H41C-1321
Interpreting detailed brine chemistry changes during early periods of in-zone CO2 storage at Cranfield site, Mississippi, USA

Thursday, 17 December 2015
Poster Hall (Moscone South)
Akand Islam, University of Texas at Austin, Austin, TX, United States and Alexander Y Sun, Univ. of Texas, Austin, United States
Abstract:
Geochemical reactions can play important role on the long-term geological storage of CO2 in sites where the target formations have reactive minerals. Although the use of batch models (experimental or theoretical) is expedient, it leaves questions about how to interpret the results from the context of field scale injection. The goal of this study is to investigate changes in fluid compositions using a detailed reactive transport model. Most published CO2 geochemical studies tend to consider only a small number of components because of expensive calculations and therefore simultaneous mobility of large number of heavy metals is not clearly known. In this study we present results of coupled multiphase, multicomponent reactive transport simulations of Cranfield site, Mississipi, USA at relatively fine scale, which are obtained using the parallel PFLOTRAN code. The geochemical system consists of 22 primary or basis species, in-situ CO2 and O2 gaseous components, and 5 minerals. The number of secondary elements is 37, representing very simple to complex mineralizations occurred simultaneously in saline formation (1.81 molality). The fluid chemical compositions were measured from production fluids and mineral composition of the formation was obtained from XRD analysis of core samples. The results show brine chemistry changes in the reservoir and shed insights on the need to monitor the mobility of heavy metals such as Mg, Ca, Al, Mn, Fe, Cu, Zn, Sr, Ba, and Cd. The study provides simultaneous potential mobile inventory of these metals in the storage formations and warns possible risk through leakage into overlying zone. From storage point of view we also aim to observe the sensitivity of aforementioned constituents. Our results show pH drop from 6.91 to 3.5 and relatively small changes in HCO3- and Fe concentrations. However aqueous Ca and Al increase by orders of magnitude. The detailed geochemical effect shows trapping efficiency increased by few percent. The brine chemistry is affected within ~500 sq. meters of CO2 source during injection period.