Water-rock interactions during a CO2 injection field-test: Implications on host rock dissolution and alteration effects

We investigated the nature and rates of in-situ CO2-fluid-rock reactions during an aqueous phase CO2 injection test. Two push-pull test experiments were performed at the Lamont-Doherty Earth Observatory test site (New York, USA): a non reactive control test without CO2 addition and a reactive test with CO2 equilibrated with the injected solution at a partial pressure of 1.10(5) Pa. The injected solution contained chemical and isotopic conservative tracers (NaCl and O-18) and was injected in an isolated and permeable interval at approximately 250 m depth. The injection interval was located at the contact zone between the Palisades sill (chilled dolerite) and the underlying metamorphic Newark Basin sediments and the injected solution incubated within this interval for roughly 3 weeks. Physico-chemical parameters were measured on the surface (pH, temperature, electrical conductivity) and water samples were collected for chemical (Dissolved Inorganic Carbon - DIC, major ions) as well as for isotopic (delta C-13(DIC), delta O-18) analyses. For the control test, post-injection chemical and isotopic compositions of recovered water samples display mixing between the background water and the injected solution. For the reactive CO2 test, observed delta C-13(DIC) and DIC both increase, and enrichment in Ca2+, Mg2+, K+ allow for quantification of the chemical pathways through which aqueous CO2 and subsequent H2CO3 were converted into HCO3-. Dissolution of carbonate minerals was the dominant H2CO3 neutralization process (approximate to 52 +/- 7%), followed by cation exchange and/or dissolution of silicate minerals (approximate to 45 +/- 10%, for both processes), and to a minor extent, mixing of the injected solution with the formation water (approximate to 3 +/- 1%). The results confirm the rapid dissolution kinetics of carbonate minerals compared to those of basic silicate minerals. However, our results remain marked by uncertainties due to the natural variability of the background water composition, in mass balance calculations. These experiments imply that the use of accurate DIC measurements can quantify the relative contribution Of CO2-fluid-rock reactions and evaluate the geochemical trapping potential for CO2 storage in reactive reservoir environments. (C) 2009 Elsevier B.V. All rights reserved.