 mm
(α) Global Angular distortion
Exp = 17.7°; Calc. = 16.3°
Final FEM Configuration
Fig. 3.7 – Coupon A. Theoretical analysis compared with experimental results
Initial FEM Configuration
Distortion comparison between Experience
and FE Calculation:
(α) Global Angular distortion
Exp = 2.6°, Calc. = 2.2°
Final FEM Configuration
Fig. 3.8 – Coupon B. Theoretical analysis compared with experimental results
78

3. Technology Program
• To consider different welding process types for the numerical simulations, in order to better
choose the welding process to be used for the fabrication of the actual components of TF coil cases.
Initially, cold–wire Tungsten Inert Gas (TIG) welding processes had been considered for the
numerical simulation. Subsequently, on the basis of the experimental results, another welding
process with filler material, that is to say the SAW process, has been utilized and therefore
numerically simulated.
The results obtained from the FEM models have been compared with a number of “ad hoc”
welding specimen prepared by ENEA Faenza Labs and Belleli (figs. 3.7,3.8).
The conclusions of this part of studies, experimentations and calculations are as follows [3.11]:
• It is possible to simulate numerically TIG welding processes also for experimental models of
significant dimensions. In fact, the simulation results are in good agreement with the
experimental ones.
• The calculation procedures, singled out and set up for the simulations of TIG welding
processes, are absolutely general and can be well applied, with minor changes, to the simulation
of other welding processes with filler material, see SAW process. However, these procedures are
too expensive as far as the computer disk space request and the computing times are concerned.
• It has been possible to find and set up new calculation procedures, which have allowed the
required mechanical response to be obtained in sufficient agreement with the experience
(maximum percentage difference for the final displacements of the lateral unconstrained
boundary specimen less than 8%), by using much shorter computing times.
However, also the latter and new methodology, if applied to models of much larger dimensions
(as, for example, ITER real coil case components), results in excessive expense, and is therefore
not practicable.
As a consequence, the activities will continue, preparing equivalent models which, starting from
those already set up, could speed up the computation time. Very promising methods are already
being envisaged.
3.2.6 Development of new calculation codes for cable-in-conduit conductors
In the framework of the EU, ENEA is the Italian co-ordinator for the development of a new
calculation code, planned to include all the Cable in Conduit Conductor (CICC) relevant physics
(Super Model) in order to obtain a predictive calculation tool on the behaviour of force-flow
cooled superconducting magnets. Besides the thermo-hydraulics of the superconducting cables,
(already taken into account by existing codes), such a code has to include, also the electromagnetic
description of conductor, joints and terminations. Being the conductor formed by
several hundred strands, each with its own inductance and resistance if quenched, the model is
quite complicated. Several contributions are coming from different institutions to this purpose.
University of Bologna, Dipartimento di Ingegneria Elettrica, is in charge of the electromagnetic
modelization of the conductor; University of Udine of modeling the electrical joints and
terminations; University of Padua of mechanical modelization of the strands; and Politecnico di
Torino is in charge of coupling the set of models with existing thermo-hydraulic codes.
ENEA co-ordinates the work and takes care of the matching between codes and experimental
results. In this framework a new experiment, to be carried-on at our labs, aimed at measuring the
magnet minimum quench energy as a function of the current unbalance, is presently being
considered.
The code will be ready next year.
79

3. Technology Program
3.2.7 Development of NbTi conductors for the poloidal field coils of ITER
(ITER Task M50 and EFDA Task Two-T405/1)
Part of the NbTi strand to be used in the framework of the EU program has been manufactured
by the Italian Company Europa Metalli. The basic layout is similar to that of the strand
developed for the magnets of the CERN accelerator Large Hadron Collider.
The surface of the strand has been coated with a Ni layer. In fact, comparative measurements
made at Commissariat à l’Energie Atomique (CEA) and University of Twente on sub-size 36
strand CICCs have shown that Ni coating is the best candidate to reduce coupling losses, while
allowing current redistribution to take place inside the cable.
The main characteristics of the strand are given in table 3.I.
Two cables, each made up of 108 strands, have been manufactured at Brugg Kabel (Switzerland) in
lengths of about 10 m each. The strands used are the EM Ni–coated, and the Alstom strand with
internal CuNi barriers. The cables have been inserted into AISI 304 SS tubes and shipped to Europa
Metalli, where reduction to final dimensions is foreseen for spring 2001. The conductors will be used
at CEA for manufacturing and testing of sub-size joints.
A contract has been assigned to Europa Metalli to cover all the remaining cabling and jacketing
of sub-size conductors. In particular, two 36–strand CICCs, made with EM and Alstom strand,
will be manufactured in 2001 in lengths of 100 m each. The conductors will be used to
manufacture two test solenoids for testing pulsed operation, stability, alternating current (AC)
losses and current distribution.
The conceptual design of the test solenoids has been completed. Special attention has been paid
to the design of the terminations, in order to allow control and measurement of current imbalance
to be performed within the sub-elements of the superconducting cable.
In the framework of this Task, ENEA is also in charge of the preparation of a sample to be tested
at the EU SULTAN facility, using the full size NbTi conductor, manufactured for the busbars of
the ITER TFMC. The spare conductor will be released only after installation of the TFMC in the
TOSKA facility, foreseen for the first half of 2001.
3.2.8 Stability and quench propagation on an NbTi CIC test conductor
This experiment belongs to a family of experiments aimed at the following:
• Testing performances of sub-size superconducting magnets, similar to those that will be used
for the next generation of fusion machines. In particular, the experiment is aimed at studying
stability, viz. sensitivity, of a superconducting magnet to externally-induced electro-magnetic
(em) disturbances;
Table 3.I - EM NbTi strand for the EU conductor R&D
Strand diameter
0.81 mm
Cu non Cu 1.90
Twist pitch
8 mm
Number of filaments 6534
Average filament diameter 6 µm
Thickness of Ni coating 1.0 µm
Guaranteed critical current density (J c ) at 6T, 4.2K 2130 A/mm2
Typical J c at 6T, 4.2K 2240 A/mm 2
• Deeper understanding of the physics
of the superconducting composite
conductor, of the thermo-hydraulic and
em phenomena taking place in the
CICC used for fusion magnets;
• Cross-checking of the experimental
results with those obtained from
calculation codes, in order to validate
them as predictive tools.
The results obtained from the stability
and quench propagation experiment,
80

3. Technology Program
nick-named SEX, for Stability EXperiment (fig. 3.9), were
analyzed during the first half of the year. The typical
experimental conditions were obtained by producing em
disturbances, and measuring their effect on the
superconducting magnets. Above a given threshold, defined
as the stability margin, this leads the magnet to quench.
Several experimental runs have been made to explore the
different guidelines. The first [3.12] starts by discussing
the conditions under which such experiments, aimed at
simulating the working conditions foreseen in large-scale
magnets, are really representative. From the analysis of the
experimental data and their comparison with the results of
the calculation codes, it is possible to assess that one of the
main requirements historically imposed on experiments of
this type is over-conservative. In detail, the requirement
imposed to simulate a long-wound, full-size conductor is
that a minimum sub-size conductor length should be
present. This is so that the travelling time of a sound wave
in pressurised liquid helium be longer than the time
duration of the external em disturbance. By analysing the
physical behaviour of the system, it is possible to assess
that the important characteristic is the comparison between
the cooling helium pressure relief time constant and the em
disturbance time. If the latter is shorter, the conductor, acts
as a long-length conductor. This is representative of what
happens in large-scale conductors.
Fig. 3.9 – Overview of the stability experiment
Once assessed whether the conductor is
relevant, the main point is how to measure
the energy released inside the cable by the
external em disturbance. Usually, it is
measured by means of calorimetric
techniques, where the helium temperature
increase and mass flow are monitored. A
new technique has been proposed, where the
helium temperature increase is directly
measured on the conductor and the energy
absorbed simply calculated as soon as the
thermal equilibrium is reached.
The second [3.13] guideline is aimed at the
comparison between the experimental
results and the thermo-hydraulic calculation
code Gandalf. The comparison has been
made mainly by obtaining the same
minimum quench energy, viz. the minimum
energy that, inductively given to the
conductor by the em disturbance, causes the
magnet to quench. Then, quench
propagation speeds (fig. 3.10) in different
experimental conditions, critical current and
Fig. 3.10 – Conductor quench propagation. Comparison
between measurements and calculations
propagation of a helium heat step (fig. 3.11) have been compared. Good agreement has been
found in all these comparisons.
81

3. Technology Program
In the second half of the year, a new
measurements campaign SEX2 has been
defined and prepared, with the following
objectives:
• To study the magnet behaviour under
different em disturbances (in intensity,
shape and duration);
• To investigate the transport current effect
on the em-induced losses;
• To measure the dependence of the
minimum quench energy on the helium
flow;
• To compare calorimetric and electric
calibration techniques.
Fig. 3.11 - Comparison between measured and calculated
results during a heat step
Fig. 3.12 - ITER blanket earth strap
Fig. 3.13 - ITER blanket strap: test facility
The experiment has been cooled down and
the results are expected early in 2001.
3.3. VACUUM VESSEL AND
SHIELD
3.3.1 ITER blanket modules: testing
on the earth strap connections
(EFDA task Two–BM/STRAP)
During plasma shots in the ITER Tokamak,
very severe electromagnetic stresses (due to
I=250 kA, lasting about 0.3 s, under B=9 T)
may affect the blanket earth connections to
the vacuum vessel (fig. 3.12). The FTU
power supply system is probably the only
one featuring the electromagnetic
performances needed to carry out the tests
required. For this reason, a specific EFDA
task (Two–BM/STRAP) has been carried
out, consisting of the following:
• Defining the design of the strap
connection, also by performing strap
detailed stress analyses and optimization of
the manufacturing procedures;
• Checking the strap capability of
withstanding the stresses foreseen, by
performing proper thermomechanical
fatigue tests and electromagnetic stress
tests on an actual strap connection.
The whole activity has been successfully
completed by December 2000 (fig. 3.13). A
new task to design, manufacture and test
the new strap connection has been entrusted
by EFDA to ENEA for the year 2001.
82

3. Technology Program
3.3.2 Design of the Plasma–Facing Component (PFC) for the divertor of ITER
FEAT (EFDA Contract/00-544) and EM analyses of shielding blanket for ITER-
FEAT design options, during plasma disruptions (EFDA Contract/00–570)
In figure 3.14, the Electromagnetic Model (EM) developed within the frame of the EFDA
Contracts 99-504 and 00-570 is shown, to compare the EM loads in the two to main design
options for the ITER FEAT.
The EM analyses performed under the same reference radial and parallel field excitations (of
0.25 T and 1.7 T in 20 ms respectively) have shown that the loads on option A are about 20%
lower than the corresponding loads on option B.
To reduce the EM loads on the in-vessel components of ITER-FEAT (shielding modules, divertor
cassette, first wall panels, divertor targets and dome), the design complexity of a large number
of components has been greatly increased through the introduction of various slots (see
fig. 3.14). The real working condition of these components would imply including the whole
plasma region, as well as the vacuum vessel and the nearest conducting structures in the EM
model. Therefore, modeling all the main geometrical features of the particular component under
investigation would require a prohibitive number of Degrees of Freedom (DOF) in the model.
For this reason, such models could not be used for any sensitivity analysis of the effects on the
EM loads of the geometrical features of the various design options. To overcome this difficulty,
a zooming procedure has been now developed in Frascati, which allows for an agile investigation
of complexes geometries even in the presence of phenomena involving wide regions, and this
without loosing any relevant detail in the description of excitation and geometry.
The zooming approach is performed in two steps.
In the first step, the whole region involved in the electromagnetic event is analysed by a 2-D
model, which allows to determine a set of excitations surrounding the much more restricted
region that includes the component under investigation. The excitations found by this model will
reproduce the exact EM conditions of the whole phenomena inside the zoomed region.
In a second step, the 3-D detailed analysis of the zoomed region is performed by using the
excitations determined in the previous step. In figure 3.15, the application of the zooming
Fig. 3.14 – EM model of the two main design options for the ITER-FEAT shield blanket. The option A
is on the left and the option B is on right. To reduce the DOF number of the two models, only the upper
half of the modules has been modeled. Some parts of the two models have been “blanketed”, or made
transparent to show the internal geometrical complexity of the models
83

3. Technology Program
a) b)
Fig. 3.15 – a) 2-D model of ITER-FEAT vacuum vessel showing the inboard target region surrounded
by a superconductor array (in magenta). This array of filaments cancels the field variation inside the
surrounded region, thus reproducing any EM excitation occurring outside the region. b) the
correspondent 3-D model used for the analysis of the inboard divertor target is shown
approach for the analysis of the ITER-FEAT inboard target is shown. Since a complete run of the
3-D model of this reduced region lasts about 30’ on a 700 MHz PC, a lot of runs have been
performed, thus allowing the selection of the most appropriate design options.
3.4 FIRST WALL AND DIVERTOR
3.4.1 Runaway electrons on ITER PFCs (EFDA Contract /00-520)
This activity started in July 2000 in the framework of the Contract EFDA/00-520. The aim of the
contract is the assessment of the thermal effects of runaway electrons on ITER PFCs.
Runaway energy deposition profiles have been estimated by using the FLUKA Montecarlo code.
The profiles have then been used as an input for the temperature pattern evaluation by means of
the finite element heat conduction code ANSYS. Five first-wall geometries have been
investigated, differing from one another in material, pitch between cooling tubes, armour
thickness etc. In one geometry, poloidal limiters protecting the inner first wall have also been
considered. Two different runaway energies have been tested (viz.10 and 50 MeV), as well as
two different energy deposition times: (viz.10 and 100 ms).
Energy deposition profiles reflecting the dependence of the electron energy loss mechanism on
the material have been provided by the FLUKA code. Presently, the optimization of the mesh
used by the ANSYS code is being performed: possible damages to the cooling tube walls as well
as metallic armour melting will be accurately investigated.
3.4.2 Deuterium desorption measurements of W-1% La 2 O 3 (EU Task DV 7A-
ITER Task EU-T438)
W-1% La 2 O 3 is a candidate material to be used for the construction of PFC of the ITER
machine. Deuterium trapping and release behaviours in tungsten material are important for
determining the tritium inventory and controlling the fuel recycling during the plasma
discharges.
84

3. Technology Program
In order to estimate the deuterium retention, an experimental activity was carried out, by measuring
the thermal desorption of samples previously exposed to deuterium plasma (see table 3.II).
The specimens were bombarded in a plasma generator which has an ion flux density and energies
relevant for tokamak machines.
The sample surface exposed was 15×15 mm2.
After the exposure, thermal desorption measurements of samples were carried out by measuring
the outgasing flow rate of the gases released, as a function of temperature and time.
The measuring apparatus, designed to work in Ultra-High Vacuum (UHV) conditions, allows
outgasing measurements to be made by means of a quadrupole mass spectrometer (1-200 AMU),
thus exploiting the known conductance method. Quantitative flow rates of the gases released are
estimated by measuring total and partial pressures and considering the effective pumping speed
in the measuring chamber through the known conductance, obtained by a diaphragm of the
known diameter.
Samples were heated from room temperature up to 900°C with a ramp of 15°C/min and then,
once the maximum temperature was reached, it was kept constant for further 60 min. For the
purpose of heating the sample, the RF inductive system was chosen, in order to avoid heating of
any other component in the vacuum chamber.
The results of thermal desorption measurements, in terms of total amount of released deuterium
and desorbed/implanted ratio of deuterium atoms, are shown in table 3.III.
Both D 2 and HD molecules are desorbed from the samples, as shown in a typical diagram (see
Table 3.II - Thermal desorption of samples previously exposed to
deuterium plasma
Sample No 2 3 6 7
Ion flux density (m -2 s -1 ) 7,9×10 21 7,6×10 21 7,5×10 21 7,5×10 21
Electron temp. (eV) 5 10 10 10
Exposure time (h) 4 4 4 4
Target temp. (°C) 600 500 500 650
Electron density (m -3 ) 3,7×10 17 4,2×10 17 4×10 17 4×10 17
Bias voltage (V) 200 74 70 100
Table 3.III - Results of thermal desorption measurements
Sample Amount of desorbed D Specific amount of Desorbed/Implanted
No atoms (D) desorbed D atoms (D/m 2 ) ratio (D/D)
2 7×1017 3,1×1021 2,73×10-5
3 2,1×10 17 9,3×10 20 8,5×10 -6
6 3×10 17 1,3×10 21 1,2×10 -5
7 2,4×10 17 1,1×10 21 1×10 -5
85

3. Technology Program
fig. 3.16). To estimate the total amount
of D desorption, the sum of D 2 and HD
deuterium atoms have been taken into
account.
Within a factor of three, the amount of
deuterium is the same for all the samples,
as it was expected, since the ion flux
density and exposure time are the same.
The very low ratio of desorbed and
implanted deuterium, about 10-5, can be
explained on the basis of the very high
(close to 1) re-emission coefficient
during implantation for high fluences
and high target temperature.
Fig. 3.16 - Flow rate vs time and temperature of sample 3
Fig. 3.17 - Flow rate vs time and temperature of sample 2
By comparing the desorption profile of
sample 3 (see fig. 3.16), implanted at
500°C, with sample 2 (see fig. 3.17),
implanted at 600°C, it can be seen that,
in the first case, three desorption peaks
are visible (180°C, 430°C and 700°C
both for D 2 and HD) while, in the
second case, only two peaks, namely at
700°C and 800°C for D 2 , and 700°C and
900°C for HD, are shown.
This rather complex profile of desorbed
deuterium versus temperature can be
better understood if the implantation
temperature is taken into account.
Samples implanted at lower temperature
show desorption peaks at lower
temperature as compared to samples
implanted at higher temperature since, in
this case, part of deuterium is desorbed
during the implantation phase.
3.4.3 Design of a welded divertor cassette (NET Contract/98-488)
Among the components of a fusion machine, the divertor is one of the most challenging. This
component, aimed at reducing impurity in the plasma, has the task of withstanding a very high
surface heat load. At the same time, it has to shield the Vacuum Vessel (VV) and the Toroidal
Field Coils (TFC). Because of its double target, it consists of two parts: the High Heat Flux
Components (HHFC), which works as an actively cooled thermal shield, and a cassette, which
is a massive supporting structure made of 316 LN steel. It has to resist the high stress expected,
due to the differential thermal dilatation and the forces induced by the disruption events.
A new cassette concept, based on a welded box structure made of stainless steel plates, has been
proposed for ITER FEAT (fig. 3.18). In the frame of a NET contract (NET 98-488) [3.14, 3.15],
ENEA has developed the design cassette sample, for the design of which a team of several
professionals from three different factories have been co-ordinated by ENEA.
86

3. Technology Program
Three cassettes can be found in each sector of ITER
FEAT, for a total of 54 cassettes in the entire machine.
There is 1 cm gap between the different cassettes.
Figure 3.18 shows the poloidal section of the cassette
and the HHFC.
Thermal-hydraulics, neutronics, electromagnetic stress,
manufacturing and cost analyses have been performed.
The total water mass flow is below 1000 kg/s. The
water velocity in the vertical targets does not exceed 12
m/s. The pressure drop is limited (1.11 MPa), with a
temperature inlet of 100˚C and an outlet of 142˚C.
Therefore, these cooling layout features are compatible
with the requirements.
The cassette gives enough shielding to the TFC system.
The thickness considered of 25 cm, with three steel plates
5 cm thick with two water layers (5 cm thick) interposed,
fulfils the requirements with a good safety margin. Some
concern may arise from the streaming through the ports
facing the cassette. This is a problem which concerns port
problems, rather than the cassette itself.
A large amount of stress analyses have been performed
Fig. 3.18 - Transparent view of divertor cassette
to study different operation conditions and loads
applied. In the normal operating conditions, viz. the reference nominal case with two fixed
hinges on the outboard and two sliding hinges on the inboard, the largest radial displacement
inward is caused by thermal expansion (18 mm). Some peak stresses can be locally reduced
through proper optimisation and stiffening of the inboard ribs. These are close to the limit
allowed (20 mm). The EM loads enhance the radial inward displacement up to a value (34 mm)
beyond the limit. Peak stresses under EM loads are localised in the surroundings of the HHFCs
attachment. The heavy loads suffered by the fingers (90 t in the outboard) can be withstood by
increasing the number of fingers, or their poloidal length.
The work showed that all the design requirements are fulfilled in normal operational conditions.
The halo currents induce deformations close to the allowable limits. Therefore, some
reinforcement of the cassette or some modifications to the VV attachment system should be
studied, in order to reduce the displacement.
3.4.4 Neutron diffraction study of high-temperature stresses in brazed divertor
mockups
This activity has been carried out in the frame of the Underlying Technology Program. Its final
goal is to develop a tool aimed at forecasting the residual stress pattern in brazed components.
As a matter of fact, neutron diffraction provides for stress field values experimentally determined
in the bulk of the investigated samples, which are necessary to validate the theoretical
predictions obtained by numerical tools, such as FEM calculations.
The investigated sample was a mockup (23×23×8 mm3), obtained by brazing a Glidcop and a W
platelet by an intermediate Cu interlayer and a TiCuAg filler layer. Both the Glidcop and W were
thermally annealed before brazing, in order to release the stresses arising from fabrication or
machining, and to get genuine information on the stress field associated to the brazing process.
The neutron diffraction measurements were carried out in 1999 at the High Flux Reactor of
Institut Laue Langevin in Grenoble, using the high-precision D1A diffractometer equipped with
an infrared furnace, where the sample is homogeneously heated and rotated to determine the
87

3. Technology Program
stress tensor. The measurements were carried out at
different temperatures, between room temperature
and 500°C. The unstrained lattice parameters of both
Glidcop and W were determined in this same
temperature range on two unstrained samples of
these two materials.
Fig. 3.19 – Normal stress on W. Note that the brazing is
at 10 mm from the sample edge
Although the experimental results gave a reliable
measurement of the stress field throughout the
sample, the stress patterns obtained were difficult to
be interpreted according to the hypothesis of a
homogeneous lattice, as it can be seen in the annexed
graphs (fig. 3.19, 3.20). In these graphs, the normal
stress in the W and Glidcop platelets is shown at
different temperatures. It can be easily seen that the
trend of the stresses is not monotonic with the
distance from the brazing (at 10 mm from the edge).
This behaviour cannot be easily explained by
theoretical considerations, therefore in 2001 new
tests will be performed on a slightly different
sample, viz. a sample with no intermediate Cu
interlayer. These tests will hopefully clarify the
relevance of this interlayer for the behaviour of the
whole brazed structure.
Fig. 3.20 – Normal stresses on Cu - Note that the
brazing is at 10 mm from the sample edge
vertical target, respectively.
3.4.5 Mechanical and electrical tests on the
attachment keys for the divertor vertical
target
The vertical target attachment system consists of two
so-called “dumbell” keys. Each dumbell key
consists of two cylindrical hinges joined by a
ligament. Each cylindrical key is formed by two
cylindrical wedges sliding on a tilted surface of the
ligament. The two cylindrical wedges are pushed
against the cassette holes and ligament surface by
means of bolts.
By screwing the bolts, it is possible to simply put the
wedge in contact with the holes, thus leaving the key
free to rotate, or locking it, if necessary, in order to
avoid any rotation.
In order to test the keys, they have been inserted into
two steel mockups simulating the divertor and the
The choice of the candidate material for the vertical target can be summarised as follows:
• Wedges: Bronzal 7
• Ligament: Stainless Steel ASTM 453 Gr 660
The mockups, used to test the “dumbell” keys mechanically are made of stainless steel very
similar to that used to build the divertor and the vertical target.
A 100 kN MTS machine has been used for the mechanical tests. The test section, suitably
88

3. Technology Program
equipped, has been inserted into a vacuum chamber. The chamber was supported by the load cell
of the machine, while the mobile shaft was supported through a set of bellows by one of the two
steel mockups connected by the dumbell keys (fig. 3.21).
The electrical discharge tests have been carried out at CESI in Milan, connecting the middle of
the two mockups of the test sections, by means of two electrodes, to the secondary of a mediumlow
137 MVA voltage transformer. The primary was fed by the network to 23 kV (600 MVA).
The results are shown in fig. 3.22 and 3.23, where it is possible to notice the following:
• in the first case (slow speed of load application), the load corresponding to the first
displacement perceivable (s 1 =0.6 mm) has been of: L 1 =6.88 kN.
• in the second case, the first movement
(s 2 =0.18 mm) has been obtained with a load
L 2 =4.37 kN.
Therefore, the test has been articulated as
follows:
• Performance of a series of electrical
discharges with gradually increasing current
from 5000A until a maximum of 75000A,
with discharge times of the order of the
second;
• Relief of the electrical discharge
parameters and disassembling of the system,
to control the wedges surface of the
connection at each step;
• Therefore, the testing phases have had the
following pace: 5000A - 10000A - 20000A -
30000A - 50000A - 75000A;
• The 75000A test has been repeated ten
times.
Fig. 3.21 - Mockup inside the vacuum chamber
Fig. 3.22 -Low-speed mechanical test
Fig. 3.23 – High-speed mechanical test
89

3. Technology Program
Table 3.IV - Screwing torque after
discharge tests
1 5000 A ≤ 15Nm
1 10000 A ≤ 15 Nm
1 15000 A ≤ 20 Nm
1 20000 A ≤ 20 Nm
1 30000 A ≤ 20 Nm
1 50000 A ≤ 18 Nm
10 75000 A ≤ 10 Nm
The key locking has been
carried out with a 25 Nm
couple before every test. A
decrease in the couple has
taken place, according to the
values of table 3.IV. The
diagram relating to one
individual discharge is shown
in fig. 3.24.
Inspections of the bronze
wedges have been carried out after every single discharge until 50 kA, and after ten discharges
at 75 kA. No change has been noticed within 50 kA; after ten discharges at 75 kA, a light
oxidation halo has appeared in the zone of the dumbell keys corresponding to the electrodes.
The results of the mechanical test show that the sliding of the keys occurs when a torque is
applied of about ten times less than the one originated by halo current and dead weight loads.
This implies that the vertical target cannot be held by the friction solely, but must be fixed on the
dumbell key.
The electrical tests show that no damage occurs at up to 50 kA (reference condition),
corresponding to a peak of ~112 kA, in spite of a reduction of the unscrewing torque (~30%).
The mean value of the contact pressure in the test conditions is ~5 MPa, enough to guarantee a
safe current flow.
During the test performed at 75 kA (~173 kA peak value), a remarkable vibration occurred and
some arching took place in the region where sharpened edges are present. This condition can be
taken as the upper limit allowable, unless a system to increase the screwing torque is provided.
In the final analysis it can be stated that the dumbell keys are capable of carrying safely the
expected halo current fraction.
3.4.6 Non-destructive testing of permanent components with calibrated defects
(ITER task T222–14)
This activity is part of a larger project in which a vertical target mockup of the ITER tokamak,
manufactured by Metallwerk Plansee, was used for a “Round Robin test” aimed at qualifying the
Non Destructive Testing (NDT) methods for divertor and PFCs. This activity started in 1998 (see
Progress Report 1998) and was planned in several phases, as follows:
1. Manufacturing of the mockup with calibrated defects;
Fig. 3.24 - 75 kA discharge test
90

3. Technology Program
2. NDT performed by the manufacturer and the associations involved in the project;
3. Thermal fatigue testing of the component in the JUDITH e-beam facility;
4. NDT after the thermal fatigue testing, performed by the associations involved in the project;
5. Destructive Examination (DE) to confirm position, size and shape of the present and the
calibrated defects.
The ENEA labs were involved in phases 2, 4 and 5.
The component has two zones, each with two different armor materials: the almost linear zone,
with a copper alloy tube, protected by a Carbon Fiber Composite (CFC) ‘tile’, acting as an armor
material; the zone with two radii of curvature, which is a square copper alloy tube, protected by
several cubic tungsten tiles on the plasma-facing side.
The CFC-copper monoblock tube was inspected from inside; the tungsten-copper interface, from
the tungsten external face.
The results of the NDTs, obtained before and after the fatigue tests, were compared in order to
determine possible size increases, due to them. Figure 3.25 shows the results obtained by
ultrasonic NDT for CFC sections of the mockup, while fig. 3.26 reports the results obtained for
the W tiles part.
Destructive examination took place after HHF testing, in order to confirm NDT results. Figure
3.27 shows two typical calibrated defects for the two different zones (W and CFC tiles).
The best NDT results have been obtained by using the ultrasonic technique: the IR-technique is
not so straightforward for NDT inspection of this type of components; tomography is timeconsuming
and not so sharp as the Ultra Sonic (US); it can be effectively used only if coupled
to other techniques.
Expansion of defects, due to the fatigue testing, has been found only in the transition zone CFC//W.
Therefore, more attention should be paid during design/manufacturing of the component.
Fig. 3.25 - NDT by ultrasonic technique- comparison before/after HHF testing of the CFC tiles
91

3. Technology Program
The presence of defects (be they
artificial or natural) found by the ND
techniques has been confirmed by
destructive examina-tions.
There is good correspondence between
the size of the defect found by means
of the US technique and the actual
dimensions found through final
destructive examination.
3.4.7 Thermal fatigue testing of
vertical target mockups
manufactured by diffusion
bonding (ITER task DV1/01)
Fig. 3.26 - NDT by ultrasonic technique-comparison before/after
HHF testing of the W zone
The aim of this activity was the
thermal fatigue testing of mockups of
the vertical target of the ITER
machine, manufactured in the
framework of task T232.6.
The manufacturing technique used was the axial diffusion bonding, in which the armour material
(i.e. W, CFC and Be) is faced to the copper alloy heat-sink; bonding is obtained by applying the
necessary axial load, temperature and time. A ductile material is usually interposed between the
tile and the heat-sink, in order to keep residual thermal stresses as low as possible.
Six samples were manufactured: two with CFC armour tiles, two with W armour tiles and two
with Be armour tiles. Testing was performed in the Russian e-beam facility TSEFEY (St.
Petersburg).
For each mockup, the thermal fatigue-testing plan foresees 1000 cycles, starting from
a) b)
Fig. 3.27 – Typical calibrated defects for the two diverse zones CFC and W tiles
92

3. Technology Program
0.5 MW/m 2 to reach the component failure
(fig. 3.28). A screening test was also performed,
before each 1000 cycle phase.
The best results obtained for each type of armour
material were the following:
• CFC armour tiles 1000 cycles at 7 MW/m 2
• W armour tiles 1000 cycles at 9 MW/m 2
• Be armour tiles 1000 cycles at 4 MW/m2.
This test confirmed the good reliability of this
joining technique, when applied to manufacturing
of plasma facing components, which have to
withstand power up to 7 MW/m2.
Fig. 3.28 - The mock-ups after the fatigue testing
3.5 REMOTE HANDLING
3.5.1 Overview of the ENEA contribution to the implementation of the ITER
L-7 project
The purpose of the ITER large R&D L-7 project is to allow for demonstration of basic feasibility
of divertor replacement and refurbishment operations. Two important facilities, viz. the divertor
test platform and the divertor refurbishment platform, have been built by ENEA to support
experimental programs on this project. Significant changes have occurred between the 1998
ITER and the new ITER-FEAT designs.
The divertor handling scheme has not changed, but has been modified to suit the new geometry.
The total number of cassettes has been reduced from 60 to 54, as well as their weight (now 12 t)
and the room available between cassette and vessel floor (70 mm). This modification has led to
the adoption of a Cantilever Multifunctional Mover (CMM) for all radial transport operations.
The Cassette Toroidal Mover concept (CTM), however, remains unchanged.
These changes do not impair the results obtained so far, but require upgrading of both Divertor Test
Platform (DTP) and Divertor Refurbishment Platform (DRP): major work for future programmes.
3.5.2 Divertor test platform (Tasks T308/1 and TW0/DTP01)
The main objectives of the divertor test platform are the validation and optimisation of the
divertor maintenance scenario.
The major achievements of the handling operations performed last year include the following:
In previous tests, the CTM (fig. 3.29) had proven capable of handling any realistic cassette
misalignment (up to 10 mm in a toroidal direction) and accommodating horizontal gaps between
adjacent VV rails (up to 2 mm). Analysis and further testing allowed for the extrapolation of
these results to other relevant situations, such as: i) presence of vertical gaps; ii) occurrence of
similar gaps in other movers (Toroidal Radial Carrier (TRC), Central Cassette Carrier (CCC),
etc.); iii) undulations in the rails instead of (or at the same time as) steps and gaps.
The CCC tests pointed out a mechanical problem at the cassette-VV interface. This problem was
detected and solved with the aid of the test equipment and data charting, but further investigation
is still required. Automatic procedures for CCC movements were also implemented and tested.
The most significant data logged, and the main abnormal situations occurring during the tests
93

3. Technology Program
have been retrieved, analysed, and entered into
a logbook. This will enable the construction of
a database to back up the final design of the
Remote Handling External (RHE) devices that
will operate in ITER.
An important part of activities was also
performed on the auxiliary devices.
The final commissioning of the Duct Equipment
(DE) was completed in Sept. 98 but, soon after,
a failure occurred at the central motion
controller board, thus blocking any further
operations. This failure has been identified and
repaired in the current year, and DE testing has
thus been resumed.
The Bore Tooling System (BTS) (supplied by
CEA/COMEX Nucléaire) was partially
Fig. 3.29 - Limit condition testing on the CTM
commissioned in January 1999. The supplier
has carried out additional commissioning in
June 2000. Trials have been performed with BTS head insertion/extraction sequences into/out of
a pipe mockup, welding cycles and ultrasonic inspections. Some possible improvements both to
mechanics and software have been identified, to reduce risk of jamming and operation failures.
A complete test campaign is still to be carried out, to verify the basic feasibility of operations
with the existing hardware. More recently, however, the design has changed to incorporate
curved cooling pipes, thus giving rise to a new task with CEA and the Italian company RTM.
The MAESTRO Manipulator device, which is currently being developed by CEA, was due to be
delivered during the year for installation on board the CTM, but its delivery has been postponed
to 2001. The efforts of the DTP team have focused on the development of the infrastructure
needed to integrate MAESTRO into the DTP.
Two more items concerning optimisation of the operating procedures have to be mentioned
among the year’s activities, namely:
• Several important elements of the Supervisory Control and Data Acquisition system
(SCADA), used for coordination of the different movers, have been finalised. The remaining
components are still under development;
• Development of the computer kinematics simulator system (TELEGRIP) is underway.
3.5.3 Divertor refurbishment platform (Tasks T308/5, T308/8, and TW0/DRP01)
The DRP was set up in Brasimone, to demonstrate the feasibility of refurbishment operations on
the ITER divertor cassette under simulated hot-cell conditions, and to optimise the relevant
equipment design.
This facility comprises a hot-cell area with handling equipment (bridge crane, transport trolleys,
manipulators), a 3-D metrology system, specialised PFC handling and replacement tools,
cassette component mockups, viewing systems, data acquisition system, and a control room.
The first handling trials, which started in 1998, demonstrated feasibility of the main critical
tasks, by means of a basic hands-off system with direct viewing and simple camera support. It
soon became clear, however, that substantial enhancements were needed to achieve an entirely
remote handling environment with segregated hot-cell and operator control areas. Furthermore,
94

3. Technology Program
this environment needed to be updated to the latest attachment concept (Multilink), whose
development as an alternative to the original shear-key concept was about to be started.
The upgrading has concerned the following main items:
Separation of operator and hot-cell areas. The areas have been separated by a series of opaque
screens of adequate dimensions, to completely obscure the hot-cell area from any direct viewing
by the operators.
Improvements to the remote viewing system. General task viewing gives a closer view of a
specific task (viz., picking up a tool), and relies on two new Sony cameras, controlled by a PC
via a standard RS232 interface. Detailed task-viewing (viz., inserting a pin in a hole) uses four
miniature cameras, which are either directly mounted on the tools, or handled by one arm of the
light manipulator. Viewing support is now provided by five general-purpose PAL monitors, an
ELCA video matrix for routing functions, and a touch-screen PC.
Data acquisition. Since upgrading of the DRP has coincided with the shift from the Shear Key
to the Multilink concept, a new LabView operator display has been designed and implemented
to reflect this new environment.
Mechanical RH interfaces. Special mechanical inter-faces have been designed and manufactured
to optimise handling of the DRP components through manipulators.
Safety interlocks. Electrical interlocks have been incorporated to prevent any accidental release
of tools (with weights of up to 100kg) by the heavy manipulator. First investigations on the
Multilink attachment (fig. 3.30) have also been carried out throughout the year. As previously
reported, this new concept involves a series of interconnecting links, secured to the target and
cassette by internally expansible hollow link pins. Expansion is achieved by means of a water
hydraulic tool, and removal by drilling out the central core in order to let a thin shell be easily
pushed out of the hole.
The reference geometry had been laid down during a previous design work; subsequent
experimental investigations led to the following:
• Values of clearances needed to achieve prescribed joint characteristics (viz. uniform contact
between pin and holes, rotation of the link around the pin instead of pin rotation in cassette-side
hole). Tests showed that, for 14 mm thick plates, pin/hole diametrical clearances 0.50±0.02 mm
and 0.45±0.02 mm resulted in post-expansion torques of 15 Nm and 55 Nm.
• Good behaviour of the joint, which remained elastic over a test duration of 100 cycles, although
some tribological issues still require further investigation.
3.5.4. Laser in-vessel viewing & ranging systems
(JET order/JWO-OFT-ENEA-02) (EFDA TWO
DTP/01-6-7-11)
The system is based on an Amplitude Modulated (AM),
properly focused, laser beam, which sounds the target,
whose viewing & ranging are then performed by means of
the intensity analysis and the phase-shifting of the
backscattered beam. The activities, that are now foreseen to
be extended to December 2002, have been carried out in
collaboration with the INN-FIS Division, which mainly
contributed for the optical part of the system. During the
year 2000, this activity has mainly included the following:
• Probe to be installed and tested on Joint European Torus
Fig. 3.30 - Multilink scheme 1:target 2:cassette
3:pin 4:link
95

3. Technology Program
Fig 3.31 - Image sample, taken by a digital camera
Fig. 3.32 - Same image as in Fig. 3.31, taken by LIVVS
Fig. 3.33 – The row range data about the staircase 1
(stair height 1 mm) have been filtered and shown in
Fig. 3.34
(JET) (JET Laser in-vessel Viewing System
LIVVS); JET order JWO-OFT-ENEA-02).
Testing of the system performances has been
almost completed, while vacuum (leakage,
outgasing, RGA, moving under inert gas
atmosphere) and reliability tests are foreseen
to be completed by July 2001. Figures 3.31,
3.32 allow for a comparison of the resolution
of the same image taken by a digital camera
and by LIVVS.
• In-Vessel Viewing System to be installed on
ITER FEAT. The system is based on the
experience gained during the JET LIVVS probe
and has the main aim of experimentally
demonstrating the feasibility of a laser in-vessel
system, able to perform both viewing and
ranging functions, under ITER FEAT operating
conditions. An ITER Task has been issued in the
framework of the EFDA Technology Work
Program 2000 (EFDA TWO DTP/01-6-7-11).
The main performances expected from the probe
are as follows:
• Viewing: sensitivity of ~1 mm at 10 m, with an acquisition speed of ~30 µs x pixel;
• Ranging: accuracy of 10 -4 (~0,5 mm at 5 m), with an acquisition speed of 0,1 ms x pixel.
During the year 2000, the system design has been completed and the supply of the components
has started, together with the development of special components/techniques. In particular, one
of the main items is to demonstrate that a coherent bundle of optical fibers can allow for undermillimetric
ranging, also by using AM laser technology, together with-high quality viewing and
reasonable timing. First relevant results have been obtained, as shown in figs. 3.33, 3.34.
3.5.5 IVROS articulated boom
In the first part of the year 2000, the control system and the mechanics of the In Vessel Remote
Operating System (IVROS) (fig. 3.35) have been fully tested on an FTU vacuum vessel sector.
96

3. Technology Program
Fig. 3.35 - IVROS control panel
Fig. 3.34 – A resolution of 200 microns confirms the
accuracy design target of 10 -4
The test campaigns included the study of robot
accuracy and trajectories optimisation aimed at
the definition of operating procedures, to be
followed during the FTU maintenance. The
IVROS articulated arm has been designed to
replace the vacuum chamber inboard limiters by
matching a special gripper tool (fig. 3.36) to the
limiter support structure. The tool is equipped
with two-expansion grippers and four Allen keys.
The keys must be simultaneously introduced in
the limiter fixing bolts. A torque control system,
based on dc motor current control, guarantees a
correct torque delivery. During the last FTU
shutdown, the IVROS system has been
successfully operated. Two limiter structures
have been remotely replaced. The removal of
delicate plasma diagnostics has thus been
avoided. The limiter removed has been checked
versus surface damages, conditioned and then reassembled
in the machine. The complete
operation took place within two working weeks.
3.5.6 Multilink general purpose boom
Fig.3.36 - Gripping tool close-up
The Multilink boom (fig. 3.37) is a modular robotic
carrier, to be introduced in the FTU vacuum
chamber during the machine shutdown. The task the boom has been designed for is to supply an
accurate positioning system, able to place a 100 N payload inside the vacuum chamber. The first
experimental campaign will consist of a systematic first-wall inspection, in order to determine the
surface status and the maximum alignment step between adjacent limiter tiles. During the year 2000,
the control algorithms have been implemented and a first procedure, intended to survey the tile step
by means of a laser device, has been preliminarly tested. The first test campaign on the FTU
machine has been postponed to 2001, to fulfil FTU safety requirements, which include a detailed,
fail safe analysis and rescue procedures for the remote-handled equipments.
97

3. Technology Program
Fig.3.37 - Multilink module
Fig. 3.38 - Weight increase of EUROFER specimens exposed
to Li 4 SiO 4 or Li 2 TiO 3 pebble bed
3.6 BREEDING
BLANKET
3.6.1 Compatibility test
between EUROFER and
Li 2 TiO 3 or Li 4 SiO 4 pebble
bed
Accelerated tests on compatibility
between EUROFER and breeder
materials were performed in a
thermobalance, in the temperature
range from 700 to 900°C in Li 2 TiO 3
pebble bed, and from 700 to 800°C
in Li 4 SiO 4 pebble bed, using a flow
of He+0.1% H 2 for an exposure
time up to a maximum of 200 h.
The thermobalance (NETZSCH
mod. STA 409) was part of a
specially designed gas loop, in
which it was possible to reduce and control
water content down to 10 ppm just before
exposure of EUROFER to breeder materials.
EUROFER disk (Heat D83344 01 supplied by
Planzet), lithium titanate pebbles (supplied by
CEA reference code CTI 30c7 Ti 1100) and
lithium orthosilicate pebbles (supplied by
Forschungszentrum Karlsruhe (FZK) batch
98/2-1 as-stabilised by annealing in air at
1000°C for two weeks) were tested. After
testing, the specimens were weighed and
analysed by X-Ray Diffraction (XRD)
technique, in order to determine the phases
present on their surface. Finally, Scanning
Electron Microscopy (SEM) observation and
Energy Dispersive X-Ray (EDX) analysis were
performed on the surface and a cross-section of
each specimen.
3.6.2 Exposure to lithium titanate
During the tests, water was generated by the system outgasing, and produced by the reduction of
lithium titanate at a rate that kept its concentration constantly under 100 ppm. Water was able to
promote oxidation of EUROFER, as reported in figure 3.38, where curves relative to weight gain
per square centimetre are shown. No clear kinetic law was observed during the first period for
the three temperatures, while a parabolic behaviour was observed after ten hours of exposure
time for the whole temperature range, indicating that a protective mechanism of corrosion was
operating. Parabolic rate constant values were calculated for each temperature after ten hours;
these data are shown in table 3.V. Their values fall within the range observed for oxidation of
chromium-bearing alloys, indicating that if any lithium species were operating in the oxidation
of EUROFER, their effect should not practically affect the corrosion process.
The XRD analysis performed after testing evidenced the presence of substituted Fe 3 O 4 and
98

3. Technology Program
Cr 2 O 3 oxides, as foreseen when using
the measured gas composition and
thermochemical data. The EDX
analysis identified the presence of Mn
and V as substituting metals at a
different concentration. In
(Fe,Cr,Mn,V)3O4, oxide, chromium
was about 25%, while Mn and V were
present in much lower amounts. In
(Fe,Cr,Mn,V)2O3 oxide iron was
about 5% and minor amounts of Mn
and V were present. Two different
morphologies of oxide were detected,
which were both acicular though of
different dimensions, as observed in
fig. 3.39; each was predominant in a
different area. No lithium compounds
were detected by the XRD analysis
performed on all the specimens.
Table 3.V – Parabolic kinetic constant as mg 2 cm -4 h -1
Exposure pebbles 700°C 800°C 900°C
Li 2 TiO 3 9.0±0.8×10-4 1.5±0.1×10-3 5.6 ± 0.3×10-2
Li 4 SiO 4 1.6±0.1×10 -2 9.6 ± 0.4×10 -2 -
The scale was always compact on
specimens tested at 700 and 800°C, but
a cracked scale was observed after 10 h
on specimens tested at 900°C.
3.6.3 Exposure to lithium
silicate
Water was released in large amounts
mainly during the heating step up to
the final temperature, together with a
small amount of CO 2 coming from the
decomposition of Li 2 CO 3 . During
tests, water content in the range from
200 to 1000 ppm was detected in the
gas phase, thus indicating a pebble
Fig. 3.39 - SE image of EUROFER specimen exposed at 700°C for
200 hours in Li 2 TiO 3 pebble bed
reduction. Such reduction can be due to metallic oxides present as impurities in the Li 4 SiO 4
pebbles; water coming from system outgasing was negligible.
The data concerning the weight increase of EUROFER disks, as expressed in mg/cm 2 , are shown
in fig. 3.38. Even in this case, a parabolic behaviour was observed after ten hours of exposure
time for both temperatures tested. Parabolic kinetic constants were calculated and reported in
table 3.V [3.16].
The corrosion rate was much faster as compared to the one measured in Li 2 TiO 3 pebbles. As an
example, it must be noticed that the parabolic rate constant in Li 4 SiO 4 at 700°C was ten times
higher than that observed in Li 2 TiO 3 at 800°C.
By using XRD analysis, FeO and LiCrO 2 were detected. The latter indirectly confirms the higher
volatility of lithium compound over silicate than over titanate pebbles. EDX analysis evidenced
a minor presence of Mn and Fe in LiCrO 2 , whilst Cr was detected as a substituting element in
FeO. The two compounds are separated in two layers: the first is in contact with the FeO alloy,
the upper one is constituted by LiCrO 2 . The scale was cracked, and in some regions near the
edges LiCrO 2 had detached itself, leaving free cubic FeO oxide (see fig. 3.40).
99

3. Technology Program
Fig. 3.40 - SEM image of EUROFER specimen exposed at 800°C for
100 hours in Li 4 SiO 4 pebble bed
3.6.4 Li 2 TiO 3 pebbles
reprocessing, recovery
of 6 Li as Li 2 CO 3
The capability of
producing tritium in a
blanket under neutron
irradiation, is proportional
to the concentration of 6Li
isotope; on the other hand,
since the natural
abundance of this isotope
is only 7.5%, an
enrichment process based
on the 6Li- 7 Li separation,
must be planned for every
type of breeding material.
Lithium titanate is one of
the most promising
candidates for tritium
breeding. Since the
amount of 6Li remaining
at “end-of-life” is much
higher than its natural
abundance, reprocessing of burned Li 2 TiO 3 pebbles can be considered very interesting from the
economic point of view [3.17].
The objective of this activity was to investigate about the feasibility of reprocessing Li 2 TiO 3
pebbles, at their “end-of-life”, when they still contain 40-50% of 6Li, in order to: i) recover 6Li
isotope as Li 2 CO 3 with the suitable chemical and morphological characteristics and ii) increase
the depleted 6Li concentration up to the value foreseen for this type of ceramic breeder, by
ensuring also a fully homogeneous distribution of 6Li isotopes.
The set-up of the process parameters was done on a Li 2 TiO 3 powder, obtained by solid-state
reaction between Li 2 CO 3 and TiO 2 . The optimisation of the procedure focused on two main
steps: a) total separation of Li from Ti by means of wet chemistry producing a Li-solution, and
b) precipitation of Li 2 CO 3 from the above solution. The full reprocessing flow chart is shown in
fig. 3.41.
Table 3.VI shows some characteristics of the Li 2 CO 3 produced.
Table 3.VI - Some characteristics of the obtained Li 2 CO 3 powders
Property Requested Type of chemical attack
A-1 A-2
Total yield (%) - 70±5 75±5
Recovered Li (%) - 62±5 70±5
BET surface area (m 2 /g) 1÷1.5 2.3±0.1 2.8±0.1
Bed apparent density (g/cc) 0.45 0.41±0.02 0.39±0.02
True density by He picnometry (g/cc) - 2.08±0.01 2.09±0.01
100

3. Technology Program
Fig. 3.41 – Reprocessing flow sheet
3.7 IFMIF
3.7.1 Activities on IFMIF optimisation and cost reduction (EFDA Contract
EFDA 99/506)
Safety analysis review and preliminary validation of occupational radiation exposure
The assessment of consequences due to failures in IFMIF was performed by applying the Failure
Mode and Effect Analysis (FMEA) technique to the latest design review. Through a first
screening of the design modifications, it is possible to anticipate that the environmental impact
of the facility should not be substantially affected by the design updating and will remain
negligible, as assessed in the Conceptual Design Activity (CDA) design.
A preliminary assessment of the ORE was also performed, by evaluating both the dose rates and
the estimated times of operation in the areas where the radiological risk is concentrated. The
results obtained from this analysis clearly point out that additional efforts are needed in reducing
both dose levels in plant-operating areas and maintenance time.
101

3. Technology Program
Design review and cost assessment of IFMIF conventional facilities
The cost reduction was carried out by taking into account the latest design solutions reported in
the conceptual design evaluation (FZK 1999). The reduced cost version of the facility (JAERI
00) was considered, which foresees a staged construction of IFMIF (in three stages).
The overall cost thus estimated was compared with the actualised CDA costs. The cost reduction
for the three conventional facilities analysed was 38% for the first stage and 25% for the third
stage.
Design of sub-miniature fission chambers for the IFMIF high flux test module
ENEA-Frascati was in charge of the work on neutron dosimetry in the IFMIF CDA and
Conceptual Design Evaluation (CDE) phases: the multi-foil activation technique was selected
for IFMIF neutron fluence and energy spectrum determination, as well as for the use of subminiaturized
fission chambers for on-line neutron fast monitoring. The definition of the final
characteristics of sub-miniaturized fission chambers monitors for IFMIF and the construction of
a prototype monitor will be performed by CEA.
3.7.2 Tasks of the key engineering phase
Lithium target
The reduction of IFMIF costs was partially obtained by means of changes in the parameters
which affect the liquid lithium jet behaviour, such as the beam power distribution and the static
pressure in the loop. For this reason, a first task is being devoted to defining the thermalhydraulic
conditions of the liquid lithium jet for the present design requirements. The work will
be performed by means of the RIGEL code, developed at ENEA in the frame of the IFMIF-CDA
and already used during that preliminary design phase. The main parameters for the definition of
the jet stability tests will be the result of this activity.
The reference IFMIF design is based on the concept of a replaceable back-plate. This solution
implies the presence of a joint in the nozzle inlet region. This discontinuity certainly affects the
jet surface stability. An experimental activity is ongoing in order to investigate the jet stability in
the presence of the discontinuity induced by the replaceable back-wall. Moreover, one of the
major factors of the Li loop cost reduction plan, is to reduce the Li loop height, and thus the depth
of the underground Li loop building. The height reduction decreases static pressure on every plan
inside the Li loop; consequently, the cavitation risk increases on every Li loop component. In
this frame, JAERI-J has planned some cavitation tests in small Li loops. During the tests, the
occurrence of the cavitation will be detected by the ENEA CASBA equipment. The results will
give an evaluation of the actual cavitation risk and the new requirements for experiment in a
dynamic simulation.
The construction of the IFMIF loop and target requires the development of suitable systems and
materials, able to operate in flowing Li. The lithium corrosion is strongly influenced by the
presence of non-metallic impurities in the liquid metal. Those impurities, especially N and C, are
able to form Li-compounds, which may increase corrosion effects on steels. For the above
reasons, a specific task is ongoing with the following aims:
• To develop a Li purification strategy, including monitoring and removal systems;
• To evaluate the corrosion rate of different materials (mainly, steels), in conditions relevant for
the IFMIF loop and target.
An activity is also foreseen for assessing remote maintenance of the removable back-plate. In
fact, for the IFMIF-CDA, ENEA has proposed a back-plate design, based on the so-called
Bayonet concept. The main advantages of such a solution concern the design simplicity and the
102

3. Technology Program
possibility to replace the back-plate without removing the Vertical Test Assembly (VTA). A
removable back-plate mockup is being manufactured to check the remote handling operation.
Design integration
Since some important modifications have been introduced at the plant facilities, a new safety
analysis has been launched, aimed at studying the sequences of an accidental event which might lead
to a dangerous situation. This activity may be also used for the preparation of the documents required
for the preliminary safety report. The IFMIF facility has to be operated with a reduced ORE. Safety
requirements have to be fixed for the overall station dose and by the individual operations on
systems and sub-systems, according to international standards and ALARA process. The respect of
such requirements will be verified by the assessment of design solutions both during the design
development–in order to assist the designers in making the right choice–and as soon as the final
design is “frozen”, in order to achieve consensus by the licensing authorities.
Test cell
The ENEA contribution to the IFMIF test cell concerns: neutron dosimetry, and the study of
activation and shielding requirement for the test facility.
Various dosimetry foils (Co, Au, Mn, Rh, Lu, Y and Zr) will be irradiated at the Nuclear Physics
Institute (NPI) Cyclotron Rez near Prague (Czech Republic); the gamma activity from the
samples will be subsequently counted in HPGe detectors. After corrections, the saturated
activities measured will be used for spectrum unfolding and cross-section optimisation. Spectra
measured with scintillators will be used as reference spectra. As a final step of the work, a test
campaign of the prototype IFMIF neutron on-line monitor and the activation foils will be
performed in the upgraded NPI Fast Neutron Facility (FNF).
Since the primary safety hazard associated with the test cell is radioactivity, an activity is
ongoing to estimate the following:
• The dose rate (gamma and neutron) in the operative areas of the IFMIF Test Facilities;
• The neutron-induced gamma activation of test modules and specimens, to assess shielding
requirements for the entire Test Facilities Complex (TCF).
3.8 NEUTRONICS
3.8.1. Experimental validation of shutdown dose rates for ITER
Neutronics activity is mainly devoted to validating nuclear analysis and predictions on designoriented
experiments in the frame of the ITER project, as well as to improving evaluations of
nuclear data and their experimental validation. Experiments with fusion neutrons are being
carried out at the Frascati Neutron Generator (FNG) that is, at the moment, the only 14 MeV
neutron source (intensity=1011 n/s) available in Europe for fusion studies. Nuclear analysis and
design is routinely performed also for other projects, with constantly updated computational
[3.18-3.25]. Neutronics experiments are important in order to validate nuclear analyses for ITER
[3.26] which are based on code and nuclear data with inherent uncertainties. Since it is essential
to guarantee occupational safety during hands-on maintenance inside the cryostat, experimental
validation is particularly necessary for dose-rate calculations of complex geometries, which still
suffer from uncertainties. Therefore, a neutronics experiment has been performed at FNG on a
stainless steel/water assembly (fig. 3.42), in which a neutron spectrum was generated, similar to that
occurring in the ITER vacuum vessel. The experiment was performed in a collaboration framework
among ENEA, Technical University of Dresden (TUD) and FZK Research Centre Karlruhe.
The mockup was irradiated at FNG with 14 MeV neutrons for sufficiently long time. A sufficiently
103

3. Technology Program
Fig. 3.42 - Schematic view of set-up for the Shutdown dose rate experiment (vertical
cut) with the positions of the various detectors employed during the two experimental
campaigns
high level of activation
was created, which
allowed to measure the
resulting dose rate for
more than two months of
cooling time after
shutdown, using two
independent
experimental
techniques. Other useful
measurements were
performed, such as
neutron spectrum, decay
gamma ray spectrum,
dose rate distribution
and some relevant
activation reaction rates
inside the mockup.
The experiment was
then analysed by using a
rigorous, two-step
method (R2S), viz. the neutron transport
code MCNP-4-B, and the activation
code FISPACT; and a direct, one-step
method (D1S), which uses an ad hoc
modified version of MCNP, used in the
nuclear analysis of ITER. FENDL-2
nuclear data libraries were used in both
cases.
The experimental analysis showed that
the dose rate measurement inside the
mockup is well predicted by the R2S
method: all Computed/Experimental
(C/E) values between 1 day and 2
months decay time are close to unity
within an 11% error margin. The
approximate D1S method is also in
Fig. 3.43 - Comparison between measured and calculated dose rate good agreement with measurements,
at the cavity centre
when using the same cross section file
(i.e. FENDL/A-2.0), and gives values
slightly but systematically lower than
the R2S method. This may be due to the fact that minor nuclides, contributing to the total dose
rate at the percent level, are not considered by the D1S method. The results obtained in the
analysis of dose-rate measurements are confirmed by direct measurements of Ni activation,
which gives rise to most of the dose-rate in the relevant decay time. In the analysis of Ni-
58(n,p)Co-58 measurements, the R2S method gives C/E values slightly higher than unity, while
the C/E values obtained with the D1S method using IRDF-90 and FENDL/MC–2 are generally
lower, and in better agreement with the measurements. As far as the Ni-58(n,2n)Ni-57 reaction
is concerned, both methods give C/E values slightly lower than unity, the underestimation being
more pronounced by D1S. These results are coherent with those found in the analysis of the dose
rate by R2S and D1S methods (fig. 3.43).
104

3. Technology Program
3.8.2 Evaluation of neutron cross-sections for fusion-relevant materials (EFF
project)
The correct design of the fusion reactor requires the availability of a complete nuclear data base,
extending up to 20 MeV in the neutron energy. The ENEA Fusion Division, through a
collaboration with the Applied Physics Division, participates in the EFF project, with the task of
re-evaluating the neutron cross-sections of carbon and oxygen in the years 1998-2002, on the
basis of recent experimental and theoretical findings. During the year 2000, the 12C(n,γ) 13 C
cross section has been calculated from thermal energies up to 12 MeV. For the 16O(n,γ) 17 O case,
the neutron energy range for the new evaluations has extended up to 2 MeV. Capture γ-ray
spectra for the two reactions under consideration have also been calculated.
A comparison of the results obtained by means of these model calculations with the presently
available data files has been produced. The most important result of this analysis is the
elucidation of the role played by the Direct Radiative Capture process (DRC), which is
responsible for the most part of the capture strength in the keV neutron energy region. The intercomparison
showed a large underestimation of the present data for evaluated 16O (up to a factor
of 100), contained in the ENDF/B-VI library.
3.8.3 Neutronics benchmark experiment on SiC (EFF project)
Silicon carbide is one of the candidate structural materials for the fusion reactor which is
presently under study, because of its excellent low-activation properties. However, there is a lack
of experimental data for silicon cross-sections. To fill the gap, a benchmark experiment on SiC
has been launched and set up in 2000, by using a block of sintered SiC (457 mm×457 mm,
711 mm in thickness, total weight 470 kg, 127 pieces), (fig. 3.44) lent to ENEA by JAERI. In the
experiment, that will be completed in 2001 in a collaboration framework among ENEA, TUD
and the Research Centre Karlsruhe (FZK), several nuclear quantities, including neutron and
γ–ray spectra, nuclear heating and activation rates will be measured at different penetration
depths inside the block irradiated with 14 MeV neutrons (up to about 58 cm, corresponding to
about 10 mean free path for 14 MeV neutrons). Measurements will be used to validate the EFF
cross-sections for SiC, including the new evaluations now in progress.
3.8.4 Experimental validation of
neutron cross-sections for fusionrelevant
materials (EAF project)
An important aspect in the development of
the fusion reactor is the capability of
developing materials with low neutroninduced
radiation levels, suitable codes and
nuclear databases to predict the nuclear
properties of such materials [3.27].
In the frame of the EAF, benchmark
experiments are being carried out on samples
of structural materials with low activation
properties. These samples are irradiated by
using the 14 MeV neutron generator FNG,
and the decay power induced is measured by
using a specially developed calorimeter, able
to discriminate beta and gamma heat. The
calorimeter is made of a large (9”×9”) CsI and
a small (1”×1”) BC400 scintillator, assembled
Fig. 3.44 - Schematic layout of the benchmark experiment on
SiC at FNG, showing the SiC block size and the four detector
locations inside the block
105

3. Technology Program
Fig. 3.45 - Schematic view of the calorimeter,
showing the large CsI and the small BC400
scintillators, assembled and enclosed in the
shield. Small samples of material irradiated with
14 Mev neutrons are located between the two
scintillators, which are able to measure the
gamma- and the beta-decay heat separately
as sketched in fig. 3.45. The CsI
scintillator detects γ-rays with highefficiency,
while it is shielded against
β-rays; the BC400 scintillator detect β-
ray (≈2π efficiency), but is almost
insensitive to γ-rays.
Fig. 3.46 - Ratio of calculated (C) over measured (E) decay heat
for EUROFER irradiated with 14-MeV neutrons in a) a short
irradiation, to study radioisotopes with short half-lives, and b) a
long irradiation, to study radioisotopes with longer half-lives
During the year 2000, samples of
EUROFER 97 have been irradiated
inside a perspex/ polyethylene
reflector. This reflector was designed to mimic the fusion reactor first-wall spectrum, which
contains a tail of slowed-down neutrons, together with the 14 MeV D-T fusion neutron peak.
Two thin disks (18 mm diam., 25 mm thick) of EUROFER 97 have been irradiated for a short
time (≈10 min) and a long time (≈ 6 h) respectively. Thin samples have been used to mitigate
beta self-shielding effects. The shortly-irradiated sample was measured for decay times ranging
from 2 min up to 2 h, while the long-irradiated sample was measured for decay times from 1 h
up to 800 h.
The decay heat was measured, and then compared with the predictions of the European
Activation Code System (EASY-99). The results are reported in fig 3.46.
Some main discrepancies between calculated (C) and experimental (E) values are found for
decay time >30 h, when C/E≤0.5 is obtained for beta heat. At this decay time, 186W(n,γ) 187 W
is the dominant reaction for gamma heat, while 51Cr dominates the beta heat. To find out
whether other radionuclides, not predicted by the calculation, were present in the activated
EUROFER 97 samples, and were responsible for the underestimation, the samples have been
106

3. Technology Program
measured also at the Gran Sasso laboratory, by using the ultra-low background HPGe detectors.
No significant difference between calculated and measured activities was found, indicating that
the element composition input in the calculation is correct. The discrepancy between the
calculated and the measured beta heat for decay time >30 h will be further investigated.
3.8.5 Activation foils and real-time neutron/gamma detectors for IFMIF
In the frame of the EFDA Technology Work Program 2000, a task has been entrusted to ENEA-
Frascati for the development of IFMIF fast-neutron monitors, and their testing on a cyclotron.
This activity (2000-2002) aims at the realisation of prototype on-line neutron monitors for the
High Flux Test Module (HFTM) of IFMIF: the detectors will be tested on an IFMIF-like neutron
field, provided by the upgrade of the cyclotron-based FNF, located at the NPI. At the same time,
integral tests on activation foils selected for IFMIF neutron dosimetry will also be performed.
The final design of the IFMIF candidate fast neutron monitors (sub-miniature fission chambers)
has been completed [3.28,3.29] by using various codes as follows:
• MCNP, to condense the fissile materials cross-sections, (n,f), (n,2n) and (n,γ), to one-energy
group with the detailed IFMIF spectrum (choice of fissile material);
• EVO77, to study the evolution of the isotopic composition of the fissile deposit during the
irradiation in a typical IFMIF run (determination of burn-up of sensitive material after prolonged
irradiation)
• FCD, to evaluate sensitivity and range of operation of the detectors (neutron sensitivity,
saturation domain).
On the basis of these evaluations, the best candidate fissile deposits have been found to be
237Np – mainly because of the higher number of fissions above 1 MeV and the type of nuclides
produced in the fission reactions – and 238U. It has been estimated that, after 9 months of IFMIF
irradiation, the burn-up of the fissile material is 3.8% and 1.5% of the initial content, respectively
for 238U and 237Np, and the production of parasitic fissile nuclides is very low, almost
independently of the fissile coating chosen. The fission chambers should have a 1.5 mm diam.,
with a total mass of 528 µg and argon filling gas at a 2 bar pressure. However, cyclotron larger
chambers (8 mm diam.) will be produced for testing in the cyclotron, in order to cope with the
neutron emission expected at the cyclotron (~3×1012 n/s/sterad), which is lower than the
emission foreseen in IFMIF (~1015 n cm-2 s-1).
3.8.6 Development of Chemical Vapour Deposition (CVD) diamond detectors
for nuclear radiation
Diamond detectors are of a particular interest as neutron detectors in fusion environments, as
they present much higher radiation resistance as compared to silicon detectors, and good energy
resolution properties. In the frame of the collaboration established with the Faculty of
Engineering of Tor Vergata University in Rome, diamond films, produced with the CVD method,
are being developed, and their characteristics for nuclear detection analysed.
During the year 2000, several new samples have been tested with nuclear particles (α–particles
and electrons). In this way important parameters, such as grain dimensions, film purity and
lattice properties have been analysed.
In particular, an analysis of the time pulse-shape of particles detected has been carried out. It has
been found that the pulse amplitude and shape depend on the field polarity and are dramatically
affected by pumping. These changes have been interpreted in the framework of the trapping/detrapping
model, originally applied to Si-based detectors. To explain the particular features found
in this work, the original model had to be modified to reflect the higher complexity of CVD
107

3. Technology Program
diamond. A computer simulation, based on the model, gives a pulse shape which agrees well
with the ones observed. Valuable information concerning the nature of trapping centers for
electrons and holes is provided by the analysis of pulse shapes. The effect of pumping, which
results in an increased amplitude and in the development of a slow component for positive field
polarity is explained, allowing to enlighten the role of pumping in the detector’s performance and
to quantitatively estimate the de-trapping time constant.
A remarkable feature is the apparent scaling (within the experimental noise) of pulses measured
in a given condition, which have all the same shape, independently of their amplitude.
The behaviour of the response under increasing electric fields leads to the conclusion that fieldenhanced
de-trapping occurs at fields close to 104 V/cm.
3.8.7 Participation in the Astrorivelatore Gamma ad Immagini LEggero
(AGILE) project - The collimator and the coded mask of the SuperAGILE
detector
The expertise in Monte Carlo techniques and in nuclear instrumentation, acquired in neutron
diagnostics design, has been profitably employed in space applications. Satellites in orbit around the
Earth are in a hostile environment. Energetic cosmic rays incessantly hit the spacecrafts and produce
a wide spectrum of undesired effects. If the satellite is devoted to x or γ-ray astronomy, energetic
photons directly increase the detectors background. High energy protons activate the payload either
directly or through spallation neutrons, thus slowly but inexorably increasing the background level.
And all contribute to the degradation of the onboard electronics. Thus, the prediction of the expected
x-ray background is of a tremendous importance in the project of a scientific spacecraft, as far as the
design of its sensitivity and performance is concerned.
AGILE [3.30,3.31] is the first mission of the Small Missions Program (SMP) of the Italian Space
Agency (ASI). Its main goal is to monitor the gamma-ray sky in the energy range between 30
MeV and 50 GeV, with a large field of view (~3 sr), good sensitivity, good angular resolution
and good timing. The satellite is presently under construction and is scheduled for launch at the
beginning of 2003 in an equatorial orbit, for a lifetime of 2 years. SuperAGILE [3.32] is the x-
ray monitor added on top of the γ-ray tracker. It will have a large field of view, providing hard
x-ray imaging thanks to its division in four mutually orthogonal detectors, each coupled to a onedimension
coded mask through a collimator. SuperAGILE will enable a wide variety of cosmic
x-ray sources, including γ-ray bursts, persistent and transient Galactic x-ray sources to be
studied, as well as many of the brightest extragalactic sources. The Neutronics section of ENEA
Fusion Division takes part in the AGILE project. It contributed to the design of the SuperAGILE
masks and collimators through a Monte Carlo study, which permitted the minimisation of the
noise to signal ratio and the optimisation of the detectors response. At present, it is involved in
its technical realisation.
3.9. FUEL CYCLE
3.9.1 Development of palladium-ceramic membranes
A wide work aimed at developing palladium-ceramic membranes has been carried out in the
framework of the task ITER TR6 [3.33-3.36]. These composite membranes have been produced
by means of the three following techniques, consisting of coating a ceramic porous tube with a
Pd-Ag thin film: electroless deposition, sputtering and rolling [3.37].
Both electroless and sputtered palladium-ceramic membranes (Pd-Ag film thickness 1-10 µm)
have shown a not complete hydrogen selectivity and a limited durability. In fact, in order to
obtain a complete hydrogen selectivity, all the pores of the ceramic support have to be closed by
108

3. Technology Program
increasing the film thickness. But the stresses at the ceramic/metal interface, due to the
expansion of the metallic layer under heating and to hydrogen loading, produce a shear stress τ,
at the metal/ceramic interface, which is proportional to the film thickness s:
where:
E Pd-Ag
ε H/Pd-Ag
ε cer =1+α cer ∆T
L
( )
E ε − ε s
Pd−Ag H/Pd-Ag cer
τ =
L
is the Young’s modulus of the Pd-Ag, Pa
is the thermal strain of the hydrogenated Pd-Ag
is the thermal strain of the ceramic
is the membrane length, m
These shear stresses give rise to formation of cracks and peeling of the metal coating in the
electroless and sputtered membranes.
The rolled membranes have been obtained by cold-rolling and annealing Pd-Ag thin foils. Then,
the rolled thin foils (thickness in the range of 50-70 µm) have been wrapped round the ceramic
porous support and joined, in order to obtain permeating tubes of length 150 mm, internal diam.
10 mm. To the purpose of joining the metal foils, a diffusion bonding procedure, based on the
high mobility of the Ag atoms through the Pd lattice, has been realised and patented [3.38]. In
these rolled membranes, an annular space existing between the metallic membrane and the
ceramic porous tube ensures the minimum clearance (20-40 µm) required for easily inserting and
moving the ceramic tube inside the metallic membrane. This clearance produces a floating fit
between metal and ceramic, thus avoiding the presence of any interfacial stresses and membrane
failures under thermal cycling and hydrogen loading.
A plant equipped with temperature, pressure and flow rate on-line measuring and controlling
devices has been set up both for testing and characterizing membranes and membrane reactors.
The experimental apparatus consists of two mass-flow controllers at the inlet of the membrane
tube (feed side) and shell side, three pressure gages at inlet and outlet of the membrane tube and
at shell outlet, several thermocouples inside the reactor and in the circuit line, two heating
systems for controlling the membrane tube temperature and vaporising the water fed inside the
membrane reactor.
During the tests, the rolled membranes have shown high and stable hydrogen permeability
values, besides hydrogen selectivity and chemical stability [3.39,3.40]. These results make these
membranes suitable for applications in the fusion fuel cycle (tritium recovery from tritiated water
via gas shift) as well as in industrial processes, where high pure hydrogen is produced (i.e.
hydrocarbon hydrogenation/dehydrogenation reactions).
3.10 SAFETY AND ENVIRONMENT
3.10.1 Assessment of ORE (Task SEA2)
A specific study has been dealing with the preliminary assessment of the collective dose due to
the scheduled maintenance and inspection activities for some of the main systems of ITER-FEAT
plant [3.41]. The preliminary collective dose results were provided for related to the hands-on
activities for maintaining, inspecting and/or replacing the following items: blanket/limiter;
109

3. Technology Program
electron cyclotron heating system; ion cyclotron heating system; cryopumps; divertor cassettes;
three loops of the Tokamak Cooling Water System (TCWS). The radiological sources,
considered by calculating the dose rate, are neutron activation due to the plasma burning and
activated corrosion products on the inner surface of the TCWS pipes and components. The
preliminary results give a collective dose for the hands-on assistance activities at the tokamak
ports of 190 person-mSv/y. A better situation was obtained for the three loops of the TCWS with
8,5 person-mSv/y, about 3 person-mSv/y per loop. The latter could be compared with that
obtained in the latest work related to the collective dose assessment for ITER Final Design
Report (FDR), which estimates a collective dose per cooling loop in the range from 39 to 16
person-mSv/y.
3.10.2 Plant safety assessment (Task SEA4)
In the frame of the ITER Task Plant Safety Assessment (PSA), a set of activities have been
performed to prepare the ITER Generic-site Specific Safety Report (GSSR).
Activation calculation support for safety analysis
Radiation transport and activation calculation for ITER have been performed to support safety
analyses design (ITER Task D451).
The activation calculation results include specific activities, such as the following: decay heat;
contact dose; clearance index; list of isotopes at shutdown and dominant isotopes vs cooling time
up to 1×106 years, related to each material for all the ITER radial zones. From the radiation
transport assessment, the nuclear heating for each zone has also been obtained.
The ITER Joint Central Team supplied the input data (geometry, material data, irradiation
characteristics and operation scenario). The machine is radially described by 87 zones, which
include the following regions: Centre Void (CV); Tie-Plates and Central Solenoid (CS); Thermal
Shields (TS); TFC; Vacuum Vessel (VV); First-Wall/Blanket Plasma (FW/BP); Cryostat;
Biological Shield (BS).
The overall integrated computational approach is the same used in the past by ENEA for scoping
analyses during ITER Engineering Design Activity (EDA). The neutron and γ flux distributions
have been calculated by means of the Bonami-Nitawl-Xsdnr Sn Model coupled to n-γ onedimension
discrete ordinates transport method, using a Scale-4.4a computer code system. The
Vitamin-ENEA-J (175n-42γ groups) transport library, based on FENDL/E-2 data library, is used
for transport calculation. The nuclear heating related to the different materials for all the radial
zones has been obtained from the transport calculation sequence by using Kerma factors,
obtained by processing nuclear basic data from the EFF-2.4. The EASY-99 activation package
(with the Fispact99 activation code) has been used to obtain the activation characteristics of all
the materials/zones of ITER. The package was supplied to ENEA by UKAEA.
The results have been used by ITER-Joint Central Team (JCT) for GSSR Volume III [3.42] and
Volume V [3.43] which provide, respectively, relevant information for ITER safety analyses
(definition of source terms for accident sequence assessment), waste management and disposal
strategies.
All the results of radiation transport and activation calculations have been included in a CD-
ROM, while, according to the requirements defined by ITER-JCT, only the following
compressed data are presented in [3.44]:
Normal irradiation scenario SA1: specific decay heat up to 1 year, for all zones up to and
including the VV; dose rate for all time steps for FW, back of shield, VV surfaces, magnet
surfaces, cryostat surfaces, bioshield surfaces; clearance index for all zones and all time steps;
complete isotope listing of most activated regions for each material of in-vessel components, at
110

3. Technology Program
shutdown {Be, Cu, SS316 of FW}; list of most significant isotopes (at least 1% contribution to
decay heat, activation, or clearance index) for Cu-FW, SS316-FW, backplate, and outboard VV,
as a function of time; nuclear heating in all zones.
Divertor irradiation scenarii SA1-DV1 and SA1-DV: tungsten specific decay heat up to 1 year;
tungsten clearance index for all the cooling times; complete isotope listing of tungsten at
shutdown; tungsten nuclear heating.
Deterministic accident analysis
Two reference accident sequences have been assessed by ENEA as a part of the European Home
Team EU-HT contribute to the ITER GSSR Volume VII [3.45]: one loss-of-flow event in the
divertor primary heat transfer system Divertor Vertical-Primary Heat Transfer System
(DV–PHTS), and a loss-of-coolant event in the DV PHTS.
The pump seizure in divertor Heat Transfer System (HTS) Loss of Flow Accident ((LOFA)
Category III) is described as an example of accident analysis (table 3.VII).
The postulated event is a pump seizure in a cooling loop of the Divertor/Limiter Primary Heat
Transfer System (DV/LIM PHTS) during a plasma burn. Once the coolant flow in the failed loop
drops to 80% its nominal value, the Fusion Power Shutdown System (FPSS) will stop the plasma
burn in three seconds (active fast train). The subsequent plasma disruption, which delivers 0.4
GJ of energy to the DV, is postulated to cause failure of a DV cooling pipe inside the plasma
chamber (double-ended pipe break of the DV/LIM HTS loop; equivalent size of 0.32 m2) due to
melting of the cooling tubes (copper) of the lower vertical target, which reaches more than
1000°C.
During operation at 500 MW fusion power with the HTS in steady-state condition, a pump
seizure in the divertor primary cooling loop results in a quasi-instantaneous coolant flow
reduction. Once the plasma burn stops, the decay heat of the divertor plate materials is
considered as the heat source for the DV/LIM HTS. The DV/LIM HTS coolant in-leakage
pressurises the VV and, as soon as VV pressure exceeds 80 kPa, the bleed lines to both VVPSS
and drain tank open.
The DV HTS coolant inventory is discharged (~160,000 kg: ~6,800 kg steam, ~153,000 kg
water) into the plasma chamber. The outlet flow is practically zero after 2,747 s. The VV
pressurises, and the set point for
bleed lines opening to the
VVPSS and to the Drain Tank is Table 3.VII - Time sequence of events for pump trip in divertor HTS
reached at t= 9 s from the
beginning of the accident. The
set point for rupture disks
opening to the VVPSS is
reached at t=111 s. The
maximum VV pressure is about
150 kPa, below its design value.
Event sequence
Total fusion power 500 MW (nominal value)
Pump seizure in a divertor primary heat transfer loop
Time

3. Technology Program
highest divertor temperature less than 125 °C after about 400 s from the in-vessel break.
The radioactive inventories involved in the accident are the following: tritium and the activated
corrosion products of the failed DV HTS loop; tritium in the co-deposited layer; in the PFC bulk;
in the cryopumps; and the activated dust inside the VV. The tritium content per DV/LIM HTS
loop is 0.85 g, which corresponds to a concentration of 0.005 g/m3. The corrosion product
inventory per loop is estimated at 10 kg (0.2% in the suspended form and 99.8 % as wall
deposits). About 100% of the suspended and deposited Activated Corrosion Product (ACP) are
released from the in-vessel break.
The mobilised inventories, transport and environmental release of tritium, as well as the
activated aerosol related to the DV pump seizure accident, are summarised in the following:
• Tritium: controlled release 8.9×10 -5 g-T and ground release 3.4×10 -5 g-T, for a total release
of 1.2×10-4 g-T
• Dust: controlled release 1.1×10-5 g and ground release 2.4×10-4 g, for a total release of
2.5×10-4 g
• ACP: controlled release 8.9×10-9 g and ground release 8.9×10-7 g, for a total release of
9.0×10-7 g.
Probabilistic accident analysis
A component level Failure Mode and Effect Analysis (FMEA) has been applied in order to
identify Postulated Initiating Events (PIEs), and possible accident consequences for the water
cooling systems of the ITER-FEAT reactor [3.46]: FW/BLK; Divertor and Limiter (DIV/LIM);
Vacuum Vessel (VV); and Neutral Beam Injector (NB Injector) primary heat transfer systems;
and, Heat Rejection System (HRS). The FMEA is a bottom-up methodology, which allows
detection of PIEs (see the sample of table
Table 3.VIII - List of PIEs related to failures in FW/BLK
cooling loop
PIEs
FF1
FF2
FF99
HF1
HT99
LF01
LF02
LF03
LFVI
LFV2
N/S
Description
Loss of flow in a FW/BLK coolant circuit
because of pump seizure
Loss of flow in a FW/BLK coolant circuit
because of pump trip
Loss of all FW/BLK ccoling pumps (with
coastdown)
Loss of heat sink to FW/BLK loop
Total loss of heat sink to all primary loops
Large rupture of FW/BLK coolant loop outside
VV (in cooling room or service shaft
Small rupture of FW/BLK coolant loop outside
VV (in cooling room or service shaft
Rupture of tubes in a primary FW/BLK heat
exchanger
Rupture of one FW/BLK segment coolant loop
inside VV
Small FW/BLK in vessel LOXA. Equivalent
break size: a few cm 2
Not safety relevant initiator
3.VIII) by grouping and classifying elementary
basic failures. Each defined PIE is, in fact,
characterised by the following:
• A set of elementary accident initiators,
grouped under this PIE, taking into account the
similarity of accident development in terms of
mitigating features and possible consequences;
• A representative event, which is usually the
most challenging one from the safety point of
view, in terms of expected frequency and
radiological consequences among several
individual component failures, which could
produce equivalent fault-plant conditions;
• An overall frequency, obtained by adding for
that elementary initiators.
To report the FMEA assessment, a specific tool
was developed in the frame on an Excel
spreadsheet. Through such a tool specific
routines, written in Visual Basic language, allow
the analyst to easily manage information and
reduce the time-consuming study reporting.
Useless writing of repetitive words is avoided,
as well as writing of same context in different
ways. Mistakes due to reporting are reduced.
112

3. Technology Program
Table 3.IX - FMEA during NO for large piping inside vault of the FW/BLK loop
Failure Causes Prev. Action Consequences Corr./Prev. Act. PIEs Comment
Mode on causes on consequence
Large break
Material defects and Adequate welding LOCA in Vault; Vault pressure relief LF01 Design solution have
aging; Corrosion; process quality; Pressurization of Vaul; to expansion Vol. in case to be taken to avoid
Abnormal operat. Water chemistry Release of RadP_in press. gets safety limits; Service Shaft flooding;
conditions; Vibrations; control; InService PW to Vault Isolate the breached Water from all PFW/BLK
Local. stresses inspect.; Leak cooling loop; Assure Vault pr. loops could be discharged
monitoring leaktightness; Air treatment if the CVCS is not promptly
of Vault by ADS; Drainage isolated (a by-pass between
& detritiation of Vault
the different pr. loops
° Impact of heavy Design against Release of RadP_in_PW Air treatment of exp. vol. ° In case pressure relief device
loads (missile) missile generation to Expansion Volume by ADS; Drainage & opens to Exp. Vol.
detritiationof Exp. Vol.
° ° ° Overheating of affected Plasma shutdown ° °
PFCs because loop
emptying in some tens
of seconds
° ° ° PFCs break and LOCA ° ° °
in-VV
° ° ° Be-Steam reaction Assure dense Be conditions ° °
on PFC; Assure low Be
temperature with the
operating loops
° ° ° Pressurization of VV VV pressure relief to ° °
Suppression tank and Drain
tank
° ° ° Release of VV_RadP Isolate the breached ° Main valves capable to
after differential pressure cooling loop; Assure Vault isolate main components
inversion to Vault leaktightness; Air treatment (HX, PZ, in-VV PFCs) could
of Vault by ADS
help in reducing released
inventory
Data on the main sheet (FMEA table) are organised by the way of relations with other sheets,
where individual pieces of information are listed and classified. The FMEA table (see a sample
in table 3.IX) is structured in order to report the following items for each component: all possible
failure modes, which might occur at the different operating stages; related accident frequencies
and category classification, failure causes and possible actions to prevent the failure;
consequences and actions to prevent and mitigate the consequences; the PIE in which the
elementary failure mode - relevant from a safety point of view - is grouped; and, eventual notes.
In the auxiliary sheets related to data reported in the FMEA table as codes or short strings, a
detailed description of the data itself is reported. By matching the component code it is possible
to find out detailed description, useful data considered to evaluate frequencies of failure modes,
number of units in the system and number of systems in the plant.
A specific routine of the FMEA tool allows the analyst to automatically obtain, in a separate file,
113

3. Technology Program
Table 3.X - Elementary failures grouped on PIEs
PIEs Component Component Failure Freq. Cat Unit/ N° of Total Tot
code description mode system systems freq. cat
LF01
FB-LpipeV-PL FW/BLK pr. loop-Large Large break 8.1×10-6 IV 1 3 2.4×10-5 IV
piping inside Vault
FB-PZ-PL FW/BLK pr. loop-Pressurizer Large break 2.5×10 -7 V 1 3 7.5×10 -7 V
FB-MP-PL FW/BLK pr. loop-Main
circulation Pump
Case large
Break
7.5×10 -7 V 1 3 2.3×10 -6 IV
FB-CVPipe-PL FW/BLK pr. loop-Piping
inside cryostat volume
Large break 1.6×10 -5 IV 1 3 4.7×10 -5 IV
FB-LPipeS-PL FW/BLK pr. loop-Large Large break 1.8×10 -5 IV 1 3
Piping inside Service Shaft
5.3×10 -5 IV
the list of PIEs identified for the systems being assessed. Data reported in the list are ordered for
plant-operating condition. For each PIE, a complete list of elementary failures, which might
contribute to cause the event and the related total frequencies are included too (see a sample in
table 3.X).
The total number of PIEs, pointed out by assessing elementary failures related to the different
water cooling sub-systems, is 47. Accident sequences arising from each PIE have been
qualitatively defined by assessing the PIEs. Deterministic analysis will have to demonstrate the
plant capability of mitigating and, in any case, keeping the consequences, arising from the
overall set of PIEs, below fixed safety limits.
All elementary failures not inducing safety-relevant consequences have been classified in a PIE
named Not Safety relevant (N/S). Even though such failures are not important from a safety
point of view, they are important in defining plant operability and maintenance strategy.
Corrosion product modeling and inventories
ACPs inventory of three cooling loops of the ITER Tokamak Water Cooling System (TCWS)
was calculated by the PACTITER code for two different operating scenarios [3.47]. The main
findings obtained from the present analysis are as follows:
• The total ACP deposit mass at shutdown is below 2 kg, regardless of the scenario;
• The ACP deposit mass evolution during the scenarios shows peaks in correspondence with the
plasma burns, whenever the transfer of ACP deposits is elevated from under-flux zones to the
out-of-flux ones. The ACP deposit mass for one scenario reaches the maximum value after the
first 155-day burn period (~8.3 kg, 8 kg onto the out-flux zones). This confirms that the 10-kg
ACP, chosen for the accident analysis, is a conservative value, as 8.3 kg of ACP is reached after
a not-realistic burning period of 155 days.
• The piping base metal release increases during plasma burns, especially for the under-flux
regions, due to the coolant temperature gradient existing there. The specific material release per
unit of surface is about three times larger for the under-flux zones than for the out-of-flux ones.
• The ACP radioactive inventory at shutdown of the under-flux is mostly due to short-lived
radionuclides. On the contrary, the out-of-flux wall activity is composed for the most part of
long-lived radionuclides, as it slowly builds up with the time.
• The coolant activity is also dominated by short-lived radionuclides at shutdown (86 –95 % of
the total). It reduces by about two orders after 1 day of decontamination mode for the combined
effect of the improved CVCS performance and the decay of the short-lived one.
114

3. Technology Program
• The out-of flux wall activity decreases, as compared to the previous value assessed for the
ITER-FDR PFW/IBB loop;

3. Technology Program
The results also confirm the effectiveness of
the ISAS tool for accident analyses.
In the frame of the validation of the
CONSEN code against ICE experiments, the
models relating to the critical flow model
and the jet impingement heat transfer model
have been activated. CONSEN allows for
heat transfer mechanisms, such as nucleate
and film boiling; critical heat flux
evaluation; evaporation at gas-liquid
interface; condensation; natural convection;
and thermal conduction inside the structures.
CONSEN [3.49] showed a good capability to
simulate the cases chosen for the blind pretest
calculations. The difference between the
calculated flow rates and the experimental
ones are in the range of 1% for almost all
cases. One of the most important parameters
in the accident evolution is the maximum
pressure resulting in the plasma chamber: the
difference between calculated and
experimental values is in the range of 15%.
3.10.4 Occupational dose and
development of requirements for
environmental releases (Task TRP1)
This study developed high-level
requirements for worker and public safety,
Fig. 3.47 - Validation of computer codes and models: which are consistent with the international
comparison between computed results and experimental data regulations and the American and Japanese
safety and environmental fusion power plant
objectives. The European Utility
Requirements (EUR) for station dose were proposed for the Power Plant Conceptual Studies
(PPCS), as a first step in demonstrating compliance to international regulation. However, the
analysis carried out [3.50] showed, that the EUR requirements would hardly be satisfied by any
reactor design utilizing current fusion materials and technology. A low–activation stainless steel
cooling system and a water coolant combination would show problems relating to the formation
of the corrosion products. Such a combination would lead to high cooling system doses.
Furthermore, the need to open the reactor every 30 months in order to remove PFC represents a
serious problem. This activity contributes for over 70% of the estimated average annual station
dose, as compared to an allowable 17% of the target based on EUR. Potential solutions
envisaged are the following: to avoid such a combination of material/coolant, and to develop
PFC materials which can survive for at least five full-power years of operation, thus reducing at
the same time the PFC replacement outage duration, currently assumed to be four months.
From a public safety perspective, a dose target of 50 µSv/a was proposed to satisfy ICRP
recommendations, including ALARA. The corresponding release targets for the reactor are as
follows: 0.9 g/a of T and 7.1 g/a for tungsten tokamak dust. Generally, it is possible to
demonstrate that all release targets can be satisfied. Tokamak dust could be a potential problem,
particularly if the PFC material is tungsten. It is possible to improve the situation by reducing
the PFC replacement outage frequency from once every 30 months to once every 60 months, also
required to satisfy the worker dose targets.
116

3. Technology Program
3.10.5 Waste management
Material optimisation and implications
The minimisation of active waste from the operation and decommissioning of a fusion power
plant must be one of the main scopes for fusion waste management studies. Clearance (release
from radioactive material regulatory control) is one of the ways to achieve this goal.
The problem of defining operative clearance levels was addressed [3.51] by proposing two
different approaches: clearance for non-active disposal, and free-release recycling. Some
operative limits were proposed and compared for both approaches.
The proposed clearance approaches were applied to the case of two SEAFP Plant Models, using
new activation ex-vessel material data, with the aim of a possible optimisation of ex-vessel
components composition, in order to allow clearance to the largest possible extent [3.52].
Waste and decommissioning strategy
Previous studies on waste management dealing with the Safety and Environmental Assessment
of Fusion Power (SEAFP), Safety and Environmental Assessment of Fusion Power Long-Term
(SEAL), SEAFP-2, SEAFP-99 studies were reviewed [3.53]. For SEAFP, the reference waste
management strategy deals with the disposal of reactor waste, mostly coming from
decommissioning plus blanket and divertor replacement, into fission waste disposal sites. Waste
volumes turned out to be considerably high. However, since most of SEAFP waste comes from
relatively low–activated material, in a shielded position from the plasma, it appeared appropriate
to explore the possibility of finding alternative pathways for the management of such waste, in
order to minimise the use of final repositories. For this purpose, an alternative management
strategy was developed during the SEAL, SEAFP–2 and SEAP-99 studies, based upon two main
concepts, as follows:
• Recycling of moderately radioactive materials within the nuclear industry;
• Declassification of the lowest activated materials to non-active material (Clearance).
The Canadian radioactive waste management experience has been reviewed because of its
potential similarities with fusion. The conclusion from this study is that some of the CANDU
waste management experience would be relevant, and perhaps useful, in fusion power-reactor
studies.
A radioactive waste management scenario for future commercial fusion power reactor was analysed
on a programmatic basis, in the context of a mature, global fusion-power industry, to identify plant
design parameters and material choices which may significantly reduce the waste disposal burden
[3.54]. The main focus was on material selection, especially for the breeder blanket.
A comparison with the fission power industry showed that, in general, there are many similarities
between fusion and fission waste. In fact, according to the current knowledge and technology,
the specific activity of fusion reactor waste seems not to be significantly lower than that of
fission reactor components, even though it has to be pointed out that radiological effects of
fusion waste are very different from the fission one, namely: fusion produces no heavy elements,
such as the transuranics, which are much more radiotoxic than the activation elements in fusion
reactor components, and potentially in weapons material.
The main findings from this study were the following:
• Operational waste is larger than decommissioning waste;
• Operational waste can be strongly minimised by a blanket design utilising Li-Pb as a
breeder/multiplier, and vanadium or low-activation steel as a structural material;
117

3. Technology Program
• Operational waste can also be strongly reduced for ceramic breeding blankets by processing
the breeder material and by recycling its structure;
• Extension of the life of the blanket structure, possibly by annealing in order to repair neutroninduced
damage, may also reduce the quantity of recycled material.
This study highlighted the importance of in-vessel materials selection; the need for breeder
reprocessing; and the influence of in-vessel component and plant design life on the quantity of
high-level waste production. The lithium-lead is the only breeding material under consideration
that can be re-use. Re-used would eliminate a potentially large source of high-level waste, and
significantly improve the environmental advantages of fusion power.
3.10.6 Socio economics studies
The research program developed by SERF2 is the natural progress and the completion of SERF1.
The methodological model is as follows:
• Widespreading of information and raising the public awareness concerning energetic issues
and the use of fusion;
• Raising the public awareness concerning socio-economic effects on the region linked to the
realisation of a Research Centre on nuclear fusion. Analysis and widespreading of JET
experiences in Culham;
• Direct experiences of a delegation of citizens of Porto Torres through visit/lab at JET in
Culham;
• Creation and widespreading of a simulated scenario about the realisation of a Research Centre
on fusion in Porto Torres, which would consider the possible effects on local economy;
• Socio-economic data processing on the achievement of the thermonuclear fusion energy,
involving the public participation though the Strategic Scenario Workshop (an adaptation of the
European Awareness Scenario Workshop methodology made by the National Monitor EASW,
Mr. Bastiani).
The first step of the research on social impacts of the fusion of the SERF2 had the objective of
collecting and elaborating materials to describe the present situation of the towns of Porto Torres
and Abingdon, to the purpose of showing the socio-economic and environmental aspects of these
regions.
The second step of the research was the realisation of a meeting at Culham and Abingdon, UK.
The objective was as follows:
• To analyse and directly ascertain the impacts of JET on the surrounding region; consequences
for the environment and social acceptancy, the direct and induced development of the local
economy;
• To bring together expert and common knowledge on issues normally reserved to experts;
• To acquire knowledge on technology of nuclear fusion as well as on the experimentation
presently underway;
• To find answers to frequently recurring problems and questions regarding the acceptancy and
the security of a high technology plant;
• To spread the information acquired and to open a discussion at the local community of Porto
Torres.
The third step was the local EASW. On December, 2nd, 2000, a Strategic Scenario
Workshop–which represents an adaptation of the European methodology of participation –
EASW, took place in Porto Torres.
118

3. Technology Program
The simulated scenario, which lies at heart of our Strategic Scenario Workshop, allows us to
imagine the kind of scenario that might arise if a certain program or project actually took place.
The choice of this methodology, as a final part of the research on the socio-economic acceptancy
parameters of nuclear fusion, is based on its wide possibility of application and reiteration; its
ability to outline barriers and succeeding elements and to make hypotheses for the future,
supported by local people.
The 50 participants in the workshop, were chosen by a technical staff and the local
Administration of Porto Torres. Young people and students were mainly chosen.
The lab program was focused on three main points: the first part was an introduction, made by Mr.
Borrelli, to subjects and aims of SERF2; then came a lecture on fusion technology, held by Mr.
Pizzuto and Mr. Valli (ENEA engineers); finally, explanations on a simulated scenario of the Porto
Torres Research Centre on fusion were given by Mr. Bastiani (National Monitor EASW).
3.11 MATERIALS
3.11.1 Manufacturing of improved Polymer Impregnation and Pyrolysis (PIP)
composites
The aim of this manufacturing campaign was to realise and characterise composites with a real
3-dimension texture, in order to assess its influence on thermal and mechanical properties.
Currently, 3-D SiC/SiC f composites are under development worldwide, but it is difficult to find
composites with a fibre yarn crossing the entire thickness: this is the case of the SiC fibre
performs used in the present work. The above mentioned activity was devoted to realising 2-D
and 3-D composites with the purpose of drawing a figure of the density and the mechanical
properties obtained.
Fibres used were High Nicalon from Nippon Carbon. The typical dimensions of fibre panels
were 100×100×3.5/ 4 mm3. The 2-D texture was performed by Nippon Carbon in a satin 8 hs
fashion; the 2-D panels were realised by overlapping 6 layers.
The 3-D samples fibre bundles were woven in a three dimensional way by Techniweave (USA),
with two different fibre percentage through the thickness (z direction) (panel performs T1 and
T2). The total fibre volumetric percentage was about 40%; the relative fibre percentage in the
thickness was 25% and 50% of the fibre amount.
The SiC performs underwent a 0.1 µm carbon deposition, and a 0.2 µm SiC deposition by
Chemical Vapour Infiltration (CVI). The final densification was carried out by using several PIP
cycles: Polycarbosylane (PCS) polymer from Nippon Carbon and, for the latest PIP cycles,
Allyhidropolycarbosylane (AHPCS) polymer from Starfire Company (US) were used.
The final density reached, measured on rectified and cut 3-D panels, gave values of up to 2.4 g/cm3.
Three-point bending testing was performed to evaluate the Modulus of Rupture (MOR) (span 40
mm, cross-head speed 1mm/min). The mean results obtained with the 3 point bending test at RT
are collected in table 3.XI.
The bending strength of 3-D panels is remarkable, especially if compared to the density values.
The strength improvement, as compared to the results obtained by testing previous composites
manufactured by us, is due to the following:
• Better properties of SiC fibres; Hi Nicalon respect Nicalon CG;
• Lower thickness: 4 mm against 6 and 10 mm;
119

3. Technology Program
Table 3.XI - Hi-Nicalon composite mechanical test
results
Sample name MOR (MPa) Deflection (mm)
T1-2 627 0.8
T2-2 701 0.85
2D-1 555 0.8
• Intrinsic characteristics of 3-D texture which,
initially conceived to improve thermal diffusivity,
also leads to a good mechanical behaviour;
• Final densification with AHPCS polymer, which
provides an almost stoichiometric SiC matrix after
pyrolysis.
3.11.2 Development of high performance SiC
fibres composites
In the year 2000, a new manufacturing campaign has
been undertaken, aimed at realising composites with superior properties. The basic idea of this
activity is to improve the performances obtained with the fully three-dimensional Hi-Nicalon-
Polymer infiltrated and pyrolised composites within the task TTMA-001.7, by using advanced
stoichiometric SiC fibres produced by UBE-Japan and commercially named Tyranno SA. A
mixed CVI-PIP technique will be used to realise the matrix. In particular, for the PIP process,
allyl-hydrido-polycarbosilane from Starfire-USA will be used to realise a stoichiometric
polycrystalline SiC matrix. A considerable number of fibre performs (2-D and 3-D) have been
realised. In particular, the 3-D performs have approximately 25 and 40 % fibre percentage
through the thickness, in order to obtain a composite with relevant thermal conductivity through
the thickness by taking advantage of the high conductivity of the Tyranno SA fibres (about 60
W/mK). The overall manufacturing campaign is expected to be completed by the end of 2001.
3.11.3 Development of polymer-based joining techniques for SiC/SiC f
composites
SiC/SiC f composites were joined by using a preceramic polymer and fillers. Various parameters
were considered and varied.
The experimental work was carried out by using two SiC/SiC f composites, namely.
CERASEP N3–1 and a bi-directional ENEA PIP composite. The preceramic polymer used for
the joining experiments was the (allyl-) hydrido-polycarbosilane (HPCS, Starfire Systems). Its
pyrolysis in inert atmosphere yields a nearly stoichiometric amorphous SiC ceramic, with a
measured ceramic yield of about 79 % (at 1200°C).
Some inert filler powders (β–SiC powder, mean diameter=0.4 to 0.6 µm) were introduced.
Powders were added in various amounts to reduce shrinkage.
The joint quality was determined by microstructural examination, using SEM and by shear tests,
performed following a modification of the ASTM D905–89 test procedure. The x-ray diffraction
(XRD) analysis shows that pyrolysis at 1200°C of pure preceramic polymer yields nanocrystalline
β-SiC, and the introduction of β-SiC filler powders significantly increases the
crystallinity of the ceramic joining mixture. No phases containing SiO 2 were detected.
The shear strength for joints obtained with pure HPCS (without fillers) is very low, even though
it is higher than what observed when joining the same samples with a different preceramic
polymer (SR350 silicone resin). On the contrary, the maximum shear strength for joints obtained
with HPCS with SiC powder fillers is 16 MPa.
3.11.4 Development of sealing coating for SiC/SiC f composites
A double-layer glass ceramic coating was optimised and proposed as a sealing and self-healing
material of fusion reactor components made of SiC/SiC f composites. The composition of the
first layer (facing the SiC/SiC f substrate), called SAMg, is 30% in weight Al 2 O 3 , 20 % MgO
120

3. Technology Program
and 60% SiO 2 , while that of the second layer, called SABC, is 59.4w % SiO 2 , 16.1% Al 2 O 3 ,
13.2 % BaO, 11.3 % CaO. Depending on the coating thickness, the above compositions can be
considered as a low activation one, in the sense that the neutron-induced activation of the coating
thus formulated doesn’t exceed that of the SiC/SiC f composite itself.
The first coating was conceived as a composite sealer while, in principle, the second one is able
to self-heal cracks which may appear during reactor operation because of the temperature
increase. Both the coatings showed good chemical and physical compatibility with the SiC/SiC f
substrate, and stability at high temperature (1200°C). The SAMg coating (sealer) can be
produced by using both the traditional high-temperature glass fabrication method and the
mechano- synthesis process (ENEA NUMA).
Room temperature permeability tests, carried out on SiC/SiC f samples coated by the SAMg
layer, have shown a two-order magnitude reduction of the permeability; but the values measured
are not yet adequate for application of such coatings to SiC/SiC f composite fusion reactor
blankets. Possible improvements by changing the basic composition are under study.
Small irradiation samples have been produced and sent to ECN-Petten to investigate the coating
stability under neutron irradiation during an experimental campaign to be performed in the High
Flux Reactor (HFR).
3.11.5 Development of a design methodology for components made of
SiC/SiC f composites
The SiC/SiC f composite progressive matrix cracking makes use of this material only for
applications in which the material can operate in the elastic field of the stress-strain curve. For
this reason, a statistical methodology of design has to be used to post-process stress data
determined by a FEM code. The reference material for fusion applications is the SIC/SiC f
CERASEP N3-1® (3-D) with NICALONΤΜ CG fibers. Four-point, three-point and short-span
bending tests have been carried out to determine the mechanical properties and the statistical
parameters of the matrix first cracking. The Mosaic Model and the Fiber Undulation Model have
been used to predict the composite elastic constants.
The design methodology can be summed up in three steps. The material characteristic
parameters are input values to the FEM calculation and have to be evaluated at the first step. The
second step can be performed by a computer code, modeling the material orthotropic behaviour
and simulating the component operational conditions. Finally, the component reliability to the
matrix first cracking stress can be evaluated by the two-parameter Weibull distribution of the
probability of fracture. The Non Interactive Model (NIM), based on the Weibull distribution for
each critical failure mode, was used to perform the matrix first cracking reliability analysis. The
statistical parameters (two-parameter Weibull distribution) were evaluated by the Linearization
Method and the Maximum Likelihood Estimation Method (MLEM), from failure data of
standard specimens. The material non-linear behaviour and the threads damage evolution were
described from a micro-mechanical point of view, thus defining the analytical method to perform the
failure analysis. The three-parameter Weibull distribution of the ultimate tensile strength of the fibers
was determined by the MLEM. The Global Load Sharing (GLS) method, along with the Cumulative
Weakening Model (CWM), was finally used to describe the evolution of the threads damage in the
matrix crack plane.
3.11.6 Compatibility of SiC/SiC f composites with Pb17Li
In the year 2000, a new activity was launched in order to study the compatibility between
SiC/SiC f composite and liquide Pb17Li at a temperature of about 550°C for a significant
exposure time (100, 1000 and 6000 hs) in physico-chemical conditions representative of those
121

3. Technology Program
Table 3.XII - Young dynamic module and density of
SiC / SiC f composites
Sample E long (GPa) E flex (GPa) Density (gcm -3 )
A-S3-1 233 2.31 2.45
A-S3-2 240 242 2.50
A-S3-3 248 253 2.53
A-S3-4 238 243 2.53
A-S3-5 229 231 2.47
A-S3-6 242 243 2.51
A-S3-7 246 251 2.54
A-S3-8 235 238 2.49
A-S4-9 315 328 2.73
B-S3-1 245 2.48 2.49
B-S3-2 236 238 2.49
B-S3-3 243 245 2.52
B-S3-4 245 253 2.54
B-S3-5 240 244 2.52
B-S3-6 241 252 2.54
B-S3-7 232 236 2.48
B-S3-8 242 252 2.56
B-S4-9 308 317 2.71
C-S3-1 237 238 2.49
C-S3-2 225 221 2.43
C-S3-3 238 238 2.47
C-S3-4 238 242 2.50
C-S3-5 244 245 2.53
C-S3-6 241 248 2.52
C-S3-7 240 243 2.52
C-S3-8 236 245 2.55
C-S4-9 311 317 2.72
of the TAURO blanket. The aim of the activity is
to quantify the degradation of mechanical and
elastic properties of composite due to corrosion
damage, if any, by using non-destructive
techniques that include geometrical dimensions,
mass variation, longitudinal and torsional
dynamic module of elasticity (by the longitudinal
and torsional fundamental resonant frequency
method).
The list of the materials to be investigated
includes CERASEP N3–1 and N41 and ENEA
PIP composites. Large variations in geometrical
and physical properties of the as-received
composite were observed. In order to evaluate the
interaction between the composite fibres and
liquid Pb17Li, the CVD SiC coating, when
present, was abraded and a specimen was added
to each batch.
Then, since those large variations could mask the
change of properties when considering their
average values, characterisations will be
performed on each individual sample before and
after exposure (table 3.XII).
The exposure will take place at the LiFus2 facility
located at ENEA Brasimone. The design of the
facility upgrading devoted to increasing the
exposure temperature has been completed, and
the real set up is going on. The exposure phase
will be carried out in the year 2001.
3.11.7 Low cycle fatigue (LCF) of
Reduced Activation Ferritic Martensitic
(RAFM) steel in water with additives
The objective of the testing campaign carried out
during 2000 has been the study of the effects of
the mechanical parameters (stress amplitude and
stress rate) on the LCF behaviour of F82H mod.
steel in a high-temperature water coolant environment [3.55]. The results of this investigation
complete the exploratory work performed in 1998-99 about the influence of water chemistry and
material microstructure [3.55] on fatigue performances of steel.
The experimental environment conditions selected were the following: flowing air at 240°C (for
base-line data); flowing oxygen-free water (9 l/h, O 2

3. Technology Program
Table 3.XIII - cycles to specimen rupture and fracture modes obtained on F82H groups I, II, III after tests at 240°C
GROUP I GROUP II GROUP III
407 MPa 407 MPa 385 MPa 407 MPa 407 MPa 385 MPa 407 MPa 407 MPa 385 MPa
0.03 Hz 0.01 Hz 0.03 Hz 0.03 Hz 0.01 Hz 0.03 Hz 0.03 Hz 0.01 Hz 0.03 Hz
AIR 4300±100 4700±100 11500±300 4000±100 4500±100 11300±200 4200±100 4300±100 11600±200
Fatigue Fatigue Fatigue Fatigue Fatigue Fatigue Fatigue Fatigue (11500±200)
Fatigue
WATER 3300±50 2550±20 8000±100 1250±30 1250±20 4200±100 1750±20 1750±20 4950±50
Fatigue Plastic Fatigue Fatigue Plastic Fatigue Fatigue Plastic (5600±100)
Fatigue
Fig. 3.48 - Fracture overviews and details of groups I, II, III specimens after LCF tests at 18kN load
amplitude and 240°C
reduction observed in high purity-oxygen-free water at 240°C, as compared to air data, could
derive from a classic fatigue process (favoured at lower stress amplitude and higher stress rate)
or from plastic collapse (favoured at higher stress and lower stress rate).
• The frequency-dependent lifetimes associated with fatigue failures were due to crack
nucleation/growth enhancement induced by water.
• Plastic instability, determining cycle-dependent lifetimes, appeared to be related to tensile strain
accumulation and could be due to an environment-induced Bauschinger or ratcheting effect.
123

3. Technology Program
Fig. 3.49 - a, b, c, d overviews of group I, II, III specimen fracture surfaces after LCF tests at 17kN
load amplitude aand 240°C; a', b', c', d' magnified views near crack origins
• The increasing susceptibility to plastic instability, as tensile residual stresses introduced by
machining increased, denoted the importance of surface state, and the necessity to remove any
specimen surface variability in the future testing campaigns.
3.11.8 Mechanical properties of RAFM steel base material and joints
During the year 2000, some relevant results have been achieved in this field, in spite of a pending
supply of specimens. The Thermo-Mechanical Fatigue (TMF) machine was modified in order to
be able to perform thermal cycling also on non-magnetic alloys; very good results were obtained
in testing the AISI 316 LN steel (the so-called ITER heat).
From the standpoint of the activities related to fatigue and creep fatigue, a few tests were carried
out on already available, “spare”, specimens of F82H mod. for completing the study of
temperature boundaries effects on TMF resistance of that material. An effective semi-empirical
correlation between testing parameters and safe-life was validated.
Structural investigation of welded joints (carbide and precipitates on molten and HAZ of F82H
mod) were carried out by means of x-ray diffraction (in collaboration with University of Tor
Vergata, Rome). Electron Beam joints will be manufactured at ENEA Casaccia in order to have
a comparison with weldments made by CEA.
Finally, tensile and impact tests were made on a commercial Oxide Dispersion Strengthened
(ODS) steel, the PM 2000 alloy, produced by Plansee GmbH. From the point of view of tensile
strength, this alloy seems rather more resistant than RAFM steels already developed, at least
concerning the test temperatures investigated up to now, which present an acceptable ductility.
Anyway, since the alloy is not optimised as to its toughness, energies absorbed at room
temperature are roughly one tenth the RAFM steel already tested.
3.11.9 Microstructural investigation of the effects in RAFM steels using Small-
Angle Neutron Scattering (SANS)
In 2000, the study of He-effects in irradiated martensitic steels by SANS was continued in the
frame of Subtask TTMS001.11 [3.56]. The samples investigated were provided by the Research
Centre Karlsruhe. F82H-mod. steel samples He-implanted (400 appm) and subsequently
124

3. Technology Program
tempered at high temperature, which had already been studied in 1999, have been investigated
over a wider experimental range, thus allowing for a more accurate evaluation of bubble volume
fraction and average size. It appears that a uniform distribution of bubbles, approximately 15Å
in average size, produced by implantation at 250°C, evolves into a bimodal one during postimplantation
tempering, with bubbles as large as 100-200Å. OPTImized FERricic (OPTIFER),
Oak Ridge National Laboratory-Tennessee (ORNL) and MArtensitic NET (MANET) steels,
neutron-irradiated at 250°C up to 0.8 dpa, have also been investigated. The results indicate that,
in the first two cases, there is poor or negative difference between irradiated and reference
samples, which would imply microstructural stability for such irradiation conditions in these
samples. In MANET, the sample irradiated has a much stronger SANS intensity, possibly
relating to Cr effects. Unirradiated EUROFER 97 samples, submitted to relevant metallurgical
treatments, have been investigated as well.
3.11.10 Mechanical characterisation of materials with miniaturised specimens
The development of a portable apparatus called: Flat-top Indentor for Mechanical Characterisation
(FIMEC) was continued in the year 2000. On the basis of the encouraging results obtained by using
indentors of smaller diameters (up to 0.5 mm), in order to reduce the penetration load and thus the
size of the counteracting structure, a demonstrative apparatus for “in situ” testing has been designed,
realised and tested. The results, in terms of load indentation curves, showed a sufficient
reproducibility and are in good agreement with those obtained by using the full-size equipment. A
significant effort was paid in order to study an attachment system, able to provide a sufficient stiff
connection with the metallic structure; in particular, two different solutions were adopted to connect
the indenter to flat and cylindrical structures (such as tubes).
Development is still going on to realise a prototypical apparatus.
3.12 LIQUID METAL AND HYDROGEN/MATERIAL INTERACTION
TECHNOLOGY
3.12.1 Interaction between lead-lithium alloy and water in conditions relevant
for DEMO
In the reference design of WCLL blanket for DEMO, the use of double wall cooling tubes is
envisaged. The adoption of this solution allows to lower the probability of a water leak into the
Pb17Li breeder and, at the same time, to reduce the tritium permeation rate from the blanket
towards the cooling system. Nevertheless, the leakage probability is still not negligible and the
interaction between water and lithium lead eutectic alloy still remains of the biggest concern for
this concept of blanket. In the frame of the EU long-term tasks TTBA-5.1 and TTBA 5.2, ENEA
is involved in experimental activities concerning the interaction between lithium lead eutectic
alloy and water, due to micro-leaks from the cooling system and to large leaks as a consequence
of the rupture of a cooling tube inside the blanket.
With reference to the study of the effects of water large leaks, the LIFUS5 apparatus has been
operated throughout 2000 at the ENEA site of Brasimone.
The pressure and temperature evolution in the reaction, as well as in the expansion vessels
S1–S5, are detected by means of fast sensors, in order to create an experimental basis to study
the violence and the effect of the interaction. In 2000, two experiments have been carried out,
namely test No 2 and test No 3.
Test No 2 was performed to the purpose of better understanding the behaviour of water-cooled
pressure sensor placed in S1, and of trying to identify the reasons for some unclear points in the
pressure and temperature evolution during the first test, performed at the end of ‘99. Test No 2,
125

3. Technology Program
in which the water injection pressure was 100
bar, demonstrated that the behaviour of the
pressure transducers, when cooled with water
(which is necessary when the temperature
exceeds 350°C), was not perfectly reliable,
because of the formation of plugs of frozen
alloy around the transducers membrane,
which impaired their performance.
On the basis of the results achieved in the
second experiment, the following technical
modifications on LIFUS5 were executed
before carrying out test No 3:
• A fast pressure transducer, of the same
Fig.3.50 - Pressure evolution in the reaction vessel (PT2-PT9) type as those installed in the reaction vessel,
and the expansion vessel (PT1) of LIFUS5
was added at the top of the expansion vessel;
• New thermocouples with a larger diameter
(1 mm) were installed in the reaction vessel,
in order to ensure a better mechanical resistance to the jet impact;
• A new data acquisition software in LabWindow environment was set up.
Test No 3 was carried out in conditions relevant for DEMO reactor. Particularly, the water
injection pressure was 155 bar, while the temperature of liquid lead lithium alloy was kept at
330°C, in order to avoid the need for cooling down the pressure transducers. In this experiment,
the pressurised water was injected at 155 bar, which is the reference value for the primary
coolant in the WCLL DEMO blanket. The time of water injection is six seconds, while the whole
duration of the acquisition system is thirty seconds. It must be pointed out that the water pressure
injection is kept constant during the whole test. The results, in terms of pressure evolution in the
reaction and expansion vessels, are shown in fig. 3.50.
During the first phase, the steam-expanding jet forces the liquid metal up into the expansion
tubes and into the other sectors of the reaction vessel. The first phase is over after about 200 ms,
when the pressure reaches its maximum at around 105 bar.
The second phase is characterised by a pressure decrease in all sectors of the reaction vessel,
because of the free flow of gases into the expansion vessel, which is not balanced by an
equivalent injection of water from the injection device. As a matter of fact, the pressure decrease
in the reaction vessel takes place when the pressurisation of the expansion vessel starts. As soon
as the pressures are balanced in the two tanks, a further pressure evolution in the reaction and
expansion vessels is observed. This second phase lasts about 400 ms.
Once the free volume of the expansion vessel is pressurised, the third phase starts, in which a
further pressure increase takes place in both the reaction and the expansion vessels, due to the
further water injection and hydrogen production. The rate of pressure increase is lower than in
the first phase because of the reduced ∆P between pressurised water and reaction vessels and,
consequently, the lower water injection rate.
In this test, also a significant temperature increase took place. At the central point of the coneshaped
reaction zone, a temperature increase of about 160°C was detected. The next
experimental campaigns will be aimed at evaluating the role of water temperature and system
geometry on the pressure and temperature evolution.
In order to reach a deeper understanding of the effect of water micro-leaks into flowing Pb17Li
under operative conditions relevant for WCLL DEMO blanket, the RELA III loop was designed
126

3. Technology Program
and built in ‘99. Two tests have been carried out in 2000, aimed at achieving final answers in
terms of the following:
• Release dynamics of generated hydrogen;
• Characteristics and behaviour of the solid reaction products;
• Possible deterioration of the heat exchange properties between liquid metal and coolant,
following the growth of solid reaction products around the cooling system tube bundle.
Fresh Pb17Li was used in these two tests in order to have a more accurate determination of
hydrogen generated by the reaction of lithium with water.
Both tests confirmed the following:
• The kinetics of hydrogen generation is fast and the hydrogen produced is quickly detected in
the free volume of the pump vessel. As a consequence, it is possible, in principle, to detect a
small water leak in the WCLL blanket, just as an increase in the hydrogen concentration in the
drain/expansion vessel foreseen in DEMO;
• The solid reaction products, mainly Li 2 O and LiOH, aggregate around the tube bundle of the
cooling system. This is due to the low mean velocity (about 5 mm/s) of the liquid metal in the
blanket, as foreseen in the reference design of WCLL DEMO blanket. Due to this aggregation
and formation of corrosion products, in some cases the water micro-leak decreases.
• No hot spots in the liquid metal were detected, because of the degradation of the global heat
exchange coefficients between the liquid metal and the cooling fluid.
3.12.2 Qualification of tritium permeation in Pb17Li/gas
The control of tritium losses is an important issue in fusion technology because of its safety and
operational implications. A reduction in hydrogen permeation rate can be obtained by using
Tritium Permeation Barriers (TPB).
The coating materials on the steel surface should have a low tritium permeability and a good
compatibility with the aggressive environment in which their operation is foreseen (liquid metal,
irradiation, etc.). Aluminium-rich coatings, forming alumina as a top layer, appear to be a
promising solution as TPB, on the basis of previous results concerning the reduction of hydrogen
permeation in gas phase. Chemical vapour deposition and hot dipping are the candidate
techniques to produce the blanket segment coating, but their behaviour in liquid metal is not yet
fully characterised. Hydrogen permeation through TPB in liquid metal, which is a study of basic
importance in order to complete the final selection of the coating technique, will be investigated
on the experimental apparatus VIVALDI, designed and installed in 2000 at the ENEA site of
Brasimone. Two hollow cylinders of EUROFER 97 low-activation martensitic steel of 10 mm
diameter and 1 mm thickness with a length of 250 mm are used in VIVALDI. One of them is
aluminised in Hot-Dipping or CVD, while the other one, not coated, is the reference specimen.
The VIVALDI apparatus was designed to directly measure the Permeation Reduction Factor
(PRF) of the coated specimen as compared to the bare one, in the same experimental condition
(hydrogen pressure and composition, temperature, etc.), thus easily comparing the gas flux in the
specimens.
The internal surface of the specimen is initially in contact with vacuum (10-5 Pa) while the
external surface is exposed to hydrogen gas with a nominal purity of 99.9999%. When
performing a measurement in Pb17Li phase, the hydrogen continuously bubbles through the
liquid. Hydrogen permeates the sample, and causes a pressure rise in the inner volume. The
pressure rise can be converted into the amount of gas in moles which permeate the unit area of
the sample per second. The procedure can be repeated for different temperatures and gas flowrates.
The experimental procedure is the same in gas and in Pb17Li phase.
127

3. Technology Program
Fig. 3.51 - Permeability of two tubular specimens of
EUROFER 97 as a function of temperature
The hydrogen permeation measurements in
two EUROFER 97 bare specimens in gas
phase, in comparison with literature data of
permeation in disk-shaped specimens, are
shown as Arrhenius plot in fig 3.51. The
permeation rates were determined by using
a hydrogen driving pressure of about 105
Pa. The results are quite similar and in good
agreement, confirming the symmetry of
measurement lines and reliability of the
system. The experimental campaign on
coated samples in liquid metal will be
completed in 2001.
3.12.3 Transport parameters and
solubility of hydrogen in Pb17Li
The knowledge of hydrogen isotopes mass
transfer parameters in the liquid Pb17Li
alloy is of basic importance for the design
of several tritium processing systems in DEMO fusion reactor, particularly for the tritium
extractors from Pb17Li, based on the technology of gas-liquid contactors.
The mechanisms of the overall hydrogen mass transfer from the molten liquid Pb17Li, in contact
with a gaseous atmosphere, can be described by the following sequence of steps:
• Transport of hydrogen by diffusion in the bulk liquid metal;
• Transport of hydrogen by diffusion through the liquid transition layer adjacent to the liquidgas
interface;
• Reaction involving hydrogen recombination at the gas-liquid interface;
• Transport of hydrogen molecules by diffusion through the gas transition layer;
• Transport of hydrogen by diffusion and convection in the bulk gas.
From previous experiments and theoretical considerations, steps involving diffusion in gas phase
are considered to be fast as compared to the transport phenomena through bulk liquid, liquid
transition layer and gas-liquid interface. However, it is difficult to say what the controlling
mechanism will be like in the overall desorption kinetics. Of course, it depends on many
parameters, such as hydrodynamic conditions of the liquid-gas system, gas composition, and liquid
metal purity level.
Researchers from different laboratories achieved results which often do not agree with each
other, based on different experimental set-up and conditions. In some cases, it was found that the
main resistance to hydrogen mass transfer was located in the bulk liquid; in other cases, the
controlling step was claimed to be the recombination of hydrogen atoms at the liquid-gas
interface and values of absorption; therefore, recombination coefficients were determined.
Generally, all the tests carried out in the past came from two kinds of experimental methodology,
namely:
• The desorption method, based on the pressure increase evolution in a closed chamber, caused
by the desorption of hydrogen from a previously hydrogen-saturated specimen of liquid metal;
• The absorption method, in which the hydrogen pressure decreases or the liquid metal weight
increases, takes place as a consequence of the gas absorption by the liquid metal kept in contact
with hydrogen at an initial given pressure.
128

3. Technology Program
In order to follow a different approach for
the experimental set-up, the device LEDI
(fig. 3.52) was designed and installed at the
ENEA site of Brasimone, in the framework
of the EU task TTBA-4.1. It is based on
hydrogen/deuterium permeation through a
thin layer of Pb17Li, stagnant over a
metallic membrane. As for any permeation
experiment, the mass transport parameters
can be determined by fitting the
experimental pressure evolution in a
vacuum chamber by means of a suitable
mathematical model. This system seems to
be more flexible, thus giving the
opportunity to vary the thickness of the
liquid metal as well as the surface
conditions, which can strongly affect the
kinetics of the whole transport mechanism.
The experimental activity is expected to be
performed in a wide range of operative
conditions in 2001.
3.12.4 Feasibility study of a
modified concept of WCLL DEMO
blanket
In the framework of EU task TTBA-4.2, an
extensive analysis of a WCLL DEMO
blanket is being carried out, in which
tritium permeation barriers on the external
surface of the breeder cooling pipes are not
used, or are reduced during the required
performance.
The activity in this field started two years
ago by evaluating the technical and
economical feasibility of such concept of
WCLL blanket, which provides for the Fig. 3.52 - Schematics of the experimental device LEDI
variation of the tritium permeation rate
allowed into the primary cooling system
and the technology to extract tritium from tritiated water (electrolysis, distillation coupled to
vapour phase chemical exchange, CECE process). The results concerning this first part of the
activity were positive and indicated a good technical economical feasibility of the tritium
management system, up to a tritium permeation rate of 10÷15 g/day, accepting a higher tritium
activity in the primary coolant, in order not to over load the water detritiation system. The study
was continued in 2000, to evaluate the impact of a well higher tritium specific activity in the
primary coolant on safety aspects, both under normal running conditions and in accident
conditions. A wide parametric analysis was carried out by varying the type of the reactor
containment system, the tritium specific activity, the enthalpy of the coolant (different positions
for a LOss of Coolant Accident (LOCA) were assumed). Tritium release to the environment in
case of a LOCA event was found to satisfy the limits recommended by the ITER Project Design
Guidelines for tritium release accident, while the respect of the limits in normal operation
strongly depends on the water leak rate from the primary to the secondary circuit.
129

3. Technology Program
3.12.5 Corrosion and mechanical tests on EUROFER 97 in Pb17Li
The investigation of the mechanical property degradation of the RAM-F steels is of major
concern for the design and development of fusion reactors cooled by liquid metals. The activities
foreseen to this purpose, developed in the framework of EU task TTMS-003-D13, consist of
tensile tests carried out on specimens of EUROFER 97 steel, pre-exposed to flowing Pb17Li for
1500-5000h, at 480 °C.
The first tensile tests were performed during 2000 on specimens exposed to flowing Pb17Li, at a
velocity of 0.6 l/h, in the LIFUS II loop, at 480 °C for, 1500 and 3000 h. It must be pointed out
that the tensile tests were performed at the same temperature at which the specimens were exposed
to liquid metal. Moreover, besides the tensile specimens, also corrosion specimens were exposed
to Pb17Li under the same conditions, in order to evaluate the corrosion rate.
The weight variation measurements, performed after dipping the corrosion samples in a solution
of acetic acid–hydrogen peroxide–ethanol, are reported in table 3.XIV.
These results clearly show that dissolution of the steel elements into the liquid metal occurs. In fact,
the corrosion mechanism is evidenced in the SEM-micrograph of fig. 3.53.
The tensile tests results are summarised in table 3.XV.
As it can be seen from these
results, a degradation of the
mechanical properties of the
EUROFER 97 steel did not occurr.
The continuation of the activity is
foreseen throughout 2001,
together with a new experimental
campaign on SiC/SiC f material,
which will be carried out after a
modification of LIFUS II loop.
3.12.6 Lithium corrosion
and chemistry for IFMIF
target
Fig. 3.53 - Cross section of the EUROFER 97 sample
exposed to flowing Pb17Li at 480°C for 3000 h. The sample
was not treated with the rinsing solution and the whitecoloured
layer is Pb17Li
In the framework of IFMIF
activities (task TTMI-002-D4),
lithium corrosion and chemistry
studies will be conducted by
means of the following actions,
foreseen for 2001:
Table 3.XIV - Weight change
measurements
Exposure time Weight change
(h) (mg/mm 2 )
1500 -4.4×10-2
3000 -5.9×10-2
Table 3.XV - Tensile tests results
Exposure RP 0.2% RM A5 % Z%
(h) (N/mm 2 ) (N/mm 2 )
0 433±16 482±28 22±2 81±2
1500 413 ±22 458±24 25±1 79±3
3000 399 ±11 438±16 26±1 83±2
130

3. Technology Program
• Evaluation of the most promising techniques for monitoring impurities in liquid Li;
• Development of monitoring systems suitable for measuring N, C, H in liquid Li;
• Selection of trapping material and suitable warm-trap temperatures.
These actions will be performed in collaboration with University of Nottingham. The design of a
lithium loop, in which corrosion tests under controlled conditions are foreseen, will be made by
taking into account the possibility of re-adapting the loop existing at the ENEA-Brasimone centre.
3.12.7 Hydrogen permeability and embrittlement in EUROFER 97 martensitic
steel
To be used as a structural material in the DEMO blanket, EUROFER 97, which belongs to the
family of 7÷10% Cr reduced-activation martensitic steels, must be adequately characterised as
far as its compatibility with a hydrogen environment is concerned.
For this reason, the experimental activities were focused on the following:
• Determination of hydrogen/deuterium permeability in the temperature range 473÷723K;
• Determination of hydrogen permeation diffusivity at room temperature and density of
trapping sites using Devanathan’s technique;
• Evaluation of the hydrogen embrittlement susceptibility at room temperature.
In order to gather data on all of these phenomena, permeation experiments have been performed
on EUROFER 97 in the temperature range 473÷723 K with the device “PERI” (gas phase
technique), by using a hydrogen or deuterium upstream pressure of about 75000 Pa, at room
temperature with a Devanathan’s electrochemical cell.
Moreover, mechanical tests on hydrogen-charged EUROFER 97 steel specimens at room
temperature have been carried out to determine the threshold concentration of hydrogen for HE.
The study was based on tensile low strain rate tests, conducted on notched and smooth
cylindrical specimens, which had previously been electrochemically charged with hydrogen
(contents up to 3 wppm were employed).
The results of permeability F and effective diffusivity D for EUROFER 97, in comparison with
F82H steel, are presented in Arrhenius plots in figures 3.54 and 3.55, respectively.
On the basis of the experimental results, permeability, lattice diffusivity and Sieverts constant
K s,l for deuterium in EUROFER 97 were calculated as follows:
14470
−
−7
D = × e RT 2 −1
15 . 10 m × s
l
( )
⎛
⎞
−
K = × e − 23810
1
RT ⎜
−
3
−3
⎟
102 . 10 mol × m × Pa 2
sl ,
⎜
⎟
⎝
⎠
38280⎛
1 ⎞
−
= × e RT ⎜
−
−8
−1 −1
⎟
Φ 153 . 10
mol × m × s Pa 2
⎜
⎟
⎝
⎠
Significant hydrogen trapping was observed below 573K. In this region, the diffusion coefficient
drops sharply below the values obtained by a direct extrapolation at lower temperature. It is worth
noting that, in F82H, the trapping phenomena were evident only at temperature below 523 K.
131

3. Technology Program
Fig. 3.54 - hydrogen and deuterium permeability through
EUROFER 97 as a function of temperature
Fig. 3.55 - hydrogen and deuterium diffusivity in EUROFER
97 as a function of temperature
Diffusivity and density of trapping sites
were also measured by experiments of
hydrogen permeation at room temperature
using the Devanathan’s technique. The
effective diffusivity measured is in the range
6×10–12÷4.9×10 –11 m 2 /s. A concentra-tion
of trapped hydrogen of 0.48 wppm was
measured, which corresponds to a density of
irreversible traps of 2.2×1024 m -3 . This
value is in quite good agreement with the
density of irreversible traps, calculated by
fitting the diffusivity data extrapolated at
low temperature.
Phenomena of hydrogen embrittlement on
EUROFER 97 were studied at room
temperature by performing tensile tests on
Fig. 3.56 - Area reduction coeffients for notched and smooth specimens previously charged with
specimens charged with hydrogen
different hydrogen concentrations. All the
tests were conducted under displacement
control by using smooth and notched
specimens. All the experimental activities were performed in collaboration with University of
Pisa.
A marked decrease in the area reduction coefficient (fig. 3.56) was found at a low hydrogen content,
viz. about 1.8wppm for smooth specimens and 1.1wppm for notched ones. Therefore, hydrogen
susceptibility of EUROFER 97 is similar to that of other martensitic steels of the 7-8% Cr family,
developed in the frame of materials for fusion applications, but significantly more pronounced than
that shown by Manet II. The experimental activity on hydrogen embrittlement will continue
throughout 2001, with the aim of evaluating the hydrogen susceptibility in the temperature range
RT÷200°C.
132

3. Technology Program
3.12.8 Measurements of H/D diffusivity in and solubility through tungsten and
tungsten alloys in the temperature range of 600°C to 800°C (ITER task 436)
In the framework of ITER task 436, ENEA was charged of evaluating hydrogen transport and
solubility parameters in tungsten, which is one of the candidate materials for first wall and divertor
of ITER. In considering materials for fusion reactors, a detailed understanding of hydrogen
transport and solubility parameters is an important issue, because they strongly affect safety and
blanket performance aspects. Numerical codes have been developed for the calculation of
recycling, inventory and permeation of deuterium and tritium in fusion reactor design concepts in
non-steady-state conditions. Essential input data for these codes are permeability, diffusivity and
solubility of deuterium and tritium in the structural material involved.
Because of their refractory nature and good thermal properties, tungsten and tungsten-alloys are
considered to be alternatives to graphite as plasma-facing materials for ITER. Transport and
inventory parameters of hydrogen and its isotopes through these materials, determined in the past
and available in literature, are questionable because there is no agreement among different
investigations. Previous experiments were conducted at ENEA in the temperature range between
350°C and 500°C, indicating the low permeability of tungsten. The hydrogen permeability
obtained through tungsten was about 10-14 (mol m-1 s-1 Pa-1/2) at 500°C. A new experimental
device, named PERI 2, was designed to increase the maximum experimental temperature,
performing permeation experiments in the temperature range between 423 K and 873 K.
The method chosen for the determination of the hydrogen/deuterium transport and inventory
parameters through tungsten and tungsten-alloys is a gas-phase technique. The permeation
apparatus is made of standard stainless steel UHV components, except for the specimens
housing. The specimens flanged housing is realised in Tungsten-Zirconium-Molybdenum
(TZM), a molybdenum alloy, with a heating-refrigerating system able to guarantee a specimen
temperature in the range of 300°C–800°C, while keeping the external flanges at a maximum
temperature of 450°C; a temperature which is necessary in order to use standard Cu O-ring. The
specimen is sealed between flanges by using two gold O-rings.
TZM was chosen because of its excellent thermal resistance, high melting temperature, adequate
mechanical properties in the temperature range required for the experiment. As compared to pure
molybdenum, it presents a better creep resistance, higher recrystallisation temperature and better
high-temperature strength.
Also the permeability of hydrogen in TZM was evaluated. Data available in literature
demonstrate the low permeability of TZM as compared with pure molybdenum or nickel alloys.
Hydrogen leak through and solubilisation in the housing were also estimated; they resulted to be
negligible as compared with the hydrogen permeation through the specimen. The apparatus will
be available for operation in June 2001.
3.13 THERMAL FLUID-DYNAMICS
3.13.1 Tests on beryllium pebble bed by Small Rectangular Test Sections
(SMARTS)
The ENEA experimental test activity was focused on the interactions between the beryllium
pebble bed and the steel structure on a mockup, called SMARTS, simulating the thermo
mechanical loads by a flat electrical resistor. The final assembly of the SMARTS test section has
been completed by the end of 2000. Particular attention was devoted to the insertion of the 70
thermocouples, welded onto the cooling plates and onto both sides of the electrical heater, and then
inserted in the bed by means of a special supporting device. Seven load cells (three lateral ones per
side, and one on the upper plug of the cell) and a Linear Voltage Displacement Transducer (LVDT)
133

3. Technology Program
are also part of the instrumentation installed. The stresses induced by the differential thermal
expansion in the axial direction will be controlled by an appropriate elastic device, that will be
applied after filling of the pebbles, and which therefore cannot be seen in the pictures.
As far as the reliability of the electrical resistor assembly is concerned, the tests on a reduced
scale mockup started in mid 2000. The resistor worked with no ruptures for more than 1000
hours at a temperature above 900°C.
All the safety issues are being analysed in the safety report, which has been completed in 2000
and will be released at the beginning of 2001.
In this document, all components, circuits and possible accident sequences are analysed in detail,
with particular attention to any sequence which might lead to beryllium particles release and the
possibility of hydrogen formation by contact of cooling water with high-temperature beryllium.
The design of an oil scrubber, acting as an ultimate filtering unit able to trap the sub-micrometric
beryllium particulate, was started. The final testing on SMARTS test section should start at the
beginning of 2001.
3.13.2 Non-nuclear tests for the solid breeder blanket in the HE-FUS3 facility
In the frame of the EU fusion program, ENEA has the responsibility to carry out thermomechanical
tests on two experimental mockups, HE-FUS3 Lithium Cassette (HELICA), and
HE-FUS3 Experimental Cassette of Lithium Beryllium Pebble Beds (HEXCALIBER), [3.58],
reproducing a portion of the Helium-Cooled Pebble Bed (HCPB) TBM to be tested in ITER. In
particular, HELICA is a mockup with a single cell with lithiate ceramic breeder pebble bed,
whilst HEXCALIBER is a medium-scale mockup
with two lithiate ceramic breeders and two
beryllium pebble beds. The relevant manufacturing
procedures will be preliminarly checked in
HELICA, anticipating the final manufacturing to
be adopted for HEXCALIBER. The mockup
reference structural material is ASME SA 387
grade 91 ferritic-martensitic steel. The first
ceramic breeder to be tested will be Li 4 SiO 4 ,
whilst other lithiate ceramic such as Li 2 ZrO 3 and
Li 2 TiO 3 pebbles will be considered for future
tests. The real volumetric heat sources inside the
breeding zone will be simulated by flat electrical
heaters, located inside each pebble bed. The
heating plates will be constituted by a 1 mm tick
KANTHAL A-1 resistor plate, closed by two 1.6
mm tick INCONEL 718 cladding sheets. The
resistor and the internal surfaces of the cladding
sheets are coated by 0.15 mm alumina layers,
deposited by plasma spray.
Fig. 3. 57 - Electrical resistor and its simulacrum tested
at 1000°C
As far as the qualification of the design of the
electrical resistor assembly is concerned, a specific
test campaign on a reduced scale mockup (fig. 3.57)
has been performed throughout 2000 with the
purpose of verifying its reliability and performance
at high temperature. The resistor worked without
ruptures for more than 1000 hs at a maximum
internal temperature above 900-1000°C.
134

3. Technology Program
Fig. 3.58 – Tazza test section
Fig. 3.59 – Preliminary UCT result on Tazza
Fig 3.60 – Optic microscope (left) and SEM (right) images of Li 4 SiO 4 pebbles
At the beginning of 2000, a test campaign aimed to define an effective filling method and to
determine the main mechanical properties of the related beds, was launched by using the Tazza
test section, fig. 3.58 [3.59]. This campaign consisted in a set of Un-axial Compressive Tests
(UCT), performed both on beryllium (monosize and binary beds) and ceramic breeders (lithium
titanate and orthosilicate) pebble beds of different heights. A new effective filling method for
binary beds will be assessed. Very high packing factors were achieved by using an Ultra Sound
bath for the bed compaction. Over 50 beds have been built and tested by UCT, and the whole
campaign is now nearly over [3.60]. The main experimental results of the first tests performed
in Tazza are shown in fig. 3.59. The ratio between the Un-axial Deformation Module (UDM),
defined as the apparent Young’s module of the bed, and the Young modulus related to the full
dense material, is reported versus the Packing Factors (PF). These results also show that the polidispersed
beds behave more as a mono-size bed than as a binary bed. Furthermore, the roughness
of the pebbles seems to have a sensible influence on the PF, figs. 3.60, 3.61, whilst the higher is
PF, the higher is UDM.
135

3. Technology Program
Fig 3.61 – Optic microscope (upper) and SEM (down) images
of Li 2 TiO 3 pebbles
During 2000, after the results obtained from
the preliminary tests, ENEA has completed
and revised the previous Technical
Specification for the fabrication of HELICA
and HEXCALIBER, issued in 1999 [3.58],
including the drawings of the details for both
mockups. The contract for the relevant
fabrication was suspended in order to define
a new offer for the manufacturing of both
HELICA and HEXCALIBER mockups. The
manufacturing of the heating plates and their
relevant qualification has been successfully
concluded in mid 2000. A new 200 kVA
electrical supply unit, for both HELICA and
HEXCALIBER mockup resistors, has been
successfully tested on the HE-FUS3 facility
in mid 2000. The design of a special safety
tank for the HEXCALIBER mockup
containment and the beryllium safety
concerns has been concluded at the end of
2000 [3.61].
The HELICA manufacturing should be
launched by early 2001 and the relevant
testing activity will be finished by the end of
2001. The HEXCALIBER manufacturing
will start by middle 2001, to be finished at the
end of 2001.
3.13.3 Fabrication and testing of a
full-scale ITER divertor outboard
mockup
The full-scale ITER divertor outboard
mockup is constituted by a Cassette Body
(CB) and its actively cooled Dummy Armour
Prototype (DAP). This DAP consists of the
Vertical Target (VT), the Wing (WI) and the Dump Target (DT), which will be integrated by the
Gas Box Liner (GBL).
The DAP, with two significant parts, namely the VT and the couple WI-GBL, is mounted on the
CB. The temperature values and profiles, arising in the actual divertor due to the plasma heat
flux, are simulated by radiative electrical heaters facing the DAP first wall. The hydraulic
scheme of both parts of the test section, is organized in parallel channels in the toroidal direction.
The integration of the DAP onto the CB was finally positively checked. Mounting and
assembling of the DAP by their key-wedges, showed enough reliability to ensure the proper
mechanical constraints to the CB. The final dimensional tolerances of the facing parts of the
DAP-CB manifolds are too large to allow an easy final welding from the inlet side. Therefore,
the connection of these manifolds was fixed by using special screwed flanges of reduced
dimensions, fig. 3.62. However, these detachable fittings can ensure the final welding of the
manifolds from the external side.
Both thermal-hydraulic and thermo-mechanical theoretical analyses were carried out prior to the
136

3. Technology Program
Fig. 3.62 – DAP manifolds and their special screwed connection flanges
Fig. 3.63 - Theoretical and experimental velocity in VT
coolant channels
start of the testing campaign. The
hydraulic analyses have been
mainly performed in order to
determine the coolant flow rate
distribution inside the VT channels
and the heat transfer coefficients.
The comparison between the VT
theoretical and experimental
coolant velocity distribution is
shown in fig. 3.63. Therefore, a
significant flow rate variation is
evident, due to the manifold
asymmetry and the by-pass effect
on the furthest channels from the
coolant inlet/outlet [3.62]. A
thermo-mechanical analysis has
also been carried out, by taking
into account the theoretical
velocity distribution along the VT
channels, in order to know
temperature and stress
distributions on the mockup during
the cycling fatigue tests [3.63].
At the end of 2000, ENEA starts
the experimental characterisation
of the ITER Divertor Cassette
Experiment (IDICE) (fig. 3.64)
mockup by using the CEF 1-2
hydraulic facility at Brasimone.
The thermal hydraulic
experimental tests at steady-state
will be devoted to determining the
coolant flow rate and pressure
Fig. 3.64 - IDICE mockup with
instruments and electrical heaters
137

3. Technology Program
drop characteristic of the DAP components,
fig. 3.65.
During thermal fatigue cycling tests, the
DAP was heated by electrical radiative
resistor at a maximum flux of 0.14 MW/m2.
The thermal fatigue cycling stress was
obtained by regulating the heater heat flux
with a period of 900 s (120 s linear ramp-up,
720 s at a constant power and 60 s of linear
ramp–down). The thermal stress was
increased by superimposing the inlet water
temperature and switching it from 20 °C to
120 °C during the heat flux cycle, fig. 3.66.
After a few thermal fatigue cycles, a
consistent water leakage was detected from
a GBL tube. The rupture, monitored by an
endoscope, was localised in
correspondence of a previous defect, which
had already been repaired by brazing. After
the tube plugging and by-passing, the
fatigue test restarted, aiming at reaching a
total of 1000 cycles by February 2001.
3.13.4 Fatigue tests on six mockups
of the primary first wall panel
prototype (EFDA Contracts 00/529
and 00/533)
Fig. 3.65 - Experimental VT hydraulic characteristics
The objective of the contract EFDA 00/529
is to perform thermal fatigue tests on six
mockups of the primary first wall panel
prototype, realised in the frame of the ITER
Fig. 3.66 - DAP temperatures during the thermal fatigue cycling
138

3. Technology Program
EDA Task T216+, with dimensions of 250 (h)×66 (w)×80 (t) mm or 250 (h)×110 (w)×80 (t) mm.
The objective of the contract EFDA 00/533 is to perform thermal fatigue tests on two mockups
of the primary first wall panel prototype, realised in the frame of the ITER EDA Task T420, with
dimensions of 900 (h)×250 (w)×90 (t) mm.
All the mockups are representative of the
reference Be/DS-Cu/SS materials,
geometries and joining procedures,
fig. 3.67. The main aim of the
experimental fatigue tests is related to the
performance first wall beryllium of the tile
armour layer, joined to the Cu alloy rear
heat sink, as far as the Hipping
procedures, performed at different
conditions, are concerned.
The performances of the Be/DS-Cu/SS
joints will be checked by thermal fatigue
tests, reaching a maximum heat flux of 0.8
MW/m2 with a period of about 300 s up to
30.000 cycles. The thermal heat flux on
the first wall of the mockups will be
imposed by radiative electric heaters.
These resistors, made of CFC sheets, will
be mounted between the couple of facing
mockups to be tested. The mockups will
be layered on the Beryllium front surface
with a black pigmented, high emissivity
special paint, able to withstand up to 1000
°C. The mockups, properly assembled on
frames, will be hosted in two special glove
boxes (EDA-BETA and THESIS),
Fig. 3.67 – EDA mock-ups
Fig. 3.68 – EDA mock-ups temperature°C (a) and stress
calculations Pa (b)
139

3. Technology Program
Fig. 3.69 – PFW mock-ups temperature °C (a) and stress calculations Pa (b)
provided with piping and electrical feed-through, inert gas and vacuum system, and Beryllium
particulate filtering and cleaning system.
In 2000, ENEA has completed the thermo-mechanical calculations for determining temperatures
and stress in the mockups under real testing conditions, figs. 3.68,3.69.
ENEA has also performed the designs of the electrical power supply systems for mockup heating
[3.64, 3.65], instrumentation, external piping equipment and interfaces with the thermal-hydraulic
facility. The design EDA-PFW mockup, which hosts EDA-BOX and THESIS vacuum vessels, have
also been performed. At the beginning of 2001, ENEA will issue the orders for the main components,
such as the electrical power supply system. By spring 2001, both the experimental activities on the
EDA and PFW mockups will start and will continue throughout 2001.
140

References
[3.1] T. Kato et al., First test results for the ITER central solenoid model coil, presented at the
21st Symp. on Fusion Technology. (Madrid 2000)
[3.2] Y. Takahashi et al., Cryog. Engineer. 35, 7, 357 (2000)
[3.3] N. Martovetsky et al., CSMC and CS insert test results, presented at the 2000 Applied
Superconductivity Conference (Virginia Beach 2000)
[3.4] N. Martovetsky et al., First results on ITER CS model coil and CS insert, presented at the
14 th Topical Meeting on the Technology of Fusion Energy (Park City 2000)
[3.5] H. Tsuji et al., Progress of the ITER central solenoid model coil program, presented at the
18th IAEA Fusion Energy Conference (Sorrento 2000)
[3.6] E.P. Balsamo et al., Physica C310, 258 (1998)
[3.7] L. Savoldi and R. Zanino, Cryogenics 40, 179 (2000)
[3.8] L. Savoldi, P. .Michael and R. Zanino, Int. J. Mod. Phys. B14, 3183 (2000)
[3.9] L. Savoldi and R. Zanino, Cryogenics 40, 593 (2000)
[3.10] L.Savoldi and R.Zanino, Cryogenics 40, 539 (2000)
[3.11] B. Carmignani, G. Toselli, 1st Report on the work developed by ENEA team (preliminary
studies concerning phase 1 of work), Intermediate Report 1–ENEAMS-M-R-001, (February
1999)
[3.12] L. Bottura et al., Stability in a long length NbTi CICC, to be published in IEEE Trans.
Appl. Supercond..
[3.13] P. Bellucci et al., Comparison between the predictions of the thermo-hydraulic code
Gandalf and the results of a long length instrumented CICC module experiment, to be published
in Cryogenics.
[3.14] L. Petrizzi et al., Design of a welded box divertor for ITER FEAT, presented at the 21 st
Symp. on Fusion Technology (Madrid 2000)
[3.15] G. Brolatti et al., Final report on design of a welded box divertor for ITER FEAT,
private communication
[3.16] C. Alvani, P. Carconi and S.Casadio, J. Nucl. Mater. 280, 372 (2000)
[3.17] C. Alvani et al., Lithium titanate pebbles reprocessing by wet chemistry, to be published
in J. Nucl. Mater.
[3.18] U. Fischer et al., Fusion Eng. Des. 51-52, 663 (2000)
[3.19] S. Rollet, P. Batistoni and R. Forrest, Fusion Eng.& Des. 51-52, 599 (2000)
[3.20] M. Angelone et al., Fusion Eng. Des. 51-52, 653 (2000)
[3.21] L. Petrizzi, P. Batistoni ans I. Kodeli, Fusion Eng. & Des. 51-52, 843 (2000)
[3.22] K. Seidel et al., Fusion Eng. Des. 51-52, 855 (2000)
[3.23] I. Kodeli, L. Petrizzi and P. Batistoni, J. Nucl. Sci. Technol. Suppl.1, 713 (2000)
[3.24] M. Angelone, P. Batistoni and M. Pillon, Effect of encapsulating material on the
peak3/peak5 response ratio of TLD-300 irradiated with neutrons of various energy, presented at
the 8 th Symp. on Radiation Physics (Prague 2000)
141

References
[3.25] S. Rollet, M. Angelone and P. Batistoni, Nucl. Instrum. Methods in Phys. Res.
B166–167, 826 (2000)
[3.26] P. Batistoni et al., Experimental validation of shutdown dose rate neutron induced
gamma-ray radioactivity in various structural materials, presented at the 21st Symp. on Fusion
Technology (Madrid 2000)
[3.27] M. Pillon et al., J. Radioanal. Nucl. Chem. 244, 2, 441 (2000)
[3.28] B. Esposito et al., Fusion Technol. 51-52, 331 (2000)
[3.29] C. Blandin and B. Esposito, Design of sub-miniature fission chambers for the IFMIF
high flux test module, ENEA Report ERB5005 CT990059 EFDA/99-506 (November 2000)
[3.30] M. Tavani et al., Proc.5 th Compton Symp., AIP 510, 746 (2000)
[3.31] G. Barbiellini et al., Proc. 5 th Compton Symp., AIP 510, 750 (2000)
[3.32] M. Rapisarda et al., SuperAGILE - The x-ray monitor of AGILE, Proc. of the Frontier
object in astrophysics and particle physics (Vulcano 2000), p. 539
[3.33] S. Tosti, L. Bettinali and V. Violante, Development of Pd-Ag coating technique on
ceramic porous tube and preliminary ceramic membrane reactors (CMR) tests, ENEA Internal
Report FUS/TN/TS 007/00 (March 2000)
[3.34] S. Tosti and V. Violante, Production of membrane formed by sputtering of Pd-Ag on a
ceramic porous tube, ENEA Internal Report FUS/TN/TS 008/00 (March 2000)
[3.35] G. Chiappetta et al., Design and functional analysis of a closed loop pilot plant for
testing catalytic membrane reactors, ENEA Internal Report FUS/TN/TS 023/2000 (April 2000)
[3.36] S. Tosti and V. Violante, Long term performance testing of CMR in TBM tritium recovery
application, ENEA Internal Report FUS/TN/TS 052/00 (December 2000)
[3.37] S. Tosti et al., Sputtered, electroless, and rolled palladium-ceramic membranes,
submitted to J. Membr. Sci.
[3.38] S. Tosti et al., Procedimento di saldatura di lamine sottili di leghe metalliche selettivamente
permeabili all’idrogeno, in particolare per la realizzazione di dispositivi a membrana e apparato
per la sua realizzazione, Domanda di brevetto n. RM2000A000412 del 25.7.00
[3.39] S. Tosti et al., Fusion Eng. Des. 49–50, 953 (2000)
[3.40] S. Tosti et al., Korean Membrane J. 1,1, 1 (1999)
[3.41] S. Sandri, Radiological safety of the scheduled working activities at the main ITER
FEAT system components, FUS TN SIC 13/2000, Rev. 2 (November 2000)
[3.42] ITER generic site safety report, vol III, ITER JCT, Garching (Germany)
[3.43] ITER generic site safety report, vol V, ITER JCT, Garching (Germany)
[3.44] D.G. Cepraga et al., Radiation transport and activation calculation in support for safety
analyses of ITER-FEAT, ENEA FUS TN SIC 07/2000 (August 2000)
[3.45] ITER generic site safety report, vol VII, ITER JCT, Garching (Germany)
[3.46] T. Pinna and L. Burgazzi, Failure mode and effect analysis for water cooling system of
ITER FEAT, ENEA Internal Report FUS TN SIC (October 2000)
142

References
[3.47] L. Di Pace and D.G. Cepraga, Interim report on activated corrosion products evaluation
for the ITER TCWS, ENEA Internal Report FUS TN SIC TR 11/00, Rev. 1 (December 2000)
[3.48] M.T. Porfiri, P. Meloni, ISAS validation against ICE – experimental campaign 2000,
ENEA Internal Report FUS-TN SA/SC/R 09-00 (December 2000)
[3.49] G. Caruso, CONSEN validation against ICE – experimental campaign 2000, ENEA
Internal Report FUS-TN SA/SC/R 10-00 (December 2000)
[3.50] A. Natalizio, L. Di Pace and T. Pinna, Occupational dose and environmental releases
development of requirements for the power plant conceptualstudy, Task TRP1 Deliverables D3
and D4, ENEA Internal Report FUS TN SIC TR 5/00 (April 2000)
[3.51] M. Zucchetti and L. Di Pace, Clearance of activated materials: the De minimis problem
WMS/TSW1D5/ENEA/1 (Rev.2) (November 2000)
[3.52] M. Zucchetti and L. Di Pace, Clearance of activated materials: optimisation of ex-vessel
materials composition, WMS/TSW1D5/ENEA/2 (Rev.1) (November 2000)
[3.53] M. Zucchetti, L. Di Pace and P. Rocco, A review of SEAFP waste management studies,
private communication
[3.54] A. Natalizio and L. Di Pace, Waste management aspects of fusion power plants,
TSW2D5/ENEA/2 (Rev. 1) (December 2000)
[3.55] M.F. Maday, J. Nucl. Mater. 283-287, 689 (2000)
[3.56] R. Coppola et al., J. Nucl. Mater. 283-287, 183 (2000)
[3. 57] G. Dell'Orco et al., Optimization of the filling for the improvement of the performance of
reference ITER/DEMO ceramic and beryllium pebble, presented at the 14th topical Meeting on
the Technology of Fusion Energy (Park City 2000)
[3.58] G. Dell’Orco et al., Technical specification for the supply of two experimental mock-ups
called HELICA (HE-FUS3 Lithium Cassette) and HEXCALIBER (HE-FUS3 Experimental Cassette
of Lithium Beryllium Pebble Beds, ENEA Internal Report BB-F-S-002, Rev. 0 (December 1999)
[3.59] G. Dell’Orco et al., TTBB003.2 fabrication and testing of test modules in He-Fus3,
presented at the 1st HCPB Task Co-ordinator Group Meeting (Karlsruhe 2000)
[3.60] G. Dell'Orco et al., Optimization of the filling for the improvement of the performance of
reference ITER/DEMO ceramic and beryllium pebble beds, presented at the 14th Topical
Meeting on the Technology for Fusion Energy (Park City Utah, 2000)
[3.61] G. Dell’Orco et al., Specifica tecnica per la fornitura di un serbatoio di sicurezza per i
mock-ups sperimentali dell’impianto He-Fus3, ENEA Internal Report BB-F-S-003 Rev.0,
(October 2000)
[3.62] D. Zito, Test specification for the component integration of the dummy armour prototype
of the divertor cassette model IDICE, SIET Internal Report 1(May 2000)
[3.63] G. Dell’Orco et al., Tests on the integration of the ITER divertor dummy armour
prototype on a simplified model of cassette body, presented at the 21st Symp. on Fusion
Technology (Madrid 2000)
[3.64] G. Dell’Orco et al., Specifica tecnica per la fornitura di un quadro elettrico in BT per
l’alimentazione controllata di resistori per mock-up del pannello di prima parete per ITER,
ENEA Internal Report SB-E-S-002, (June 2000)
143

References
[3.65] G. Dell’Orco et al., Specifica tecnica per la fornitura di resistori elettrici in CFC per le
sezioni di prova del pannello di prima parete per ITER, ENEA Internal Report SB-E-S-003 (July
2000)
144

4. Inertial Confinement
4.1 INTRODUCTION 147
4.2 TARGET CHAMBER & DIAGNOSTIC UPGRADING 147
4.3 PREPARATION OF THE NEW EXPERIMENTAL CAMPAIGN 147
4.4 THEORY 147
4.5 DPSSL DESIGN ACTIVITY 151
REFERENCES 151

4. Inertial Confinement
4.1 INTRODUCTION
During the reference period the most relevant activities were those related to (i) the target
chamber and diagnostics upgrading, (ii) the preparation of the new measurement campaign, (iii)
the theoretical activity, and (iv) the design of the diode pumped amplifier for the ABCD laser.
4.2 TARGET CHAMBER & DIAGNOSTIC UPGRADING
During the year 2000 the ABC installation started to be implemented by a new automatic system
for target replacing and alignment. The assembly was designed and acquired in cooperation with
National Industry. The mechanical component and the electronics have been installed and the
ICF Physics & Technology Laboratory started the final fixing and testing of the system.
With regard to the diagnostic system, an additional streak camera has been associated to a
spectrograph, for time resolving spectra of laser harmonics produced in the interaction. The
diagnostic system is also being implemented by an additional x-ray framing camera assembled
at one of the vacuum chamber ports. The role the two framing cameras will be to produce timeresolved
imaging and space-time resolved spectra in the soft x-ray region.
4.3 PREPARATION OF THE NEW EXPERIMENTAL CAMPAIGN
We started the preparation of the targets for a new experimental campaign. This includes new
measurements on basic laser light interaction with low density, structured targets. The interest for
this study is twofold. First of all this materials provide a unique way to study relaxation processes
in plasmas, since in the interaction situations far from equilibrium are created. Previous
experiments performed on ABC in Frascati and elsewhere seem to indicate that the light
absorption occurs according to non-classical laws. Also, based on these laws, new kind of
modest-to-high-yield targets have been proposed. This activity is performed also in cooperation
with the Lebedev Institute of the Russian Academy of Sciences.
4.4 THEORY
The study on the injected entropy method (IE) [4.1] to design high-yield target was continued
through the exploration of a new approach for energy injection. Laser-generated light ion beams
(LIB) were considered for the ignition of imploding cylinders in the IE mode [4.2]. Pulses of
deuterium ions at 6 MeV were considered. For aperture F/2, ∆=R 1 and duration 40 ps, ignition
and high burn were found for 20 kJ total energy. The short pulse duration adopted in this case
strictly was not necessary for good coupling and derived from consideration on the LIB source.
In figure 4.1 the evolution of a target ignited by LIB in the IE mode is shown.
Thin foils exploded by short laser pulses are natural candidates as sources of fast ions. This
possibility was recognized early at the end of the sixties [4.3] when the first interaction
experiments with ultra-short laser pulses were started. The general idea is as follows: if a finite
fraction (η abs ) of the laser pulse energy is coupled to the electrons, an energetic ion flow follows
in the subsequent expansion, as the electronic pressure acts on the ions by electrostatic coupling
(quasi-neutrality).
The time available for energy transfer from the laser to the electrons is limited by the plasma
expansion, since at densities substantially lower than the critical (ρ c ) the system becomes
transparent to the laser radiation. A conveniently dimensioned foil is that represented in fig. 4.2,
where the initial and the exploded configuration are sketched. In this case the initial thickness is
chosen such that
147

4. Inertial Confinement
Light ion
beam source
30 ps 150 ps 370 ps 530 ps
729 ps 800 ps 980 ps 1002 ps
1010 ps 1016 ps 1020 ps 1038 ps
1049 ps 1054 ps 1061 ps
6 7 8 9
Log[Ti (°K)]
1 mm
Fig. 4.1 - Ignition of an imploding DT cylinder in the injected entropy mode. A 20 kJ, 40 ps, 6 MeV
deuterium ion pulse is injected at one of the open ends of the imploding fuel, to set a portion of material
on a high adiabat. The ion beam is diverging (F/2, see drawing). The ionic temperature maps show the
Fig. 4.01
ignition spark formation due to compression. The simulation code COBRA was used
2 R 0
ρ ρ c
2 Z 0
Fig. 4. 2 – The exploding foil geometry
148

4. Inertial Confinement
Zo
= ρ c Ro
ρo
(1)
In the previous equation ρ o is the solid state density (for solid deuterium and λ=1.054 µm,
ρ o /ρ c ≈50). When the plasma thickness becomes of the order of the diameter, the density is about
ρ c . At this time 3D expansion becomes effective and makes the system rapidly transparent to the
laser radiation. This occurs at the time
tint
R
≈ 2 o
Vi
(2)
Here V i is the final ion velocity (for deuterium ions at 6 MeV, V i ≈2.4×10 9 cm/s). The time t int
represents the order of magnitude allowed for the laser pulse duration. To be noted that this time
is substantially longer than the explosion time t expl since it is found that
tint
ρo
2/ 3 R
t o 2/
3
≈ ( ) expl = ( ) texpl
ρc
Zo
(3)
The foil dimensioning can be given in terms of the total energy in the ion flow, E tot . It is found:
E
R tot 13
o = ( ) / 23
∝ / ρ
; Z c
o = Ro
∝ − 43
λ
λ
/
2
πρcVi
ρo
(4)
tint
E
≈2 1 tot 13 23
( ) / ∝λ
/
π 5 ρ cVi
(5)
φabs ≈
1 3
ρcVi
∝λ
− 2
2
(6)
In equation 6 φ abs represents the power density to be absorbed on the target surface. The
impinging flux would be φ=φ abs /η abs . It is to be noted that φ abs do not depend on E tot .
Furthermore the only quantity depending on the initial solid density is Z o . For λ=1.054 µm,
ρ o =0.169 g/cm3 and E tot =50 kJ we get from eq. 4-6, R o ≈200 µm, 2Z o ≈10 µm,
φ abs λ µ 2 ≈3×1018 W×µm2/cm2, t int =20 ps. If the laser wavelength is decreased the target aspect
ratio q=R o /Z o decreases, the power density increases, the pulse duration decreases.
Let us assume η abs =0.5. From the previously considered dimensioning follows φ×λµ2≈6×1018
W×µm2/cm2 and a quiver electronic kinetic energy K e ≈0.7 MeV. Since the number of electrons
is equal to that of ions (Z=1), electrons must achieve an energy of the order of that required for
the ions, that is 6 MeV. The situation can be described by saying that each electron must receive
about 8.6 “kicks” with energy equal to the quiver one. The average round trip for an electron in
the plasma is estimated as 2R o /c (c is the speed of light), so that, on the average, each electron
can see the laser field a number of times t int c/2R o =c/V i ≈13.
In the previously dimensioned system, the electronic motion is practically collisionless, and ions
149

4. Inertial Confinement
can take energy by electrostatic coupling during the system expansion. Consistently with the
previous scheme global, forward collimation effects due to transfer of electromagnetic
momentum are expected to be negligible. Actually a simple estimate shows that
∆Vi
Vi
1 2 V
≈ ( −1)
i
2 ηabs
c
.
(7)
In the previous equation ∆V i is the additional velocity due to radiation pressure. For V i ≈2.4×109
cm/s and η abs =0.5, ∆V i /V i ≈0.12. However, for η abs .=0.08, it is found ∆V i /V i ≈1. In practice the
effect is substantial when the energy transfer is inefficient. More effective can be the mechanism
of collimation by pressure gradient. Since the gradient of pressure is greater in the direction
normal to the thin foil surface, a bilateral collimation effect can result for thin foils. The
asymptotic value of the F/Number can be numerically found by integrating the fluid equations
for a 2D, gaussian density distribution representing initially cold foils with cylindrical symmetry.
In our calculations an inner power source representing the energy given by the laser to the
electrons started the hydrodynamic motion. The duration of the energy source in units of the
explosion time t expl was taken in the ratio (R o /Z o )2/3, according to the prescription given by
Eq. 3. The results of these calculations are represented in fig. 4.3, where the F/Number is given
as function of the foil aspect ratio. It is seen that for the aspect ratios of interest (20÷50)
F/Number in excess of 2 are possible.
In the previous considerations it was tried a framing of the global behavior of the exploding foil.
The individuated regime is quite different from those up to now studied in experiments. The
distinctive feature is the long duration of the pulse (≈15÷20 ps) associated to high intensity
(possibly 1÷5×1018 W/cm2). For these power densities, theory and experiments normally
consider 0.5÷1 ps duration [4.4, 4.6]. Moreover, to reach in the experiments high power density,
the pulse energy was deposited in spots around 10 µm wide, a few light wavelengths. In the case
here considered systems with diameters several 100 µm wide are assumed and the target
substantially expands during the interaction. Key information for the application of thin foil
explosion to usable production of LIB is related to the value of η abs and to the way the absorbed
energy is shared between the foil particles. For instance, a small value of η abs would affect the
efficiency of the method. The self-organization of a fraction of the plasma in high-energy jets
2.5
2.0
F
1.5
1.0
0.5
0 20 40 60 80 100
R 0 /Z 0
Fig. 4.3 – The asymptotic F/Number for exploding gaussian thin foils as function of the foil aspect ratio
150

4. Inertial Confinement
during the interaction could have the same effect or, at least, would imply changes in the model
to be used for foil dimensioning.
4.5 DPSSL DESIGN ACTIVITY
The activity on diode pumped solid state laser (DPSSL) was continued for the design of one of
the sub-amplifiers of the laser ABCD. Most of the recent activity was devoted to the optimization
of the energy transfer from the diode assembly to the active material.
The Frascati design was discussed in a two-day ad hoc meeting at LLNL, last July.
151

References
[4.1] A. Caruso, C. Strangio, Laser Part. Beams, 18, 35 (2000)
[4.2] A. Caruso, C. Strangio, “Studies on nonconventional high-gain target design for ICF” to
appear on Laser Part. Beams
[4.3] A. Caruso, R. Gratton, Phys. Letters, 36A (4), 275 (1971)
[4.4] F.N. Beg et al., Phys. Plasmas, 4 (2), 447 (1997)
[4.5] M.H. Key et al., IFSA 99 Proceedings and Preprint UCRL-JC-135477 (1999)
152

5. Miscellaneous
5.1 ADVANCED SUPER-CONDUCTING MATERIALS AND DEVICES 155
5.1.1 Influence of YBCO film thickness on critical current 155
5.1.2 MgO-based buffer-layer on Ni-V substrates 156
5.1.3 Buffer-layer deposition on inclined substrates 158
5.1.4 Development of high-field Nb 3 Al multifilamentary strands 159
5.2 CRYOGENIC TESTING 159
5.2.1 Cryogenic testing of diode stacks for CERN 159
5.3. NEW HYDROGEN ENERGY 160
5.3.1 New electrolytic cell experimental campaign 160
5.4 CRYOGENICS 161
5.4.1 Liquid helium service 161
5.4.2 Cryogenic technologies 161
5.5 ACCELERATOR-DRIVEN SUBCRITICAL (ADS) 162
5.5.1 Long – term compatibility tests on the AISI 316L steel and the MANET II steel
in stagnant PbBi 162
5.5.2 Study of the PbO/H 2 reaction for the control of oxygen in Pb-Bi with the ORE device 163
5.5.3 LECOR and CHEOPE III loops 164
5.5.4 Development of an oxygen sensor and validation of the oxygen probe delivered
by IPPE 165
5.6 PLASMA FOCUS 165
5.6.1 Development of a mobile and repetitive plasma focus (PF) 165
5.6.2 The apparatus 166
5.6.3 Measurements of the neutron yield 167
5.6.4 Neutron yield from single shots 167
5.6.5 Neutron yield from a sequence of shots 168
5.6.6 Neutron yields from individual shots within a sequence 168
REFERENCES 169

5. Miscellaneous
5.1 ADVANCED SUPERCONDUCTING MATERIALS AND DEVICES
5.1.1 Influence of YBCO film thickness on critical current
The research activity is mainly focused on the development of YBa 2 Cu 3 O 7-x (YBCO)-based
coated conductors, viz. the fabrication of bi-axially textured YBCO thick films on flexible metal
tapes, useful for large-scale power applications. The use of appropriate materials as an
intermediate buffer layer between metallic substrates and YBCO films is needed in order to
obtain high critical current density (J C ) coated conductors.
Up to now, samples using non-magnetic cube textured Ni 89 V 11 (Ni-V) alloy as a metallic
substrate (Italian Patent AN: RM98A000395) have been developed: the best results were
obtained with a bi-axially textured NiO/CeO 2 buffer layer architecture, reaching J C as high as
0.6 MA/cm2 at 77 K and zero magnetic field for 0.7 µm of YBCO thickness [5.1]. In order to
improve the engineering current density of coated conductors (ratio of critical current (I C ) to
total tape cross-section), a study regarding the relationship between the thickness of the
superconducting film and its critical current was performed on bi-axially aligned YBCO films,
grown on epitaxial CeO 2 /NiO structure on Ni-V.
The research activities have been carried out in collaboration with Pirelli Cavi & Sistemi S.p.A.,
Unità INFM of Salerno, Technical University of Cluj (Romania) and in the framework of CNR
5% Funding Italian National Programme.
The YBCO/CeO 2 /NiO/Ni-V structures were obtained in-situ, with no vacuum breakdown.
Details on deposition conditions and sample preparation procedures can be found elsewhere
[5.1]. During this study, the YBCO growth rate was kept constant at about 11 nm/min; samples
with a different YBCO thickness, ranging from 0.2 to 2 µm, were produced by varying the
deposition time from 30 to 180 minutes.
Figure 5.1 shows the zero-field critical current density (J C ) at 77 K as a function of YBCO film
thickness. The J C curve shows a monotonic decreasing trend with an increasing film thickness.
Up to 0.7 µm, the J C ranges between 6×105 and 4×105 A/cm2, while a sharper decrease is
observed for thicker films. This feature is more emphasized at the inset of fig. 5.1, where the J C
dependence is reported on a logarithmic scale.
For the same series of samples, the product of
J C and YBCO thickness, corresponding to a
1 cm wide tape I C , is plotted in fig. 5.2. As it
can be seen, I C drops in correspondence of the
J C reduction for YBCO films thicker than about
0.7 µm. Hence, a qualitative indication can be
deduced: the deposition of thicker YBCO films
has, as a consequence, the degradation of the
entire film in terms of superconducting
transport properties [5.2].
In order to investigate the decrease of YBCO
current-carrying capability with increasing film
thickness, structural and morphological
properties of the samples are being analysed, as
well as the effect of the YBCO deposition time.
XRD measurements reveal that thinner YBCO
films, grown on a CeO 2 /NiO/Ni-V architecture,
have a sharp bi-axial texture. The YBCO film is
c-axis oriented, and no other reflections can be
detected in the θ-2θ diffraction pattern. The out-
Fig. 5.1 - J C as a function of YBCO film thickness. YBCO
films were grown on CeO 2 /NiO/Ni-V architectures. At the
inset, the J C dependence is reported on a logarithmic scale.
Empty symbols refer to YBCO films grown on SrTiO 3 singlecrystal
substrates
155

5. Miscellaneous
Fig. 5.2 - DC-transport I C measured for a 1 cm wide tape as
a function of YBCO film thickness for the same series of
samples as Fig. 5.1 Values are obtained as products of the
critical current density and corresponding YBCO film
thickness
of-plane and in-plane crystalline distributions,
evaluated by (005)YBCO ω- and (113)YBCO
φ-scans, reveal Full Width at Half Maximum
(FWHM) value of 8° and 9°, respectively.
Some structural changes can be observed with
increasing YBCO film thickness. The fraction
of a-axis oriented grains progressively
increases with YBCO films thicker than about
0.7 µm. Pole figure analyses reveal a pole
broadening with thickness indicating a
sharpness reduction of YBCO film texture.
The FWHM values of (005)YBCO ω- and
(113)YBCO φ-scans of about 11° and 14°,
respectively, is reached for 2 µm thick films.
YBCO morphological evolution with
increasing film thickness was analysed by
Scanning Electron Microscopy (SEM)
investigations. Dense coating and smooth
surface, with good grain coalescence and
uniform dimensions, were observed for
YBCO films up to about 1 µm film thickness.
For thicker films, the YBCO surface appears
rough, with poor coalescence among adjacent
grains. A non-homogeneous grain size can
also be observed. The cross-sectional analyses
reveal the compact and dense nature of the
YBCO films for thickness up to 1 µm while,
for thicker samples, a granular region close to
the film surface is observed, as it can be seen
in fig. 5.3, where a 1.4 µm thick film crosssection
is reported.
In order to check the influence of deposition
time on YBCO tape properties, a simulation
of YBCO deposition for 90 minutes (time
required for a 1 µm thick film deposition) was
performed on a 0.2 µm thick YBCO film.
Such a heat treatment in highly oxidising
conditions leads to a drastic reduction
Fig. 5.3 - Cross section SEM picture of a fractured of transport properties. In fact, after the heat
YBCO/CeO 2 /NiO/Ni-V structure. The region near the YBCO treatment, the resistive transition does not
film surface presents a less compact feature as compared to show any remarkable change. Nevertheless,
the inner one (magnification 25kX)
the annealed film shows a residual resistance
below the critical temperature (T R=0 ≈88 K)
of the as-deposited film. This leads to
an ohmic behaviour of the voltage-current characteristic at 77 K.
5.1.2 MgO-based buffer-layer on Ni-V substrates
Development of alternative buffer layer architectures, suitable for YBCO epitaxial growth using
materials such as MgO and Pd on Ni-V substrates, was performed.
MgO films were deposited in vacuum on Ni-V by e-beam evaporation [5.3]. Pole figure analyses
156

5. Miscellaneous
for as-deposited MgO films, reveal a main (001)[110] texture and other features compatible with
a (221) orientation, fig. 5.4. The presence of (221) orientation depends on film thickness and
deposition temperature. Films deposited at 600°C and 150-180 nm thick only show the main
texture, with (202) ϕ and (002) ω-scan FWHM of about 12° and 10°, respectively. The surface,
shown in fig. 5.5, appears to be compact and free of cracks.
The structures were deposited in-situ by e-beam evaporation. Pd films were grown in a
temperature range from room temperature to 350°C, while MgO films at 580°C [5.3].
Fig. 5.4 - (002) pole figures for MgO films on Ni-V substrates: a), b) and c) 150 nm thick films deposited
at 525, 600 and 700°C, respectively; d), e) and f) films deposited at 600°C and 550, 230 and 180 nm
thick, respectively. Poles at χ =55° and ϕ = 45+nπ/2 were an artefact due to (111) Ni-V reflections
Fig. 5.5 - Surface SEM micrograph (magnification
10 kX) of a 150 nm thick MgO film on Ni-V substrate,
deposited at 600°C
Fig. 5.6 - a) (002) and b) (202) pole figures for 150 nm thick
MgO film on Pd/Ni-V structure. In (002) pole figure, poles at
χ = 55° and ϕ = 45+nπ/2 are due to (111) Ni-V reflections
157

5. Miscellaneous
X-ray diffraction analyses reveal that, when using the intermediate Pd buffer-layer, MgO films
develop a sharper (001)[100] texture, with an out-of-plane FWHM of about 5° and in-plane
FWHM of about 8° (fig. 5.6). In order to evaluate the effect of the interdiffusion of Pd and
metallic substrate, the MgO/Pd/Ni-V structure was subjected to a thermal cycle at temperatures
typically required for YBCO deposition. It was observed that MgO films, annealed at 850°C for
30 minutes, preserve their structural and morphological properties.
5.1.3 Buffer-layer deposition on inclined substrates
A different approach to the fabrication of YBCO-coated conductors has also been investigated:
texture of a MgO layer is induced by depositing the film onto an inclined randomly-oriented
metallic substrate. Bi-axially oriented buffer-layer fabrication with an inclined substrate
deposition (ISD) is a very promising technique for industrial scalability of coated conductors
production. ISD technique is a potential low-cost approach, because of the use of randomlyoriented
metallic substrates and the simplicity of the procedures
required.
Fig. 5.7 - (002) pole figure of MgO
obtained by ISD technique. The vertical
arrow indicates the vapor beam direction
MgO films were obtained by e-beam evaporation in vacuum at
room temperature. The substrates were inclined as compared to
the direction of the incident vapour, thus forming an angle α
with the substrate normal direction. The film texture mainly
depends on both film thickness and α value. The optimal
inclination angle α of the substrate was found to be 55° and the
deposition rate, typical of the process, was about 5 nm/s [5.4].
Pole figure (fig. 5.7) reveals that the [001] axis is tilted by an
angle of about 30° from the substrate normal towards the
deposition direction. Thicker MgO layers are better aligned,
reaching an FWHM of less than 10° for thickness of about 3 µm,
as reported in fig. 5.8. The MgO has a columnar morphology
with a very fine grain size in the direction normal to the column
axis, according to the SEM micrograph of fig. 5.9.
These results show that ISD MgO films are suitable as a bufferlayer
for YBCO deposition, even though the effects of the MgO
Fig. 5.8 - FWHM of (002) pole vs MgO film thickness
Fig. 5.9 - Cross section SEM picture of a fractured MgO
film deposited by ISD (magnification 25 kX
158

5. Miscellaneous
surface roughness, - which is due to the columnar morphology - on the superconducting film
growth will have to be taken into account.
5.1.4 Development of high-field Nb 3 Al multifilamentary strands
A collaboration contract has been signed by ENEA and the Italian Company EDISON
Termoelettrica S.p.A for a two year-programme aimed at demonstrating the feasibility of an
industrially scalable process. The new apparatus for the manufacturing of multi-filamentary
Nb 3 Al strands in a Nb matrix with the rapid quenching approach has been recently terminated.
A first billet of single-strand wire has been processed, but enbrittlement of the wire, just after the
thermal quench, has led to wire rupture. Several combinations have been explored for the
experiment conditions, but the problem was not overcome. Therefore, a new wire billet, with a
different relative percentage of Nb and Al, will have to be tested.
Short samples of mono-filamentary “rapid quenching” Nb 3 Sn, produced by CNR Lecco, have
been characterized in our labs, where a critical temperature of 17.8 K has been measured.
5.2 CRYOGENIC TESTING
5.2.1 Cryogenic testing of diode stacks for CERN
ENEA developed a collaboration with OCEM, an Italian firm operating in the field of electronic
devices and power supplies, for manufacturing and testing of diodes for the quench protection of
the dipole and quadrupole magnets of Large Hadron Collider (LHC), the new particle accelerator
under construction at CERN, which should come into operation in the year 2005. It will mainly
accelerate and collide 7 TeV proton beams but also heavier ions up to Pb. The accelerator design
is based on superconducting tin-aperture magnets, cooled at 1.9 K in a superfluid helium bath.
In the case of a quench, the magnets will be protected by turning on by-pass diodes. The
operating current, namely 11.8 kA, will then exponentially decay in the diode and the magnetic
energy removed will cause a rise up to 300K in the diode-stack temperature.
OCEM will produce the components (copper heat sink and clamping systems) and assembly the
stacks, while ENEA will be testing all of them (1250 for the dipoles and 400 for the quadrupoles)
under CERN supervision by December 2004.
The cryogenic tests requested by CERN consist
of pre-test measurements of the reverse and
forward voltage diode characteristics at 77 K
and at 300 K, and in endurance test cycles at
liquid helium temperature.
To this purpose, ENEA has designed the
experimental set-up, developed the data
acquisition system, and prepared a dedicated
facility in Frascati.
A diode stack, connected to the current leads
for low-temperature tests, is shown in fig. 5.10.
At the end of 2000, after the successful official
commissioning of the new LAB and the
approval by CERN, the diode stacks testing
will be able to start.
Fig. 5.10 - The NbAl rapid quench facility
159

5. Miscellaneous
5.3. NEW HYDROGEN ENERGY
5.3.1 New electrolytic cell experimental campaign
In the final months of 1999, a new ENEA Project started, called “New Hydrogen Energy”. It
has the goal of building an electrolytic cell prototype, able to clearly prove the existence of the
phenomenon known as “cold fusion”.
The project scope is the investigation of the physical nature of the heat excess which has been
measured in the device under construction. The first ten months of experimentation were focused
on the characterisation of a basic-device, viz. a very simple electrolytic cell, easy to assemble
and use (see fig. 5.11), which will be the basic element of a more complicated device, able to
produce larger quantities of heat. The research activities focused on the properties of suitablydesigned
thin Palladium films to adsorbs hydrogen and its isotopes at very high density. It had
been assessed in past years that the cold fusion phenomenon, viz. the capability of deuterated
palladium of producing energy, only starts when a threshold concentration of deuterium inside
palladium lattice is reached. The use of thin film (from thousand Angstrom to a few microns)
seemed to be a promising technique to obtain very high loading and bypass the very long loading
time due to the small diffusion of deuterons in palladium lattice. However, it was recently found
in experiment and then proposed that a voltage drop along the palladium sample can affect the
chemical potential of deuterons inside the metal, thus decreasing it dramatically. The ultimate
effect is to accommodate very high loading. The first attempts showed a very promising increase
in the average loading obtainable and a quite comfortable reproducibility of the results. The new
electrolytic cell was then designed around a thin film cathode with an appropriate geometry.
Fig. 5.11 - The electrolytic cell
Table 5.I - Heat excess produced by the cathode of the new
electrolitric cell
Total number Heat excess Malfunctioning
measured in (broken cathode)
Static calorimetry 13 7 6
Flux calorimetry 8 0
Calibrations 29
The results of the study on the heat
excess produced by these cathodes
are summarised in table 5.I.
Because of the very small amount
of material involved (V cathode =
25×10-6 cm3), the heat excess
produced is in the order of tens of
milliwatt, corresponding to a
density power of some kW per
cubic centimeter. The power
produced (in terms of increase in
the cell temperature) ranges from
30 to 200% the electrical input
power.
The energy produced can be
increased by increasing the number
of elementary cells, so that a more
complex device operating with six
cells in parallel has been designed
and is presently under construction.
During 2000, the experimental
activities have moved into a new
and better equipped laboratory.
In May 2000, The VIII
International Conference on Cold
Fusion was held in Lerici (La
Spezia), with the sponsorship of
160

5. Miscellaneous
ENEA. The Frascati ENEA group very actively participated in the logistic and scientific
organization, and contributed with five communications.
A first experimental campaign was carried out with the high resolution Quadrupole Mass
Spectrometer (QMS), in order to detect any possible production of 4He during excess heat
phenomena. The experimental set-up, in its first arrangement, includes an activated charcoal
trap, cooled down to liquid nitrogen temperature, and a Non-Evaporable Getter (NEG) pump,
operating at room temperature in order to selectively remove all components from the gas
mixture, with the exception of noble gases. More than sixty samples were analysed, from both
blank and “black” cells. The preliminary results indicate, on a statistical base, an augmentation
of the 4He content in black experiments (three to ten times that observed for blank cells). In at
least four cases, there is clear evidence of a correlation between helium and excess heat
production.
In order to improve our confidence in the experimental results, some changes in the arrangement
are underway. A Pd alloy catalyst will be introduced, to remove most of oxygen and deuterium,
while the cryo-sorption pump will be replaced by a Capacitor NEG pump, operating at high
temperatures (~300°C), which will strongly reduce the content of all active components, before
sending the gas sample to the room temperature NEG pump. This would lead to the suppression
of the spread observed in the background content of helium, introduced by the activated charcoal
cryo-pump. A new vacuum system, based on two turbomolecular drag pumping stations and a
membrane backing pump, has been designed; furthermore, in order to improve the clearness of
the ultra-high vacuum (UHV) system, ultra-high purity nitrogen (N70) will be introduced as a
purging gas.
5.4 CRYOGENICS
5.4.1 Liquid helium service
The new helium liquefier, installed at the and of April 1999, has been regularly operating during
the year 2000, with an overall liquid helium production of about 28,000 litres. Furthermore,
23,000 litres of liquid helium have been delivered to the users, and about 9,000 litres have been
acquired to reintegrate the helium inventory.
This facility consists of a TCF20 cold box (fig. 5.12), equipped with an internal autopurifier and
two turbine expanders for gas pre-cooling; a screw
driven DS220 recycling compressor equipped with
an oil removal system; a line drier; a pressure
control panel; two 1000-litre Dewars for liquid
helium storage; two transfer and decant lines; a 7
m3 pure helium buffer tank; and an analytical panel
equipped with a purity monitor and a moisture
meter. The plant, supplied by Linde Cryogenics
Ltd. (UK), is automatically controlled by an Allen
Bradley Programmable Logic Controller (PLC).
The liquid helium production rates, with or without
liquid nitrogen pre-cooling, are turning out to be
better than the nominal rates of 60 and 30
litres/hour.
5.4.2 Cryogenic technologies
Substantial progress has been achieved in the
development of a low-noise single shot 3He/ 4 He
Fig. 5.12 - The Linde TCF20 helium liquefier
161

5. Miscellaneous
dilution refrigerator, based on the use of cryosorption pumps.
This activity was carried out in the context of a two-year cooperative
agreement between ENEA and the Physics
Department of University of Rome 3, aimed at demonstrating
the feasibility of a dilution refrigerator capable of achieving
temperatures below 100 mK, with an operating cycle of at
least 10 hours. The use of cryosorption pumps allows for the
absence of vibrations and mechanical noise, which is a main
requirement for many space-and earth-based applications.
The prototype (fig. 5.13) has been completely designed and
assembled during 2000. It consists of three refrigerating
stages: the first can be cooled down to about 1.5 K, by
pumping on a liquid 4He bath, thus allowing to liquefy the
3He and 3 He/ 4 He mixture stored in the second and third
stage, respectively. Pumping on liquid 3He in the second stage
allows temperatures below 300 mK to be achieved, thus
ensuring a good phase separation of the gas mixture in the
mixing chamber. The final cooling down is achieved by
pumping on the 3He-rich phase.
Preliminary tests have been carried out to separately check the
performance of both the first and the second stage. Excellent
results have been achieved with the 3He refrigerator, featuring
Fig. 5.13 - The 3 He/ 4 He dilution refrigerator a minimum temperature of 285 mK, with a stability better
than ±5 mK for over 24 hours. The behaviour of the first stage
has turned out to be less satisfactory, since it was able to keep its minimum temperature of 1.5
K for about 3 hours only; further work is in progress to improve its performance.
5.5 ACCELERATOR-DRIVEN SUBCRITICAL (ADS)
5.5.1 Long – term compatibility tests on the AISI 316L steel and the MANET II
steel in stagnant PbBi
The compatibility of the AISI 316L and the MANET II steels in stagnant molten Pb-55.5Bi was
evaluated at constant temperatures of 573, 673 and 823 K over 1500 h, 3000 h and 5000 h
immersion time. The results obtained can be summarized as follows:
• At 573 K and up to 5000 h, both steels suffered damages due to corrosive attacks of the lead
alloy, and a thin oxide layer could be detected only after 5000 h exposure.
• At 673 K, after exposure for 5000 h, the AISI 316L steel still had a thin oxide scale (1 µm) and
the MANET II steel, tested in the same conditions, exhibited an oxide scale – with a thickness
of about 5 µm – formed by two sub-layers.
• At 823 K, the behaviour of the two types of steels completely changed: in fact, no more
corrosive products due to oxidation were detected, while severe corrosive liquid metal attacks
could be observed, as shown in the SEM – micrographs of figs. 5.14 and 5.15.
Taking into consideration the results reported in the refs. [5.5, 5.6, 5.7], as well as those obtained
from the long-term tests, the following could be assumed:
• Thin oxide growth, following the logarithmic kinetic [5.8], was observed on the AISI 316L
steel, exposed to oxygenated Pb-Bi from 573 K to 750 K and on the MANET II steel, in the
temperature range between 573 K and 623 K.
162

5. Miscellaneous
Fig. 5.14 - AISI 316L in Pb-Bi at 823 K for 3000 h
Fig. 5.15 - MANET II in Pb-Bi at 823 K for 5000 h
• At temperature up to 753 K, it was observed that the oxidation of martensitic steels follows a
parabolic law [5.9]. This observation is also in accordance with the oxidation kinetics, observed
in gas phase at higher temperatures [5.7].
• For the two types of steels between 750 K and 823 K, a transition temperature or temperature
range occurs where a change from oxidation to dissolution in the corrosion mechanism is
expected. Future work is aimed at evaluating this transition range.
5.5.2 Study of the PbO/H 2 reaction for the control of oxygen in Pb-Bi with the
ORE device
In its basic configuration, Occupational Radiation Exposure (ORE) consists of a cylindrical
crucible in AISI 316 stainless steel, containing a well-known amount of Pb-Bi eutectic alloy. The
crucible is externally heated by electric wires and thermally insulated. It is supplied from the
bottom with an Ar + H 2 gas mixture at a given flow-rate. A detailed description of the ORE
device is given in ref. [5.10]. After reacting with the oxygen present in the liquid metal, the gas
mixture, containing water as a reaction product, is analysed by a capacitance hygrometer type
CERMET, supplied by Michell Instruments Company. A detailed description of the
measurement system and the relation used for the evaluation of the oxygen extraction rate are
given in ref. [5.11].
Two types of tests have been carried out during the latest experimental campaign. In the first one,
no oxygen was added to the liquid metal in the form of lead oxide, and the presence of lead oxide
in the crucible was only due to oxygen concentration above its saturation limit.
The second kind of experiments was mainly focused on the determination of the PbO reduction
rate. In this case, 50 grams of PbO were initially added to the liquid metal, and the total amount
of oxygen in the crucible was about 830 wppm at the beginning of the tests.
For the evaluation of the oxygen extraction rate with no PbO addition, two runs have been
carried out on ORE apparatus, at temperatures of 623 and 673K.
At 623 K, the average oxygen extraction rate obtained was 1.4 wppm/h and at 673 K it was 1.7
wppm/h.
Following the Orlov’s relationship [5.10] for the oxygen solubility limit, which is 0.55 wppm at
623 K and 1.4 wppm at 673 K, from the results achieved in terms of oxygen extraction rate it is
evident that the liquid metal was in oxygen saturation conditions at the beginning of the run.
163

5. Miscellaneous
Table 5.II - Oxygen solubility results in the first kind of experiment
Temp. of the liquid metal Mean PbO reduction rate Mean PbO reduction rate per unit area Mean O extraction
(K) (wppm/h) (gm -2 h) (wppm/h)
523 14 9 1.0
573 45 28 3.2
623 104 94 7.5
673 320 288 23
Table 5.III - Oxygen solubility results in the second kind of
experiment
Temp. of the liquid metal
(K)
Time to reach stationary state
(s)
523 1020
573 720
623 200
673 60
In the second type of experiments, the
temperature range investigated was 523÷673
K. The results are shown in tables 5.II and
5.III and in fig. 5.16.
As it can be seen in table 5.I, the PbO
reduction rate varies more than one order of
magnitude, increasing from 523 to 673 K.
The correlation of the results with an Arrhenius-type equation is quite good; this is an indication
for the reliability of the experimental points.
The next experimental campaigns on ORE will be focused on the extraction of oxygen from Pb-
Bi with a small oxygen content, with no addition of lead oxide. The oxygen extraction rate will
be determined in a temperature range of 523÷673 K, passing from a gas sweeping to gas a
bubbling regime.
Moreover, an oxygen control method, based on the contact between the liquid metal and a
calibrated H 2 /H 2 O mixture, will be studied.
5.5.3 LECOR and CHEOPE III loops
Fig. 5.16 - Reaction rate vs temperature
The main objective of the experiments to be carried out in LECOR is the quantitative study of
corrosion phenomena and related mechanical effects affecting steels and structural materials of
new concepts for ADS application, in presence of flowing lead bismuth alloy, under different
operative conditions, such as temperature, velocity of the flowing liquid metal, and oxygen
concentration. More in detail, the main task of this loop is to test the behaviour of materials and
coatings at low oxygen activity, simulating the presence of a reducing environment in the
spallation target of an ADS system.
On the other hand, with the CHEOPE-3 loop, corrosion tests were performed by exposing
164

5. Miscellaneous
structural materials to Pb-Bi with a high oxidzsing potential. The description of the two loops is
reported in ref. [5.11]. Oxygen probes delivered from IPPE were installed on both loops for
oxygen monitoring.
The materials to be tested in LECOR are the following: AISI 316L and T91 steels; tungsten;
molybdenum; tantalum and tungsten coatings deposited on AISI 316L with the CVD technique.
During the firs run, the test section is at 723 K, the liquid metal velocity 0.8 m/s, and the oxygen
concentration in the range of 10-8÷10 -10 wt %.
As far as the thermal regime in CHEOPE-3 is concerned, the reference temperatures in
CHEOPE-3 is 623 K in the cold leg and 723 K in the test section. The velocity of the liquid metal
is in the range of 0.5÷1 m/s. An oxygen control system in the whole loop is envisaged, based on
the free surface conditioning of the liquid metal in the feed tank of the loop. The oxygen control
system is described in ref. [5.11]. The oxygen concentration in the loop is around 10-6 wt%. The
main goal of corrosion tests in CHEOPE-3 is to assess the protection efficiency of the oxides
growing on the steel, with a controlled oxygen content in presence of flowing lead bismuth alloy.
5.5.4 Development of an oxygen sensor and validation of the oxygen probe
delivered by IPPE
During the year 1999, a first prototype of oxygen probe with a solid electrolyte was developed,
as reported in ref. [5.11]. The solid oxide electrolyte used was the YSZ ceramic, which has the
ability of exchanging oxygen between the two sides of a galvanic cell in the form of a negative
ion (O2-). As a reference electrode, the red-ox pair Sn/SnO 2 was used.
The new oxygen probe prototype has YSZ as a solid electrolyte, while the red-ox pair was
changed from Sn/SnO 2 into In/In 2 O 3 , and the solid electrolyte’s housing was changed from
stainless steel to ceramic.
The experimental device to test the sensor is composed by a stainless steel cylindrical crucible,
which contains the Pb-55Bi alloy. A thermocouple connection is put in the crucible for
temperature monitoring within the liquid alloy. The container heating is performed through a
heating coil. To measure the electromotive force, the Pt wire of the sensor is connected to a highimpedance
voltmeter.
The testing procedure foreseen for analysing the response of the sensor consists in adding a
quantity of PbO to Pb-Bi, in order to obtain about 1×10-3 wt. % of oxygen. The cell voltage is
registered for a temperature of 673 K in the test section and, if necessary, the temperature
remains at this value until the cell signal measured with the voltmeter reaches a steady value. The
temperature is then increased stepwise by 50 K up to 773 K. At each temperature (673, 723 and
773 K) the voltage is registered. The temperature dependency on the voltage measured will be
compared with the theoretical evaluations performed in oxygen-saturated conditions.
In the frame of the collaboration between ENEA and IPPE, studies on the validation of an
oxygen meter developed by IPPE are under progress. IPPE has supplied two oxygen meters to
ENEA, which have been installed in the LECOR and CHEOPE III loops.
The solid electrolyte is a ceramic of ZrO 2 , stabilised with Y 2 O 3 . A reference electrode made of
Bi/Bi 2 O 3 is located in contact with the ceramic electrolyte. The working temperature range of
the meter is between 553 K and 873 K. The oxygen meters were calibrated at IPPE and the
relationship between the electromotive force signal and the oxygen activity is the following:
E=0.088-0.178×10-4×T-9.917×10 -5 ×T×log(a)
where the electromotive force E is measured in [V] and the temperature T in [K].
165

5. Miscellaneous
5.6 PLASMA FOCUS
5.6.1 Development of a mobile and repetitive Plasma Focus (PF)
ENEA, University of Ferrara and University of Bologna are jointly working on a PF project, with
the aim to exploit the characteristic emission released by these devices (neutrons, x-rays, gamma
rays, ion and electron beams) for marketable applications.
PF’s are essentially pulsed generators producing bursts of neutrons lasting for about 100 ns, with
energies of 2.5, 10-11 or 14 MeV depending on which gas is fuelling the machine
(deuterium-deuterium (DD) with a lithium target or a deuterium-tritium (DT) mixture).
In a PF working in DD, the neutron emission is typically of about 5×108 neutrons/shot for 10 kJ
capacitors bank energy and 1010 neutrons/shot for 50 kJ. In deuterium-lithium, the emission is
about the same, whilst in D neutron emission is about one hundred times higher, due to the larger
fusion cross section.
From the point of view of the applications, PF must be compared to the other neutron sources
present on the market, viz. radioactive sources and accelerators, and show to be competitive.
This analysis indicates that PF machines emitting between 108 and 109 neutrons per second,
fuelled with deuterium gas and easily transportable, can be of great interest in applications where
intrinsic safety and a rather high neutron emission are relevant. The aforementioned neutron
production is at least one order of magnitude higher than a commercially available DD
accelerator source, and is obtainable with a bank of capacitors of about 10 kJ which in size and
weight, is well compatible with a mobile machine. A passive radioactive source of a similar
emission (for instance, about 10 GBq of Cf or several hundreds of GBq for Po/B or Am/Be)
never stops emitting and raises serious handling and storing problems due to the need for it to be
shielded. Furthermore the PF machine does not raises any activation problems as to storage and
handling.
A PF device, aimed to work in environmental conditions not tailored for nuclear equipment has
been developed by ENEA. It can operate in deuterium at 1 Hz shot-rate, with an average neutron
production of 3×108 n/shot at 6 kJ of capacitors energy. It is easy to transport and conceived for
those industrial purposes where a rather intense neutron generator is required, but the use of
tritium is forbidden. The machine is designed to be reliable and uniform in emission, and to be
operated in the field by not specialised personnel, to meet the requirements of the possible
transformation into a commercial product.
5.6.2 The apparatus
The machine (called PF4 ) has been designed and built at the ENEA Center of Brasimone. Its
aim is to demonstrate the feasibility of a low cost Plasma Focus device, easy to be run and
maintained.
The system that is being built, though fully operative, only has demonstration purposes and,
therefore, is not technologically perfect. Although our device can already be transported in an
ISO 10 container, we know that it is possible to obtain a considerable reduction of its weight and
volume. In fact, these two parameters have not been minimised at all in our test machine, and we
believe that it will be rather easy to do so for possible future industrial applications.
A block picture of the machine is reported in fig. 5.17. Here, the characteristics of the various
subsystems are illustrated and the core information on the performances of the machine is
supplied (table 5.IV).
Power supply. The power supply is an Italian commercial product, made by OCEM, designed for
166

5. Miscellaneous
Table 5. IV - PF4 characteristics
Energy
Voltage
Capacity
Repetition rate
6 kJ
21 kV
3×10 8 n/shot
1 Hz
repetitive use. It can provide a constant capacitor
bank charging current of 1 A up to a voltage of 35 kV.
Capacitor bank. The bank is made of five 6 µF, 60 kV
low-inductance capacitors. The capacitor bank is
connected to the spark gap by two flexible copper
plates 2 mm thick, 25 cm wide and about 1 m long,
insulated by mylar foils.
Spark gap. The spark gap, made by Maxwell (Model
40200 Rail-gap switch driven by a trigger generator
Model 40151-B), is commercially available. It has
been modified to add an automatic system to replace
the gas and is pressurized with Ar-O 2 .
Plasma Focus head. The two electrodes have radii
respectively of r 1 =40 mm and r 2 =90 mm. The length
of the inner electrode is 75 mm, plus 60 mm
Fig. 5.17 - Block picture of the PF4 device
insulator. The metal employed for the anode is
tungsten, to minimize sputtering on the insulator. The
filling gas used is deuterium at a pressure of 500 Pa.
No renewal of the filling gas was accomplished during the repetitive operation of the machine.
The anode is cooled by a heat-pipe (designed and realized at ENEA), machined in the anode up
to the plasma facing surface. A heat sink protruding outside the electrode ensures that the needed
dissipation takes place.
5.6.3 Measurements of the neutron yield
Measurements of the radiation emitted by a PF machine are not trivial, due to the short burst
duration (≈ 100 ns) and the strong electrical noise generated. Thus, the whole acquisition chain
must be shielded and possibly electrically decoupled from the rest of the machine. PF bursts,
moreover, also produce x and γ-rays, and since the majority of the neutron detectors are also
photon-sensitive, they cannot be operated in current mode (viz. without neutron/gamma
discrimination). These constraints require the use of detection techniques which either foresee a
passive method (activation), or employ a detector which is insensitive to photons.
Three independent methods were chosen to measure the neutron production of our PF. The first
method was aimed to measure the absolute neutron yield of individual shots; the second one, to
measure the total yield of a repetitive sequence of shots, and the third one attempted to measure
the neutron yields of individual shots within the sequence.
5.6.4 Neutron yield from single shots
The detector for single shots consists of four Geiger Müller tubes, wrapped in silver foils and
embedded in a polyethylene moderator. This counter was developed and calibrated at the Los
Alamos scientific laboratory. Its operational principle is based upon the Ag activation by low-
167

5. Miscellaneous
Fig. 5.18 - Neutron signal measured by the scintillator counter during
40 shots at 0.5 Hz
energy neutrons, obtained by
moderating the 2.5 MeV neutrons
emitted from the PF in the moderator
surrounding the GM tubes. The tubes
measure the beta decay from the
silver activated by neutrons, the
activity being proportional to the
neutron yield of the discharge. Since
the counting starts slightly after the
shot, this practice avoids the
electrical noise produced by the PF.
The silver counter was used to
measure the neutron yield operating
the PF in single-shot mode.
5.6.5 Neutron yield from a
sequence of shots
To measure the total yield of a
sequence of shots, the method of
neutron activation was used. The
197Au(n,γ)198Au reaction was
chosen, usually employed for thermal neutron detection. Gold foils were first irradiated with a
reference neutron source (FNG); then, their activity was measured. Subsequently (after their
total decay) the same foils were exposed to the sequence of PF shots and their activity measured
again. The comparison of the two activities allowed the unknown PF yield to be measured.
5.6.6 Neutron yields from individual shots within a sequence
If the sequence of shots has a frequency higher than 0.01 Hz, the silver counter for individual
shots cannot be used, because the decay time of silver becomes comparable to the repetition
time. To avoid the constraints due to the silver decay time, a scintillator detector was used to
record the neutron emission produced during the repetitive PF operation.
A 2” lithium-loaded NE422 scintillator coupled to a RCA 8875 photomultiplier was used. The
NE422 is sensitive to thermal neutrons and almost insensitive to photons, a good choice to the
purpose of minimizing x and gamma-ray pollution. A 6 cm thick polyethylene cylinder was
placed opposite the scintillator, to moderate the incoming neutrons, and the whole system was
inserted in a double cylindrical electromagnetic shield, made respectively of µ-metal and steel.
While slowing down the neutrons, the polyethylene cylinder spreads the burst over a few
hundreds µs, enough to be recorded by the scintillator, which has a recovery time of about 200 ns.
The scintillator was placed about 1 meter away from the PF head. The HV and the output signal
were sent to the control room through 80-meter double-screened cables. The output signal was
fed into a fast discriminator and then sent to a Multi-Channel Scaler (MCS). Contingent
conditions at the time of the experiment imposed a maximum resolution time of 50 µs.
Several measurements of shots with and without neutrons were made to verify the influence of
the electromagnetic noise on the counting; the conclusion was that the noise could not contribute
for more than ± 10 % to our typical signal level. An example of measurement performed during
a sequence of 40 PF shots at 0.5 Hz is shown in fig 5.18. The sporadical high variability shown
by some particular shots does not affect the practical use of the machine, which will operate for
hundreds of shots to reach typical irradiation levels.
168

References
[5.1] V. Boffa et al., Supercond. Sci. Technol. 13, 1467 (2000)
[5.2] V. Boffa et al., Influence of film thickness on the critical current of YBa 2 Cu 3 O 7-x thick
films on Ni-V biaxially textured substrate, 2000 Appl. Superc. Conf. (Virginia Beach 2000) to be
published on IEEE Trans. Appl. Supercond.
[5.3] A. Mancini et al., Int. J. Mod. Phys. B14, 3128 (2000)
[5.4] G. Giunchi et al., Int. J. Mod. Phys. B14, 3134 (2000)
[5.5] C. Fazio et al., Compatibility Tests of steels exposed to stagnant molten lead and lead
bismuth, IV Int. Workshop on Spallation Materials Technology (Schruns 2000) to be published
on J. Nucl. Mater.
[5.5] F. Barbier, A. Rusanov, Corrosion behavior of steels in flowing lead-bismuth, IV Int.
Workshop on Spallation Materials Technology (Schruns 2000) to be published on J. Nucl. Mater.
[5.7] F. Barbier et al., Compatibility tests of steels in flowing liquid lead-bismuth, to be published
on J. Nucl. Mater.
[5.8] ASM Handbook, Vol. 13, “Corrosion”, 1994
[5.9] G. Palombariniet al., Giornata di studio dei materiali in piombo e piombo - bismuto,
ENEA doc. ERG FUS ISP CMAT 008 (1997)
[5.10] I. Ricapito, F. Salvi, ENEA RTI NdL 1, July 2000
[5.11] C. Fazio, I. Ricapito and G. Benamati, Activities carried out by FUS ISP at ENEA-
Brasimone in the framework of TRSCO I National Programme, (to be published)
169

Publications and Conferences
Publications 173
Contributions to Conferences 179
Conferences and Seminars Held at Frascati in 2000 184

Publications and Conferences
PUBLICATIONS
00/03 S.E. SEGRE: Evolution of the polarization state for radiation propagating in a nonuniform,
birefringent, optically active, and dichroic medium: the case of a magnetized plasma
J. Opt. Soc. Am. A 17, 1, 95
00/06 S. TOSTI, L. BETTINALI, V. VIOLANTE: Rolled thin Pd and Pd-Ag membranes for
hydrogen separation and production
Int. J. Hydrogen Energy 25, 319
00/07 S. BRIGUGLIO, L. CHEN, J.Q. DONG, G. FOGACCIA, R.A. SANTORO, G. VLAD,
F. ZONCA: High and low frequency Alfvén modes in tokamaks
Nucl. Fusion 40, 3Y, 701
00/08 V. BOFFA, C. ANNINO, D. BETTINELLI, S. CERESARA, L. CIONTEA, F.
FABBRI, V. GALLUZZI, U. GAMBARDELLA, G. CELENTANO, G. GRIMALDI, A.
MANCINI, T. PETRISOR, P. SCARDI: Epitaxial growth of heterostructures on biaxially
textured metallic substrates for YBa 2 Cu 3 O 7-x tape fabrication
Philos. Mag. B 80, 5, 979
00/09 M. PILLON, M. ANGELONE, P. BATISTONI, R.A. FORREST, J.-CH. SUBLET:
Benchmark experiments of fusion neutron induced gamma-ray radioactivity in various structural
materials
J. Radioanal. Nucl. Chem. 244, 2, 441-445
00/010 S.E. SEGRE: Evolution of the state of polarization of radiation propagating in a
magnetized plasma including particle collisions
Plasma Phys. Controll. Fusion (Letts to the Editor) 42, L9-L12
00/015 D. PACELLA, K.B. FOURNIER, M. ZERBINI, M. FINKENTHAL, M. MATTIOLI,
M.J. MAY, W.H. GOLDSTEIN: Temperature and impurity transport studies of a heated
Tokamak plasmas by means of a collisional-radiative model of x-ray emissions from Mo 30+ to
Mo 39+
Phys. Rev. E 61, 5, 5701
00/017 T. PETRISOR, V. BOFFA, S. CERESARA, C. ANNINO, D. BETTINELLI, F.
FABBRI, U. GAMBARDELLA, G. CELENTANO, P. SCARDI, P. CARACINO: Non magnetic
Ni 100-x V x biaxially textured substrates for YBCO tape fabrication
Inst. Phys. Conf. Ser. No 167, 431
00/018 V. BOFFA, T. PETRISOR, C. ANNINO, F. FABBRI, D. BETTINELLI, G.
CELENTANO, L. CIONTEA, U. GAMBARDELLA, G. GRIMALDI, A. MANCINI, V.
GALLUZZI: High J c YBCO thick films on biaxially textured Ni-V substrate with CeO 2 /NiO
intermediate layers
Inst. Phys. Conf. Ser. No 167, 427
00/021 S.E. SEGRE, V. ZANZA: Polarization of radiation in incoherent Thomson scattering
by high temperature plasma
Phys. Plasmas 7, 6, 2677
173

Publications and Conferences
00/023 C. LO SURDO: Quasistatic evolution of a dissipative plasma column in vacuum
“Errata and addenda”
J. Plasma Phys. 63, 1, 21-41
00/025 S. BRIGUGLIO, G. VLAD, B. DI MARTINO, G. FOGACCIA: Parallelization of
plasma simulation codes: gridless finite size particle versus particle in cell approach
Future Generation Comp. Syst. 16, 541-552
00/029 S.E. SEGRE: Effect of ray refraction on evolution of the polarization state of radiation
propagating in a nonuniform, birefringent, optically active and dichroic medium
J. Opt. Soc. Am. A 17, 9, 1683
00/030 M. MARINELLI, E. MILANI, A. PAOLETTI, A TUCCIARONE, G. VERONA
RINATI, M. ANGELONE, M. PILLON: High collection efficiency in chemical vapor deposited
diamond particle detectors
Diamond Rel. Mat. 9, 998-1002
00/033 J.R. MARTÍN-SOLÍS, R. SÁNCHEZ, B. ESPOSITO: On the effect of synchrotron
radiation and magnetic fluctuations on the avalanche runaway growth rate
Phys. Plasmas 7, 9, 3814
00/034 P. TRIPODI, M.C.H. MCKUBRE, F.L. TANZELLA, P.A. HONNOR, D. DI
GIOACCHINO, F. CELANI, V. VIOLANTE: Temperature coefficient of resistivity at
compositions approaching PdH
Phys. Letts A 276, 1-5
00/035 F. ZONCA, L. CHEN: Destabilization of energetic particle modes by localized particle
sources
Phys. Plasmas 7, 11, 4600
00/037 S. ROLLET, M. ANGELONE, P. BATISTONI: Absorbed dose calculations for the
Ignitor tokamak magnet coils insulator
Nucl. Instrum. Methods in Phys. Res. B166-167, 826-830
00/038 S. ROLLET, P. BATISTONI, R. FORREST: Activation analysis for the Ignitor tokamak
Fusion Eng. Des. 51-52, 599-604
00/039 M. ANGELONE, P. BATISTONI, L. PETRIZZI, M. PILLON: Neutron streaming
experiment at FNG: results and analysis
Fusion Eng. Des. 51-52, 653-661
00/040 U. FISCHER, P. BATISTONI, Y. IKEDA, M.Z. YOUSSEF: Neutronics and nuclear
data: achievements in computational simulations and experiments in support of fusion reactor
design
Fusion Eng. Des. 51-52, 663-680
00/041 L. PETRIZZI, P. BATISTONI, I. KODELI: Sensitivity and uncertainty analysis
performed on a 14-MeV neutron streaming experiment
Fusion Eng. Des. 51-52, 843-848
174

Publications and Conferences
00/042 K. SEIDEL, M. ANGELONE, P. BATISTONI, U. FISCHER, H. FREIESLEBEN, W.
HANSEN, M. PILLON, D. RICHTER, S. UNHOLZER: Investigation of neutron and photon
flux spectra in a streaming mock-up for ITER
Fusion Eng. Des. 51-52, 855-861
00/043 R. COPPOLA, C. NARDI, B. RICCARDI: High temperature residual strain
measurements in a brazed sample for NET/ITER
J. Nucl. Mat. 283-287, 1243-1247
00/044 A. HASEGAWA, A. KOHYAMA, R.H. JONES, L.L. SNEAD, B. RICCARDI P.
FENICI: Critical issues and current status of SiC/SiC composites for fusion
J. Nucl Mat. 283-287, 128-137
00/045 PH. CHAPPUIS, F. ESCOURBIAC, M. CHANTANT, M. FEBVRE, M.
GRATTAROLA, M. BET, M. MEROLA, B. RICCARDI: Infrared characterization and high heat
flux testing of plasma sprayed layers
J. Nucl. Mat. 283-287 1081-1084
00/046 F. FABBRI, C. ANNINO, V. BOFFA, G. CELENTANO, L. CIONTEA, U.
GAMBARDELLA, G. GRIMALDI, A. MANCINI, T. PETRISOR: Properties of biaxially
oriented Y 2 O 3 based buffer layers deposited on cube textured non-magnetic Ni-V substrates for
YBCO couted conductors
Physica C 341-348, 2503-2504
00/047 G. CELENTANO, C. ANNINO, V. BOFFA, L. CIONTEA, F. FABBRI, U.
GAMBARDELLA, V. GALLUZZI, G. GRIMALDI, A. MANCINI, L. MUZZI, T. PETRISOR:
Superconducting and structural properties of YBCO thick films grown on biaxially oriented
CeO 2 /NiO/Ni-V architecture
Physica C 341-348, 2501-2502
00/048 A. CARUSO, C. STRANGIO, S.YU. GUS’KOV, V.B. ROZANOV: Interaction
experiments of laser light with low density supercritical foams at the AEEF ABC facility
Laser Part. Beams 18, 25-35
00/049 A. CARDINALI: Quasilinear absorption of the lower hybrid wave in tokamak plasmas
Recent Res. Devel. Plasmas, 1, 185-197
00/050 S. ROLLET, M. ANGELONE, P. BATISTONI: Absorbed dose calculations for the
Ignitor tokamak magnet coils insulator
Nucl. Instrum. Methods Phys. Res. B 166-167, 826-830
00/051 R. ANDREANI: What is lacking in order to design and build a commercially viable
fusion reactor
Nucl. Fusion 40, 6, 1033-1046
00/052 T. PINNA, L.C. CADWALLADER: Component failure rate data base for fusion
applications
Fusion Eng. Des. 51-52, 579-585
00/053 M.T. PORFIRI, G. CAMBI: Integrated safety analysis code system (ISAS) application
175

Publications and Conferences
for accident sequence analyses
Fusion Eng. Des. 51-52, 587-591
00/055 J.R. MARTÍN-SOLÍS, B. ESPOSITO, R. SÁNCHEZ, L. BERTALOT, S. ROLLET,
Y.A. KASCHUCK, D.V. PORTNOV: Runaway electron measurements in FTU tokamak
Europhys. Conf. Abstracts (ECA) 24B, 165-168
00/056 F. CRISANTI, B. ESPOSITO, L. BERTALOT, C. GIROUD, C. GORMEZANO, C.
GOWERS, R. PRENTICE, A. TUCCILLO, K.-D. ZASTROW, M. ZERBINI: ExB flow shearing
rate evalution in JET ITB discharges
Europhys. Conf. Abstracts (ECA) 24B, 153-156
00/057 B. ESPOSITO, L. BERTALOT, G. MARUCCIA, L. PETRIZZI, G. BIGNAN, C.
BLANDIN, S. CHAUFFRIAT, A. LEBRUN, H. RECROIX, J.P. TRAPP, Y. KASCHUCK:
Results from the CDE phase activity on neutron dosimetry for the international fusion materials
irradiation facility test cell
Fusion Eng. Des. 51-52, 331-338
00/058 F. ZONCA, S. BRIGUGLIO, L. CHEN, G. FOGACCIA, G. VLAD: Theoretical
aspects of collective mode excitations by energetic ions in tokamaks
ISSP-19 “Piero Caldirola” Theory of Fusion Plasmas (J.W. Connor, O. Sauter and E. Sindoni
(Eds.)) SIF, Bologna 2000, p. 17-30
00/063 E.P. BALSAMO, P. GISLON, G. PASOTTI, M.V. RICCI, M. SPADONI, J.V.
MINERVINI, V.S. VYSOTSKY: Experimental study of the current redistribution in pulsed
operation inside the Nb 3 Sn CICC of an ITER relevant magnet
IEEE Trans. Appl. Supercond. 10, 2, 1598-1602
00/064 M. CIOTTI, P. GISLON, M. MORONI, M. SPADONI, T. PETRISOR, V. POP: Test
and qualification of a variable temperature vibrating sample magnetometer system for the
measurement of magnetization cycles up to ± 12T in the 4.2K - 300K temperature range
Int. J. Modern Phys. B 14, 25-27, 2914-2919
00/065 A. MANCINI, V. BOFFA, G. CELENTANO, L. CIONTEA, M. DAMASCENI, F.
FABBRI, U. GALLUZZI, G. GRIMALDI, T. PETRISOR: Development of buffer layer
structures for Yba 2 Cu 3 O 7-δ coated conductors on textured Ni-V substrate
Int J. Modern Phys B 14, 25-27, 3128-3133
00/066 G. GIUNCHI, S. CERESARA, R. CORTI, T. PETRISOR, A. MANCINI, G.
CELENTANO: A new metallic non magnetic substrate for couted tape superconductors
Int. J. Modern Phys. B 14, 25-27, 3134-3138
00/067 P. BELLUCCI, M. CIOTTI, P. GISLON, M. SPADONI, L. BOTTURA, L. MUZZI, S.
TURTU’: Comparison between the predictions of the thermo-hydraulic code Gandalf and the
results of a long length instrumented CICC module experiment
Cryog. 40, 555-559
00/068 A. DE SANTIS, G. GRIMALDI, U. GAMBARDELLA, S. PACE, A.M. CUCOLO,
M.C. CUCOLO, V. BOFFA, G. CELENTANO, F. FABBRI: Voltage-current characteristics of c-
axis oriented YBa 2 Cu 3 O 7-δ , films deposited by dc sputtering
176

Publications and Conferences
Physica C 340, 225-229
00/069 V. BOFFA, T. PETRISOR, G. CELENTANO, F. FABBRI, C. ANNINO, S.
CERESARA, L. CIONTEA, V. GALLUZZI, U. GAMBARDELLA, G. GRIMALDI, A.
MANCINI: Epitaxial growth of YBa 2 Cu 3 O 7-δ on Ni 89 V 11 non-magnetic biaxially textured
substrate using NiO as buffer layer
Supercond. Sci. Technol. 13, 1467-1469
00/070 F. FABBRI, G. PADELETTI, T. PETRISOR, G. CELENTANO, V. BOFFA: Surface
morphology of pulsed laser deposited YBa 2 Cu 3 O 7-δ , and NdBa 2 Cu 3 O 7-δ thin films on
SrTiO 3 substrates
Supercond. Sci. Technol. 13, 1492-1498
00/071 A. DELLA CORTE, R. BRUZZESE, S. CHIARELLI, M. SPADONI, V. CAVALIERE,
M. MARIANI, G. MASULLO, A. MATRONE, E. PETRILLO, R. QUARANTIELLO: Design
and manufacture of a Nb 3 Sn 16T demountable insert solenoid for a high field testing system
IEEE Trans. Appl. Supercond. 10, n. l, 458-461
00/072 E. BALSAMO, O. CICCHELLI, M. CUOMO, A. DELLA CORTE, P. GISLON, G.
PASOTTI, M. RICCI, M. SPADONI: Performance tests on an ITER relevant CICC Nb 3 Sn coil
IEEE Trans. Appl. Supercond. 10, n. l, 572-575
00/073 R. MAIX, H. FILLUNGER, F. HURD, J. PALMER, E. SALPIETRO, N. MITCHELL,
P. DECOOL, P. LIBEYRE, A. ULBRICHT, G. ZAHN, A. DELLA CORTE, R. GARRE’, B.
SCHELLONG, A. LAURENTI, N. VALLE, A. BOURQUARD, D. BRESSON, E. THEISEN:
Manufacture, assembly and Q A
of the ITER toroidal field model coil
IEEE Trans. Appl. Supercond. 10, n. 1, 584-587
00/075 S. BRIGUGLIO, G. VLAD, B. Dl MARTINO, G. FOGACCIA: Parallelization of
particle codes for the simulation of Alfvénic turbulence: gridless finite size partive versus PIC
approach
Estratto dal Proc. 17th Int. Conf. on the Numerical Simulation of Plasmas (2000) p. 37-41
00/076 B. Dl MARTINO, S. BRIGUGLIO, G. FOGACCIA, G. VLAD: Programming shared
memory architectures with OpenMP: a case study
Estratto dal Proc. 2nd European Workshop on OpenMP (EWOMP 2000) p. 9- 13
00/077 B. Dl MARTINO, S. BRIGUGLIO, G. FOGACCIA, G. VLAD: Parallel PIC codes for
distributed and shared memory architectures with HPF and OpenMP
Estratto dal Proc. of the Int. Conf. on Parallel and Distributed Processing Techniques and
Applications (PDPTA 2000) (Editor H R Arabnia, World Scientific Eng. Soc.) Vol. IV, p. 2233-
2239
00/078 A. BADALÀ, R. BARBERA, F. LIBRIZZI, A. PALMERI, G.S. PAPPALARDO, F.
RIGGI, S. DI LIBERTO, F. MEDDI, S. SESTITO, D. LOI, M. ANGELONE, M. PILLON:
Irradiation measurements on the 0.25 mm CMOS pixel readout test chip by a 14 MeV neutron
facility
ALICE/ITS 2000-24 (CERN/Internal Note-ITS)
00/079 V. VIOLANTE, G.H. MILEY, P. TRIPODI, D. Dl GIACCHINO, C. SIBILIA: Recent
177

Publications and Conferences
results from collaborative research at ENEA-Frascati on reaction phenomena in solids
Low-Energy Nucl. Reactions I,361-362
00/080 G. BENAMATI, P. BUTTOL, V. IMBENI, C. MARTINI, G. PALOMBARINI:
Behaviour of materials for accelerator driven systems in stagnant molten lead
J. Nucl. Mater. 279, 308-316
00/081 E. SERRA, E. RIGAL, G. BENAMATI: Hydrogen and deuterium permeation
measurements on the double-wall tubes material for the water-cooled Pb-17Li DEMO blanket
Fusion Eng. Des. 49-50, 675-679
00/083 G. VERRI, F. MEZZETTI, A. DA RE, A. BORTOLOTTI, L. RAPEZZI, V.A.
GRIBKOV: Fast neutron activation analysis of gold by inelastic scattering,
197 Au(n,n’γ) 197 Au m , by means of Plasma Focus devices
NUCLEONIKA 45, 3, 189-191
00/084 B. RICCARDI: EU activities on SiC f
/SiC composites for fusion
Fourth International Energy Agency Workshop on SiC f
/SiC Ceramic Composites for Fusion
Application, (Frascati, October 12-13) p.9
00/085 C.A. NANNETTI, B. RICCARDI, A. ORTONA, A. LA BARBERA, E. SCAFÈ:
Manufacturing of advanced three-dimensional texture SiC f /SiC composites
Fourth International Energy Agency Workshop on SiC f
/SiC Ceramic Composites for Fusion
Application, (Frascati, October 12-13) p.92
00/086 M. FERRARIS, B. RICCARDI, M. SALVO: High characteristic temperature SiC f
/SiC
composite glass-ceramic coutings for fusion application
Fourth International Energy Agency Workshop on SiC f
/SiC Ceramic Composites for Fusion
Application, (Frascati, October 12-13) p.213
00/087 L.C. ALVES, E. ALVES, A. PAUL, M.F. DA SILVA, J.C. SOARES, A. LA
BARBERA, B. RICCARDI: Surface renctions of SiC/SiC f
composites studied with microbeams
Fourth International Energy Agency Workshop on SiC f
/SiC Ceramic Composites for Fusion
Application, (Frascati, October 12-13) p. 220
00/088 M. ANGELONE, M. CHITI, A. ESPOSITO, A. GENTILE: Mensurement of the γ-ray
dose around DAΦNE collider using different types of high sensitivity TLDs
J. Nucl. Science Techn. Suppl. 1, 758-761
00/092 P. COLOMBO, B. RICCARDI, A. DONATO, G. SCARINCI:Joining of SiC/SiC f
Ceramic matrix composites for fusion reactor blanket applications
J. Nucl. Mater. 278, 127-135
00/093 B. RICCARDI, P. FENICI, A. FRIAS REBELO, L. GIANCARLI, G. LE MAROIS, E.
PHILIPPE
Status of the European R&D activities on SiC f
/SiC composites for fusion reactors
Fusion Eng. Des. 51-52, 11-22
178

Publications and Conferences
CONTRIBUTIONS TO CONFERENCE
M.PILLON, M. ANGELONE, R.A. FORREST: A new detector to measure gamma and beta
decay power from radionuclides
Presented at the 8th Pisa Meeting on Advanced Detectors (La Biodola, Isola d’Elba, May 21-27,
2000)
M. ANGELONE, P. BATISTONI, M. PILLON: Effect of the encapsulating material on the
peak3/peak5 response ratio of TLD-300 irradiated with neutrons of variuos energy
Presented at the 8th Int. Symp. on Radiation Physics (ISRP-8) (Praga, June 5-9, 2000)
V. VIOLANTE, G.H. MILEY, P. TRIPODI, D. Dl GIACCHINO, C. SIBILIA: Recent results
from collaborative research at ENEA-Frascati on reaction phenomena in solids
Presented at the Winter Meeting of the American Nuclear Society (ANS) (Washington,
November 12-17, 2000)
R. DE ANGELIS, S. E. SEGRE, N. TARTONI, V. ZANZA: The motional stark effect diagnostic
in FTU
Presented at the Thirteenth Topical Conference on High-Temperature Plasma Diagnostics
(Tucson, Arizona June 18-22, 2000)
M.L. APICELLA, G. APRUZZESE, R. DE ANGELIS, G. GATTI, M. LEIGHEB, D. PACELLA,
G. MAZZITELLI, V. PERICOLI-RIDOLFINI: Effects of wall titanium couting on FTU plasma
operations Presented at the EPS-27 Conference on Controlled Fusion and Plasma Physics
(Budapest, June 12- 16, 2000)
D. PACELLA, M. LEIGHEB, M.J. MAY, M. FINKENTHAL, M. MATTIOLI, L.
GABELLIERI, K.B. FOURNIER: Peculiarities of space and time evolution of iron impurity
injected in the FTU tokamak plasmas
Presented at the EPS - 27 Conference on Controlled Fusion and Plasma Physics (Budapest, June
12- 16, 2000)
M. MCKUBRE, F. TANZELLA, P. TRIPODI, D. Dl GIOACCHINO, V. VIOLANTE: Finite
element modeling of the transient calorimetric behaviour of the MATRIX experimental
apparatus: 4 He and excess of power production correlation through numerical results
Presented at the ICCF-8 - VIII International Conference on Cold Fusion (Lerici, La Spezia,
Maggio 21-26, 2000)
R. GARRE’, S. ROSSI, R. BRUZZESE, S. CHIARELLI, P. GISLON, M. SPADONI: The
multifilamentary internal TIN Nb 3 Sn strand for the ITER toroidal field model coil
Presented at the ICMC 2000 (Rio de Janeiro, Brasile June 5-9, 2000)
E.VISCA, B. RICCARDI, A. ORSINI, C. TESTANI: Manufacturing and testing of monoblock
tungsten small-scale
Presented at the 21st Symposium on Fusion Technology (Madrid, September 11-15, 2000)
B. RICCARDI, R. MONTANARI, L.F. MORESCHI, A. SILI, S. STORAI: Mechanical
characterisation of fusion materials by indentation test
Presented at the 21st Symposium on Fusion Technology (Madrid, September 11-15, 2000)
179

Publications and Conferences
M. FERRARI, L. GIANACARLI, K. KLEEFELDT, C. NARDI, M. RODIG, J. REIMANN:
Evaluation of divertor conceptual designs for a fusion power plant
Presented at the 21st Symposium on Fusion Technology (Madrid, September 11-15, 2000)
T. PINNA, R. CAPORALI, L. BURGAZZI: Selection of accident sequences for the new design
of ITER
Presented at the Int. Conf. on Probabilistic Safety Assessment and Management PSAM5 (Osaka,
November 27-December 1, 2000)
G. CELENTANO, A. CAPRICCIOLI, A. CUCCHIARO, M. GASPAROTTO, A. BIANCHI, G.
FERRARI, B. PARODI, G.P. SANGUINETTI, F. VIVALDI, S. ORLANDI, B. COPPI:
Engineering evolution of the ignitor machine
Presented at the 21st Symposium on Fusion Technology (Madrid, September 11-15, 2000)
A. CARDINALI: Complex ray tracing method in high harmonic fast wave propagation and
absorption
Presented at the Int. School of Plasma Physics “Piero Caldirola” (Varenna, August 28-September
1, 2000)
L. BARTOLINI, A. COLETTI, M. FERRI DE COLLIBUS, G. FORNETTI, C. NERI, M. RIVA,
L. SEMERARO, C. TALARICO: Experimental results of laser in vessel viewing system
(LIVVS) for JET
Presented at the 21st Symposium on Fusion Technology (Madrid, September 11-15, 2000)
A. BERTOCCHI, G. BRACCO, G. BUCETI, C. CENTIOLI, F. IANNONE, G. MANDUCHI,
M. PANELLA, U. NANNI, C. STRACUZZI, V. VITALE: Recent developments and objectoriented
approach in the FTU base
Presented at the 21st Symposium on Fusion Technology (Madrid, September 11-15, 2000)
G. FERMANI, M. ZARFINO: A software environment to execute automatic operational
sequences on the ITER-FEAT DPT facility
Presented at the 21st Symposium on Fusion Technology (Madrid, September 11-15, 2000)
F. ALLADIO, L.A. GROSSO, A. MANCUSO, S. MANTOVANI, P. MICOZZI, G.
ABRUZZESE, L. BETTINALI, P. BURATTI, R. DE ANGELIS, G. GATTI, G. MONARI, M.
PILLON, A. SIBIO, B. TILIA, O. TUDISCO: Results of proto-pinch test bench for the protosphera
experiment
Presented at the 27th Conf. on Controlled Fusion and Plasma Physics (EPS) (Budapest, June 12-
16, 2000)
L. BOTTURA, M. CIOTTI, P. GISLON, M. SPADONI, P. BELLUCCI, L. MUZZI, S. TURTU’
A. CATITTI, S. CHIARELLI, A. DELLA CORTE, E. DI FERDINANDO: Stability in a long
length NbTi CICC
Presented at the 2000 Applied Superconductivity Conference (Virginia Beach, Sept. 17-22,
2000)
P. BELLUCCI, M. CIOTTI, P. GISLON, M. SPADONI, L. BOTTURA, L. MUZZI, S. TURTU’:
Comparison between the predictions of the thermohydraulic code Gandalf and the results of a
long length instrumented CICC module experiment
Presented at the SATT 10 Congresso Nazionale di Superconduttività (Frascati, Sept. 6-8, 2000)
180

Publications and Conferences
A. DE NINNO, A. FRATTOLILLO, A. RIZZO, F. SCARAMUZZI, C. ALESSANDRINI: A new
method aimed at detecting small amounts of helium in a gaseous mixture
Presented at the ICCF-8 - VIII Int. Conf on Cold Fusion (Lerici, La Spezia, Maggio 21-26, 2000)
E. DEL GIUDICE, A. DE NINNO, A. FRATTOLILLO, G. PREPARATA, F. SCARAMUZZI,
A. BULFONE, M. COLA, C. GIANNETTI: The Fleischmann-Pons effect in a novel electrolytic
configuration
Presented at the ICCF-8 - VIII Int Conf on Cold Fusion (Lerici, La Spezia, Maggio 21-26, 2000)
E. DEL GIUDICE, A. DE NINNO, A. FRATTOLILLO, G. PREPARATA, F. SCARAMUZZI, P.
TRIPODI:Looding palladium with deuterium gas while lowering temperature
Presented at the ICCF-8 - VIII Int. Conf. on Cold Fusion (Lerici, La Spezia, Maggio 21-26,
2000)
D. PACELLA, M. LEGHEB, L. GABELLIERI, L. PANACCIONE, M. ZERBINI, M.
MATTIOLI, M. MAY, M. FINKENTHAL, K.B. FOURNIER, W.H. GOLDSTEIN: Mid-high z
impurities as diagnostic tools in tokamak plasmas
Presented at the 18th IAEA Fusion Energy Conference (Sorrento, October 4-10, 2000)
G. BRACCO, A. BRUSCHI, P. BURATTI, S. CIRANT, F. CRISANTI, B. ESPOSITO, D.
FRIGIONE, E. GIOVANNOZZI, G. GIRUZZI, G. GRANUCCI, V. KRIVENSKI, C. SOZZI, O.
TUDISCO, V. ZANZA, F. ALLADIO, B. ANGELINI, M.L. APICELLA, G. APRUZZESE, E.
BARBATO, L. BERTALOT, A. BERTOCCHI, G. BUCETI, A. CARDINALI, S. CASCINO, C.
CASTALDO, C. CENTIOLI, R. CESARIO, P. CHUILLON, S. CIATTAGLIA, V. COCILOVO,
R. DE ANGELIS, M. DE BENEDETTI, E. DE LA LUNA, F. DE MARCO, B. FRANCIONI, L.
GABELLIERI, G. GATTI, C. GORMEZANO, F. GRAVANTI, M. GROLLI, F. IANNONE, H.
KROEGLER, M. LEIGHEB, G. MADDALUNO, G. MAFIA, M. MARINUCCI, G.
MAZZITELLI, P. MICOZZI, F. MIRIZZI, S. NOWAK, F.P. ORSITTO, D. PACELLA, L.
PANACCIONE, M. PANELLA, F. PAPITTO, V. PERICOLI RIDOLFINI, L. PIERONI, S.
PODDA, F. POLI, G. PULCELLA, G. RAVERA, G.B. RIGHETTI, F. ROMANELLI, M.
ROMANELLI, A. RUSSO, F. SANTINI, SASSI, S.E. SEGRE, A. SIMONETTO, P.
SMEULDERS, S. STERNINI, N. TARTONI, P.E. TRAVISANUTTO, A.A. TUCCILLO, V.
VITALE, G. VLAD M. ZERBINI, F. ZONCA: ECRH results during current ramp-up post-pellet
injection in FTU plasma
Presented at the 18th IAEA Fusion Energy Conference (Sorrento, October, 4-10, 2000)
C. GORMEZANO: Overview of JET results in support of the ITER physics basis
Presented at the 18th IAEA Fusion Energy Conference (Sorrento, October 4-10, 2000)
P. BURATTI AND JET TEAM: High beta plasmas and internal barrier dynamics in JET
discharges with optimised shear
Presented at the 18th IAEA Fusion Energy Conference (Sorrento, October, 4-10 2000)
F. ALLADIO, B. ANGELINI, M.L. APICELLA, G. APRUZZESE, E. BARBATO, L.
BERTALOT, A. BERTOCCHI, G. BRACCO, A. BRUSCHI, G. BUCETI, P. BURATTI, A.
CARDINALI, S. CASCINO, C. CASTALDO, C. CENTIOLI, R. CESARIO, P. CHUILLON, S.
CIATTAGLIA, S. CIRANT, V. COCILOVO, F. CRISANTI, R. DE ANGELIS, M. DE
BENEDETTI, E DE LA LUNA G. GIRUZZI, F. DE MARCO, B. ESPOSITO, M.
FINKENTHAL, B. FRANCIONI, D. FRIGIONE, L. GABELLIERI, F. GANDINI, G. GATTI,
181

Publications and Conferences
E. GIOVANNOZZI, C. GORMEZANO, F. GRAVANTI, G. GRANUCCI, M. GROLLI, F.
IANNONE, V. KRIVENSKI, H. KROEGLER, E. LAZZARO, M. LEIGHEB, G.
MADDALUNO, G. MAFFIA, M. MARINUCCI, G. MAZZITELLI, M. MAY, P. MICOZZI, F.
MIRIZZI, S. NOWAKI, F.P. ORSITTO, D. PACELLA, L. PANACCIONE, M. PANELLA, P.
PAPITTO, V. PERICOLI-RIDOLFINI, A.A. PETROV, L. PIERONI, S. PODDA, F. POLI, G.
PUCELLA, G. RAVERA, G.B. RIGHETTI, F. ROMANELLI, M. ROMANELLI, A. RUSSO, F.
SANTINI, M. SASSI, S.E. SEGRE, A. SIMONETTO, P. SMEULDERS, E. STERNINI, C.
SOZZI, N. TARTONI, B. TILIA, P. E. TREVISANUTTO, A. A. TUCCILLO, O. TUDISCO, V.
VERSHKOV, V. VITALE, G. VLAD, V. ZANZA, M. ZERBINI, F. ZONCA: Overview of the
FTU results
Presented at the 18th IAEA Fusion Energy Conference (Sorrento, October 4-10, 2000)
L. CHEN, Z. LIN, R.B. WHITE, F. ZONCA: Nonlinear zonal dynamics of drift and drift-Alfvén
turbulences in tokamak plasmas
Presented at the 18th IAEA Fusion Energy Conference (Sorrento, October 4-10, 2000)
F. ZONCA, S. BRIGUGLIO, L. CHEN, G. FOGACCIA, G. VLAD, L.-J. ZHENG: Energetic
particle mode dynamics in tokamaks
Presented at the 18th IAEA Fusion Energy Conference (Sorrento, October 4-10, 2000)
V. PERICOLI RIDOLFINI, Y. PEYSSON, R. DUMONT, G. GIRUZZI, G. GRANUCCI, L.
PANACCIONE, L. DELPECH, B. TILIA, FTU TEAM, ECH TEAM: Combined LH and ECH
experiments in the FTU tokamak
Presented at the 18th IAEA Fusion Energy Conference (Sorrento, October 4-10, 2000)
A. NATALIZIO, L. DI PACE, T. PINNA: Assessment of occupational radiation exposure for two
fusion power plant designs
Presented at the IAEA Technical Committee Meeting on Fusion Reactor Safety (FI-TC-1165)
(Cannes, France, June 13-16, 2000)
A. NATALIZIO, T. PINNA, L. DI PACE: Impact of plant incidents on worker radiation exposure
for the SEAFT design
Presented at the 21st Symposium on Fusion Technology (Madrid, September 11-15, 2000)
D.G. CEPRAGA, G. CAMBI, M. FRISONI, L. DI PACE: Dose rate outside cryostat of the
SEAFP-2 fusion plant
Presented at the 21st Symposium on Fusion Technology (Madrid, September 11-15, 2000)
F. DE MARCO, P. PAPITTO, G. MARROCCO, F. BARDATI: Modellistica elettromagnetica di
una camera a risonanza elettronica ciclotronica con il metodo FDTD
Presented at the RINEM 2000 (Riunione Nazionale di Elettromagnetismo) (Como, Settembre
18-24, 2000)
M. CIOTTI, P. GISLON, M. MORONI, M. SPADONI, T. PETRISOR, V. POP: Test and
qualification of a variable temperature vibrating sample magnetometer system for the
measurement of magnetization cycles up to ±12T in the 4.2K - 300K temperature range
Presented at the SATT-10 (Frascati, Maggio 9-12, 2000)
V. BOFFA, G. CELENTANO, L. CIONTEA, F. FABBRI, V. GALLUZZI, U.
GAMBARDELLA, G. GRIMALDI, A. MANCINI, T. PETRISOR: Influence of film thickness
182

Publications and Conferences
on the critical current of YBa 2 Cu 3 O 7-x thick films on Ni-V biaxially textured substrates
Presented at the 2000 Applied Superconductivity Conference (Virginia, September 17-22, 2000)
D. FRIGIONE, E. GIOVANNOZZI, C. GORMEZANO, F. POLI, M. ROMANELLI, O.
TUDISCO, F. CRISANTI, B. ESPOSITO, L. GABELLIERI, L. GARZOTTI, M. LEIGHEB, D.
PACELLA, AND FTU TEAM: Steady improved confinement in FTU high field plasmas
sustained by deep pellet injection
Presented at the 18th IAEA Fusion Energy Conference (Sorrento, October 4-10, 2000)
C. NERI, L. BARTOLINI, A.COLETTI, M. FERRI DE COLLIBUS, G. FORNETTI, M. RIVA,
L. SEMERARO, C. TALARICO: AM laser in-vessel viewing system for thermonuclear fusion
machine: First experimental results and extension to ranging
Presented at the 14th ANS Topical Meeting on the Technology of Fusion Energy (Park City, Utah
October 15-19, 2000)
M. COLA, E. DEL GIUDICE, A. DE NINNO, G. PREPARATO: A simple model of the “Cohn-
Aharonov” effect in a peculiar electrolytic configuration
Presented at the ICCF-8 - VIII Int Conf on Cold Fusion (Lerici, La Spezia Maggio 21-26, 2000)
V. VIOLANTE, C. SIBILIA, D. Dl GIOACCHINO, M. MCKUBRE, F. TANZELLA, P.
TRIPODI: Hydrogen isotopes interaction dynamics in palladium lattice
Presented at the ICCF-8 - VIII Int Conf. on Cold Fusion (Lerici, La Spezia Maggio 21-26, 2000)
P. BATISTONI, M. ANGELONE, L. PETRIZZI, M. PILLON: Experimental validation of shut
down dose rates calculations inside ITER cryostat
Presented at the 21st Symposium on Fusion Technology (Madrid, September 11-15, 2000)
L. PETRIZZI, G. BROLATTI, A. DAL SANTO, F. LUCCA, A. MARIN, G. MAZZONE, M.
MEROLA, M. ROCCELLA, L. SEMERARO, G. VIEIDER: Design of a welded box divertor
cassette for ITER FEAT
Presented at the 21st Symposium on Fusion Technology (Madrid, September 11-15, 2000)
G. VLAD, F. ZONCA: High-n ideal TAE stability of ITER
Presented at the 18th IAEA Fusion Energy Conference (Sorrento, October 4-10, 2000)
E. BARBATO, A. BRUSCHI, C. CANDELA, R. CESARIO, P. CHUILLON, S. CIRANT, R.
CMAESEN, V. COCILOVO, A. COLETTI, C. CRESCENZI, F. CRISANTI, A. CUCCHIARO,
R. DE ANGELIS, G. FERMANI, M. GASPAROTTO, G. GRANUCCI, E. LAZZARO, G.
MADDALUNO, M. MARINUCCI, G. MAZZITELLI, S. NOVAK, S. PAPASTERGIOU, V.
PERICOLI, L. PIERONI, G. RAMPONI, U. ROCCELLA, F. ROMANELLI, M. SANTINELLI,
L. SEMERARO, C. SOZZI, F. STARACE, A.A. TUCCILLO, O. TUDISCO, V. VITALE, L.
ZANNELLI, R. ALBANESE, G. AMBROSINO, M. ARIOLA, A. PIRONT, F. VILLONE: The
FTU-D project
Presented at the l8th IAEA Fusion Energy Conference (Sorrento, October 4-10, 2000)
M. ANGELONE, A. ESPOSITO, M. CHITI, A. GENTILE: Measurement of total absorption
coefficients for four mixtures using x-rays from 13 keV up to 40 keV
Presented at the 8th Int. Symp. on Radiation Physics (Praga, June 5-9, 2000)
183

Publications and Conferences
CONFERENCES AND SEMINARS
The Nuclear Fusion Department promotes the dissemination of information on plasma physics
and fusion technology, both nationally and internationally.
Conferences organized at the ENEA Frascati in 2000
Sorrento, 4-10/10/00:
Frascati, 11-13/10/00:
Frascati, 11-13/10/00:
Frascati, 11-13/10/00:
Frascati, 11-13/10/00:
Operation
18th IAEA Fusion Energy Conference
International Workshop on Heating and Transport in Tokamaks
International Workshop on Confinement Database & Modelling
International Workshop on Transport and Internal Barrier Physics
International Workshop on Fast Particle Heating and Steady State
Seminars organized and held at Frascati in 2000
14 02 2000 BARTIROMO R. - Consorzio RFX - Padova, Italy
Reversed field pinch physics at the RFX experiment
22 02 2000 XING ZHONG LI - Tsinghua Univ. - Beijing, China
Sub-barrier and selective resonant tunneling
27 03 2000 ZONCA F. - ENEA Frascati, Italy
Theoretical aspects of collective mode excitations by energetic ions in tokamaks
08 05 2000 TESSAROTTO M. - Università di Trieste, Trieste, Italy
Teorie cinetiche inverse in fluidodinamica e loro applicazioni
08 06 2000 SHOUCRI M. - CCFM Montreal, USA
The numerical integration of the Vlasov equation possessing an invariant: application to the
study of the formation of a charge separation and an electric field at a plasma edge
25 10 2000 BICKFORD HOOPER E. - Lawrence Livermore Nat. Lab. Livermore, USA
Plasma sustainment and confinement in the sustained spheromak physics experiment, SSPX
184

Organization Chart
FUSION - DIVISION DIRECTORATE
FRASCATI
Projects
M.Samuelli
Assoc. Directors: F. De Marco
G.B. Righetti
Plasma Physics Application
L. Rapezzi
Scientific Secretariat
F. De Marco (acting)
Intense Neutron Source
B. Riccardi
Administration & Control
N. Manganiello
Radiofrequency
G.B. Righetti (acting)
JET/NET Personnel
M. Samuelli (acting)
Conceptual Reactor Studies
A. Pizzuto (acting)
Research Center Brasimone
D. Cassarini
NET/ITER
A. Pizzuto
Electrical Engineering
A. Coletti
Deputy Director Experimental Engineering
G. Benamati
Deputy Director Fusion Technology
A. Pizzuto
Inertial Confinement Fusion
A. Caruso
Deputy Dir. Magnetic Confinement Fusion Physics
F. Romanelli

List of Personnel
Fusion Division Directorate Frascati 189
Projects 189
Scientific Secretariat 189
Inertial Confinement Fusion 189
Electrical Engineering 189
Technical and Administrative Support 190
Magnetic Confinement Fusion Physics 190
Fusion Technology 191
Experimental Engineering Brasimone 192
Research Center Brasimone 193

List of Personnel
The following list of Nuclear Fusion Division personnel is ordered according to the divisions and units
shown in the Organisation Chart. Some of the staff belong to the EURATOM-ENEA Association.
Superscripts: 1) EURATOM personnel; 2) On leave at JET; 3) On mission at NET; 4) On leave
FUSION DIVISION DIRECTORATE
FRASCATI
Samuelli Maurizio (Director)
De Marco Francesco (Associate Director)
Righetti Giovan Battista (Associate Director)
Albanese Angelina
Lazzarini Giovanna
Novelli Maria Rita
Scientific Advisors and Assistants to the
Director
Sacchetti Nicola
Santini Franco 1
JET/NET Personnel
Malavasi Battista 3
Nannetti Morena 3
Salpietro Ettore 3
Tanga Arturo 2
Tesini Alessandro 2 PROJECTS
Radiofrequency
Righetti G. Battista (acting)
Sassi Mauro
Plasma Physics Application
Rapezzi Luigi
Conceptual Reactor Studies
Pizzuto Aldo Maria
Intense Neutron Source (IFMIF)
Riccardi Bruno
NET/ITER
Samuelli Maurizio (acting)
SCIENTIFIC SECRETARIAT
De Marco Francesco (acting)
Arcangeli Donatella
Bocci Laura Zita
Bottomei Mauro
Cecchini Marisa
Crescentini Lucilla
Ghezzi Lucilla
Polidoro Maria
Riske Hans Peter
Vendetti Lucia
INERTIAL CONFINEMENT
FUSION
Caruso Angelo
Andreoli Pierluigi
Cristofari Giuseppe
Dattola Antonino
Fioravanti Laura
Montanari Enrico
Strangio Carmela
ELECTRICAL ENGINEERING
Coletti Alberto
Aquilini Massimo
Baccarelli Gianfranco
Berardi Beniamino
Callegari Mauro
Candela Guido
Claesen Renier 1
Ciccone Giovanni
Costa Pietro
Del Prete Enrico
Di Domenincantonio Mario
Di Giovenale Sergio
Domenicone Antonio Aldo
Emiliozzi Vincenzo
Fermani Giovanni
Fortunato Tullio
189

List of Personnel
Lupini Sergio
Maffia Giuseppe
Mirizzi Francesco
Neri Carlo
Papalini Massimo
Papitto Paolo
Pollastrone Fabio
Pretolini Piero
Ravera Gian Luca
Riva Marco
Roccon Mario
Santinelli Maurizio
Sodani Franca
Starace Fabio
Zampelli Pietro
Zannelli Luigi
TECHNICAL AND
ADMINISTRATIVE SUPPORT
Pecorella Francesco
Miglietta Flavio
Associations and Contracts
Manganiello Nicola
Carocci Franco
Ciavarella Angela,
Foschini Simona
Fraboni Gianpiero
Genangeli Emilia
Misano Guglielma
Perez Laura
Spignese Giulia
Vinciguerra Franca
Zaccardi Mario
Technical Support and Scientific
Publications
Leprai Filippo
Antonelli Laura
De Santis Maria Grazia
Di Natale Lucia
Venettoni Patrizia
MAGNETIC CONFINEMENT
FUSION PHYSICS
Romanelli Francesco (Deputy Director)
Barbato Emilia Orsitto (Assistant Deputy Director)
Nardone Daniela
FTU Operation Coordinator
Crisanti Flavio
FTU Data Analysis
Buratti Paolo
Bracco Giovanni
Low-Aspect-Ratio Tokamak Project
Alladio Franco
Mancuso Alessandro
Micozzi Paolo
Pieroni Leonardo
Experimental Physics and Tokamak
Operation Section
Tuccillo Angelo Antonio
Cantarini Luciano
Cesario Roberto
Colombi Salvatore
Conti Bruno
De Benedetti Massimo
Frigione Domenico
Grosso Luigi Andrea
Panaccione Luigi
Podda Salvatore
Rocchi Giuliano
Smeulders Paul
Tudisco Onofrio
Zerbini Marco
Experimental Physics and Radiation
Diagnostics Section
Zanza Vincenzo
Apruzzese Gerarda Maria
Castaldo Carmine
De Angelis Riccardo
Gabellieri Lori
Gatti Gerardo
Giovannozzi Edmondo
Gormezzano Claude
Grolli Mario
Kroegler Horst 1
Leigheb Mario
Mantovani Sergio
Monari Giancarlo
Orsitto Francesco
Pacella Danilo
Pericoli-Ridolfini Vincenzo
Pizzicaroli Giuseppe
Sibio Alessandro
Tartoni Nicola
Tilia Benedetto
Theoretical Physics Section
Zonca Fulvio
Briguglio Sergio
Cardinali Alessandro
Fogaccia Giuliana
Marinucci Massimo
Vlad Gregorio
190

List of Personnel
FTU Machine Units
Ciattaglia Sergio
Mazzitelli Giuseppe
Angelini Bianca Maria
Apicella Maria Laura
Bertocchi Alfredo
Bozzolan Walter
Buceti Giuliano
Brunetti Alessandra
Cefali Paolo
Centioli Cristina
Cesarini Bruno
Chuilon Pierre 1
Ciaffi Massimiliano
Cocilovo Valter
Gravanti Filippo
Iannone Francesco
Mazza Giulio
Mori Maria Luisa
Pambianchi Lamberto
Panella Maurizio
Sternini Enrico
Torelli Canzio
Tulli Rosario
Vitale Vincenzo
Zannetti Danilo
FUSION TECHNOLOGY
Pizzuto Aldo Maria (Deputy Director)
Cucchiaro Antonio (Assistant Deputy Director)
Giovagnoli Laura
Melorio Catia
Sansovini Maria Laura
Demo Technologies Project
Riccardi Bruno
Electromagnetic Computations
Roccella Massimo
Lattanzi Daniele
Safety and Environmental
Pizzuto Aldo Maria (acting)
Di Pace Luigi
Pinna Tonio
Porfiri M Teresa
Mechanical Engineering Section
Semeraro Luigi
Angelone Giuseppe
Baldarelli Massimo
Brolatti Giorgio
Capobianchi Mario
Celentano Giulio
Crescenzi Claudio
De Vellis Attilio
Ferrari Marco
Iorizzo Angelgiorgio
Lo Bue Alessandro
Macklin Brian 1
Marcelli Michele Antonio
Marra Antonio
Massimi Alberto
Mazzone Giuseppe
Moriani Andrea
Nardi Claudio
Nuvoli Marcello
Papastergiou Stamos 1
Polinari Paolo
Neutronics Section
Batistoni Paola
Angelone Maurizio
Bertalot Luciano
Borelli Rodolfo
Esposito Basilio
Pagano Guglielmo
Pensa Aldo
Petrizzi Luigino
Pillon Mario
Rapisarda Massimo
Spatafora Giuseppe
Advanced Fusion Technologies Section
Violante Vittorio
Alessandrini Carlo
Bettinali Livio
Borelli Rodolfo
Burul Mauro
De Ninno Antonella
Di Pietro Enrico
Frattolillo Antonio
Giacomi Giuliano
Galifi Giuseppe
Gallina Mauro
Lecci Domenico
Libera Stefano
Maddaluno Giorgio
Marini Fabrizio
Martinis Lorenzo
Mori Luciano
Orsini Aldo
Rizzo Antonietta
Sacchetti Marcello
Sorgi Mauro
Tosti Silvano
Tripodi Paolo
Verdini Luigi
Veschetti Mirian
Visca Eliseo
Superconductivity Section
Spadoni Maurizio
Annino Carmela
Catitti Aldo
191

List of Personnel
Celentano Giuseppe
Chiarelli Sandro
Ciotti Marco
Cristofori Pietro
Della Corte Antonio
Di Ferdinando Enzo
Giammatteo Fabio
Gislon Paola
Mazza Luciano
Moroni Manrico
Novelli Alessio
Pioli Fabrizi
Ricci Mario
Rufoloni Alessandro
EXPERIMENTAL ENGINEERING
BRASIMONE
Beneamati Gianluca (Deputy Director)
Rapezzi Luigi (Assistant Deputy Director)
Rossi Elia (Assistant Deputy Director)
Aiello Antonio
Fazio Concetta
Groppalli Carla Luisa
Masinara Annamaria
Miosotidi Emanuela
Miosotidi Silvana
Thermofluidodynamics Project
Dell’Orco Giovanni
Polazzi Giuseppe
Remote Handling Project
Damiani Carlo
Baldi Luciano
Irving Michael 1
Lorenzelli Luciano
Miccichè Gioacchino
Plasma Physics Applications Project
Rapezzi Luigi (acting)
Nivazzi Tiziano
Sammarco Gianfranco
Advanced Materials Project
Benamati Gianluca
Agostini Massimo
Degli Esposti Luciano
Guccini Massimo Andrea
Rapezzi Luca
Ricapito Italo
Conventional Safety Service
Cucumazzi Romolo (acting)
Barbi Bruno
Beccaglia Sergio
Civerra Aldo
Gamberini Sergio
Giardini Paolo
Marinaci Silvio
Vitali Silvano
Vitamia Agostino
Experimental Plants Management
Section
Moreschi Luigi Filippo
Alessandrini Italo
Armeni Maurizio
Arrighi Carlo
Barbi Angelo
Bichicchi Amabilio
Bichicchi Renzo
Caliolo Antonio
Canneta Angelo
Collina Giuseppe
Collina Graziano
Desideri Fabrizio
Fasano Giuseppe
Fogacci Guido
Frascati Fabrizio
Gamberini Sergi
Giacomelli Silvano
Guzzini Enzo
Lamma Giuseppe
Lepri Giovanni
Malavasi Andrea
Muro Luigi
Nivazzi Tiziano
Nucci Sergio
Panichi Enrico
Pazzaglia Giorgio
Pazzaglia Sergio
Pazzaglia Ugo 51
Pierucci Giampiero
Querci Messero
Rapezzi Danilo
Rapezzi Emilio
Romagnoli Giuseppe
Sacchetti Jader
Salvi Federico
Sansone Lorenzo
Serra Massimo
Simoncini Massimiliano
Storai Sandro
Varocchi Giuseppe
Components Design Section - Bologna
Antonucci Carlo Maria
Colaiuda Antonio
Gaggini Pierantonio
Poli Maurizio
Sabioni Elia
192

List of Personnel
RFX Support - Padua
Monari Demetrio
Antonelli Angelo
Apolloni Loris
Bernardi Guglielmo
De Biagi Arturo
Lorenzini Rita
Cassarini Domenico
Ferretti Floriano
Campori Paola
De Roit Renata
Filotto Francesco
Gatti Gabriele
Giordano Salvatore
Panichi Saverio
Ruggeri Ivana
RESEARCH CENTER -
BRASIMONE
Prevention and Protection
Cucumazzi Romolo
Fabbri Claudio
Martinelli Roberto
Administrative Support
Cassarini Domenico (acting)
Campori Fiorella
Fabbri Stefano
Lancetti Sonia
Morganti Maria
Nuzzi Emanuela
Pazzaglia Alessandra
Pazzaglia Bruno
Salvi Anna
Tonelli Elisabetta
Tonelli Graziella
Tonelli Luciano
Technical Support
Corvalli Giordano
Agostini Roberto
Aldrovandi Mara
Ballerini Graziano
Barbi Giuliano
Benassi Gisberto
Benassi Marisa
Brunetti Silvano
Carpani Bruno
Cristalli Luigi
Fabbri Ludovico
Fabbri Moreno
Giannerini Mario
Giorgi Franco
Gomedi Mauro
Muzzarelli Colomba
Nerattini Giuseppe
Nucci Germano
Panichi Riccardo
Rapezzi Giuseppe
Taulli Vincenzo
Varocchi Liliana
Vitali Guido
Zagnoli Antonio
193

Abreviations and acronyms

Abreviations and acronyms
ac
ACP
ADPAK
ADS
AES
AGILE
AHPCS
AHG
AITG
AM
APS
ASDEX-U
ASI
alternating current
activated corrosion product
Atomic Data Package
accelerator-driven subcritical systems
Auger electron spectroscopy
Astrorivelatore Gamma ad Immagini LEggero
Allyhidropolycarbosylane
Ad-hoc group
Alfvén ion-temperature gradient
amplitude modulated
air plasma spraying
Asdex Upgrade - Germany
Agenzia Spaziale Italiana
BAE
BB
BLK
BP
BPSX
BS
BTS
B-induced Alfvén eigenmode
breeding blanket
blanket
Burning plasma
Burning plasma science experiment
Biological shield
Bore tooling system
CB
CBR
CCC
CCHEN
CD
CD
CDA
CDE
C/E
CEA
CERN
CFC
CFK
CICC
CMCS
CMM
CMR
CNR
CNR-TEMPE
cassette body
cosmic background radiation
central cassette carrier
Comisiòn Chilena de Energia Nuclear - Chile
centered disruption
current drive
Conceptual Design Activity
Conceptual Design Evaluation
Computed/Experimental
Commissariat à l’Energie Atomique - France
European Organization for Nuclear Research - Geneva
carbon fibre composite
Chandrasekhar-Kendall-Furth
cable in conduit conductor
CERN Montecarlo subcritical code
cantilever multifunctional mover
catalytic membrane reactor
Consiglio Nazionale delle Ricerche - Italy
Istituto per la Tecnologia dei Materiali e dei Processi Energetici - CNR - Milan
197

Abreviations and acronyms
CP
cr
CR
CRNL
CRX
CS
CSMC
CTM
CV
CVD
CVI
CWIE
CWM
cooling plate
collisional radiative
cryopump room
Chalk River National Laboratory - Canada
crystal x-ray spectrometer
central solenoid
central solenoid model coil
cassette toroidal mover concept
centre void
chemical vapor deposition
chemical vapor infiltration
cutting/welding/inspecting equipment
cumulative weakening model
DA
DAP
DBTT
dc
D-D
DDD
DE
DOE
DOF
dpa
DRC
DRFC
DRP
D-T
DT
DTE
DTP
DV/LIM PHTS
DVT
divertor assembly
dummy armor prototype
ductile-to-brittle transition temperature
direct current
deuterium-deuterium
design description document (ITER)
destructive examination
Department of Energy - U.S.A.
Degrees of freedom
displacement per atom
direct radiative capture
Département de Recherches sur le Fusion Controlée
Divertor Refurbishment Platform - ENEA - Brasimone
deuterium-tritium
dump target
Deuterium-Tritium Experiment
Divertor Test Platform - ENEA - Brasimone
Divertor/Limiter Primary Heat Transfer System
divertor vertical target
EA
EAC
EAF
EASY
EBP
EBW
EC
energy amplifier
environmentally assisted cracking
European Activation File
European Activation Code System
European blanket project
electron beam welding
electron cyclotron
198

Abreviations and acronyms
ECCD
ECE
ECH
ECN
ECRH
EDA
EDS
EDX
EFDA
EFET
EFF
EFTP
ELM
EM
em
EMAS
EOL
EOS
EPFL
EPM
EPP
EST
ET
ETL
EU
EU-HT
EUR
electron cyclotron current drive
electron cyclotron emission
electron cyclotron heating
Energy Research Foundation - Petten - The Netherlands
electron cyclotron resonance heating
Engineering Design Activities
energy dispersion spectroscopy
energy dispersion x-ray
European Fusion Development Agreement
European Fusion Engineering & Technology
European Fusion File
European Fusion Technology Programme
Edge localized modes
electromagnetic model
electromagnetic
electromagnetic analysis system
end of life
equation-of-state
École Polytechnique Fédérale de Lausanne - CH
energetic particle mode
Enhanced Performance Phase
environment source term
event tree
equatorial toroidal limiter
European Union
European Home Team
European Utility Requirements
FDR
FEAT
FEM
FFMEA
FIMEC
FLR
FMEA
FN
FNF
FNG
FNS
FOW
FPC
Final Design Report
Fusion Energy Advanced Tokamak
finite-element method
functional failure mode and effect analysis
flat-top indentor for mechanical characterization
finite Larmour radius
failure mode and effect analysis
Fabbricazione Nucleare - Italy
fast neutron facility
Frascati Neutron Generator - ENEA - Frascati
Fusion Neutronics Source - JAERI - Japan
finite drift-orbit width
Fusion Programme Committee
199

Abreviations and acronyms
FPSS
FRDF
FSP
FSR
FTS
FTU
FW
FWHM
FZJ
FZK
fusion power shut-down system
failure rate database
finite-size particle
first stability region
Fourier transform spectrometer
Frascati Tokamak Upgrade - ENEA - Frascati
first-wall
full width at half maximum
Forschungszeuntrum - Jülich - Germany
Forschungszeuntrum - Karlsruhe - Germany
GBL
GDRD
GLS
GRBM
GSSR
gas box liner
General Design Requirement Document
Global load sharing
gamma-ray burst monitor
generic-site specific safety report
HC
HCPB
HE
HELICA
HFR
HFTM
HHFC
HIDIF
HIP
HIP
HMGC
HPF
HRS
HT
HTc
HTS
HTS
HWHM
HXR
halo current
helium-cooled pebble bed
hydrogen embrittlement
HE-FUS3 Lithium Cassette
high-flux reactor
high-flux test module
high heat flux component
heavy-ion-driven inertial fusion
heavy ion pulse
hot isostatic pressing
hybrid MHD gyrokinetic code
high performant Fortran
heat rejection system
home team
high critical temperature
heat transfer system
high-temperature superconductor
half width at half maximum
hard x-ray
IAS
IBB
IBW
ICE
Istituto Astrofisica Spaziale - CNR
inboard baffle
ion Bernstein wave
inlet coolant events
200

Abreviations and acronyms
ICE/LOVA
ICF
ICRF
ICRH
ICS
id
IDICE
IEA
IFE
IFMIF
IFP
IHTS
INFM
INFN
INTRA
IP
IR
ISAS
IT
ITB
ITER
ITG
ITM
IVROS
ingress of coolant event/loss of vacuum accident
inertial confinement fusion
ion cyclotron resonance frequency
ion cyclotron resonance heating
ITER-coil structure
inner diameter
ITER Divertor Cassette Experiment
International Energy Agency
inertial fusion energy
International Fusion Materials Irradiation Facility
Istituto di Fisica del Plasma - CNR
intermediate heat transfer system
Unità di Ricerca, Department of Physics, University of Salerno - Italy
Istituto Nazionale di Fisica Nucleare - Italy
In Vessel TRansient analysis
inter-pancake
infrared
Integrated Safety Analysis Code System
inter-turn
internal transport barrier
International Thermonuclear Experimental Reactor
ion-temperature gradient
ITER test module
in vessel remote operating system
JAERI
JCT
JET
JHU
JRC
Japan Atomic Energy Research Institute - Japan
Joint Central Team
Joint European Torus - Culham - U.K.
Johns Hopkins University - Maryland - U.S.A.
Joint Research Centre - Ispra - Italy
KAW
KBM
KFA IPP
KEP
kinetic Alfvén wave
kinetic ballooning mode
Forschungszentrum - Jülich - Germany
Key Element Technology Development Phase
LBO
LCF
LCMS
LCT
LH
laser blow-off
low-cycle fatigue
last closed magnetic surface
large coil task
lower hybrid
201

Abreviations and acronyms
LHC
LHCD
LHP
LHW
LIM
LIVVS
LLNL
LOCA
LOFA
LOVA
LSTAE
LTL
LVDT
LULI
large hadran collider
lower hybrid current drive
lower hybrid power
lower hybrid wave
limiter
laser in-vessel viewing system
Lawrence Livermore National Laboratory - California - U.S.A.
loss-of-coolant accident
loss-of-flow accident
loss-of-vacuum accident
low shear toroidal Alfvén eigenmode
lower tokamak limiter
linear voltage displacement transducer
Laboratoire pour l’Utilisation des Lasers Intenses - National Laser Facility - France
MAGS
MANET
M&M
MHD
MI
MJ
ML
MLD
MLEM
MOR
MOU
MSE
MSGC
MTTR
magnetic system
martensitic NET
Multi-conductor Mithraudir
magnetohydrodynamic
mobile inventory
multijunction
multilink
master logic diagram
method and the maximum likelihood estimation method
modulus of rapture
matching optics unit
motional Stark effect
microstrip gas chamber
mean time to repair
NBI
NDT
NEG
NFPP
NIM
NL
NPI
N/S
neutral beam injection
nondestructive testing
non-evaporable getter
nuclear fission power plant
Non-interactive model
Non-inear
Nuclear Physics Institute - Rez near Prague Czech. Republic
not safety
OBB
od
outboard baffle
outer diameter
202

Abreviations and acronyms
ODE
ODS
OPTIFER
ORE
ORNL
ordinary differential equation
oxide dispersion strengthened
optimized ferritic
occupational radiation exposure
Oak Ridge National Laboratory - Tennessee - U.S.A.
PAM
PBX-M
PCS
PEM
PF
PFC
PGH
PHTS
PIC
PIE
PIP
PLC
PLD
POLITO
PPCS
PPPL
PRF
PRM
PSA
PST
PVD
PWR
passive-active multijunction
Beta Experiment-Modified at PPPL
polycarbosylane
photo elastic modulator
packing factors
plasma-facing component
purge gas distributor/collector
primary heat transfer system
particle-in-cell
postulated initiating event
polymer impregnation and pyrolysis
programmable logic controller
pulsed-laser deposition
Politecnico di Torino
power plant conceptual studies
Princeton Plasma Physics Laboratory - New Jersey - U.S.A.
permeation reduction factor
pressure rise method
plant safety assessment
process source term
physical vapor deposition
pressurized water reactor
QMS
QOG
quadrupole mass spectrometer
quasi-optical grill
RAF
RAM
RF
RH
RHE
RR
rr
RT
RTM
reduced-activation ferritic (steels)
reduced-activation martensitic (steels)
radiofrequency
remote handling
remote handling external
radiative recombination
repetition rate
room temperature
Istituto per le Ricerche di Tecnologia Meccanica e per l’Automazione S.p.A. - Turin - Italy
203

Abreviations and acronyms
SANS
SB
SC
SCADA
SCK/CEN
SEAFP
SEM
SERF
SM
SMART
SMP
SMPS
SNR
SOL
SP
SPHERA
SPND
SS
SSR
ST
START
small-angle neutron scattering
shielding blanket
Starfire Company - US
supervisory control and data acquisition system
Nuclear Research Centre - Mol - Belgium
Safety and Environmental Assessment of Fusion Power
scanning electron microscopy
socio economics research for fusion
superconducting magnet
small rectangular test sections
small missions programme
shared memory multiprocessor systems
signal-to-noise
scrape-off layer
screw pinch
Spherical Plasma for Helicity Relaxation Assessment
self-powered neutron detectors
stailess steel
second stability region
Spherical torus
Small Tight Aspect Ratio Tokamak - Culham - U.K.
TAE
TBR
TCV
TCWS
TEKES
TF
TFC
TFMC
TFTR
TIG
TMF
Tore-Supra
TOSKA
T-P
TPB
TPD/TPR
TRC
TS
TWCS
toroidal Alfvén eigenmode
tritium breeding ratio
Tokamak Condition Variable - Lausanne - CH
tokamak cooling water system
Technology Center - Finland
toroidal field
Test facilities complex
toroidal field model coil
Tokamak Fusion Test Reactor -PPPL - U.S.A.
tungsten inert gas (welding)
thermomechanical fatigue
Tokamak at Cadarache - France
Toroidal Spulenanlage Karlsurhe
Tie-Plates
tritium permeation barrier
temperature-programmed/desorption/reduction
toroidal radial Carrier
thermal shield
tokamak water cooling system
204

Abreviations and acronyms
TWO
TUD
TZM
Technology work programme
Technical University of Dresden
tungsten-zirconium-molybdenum
UCT
UDM
UFA
UHV
UKAEA
ULART
US
UTL
UTP
Un-axial compressive tests
Un-axial deformation module
University Fusion Association Group
ultrahigh vacuum
United Kingdom Atomic Energy Agency
Ultralow Aspect Ratio Torus
ultrasonic
upper toroidal limiter
Underlying Technology Program
VDE
VPS
VT
VTA
VUV
VV
vertical displacement event
vacuum plasma spraying
vertical target
vertical target assembly
vacuum ultraviolet
vacuum vessel
WCLL
WGSR
WI
WKB
XEXCALIBER
XRD
water-cooled lithium lead
water-gas shift reactor
wing
Wenzel, Kramer, Brillouin code
experimental cassette of lithium beryllium pepple beds
x-ray diffraction
205