WELL MATERIALS AGEING IN A S-CO 2 STORAGE

Transcription

1 Nicolas Jacquemet WELL MATERIALS AGEING IN A CONTEXT OF H 2 S-CO 2 GEOLOGICAL STORAGE Lab. G2R Univ. H. Poincaré (Nancy, France) 1

2 One issue: the leakage through deteriorated wells RESERVOIR WELL RESERVOIR CEMENT Cement Steel SC H 2 S-CO 2 Biphasic fluid = SC H 2 S-CO 2 + Aqueous liquid Aqueous liquid ~ 10 cm 4 ph ph 10 Different cement-fluids contacts High P-T conditions Alteration of the materials Pathway for leakage to the surface SCOPE OF THE STUDY: Evaluation of the well materials durability in such environment INTRODUCTION 2

3 Plan 1 - EXPERIMENTAL AND ANALYSIS PROTOCOL 2 - STARTING MATERIALS 3 - EXPERIMENTAL APPROACH 4 - NUMERICAL APPROACH INTRODUCTION 3

4 1-Experimental and analysis protocol 1. Experimental protocol 2. Analysis protocol Experimental protocol 4

5 Experimental apparatus AUTOCLAVES AND REACTORS Systems at geological relevant P-T Autoclave Engineer Hydraulic pressure autoclaves (Max. cond.: 1000 bar-450 C) Systems with high concentration of gas Safety rules Use of micro-reactors (gold capsules of 2 cc) H 2 S-CO 2 Cement + Steel Brine GAS LOADING IN THE CAPSULES Specific apparatus under the form of a gas line Allows to introduce safely a precise mass of gas in the capsules Experimental protocol 5

6 1-Experimental and analysis protocol 1. Experimental protocol 2. Analysis protocol Analysis protocol 6

7 Analysis of cement and steel 1. STEEL Optical microscopy SEM TEM 2. CEMENT X rays diffraction SEM elemental mapping Raman micro-spectroscopy mineralogical mapping TEM Water porosimetry Analysis protocol 7

8 Analysis of the fluids 1. Recovery and quantitative analysis of the residual gas using the gas line --> Mass balance of the gaseous species --> amount of mineralized gas 2. Synthetic fluid inclusions --> in situ sampling of fluids surrounding the materials --> visualisation of the number and state of fluid phases --> Composition of the phases (speciation of liquid, concentration, ) Analysis protocol 8

9 2-Starting materials Starting materials 9

10 The starting materials: analogues of deep well materials CEMENT (slurry hardening) The slurry has a typical composition of deep well cement slurry It hardened in hydrothermal conditions during the cure Bicalcium Silcate Assemblage of Tobermorite + Quartz Quartz Tobermorite (hydrothermal CSH) Ferrite 1 mm STEEL: used for the casing low C content --> Fe (98 mol%) Starting materials 10

11 3-Experimental approach 1. Experimental conditions 2. Results Experimental approach 11

12 Geologically relevant conditions Cement under the form of bar H 2 S+CO 2 Brine 1 mm Steel Bar of cement 5 mm RESERVOIR EXPERIMENTS FORMATION WATER: TOTAL PRESSURE: NaCl brine (150 g/l) 500 bar TEMPERATURE: 120 C, 200 C GAS: 66mol% H 2 S + 34mol% CO 2 Experimental approach 12

13 Three fluids at the cement-external media interface 3 types of ageing exposures LIQUID (L) BIPHASIC (L+S) DRY SC (S) H 2 S+CO 2 Brine Cement bar Porous volume Brine SC phase SC phase Mineral Experimental approach 13

14 3-Experimental approach 1. Experimental conditions 2. Results Experimental approach 14

15 Different carbonation profiles according to the exposures L L+S S Thin front of constant thickness Front of variable thickness No front 1 mm 1 mm 1 mm Internal zone Intermediate zone Internal zone Intermediate zone Thin external deposit CaCO 3 deposit Thin external deposit 100 µm 25 µm 100 µm Constant thickness front Massive deposit of CaCO 3 Variable thickness front Total homogeneous alteration Experimental approach 15

16 Schematization of the carbonation profiles L L+S S Massive calcite Massive calcite Aragonite (120 C) OR Calcite (200 C) µm NO ALTERED CEMENT DECALCIFIED CEMENT CARBONATED CEMENT Diffusion of C and Ca species Massive deposit of calcite Diffusive blockage Alteration stopping Intrusion of the SC into the porous volume digitations Experimental approach Total carbonation (favorised by the absence of liquid water and optimal diffusion of the SC CO 2 ) 16

17 The sulfidation of the iron bearing phases 1. IN THE CIMENT: THE SULFIDATION OF THE FERRITES 200 C Pyrite Ca 2 AlFeO H 2 S FeS µm 2. THE SULFIDATION OF THE STEEL Fe + H 2 S FeS + H 2 Before After pyrrhotite Fe sulfides 1 mm No altered steel 200 µm Iron sulphide crust 30 µm Group of pyrrhotites Experimental approach 17

18 4 Numerical approach Numerical approach 18

19 Cement alteration: a result of coupled processes H 2 S+CO 2 BRINE CEMENT L Dissolution Reaction Diffusion Diffusion Need of a coupled chemistry-transport code (HYTEC) SCOPE: Reproduction of the experimental results by numerical simulation Calcite L 200 C Decalcified zone No altered cement STOPPING OF ALTERATION BY DIFFUSIVE BLOCKAGE 100 µm µm Numerical approach 19

20 Modelling results EXTERNAL MEDIA CEMENT Concentration (mol/l) Name Calcite Ca/Si +inf. Foshagite 1,33 Tobermorite-11A 0,83 Gyrolite 0,66 SiO 2(am) 0,00 MODELLING 0-0,2 0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 X(mm) Stopping at 5-10 d. Crust Decalcified zone No altered cement EXPERIMENT Confirm the calcite deposit as the responsible of the stopping of the alteration that we observed experimentally Refine the model to fit the experimental and numerical front thicknesses Numerical approach 20

21 Conclusion CONCLUSION 21

22 Distinctive reactions according to the material WELL STEEL CEMENT H 2 S Legend: Steel Ferrite C-S-H CO 2 Sulfidation Carbonation The cement highly reacts with the CO 2 weakly with the H 2 S The steel strongly reacts with the H 2 S weakly with the CO 2 1st order parameter which controls the intensity degree of carbonation = fluid phase at the contact of the minerals carbonation favoured without liquid water, max. within dry SC CONCLUSION 22

23 Evaluation of the well durability (petrograph point of vue) THREE CRITERIONS OF DURABILITY OF THE CEMENT: Mineralogic evolution: Mf = Mi Textural evolution: Tf = Ti Porosity evolution : Pf < Pi or Pf = Pi INITIAL WELL FINAL WELL (120 C) CEMENT STEEL Steel Cement SC H 2 S-CO 2 INJECTION ZONE Mf Mi Tf = Ti Pf < Pi Mi Ti Pi Biphasic fluid GAS ZONE Mf = Mi Tf = Ti Pf < Pi Aqueous liquid WATER ZONE Mf = Mi Tf = Ti Pf < Pi CONCLUSION 23

24 Limits and perspectives Ageing at 200 C: experimental T is superior to the cure T of the cement --> superposition of two processes (re crystallisation and alteration) Ageing T curing T of the cement At reservoir scale: 1) volumic predominance of the reservoir rock / well, 2) flowing. the fluids which interact with the well are: 1) at equilibrium with the reservoir rock, 2) renewed. Reactive percolation experiments where the fluid is pre-equilibrated with the reservoir rock Exhaustive study of durability Must take in account the mechanical properties variations of the materials Large spatial and time scale view on the alteration of well --> Extrapolate the numerical results CONCLUSION 24

25 COMPLEMENTARY SLIDES Complementary slides 25

26 The micro-reactors and autoclaves Respect of geological relevant P-T Hydraulic pressure autoclaves (Max. cond.: 1000 bar-450 C) Systems with high concentration of gas Safety rules Use of micro-reactors (gold capsules of 2 cc) Hydraulic liquid Gauge Hydraulic HP pump Autoclave Engineer H 2 S-CO 2 Autoclave Solids Furnace Liquid Temperature controler Complementary slides 26

27 The secured gas loading in gold capsules Vacuum pump Pressure displayer Gas inlet (1-2 bar) Calibrated volume Valve capsule Liquid N 2 Gas inlet and apparatus placed in hood in safe area The cryo-condensation allows a low pressure loading Precise mass of gas in the capsules Complementary slides 27

28 Terminology in cement chemistry and specific mineralogy Oxide Designation CaO SiO 2 H 2 O Al 2 O 3 Fe 2 O 3 C S H A F Mineral name Simplified formula Mineralogic formula Hydrothermal C-S-H Tobermorite C 5 S 6 H 5 Ca 5 Si 6 O 16 (OH) 2 4H 2 O Xonotlite C 6 S 6 H Ca 6 Si 6 O 17 (OH) 2 Other minerals Quartz S SiO 2 Bicalcium silicate C 2 S Ca 2 SiO 4 Ferrite C 4 AF Ca 2 AlFeO 5 Complementary slides 28

29 The starting materials: analogues of deep well materials CEMENT (slurry hardening) The slurry has a typical composition of well cement slurry : Portland class G-HSR + silica flour (quartz grains) + water It hardened in hydrothermal conditions during the cure: 210 bar, 140 C, 8 days C 2 S Assemblage of Tobermorite + Quartz Porosity = 0.4 Quartz Tobermorite C 4 AF 1 mm STEEL: used for the casing low C content --> Fe (98 mol%) Complementary slides 29

30 Two forms of cement and geological relevant conditions Cement under the form of powder Cement under the form of bar H 2 S+CO 2 H 2 S+CO 2 Brine 1 mm Brine 1 mm Steel Steel Powder of hardened cement Bar of cement 1.5 cm 5 mm Optimisation of the reactive surface --> homogeneous reactivity AND advanced reaction state Proper minerals reactivity Real texture of the well cement RESERVOIR EXPERIMENTATION (times from 15 to 60 days) FORMATION WATER: NaCl brine (150 g/l) TOTAL PRESSURE: 500 bar TEMPERATURE: 120 C, 200 C GAS: 66mol% H 2 S + 34mol% CO 2 Complementary slides 30

31 Two degrees of carbonation and sulfidation SIMPLIFIED REACTIONS: - Carbonation - CaO.SiO 2.H 2 O + CO 2 CaCO 3 + SiO 2.H 2 O - Sulfidation - Fe + H 2 S FeS + H 2 -Ca 2 AlFeO H 2 S FeS 2 + (200 C) Mineralised gas/total gas (mol%) 100% 90% 80% 70% 60% 50% 40% 30% 20% CO 2 H 2 S Moderate carb. Sulf. ~ 10% Intense carb. Sulf. ~ 10% Intense carb. Sulf. min. 10% 0% LIQUID BIPHASIC SC Complementary slides 31

32 Schematization of the alteration profiles L L+S S Massive calcite Massive calcite Aragonite (120 C) OR Calcite (200 C) µm NO ALTERED CEMENT Hydrothermal CSH Quartz (dissolved at 200 C) Diffuse calcite Diffusion of C and Ca species Massive deposit of calcite Diffusive blockage Alteration stopping DECALCIFIED CEMENT Silica +/- Ca Quartz Traces of calcite Intrusion of the SC into the porous volume digitations Complementary slides CARBONATED CEMENT Silica +/- Ca CaCO 3 (aragonite+calcite) Quartz Total carbonation (favorised by the absence of liquid water and optimal diffusion of the SC CO 2 ) 32

33 Analogy experimental system numerical system INITIAL EXPERIMENTAL SYSTEM INITIAL NUMERICAL SYSTEM FILM BRINE CRUST CEMENT H 2 S (aq) (=exp.) CO 2(aq) (=exp.) Na, Cl (=exp.) Ca, Si (= exp.) Na, Cl (= exp.) Porosity = 1 Dwater Porosity = 1 Dwater Porosity = 0.4 (= exp.) Dwater 0.1 mm 2 mm 0.06 mm 1.5 mm Complementary slides 33

34 Modelling results WITHOUT THE DIFFUSIONAL BARRIER EFFECT WITH THE DIFFUSIONAL BARRIER EFFECT EXT. MEDIA CEMENT EXT. MEDIA CEMENT Concentration (mol/l) Name Calcite Ca/Si +inf. Foshagite 1,33 Tobermorite-11A 0,83 Gyrolite 0,66 SiO 2(am) 0, Name Calcite Ca/Si +inf. Foshagite 1,33 Tobermorite-11A 0,83 Gyrolite 0,66 SiO 2(am) 0, ,2 0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 Crust Decalcified zone X(mm) No altered cement MODELLING EXPERIMENT 0-0,2 0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 X(mm) Confirm the calcite deposit as the responsible of the stopping of the alteration that we observed experimentally Refine the model to fit the experimental and numerical front thicknesses Extrapolate this results to large spatial and time scales (well during several 1000 years) Crust Stopping at 5-10 d. Decalcified zone No altered cement Complementary slides 34