Tokamak edge physics and plasma- surface interactionspitts/pitts/pitts_varenna_27_09_2007.pdf ·...

Tokamak edge physics and plasma-surface interactions

Richard A. Pitts

Centre de Recherches en Physique des PlasmasEcole Polytechnique Fédérale de Lausanne, Switzerland

Association EURATOM-Swiss Confederation

thanks toAndre Kukuskin, Philip Andrew, ITER Organisation

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/20072 of 32

“Mission statement” for this talk …

“The interaction of plasma with first wall surfaces will have a considerable impact on the performance of fusion plasmas, the lifetime of plasma-facing components and the retention of tritium in next step burning plasma experiments”

Progress in the ITER Physics Basis, Chap. 4: “Power and particle control”, Nucl. Fusion 47 (2007) S203-S263

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/20073 of 32

Outline• The scrape-off layer (SOL) and divertor

− SOL power width

− Divertor detachment• Plasma-surface interactions

− Material lifetime – erosion and migration− Tritium retention− Dust

− Mixed materials• What to diagnose ….

CAVEAT: Edge plasma physics and PSI is a vast domain. Can only scratch the surface in this talk. Work referenced throughout the talk is listed at the end.

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/20074 of 32

Divertor and SOL physics

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/20075 of 32

Terminology: limiters and divertors

Core plasma

Core plasma

Scrape-off layer (SOL) plasma: region of open field lines

Divertor targets

Limiter

Vessel walls

Private plasma

Separatrix

LCFS

X-point

“Upstream”

Outermidplane

CL

OuterInner

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/20076 of 32

•JET #62218, t = 3.0 s

Limited

Ramp-up and ramp-down phases in ITER will be in limited phase, ~30 s long [5]. Full burn divertor phase of ~400 s for the QDT = 10 inductive scenario

Diverted

JET #62218, t = 15.2 s

Limiter and divertor phases in most JET shots

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/20077 of 32

Basics – SOL width, λλλλn [1]• Any solid surface inserted into

a plasma constitutes a very strong particle sink

• In the high tokamak B-field:Γ⊥ << Γ||

• Thin Debye sheath (λD few 10’s µm thick ) forms at the surface � controls flow of particles and energy || B

e.g. L ~ 30 m, TLCFS ~ 100 eV, cs ~ 105 ms-1, D⊥~ 1 m2s-1 (near SOL)� λn~ 1.7 cm!!cf. a = 2.0 m for ITEREven worse for energy – see next ……

• Quick and dirty estimate of λn with diffusive approx. for cross-field particle transport (all ionisation inside LCFS):Γ⊥ ≡ nv⊥ = -D⊥dn/dr ~ D⊥n/ λn

� v⊥ ≈ D⊥/λn , λn= τ⊥v⊥ �

v|| ≈ cs ~ (kT/mi)1/2 �

Then, if τ⊥ = τ||,

⊥⊥ = Dn /2λτ

scL /|| =τ

( ) 2/1/ sn cLD⊥=λ

λq, λT, λn

WALL

Limiter or

divertorplate

SOL

Main plasmaB

2L“Connection length”

“upstream”

Adapted from [1]

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/20078 of 32

The problem with λλλλq• SOL width for power, λq, is also small and is an important parameter of the edge plasma• As for particles, λq is determined by the ratio of ⊥ to || transport (e.g. cross-field ion

conduction and parallel electron conduction: ie ∝ (χ⊥/χ||)1/2 ), where χ⊥ is anomalous• Scalings for λq can be derived from models and experiments, e.g.:

− “2-point” analytic modelling: PSOL = power into SOL [1]

− Scaling from H-mode experiments on JET [6]:− ITER modelling [7] assumes λq = 5 mm, JET scaling gives λq = 3.7 mm (cf. a=2.0 m)− Very recent multi-machine scaling [8] gives λq/R ~ constant

• Note also that the parallel power flux, q|| ∝ PSOL/λq ~ as much as 1 GWm-2 in ITER

9/5−∝ SOLq Pλ15.04.0

959.05.0

uSOLq nqBP −−∝ ϕλ

Stored energy scales strongly with tokamak major radius, W ~ ∝ R4 [9]But power deposition area in the divertor ∝ Rλq only (~6 m2 in ITER)

Bottom line is that despite its increased physical size, ITER will concentrate more power into a narrower channel at the plasma edge than today’s devices. The use of divertor detachment, radiation and geometry will be used to reduce the surface power flux densities to manageable levels, but careful monitoring will be critical � see talk by Albrecht Herrmann.

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/20079 of 32

Power handling – ITER case (approx)

• Max. steady-state power flux density permitted at ITER divertor targets: q⊥ ≤ 10 MWm-2

• Magnetic and divertorgeometry alone cannot reduce the power to tolerable levels

• Most of the parallel power flux must be prevented from reaching the plates� divertor detachment and high radiative loss

SOL

CORE PLASMA

(adapted from [10])

~100 MW

22.0~)/(4~ mBBRArea uq θλπq||,u ~ 500 MWm-2

α

Magnetic flux expansion ~(Bθ/B)u/(Bθ/B)t ~4 for ITER outer divertor � low field line angles at strike points (~3º)

+Target tilting in poloidal plane (α ~ 25º for ITER outer target)

λq = 5 mm

20.3~)/)(sin(/)/(4~ mBBBBRArea tuq θθ αλπ per target

q⊥ ~ 16 MWm-2

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200710 of 32

ν* rises as nu rises, finite electron heat conductivity:

(note: κ0,e » κ0,i)allows parallel T gradients to develop � Tt

decreases, but pressure balance maintained (∇p|| ~ 0) so that nt rises strongly ( )λion (∝ 1/nt) decreases so that target recycling increases strongly � flux amplificationAs Tt ↓, radiation loss increases � Tt ↓ further

2/50||||||||, ,/ TKdsdTKq cond κ=−=

2ut n∝Γ

The route to detachment (1)Mean free paths for particle collisions are long: SOL collisionality: is low Power flow to surface largely controlled by target sheath: γ = sheath heat transmission coefficientεpot = potential energy per incident ion

iieieeieuuucoll TTTnT λλλλ ~~,~~,/2∝

collL λν /* =

potstttsttt cnTcnq εγ +=||,

Target

T

n

Lu t

Low n, high T (high P SOL)“Sheath limited”

ut TT ∝

2/ut nn ∝

Target

T

n

Lu t

Moderate n, T“High recycling”

3ut nn ∝

2/1 ut nT ∝Region of strong radiation losses

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200711 of 32

The route to detachment (2)At sufficiently low Tt, (< 5 eV), neutral ionisation rate < ion-neutral friction processes (CX, elastic scattering). Momentum transferred from ions to dense cloud of neutrals in front of the plate (recycle region) � begins to reduce nt, ∇p|| ≠ 0 and plasma pressure falls across recycle region.Once Tt ~1-2 eV (and if nt high enough), volume recombination locally “extinguishes”plasma, reducing target power flux

Detachment seen experimentally in many devices, but complex “volumetric” process and relative importance of ion-momentum friction vs. recombination still unclear. X-point geometry � long connection lengths � high residence times in low Teplasma � efficient radiative loss favouringpower reductions where q|| is highest (i.e. on flux surfaces near separatrix).

Target

T

n

Lu t

High n “Detached”

Recycle region

C-Mod , B. Labombard, et al., [11]

Adapted from [2]

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200712 of 32

Full detachment is a problem

JET, A. Huber, et al. [12]

• Detachment which is too “strong” (particle flux reduced across the whole target) is often associated with zones of high radiation in the X-point region and confined plasma (MARFE)

• MARFE formation can drive a transition from H to L-mode (H-mode density limit) or disruption

• MARFE physics still not well modelled

Limit detachment to regions of highest power flux (where it is needed most).Maintain remainder of SOL in high recycling (attached)A few ways to arrange that this happens more readily:

Divertor closure Target orientation Impurity seeding

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200713 of 32

Divertor closure

• Increased closure significantly improves divertor neutral pressure � increased neutral density (nn), promoting earlier detachment

• Closing “bypass” leaks important for increasing nn

• Divertor closure also promotes helium compression and exhaust – very important for ITER and reactors

JET, R. D. Monk, et al. [13]

Increasing closure

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200714 of 32

Target orientation• Parallel heat fluxes

significantly reduced for vertical cf. horizontal targets

• Underlying effect is preferential reflection of recycled deuterium neutrals towards the separatrix

Hotter plasma near separatrix

Increased ionisation near sep.

Higher nt, lower Tt

Higher CX losses

AUG, A. Kallenbach, et al. [14]

Cooler, less dense plasma

Separatrix

Pressure loss, q|| ↓

Separatrix

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200715 of 32

Impurity seedingDIII-D, C. J. Lasnier, et al. [15]

JET, G. F. Matthews et al. [16]

D2 puff92 torrls -1 for 1.8 s

Ne puff12 torrls -1 for 0.1 s

Unfuelled

Strong D 2 puff

Strong D 2 +N2puff

Strong impurity seeding reduces ELM size but price is paid in confinement

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200716 of 32

ITER divertor achieves partial detachment

ITER Divertor DDD 17, Case 489 (SOLPS5 runs by A. Kukushkin)

Deep V-shaped divertor, vertical, inclined targetsDome separating inner and outer targets – also helpful for diagnostics, neutron shielding and reducing neutral reflux to the core

Inner strike pt.

Kirschner et al. [17]

Pow

er lo

ad (

Wm

-2)

Outer strike pt.

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200717 of 32

Divertor exhaust

Apart from power handling, primary function of divertor is to deal with He from fusion reactions � compress D, T, and He exhaust as much as possible for efficient pumping (and therefore also good density control).

To cryopumps

Critical criterion for an ITER burning plasma is that He is removed fast enough such that: is satisfied.

is the global helium particle residence time – a function of τp, the He neutral density in the divertor and the pumping speed (conductance) [18].Helium enrichment:is the ratio of He concentration in the divertor compared to the main plasma.

105/*, −≤EHep ττ

plasma

pump

eplasmaHe

pumpD

pumpHe

He C

C

nn

nn ==/

2/ 2η

*,Hepτ e.g. ITER: He prod. rate ~2×1020s-1

Max. divertor pumping speed ~200 Pa m3s-1 ~ 1×1023 He atom s-1

� Cpump ~ 2×10-3 = 0.2%Typical acceptable He conc. in the core: ~4% � ηHe = 0.2/4 = 0.05 is minimum required. The values of

and required for ITER have been achieved experimentally

*,HepτHeη

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200718 of 32

Plasma-surface interaction

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200719 of 32

ITER materials choices

• Be for the first wall−Low T-retention−Low Z−Good oxygen getter

Driven by the need for operational flexibility

• C for the targets−Low Z−Does not melt−Excellent radiator

• W for the dome/baffles−High physical sputtering

threshold

• Possible alternative:−Be wall, all-W divertor

To avoid problem of T-retention

W

CFCWhat are the issues associated with plasma-surface interactions?

Beryll ium

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200720 of 32

Critical issues

Steady state erosion

Long term tritium retention

Material lifetime

Transient erosion(ELMs, disruptions)

Material mixing

Short and long range material migration

Redeposition

All strongly interlinked

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200721 of 32

Impurity migration

TransportErosion DepositionRe-erosion

=Migration

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200722 of 32

Chemical (carbon)

• Energy threshold � higher for higher Z substrate

• Much higher yields for high Z projectiles – important if using impurity seed gases

• No threshold• Dependent on bombarding energy,

flux and surface temperature

ITER divertor fluxD impact

Erosion: Physical and chemical sputteringPhysical

Adapted from Eckstein et al. [20] Roth et al. [21]

Current steady state divertor target erosion rates (ERO modelling) due to Yphys and Ychemestimated at ~0.4 - 2 nms-1 for ITER [17]

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200723 of 32

Erosion: transients, e.g. ELMs on the divertorImportant factor is max. ∆Tsurf due to arrival of short heat pulse (duration, t):

f = fraction of Wth lost during transientAdiv, divertor wetted area (~6m2)ρ,k,C=density,conductivity,heat capacity ELM energy losses must stay below melting/sublimation/evaporation limit to avoid fast erosion (e.g. melt later loss)

2/1

max 2

≈Ck

t

A

fWT

div

thsurf πρ

Time (µs)

� Important to measure Tmax

Federici et al. [22]

If ∆transient ~ few µm and target thickness ~ cm � lifetime ~104 events~103 Type I ELMs/discharge �lifetime ~ 10 ITER pulses!!

Tests on ITER target mock-ups with realistic energy fluxes show that damage threshold ~2x lower than for ideal materials (crack formation) [23,24]� ELM energy flux ≤ 0.7 MJm-2 for W and CFC (~1.5% of Wth @ QDT = 10)

This is the very lower limit for Type I ELMs observed today � need to mitigate ELMs or find small ELM regimes and provide best possible monitoring of target erosion � see talk by E. Gauthier (Thurs. morning)NB: plasma reattaches during ELMsand strong inherent ELM variability!!

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200724 of 32

Ions:Cross-field transport – turbulent driven ion fluxes can extend into far SOL� recycled neutrals� direct impurity releaseELMs can also reach first walls

Eroded Impurity ions “leak” out of the divertor (∇Ti forces)

SOL and divertor ion fluid flows can entrain impurities

Neutrals:• From divertor plasma leakage, gas

puffs, bypass leaks � low energy CX fluxes � wall sputtering

• Lower fluxes of energetic D0 from deeper in the core plasma

• A problem for first mirrors � see talk by Vladimir Voitsenya (Thurs. morning)

Transport creates and moves impurities

EDGE2D/NIMBUS

Bypass leaks

Escape via divertorplasma

Ionisation

D0 from wall ion flux or gas puff

CX event

Courtesy G.F. Matthews

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200725 of 32

20g Be (BeII)

450g C (CIII)

~250 kg/year if JET operated full time!Carbon migrates to remote locations forming D-rich soft layers (high T-retention)

Migration balance – example from JET• Make balance for period 1999-2001 with

MarkIIGB divertor: 14 hours plasma in diverted phase (50400 s, 5748 shots)

• Use spectroscopy and modelling to estimate main chamber sources

~400g C22g Be

Likonen et al. [26]Coad et al. [27]

Main chamber: source of

net erosion

• Post mortem surface analysis− Deposition almost all at inner divertor− Surface layers are Be rich � C chemically

eroded and migrates, Be stays put− Outer divertor – region of net erosion or

balanced erosion/redeposition – BUT mostly attached conditions (not like ITER)

Strachan et al. [25]

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200726 of 32

Tritium retention (1)• One of the most challenging operational

issues for burning plasmas

• If carbon present, complex interplay between erosion � hydrocarbons �dissociation/ionisation � transport �re-deposition � migration to remote areas with high sticking coefficients and retention in co-deposits

− Carbon traps D, T very efficiently− D/C ratio can be in the range ~0.4 � > 1

depending on the type of re-deposited layer

• Retention very hard to characterise in today’s mostly carbon dominated devices

• Dependent on materials, Tsurf, geometry (limiter/divertor), operating scenarios (H-mode, L-mode, low/high dens.)

Reported measurements range from 3-50% retention [28]!e.g. on JET, ~3% obtained from long term, post mortem surface analysis, ~10-20% from gas balance

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200727 of 32

Roth et al. [13]Kirschner et al. [30]

C targets, Tsurf = 800ºC, chemical+physical sputtering

~0.02g/discharge

~2g/discharge

Tritium retention (2)• A 400 s QDT = 10 ITER discharge

will require ~50 g of T fuelling(cf. 0.01-0.2 g in today’s tokamaks)

• Working guideline for max. in-vess. mobilisable T in ITER ~1kg [29,30]

• World supply of T is also limited− Must avoid build-up in inaccessible

locations • Predicting the expected retention in

ITER is notoriously difficult

• Very recent estimates (ERO code 2007 including Be main chamber influx) show that the in-vessel limit could be reached after only ~140 shots [17]

− Modelling does not yet contain effects of transients (ELMs disruptions)!

− No account taken for trapping in tile gaps

ITER target [29] is a retention level of ~0.05 g/discharge � ~7000 shots before major shutdown for T-removal

Accurate measurement of T-retention and the development of efficient T-removal methods will be critical for the success of ITER

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200728 of 32

Dust

Carbon dust collected from tokamaks after operation periods is usually micron sized. Formed from flaking and degradation of deposited films, unipolar arcing, brittle fracture (e.g. due to transients) etc.

• Dust is seen in all tokamaks, especially with C walls• Not generally a concern in today’s devices …• But is potentially very important in ITER � see talk

by Sandrine Rosanvallon (Wednesday morning)− As an inventory for trapped tritium in areas difficult to

access− As an explosive safety hasard – water leak � hot

surfaces � steam � hydrogen (by oxidation) �possible explosion if enough air also present [31]

− As a radiological or toxic hasard (activation products of W, tritium contained in Be, C dust, toxicity of Be dust)

TCV: floor viewing IR camera during disruption, #33448

No real idea yet how much dust ITER will generate, where it will go or how to get it out � a big effort needed to improve this situation

DIII-D: floor viewing DiMES TV with near IR filter. 2nd shot in 2007 after “dirty vent”, #127331. Courtesy of D. L. Rudakov & W. P. West [32]

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200729 of 32

Mixed MaterialsNo fusion device operating today contains the material mix currently planned for the ITER first wall and divertor: Be, W, C. Cross contamination of the material surfaces will be unavoidable. This is likely to have several consequences [29]:

Material property changes due to mixing

Effect on H-isotope retention

Effect on material erosion

• Formation of metallic carbides � diffusion of C into bulk material at high temperatures

• Formation of Be-W alloys � melting point can be reduced by as much as ~2000ºC

• Retention of H in BeO can be as high as in C• Retention in W can be increased by C or

oxide layers but is very low in pure W or Be• Very complex – difficult to predict yet for ITER

• Can both increase and decrease erosion!• Heavy ions (e.g. BeZ+, CZ+) on C, W �

increased phys. sputt. but surface coverage (e.g. Be on C) reduces chemical sputtering.

Preparations underway at JET to test a Be/W & Be/W/C wall mix from ~2010 [33]

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200730 of 32

So, what needs to be diagnosed [34]?

Target plate heat and particle fluxes, Te, surface temperature, erosion rate

Neutral gas pressure

Cryopump inlet composition (H/D/T/He, CxHy)

Te ne Prad

Dust accumulation?

ne, Te, Prad, position of ionisation front, nHe, nD/nT, impurity and D, T influxes

Wall temperature and visible imageMain chamber gas pressure and gas compositionSOL neutral density (D/T)Impurity influxesSOL ne, Te profiles (but challenging)

See next talk by Philip Andrew!

Courtesy of A

. Kukushkin

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200731 of 32

References (1)

[1] “The plasma boundary of magnetic fusion devices”, P. C. Stangeby, IoP Publishing Ltd, Bristol, 2000[2] “Experimental divertor physics”, C. S. Pitcher and P. C. Stangeby, Plasma Phys. Control. Fusion 39 (1997) 779[3] “Plasma-material interactions in current tokamaks and their implications for next step fusion reactors”, G. Federici et al.,

Nucl. Fusion 39 (1997) 79[4] “Material erosion and migration in tokamaks”, R. A. Pitts et al., Plasma Phys. Control. Fusion 47 (205) B303

A 45 min. talk can only hope to scratch the surface of such a vast field. Some good reference sources covering aspects of the material in this ta lk are the following:

A number of additional papers have been used to pre pare the slides in this presentation. They are listed below in order of appearance in the talk.

[5] “Simulations of ITER start-up and assessment of initial power loads”, G. Federici et al., J. Nucl. Mater. 363-365 (2007) 346

[6] “Boundary plasma energy transport in JET ELMy H-modes”, W. Fundamenski and W. Sailer, Nucl. Fusion 44 (2003) 20[7] “Scaling laws for edge plasma parameters in ITER from two-dimensional edge modelling”, A. Kukushkin et al.,

Nucl. Fusion 43 (2003) 716[8] “Plasma-surface interaction, scrape-off layer and divertor physics: implications for ITER”, B. Lipschultz et al.,

Nucl. Fusion 47 (2007) 1189[9] “Steady state and transient power handling in JET”, G. F. Matthews et al., Nucl. Fusion 43 (2003) 999[10] “The plasma-wall interaction region: a key low temperature plasma for controlled fusion”, G. F. Counsell, Plasma

Sources Sci. Technol. 11 (2002) A80[11] “Experimental investigation of transport phenomena in the scrape-off layer and divertor”, B. LaBombard et al.,

J. Nucl. Mater. 241-243 (1997) 149[12] “Improved radiation measurements on JET – first results from an upgraded bolometer system”, A. Huber et al.,

J. Nucl. Mater. 363-365 (2007) 365[13] “Recent results from divertor and scrape-off layer studies on JET”, R. D. Monk et al., Nucl. Fusion 39 (1999) 1751[14] “Scrape-off layer radiation and heat load to the ASDEX Upgrade LYRA divertor”, A. Kallenbach et al.,

Nucl. Fusion 39 (1999) 901

R. A. Pitts: Burning Plasma Diagnostics Workshop, Varenna, Italy, 24-28/09/200732 of 32

References (2)[15] “Study of target plate heat load in diverted DIII-D tokamak discharges”, C. Lasnier et al., Nucl. Fusion 38 (1998) 1225[16] “Studies in JET divertors of varied geometry: II Impurity seeded plasmas”, G. F. Matthews et al.,

Nucl. Fusion 39 (1999) 19[17] “Modelling of tritium retention and target lifetime of the ITER divertor using the ERO code”, A. Kirschner et al.,

J. Nucl. Mater. 363-365 (2007) 91[18] “ITER Physics basis: Chapter 4, power and particle control”, Nucl. Fusion 39 (1999) 2391[19] “Material migration in JET”, G. F. Matthews et al., Proc. 30th EPS Conf. on Control. Fusion and Plasma Physics (St.

Petersburg, 2003) 27A (ECA) P-3.198[20] W. Eckstein et al., IPP Garching report number 9/82 (1993)[21] “Flux dependence of carbon chemical erosion by deuterium ions”, J. Roth et al., Nucl. Fusion 44 (2004) L21[22] “Assessment of the erosion of the ITER divertor targets during Type I ELMs”, G. Federici et al.,

Plasma Phys. Control. Fusion 45 (2003) 1523[23] “Transient energy fluxes in tokamaks : Physical processes and consequences for next step devices”, A. Loarte et al.,

34th EPS Conf. on Control. Fusion and Plasma Physics (Warsaw, 2007)[24] “Effect of ELMs on ITER divertor armour materials”, A. Zhitlukhin et al., J. Nucl. Mater. 363-365 (2007) 301[25] “JET carbon screening experiments using methane gas puffing and its relation to intrinsic carbon impurities”,

J. D. Strachan et al., Nucl. Fusion 43 (2003) 922[26] “Beryllium accumulation at the inner divertor of JET”, J. Likonen et al., J. Nucl. Mater. 337-339 (2005) 60[28] “Gas balance and fuel retention in fusion devices”, T. Loarer et al., Nucl. Fusion 47 (2007) 1112[29] “Progress in the ITER Physics basis: Chapter 4: Power and particle control”, Nucl. Fusion 47 (2007) S203[30] “ITER Technical Basis”, ITER EDA Documentation Series No. 24 (Vienna: IAEA) 2002[31] “The safety implications of tokamak dust size and surface area”, K. A. McCarthy et al., Fus. Eng. Design 42 (1998) 45[32] “Observations of Dust in DIII-D Divertor and SOL”, D. L. Rudakov et al., 1st Workshop on “Dust in Fusion Plasmas”,

8-10 July 2007, Warsaw, Poland[33] “An ITER-like wall for JET”, J. Pamela et al., J. Nucl. Mater. 363-365 (2007) 1[34] “Progress in the ITER Physics basis: Chapter 7: Diagnostics”, Nucl. Fusion 47 (2007) S337