$3(;�)<�����7DVN�,,�

([SORUDWLRQ�RI�+LJK�3D\RII�/LTXLG�:DOOV�2YHUYLHZ

(From the Reference APEX FY2000 Technical Plan)

Task Leader:Neil Morley

SubTask Coordinators and Contributors:N. Ghoniem, K. Gulec, R. Kaita, M. Kotschenreuther,

K. McCarthy, B. Nelson, N. Morley, T. Rognlien, D. Ruzic,S. Smolentsev, R. Woolley, A. Ying, M. Youssef, L. Zakharov, S. Zinkle

APEX MeetingArgonne National Laboratory

May 10-12, 2000

WDVN�,,�VFRSH�DQG�DSSURDFK

Scope:

• exploring high payoff liquid wall concepts that increase the attractiveness offusion energy, with emphasis on understanding the key scientific issues

• both liquid metals and Flibe, and thin and thick liquid walls that have thepotential to improve the physics performance of plasma

• Other new APEX concepts that are advanced this year

Approach:

• development and application of much-needed, generic modeling tools for liquidwalls and plasma interaction with liquid walls

• initiation of experiments that address fundamental LW issues identified in lastyear’s effort that are key to the understanding of liquid wall phenomena

• utilization of tools and experimental data to advance the conceptualization ofvarious LW designs

WDVN�,,�VXEWDVNV

II.1 Exploration thick and thin liquid wall concepts

• Bulk Plasma-Liquid metal wall interactions – Kaita and Kotschenreuther• Plasma-liquid surface interactions with lithium (under ALPS/APEX PLSI) –

Rognlien• LM-MHD numerical tool development and analysis of LM wall designs –

Smolentsev (Morley)• LM experiments (also Task I) – Morley (Woolley)

II.2 Exploration of thick Flibe liquid wall concepts

• Model development and analysis of potential thick Flibe concepts – Ying(Smolentsev, Moir)

• Plasma-Liquid Surface Interactions (under ALPS/APEX PLSI) - Rognlien• Flibe simulant experiments in basic poloidal flow LW geometries – Gulec• Mechanical configuration issues and drawings – Nelson

Blue – hereRed – elsewhereBlack - nowhere

WDVN�,,�VXEWDVNV��FRQWLQXHG�

II.3 Exploration of Liquid Walls for non-tokamak confinement schemes

• Continuation of FRC work - Moir• RFP, Stellarator – Moir

II.4 Materials, safety, and nuclear analysis for high payoff Liquid walls

• Identifying compatible liquid-structure combinations and temperature andother operating limits for a variety of applications: flexible vs. rigid, hermeticvs. non-hermetic, etc. – Zinkle

• Preliminary assessment of erosion rates for various coolant/materialcombinations as a function of temperature and coolant velocity – Ghoniem

• Analysis of safety issues for liquid walls – McCarthy

• Nuclear analysis and activation {some overlap with Task III nuclear work} –Youssef

WDVN�,,���WDVN�$��FRPELQHG�SUHVHQWDWLRQ

Part 1: Liquid Wall Hydrodynamics and Heat Transfer

• Morley (UCLA): Overall Status of Task II Work (30 min.)• Smolentsev (UCLA): Recent k-e Calculations for CLIFF and FLYHI

Experiment (20 min.)• Gulec (UCLA): FliHy Experimental Plan and Facility Design (20 min.)• Zakharov (PPPL): Intense Lithium Streams in Tokamaks (10 min.)• Ruzic (U. Illinois): Macroscopic Liquid Metal Experiments at Illinois (20 min.)• Moir (LLNL): Heat Transfer Enhancement by Liquid Droplets (10 min.)• Kotschenreuther (UT): Soaker Hose Idea Exploration (10 min.)

Part 2: Liquid Metal Wall / Liquid Flow Bulk Interaction

• Kaita (PPPL): Tokamak Simulation Code Modeling with LM Walls (10 min.)• Kotschenreuther (UT): Resistive MHD (20 min.)• Zakharov (PPPL): Proposal on Lithium Wall Experiment on PBX-M Facility

(20 min.)

�7DVN�,,��+\GURG\QDPLFV�DQG�KHDW�WUDQVIHU

&RQFOXVLRQV�IURP�1RYHPEHU�PHHWLQJ�

Only thin Flibe was ready for more detailed engineering consideration because itdoes not interact heavily with the applied field – graduated to task III

Concepts like:• Top down flow CLiFF with LM• Magnetic Propulsion from inboard to outboard• Two-layer thick LM flow• Top down Thick Flibe flow with MHD breaking• Complex LM film, jet, droplet flow in divertor

needed development of additional modeling tools or experiments for explorationand validation

LM concepts needed a better understanding of plasma bulk (MHD) interaction

Task II was created as a consolidated effort to 1st) develop the tools and facilitiesand 2nd) apply them to the analysis of concepts

7DVN�,,��+\GURG\QDPLFV�DQG�KHDW�WUDQVIHU

0DLQ�IRFXV�WKLV�\HDU�

Development of generic laminar and turbulent MHD modeling tools withcapability to analyze

• complex fields and wall geometries• effect of applied electric currents• near surface turbulent heat transfer behavior

Development of generic experimental tests that provide• insight into behavior of liquid wall flows to improve design• quantification of critical liquid flow uncertainties• data for verification of modeling tools

Apply tools and data to analysis of existing concepts leading to improvedimplementation and increased credibility.

1HZ�/0�0+'�&RPSXWDWLRQDO�0RGHOV

3-component field, approximate solution in 2.5D (axisymmetric)

1-component field, full solution in 2D (axisymmetric)Allows:• toroidal field that varies arbitrarily in time and space• applied electric currents• flexible geometries, flow obstacles

Toroidal field, full solution in 2D with VOF free surface tracking (axisymmetric)

Allows:• 3 field components that vary in main flow

direction and in time• applied electric currents• flexible geometries

Does not allow:• Strongly inductive effects• Rapid variation in flow parameters

(droplets, splashing, reverse flow, etc.)• Non-axisymmetric flows (behavior near

dividing walls or penetrations)

Allows:• Arbitrary temporal and spatial toroidal

field variations• applied electric currents• flexible geometries, internal flow

obstacles• Splashing, wave breaking, droplet

formation

Does not allow:• Multi-field components effects• Flow in toroidal direction• Non-axisymmetric flows (behavior near

dividing walls or penetrations)

180V/3600A Power Supplyrecently received from

M-Tor QuarterSection

/0�0+'�)ORZ�([SHULPHQWV

MTOR / MeGA Loop – LM flow in a toroidal magnetic field• Bi-Pb flowloop – 1.5 L/s• Inboard toroidal field – 0.6 T• Major radius 0.8 m, Aspect ratio 2• Assembly underway

Gap magnet experiments at UIUC• Small scale Magnetic Propulsion experiments

)OLEH�WXUEXOHQW�0+'�&RPSXWDWLRQDO�0RGHOV

k-e – MHD model(1.5D Hydrodynamics)

• MHD boundary conditions implemented• MHD drag terms for thick flow• developing flow version created• applied to task III problems

DNS simulations at limitedRe and High Pr

DNS-MHD simulations(still in development)

)OLEH�)ORZ�DQG�+HDW7UDQVIHU�([SHULPHQWV

FliHy facility – water/KOH discharge andflowloop

• Various open channel test sectionsplanned (curved sections, flat longsections and flow obstructions tosimulate penetrations)

• KOH to add electrical conductivity andsuppress surface vaporization

• UCLA design review last week• Flexible for IFE and Monbusho

collaboration needs

Potential to do low flowrate experiments inthe Japanese HTS loop

�7DVN�,,�)ORZ�DQG�+HDW�7UDQVIHU�0LOHVWRQHV

2/00- Design review for Flibe Hydrodynamics Simulation Facility (Fli-Hy)

: STATUS - Internal UCLA group review this week

3/00- Design review for toroidal MHD flow facility: DONE – assembly underway- Extension of 1.5-D MHD model to the case of a 2-component magnetic field,

begin analysis of cases at UCLA (toroidal+radial) : DONE – yielding results- Development of a 2-D model for analyzing local flow effects and field

gradients, begin analysis of cases at UCLA : DONE – yielding results

4/00- Complete base construction of Fli-Hy, begin tests on curved wall : STATUS

– awaiting facility completion- Modification of the current k-ε model and the code for the developing MHD

flow : DONE – yielding results

6/00- MHD facility operating at UCLA, begin first tests for field gradients and

applied currents : STATUS – awaiting facility completion

$3(;�)<�����7DVN�,,

/0�0+'�PRGHO�GHYHORSPHQW�

)UHH�VXUIDFH�IORZ�ZLWK�ILHOGJUDGLHQWV�DQG�DSSOLHG�FXUUHQWV

Presented by:Neil Morley

Contributors:D.-H. Gao, N. Morley, S. Smolentsev

APEX/ALPS MeetingArgonne National Lab

May 8-12, 2000

0+'�PRGHOLQJ�IRU�)5((�685)$&(IORZV�LQ�YDULDEOH�ILHOGV

Issues• Sensitivity of LM flows to spatial field gradients• Sensitivity of LM flows to temporal field gradients• Effect of applied electric currents on LM flow with and

without field gradients

LW Applications• Flow on first wall in 1/R toroidal field• Flow in NSTX 1/R pulsed toroidal field• Electromagnetic Restraint and Magnetic Propulsion• Jet Breakup into Droplet (cylinders) – internal velocity

profiles for heat transfer

&RGH�'HVFULSWLRQ

• Infinitely wide film in z (toroidal) direction (∂/∂z = 0)• Magnetic field in z-direction with spatial and temporal

variations in the xy-plane (Ba = Ba(x,y)k)• Electric currents applied at boundaries (Jn = ∂Bi/∂s)• VOF interface tracking methodology (∂F/∂T + v.∇F = 0)

y

x

B

Nozzle

Liquid Film

Back Plate

&RGH�'HYHORSPHQW�DQG%HQFKPDUNLQJ

Status• Applied to closed channels with Ha > 1000, N > 500 –• Agreement with Walker solutions for ∆P3D due to strong

gradients• Free surface benchmark tests with field begun, droplet

interaction with gradient region shown here

Near term improvements• Obstacles• Thin conducting walls• Alternate geometries• Better handling of non-linear terms for flow instabilities

Long term improvements• Multiple field components• 3D Hydrodynamics• Heat Transfer

([DPSOH�&DVH�

Acceleration of initially stagnant liquidwith a pseudo-free surface by

Zakharov’s Magnetic Propulsion effect

• Free inflow/outflow velocity and pressure BCs• Applied current density = 105 A/m2

• Channel size = 1 x 20 cm, Gradient region size = 1 cm• Free surface approximated by fixed, free slip wall at P=0

× B = 1 T × B = .5 T

YX

Insulated wall

Field gradientregion

Stagnant Li

"free surface"Applied Current

Fully developed results for MP simulation–velocity and current streamlines in the xy plane

YX

YX

x-velocity profiles in y-crosssection andpressure streamlines in the xy plane

YX

,PSRUWDQW�SRLQWV�JOHDQHG�IURP

0DJQHWLF�3URSXOVLRQ�WHVW�FDVH

• MP is effective in accelerating and sustaining LM flow inthe presence of field gradient and viscous drag

• Velocity profile produced by MP effect has region ofslow flow near surface - not attractive for surface heatremoval

• Actual free surface model is required to validate theseconclusion

([DPSOH�&DVH�

Deformation and deceleration of Lidroplet by a strong magnetic field

gradient

• Channel size = 2 x 6 cm, Gradient region size = 1 cm• Initial droplet radius 5 mm, Initial droplet velocity 1m/s• Free surface approximated by VOF function technique

× B = 0 T × B = 1 T

YX

Insulated wall

Field gradientregion

TranslatingLi droplet

'URSOHW�)� �ô�VXUIDFH��PRYLQJ�LQ�FRQVW�%� ��7

Uinitial = 1 m/sRdrop = 5 mmPictures at 0, 10, 20 ms

Uinitial = 1 m/sRdrop = 5 mmPictures at 30, 40, 50 ms

'URSOHW�)� �ô�VXUIDFH��PRYLQJ�LQWR�∇%� ���7�FP

Bleft = 0, Bright = 1Transition reg, x = 2 - 3 cmPictures at 10, 20, 30 ms

&XUUHQW�IORZ�LQ�WKH�GURSOHW

Bleft = 0, Bright = 1Transition reg, x = 2 - 3 cmPictures at 10, 20, 30 ms