Pore-scale investigation of residual oil displacement in surfactant–polymer flooding using nuclear magnetic resonance experiments

Pet. Sci. Pore-scale investigation of residual oil displacement in surfactant- polymer flooding using nuclear magnetic resonance experiments Zhe-Yu Liu 0 1 Yi-Qiang Li 0 1 Ming-Hui Cui 0 1 Fu-Yong Wang 0 1 A. G. Prasiddhianti 0 1 0 Edited by Yan-Hua Sun 1 EOR Research Institute, China University of Petroleum , Beijing 102249 , China Research on the Gangxi III area in the Dagang Oilfield shows that there was still a significant amount of oil remaining in oil reservoirs after many years of polymer flooding. This is a potential target for enhanced oil recovery (EOR). Surfactant-polymer (SP) flooding is an effective chemical EOR method for mobilizing residual oil and improving displacement efficiency macroscopically, but the microscopic oil displacement efficiency in pores of different sizes is unclear. Nuclear magnetic resonance (NMR) is an efficient method for quantifying oil saturation in the rock matrix and analyzing pore structures. In this paper, the threshold values of different pore sizes were established from the relationship between mercury injection curves and NMR T2 spectrums. The distribution and migration of residual oil in different flooding processes was evaluated by quantitatively analyzing the change of the relaxation time. The oil displaced from pores of different sizes after the water flood, polymer flood, and the SP flood was calculated, respectively. Experimental results indicate that (1) the residual oil in medium pores contributed the most to the incremental oil recovery for the SP flood, ranging from 40 % to 49 %, and small pores usually contributed \30 %; (2) the residual oil after the SP flood was mainly distributed in small and medium pores; the residual oil in medium pores accounted for 47.3 %-54.7 %, while that trapped in small pores was 25.7 %-42.5 %. The residual oil in small and medium pores was the main target for EOR after the SP flood in oilfields. Nuclear magnetic resonance (NMR); Surfactant-polymer (SP) flood; Residual oil distribution; Displacement mechanism; Core displacement test 1 Introduction After over 30 years of water flooding and polymer flooding, the Dagang Oilfield is now a mature oil field with its water cut reaching the economic limit. However, as much as 70 % of the original oil in place (OOIP) may remain in the reservoir after secondary recovery processes (Sorbie 1991) . A large portion of the residual oil is capillary trapped (Lake 1989) . To tackle the residual oil saturation and revitalize this reservoir, a tertiary recovery is required. Surfactant–polymer (SP) flooding has been proved to be an efficient tertiary method for most major oilfields in China. To apply SP flooding in the Dagang Oilfield and optimize the process in heterogeneous reservoirs, it is necessary to predict the residual oil after the SP flood and identify the displacement efficiency in pores of different sizes. Several experimental/numerical techniques have been proposed to measure or predict the residual oil distribution after displacement processes. For example, a widely used traditional method is to measure residual oil through analysis of cast thin sections of a reservoir core (Zao et al. 2009) . This method damages reservoir cores while obtaining slices. Furthermore, fractured cores, unconsolidated sands, and mud cannot be cut into slices using this method. Another experimental technique is to use a microvisualized model instead of a reservoir core to simulate a displacement process as well as the distribution of residual oil (Wang et al. 2010). However, this method does not take into account the influence of interstitial matter on the distribution of residual oil. X-ray computed tomography is often used to detect the rock matrix, but it is not sensitive to fluid changes (Vinegar 1986; de Argandona et al. 1999; Liu 2013) . Numerical simulations require some assumptions in order to achieve mathematical completeness. In addition, reservoir parameters are uncertain and hard to determine. Therefore, there is usually a discrepancy between simulation results and actual conditions (Li et al. 2006) . Nuclear magnetic resonance (NMR) is a quick, accurate, non-destructive, and widely used technology for core testing (Kleinberg and Vinegar 1996; Xie and Xiao 2007; Zhao et al. 2011) . In NMR measurements, the received signals originate only from fluids in pores. To differentiate hydrocarbons from brine, brine is doped with paramagnetic ions to shield the signals from water, so that the signals only come from the oil. NMR T2 relaxation time represents the fluid content in the pores of different sizes. The longer T2 relaxation time corresponds to the larger pores, and vice versa (Williams et al. 1991; Gleeson et al. 1993; Cowan 1997) . The residual oil distributions in pores of different sizes are quantified through the T2 distribution analysis, and the accurate oil saturation can be calculated to investigate the oil movement in pores of different sizes. In our study, the NMR (...truncated)