Modelling of Chemical Alteration of Cement Materials in Radioactive - PowerPoint PPT Presentation

International Workshop: 2nd Mechanisms and modelling of waste/cement interactions Modelling of Chemical Alteration of Cement Materials in Radioactive Waste Repository Environment Daisuke Sugiyama, Isao Kurashige Central Research Institute of Electric Power Industry (CRIEPI), Japan 14 October 2008, Le Croisic [1]

Cementitious materials in radwaste repository Backfill Construction Waste package - High pH (Alkaline) plume - Interaction with bentonite buffer and rock GW flow Physical and chemical Chemical alteration ： containment: - Dissolution - High pH (low solubility, high - Secondary minerals sorption) - Crystallization - Low permeability/diffusivity for a LONG TERM [2]

Contents 1. Observation of alteration of OPC monolith (Lab. Experimental) - Leaching/precipitation of components in a tank-leaching experiment - Effect of surface precipitates (CaCO 3 ) on alteration 2. Development of a reactive transport computational code (CCT-P) - Thermodynamic incongruent dissolution model of C-S-H - One-dimensional advection/dispersion/diffusion equation - Evolution of the hydraulic properties of the cement solid matrix due to the leaching and precipitation of components - Description of precipitation of secondary less-soluble phase acting as a diffusion barrier 3. Preliminary calculation of the evolution of the cementitious repository system 4. Summary and conclusions [3]

OPC monolith alteration experiments A tank-leaching experiment OPC hydrate monolith: - w/c=0.35, cured in OPC-equilibrated water for 91 days at 50 ° C - 20 x 20 x 70mm - Only one of the faces of monolith was exposed to the aqueous solution Solution: - Deionised water - NaHCO 3 solution (6e-5, 6e-4, 6e-3 M) • The solution and solid samples were separated and the monolith samples were recontacted with fresh deionised water or NaHCO 3 solution after 1, 5, 9, 13, 17, 21, 26, 30, 34, 39, 43, 47 and 52 weeks. • All experiments were carried out in an nitrogen- filled glovebox. OPC monolith Epoxy resin [4]

Results(1): Calcium leaching -3 2.5x10 In Deionised water : NaHCO 3 Experiment The dissolution of Ca(OH) 2 2 ] 0 (Water) dominated the leaching of Ca leached from OPC monolith [mol/cm 2.0 6e-5M calcium in the early stage, 6e-4M then the incongruent 6e-3M dissolution of C-S-H gel in 1.5 the altered surface region dominated the calcium leaching. 1.0 In NaHCO 3 solution : 0.5 The leaching of calcium was inhibited, significantly at the 0.0 NaHCO 3 concentrations of 0 10 20 30 40 50 6e-4 and 6e-3 M. Time [week] [5]

Results(2): Calcium concentration in solid NaHCO 3 6e-5 M 6e-4 M 6e-3 M Deionised water NaHCO 3 : 6e-5 M 500 µm NaHCO 3 : 6e-4 M 1 mm NaHCO 3 : 6e-3 M 0 1 2 3 4 mm 100 µm Initial surface of monolith Surface of monolith (Solid/liquid interface) (Solid/liquid 0 0.5 1.0 1.5 2.0 2.5 0 2 4 6 8 10 interface) Ca/Si ratio in OPC solid matrix Ca/Si ratio in OPC solid matrix In 6e-4 and 6e-3 M sodium bicarbonate solutions, little calcium was leached and a layer of calcite precipitation formed. [6]

CCT-P : A coupling transport and chemical equilibrium calculation code dC / dx=0 or C=const. dC / dx=0 or C=const. Region 1 Region 2 Region 3 Region n φ 1 ( x , t ) φ 2 ( x , t ) φ 3 ( x , t ) φ 3 ( x , t ) Porous D e 1 ( x , t ) D e 2 ( x , t ) D e 3 ( x , t ) D e 3 ( x , t ) R d 1 ( x , t ) R d 2 ( x , t ) R d 3 ( x , t ) R d 3 ( x , t ) matrix ・・・ Dissolution/Precipitation/… Diffusion/Advection x ∂ ∂ ∂ ∂ φ ⋅ ⋅ C ( x , t ) C ( x , t ) { ( x , t ) R ( x , t ) C ( x , t )} ⋅ − ⋅ = − d { D ( x , t ) } V S ( x , t ) e d eq ∂ ∂ ∂ ∂ x x x t φ : porosity, C : concentration of aqueous species, t : time, V d : velocity of flow in matrix, D e : effective diffusion constant in matrix, S eq : source term given by chemical equilibrium calculation within matrix, ⎛ − ⎞ φ 1 ( t ) R d : retardation factor ( ), ρ : density, ⎜ ⎟ = + ρ ⋅ ⋅ K d : distribution coefficient. R ( t ) 1 K ⎜ ⎟ d d φ ⎝ ⎠ ( t ) [7]

The evolution of the hydraulic properties of the solid In CCT-P, the diffusion coefficient in the altered region of the solid matrix can be described as a function of porosity; n ⎛ ⎞ φ ( t ) ⎜ ⎟ = ⋅ D ( t ) D (0) ( n =2 in this study) ⎜ ⎟ φ ⎝ ⎠ (0) The porosity of the solid matrix increases or decreases as the component minerals are dissolved and leached or precipitated, respectively; V solid : volume of solid phase, ( t ) V φ = − − φ ⋅ solid CS : molarity of component mineral, ( t ) 1 ( 1 ( 0 )) ( 0 ) V solid v mol : molar volume of component mineral = ∑ v mol Ca(OH)2 = 0.0331 dm 3 mol -1 , ⋅ + ( t ) CS ( t ) v V V i mol solid , static solid v mol SiO2 = 0.0273 dm 3 mol -1 , i i : solid v mol CaCO3 = 0.0369 dm 3 mol -1 , V solid,static : volume of insoluble residual solid phase. [8]

A hypothetical reaction layer model At the boundaries of the regions, the advection/dispersion/diffusion equations in adjacent regions are connected as follows: ∂ ∂ C ( x , t ) C ( x , t ) − ⋅ + ⋅ = − ⋅ + ⋅ D ( x , t ) V C ( x , t ) D ( x , t ) V C ( x , t ) , e upper d boundary e lower d boundary ∂ ∂ x upper x lower upper lower A hypothetical reaction layer model is introduced when a less-soluble or insoluble phase is precipitated: ⎧ ⎫ ⎛ ⎞ ∂ φ ∂ ⎪ ⎪ C ( x , t ) ( t ) C ( x , t ) ⎜ ⎟ − ⋅ = ⋅ − ⋅ ⎨ ⎬ D ( x , t ) D ( x , t ) ⎜ ⎟ e cement e HRL ∂ φ ∂ ⎪ ⎪ ⎝ ⎠ x (0) x ⎩ ⎭ cement HRL surface Hypothetical Cement solid Solution reaction layer (CH, C-S-H gel 、 CaCO 3 ) (NaHCO 3 ) - HCO 3 HCO 3 - CaCO 3 Porosity of the solid matrix in the near surface region decreases precipitates Ca 2+ Ca 2+ [9]

Incongruent C-S-H dissolution/precipitation model • C-S-H is described as a binary nonideal solid solution of Ca(OH) 2 and SiO 2 . • The notable features of the model are its good continuity and simplicity, so that the model predicts well the equilibria of the incongruent precipitation/dissolution of cementitious materials accompanying the change of Ca/Si ratio by iterative numerical calculations. 1 1 1 y 1 - y 1 - y ' ' ' 2 ・ ・ ・ logK = logK - log + [A + A ( ) + A ( ) ] + + s s0 s0 s1 s2 2 1 + y 1 + y 1 + y 1 y 1 y (1 + y ) y y y y 1 - y 1 - y ' ' ' 2 ・ ・ ・ logK = logK - log + [A + A ( ) + A ( ) ] + + c c0 c0 c1 c2 2 1 + y 1 + y 1 + y 1 y 1 y (1 + y ) ( y = Ca/Si of C-S-H ) End member SiO 2 Ca(OH) 2 A ij A s0 A s1 A s2 A c0 A c1 A c2 Ca/Si ≤ 0.833 -18.908 57.821 -58.779 36.902 -37.015 163.21 Ca/Si > 0.833 -18.933 49.633 24.582 36.923 -7.8143 -50.323 at Ca/Si ≦ 0.400 SiO 2 : logK s = logK s0 – log(1+ y ) logK s0 = -2.639 at 1.686 ≦ Ca/Si SiO 2 : logK s = -7.835 logK c0 = 22.71 (JNC-TDB) Ca(OH) 2 : logK c = 22.71 (= log K c0 ) Refs: D. Sugiyama and T. Fujita, Cem Concr Res 36 (2006) 227-237. [10] JAEA and FEPC, JAEA-Review 2007-010, FEPC TRU-TR2-2007-01, March 2007.

Modelling of the OPC experiment Calculated mineral composition in OPC hydrate [mol/kg] Ettringite Ca(OH) 2 C-S-H gel NaOH KOH (Ca/Si = 1.686) 0.31 2.5 2.7 0.052 0.064 Solution Boundary condition: C (0, t ) = C 0 Hypothetical reaction layer (for NaHCO 3 cases) Solid OPC Thickness: 5mm monolith Number of grid layers : 45 Thickness of each grid layer : 10-920µm Boundary condition: d C /d x = 0 D e (0) = 7.8e-12 [m 2 /s] within the initial solid matrix [11]

Calculation parameters for OPC modelling 6×10 -5 mol 6×10 -4 mol 6×10 -3 mol Solution Deionised dm -3 NaHCO 3 dm -3 NaHCO 3 dm -3 NaHCO 3 water － Hypothetical Thickness of 0.6 0.6 0.6 reaction region [mm] layer － Number of grid 3 3 3 layers － Thickness of each 0.2 0.2 0.2 grid layer [mm] * [m 2 s -1 ] 8.0×10 -10 8.0×10 -10 8.0×10 -10 － Initial D e － Initial porosity 1 1 1 Vicinity of Thickness of 2.7 2.7 0.4 0.4 the surface region [mm] of OPC Number of grid 40 40 40 40 monolith layers Thickness of each 0.0675 0.0675 0.01 0.01 grid layer [mm] Initial D e [m 2 s -1 ] 7.8 × 10 -12 7.8 × 10 -12 7.8 × 10 -12 7.8 × 10 -12 Initial porosity 0.138 0.138 0.138 0.138 Matrix of Thickness of 2.3 2.3 4.6 4.6 OPC region [mm] monolith Number of grid 5 5 5 5 layers Thickness of each 0.460 0.460 0.92 0.92 grid layer [mm] Initial D e [m 2 s -1 ] 7.8 × 10 -12 7.8 × 10 -12 7.8 × 10 -12 7.8 × 10 -12 Initial porosity 0.138 0.138 0.138 0.138 [12]

Modelling Results(1): Calcium leaching NaHCO 3 Experiment Modelling The effective diffusion -3 2.5x10 0 (Water) coefficient within the initial 6e-5M solid matrix was estimated by 2 ] 6e-4M Ca leached from OPC monolith [mol/cm a series of sensitivity 2.0 6e-3M analyses to fit the measured amount of leached calcium in deionised water. 1.5 D e (0) = 7.8e-12 [m 2 /s] 1.0 The modelling 0.5 calculations very accurately quantitatively predicted the 0.0 experimental results for 0 10 20 30 40 50 the leaching of calcium. Time [week] [13]