Effect of fluid rheology on enhanced oil recovery in a microfluidic sandstone device

https://doi.org/10.1016/j.jnnfm.2013.09.011Get rights and content

Highlights

  • Development of sandstone μ-fluidic device for enhanced oil recovery (EOR) testing.

  • Presents engineered shear-thickening fluid designed for specific EOR conditions.

  • Results from rheologically complex fluids collapse onto a master curve when compared as a function of shear rate dependent Capillary number.

  • Two-stage floods recover single stage flood results for both rheological fluids.

  • Sandstone μ-fluidic device greatly reduces cost, time of current testing methods.

Abstract

As global energy usage increases, maximizing oil recovery from known reserves becomes crucial to meet the rising demand. In this work, we present the development of a microfluidic sandstone platform capable of quickly and inexpensively testing the performance of fluids with different rheological properties on the recovery of oil. Specifically, in this study we utilized these microfluidic devices to examine how shear-thinning, shear-thickening, and viscoelastic fluids affect oil recovery. Initial baseline experiments were performed by displacing oil with both water and a water–surfactant solution over a wide range of flow rates. The surfactant was found to reduce the interfacial tension of the water by a factor of ten and increased oil recovery by approximately 15% when compared to oil displaced by water at the same flow rates. Flopaam, a commercially available fluid thickener that is shear-thinning and viscoelastic was also studied. It was found to displace more oil then either the water or the surfactant solution and increase oil recovery at all flow rates studied. Finally, a shear-thickening nanoparticle solution was studied which was designed to thicken at a shear rate of approximately 10 s−1. The shear rate corresponds to typical shear rates in the oil reservoirs, and values easily attainable in our microfluidic sandstone device. These shear-thickening fluids were found to be particularly effective at oil recovery. This was especially true for flowrates that closely matched the shear rates associated with the shear-thickening regime. When the appropriate choice of shear rate dependent viscosity was used to the capillary number, the oil recovery obtained from both the Newtonian and non-Newtonian was found to collapse quite well onto a single master curve. Additionally, it was shown that a two-stage recovery process that starts with an initial water flood followed by a flood with a secondary fluid can recover as much oil as a single stage recovery with that secondary fluid alone. These results clearly demonstrate that the microfluidic sandstone devices presented in this paper both reduce the time and cost required to investigate the effectiveness of enhanced oil recovery fluids using traditional methods, and can serve to quickly focus searches for customized oil recovery fluid selection.

Introduction

As the global oil supply decreases, the ability to effectively recover all of the oil from a particular well becomes increasingly important. The recovery of oil generally takes place in three stages: primary, secondary, and tertiary [1]. The primary stage occurs when the well is first accessed, and oil is produced by the internal pressures within the well. After completion of the primary stage, approximately 10% of the total oil in the field will have been recovered [1]. The secondary stage of oil recovery is characterized by the use of a driving or pumping fluid to displace oil in the reservoir. Generally water or a gas is employed as the driving fluid and typically results in an additional 20–40% of the original amount of the oil being recovered [1]. After primary and secondary oil recovery techniques have been exhausted, between 50% and 70% of the original oil remains in the oil field. It goes without saying then that there is an enormous amount of interest in developing methods to access and recover all of the remaining oil trapped within the reservoir. It is these tertiary or enhanced oil recovery (EOR) techniques that this paper will focus. Specifically, we investigate the role of the rheology of the driving fluid and shear thickening in particular on oil recovery.

The tertiary or enhanced stage of oil recover has developed much interest in more recent decades partially due to the rise in oil prices [1], [2], [3], [4], [5], [6], [7]. The goal of the tertiary stage is to be able to access and recover as much of the remaining oil in the fields as possible. The methods of tertiary oil recovery can be categorized into three main approaches: thermal, gas, and chemical. All three approaches aim to ease the recovery of the oil, either by changing the properties of the oil, the imbibing fluid, or the core material itself.

Chemical methods of enhanced oil recovery became widely popular during the 1980s [2]. Chemical methods aim to increase the amount of oil recovered by either increasing the effectiveness of water floods by modifying the water used to displace the oil, by reducing the interfacial tension between the imbibing fluid and the oil with the use of a host of different surfactants, or by modifying the wettability of the oil fields substrate to make it lyophobic [1], [2], [4]. Chemical methods use either alkali-polymers, surfactant polymers, or more recently a combination alkali–surfactant polymer system [2]. The larger challenge with chemical methods is that every possible variable with respect to fluid and substrate properties can change from one oil reservoir to the next and thus the chemistry must be tailored specifically for any given reservoir. There is a large body of work for conditions in and solutions to enhanced oil recovery in particular oil fields, summarized in a few works [2], [7], [8]. The main challenge resulting from the use of surfactants to lower the interfacial tension or polymers as a thickener is its delicate relationships to the conditions of the oil field substrate, which is not always constant, and the oil properties, which can also vary. Oftentimes, the polymers or surfactants are applied too far into the initial water flood, or they can lose effectiveness midway through the field [7]. Another challenge with chemical approaches to enhanced oil recovery is cost; depending on the fluctuating cost of oil and production, it can quickly become the prohibiting factor. While challenging, the vast amount of oil remaining within oil fields is only going to be a growing driving factor for EOR research as the easily-accessed oil is recovered and consumed. New fluid technologies will need developing, and modifications of existing technologies will need testing. In this paper, we study the impact of different rheological properties of the driving fluids using a series of microfluidic devices designed to mimic the physical and chemical properties of sandstone.

With a wide range of oil fields, testing methods are varied. It is impossible to perform in situ measurements. Instead, experiments are generally performed with core samples of the actual oil field extracted from the field. These samples can be filled with oil directly from the field, or similar man made oils. This core can then be used to test the ability of penetrating fluids to either displace the oil from pressure flow or imbibitions, which is where the displacing fluid wets the substrate by either natural wettability or by some alteration process [5]. This method does not allow for fine inspection of the processes at the pore and capillary level. Additionally, this method is prohibitively costly. This has led researchers towards the development of low cost alternatives.

Fluid testing is also performed on idealized representations of flows, often two-dimensional arrays of posts or cylinders or three-dimensional beds of packed spheres [5], [9]. This allows for dimensions to be specified for precise control of particular fluid flow properties and examine dynamics at the micro- and nano-scales [5], [9]. This method does not account for the inhomogeneous nature of field conditions or the true geometric flow constraints that exist in the field. Other researchers have used more complex micromodels to study multiphase flow in porous media composed of a network of channels etched into glass or fabricated in polymers [10], [11], [12]. These micromodels have proven to be extremely useful in studying a variety of enhanced oil processes because they provide direct visualization of a complex flow environment that can be easily modified to affect wettability, porosity or permiability. A nice literature survey on this topic can be found in Kamari et al. [12].

In this paper, a series of microfluidic devices were developed and used that were designed to precisely reproduce a two-dimensional slice from a sandstone core. Microfluidics is a relatively young and developing field that encompasses the development of devices that allow for observations of fluid phenoma at the microscale [13], [14], [15], [16], [17], [18], [19]. In microfluidics, photolithography is used to transfer a pattern onto a silicon wafer using a photoresist such as SU-8 [20]. Once developed, the two-dimensional pattern in the photoresist is used as a master from which multiple daughters can be cast in polydimethyl-siloxane (PDMS) or other cross-linking polymers, containing negatives of the pattern on the master [14], [21], [22]. This soft lithography technique has been used for more than a decade to generate microfluidic devices containing features as small as 10 μm [22], [23]. Here we utilize this technology to probe the effect of fluid rheology on oil recovery from hydrophobic microfluidic sandstone devices.

Section snippets

Experimental design

The device layout and properties are shown in Fig. 1. The mask was created from an actual cross-sectional image of sandstone which was used as a template for a microfluidic device that approximates flow through sandstone. The microfluidic device was fabricated out of PDMS using standard photolithographic techniques. The microfluidic sandstone device is 200 microns thick, and has capillaries and pores with average sizes from 200 microns to 10’s of microns. The only modification to the sandstone

Results and discussion

The most common driving fluid used for oil recovery is water. Thus, for this study, water is used as the control for which other driving fluids are compared against. The flowrates of driving water through the microfluidic sandstone device ranged between 1.5 ml/h and 22 ml/h. This corresponds to front speeds just before the sandstone features of 0.38–5.5 mm/s and capillary numbers for water that ranged between 3.8 × 10−5 < Ca = ηU/σ < 5.6 × 10−3. The results for water and the other driving fluids are

Conclusions

Enhanced oil recovery is an increasingly important field, and this work presents the efforts of developing a microfluidic platform for quickly testing fluids of different rheological properties for the recovery of oil from hydrophobic sandstone. Water was tested in the microfluidic sandstone device as a baseline for oil recovery comparison. Systematic variations of fluid properties were examined for their ability to increase oil recovery. A 5 mM CTAB surfactant solution was mixed in order to

Acknowledgements

The authors would like to acknowledge BASF for their financial support of this work, as well as Jack Tinsley, Christian Kunkelmann, Sebastian Weisse, Ravindra Aglave, and Björn Heinz for their assistance and enlightening discussions.

References (39)

  • S. Thomas

    Enhanced oil recovery – an overview

    Oil and Gas Science and Technology-Revue De L Institut Francais Du Petrole

    (2008)
  • R.P. Chhabra

    Bubbles, Drops and Particles in Non-Newtonian Fluids

    (2007)
  • S.G. Sayegh et al.

    Enhanced oil recovery by CO2 flooding in homogeneous and heterogeneous 2D micromodels

    Journal of Canadian Petroleum Technology

    (2009)
  • M. McKellar et al.

    A method of making two-dimensional glass micromodels of pore systems

    Journal of Canadian Petroleum Technology

    (1982)
  • E. Kamari et al.

    Immiscible displacement of awetting fluid by a non-wetting one at high capillary number in a micro-model containing a single fracture

    Transport in Porous Media

    (2012)
  • J. Atencia et al.

    Controlled microfluidic interfaces

    Nature

    (2005)
  • D.C. Duffy et al.

    Rapid prototyping of microfluidic systems in poly(dimethylsiloxane)

    Analytical Chemistry

    (1998)
  • L.C. Gao, T.J. McCarthy, An “artificial lotus leaf” prepared using a 1945 patent and a commercial textile, Abstracts of...
  • Z.H. Nie et al.

    Emulsification in a microfluidic flow-focusing device: effect of the viscosities of the liquids

    Microfluidics and Nanofluidics

    (2008)
  • Cited by (94)

    • Visualization of adaptive polymer flow and displacement in medium-permeable 3D core-on-a-chip

      2023, Petroleum Science
      Citation Excerpt :

      Despite this class of smart polymers has opened a new pathway for EOR in medium-to low-permeability reservoirs, how they propagate and why they displace oil in porous media with such permeability range are unclear yet, because typical porous media such as sand pack columns, artificial cores, and natural rocks are opaque to visible light, precluding the direct visualization of fluid flow dynamics inside them. During the last two decades, microfluidics-based transparent models have been developed to study the pore-scale flow of complex fluid under transient conditions (Browne et al., 2020), represented by the regular expansion-contracting geometry (Rodd. et al., 2007; Galindo-Rosales et al., 2012; Ober et al., 2013) and the disorder etching channel based on realistic core (Nilsson et al., 2013; Zhao et al., 2022) at two-dimensional (2D) scale. However, those 2D micromodels cannot reflect the complex connectivity and tortuosity of real-world porous media in the third dimension, which is always a critical factor affecting the flow of polymer solution.

    View all citing articles on Scopus
    View full text