Journal of Porous Media, 13(7), 591–599 (2010) NONWETTING PHASE RESIDUAL SATURATION IN SAND PACKS Paul Gittins, Stefan Iglauer, Christopher H. Pentland, Saleh Al-Mansoori, Saif Al-Sayari, Branko Bijeljic, * & Martin J. Blunt Department of Earth Science and Engineering, Imperial College London, Prince Consort Road, London SW7 2BP, UK * Address all correspondence to Branko Bijeljic E-mail: b.bijeljic@imperial.ac.uk Original Manuscript Submitted: 9/17/09; Final Draft Received: 11/24/09 We measure residual nonwetting phase saturation in six unconsolidated sands of different average grain sizes. We also analyze the pore structure using three-dimensional images from which topologically equivalent pore networks are extracted. The residual saturations range from 10.8% to 13.1%, which is lower than for most consolidated media. Higher porosity is associated with lower residual saturations, while there is little correlation between grain and pore shape and the degree of trapping. We also study layered packs: the residual saturations are reduced compared to comparable homogeneous systems. We discuss the results in the context of capillary trapping during carbon storage in aquifers. KEY WORDS: capillary trapping, residual saturation, carbon storage, two-phase flow, sand packs, porous media 1. INTRODUCTION It is now widely accepted that increased levels of CO 2 and other greenhouse gases (GHG) in the atmosphere will cause a rise in global temperature that, if uncontrolled, could cause considerable damage to society (Stern, 2007). Carbon capture and storage, where CO 2 is collected from point sources of emissions, such as power stations, and in- jected deep underground in geological formations, is set to play a vital role in climate change mitigation (Intergov- ernmental Panel on Climate Change, 2005). The principal public concern is that the injected CO 2 remains under- ground for hundreds or thousands of years. Capillary trapping, where the CO 2 is stored safely as pore-space bubbles in the rock, has been proposed as a rapid and effective way to render the injected fluid im- mobile (Ennis-King and Paterson, 2005; Kumar et al., 2005; Juanes et al., 2006; Obi and Blunt, 2006; Ide et al., 2007). This trapping occurs when the CO 2 is displaced by water—this process occurs as the CO 2 rises upward under buoyancy forces in an aquifer, during natural groundwater flow, or when CO 2 is injected with water to enhance the trapping process. This last concept, injecting water in ad- dition to CO 2 , may allow all the CO 2 to be trapped after only a few years of water injection (Juanes et al., 2006; Qi et al., 2009): over time, it may slowly dissolve or re- act with the host rock, but it is safely stored. This may significantly increase the storage security and capacity of saline aquifers; any impermeable cap rock provides only a secondary containment. While CO 2 would generally be injected as a supercrit- ical phase, implying injection depths of around 800 m or more, injecting much deeper is likely to be expen- sive. Hence CO 2 trapping may occur in unconsolidated or poorly consolidated formations. Recent simulation stud- ies have highlighted how the effectiveness of the trapping strategy hinges on estimates of the residual or trapped sat- urations (Qi et al., 2009). Many researchers have measured trapping in both con- solidated cores (Geffen et al., 1952; Crowell and Dean, 1966; Land, 1971; Ma and Youngren, 1994; Jerauld, 1997; Kleppe et al., 1997; Kralik et al., 2000; Suzanne et 1091–028X/10/$35.00 c  2010 by Begell House, Inc. 591 Begell House Inc., http://begellhouse.com Downloaded 2010-10-21 from IP 155.198.96.8 by Dr. Branko Bijeljic (branko)