Skip to main content
Log in

The Effect of Wettability on Waterflood Oil Recovery in Carbonate Rock Samples: A Systematic Multi-scale Experimental Investigation

  • Published:
Transport in Porous Media Aims and scope Submit manuscript

Abstract

The effect of wettability on waterflooding oil recovery and the associated pore-scale displacement mechanisms are systematically investigated during flow processes in limestone core samples with a broad spectrum of wettability. Using a miniature core-flooding setup integrated with micro-computed tomography (CT) imaging apparatus, we, for the first time, characterize in situ equilibrium wettability states and demonstrate capillary interactions of the flowing phases in strongly water-wet (SWW), intermediate-wet (IW), weakly oil-wet (WOW), and strongly oil-wet (SOW) systems. The microscale observations were then employed to explain the recovery results obtained from replicate, macroscale experiments. The waterflooding parameters, such as experimental temperature, brine composition, defending and invading phase properties, and the initial water saturation, were maintained nearly identical in all of the waterflooding tests, thereby ensuring that the wettability state was the only factor that controlled the variations in waterflood oil recoveries. We provide pore-scale evidences of various pore-scale displacement mechanisms and the consequent fluid configurations in systems with different wettability states. These findings are then linked to the recovery trends of macroscale experiments. The production from IW, WOW, and SOW cases portrayed a prolonged oil recovery owing to the gradual invasion of brine into small and intermediate-sized oil-wet pores. The IW case showed the highest oil recovery among all cases. Moreover, when the injection flow rate was increased, the oil recovery was gradually increased for the IW, WOW, and SOW systems, whereas no significant additional production was observed in the case of SWW.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14

Similar content being viewed by others

References

Download references

Acknowledgements

We gratefully acknowledge the financial support of Thermo Fisher Scientific, Hess Corporation, and the School of Energy Resources at the University of Wyoming. We also extend our gratitude to Uche Igwe, Binjuin Zhang, and Omar Elmasry of Piri Research Group at the Center of Innovation for Flow through Porous Media of the University of Wyoming for their assistance with the experimental setups and SEM image acquisition.

Author information

Authors and Affiliations

Authors

Contributions

GKE carried out conceptualization, methodology, investigation, data curation, formal analysis, writing the original draft, reviewing, and editing. MK took part in conceptualization, methodology, investigation, validation, writing reviewing, and editing. WK contributed to investigation, visualization, writing, reviewing, and editing. MP and SM were involved in conceptualization, supervision, writing, reviewing, and editing.

Corresponding author

Correspondence to Gerald K. Ekechukwu.

Ethics declarations

Conflict of interest

The authors declare no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

More details of the experimental procedure are given in the following paragraphs:

  1. 1.

    Initial preparation To initiate each experiment, each core sample was loaded inside a core holder and then vacuumed to remove bulk air from the pore space. Afterward, CO2 was injected into the core to displace the remaining air and the sample was vacuumed once more for at least 12 h to evacuate bulk CO2. At this stage, the middle section of each of the miniature core samples was imaged at a resolution of 2.7 µm (voxel size of 2.7 × 2.7 × 2.7 µm3) to obtain a dry reference scan and generate the pore map, which would be later employed to analyze images of the flooded sample (i.e., wet images). This field of view (FOV) was deliberately selected to ensure that our results and subsequent analyses were not affected by capillary end effects. This is in accordance with SCAL best practices as recommended by several researchers (Alizadeh et al. 2014; Kamath et al. 2001; Lin et al. 2019; Wang and Masalmeh 2019). Once a high-quality reference image was obtained, the core was vacuum-saturated with the aqueous phase. The pore pressure was then increased in a stepwise manner while the overburden pressure was adjusted accordingly to maintain a constant net radial confining pressure of 2.07 MPa on the sleeve surrounding the core sample. Several pore volumes of brine were injected to displace trace amounts of CO2 remaining in the pore space at the elevated pore pressure of 5.52 MPa (R.Y. Makhnenko; J.F. Labuz). Afterward, the absolute permeability to brine was measured by injecting water at multiple flow rates and recording the differential pressure data. The temperature of the system was subsequently increased to the target experimental temperature (see Table 3) while the injection of brine was continued at a very low flow rate for an extended period of time to allow for the ionic equilibration between the aqueous ions and the rock minerals.

  2. 2.

    Establishing initial water saturation Each core sample was subjected to a drainage process with the designated oil phase to establish an initial water saturation (Swi) of approximately 0.3. The maximum oil injection flow rates were 0.3 and 12 cm3/min for the micro- and macroscale experiments, respectively. Both selected flow rates corresponded to capillary numbers (Nca) in the order of E−05. The capillary number formula (Eq. 1) used throughout this study is that given by Melrose et al. (Melrose 1974):

    $$N_{{{\text{ca}}}} = \frac{{\mu_{i} u_{i} }}{{\emptyset \sigma_{ow} }} = \frac{K\Delta p}{{\emptyset \sigma_{ow} L}}$$
    (1)

    where \(\mu_{i}\) and \(u_{i}\) are the viscosity and Darcy velocity of the invading phase, \(\emptyset\) is the porosity of the porous medium, \(\sigma_{ow}\) denotes the interfacial tension between the invading and defending fluids, \(\Delta p\) is the pressure drop, and L is the length of the core.

    Once the flow reached a stable (steady-state) condition, the injection was halted to allow the system to equilibrate, and the sample was then imaged at a resolution of 2.7 µm to obtain the fluid occupancy map. The steady-state condition was assumed to be established after the differential pressure along the length of the core and pore fluid configuration after consecutive monitoring scans became stable. The above-mentioned procedure was executed whenever the sample needed to be imaged during the aging and waterflooding stages of the microscale experiments. For the macroscale experiments, fluid production and pressure drop data were used to determine the steady-state conditions. In Experiments G to J, the flow direction was reversed at the end of the drainage processes to establish uniform initial water saturation profiles across the core samples. The drainage tests in Samples A to J led to initial water saturations ranging from 0.31 to 0.34. The presence of a significant amount of microporosity in the samples is believed to be responsible for the relatively high initial water saturations observed in the cores. Similar results in carbonates have been reported by Dernaika and Masalmeh (2019b). This narrow range of initial water saturations and the use of homogenous rock samples and similar fluid properties enabled us to relate the oil recovery trends only to the wettability conditions.

  3. 3.

    Wettability Restoration After establishing Swi, for all experiments except Experiments A, E, and F, the designated polar oil was continuously injected at a very low flow rate (e.g., 1 pore volume per week) and for certain periods of time (see Table 3) to dynamically alter the wettability of the core samples. During the aging process, Samples B and D were scanned intermittently to measure the in situ contact angles and to determine the aging time, i.e., the time required to establish the equilibrium wettability state. This information was subsequently used in the macroscale counterpart experiments to set the aging time. In Experiments E to J, the flow direction was reversed during the dynamic aging process to uniformly alter the wettability over the length of the core samples. For Experiments A to D, however, the samples were relatively shorter in length and it was unnecessary to reverse the flow direction during the aging process. When the aging processes were concluded, the resident crude oil mixtures were miscibly displaced with 5 to 10 pore volumes (PVs) of decalin followed by 10–20 PVs of Soltrol 170. The sample was imaged once more to generate the distributions of contact angles prior to waterflooding. The comparison between the contact angle distributions before and after injecting decalin and Soltrol 170 confirmed that the contact angle distribution was preserved after conducting the aforementioned experimental stages.

  4. 4.

    Water looding When the core samples were aged and the crude oil mixtures were displaced by Soltrol 170, the temperature of the system was adjusted to 50 °C and the core samples were waterflooded with 20 PV of the synthetic brine at low flow rates, i.e., under a capillary-dominated flow regime. The brine was injected from the bottom face of the core while the effluent was produced from the top side. The waterfloods were conducted in stages while the number of pore volume of water injected (PVWI) was increased. The pressure drop and production data were constantly monitored during the injection process. The injection flow rates were then increased in steps (as listed in Table 4) to investigate the effect of flow rate on the remaining oil saturation in the macroscale tests. The multi-rate waterflooding experiments allowed us to overcome the capillary end effects that exist particularly in non-water-wet samples. During each step, a number of pore volumes was injected until the saturation and pressure drop along the length of the core became stable.

During the waterflooding tests, either the sample was scanned to generate fluid occupancy maps (in the microscale tests) or the effluent samples were analyzed (in the macroscale experiments) to measure residual/remaining oil saturation (i.e., Sor) values. The oil production data was converted to oil recovery factors using Equation 2 to compensate for minor differences in the initial water saturations.

$${\text{Recovery}}\;{\text{factor}} \left( {{\text{RF}}} \right) = \frac{{1 - S_{{{\text{wi}}}} - S_{{{\text{or}}}} }}{{1 - S_{{{\text{wi}}}} }}$$
(2)

Lastly, it should be noted that reservoir formation brines may be significantly different than the aqueous phase used in this study in terms of composition, ionic strength, pH, and divalent/monovalent cation ratio, which impact the equilibrium wettability state of the rock. Nonetheless, in each case, the pore-scale flow mechanisms and recovery trend would be similar to the respective system in our study.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ekechukwu, G.K., Khishvand, M., Kuang, W. et al. The Effect of Wettability on Waterflood Oil Recovery in Carbonate Rock Samples: A Systematic Multi-scale Experimental Investigation. Transp Porous Med 138, 369–400 (2021). https://doi.org/10.1007/s11242-021-01612-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11242-021-01612-3

Keywords

Navigation