1. 1 Modelling of Dispersion from Direct Injection of Carbon Dioxide in the Water Column Baixin Chen and Makoto Akai National Institute of Advanced Industrial Science and Technology (AIST), Japan

2. 2 Turbulent Multiphase Mixing and Interactions: (Mass, Momentum,Energy Exchanging and Phase Changing) Droplet- seawater interactions: drag, deformation, raising Droplets interaction (collision, coalescence, second breakup) CO 2 dissolution or shrinking CO 2 hydrate dynamics; gasification Local turbulent flow, wake, and mixing Chemical reactions of dissolved CO 2 and seawater Biological Impacts CO 2 injection : Droplets formation; Hydrate;Distributions of Initial Diameter and Number Density; Towering pipe wakes….. Mesoscale Eddies Ocean Currents Bottom Boundary Layer 100~1000 m Ocean Surface Small-scale ocean turbulence and turbulent wakes Two-fluid modeling Biological Impact Modeling CO 2 10 ~100 Km Horizontal 2-D modeling of CO2 dispersion 2000m2000m What must be handled!

3. 3 Models developed Alendal et al. (NERSC Technical Report,1998; JGR-ocean, 2002) Sato et al. (RITE report, 1998;ASME,2000; GHGT-6, 2002) Chen et al.(RITE report, 1999; ASME,2000; Tellus, 2003)

4. 4 Outline Introduction of the model developed Case investigation: Release of CO2 from fixed port Release of CO2 from a towered pipe Conclusions

5. 5 A Near-field Physical & biological impact model of CO 2 Ocean Sequestration 1. Modeling of Small scale ocean turbulent flow (Re- construction) Forced-dissipative ocean turbulent flow model CO 2 enrich-seawater dynamic model 2. Modeling of momentum and mass transfer between CO 2 droplets and seawater Sub-model of CO 2 droplet drag coefficient Sub-model of CO 2 droplet deformation Sub-model of CO 2 solubility Supported by Lab. and field Exp. 3. Modeling of biological impacts of floating-orgms. Conservative variables: Mass or Number density of organism k Non-conservative variables: Degree of Damage/Activity Index, A k. Sub-models of Damage Degree /Activity Index

6. 6 Part – I : Reconstruction of small scale turbulent ocean with basis on observation data Theories and physical model Observation data analysis and implement

7. 7 Turbulent kinetic energy spectrum in the ocean (J. D. Wood in Nature 1985 and CREIPI at Keahole Pt. Offing, 1999) Eddy resolving truncation scale (10 km) by Earth Simulator (0.1deg.) in Japan Eddy resolving truncation scale (1 m) by Small-scale two-phase model Small-scale two-phase model Eddy resolving truncation scale (100km) by year 2000 estimated by Wood in 1985 Forced-dissipative and kinetic energy cascade theories applied? N-S B-230m Frequency (CPs) Horizontal Vertical CREIPI at Keahole Pt. Offing, 1999 Meso-scale ocean model

8. 8 Theories and Techniques Applied Inside of the ocean: N-S based 3-D unsteady Governing Eq.s Forced-dissipative Energy cascade theories Adjusted by observation spectrum Large-scale information from Boundaries : Mean properties (X,t) Turbulent properties at K > K f Field Obs. Data Data analysis

9. 9 a. Forced-dissipative system of small-scale ocean: Forced term: Dissipative term: ik k t S0.2D b. Structure-function Turbulent viscosity model: )]x,x(F[xC.)xx( kk k k / k k,k k t 2 23 150 2 2 ))()((25.0 kkkkk k xxuxuF Governing Equations for simulating small-scale ocean turbulence

10. 10 1-4. Example: Hawaiian Case (small- scale): Computational domain, initial &boundary conditions X1 = 500 m; N 1 = 256 X 2 = 300m; N 2 = 128 X 3 = 300 m; N 3 =128; Periodic conditions Inlet output solid wall ρ 0 = f(T 0,S 0 ) U 1m (x 3 ), T 0 (x 3 ) S 0 (x 3 )are the observation data

11. 11 Simulations of small-scale ocean turbulence Instantaneous velocities and temperature T

12. 12 Part-II: CO 2 droplet dynamics Experimental Observations and Modeling Assumptions Assumption:CO 2 droplet with hydrate covered is a Deformable rigid droplet with Permeable Surface Experimental data adopted are those from Stewart(1970) and Kimuro (1994) for CO 2 solubility, and from Ohgaki (1993) for phase diagram. Experiment data dealing with momentum transfer between droplets and seawater are from the experiments of Dr. Ozaki (1999)

13. 13 Sub-models of drag coefficient and terminal velocities

14. 14 Model prediction of an individual droplet dissolution (model calibration) (CO 2 droplet Diameter vs Exp data by P. Brewer et al) Key Parameters: C d : Drag coefficient Sh: Sherwood Number C s : The solubility α : The effective area coefficient

15. 15 CO 2 droplet dissolution at variant depth

16. 16 Ascending /Descending of CO 2 droplet

17. 17 Governing Equations of Seawater Governing Equations of LCO 2 Two-fluid ocean turbulent flow model

18. 18 Density change of CO 2 enriched seawater s = w (1.0 + ) s : CO 2 solution density w : seawater density : CO 2 mass fraction =0.273 by Exp (Song et al. 2003)

19. 19 Part – III : Dispersion from Direct Injection of Carbon Dioxide in the Water Column Injection of CO 2 from fixed ports Injection of CO 2 from towered pipe /sec

20. 20 Dispersion from a fixed port release T=32 min T=93 min CO2 droplet plume CO 2 enriched water plume

21. 21 Plume characters from a fixed port T=93 min

22. 22 Lower Injection rate (pH plume at middle depth) Mc=0.6kg/s; D 0 =8.0mm T=100.3 min Mc=0.1kg/s; D 0 =8.0mm

23. 23 Dispersion from a towered pipe release T=1.0 min T=23 min T=70 min X=180 m X=10 m T=70 min Yc=0.01kg m -3

24. 24 Towered pipe Fixed ports Statistical characteristics of CO 2 enriched seawater plume

25. 25 Statistical characteristics of CO2 enriched seawater plume

26. 26 Conclusions Near-filed physical and chemical process created by directly injected LCO 2 into the ocean waters could be reasonably simulated. To engineering application, injection of LCO 2 from fixed ports should be carefully arranged to limit the local injection rate associated with the selection of an incline seafloor. In case of large injection rate (100kg/s) from fixed port on a flat seafloor, injected LCO 2 could yield a large pH change and an unsteady waving double-plume. Alternatively, release of LCO 2 from a towered pipe at middle-depth with a relatively large size droplets is an expectable option to practically performance of CO 2 ocean sequestration, which could be adjusted with the limitation of biological impact. Understanding of the effect of dissolved CO 2 on oceanic bio-organisms appeared to be urgently necessary for assessing the oceanic environmental impacts. …. We still have more works to be done.

27. 27 Acknowledgements This study is a part of the investigation of two projects: A research Project on Accounting Rules on CO 2 Sequestration for National GHG Inventories (ARCS) managed by National Institute of Advanced Industrial Science and Technology (AIST) and The CO 2 Ocean Sequestration Project managed by Research Institute of Innovative Technology for the Earth (RITE). New Energy and Industrial Technology Development Organization (NEDO), Japan, fund both projects.