Nanofluid of graphene-based amphiphilic Janus nanosheets for tertiary or enhanced oil recovery: High performance at low concentration

Significance

Improving crude oil recovery by 1% worldwide would result in a huge amount of crude oil resources becoming available. However, the economic and environmental concerns are too serious to ignore when chemical methods (surfactants or polymers flooding, etc.) are used for an average 10–20% enhancement for tertiary oil recovery. Simple nanofluid (containing only nanoparticles) flooding at low concentration (0.01 wt % or less) is a promising alternative, but the efficiency is below 5% in a saline environment (2 wt % or higher NaCl content). We report a simple nanofluid of graphene-based Janus amphiphilic nanosheets for enhanced oil recovery with efficiency of about 15.2%, comparable to chemical methods, which is both economically and environmentally beneficial to the petroleum industry.

Abstract

The current simple nanofluid flooding method for tertiary or enhanced oil recovery is inefficient, especially when used with low nanoparticle concentration. We have designed and produced a nanofluid of graphene-based amphiphilic nanosheets that is very effective at low concentration. Our nanosheets spontaneously approached the oil–water interface and reduced the interfacial tension in a saline environment (4 wt % NaCl and 1 wt % CaCl₂), regardless of the solid surface wettability. A climbing film appeared and grew at moderate hydrodynamic condition to encapsulate the oil phase. With strong hydrodynamic power input, a solid-like interfacial film formed and was able to return to its original form even after being seriously disturbed. The film rapidly separated oil and water phases for slug-like oil displacement. The unique behavior of our nanosheet nanofluid tripled the best performance of conventional nanofluid flooding methods under similar conditions.

Finding economically viable and environmentally friendly methods to extract the huge amount of residual oil after primary and secondary recovery remains challenging for the oil and gas industry and is also of significant importance in efforts to satisfy the world’s increasing energy demand. Nanofluid flooding as an alternative tertiary oil recovery method has been recently reported (1–5). Obviously, simple nanofluid flooding (containing only nanoparticles) at low concentration (0.01 wt % or less) shows the greatest potential from the environmental and economic perspective. Several corresponding oil displacement mechanisms have also been introduced, including reduction of oil–water interfacial tension (6, 7), alteration of rock surface wettability (8–10), and generation of structural disjoining pressure (11–13). However, the oil recovery factor is below 5% with 0.01% nanoparticle loading in core flooding tests in a saline environment (2 wt % or higher NaCl content). Here we show that an oil recovery factor of 15.2% is achieved by using a simple nanofluid of graphene-based Janus amphiphilic nanosheets. To our knowledge, this is the first report of applying nanofluid of amphiphilic Janus two-dimensional materials in tertiary or enhanced oil recovery. We found that in a saline environment, the nanosheets spontaneously approach the oil–water interface, reducing the interfacial tension. A climbing film emerges and encapsulates the oil phase and may carry it forward. Furthermore, we found that a solid-like film forms with strong hydrodynamic power. The film rapidly separates oil and water for slug-like oil displacement. Even though there are ways to achieve 20% enhanced recovery by complicated alkali/surfactant/polymer flooding (14) or by surfactants with added nanoparticles (5), the necessary concentrations of the chemicals and nanoparticles are much higher than 0.01 wt %. Our results provide a nanofluid flooding method for tertiary oil recovery that is comparable to the sophisticated chemical methods. We anticipate that this work will bring simple nanofluid flooding at low concentration to the stage of oil field practice, which could result in oil being recovered in a more environmentally friendly and cost-effective manner.

Materials and Methods

Synthesis.

The Janus amphiphilic nanosheets were produced by tuning the Janus balance of graphene oxide (GO) with alkylamine. Initially, GO was synthesized from chemical oxidation of graphite (15). Single-surface hydrophobization was then carried out (16, 17). The nanofluid was made stable to avoid agglomeration of the nanosheets. Brine used in all experiments contained 4 wt % NaCl and 1 wt % CaCl₂.

Characterization.

Atomic force microscopy (AFM; Veeco Dimensions 3000 atomic force microscope) was used to examine the morphology of GO. Measurement was conducted using silicon AFM probes (HQ:NSC15/AL BS, Mikromasch) with a resonant frequency of ∼325 kHz, a force constant of ∼40 N m⁻¹, and a tip radius of ∼8 nm. Imaging was done in tapping mode with resolutions of 512 × 512. Scanning electron microscopy (SEM; FEI Quanta 200) was used to examine the cross-section of the sandstone cores under an accelerating voltage of 20 kV. Fourier transform infrared spectroscopy (FTIR) spectra were recorded on a Nicolet iS50 FTIR spectrometer with an attenuated total reflectance accessory. Thermal gravimetric analysis (TGA) was performed on a TGA Q50 (TA Instrument) under nitrogen atmosphere at a rate of 10 °C/min. Particle size and concentration were detected and visualized using a Malvern NanoSight NS300.

Core Flooding Test.

Four man-made sandstone rock cores were tested in the flooding equipment (Fig. S1). The physical properties of the rock cores were measured and are listed in Table S1. Crude oil samples were taken from one of China’s oil fields. The viscosity of the crude oil was 75 cP at 25 °C. Nanofluids of 0.005 wt % and 0.01 wt % nanosheet concentration were injected into a saline environment (4 wt % NaCl and 1 wt % CaCl₂) to measure the enhanced oil recovery factor for rock cores with both low (samples 1 and 2) and high (samples 3 and 4) liquid permeability. The core flooding test was conducted sequentially with the following steps: cleaning of rock cores; saturating cores with brine; establishing initial brine water and oil saturation by oil injection until no more brine water was produced; brine water flooding until no more oil (i.e., 100% water cut) was produced; and nanofluid flooding until no more oil was extracted. The total injection volume of nanofluid for each flooding test was 3–4 times the pore volume (PV).

Fig. S1.

Schematic of core flooding equipment for oil recovery tests.

Table S1.

Physical properties of the rock cores

Rock core	Length, cm	Diameter, cm	Porosity, %	Average liquid permeability, mD	PV, cm3
1	3.936	2.550	24.8	54.4	4.985
2	3.896	2.541	26.0	44.5	5.145
3	4.168	2.526	27.9	130.0	5.817
4	4.074	2.515	25.8	132.0	5.228

Results and Discussion

Surface Functionalization of GO with Alkylamine.

The thickness of GO was estimated to be about 1 nm from the AFM image (Fig. S2), indicating the single-layer feature of GO, in agreement with the previous report (18). For comparison, we have also studied the effect of unmodified GO. Either the unmodified GO or our amphiphilic Janus nanosheets were injected into the heptane/brine system. The amphiphilic Janus nanosheets spontaneously accumulated at the heptane/brine interface while GO stayed only in the brine phase and instantaneously started to aggregate. When subjected to vortex-induced vibrations, the amphiphilic Janus nanosheets formed a thin interfacial film separating the heptane and brine in contrast to the GO agglomeration due to the salt screening effect (Fig. 1). After settling for 2 h, unmodified GO precipitated while the interfacial film of our nanosheets remained intact. This observation suggested successful asymmetrical functionalization of GO with hydrocarbon chains. In FTIR analysis (Fig. 2A), both GO and amphiphilic nanosheets exhibited peaks at 1,723 cm⁻¹, 1,587 cm⁻¹, and 1,230 cm⁻¹, which can be assigned to C=O carbonyl/carboxyl, C=C aromatic, and C–O–C epoxy vibrations, respectively. In addition, amphiphilic nanosheets clearly demonstrated the presence of methylene groups (strong peaks at 1,470 cm⁻¹, 2,850 cm⁻¹, and 2,925 cm⁻¹) and methyl groups (weaker signals at 1,380 cm⁻¹ and 2,960 cm⁻¹), which confirmed the successful conjugation of hydrocarbon chains onto the GO surface. Compared with GO, the TGA curve of amphiphilic nanosheets displayed an additional weight loss stage between 400 °C and 500 °C, which may be attributed to the decomposition of the carbon chains (Fig. 2B).

Fig. S2.

AFM image of graphene oxide.

Fig. 1.

Behaviors of unmodified GO and amphiphilic nanosheets in the heptane/brine system. Small pieces of interfacial film were attached to the hydrophilic glass surface in the heptane phase due to its amphiphilicity, which appeared as black dots.

Fig. 2.

(A) FTIR spectra and (B) TGA curves of GO and amphiphilic nanosheet.

Stability Evaluation of the Nanofluid.

It is crucial that the nanofluid has good stability before being injected into the reservoir. From observation, both GO and amphiphilic nanosheets have very small amounts of precipitates even after 30 d. To evaluate the stability in microscopic view, the dispersions of GO and amphiphilic nanosheets were first subjected to bath sonication for 30 s before dilution to the concentration of 0.005 wt % and then injected into the chamber in the Malvern NS300 system, in which a laser passes through. Particles in the path of the laser scattered light and were visualized by a camera. Altogether, five locations were observed, each for a period of 60 s. Fig. 3 shows the average hydrodynamic diameter distribution of GO (Fig. 3A) and amphiphilic nanosheets (Fig. 3B). It appeared that the mode of hydrodynamic diameter of amphiphilic nanosheets was rather close to that of GO, implying that no agglomeration took place after functionalization. From the captured video of nanofluid dispersion, we did not find any noticeable aggregation. Photographs of amphiphilic nanosheet nanofluid in three locations at different time points were selected as examples (Fig. 3C). Each bright dot in the photographs represented a single nanosheet in Brownian motion. No aggregation, which would display as clusters, was detected. The excellent stability of the nanofluid would ensure that no extra additives or methods are required to preserve the nanofluid before injection.

Fig. 3.

Hydrodynamic diameter distributions of (A) GO and (B) amphiphilic nanosheets in nanofluids with a concentration of 0.005 wt %. (Insets) Mode sizes. (C) Selected photographs of amphiphilic nanosheets in the nanofluid in three locations at different time points.

Oil Displacement Efficiency and Mechanisms.

Approximately 90% of the oil recovered by nanofluid flooding was extracted after the first PV injection. As shown in Table 1, under similar conditions (0.01% nanofluid concentration), our recovery was 15.2%, more than triple the best reported result of 4.7% (2). In addition, for 0.005 wt % concentration, which to our knowledge has not been previously reported, we also achieved extraordinary performance.

Table 1.

Oil recovery factor of nanofluid flooding with different concentrations

Rock core	Porosity, %	Average liquid permeability, mD	Nanofluid concentration, wt %	Oil recovery factor after brine water flooding, %	Enhanced oil recovery factor after nanofluid flooding, %	Total oil recovery factor, %
1	24.8	54.4	0.005	71.1	6.7	77.8
2	26.0	44.5	0.01	62.5	9.5	72.0
3	27.9	130.0	0.005	68.2	10.2	78.4
4	25.8	132.0	0.01	69.6	15.2	84.8

In reservoirs, the motion of crude oil can be categorized into three scenarios depending on the local underground hydrodynamic power: static, slow, and fast moving. Therefore, we investigated the behaviors of nanosheets under simulated hydrodynamic scenarios to study the oil displacement mechanisms, and our findings showed results quite different from those of the traditional nanofluid.

As shown in Fig. 4, under static conditions, the interface between heptane and brine was concave in a water-wet glass tube but convex in an oil-wet plastic tube, indicating relatively strong interfacial tension. When the nanofluid was injected into the brine, electrostatic repulsion was screened by salt and hydrophobic attraction came into play (19). As a result, nanosheets spontaneously accumulated at the interface in each case. With increasing amount of nanosheets adsorbed at the interface, the interfacial tension was further reduced (20, 21), as indicated by the increasingly flattened interface (Movie S1). As with conventional nanofluids, however, salt ions in the reservoir fluid were found to be a permeability damage factor due to increased nanoparticle aggregation. The climbing film was also observed because locally raised nanosheet concentrations induced Marangoni stress (the stress produced by the gradient of interfacial tension) to push the oil–water interface up the tube surface (22). This film climbed upward in the case of the water-wet tube surface, which helped detach the oil phase at solid surface, but downward in the case of the oil-wet tube surface. For the oil-wet surface, nanosheets may be captured by the hydrophobic tube surface with the hydrophilic side facing the water phase. Such behaviors altered the wettability of the oil-wet surface.

Fig. 4.

Behaviors of nanosheets in oil/brine system with increasing hydrodynamic power. (A–E) Behaviors of nanosheets in oil/brine system with increasing hydrodynamic power in a glass tube with water-wet surface (the heptane was dyed with Sudan Red 7B). (A) Heptane/brine mixture. (B) Nanosheets adsorbing to heptane/brine interface during nanofluid injection under static conditions. (C) Nanosheets at heptane/brine interface after injection; a climbing film appeared on the wall. (D) Growth of climbing film after gentle shaking. (E) Formation of interfacial film after vigorous shaking. (F–J) Behaviors of nanosheets in oil/brine system with increasing hydrodynamic power in a plastic tube with oil-wet surface (Sudan Red 7B was not used as the whole plastic tube would be dyed). (F) Heptane/brine mixture. (G) Nanosheets adsorbing to heptane/brine interface during nanofluid injection under static conditions. (H) Nanosheets at heptane/brine interface after injection; a climbing film appeared on the wall. (I) Growth of climbing film after gentle shaking. (J) Formation of interfacial film after vigorous shaking.

To simulate the condition of moderate hydrodynamic power, the tubes were shaken gently after injection. It was observed that for water-wet glass tubes, the climbing film grew to encapsulate the oil phase and may carry the oil forward at flow conditions, leaving very little residual oil behind. For the oil-wet plastic tube, the growth of the climbing film was not as obvious as with the glass tube, possibly due to the adsorption of nanosheets onto the wall of the tube. When subjected to vigorous shaking, the nanosheets formed flat films at interfaces for both the glass and plastic tubes. The formation of such a solid-like film was also predicted by computer simulation for near-neutral wetting spherical particles (23).

When subjected to intrusion of a glass rod, the surface of interfacial films was deformed but not ruptured (Fig. 5 and Movie S2). After removal of the glass rod, the films recovered to their original state, clearly demonstrating their elasticity. The presence of nanosheets altered not only the normal stress balance, but also the tangential stress balance, leading to a redistribution of nanosheets to form a flat shape. The elasticity of the interfacial films kept them intact even at 90 °C. To demonstrate that the elasticity of the interfacial film was unique to our amphiphilic nanosheets, we chose sodium dodecyl sulfate (SDS) and TWEEN 20 (purchased from Sigma-Aldrich) as examples of ionic and nonionic surfactants, respectively, for comparison. With the same concentration of 0.01 wt %, when the interfaces formed by SDS or TWEEN 20 were subjected to the glass rod intrusion, neither of them exhibited observable elastic deformation but simply broke through, very different from the elastic interfacial film formed by amphiphilic nanosheets (Fig. 5C). In terms of interfacial rheology, the interfacial films may resist dilation and bending as characterized by the nearly flat interface after the tubes were tilted to enlarge the area of the interfaces (Fig. 6 A and B). Vigorous agitation disrupted the films for both solid surfaces (Fig. 6 C and D). However, the films reformed immediately at the interfaces and separated the oil and brine phases after shaking. This process may be driven by the amphiphilicity of nanosheets. In contrast to emulsion flooding, the interfacial film in our case strictly separates the water and oil phases, which may push the oil to the outlet like a slug at flow conditions. This was confirmed in the core flooding tests, where the oil and brine came out sequentially at the outlet. Such oil displacement mechanism leaves less oil residue and is free of demulsification, which is a significant advantage of our nanofluid over the traditional ones.

Fig. 5.

Testing the elasticity of interfacial films. (A) Interfacial film responding to intrusion of a glass rod in a glass tube with a water-wet surface (from left to right: before intrusion, during intrusion, after intrusion). (B) Interfacial film responding to intrusion of a glass rod in a plastic tube with an oil-wet surface (from left to right: before intrusion, during intrusion, after intrusion). (C) The heptane/brine interface with amphiphilic nanosheets (Left), SDS (Middle), and TWEEN 20 (Right) responding to intrusion of a glass rod.

Fig. 6.

Testing the stability of interfacial films. (A and B) Deformation of interfacial films upon tilting. (A) Image of heptane/brine interfaces in tilted glass tubes with water-wet surfaces. (Upper) Interface with nanosheet interfacial film. (Lower) Interface without nanosheets. (B) Image of heptane/brine interfaces in tilted plastic tubes with oil-wet surfaces. (Upper) Interface with nanosheet interfacial film. (Lower) Interface without nanosheets. (C and D) Reformation of interfacial films after shaking. (C) Image of interfacial film under vigorous shaking in a glass tube with water-wet surface (from left to right: before shaking, during shaking, 1 min after shaking). (D) Image of interfacial film under vigorous shaking in a plastic tube with oil-wet surface (from left to right: before shaking, right after shaking, 1 min after shaking).

After careful examination of the existing mechanisms (6–13), we found none of them was fully applicable to the above experimental observation under different simulated hydrodynamic conditions. Therefore, we propose two oil displacement mechanisms for nanofluid flooding with nanosheets, as illustrated in Fig. 7:

• i)The climbing film (a film of nanosheets along the tube’s surface) encapsulation for water-wet surface. As shown in Fig. 7A, at t = t₀ the increased concentration of nanosheets due to the adsorption at the oil–water interface produces the concentration gradient leading to transfer of nanosheets to the three-phase (nanofluid, oil, and rock solid) region, detaching and encapsulating oil from the rock surface. When flow continues under gentle hydrodynamic condition from t₀ to t₀ + Δt, the film grows due to the ongoing supply of nanosheets from the nanofluid and carries the oil phase forward.
• ii)Slug-like displacement by the interfacial film. As seen in Fig. 7B, at t = t₀ an elastic interfacial film forms at the oil–water interface at strong hydrodynamic power condition. The film can resist bending and also reform after being disrupted. As a result, at t = t₀ + Δt oil is slug-like and displaced over a certain distance.

Fig. 7.

Schematic illustration of oil displacement mechanisms. (A) Climbing film encapsulation mechanism for water-wet surface. (B) Slug-like displacement mechanism.

Except for these oil displacement mechanisms, the amphiphilic Janus nanosheets are expected to have a lower chance to be captured by the rock surface or to plug the rock pores due to the self-accumulating at the oil–water interface, which may also contribute to the high efficiency of oil recovery. The cross-sections of rock cores were examined by SEM before and after core flooding and are displayed in Fig. 8. In each figure, the smallest and largest pore openings are both labeled. As for the rock cores with high permeability (Fig. 8 A and C), after nanofluid flooding, the pore sizes at the two ends remained at the same level, comparable to those before flooding. Similar results were also detected in rock cores with low permeability (Fig. 8 B and D), with survival of narrow pore openings of around 2 μm. The observation that there were no noticeable changes in the pore opening sizes indicated that our nanofluid caused minimal damage to the permeability of the rock pores.

Fig. 8.

SEM images of the cross-sections of (A) rock core with high permeability and (B) rock core with low permeability before core flooding tests; and (C) rock core with high permeability and (D) rock core with low permeability after core flooding tests. Arrows in each image indicate the smallest and largest pore openings.

In addition, fresh water was used to prepare the nanofluid for the tests described above. However, brine is preferred in some operations (e.g., seawater in offshore reservoirs or when fresh water is scarce) to reduce cost and conserve fresh-water resources. This requires that nanosheets be stable in a saline environment for a certain time before reaching underground reservoirs. Our study showed that sequential addition of polyvinylpyrrolidone and polyvinyl alcohol provided good stability to our nanofluid in brine. We are currently investigating the stabilization mechanism, the oil recovery efficiency in core flooding tests, and oil displacement mechanisms due to the existing polymer.

Conclusion

We have presented the first (to our knowledge) tertiary or enhanced oil recovery experiments using simple nanofluid flooding with graphene-based Janus amphiphilic nanosheets at a low concentration. The result from core flooding measurements showed that the oil enhancement efficiency of 15.2% by this nanofluid flooding is more than three times that of the previously reported best efficiency (4.7%) under similar conditions at 0.01 wt % concentration. The behavior tests of nanosheets in oil and brine system provided evidence that under a saline environment, (i) the accumulation of nanosheets at the oil–water interface, (ii) the appearance of climbing films, and (iii) the generation of elastic interfacial films may be responsible for the high oil recovery efficiency.