Ferrofluid displacement in a simple 2.5D micromodel

• https://onepetro.org/SPEATCE/proceedings-abstract/20ATCE/3-20ATCE/D031S029R008/449861
• 10.2118/201365-MS

Abstract — Nanoparticles have great potential to mobilize trapped oil in reservoirs by reducing the oil-water interfacial tension, altering the rock wettability, stabilizing foams and emulsions, and heating the reservoir to decrease the oil viscosity. However, the direct application of magnetic forces on paramagnetic nanoparticles in reservoir engineering applications has not be extensively investigated. We demonstrate the enhanced oil recovery (EOR) potential of hydrophilic magnetic nanoparticles in oil production by direct observation using microfluidics. We studied the mobilization of oil blobs by a ferrofluid (a suspension of hydrophilic magnetic nanoparticles in water) in a converging-diverging channel with varying depth (so-called 2.5D micromodel). The channel had a varying depth of 10-30 microns and a varying width of 50-200 microns, approximating a flow path in the rock. The nanoparticle suspension was injected at 0.1 microliter/hour. The channel was made of glass and thus the water-based ferrofluid was the wetting fluid. Initial ferrofluid flooding experiments were performed under a static magnetic field. This magnetic field caused oil droplet deformation, dynamic break-up into smaller droplets, and subsequent residual oil saturation reduction. Significant oil blob displacement was observed within 2 hours after the magnetic field was applied. During the flooding, the oil saturation within the observation area of the micromodel reduced from 27.4% to 12.0%. We then hypothesized that a changing field would have an even larger effect in saturation reduction. We have thus designed experiments with a magnetic field of the same magnitude slowly rotating under the micromodel. We subsequently observed a completely different phenomenon, namely self-assembly of oil droplets, indicating formation of the hydrophilic magnetic nanoparticles microstructures (chains under the magnetic field). These magnetic nanoparticle microstructures were ever-changing under the rotating magnetic field. While the ability of ferrofluid to rotate small blobs was in itself interesting, in experiments without actual flooding (and thus synergy of hydrodynamic and magnetic forces) we did not observe any additional oil recovery.

• https://github.com/dreamlot/ComputeOilSaturationSingleChannel
• 10.5281/zenodo.4884895

Abstract — Compute the bright-phase saturation in segmented images of a single channel of two phases. The image dataset is published on Digital Rocks Portal https://www.digitalrocksportal.org/projects/367. cut.py cuts images in a directory to the wanted size. imshow.py implements cv2.imshow(). split_images.py splits the images along horizontal direction. compute_So_whole computes the oil (bright phase) saturation in the unsplitted image. It detects the upper and lower boundary of the flow channel and counts the number of bright pixels in between. compute_So_whole computes the oil (bright phase) saturation in the splitted image. It detects the upper and lower boundary of the flow channel and counts the number of bright pixels in between.

• https://doi.org/10.2118/201365-PA
• 10.2118/201365-PA

Abstract — Nanoparticles have great potential to mobilize trapped oil in reservoirs because of their chemical, thermal, and interfacial properties. However, the direct application of magnetic forces on superparamagnetic nanoparticles in reservoir engineering applications has not been extensively investigated. We demonstrate the enhanced oil recovery (EOR) potential of hydrophilic superparamagnetic nanoparticles in oil production by direct observation using microfluidics. We studied the mobilization of oil blobs by a ferrofluid (a suspension of hydrophilic superparamagnetic nanoparticles in water) both in a converging/diverging micromodel channel and in a foot-long pore network micromodel, both with varying depth (so-called 2.5D micromodels). The water-based ferrofluid in all cases was the wetting fluid. Initial ferrofluid flooding experiments in single channels were performed without and then with a static magnetic field. This magnetic field caused oil droplet deformation, dynamic breakup into smaller droplets, and subsequent residual oil saturation reduction. During the flooding, after the magnetic field was applied, significant oil displacement was observed within 2 hours [6 pore volumes injected (PVI)], and 86.2% of the oil that was not mobilized without a magnetic field was mobilized within 64 hours (192 PVI). Then, in experiments in the micromodel and in a Hele-Shaw cell without flooding, we observed self-assembly of oil droplets, indicating the formation of the hydrophilic magnetic nanoparticle microstructures (chains under the magnetic field) and their interaction with the oil blobs. Further ferrofluid flooding experiments were performed in a foot-long micromodel under a rotating magnetic field. The oil saturation was reduced from 44.6 to 33.3% after 17 hours (8.5 PVI) of ferrofluid flooding after the rotating magnetic field was applied. Finally, a discussion of field application of ferrofluid flooding is presented.