نبذة مختصرة : Low salinity waterflooding is a promising enhanced oil recovery technique that has been observed to increase recovery by up to 14% in experiments over a range of scales. However, results vary dramatically and there is currently no way of predicting which systems will respond well to low salinity waterflooding. This is partly due to an absence of mechanistic explanations linking fluid-solid interactions at the microscopic scale to oil recovery at core and field scales. Pore scale observations of low salinity flooding are crucial to bridging this gap in understanding and improving predictive capabilities. In this work, we utilise X-ray micro-CT imaging and advances in image analysis techniques to make comprehensive pore scale observations of low salinity waterflooding experiments across four sandstone samples of varied pore structures. We first analyse the impact of initial wetting state on recovery during low salinity flooding by comparing observations in two Berea sandstone cores of different initial wetting states. The wetting state of one core is altered with exposure to crude oil. The second core is not altered so that it is initially water-wet. We characterise the wetting state of both samples using imagery of fluid-solid fractional wetting and pore occupancy analysis. In the unaltered rock, oil saturation, fractional mineral surface area covered by oil and size distribution of oil filled pores remain constant between high salinity flooding and subsequent low salinity floods. In contrast, in the altered sample, we observe a shift in the mineral surface area covered by oil after low salinity flooding towards decreasing coverage, consistent with a change to a less oil-wetting state. This change is also reflected in the observed variation in fluid pore occupancy. The wetting alteration results in the redistribution of 22% of oil within the rock, but an additional recovery of just three percentage points. We hypothesise that the success of low salinity water flooding depends on both a wetting alteration and a pore structure which facilitates the production of mobilised oil. We investigate the impact of pore structure on recovery during low salinity flooding by comparing the pore scale observations in the aforementioned altered Berea sandstone sample with observations in altered Bunter sandstone and Castlegate sandstone samples. This comprises the first systematic comparison of the pore scale response to low salinity flooding across multiple sandstone samples. We observe fluid saturations and characterise the wetting state of samples using imagery of fluid-solid fractional wetting and pore occupancy analysis. In the Berea sample, we observed an additional oil production of 3 percentage points during low salinity water flooding, with large volumes of oil displaced from small pores but also re-trapping of mobilised oil in large pores. In the Bunter sandstone, we observe four percentage points of additional recovery with significant displacement of oil from small pores and no significant retrapping of oil in large pores. In the Castlegate sample, we observed just 1 percentage point of additional recovery and relatively small volumes of oil mobilisation. We observe a significant wettability alteration towards more water-wet conditions in the Berea and Bunter sandstones, but no significant alteration in the Castlegate sample. We show evidence that the pore structure, specifically the connectivity of the largest pores in each sample, affected recovery. This work gives the first pore scale insights into the role of pore geometry and topology on the mobilisation and retrapping of oil during low salinity waterflooding. We assess the role of mineralogy on low salinity response. The presence of clay minerals is generally accepted to be a necessary condition for favourable response to low salinity flooding in sandstones. However, there have been no in situ pore scale observations of oil detachment from clay minerals during low salinity water flooding. Furthermore, there is a growing body of evidence that suggests that significant oil detachment can occur on quartz and feldspar mineral surfaces during low salinity flooding. We further analyse the dataset described above to make the first pore scale in situ observations of clay behaviour during low salinity flooding in oil-wetting sandstone samples. We find that clay minerals are initially significantly more oil-wetting than other mineral groups in all samples. We observe significant wettability alteration for clay minerals and non-clay minerals in both the Berea and Bunter samples. No significant wettability alteration is observed in either mineral group in the Castlegate sample. We observe pore occupancy changes consistent with wettability alteration towards more water-wetting conditions in clay-poor pores in all samples. In all samples, clay-poor pores contribute significantly more to total oil production during low salinity waterflooding than clay-rich pores. Clay-rich pores are found to account for only 15%, 5% and 3% of total additional recovery in the Berea, Bunter, and Castlegate samples respectively. We see no evidence that clays dominate the response to low salinity flooding in any of the samples. This work adds to the growing body of evidence that questions the presence of clays as a necessary condition for a favourable response to low salinity flooding in sandstones. This work gives unique insights into low salinity water flooding through a more systematic and controlled set of pore-network scale observations and analyses. We have shown that initial wetting state strongly controls behaviour during low salinity waterflooding, with more oil mobilisation and production associated with more oil-wetting initial conditions. Pore structure is also shown to impact oil mobilisation, with well connected large pores associated with lower volumes of mobilised oil during low salinity flooding. Lastly, we show that clay-poor pores contribute more to additional recovery during low salinity flooding than clay-rich pores.
Processing Request
No Comments.