projet cedal développement d’une matrice à base de ciment...
TRANSCRIPT
Projet CEDAL
Développement d’une matrice à
base de ciment phospho-magnésien
pour le conditionnement de
l’aluminium métallique
| David LAMBERTIN, Céline CAU DIT COUMES, Hugo LAHAL LE, Pascal ANTONUCCI : CEA MARCOULE/DEN/DTCD/SPDE
Laboratoire de Physico-chimie des Matériaux Cimenta ires
| Sylvie DELPECH, Céline CANNES, Quang Thuan TRAN : IPN Orsay
28 OCTOBRE 2014 | PAGE 1
1. Introduction: hardened paste of Portland cement
28 OCTOBRE 2014 | PAGE 2
Portland cement paste = porous, hydrates and alkaline medium
Alkaline pore water
SiO2 SO3 Na2O K2O pH
22 844 4430 26100 13.6
Concentrations in mg/kg of extracted solution
OPC paste (clinker 95.5% - gypsum 4.5%) - W/C 0.5 -curing at 20°C in air-tight bag for 13 months (Longuet, 1973)
C-S-H (≈ 70%)
Portlandite ( ≈ 20%)
+ Hydrated aluminates (≈10%)
Capillarypore
Journées Needs | 21 octobre 2014
1. Introduction: redox properties of aluminium
Al3+
AlO2-
Al
E(V)
pH
PASSIVATION CORROSIONCORROSION
IMMUNITY
2 Al + 2 OH - + 2 H2O →→→→ 2 AlO2- + 3 H22 Al + 6 H+ →→→→ 2 Al 3+ + 3 H2
28 OCTOBRE 2014 | PAGE 3Journées Needs | 21 octobre 2014
Portand cement
1. Introduction: ANDRA specifications
28 OCTOBRE 2014 | PAGE 4
Specifications for low- and intermediate-level and s hort lived radioactive waste (CSFMA waste disposal facility)
Maximum surface area of Al per 330L-canister
Al directly embedded in cement 0.1 m2
Al isolated from cement by a material like vinyl
0.5 m2Al in a canister to be compacted
1 m2
Journées Needs | 21 octobre 2014
1. Introduction: which strategy for Al encapsulation?
28 OCTOBRE 2014 | PAGE 5
2 complementary approaches
Aluminium (1E-4 mol/L)
0 4 82 6 10 12 14pH
-3
-2
-1
0
1
Eh (V)
Al
Al3+ DiasporeAlO2
-
H2 (g)
O2 (g)
Passivation
� Use of alternative cements, yielding pore solutions with a pH close to neutrality
� Addition of corrosion inhibitors
Journées Needs | 21 octobre 2014
1. Outline
���� Literature survey : search for alternative binders to calcium silicate cements
���� Experimental screening of binders
���� Factors controlling the cement reactivity
| PAGE 6
���� Processes responsible for setting and hardening
���� Electrochemical behavior of metallic aluminum in a magnesium phosphate binder
Journées Needs | 21 octobre 2014
2. Literature survey : search for alternative binders to calcium silicate cements
28 OCTOBRE 2014 | PAGE 7
Comparison of pore solution pH generated by differe nt cement systems
pH3 4 5 6 7 8 9 10 11 12 13
OPC
Ettringite cement
MagnesiumPhosphate cement
PlasterMagnesium
chloride cement
pH3 4 5 6 7 8 9 10 11 12 13
MagnesiumSilicatecement
210
Brushite cement
pH3 4 5 6 7 8 9 10 11 12 13 pH3 4 5 6 7 8 9 10 11 12 13
Al passivation
Apatitic cement
210
Calcium aluminate
phosphate cement
Journées Needs | 21 octobre 2014
2. Literature survey : conclusion
28 OCTOBRE 2014 | PAGE 8
Selection of binders for an experimental study
pH3 4 5 6 7 8 9 10 11 12 13
OPC
Ettringitecement
MagnesiumPhosphate cement
pH3 4 5 6 7 8 9 10 11 12 13
MagnesiumSilicatecement
210
Brushite cement
pH3 4 5 6 7 8 9 10 11 12 13 pH3 4 5 6 7 8 9 10 11 12 13
Al passivation
210
Experimental study to measure under identical condi tions the hydrogen release due to corrosion of an aluminum rod embedde d in the selected cement pastes
Journées Needs | 21 octobre 2014
PlasterMagnesium
chloridecement
Apatiticcement
Calcium aluminate
phosphate cement
Binder Reference and producer W/C ratio (g/g) Reference
Portland cement CEM I 52.5 PM ES CP2 (Lafarge Le Teil)
0.4 OPC
Calcium aluminate cement + gypsum (67 :33 g/g) Fondu (Kerneos) 0.48
CAC + gypsum
Calcium aluminate cement + gypsum + silica fume(60 :30:10 g/g)
Fondu (Kerneos) SF S95DM (Condensil)
0.60 CAC + gypsum + SF
Calcium sulfoaluminate cement (75% clinker, 25% gypsum) KTS 100 (Belitex) 0.55 CSAC
Magnesium silicate binder (40% MgO, 10% MgCO3, 50% SiO2)
MgO, MgCO3 (VWR) SF S95DM (Condensil)
0.35 MSH
Magnesium phosphate cement (MgO: KH2PO4 1:1 mol/mol)
MgO MagChem 10 CR (MAF Magnesite BV)
KH2PO4 (VWR) 0.51 MKP
Brushite cement (wollastonite + phosphoric acid solution) FOTIMINE (Sulitec) 0.80 BRU
3. Experimental screening of binders
28 OCTOBRE 2014 | PAGE 9
Simulated waste
Aluminum rod (99.99% Al) (φ = 1 cm, h = 3 cm)Depassivated by immersion in H2SO4 (20%) and rinsed with demineralized water just before the experiment
Investigated binders
Journées Needs | 21 octobre 2014
3. Experimental screening of binders
28 OCTOBRE 2014 | PAGE 10
Experimental device for H 2 measurement
Aluminium
H2
Gaschromatography
P
Cement paste
Air-tight reactor
Journées Needs | 21 octobre 2014
3. Experimental screening of binders
28 OCTOBRE 2014 | PAGE 11
H2 production
0
5
10
15
20
25
30
35
40
45
0 20 40 60 80 100 120 140
Time (days)
OPC
CAC + gypsum
CAC + gypsum + SF
CSAC
MSH
MKP
BRU
Rel
ease
d H
2(L
/m2
of A
l)
0.26 L/m2/y
Most promising binder: magnesium phosphate binder MgO + KH2PO4 + 5 H2O→ MgKPO4.6H2O
7 L/m2/y
Journées Needs | 21 octobre 2014
3. Experimental screening of binders
28 OCTOBRE 2014 | PAGE 12
H2 production
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0 20 40 60 80 100 120 140
Time (days)
MKP + H3BO3
MKP + H3BO3 + LiNO3
Rel
ease
d H
2(L
/m2
of A
l)
With LiNO3: released H2 < detection limit of the analytical methodExtrapolation over 1 year : H2 < 2.28x10-4 L/(m2.year)
Journées Needs | 21 octobre 2014
Formation of a protective insoluble layer of LiH(AlO2)2.5H2O at the surface of the metal (Matsuo et al., 1995)
3. Experimental screening of binders
28 OCTOBRE 2014 | PAGE 13
Conclusion
Selection of the magnesium -phosphate cementfor the next step of the project
Journées Needs | 21 octobre 2014
28 OCTOBRE 2014 | PAGE 14
Critical parameters for a formulation
4. Control of the cement reactivity
MgO / KH2PO4 ratio
Type and content of filler
Type and content of sand
MgO characteristics (particle size, specific surface area)
Use of admixtures
Hardburnt (S < 1 m2/g)
1
Low CaO fly ash
Siliceous sand
H3BO3, LiNO3
Journées Needs | 21 octobre 2014
| PAGE 15
4. Processes responsible for the setting and hardening
Mechanisms of setting and hardening System {MgO + KH 2PO4 + Fly ash}
-40
-20
0
20
40
60
80
0.E+00
1.E+06
2.E+06
3.E+06
4.E+06
5.E+06
6.E+06
7.E+06
8.E+06
9.E+06
1.E+07
0 1 2 3 4 5 6 7
Co
nd
ucti
vit
y (
mS
/cm
)
He
at
flo
w (
µw
/g
)
G' (
Pa
)
Time (h)
Shear storage modulus
Conductivity
Heat flow
15 min 26 jK-struvite
0 min 5 h 30 minKH2PO4
3 h 45 min 26 jOther secondary phases
(MgHPO4.7H2O, AlPO4.xH2O, ?)
Poorly crystallized phase (Mg2KH(PO4)2(H2O)15 ?)15 min 3 h 45 min
Beginningof setting
End of setting
Thèse de Hugo Lahalle en cours
5. Electrochemical behaviour of aluminium in a magnesium phosphate binder
L’électrochimie pour l’étude de la corrosion in situ
| PAGE 16Journées Needs | 21 octobre 2014
5. Electrochmical behaviour of aluminium in a magnesium phosphate binder
La spectroscopie d’impédance pour l’étude de la corros ion chimique in situ
| PAGE 17Journées Needs | 21 octobre 2014
5. Electrochemical behaviour of aluminium in a magnesium phosphate binder
Déroulement de l’étude
| PAGE 18Journées Needs | 21 octobre 2014
5. Electrochemical behaviour of aluminium in a magnesium phosphate binder
Pâtes de ciment phospho-magnésien et Portland : diagra mmes d’impédance
| PAGE 19Journées Needs | 21 octobre 2014
Evolution au cours du temps (Phospho-
magnésien)
Portland avec LiNO 3: impédance caractéristique d’un phénomène de corrosion
Phospho-magnésien avec LiNO 3: diminution de l’impedance pendant les 4 premiers jours puis
augmentation => Il est probable que la couche d’oxyde qui protège initialement Al soit progressivement
remplacée par l’aluminate de lithium, plus stable et protecteur
| PAGE 20
5. Conclusion
Magnesium phosphate cements appear to be the most promising binders (amongthose tested) for the encapsulation of metallic aluminium
Corrosion of aluminium can be reduced still further by adding a corrosion inhibitor(lithium nitrate): Brevet Français , C.Cau dit Coumes, D.Lambertin, P.Antonucci, M.Charlot,« Liant et son utilisation pour le conditionnement de déchets contenant de l’aluminium métallique», N°FR3003252A1Critical parameters controlling the cement reactivity :
- type of MgO (calcination temperature, specific surface area)- type of filler used- MgO / KH2PO4 ratio- L/S ratio- type and amount of admixtures
Journées Needs | 21 octobre 2014
The electrochemical measurements (determination of the open circuit potential and electrochemical impedance spectroscopy) confirm that aluminum is corroded when it is encapsulated in a Portland cement paste, whereas it tends to a passive state in the magnesium phosphate binder
C. Cau Dit Coumes, D. Lambertin, H. Lahalle, P. Antonucci, C. Cannes, S. Delpech, Selection of a mineral binder with potentialities for the stabilization / solidification of aluminum metal, Journal of Nuclear Materials 453 [1-3] (2014) 31-40
| PAGE 21
5. Prospects
Use of the electrochemical method developed in this project to study the encapsulation of other reactive metals
Analysis of the EIS data by using judicious equivalent electrical circuits to get more information on the aluminum corrosion (mechanism, rate)
Understanding of the influence of admixtures (H3BO3, LiNO3) on cementhydration
Design and characterization of an injection grout in view of blocking massive aluminum pieces
Journées Needs | 21 octobre 2014
3.3 m2 d’aluminium métal
CEA/DEN/DTCD/LP2CCEA/DEN/DTCD/LP2C
30L MKP+3.3 m 2 d’Al