thermo-chemical and mechanical coupled analysis of
TRANSCRIPT
G. PINAUD
ASTRIUM SPACE TRANSPORTATION, F-33165 St. Médard-en-Jalles, France
Thermo-chemical and mechanical coupled analysis of swelling charring and ablative
materials for re-entry application
5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 2
OUTLINE
Outline
AstriumNorcoat Liege
Phenomena overview and model
Experimental validationNumerical simulation
ConclusionAstrium
Norcoat Liege
Phenomena overviewAerothermodynamicsConductive heat transferPyrolysisAblationSwelling
Experimental validation
Numerical simulation
Conclusions & perspectives
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 3
AstriumAstrium: part of EADS, a global leader in aerospace and defence
Astrium
Norcoat LiegePhenomena overview and modelExperimental validation
Numerical simulationConclusionAstrium Space
TransportationThe European
prime contractorfor civil and
military spacetransportationand manned
space activities
RE-ENTRY SYSTEM &
TECHNOLOGY DEPARTMENT
ISS servicing
Thermal protection systems for
interplanetary probes
Interplanetary exploration
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 4
A material: Norcoat LiegeAstrium
Norcoat LiegePhenomena overview and modelExperimental validationHarvest of cork bark
From tree "Quercus suber"to
space
Stockpiling
drying of bark
Grinding
in granules
Mixing- Curing
Phenolic Resin Additives
NL sheets
NL molding
Back cover of ARD
Beagle2 Heat-shield
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 5
PHENOMENA OVERVIEW & MODEL
Close up view on a side cut of the heat shield
Astrium
Norcoat LiegePhenomena overview and modelExperimental validation
Numerical simulationConclusion
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 6
Aero-shell
Free stream flow
Boundary layer
Thermal protection material
Astrium
Norcoat LiegePhenomena overview and model:Side view
Experimental validationNumerical simulation
Conclusion
PHENOMENA OVERVIEW & MODEL
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 7
Free stream flow
Boundary layer
Convection
Radiation
Surface shear
Computation of the viscous boundary layer heat transfer:
convective heat fluxshock radiationwall shear stressequilibrium and non equilibrium
air chemistry
Astrium
Norcoat LiegePhenomena overview and model:Aerothermodynamic
Experimental validationNumerical simulation
Conclusion
PHENOMENA OVERVIEW & MODEL
Internal Convection Mutual Radiation
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 8
Boundary layer
Hea
t con
duct
ion
Computation of the conductive heat transfer in the material:
standard Fourier's lawmaterial state (density) default
linear dependency
Astrium
Norcoat LiegePhenomena overview and model:Conductive heat transferExperimental validation
Numerical simulation
Conclusion
PHENOMENA OVERVIEW & MODEL
[ ]cvv λλαλλ −−=
cv
v
ρρρρα
−−
=
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 9
Boundary layer
Under heating, organic material compounds start to degrade:
kinetics of the pyrolysis approached by multiple Arrhenius law
momentum conservation of the gaseous products follows Darcy's law for porous media ( and perfect gas law)
Astrium
Norcoat LiegePhenomena overview and model:Pyrolysis (1/2)
Experimental validationNumerical simulation
Conclusion
PHENOMENA OVERVIEW & MODEL
Decomposition gas blowing
Dec
ompo
sitio
n
Virgin
Pyrolysis
Charred
PKm Pg ∇−=&
T E M P E R A T U R E : [K ]
MA
SS
LOS
S:[
%]
0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
9 0
1 0 0
T G A : 5 K /m in2 A rrh e n iu s la w s: 5 K /m inT G A 2 0 K /s2 A rrh e n iu s la w 2 0 K /sT G A : 5 0 K /s2 A rrh e n iu s la w 5 0 K /s
( ) [ ] ( ) RTE
Ni
Ncv
Nvii
iiii eA −−− −−= χρρρχ 111&
∑=
Δ−=
Narrhii
v
i
v
t,1
.1)( χρρ
ρρ
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 10
Enthalpy of pyrolysis gas supposed at thermal and chemical equilibrium with charred material
Extrapolation at low temperature to keep constant endothermal heat of pyrolysis
Using quasi-steady state assumption for pyrolysis gas advective heat term
Astrium
Norcoat LiegePhenomena overview and model:Pyrolysis (2/2)
Experimental validationNumerical simulation
Conclusion
PHENOMENA OVERVIEW & MODEL
Chemica
l equ
ilibriu
m regim
e
Extrapolation
at constant ΔHpyroCV
NLCNLVgazpyro
NLgazpyro
THTHPTHPTH
THPTHPTH
CV
ρρρρ
−
−−=Δ
−=Δ
)()(),(),(
)(),(),(
)),((),(
PTHmt
PTHgaz
ggazg
&∇<<∂
∂ρ
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 11
Boundary layer
Surface material is removed by:heterogeneous chemical reactions
(oxydation with temperature dependent rate)
Heat of ablation taken into account
mechanical spalation under shear stress
Astrium
Norcoat LiegePhenomena overview and model:Ablation
Experimental validationNumerical simulation
Conclusion
PHENOMENA OVERVIEW & MODEL
Virgin
Charred
Original surface
Heatshield surface
Ablation
)( wcc Tfm =&
)( wcc THH Δ=Δ
),( wwmm Tfm τ=&
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 12
Boundary layer
Volume expansion generated by:in-depth structural and chemical
decomposition initiated by heat transfer
internal pyrolysis gas product pressure in cork cell
Prismatic cork cells membranes are made of 2 main molecules:
SuberinLignin
responsible for flexbility and rigidity
Astrium
Norcoat LiegePhenomena overview and model:Swell (1/3 )
Experimental validationNumerical simulation
Conclusion
PHENOMENA OVERVIEW & MODEL
Virgin
Charred Original surface
Heatshield surface
Swelling
Temperature increase triggers mass
exchange between the 2 structural molecules
cells wall become more brittle
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 13
Astrium
Norcoat LiegePhenomena overview and model:Swell (2/3 )
Experimental validationNumerical simulation
Conclusion
PHENOMENA OVERVIEW & MODEL
At higher temperature:cell membrane thickness becomes
thinner (mass loss)membranes start to stretchmembranes break-off
Considering:Volume phenomenonThermo-chemically decomposition linked mechanical expansionnon-elastic expansion
choice of an Arrhenius kind law for the strain rate evolution:
Temperature °C
1500
2000
2500
( ) [ ] ( ) RTE
Ns
Ncv
Nvst
ssss e−−− −−∝∂ χρρρε 111
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 14
Astrium
Norcoat LiegePhenomena overview and model:Swell (3/3)
Experimental validationNumerical simulation
Conclusion
PHENOMENA OVERVIEW & MODEL
Swelling numerically treated by a staggered coupling scheme involving :
Amaryllis for the thermal and ablative response
Mecano for the mechanical response
Supervisor as the interface and the synchronisation of the chore modules
Need to implement A.L.E (arbitrary Lagrangian Eulerian ) solution in Mecano to take into account for the mesh deformation (due to ablation) sent by AmaryllisT,ρ, ,P,X exchange at every time step and are used to solve for the stress and strain
( )
( )⎪⎪⎪
⎩
⎪⎪⎪
⎨
⎧
+−Δ−=
=∂∂
=
=+
LagrangianEulerian,,,
, and
0
tot
,
iijijij
klijkliji
i
ijij
PTFT
TCxPf
f
χραδεε
ερτ
τ
iχ
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 15
EXPERIMENTAL VALIDATIONObservation of swelling under various environment
Radiative furnaceFlat plate configurationMax cold wall heat flux: ~500 kw/m2
Duration : ~ 80 s Inner face thermocouple
Inductive plasma torch: CometeStagnation point configurationMax cold wall heat flux 800 kW/m2Internal thermocouple
Arcjet plasma: SimounAir/CO2 atmosphereMax cold wall heat flux 1.8 MW/m2Duration: ~60 sThermocouple plugsLaser recession
Astrium
Norcoat LiegePhenomena overview and model:Experimental validation
Numerical simulationConclusion
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 16
TIME: [s]
WA
LLD
ISP
LAC
EM
EN
T:[%
]
0 100 200 300 400 5000
10
20
30
40
50
Measures
Swell timeline
EXPERIMENTAL VALIDATIONRadiative furnace
Perspective view of the specimen during exposure to radiative heat flux
Astrium
Norcoat LiegePhenomena overview and model:Experimental validation
Numerical simulationConclusion
Initial thickness
T=0 s T=22 s T=74 s T=100 s
Post-test direct measurements
Specimen sheet before test
Specimen sheet after test
Final thickness
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 17
EXPERIMENTAL VALIDATIONComete inductive plasma torch
Astrium
Norcoat LiegePhenomena overview and model:Experimental validation
Numerical simulationConclusion
TIME: [s]
WA
LLD
ISP
LAC
EM
EN
T:[%
]
0 50 100 150 200 250 300 350 400 450 500-10
-5
0
5
10
15
20
Measures
T=0 s T=150 sT=45 s
Heat flux ~800 kW/m2
Specimen before test
Specimen after test
Comparison of surface scanned geometry
before/after test
Post-test direct axial
measurements
Strong lagged swelling effect on thickness
Recession
Expansion
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 18
EXPERIMENTAL VALIDATIONSimoun arc-jet plasma test
Astrium
Norcoat LiegePhenomena overview and model:Experimental validation
Numerical simulationConclusion
Specimen before test Specimen after test
Cross section:
Charred, pyrolysis and virgin layer
Top view of the sample at thermocouple plug
location
T IM E : [s]
WA
LLD
ISP
LAC
EM
EN
T:[%
]
0 50 100 150 200
-70
-60
-50
-40
-30
-20
-10
0
10
M easures Laser AM easures Laser B
Pla
sma
shut
dow
n
S atu
ratio
nLa
ser
Recession
Expansion
Laser tracked surface motion
Due to high convective exchange coefficient, recession is dominating expansion
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 19
NUMERICAL SIMULATIONSurface kinetic rebuilding including
Surface chemical recessionSurface mechanical recessionVolume swelling
Astrium
Norcoat LiegePhenomena overview and model:Experimental validation
Numerical simulationConclusion
TIME:[s]
WA
LLD
ISP
LAC
EM
EN
T:[m
]
0 50 100 150 200-0.012
-0.01
-0.008
-0.006
-0.004
-0.002
0
0.002
MeasuresLaserAMeasuresLaserBSimulationP
lasm
ash
utdo
wn
Satu
ratio
nLa
ser
TIME:[s]
WA
LLD
ISP
LAC
EM
EN
T:[m
]
0 50 100 150 200 250 300 350 400 450 500-0.002
-0.0015
-0.001
-0.0005
0
0.0005
0.001
0.0015
0.002
0.0025
0.003
0.0035
MeasuresSimulation
TIME:[s]
WA
LLD
ISP
LAC
EM
EN
T:[m
]
0 100 200 300 400
0
0.0005
0.001
0.0015
0.002
0.0025
MeasuresSimulation
Optimization of the ablation law, and the swelling Arrhenius law parameters to fit the surface motion, density gradient and thermal thermocouple history
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 20
CONCLUSIONNew development in Amaryllis and Mecano were successfully applied to model the behavior of a cork based Thermal Protection Material including:
conductive heat transferchemical decomposition (pyrolysis) kinetics and energychemical and mechanical ablationthermally-generated volume swelling
For cork based material, ablation and swelling cannot be dissociated in model rebuilding. Lagged effects of swelling must be considered for checking the reliability of post-test direct measurements
Samcef modules enable full coupling between thermal, thermo-chemical and mechanical response
various industrial issues when coupled thermal and mechanical loads play an important rule could be treated
Astrium
Norcoat LiegePhenomena overview and model:Experimental validation
Numerical simulationConclusion
G. Pinaud - 5th Ablation Workshop, Feb. 28th – Mar. 1st 2012, Lexington, Kentucky - 21
CONCLUSION & FUTURE
Future model will focus on 2D orthotropic swelling material properties
Design a specific instrumentation for in-depth characterization of swelling phenomena
Astrium
Norcoat LiegePhenomena overview and model:Experimental validation
Numerical simulationConclusion
Temperature field Density field