Evaluation of Groundwater Quality Changes in Response to CO 2 Leakage from Deep Geological Storage

Transcription

1 Evaluation of Groundwater Quality Changes in Response to CO 2 Leakage from Deep Geological Storage 1 Liange Zheng, 1 Jens T. Birkholzer, 1 John A. Apps, 2 Yousif Kharaka, 2 James Thordsen 1 Lawrence Berkeley National Lab, Berkeley, CA; 2 USGS, Menlo Park, CA Water/Energy Sustainability Symposium Salt Lake City, Utah September 13-16,

2 CO 2 Leaks? There are several key technical risk features, events and processes for geologic CO 2 storage and these include: The release of CO2 from its storage location (e.g., via wellbores, faults or fractures, etc.)... ( With appropriate site selection informed by available subsurface information, a monitoring program to detect problems, a regulatory system, and the appropriate use of remediation methods to stop or control CO 2 releases if they arise, the local health, safety and environment risks of geological storage would be comparable to risks of current activities i i such as natural gas storage, EOR, and deep underground disposal of acid gas. (Sally Benson, IPCC Special Report on Carbon Dioxide Capture and Storage, 2005)

3 Water Quality Effects from CO 2 Leakage Possibility of leakage pathways such as faults or wells CO 2 leakage into aquifers may cause mobilization of hazardous trace elements Leakage of CO 2 or brine migration may cause migration of other contaminants (e.g., organics, co-injectants) t into aquifers

4 Selected Work on Groundwater Quality Impacts Wang and Jaffe, Energy Conversion and Management, 45, 2004 Simulation of CO 2 intrusion into shallow groundwater shows increase in lead concentrations, for very simplified host rock mineralogy. Kharaka et al., Geology, 34, 2006 Strong increases in trace metal concentrations following CO 2 injection in a deep storage formation at Frio. Lewicki et al., Environmental Geology, 52, 2007 Natural analogs show acidification of groundwater and changes in chemical composition, but waters remain potable in most cases. McGrath et al., Ground Water Monitoring & Remediation, 27, 2007 Increase in cadmium concentrations in shallow groundwater (above drinking water limits), related to CO 2 releases from a municipal landfill. Smyth et al., Proceedings GHGT-9, 2008 Increases in cation concentrations measured in laboratory batch experiments of diverse aquifer rocks exposed to CO 2 -water mix. Carroll et al., 8 th CCS annual conference, Increase in concentrations of various cations in batch experiments when aquifer rocks is exposed to CO2-charged water.

5 Relevant Research at LBNL Systematic prediction of CO 2 -related mobilization of hazardous trace elements in groundwater (finalized) Field experiment with shallow CO 2 release and measurements of geochemical changes (ongoing ZERT project, funded by EPRI) Potential impact of organics and co-injectants on groundwater (ongoing project) Presented in this talk

6 CO 2 -Related Mobilization of Hazardous Constituents in Groundwater Generic study by reactive transport modeling y Confined Shallow Aquifer at 50 m Depth 500 m Groundwater flow with 10 m per year z Eh = V, ph = 7.6, P CO2 = bar Trace amounts of galena and arsenopyrite controlling lead and arsenic, respectively 200 m 10 m x Base case mineralogy representative of a mildly impure arenite (North Atlantic Coastal Plain Sandstone) 105 m Gaseous CO2 2 intrusion at at a a rate rate of of 5x kg/s -5 (1.58 kg/s tonnes/yr) (2.36 tonnes/yr) for 100 for year 100 simulation year simulation period period

7 Gaseous and Dissolved CO 2 (100 yrs) Intrusion Rate: 2.4 t/yr Z Intrusion Rate: 19 t/yr Z Y Y X X X Gas Saturation (Close-up) 20 Y Z Z SG X 200 Gas Saturation Y -5 Z Z SG Y Y X X Total Dissolved Carbon Y -5 Z X TIC X Total Dissolved Carbon Y -5 Z X TIC

8 Base Case with 2.4 t/yr ph ph Distance (m) Relative Change of Total Pb Sorption Sites sorption sites chan nge (%) 20 5 yr yr 50 yr 100 yr -40 Total Pb yr Distance (m) Lead Concentration Arsenic Concentration us Pb concentratio on (mol/l) Total aqueo 8x10-8 6x10-8 4x10-8 2x10-8 0x10 0 MCL 15 ppb 5 yr 20 yr 50 yr 100 yr on (mol/l) Total aqueo ous As concentrati 1.4x x x x x10-8 MCL 10 ppb 5 yr 20 yr 50 yr 100 yr Distance (m) Distance (m)

9 Different Aquifer Mineralogy Base case with Coastal Plain Sandstone: aquifer contains enough carbonate minerals to buffer ph. Sensitivity case with St Peter Sandstone: carbonate minerals in the aquifer are depleted and ph buffering capacity diminishes. Total aqueous As concentration (mol/l) ( Arsenic Concentration at 100 Years 4.0x x x x10 0 MCL 100 yr Coastal Plain Sandstone 100 yr St Peter Sandstone Distance (m) concentration (m mol/l) To otal aqueous Pb 8.0x x x x x10 0 Lead Concentration at 100 Years MCL 100 yr Coastal Plain Sandstone 100 yr St Peter Sandstone Distance (m)

10 Conclusions from Generic Modeling Study While substantial increases in aqueous concentrations are predicted to occur compared to the initial water composition, the maximum contaminant level for arsenic in groundwater has been exceeded in only a few simulation cases, whereas the maximum contaminant level for lead has not been exceeded at all Adsorption/desorption via surface complexation is arguably the most important process controlling the fate of hazardous constituents mobilized by CO 2 leakage. The relative importance of dissolution/precipitation versus adsorption/desorption is controlled by many factors, including adsorption parameters and aquifer mineralogy, reaction kinetics, aqueous complexation processes, and mineral solubility constants The rate of CO 2 entering an aquifer has small minor effect on the maximum contaminant concentrations. Excess CO 2 remains in the gas phase and migrates elsewhere, thereby changing the spatial distribution of possible groundwater contamination, but the contamination level increasing only slightly. Site specific studies are recommended, with model predictions and supplementary laboratory and/or field experiments.

11 Field Experiment at ZERT Shallow Release Facility ZERT Field Site (Bozeman, Montana) Facility Goals, Rationale, and Design Develop a well characterized site Apply known CO 2 injection rates for testing near-surface monitoring Use this site to establish detection limits for monitoring technologies Use this site to improve flow and transport models ~80 m ~2.8 m Slotted stainless pipe with internal CO 2 pipe & packer system for even gas distribution Activities i i to Date 2006 Characterization, vertical-well injections, horizontal well installation 2007 Year 1 Shallow-release Phase kg/day for 10 days Phase kg/day for 7 days 2008 Year 2 Shallow-release Phase kg/day for 30 days (Courtesy of Lee Spangler, MSU)

12 Wells for Groundwater Characterization Wells location Water well headspace CO CO 2 2 Concentrations in Head Space Above Wells 100 O2 vol % CO B 2B 3B 4B 5B 20 (Courtesy of Lee Spangler, MSU) 0 7-Jul 8-Jul 9-Jul 10-Jul 11-Jul 12-Jul 13-Jul 14-Jul 15-Jul 16-Jul 17-Jul date

13 Groundwater Monitoring Chemical Analyses of Groundwaters by USGS 62 complete chemical analyses between 7 July 2008 and 13 August chemical constituents, T, EC, ph, TDS, Total Cations, Total Anions Hazardous constituents include: As, Ba, B, Cd, Cr, Cu, F, Pb, Se, U 7.5 ZERT - "B" wells - water samples 1E-04 well 1B 1E-05 well 2B Trace metals ph CO 2 start CO 2 stop /07 07/10 07/13 07/16 07/19 07/22 07/25 07/28 07/31 08/03 08/06 08/09 08/12 08/15 Trace metal (mo ol/l) well 4B 1E-06 well 5B 1E-07 CO2 start CO2 stop 1E-08 7/18 rain 1E cm 7/19 rain cm 1E-10 8/7 rain.56cm 8/8 rain 1E cm ph Ba138 Co59 Cu65 Zn66 Cd113 Pb207 Concentrations of trace metals are below the Maximum Contaminant Levels

14 Geochemical Process Modeling Cation exchange could explain the evolution of trace metals. 8.E-10 7.E-09 Pb (mol/l) 7.E-10 6.E-10 5.E-10 4.E-10 3.E-10 2.E-10 1.E-10 Pb measured data model results Cd (mol/l) 6.E-09 5.E-09 4.E-09 3.E-09 2.E-09 1.E-09 Cd measured data model results 0.E+00 0.E ph ph Adsorption/desorption could explain the evolution of anions H2AsO4-2 (mol/l) 3.0E E E E E E E H -2 2 AsO 4 measured data model results ph HSeO3 - (mol/l) 1.8E E E E E E E-08 40E E E E Increase measured data HSeO 3 model results ph Decrease

15 Concluding from Field Experiment The release of CO 2 into shallow groundwater leads to a fast and systematic drop of ph and consequently an increase of the concentration of trace cations and anions. Calcite dissolution could be the main ph buffer process. The increase in the concentrations of major cations and trace metals could be explained by Ca +2 -driven exchange reactions. The release of anions from sorption site due to competing adsorption of carbonate could explain the concentration trend of most anions. Further characterization of soil mineralogy is ongoing, which will be followed by more precise geochemical interpretation and modeling. Findings may not be representative of most groundwater resources because a very shallow system is considered, strongly affected by atmospheric precipitation.

16 Acknowledgments The first study was funded by US Environmental Protection Agency, Office of Water and Office of Air and Radiation, under an Interagency Agreement with the U.S. Department of Energy at the Lawrence Berkeley National Laboratory, Contract No. DE-AC02-05CH This second work was funded by the Electric Power Research Institute, EPRI. We thank the entire ZERT team and all participating organizations for a supportive and exciting research environment Thank you for your attention