Effect of surface grafting on the oil-water mixture passing through a nanoslit: A Molecular Dynamics Simulation Study

doi:10.21203/rs.3.rs-5098627/v1

Download PDF

Research Article

Effect of surface grafting on the oil-water mixture passing through a nanoslit: A Molecular Dynamics Simulation Study

https://doi.org/10.21203/rs.3.rs-5098627/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 08 Nov, 2024

Read the published version in Advanced Composites and Hybrid Materials →

You are reading this latest preprint version

Graphene oxide-based membranes hold great promise in composite materials for applications such as wastewater treatment and oil-water separation. In this study, we employed molecular dynamics simulations to investigate the separation of water from an oil-water mixture using a two-layer graphene oxide membrane. We explored the effects of random and stripe-like grafting patterns on penetration efficiency, focusing on varying grafting densities. Our results show that increasing grafting density reduces flux and permeability of both oil and water molecules, highlighting the critical role of surface functionalization in membrane design. Notably, the stripe grafting pattern significantly enhances penetration efficiency by optimizing steric interactions around the nanoslit. These findings contribute to the development of nanocomposite materials and surface modification techniques, offering insights into the design of high-performance membranes for oil-water separation. Understanding the relationship between grafting density, surface patterning, and membrane performance is crucial for advancing hybrid materials that address industrial challenges such as wastewater treatment and oil spill remediation. The insights gained from this study can be further refined by exploring different functional groups and surface modifications, broadening the applications of these membranes in industrial separation processes.

Massive amounts of oily wastewater are continuously generated due to the rapid development of industry and daily life. For example, global wastewater production is estimated to be around 250 million barrels per day [1]. Therefore, there is a significant demand for advanced treatment methods for oily wastewater [1]. Several methods [2, 3] exist for water-oil separation, including distillation, electro-dialysis, solvent extraction, reverse osmosis, and carbon nanotubes [4]. Membrane technology offers unique advantages in treating oily wastewater due to its high separation efficiency and low energy cost [5]. Two-dimensional graphene oxide (GO)-based membranes show great promise in wastewater treatment due to their high flux [6]. In experimental studies, a group of researchers used a metal mesh membrane coated with graphene oxide nanoparticles to separate water from oil [7, 8]. They found that stainless steel mesh [9] coated with nano-graphene oxide particles was hydrophilic when exposed to air but super-oleophobic when immersed in a water environment [7]. The pore diameter of graphene is a significant factor in the water flux in the membrane; membranes with hydrophilic pores show greater water flux than those with hydrophobic pores [10]. Hydrophilic pores also increase the probability of hydrogen bonding [11] during passage, causing water molecules to pass through. Recently, two parallel graphene layers with pores were used for water purification [12]. The results showed that the interlayer spacing affects water flux, with intervals of about 5 Å resulting in almost zero water flux, while larger distances increased the flux of water [13].

The major obstacle for the membrane separation technique is the fouling issue [14]. Spreadable oil droplets are prone to coalesce and spread on membrane surfaces, leading to severe membrane fouling and lower separation efficiency [15]. Tailoring membrane surfaces to mitigate severe fouling at high flux is, therefore, a critical challenge in developing membranes for oil-water separation [2]. Recently, an amphiphilic GO membrane was constructed to regulate interfacial interactions with oil droplets and achieve ultralow fouling at high fluxes [16]. However, the impact of surface modification on water flux is not yet fully understood, and molecular-scale information is lacking. The energetic barrier of hydration layers induced by the hydrophilic domains resists the spread of pollutants, while the low surface energy of the hydrophobic domains facilitates the release of pollutants [17]. Nonpolar hydrophobic domains used to enhance fouling-release ability usually reduce surface hydration [16]. Despite these advances, there are still few simulations studying the effect of grafting density and patterns.

In this study, we perform molecular dynamics (MD) simulations to investigate the effect of surface modification of GO membranes on water and oil flux. The surface of the GO membrane was grafted with perfluoroalkyl chains at various fractional areas. Perfluoroalkyl chains are hydrophobic and favor the release of oils. We focus on the effect of grafting area fractions and the arranging patterns. Especially, two types of grafting patterns are studied: random and stripe. We found that the flux and permeability of oil and water molecules decrease with increasing grafting density. However, the penetration efficiency also depends on the grafting pattern. The random pattern is lower in efficiency than that of the stripe pattern due to the steric hindrance of inclined molecules around the nanoslit. Our findings are helpful to design more effective oil-water separation membranes through surface modification.

The schematic of the simulation box is shown in Fig. 1. One layer of graphene oxide (GO, shown in red) and one layer of graphene (shown in yellow) are used as a membrane located in the simulation box. The gap between them is about 1 nm. Each layer has a nano-slit with a width of 0.8 nm. The shift distance between the two nanoslits is 1 nm. The surface of the GO is randomly modified by perfluorocarboxylic acid (CF₃CF₂CF₂CF₂COOH, shown in green) with the -OH group close to the surface. On the right side of the membrane, there is a mixture of water and oil containing hexane molecules (C₄H₁₄). Water molecules are transparent, and hexane molecules appear as blue points. The numbers of water and oil molecules are 5000 and 64, respectively. The oil-water mixture was kept constant in the initial configuration in all systems. Two graphene sheets, shown in azure, are used as pistons, one of which is behind the water and oil emulsion. The simulation box is periodic only in the x and y directions, while the z direction is non-periodic. Two arrows indicate the direction of the applied atmospheric pressure. Different numbers of perfluoroalkyl chains (20, 30, 42) were grafted on the neutral or negatively charged surface with a charge of -10e. For the negatively charged surface, 10 Na + ions were added to the box. We mainly focused on the random modification, but systems with stripe arrangements of perfluoroalkyl chains on the surface were also studied for comparison. All systems we performed are listed in Table 1.

Table 1

Summary of the simulation systems showing the number of perfluoroalkyl chains grafted on neutral and negatively charged graphene oxide surfaces. Systems are categorized by grafting pattern (random or stripe) and by the density of grafted chains (zero, moderate, and high density). The surface charge (neutral or negatively charged) and arrangements of grafted chains are explored to study their effects on the behavior of the water-oil mixture in the simulation box.
Number of grafted chains		Zero	Moderate density		High density
Number of grafted chains		0 ^a	20	30	42
Random	Neutral ^b	√	√	√	√
Random	Negative	√	√	√	√
Stripe	Neutral	-	√	√	√
Stripe	Negative	-	√	√	√

^a Values present the number of grafted molecules, which were also used to indicate the corresponding systems.

^b The neutral and negative indicate the surface electricity. The negatively charged surface is with charge of -10e.

Our MD simulations were performed using the LAMMPS package [18]. The box size is 3.9 nm x 3.9 nm in the X and Y directions. The entire simulation was carried out at a fixed volume and constant temperature using the NVT statistical ensemble. To stabilize the temperature at 300 K, the Nose-Hoover thermostat was used with a decay ratio of 0.1 ps⁻¹ [19]. The Verlet algorithm was used to solve Newton's equations of motion for the particles at each time step of 1 fs. The cutoff radius for both the Coulombic and Lennard-Jones energy was set to 12.0 Å. Each simulation was performed for 2 ns with 1 bar applied on both sides for equilibrium, followed by 8 ns for data production with 200 bar on the right side. In this simulation, the SPC/E model was used for water molecules, and the OPLS force field was used to describe bond, angular, dihedral, van der Waals, and electrostatic interactions between hexane hydrocarbon molecules [20]. To describe the interactions between atoms in graphene oxide and the graphene piston, the force field developed by Cheng and Steele [21] was used. The PPPM method was used to correct the potential of electrostatic interactions of Coulombic and long-range Lennard-Jones interactions. Using the SHAKE algorithm [22], the O-H bond length and H-O-H angle of water molecules were kept constant at 1.0 Å and 104.59°, respectively.

We first focus on the randomly grafting systems. The distribution of oil, brush, and water in the equilibrium state without a pressure difference in the Z direction is given in Fig. 2. The height of the brush is about 1.1 nm for all cases. With the increase in grafted density, the distribution profile of the brush slightly broadens in the Z direction due to steric interactions. It can also be found that neutral surface favor the brush due to the hydrophobic effect. The oil distribution is broad. There is more oil in the brush layer for cases of low grafting densities because of the greater free space near the surface, as seen in the spatial density map of the brush (bottom, Fig. 2). For higher grafted density, the oils are mainly distributed ~ 1.0nm far from the surface, close to the brush terminal. It can be observed that there are more perfluorocarboxylic acid molecules around the nanoslit for higher grafted densities. There is a water layer near the surface, indicated by the first peaks of water density. Additionally, there appears to be more water on the negatively charged surface compared to the neutral one due to electrostatic-dipole interactions. Further, the distributions of water close to the brush terminals shows no significant difference. It should be pointed out that we do not observe significant aggregation of oil molecules at the terminal. This is in good agreement with the finding by Yang et al. [16], who studied the antifouling effect of grafted membrane.

Now we turn to the oil and water penetration behavior. Special systems without grafted perfluorocarboxylic acids were also included for comparison. The number of oil and water molecules passing through the membrane as a function of time is calculated and shown in Fig. 3 to evaluate the performance at different grafting densities. It can be seen that, without surface modification, oil and water penetrate through the nanoslit quickly. When the surface is grafted with perfluorocarboxylic acids, the penetration efficiency of oil decreases. High grafting density is more effective at blocking oil penetration (Fig. 3a). The behavior is similar for the penetration of water (Fig. 3b). Qualitatively, there is no significant difference between neutral and negatively charged surfaces. However, water molecules are not stuck at high grafting densities due to their small size. It appears that negatively charged modification is better for water penetration due to its high hydration effect.

The efficiency of the flow of oil and water molecules through the channels of various systems is shown in Fig. 4. Efficiency is defined as the ratio of the number of molecules (water or oil) passing through the second nanoslit within 6 ns. It can be observed that without modification, almost all oil and water molecules penetrate through the membrane regardless of surface charge. Moderate grafting density nearly blocks half of the molecules. For high grafting density, the efficiency is close to zero for the neutral surface. For the negatively charged surface, the efficiency for water is about 10%. This result is consistent with the number of penetrated molecules.

To study the effect of the grafting pattern, we also designed a stripe-like structure on the surface. The Z-direction density profile in the equilibrium state is similar to that of the random system (see Fig. 5a and Fig. 2c). The distribution of grafted molecules is relatively regular, with almost no grafted molecules tilting in the nanoslit region (Fig. 5b). Qualitatively, the penetration efficiency of oil and water decreases as the grafting density increases (Fig. 5c). About 80% of the molecules penetrate through the nanoslit for all cases we studied (Fig. 5d). Surprisingly, it is obvious that the penetration efficiency at high density is larger than that of the random system (Fig. 4 and Fig. 5c). For moderate grafting density, the efficiency is also slightly higher than that of the random system. This implies that grafting pattern plays a crucial role in penetration efficiency.

To explore the mechanism behind the different efficiencies at high density, we carefully examined the water penetration process by analyzing the simulation trajectories. The tilt of grafted molecules plays a crucial role in blocking oil and water. To quantitatively compare the degree of tilting, we defined a region of 1 nm × 4 nm × 3.9 nm above the nanoslit (Fig. 6a) and counted the average number of carbon atoms of perfluorocarboxylic acids in this region for the negatively charged systems (Fig. 6b) using data from the last 2 ns of the trajectories. It can be observed that the number of carbon atoms varies with different patterns. At high density, the carbon atom count for the random pattern is higher than that for the stripe pattern. The time evolution of the carbon atoms in the region is shown in Fig. 6c. It can be seen that the grafted molecules close to the nanoslit tilt more quickly for the random pattern, whereas for the stripe pattern, the carbon atom number increases gradually. This difference may be due to the varying distances between grafted molecules and the nanoslit. The tilt of grafted molecules depends not only on their stiffness, but on the hydrodynamic flows in the nanoslit due to the pressure difference. To further investigate the differences between the two systems, we also plotted the spatial density maps of the randomly and stripe-like brush (Fig. 7). Evidently, there are more grafted molecules distributed in the nanoslit region for the random pattern.

When designing membranes for oil-water separation, key properties such as pore size, slit size, and interlayer spacing are critical for determining the separation efficiency [1]. Given that filtration is primarily a physical process, these parameters must be carefully controlled to prevent contaminants from passing through the membrane. Control over these dimensions allows for enhanced separation performance, especially when functional groups are incorporated into the membrane structure [23]. The addition of functional groups can dramatically alter membrane behavior by modifying its interactions with water and oil molecules, enhancing specific properties such as water permeability, ion rejection, and pore polarity. This is particularly important for composite materials, where functionality and performance can be tailored for specific applications.

For instance, membranes with superhydrophobic-superoleophilic or superhydrophilic-superoleophobic properties have proven highly effective for oil-water separation [24]. These materials either filter molecules by blocking their passage or selectively absorb them, depending on the membrane’s surface chemistry. Hydrophilic functional groups typically enhance water permeability by facilitating water passage through the membrane, whereas hydrophobic groups tend to reduce permeability by repelling water molecules and decreasing membrane wettability. This dual behavior enables a more tailored approach to membrane design, making functionalization an essential component of next-generation nanocomposite materials.

Our findings demonstrate that grafting density plays a crucial role in determining the performance of GO-based membranes. As grafting density increases, the surface becomes more densely covered with functional groups, which can directly affect the membrane's hydrophilicity and interactions with water and oil molecules. At higher densities, steric hindrance is introduced, leading to a more densely packed surface layer that blocks the passage of molecules. This results in reduced permeability, which is consistent with prior studies showing that increased surface coverage can impede transport through the membrane [16].

Moreover, our study reveals that the nanoscale pattern of functional groups significantly impacts membrane performance. A random grafting pattern creates a heterogeneous surface with unevenly distributed functional groups, leading to varying interactions with water and oil molecules. In contrast, stripe-like patterns produce a more uniform distribution, which optimizes the interaction between the membrane and the permeating molecules, thereby enhancing separation efficiency. This suggests that the design of the grafting pattern is as important as the grafting density in optimizing membrane performance.

The differences in penetration efficiency due to grafting patterns are primarily driven by steric effects, where the spatial arrangement of functional groups around the nanoslit influences the ability of molecules to pass through. The physical arrangement of these groups can either block or facilitate the passage of water and oil molecules through the membrane. Additionally, hydrodynamic interactions during non-equilibrium transport processes can cause the grafted molecules to tilt or reorient, further affecting separation efficiency. This tilting behavior, which varies between random and stripe patterns, emphasizes the role of both molecular interactions and fluid dynamics in membrane functionality.

Graphene oxide-based membranes have shown great potential in composite materials for applications ranging from wastewater treatment to oil-water separation. In the present work, we have studied the separation of water from an oil-water mixture using a two-layer graphene oxide membrane through molecular dynamics simulations. We focus on the effects of random grafting density on penetration efficiency and compare this with a stripe-like grafting pattern. Our simulation results demonstrate that increasing grafting density leads to decreased flux and permeability of oil and water molecules, emphasizing the critical role of surface functionalization in composite membrane design. Furthermore, the grafting pattern, particularly the stripe configuration, significantly enhances penetration efficiency by optimizing steric interactions around the nanoslit.

These findings contribute to the advancement of nanocomposite materials and surface modification techniques, offering insights into the design of high-performance membranes for oil-water separation. Understanding the interplay between grafting density, surface patterning, and membrane performance is crucial for developing advanced hybrid materials that can address industrial challenges such as wastewater treatment and oil spill remediation. Future research could explore the integration of alternative functional groups and patterns, as well as the real-world application of these membranes in large-scale industrial processes. Such advancements would not only improve the efficiency of separation technologies but also broaden the scope of composite materials in environmental and industrial applications.

Author contribution Wende Tian wrote the main manuscript text and did all the simulation works; Yanwei Wang wrote some part of the manuscript text and revised the draft; Zhexenbek Toktarbay did experiments conceptualization, general management, and manuscript revision.

Funding This work was supported by the Ministry of Science and Higher Education of the Republic of Kazakhstan under the project AP19679745 “Design of mechanically strong, biodegradable membrane for separation process”.

Competing interests: The authors declare no competing interests.

Ethical Approval: not applicable.

Kommu A and Singh JK (2020) A Review on Graphene-Based Materials for Removal of Toxic Pollutants from Wastewater, Soft Materials 18:297-322. https://doi.org/10.1080/1539445x.2020.1739710
Lejars M, Margaillan A, and Bressy C (2012) Fouling Release Coatings: A Nontoxic Alternative to Biocidal Antifouling Coatings, Chem. Rev. 112:4347-4390. https://doi.org/10.1021/cr200350v
Rana D and Matsuura T (2010) Surface Modifications for Antifouling Membranes, Chem. Rev. 110:2448–2471. https://doi.org/10.1021/cr800208y
Fornasiero F, Park HG, Holt JK, Stadermann M, Grigoropoulos CP, Noy A, and Bakajin O (2008) Ion Exclusion by Sub-2-Nm Carbon Nanotube Pores, Proc. Natl. Acad. Sci. U.S.A. 105:17250-17255. https://doi.org/10.1073/pnas.0710437105
Tao M, Xue L, Liu F, and Jiang L (2014) An Intelligent Superwetting PVDF Membrane Showing Switchable Transport Performance for Oil/Water Separation, Advanced Materials 26:2943-2948. https://doi.org/10.1002/adma.201305112
Liu Y, Su Y, Guan J, Cao J, Zhang R, He M, Gao K, Zhou L, and Jiang Z (2018) 2D Heterostructure Membranes with Sunlight‐Driven Self‐Cleaning Ability for Highly Efficient Oil–Water Separation. Adv. Funct. Mater. 28:1706545. https://doi.org/10.1002/adfm.201706545
You X, Wu H, Zhang R, Su Y, Cao L, Yu Q, Yuan J, Xiao K, He M, and Jiang Z (2019) Metal-Coordinated Sub-10 Nm Membranes for Water Purification. Nat. Commun. 10:4160. https://doi.org/10.1038/s41467-019-12100-0
Song X, Chen Y, Rong M, Xie Z, Zhao T, Wang Y, Chen X, and Wolfbeis OS (2016) A Phytic Acid Induced Super‐Amphiphilic Multifunctional 3D Graphene‐Based Foam, Angew. Chem. Int. Ed. 55:3936-3941. https://doi.org/10.1002/anie.201511064
Dong Y, Li J, Shi L, Wang X, Guo Z, and Liu W (2014) Underwater Superoleophobic Graphene Oxide Coated Meshes for the Separation of Oil and Water, Chem. Commun. 50:5586-5589. https://doi.org/10.1039/c4cc01408a
Cohen-Tanugi D and Grossman JC (2012) Water Desalination across Nanoporous Graphene, Nano Lett. 12:3602-3608. https://doi.org/10.1021/nl3012853
O’Hern SC, Boutilier MSH, Idrobo JC, Song Y, Kong J, Laoui T, Atieh M, and Karnik R (2014) Selective Ionic Transport through Tunable Subnanometer Pores in Single-Layer Graphene Membranes, Nano Lett. 14:1234-1241. https://doi.org/10.1021/nl404118f
Foroutan M, Zahedi H, and Soleimani E (2018) Investigation of Water-Oil Separation via Graphene Oxide Membranes: A Molecular Dynamics Study, Colloids Surf. A Physicochem. Eng. Asp. 555:201-208. https://doi.org/10.1016/j.colsurfa.2018.07.002
Cohen-Tanugi D, Lin LC, and Grossman JC (2016) Multilayer Nanoporous Graphene Membranes for Water Desalination, Nano Lett. 16:1027-1033. https://doi.org/10.1021/acs.nanolett.5b04089
Zhang R, Liu Y, He M, Su Y, Zhao X, Elimelech M, and Jiang Z (2016) Antifouling Membranes for Sustainable Water Purification: Strategies and Mechanisms, Chem. Soc. Rev. 45:5888-5924. https://doi.org/10.1039/c5cs00579e
Molino PJ, Yang D, Penna M, Miyazawa K, Knowles BR, MacLaughlin S, Fukuma T, Yarovsky I, and Higgins MJ (2018) Hydration Layer Structure of Biofouling-Resistant Nanoparticles, ACS Nano 12:11610-11624. https://doi.org/10.1021/acsnano.8b06856
Yang C, et al. (2022) Antifouling Graphene Oxide Membranes for Oil-Water Separation via Hydrophobic Chain Engineering, Nat. Commun. 13:7334. https://doi.org/10.1038/s41467-022-35105-8
Hou Y, Xu Z, and Yang X (2016) Interface-Induced Affinity Sieving in Nanoporous Graphenes for Liquid-Phase Mixtures, J. Phys. Chem. C 120:4053-4060. https://doi.org/10.1021/acs.jpcc.5b10287
Plimpton S (1995) Fast Parallel Algorithms for Short-Range Molecular Dynamics, J. Comput. Phys. 117:1-19. https://doi.org/10.1006/jcph.1995.1039
Nosé S (1984) A Unified Formulation of the Constant Temperature Molecular Dynamics Methods, J. Chem. Phys. 81:511-519. https://doi.org/10.1063/1.447334
Jorgensen WL, Maxwell DS, and Tirado-Rives J (1996) Development and Testing of the OPLS All-Atom Force Field on Conformational Energetics and Properties of Organic Liquids, J. Am. Chem. Soc. 118:11225-11236. https://doi.org/10.1021/ja9621760
Cheng A and Steele WA (1990) Computer Simulation of Ammonia on Graphite. I. Low Temperature Structure of Monolayer and Bilayer Films, J. Chem. Phys. 92:3858-3866. https://doi.org/10.1063/1.458562
Miyamoto S and Kollman PA (1992) Settle: An Analytical Version of the SHAKE and RATTLE Algorithm for Rigid Water Models, J. Comput. Chem. 13:952-962. https://doi.org/10.1002/jcc.540130805
Fu L, Liao K, Tang B, Jiang L, and Huang W (2020) Applications of Graphene and Its Derivatives in the Upstream Oil and Gas Industry: A Systematic Review, Nanomaterials 10:1013. https://doi.org/10.3390/nano10061013
Zahedi H and Foroutan M (2019) Separation of Water–Oil Mixture on Poly Methyl Methacrylate Surface Using TiO2 Nanoparticles via Molecular Dynamics Simulation, Adsorption 25:1019-1031. https://doi.org/10.1007/s10450-019-00119-0

No competing interests reported.

Download PDF

Journal Publication

published 08 Nov, 2024

Read the published version in Advanced Composites and Hybrid Materials →

Editorial decision: Revision requested
08 Oct, 2024
Reviews received at journal
08 Oct, 2024
Reviews received at journal
06 Oct, 2024
Reviewers agreed at journal
06 Oct, 2024
Reviewers agreed at journal
05 Oct, 2024
Reviewers agreed at journal
03 Oct, 2024
Reviewers invited by journal
03 Oct, 2024
Editor assigned by journal
01 Oct, 2024
Submission checks completed at journal
27 Sep, 2024
First submitted to journal
16 Sep, 2024

You are reading this latest preprint version

Effect of surface grafting on the oil-water mixture passing through a nanoslit: A Molecular Dynamics Simulation Study

Status:

Journal Publication

Version 1

Abstract

Figures

1 Introduction

2 Model and Methods

3 Results and discussions

4 Conclusion

Declarations

References

Additional Declarations

Status:

Journal Publication

Version 1