Continuous oil–water separation with surface modified sponge for cleanup of oil spills

Abstract

A continuous in situ oil–water separation technique for cleanup of oil spills has been designed using surface modified polyurethane (PU) sponges as sorbents. By connecting PU sponges to a vacuum pump, they can continuously separate various kinds of oil and organic pollutant, including engine oil, diesel oil, peanut oil, liquid paraffin and hexane, from the surface of water with negligible water take-up. The sorbents used in the technique were prepared by dip-coating commercial PU sponges with TiO₂ sol and n-octadecylthiol successively. After surface modification, the sponges change from hydrophilic to superhydrophobic and superoleophilic surface with the water contact angle measured to be 152°. The as-prepared sponges absorb a broad variety of oils and organic solvents with high oil absorption capacity (80–110 g g⁻¹) and negligible water take-up at both static and dynamic conditions. The success of the continuous in situ oil–water separation technique is based on the excellent oil–water selectivity and oil adsorption capacity of the modified sponges.

---

1. Introduction

Oil is one of the most important energy sources. The modern economy with an increasing demand for energy has prompted offshore oil exploration and transportation, which results in unexpected disasters such as oil spills.¹ For example, on the 20th of April, 2010, the Mexico Gulf incident spilled 5 million barrels of crude oil into the sea and killed huge populations of marine animals.² Oil spills also have economical impacts on mariculture and tourism, in addition to the loss of oil.^3,4 Current cleanup technologies for oil spills include in situ burning, enhanced bioremediation, chemical dispersion, skimming, and mechanical extraction.^5–8 However, each of the above techniques has its own disadvantages. For example, in situ burning may lead to secondary pollution, skimming requires a successive oil–water separation process, and mechanical extraction is time- and energy-consuming because of the absorption-squeezing cycles. It is an urgent requirement to develop new techniques with high efficiency for continuous in situ separation of oil from the water surface.

Recently, superhydrophobic and superoleophilic filtration materials, as well as superhydrophilic and superoleophobic materials in the forms of meshes, films, and membranes, have attracted broad attention because of their capacity for direct separation of oils or organic solvents from water.^9–15 The possibility of applying such filtration materials for the in situ separation of spilled oil from water surface has also been studied. The research in this area is still in the initial stage. Significant research efforts are required to further develop the materials and technique towards real applications in cleanup of oil spills. In the latest publications, continuous in situ separation techniques have been reported and proved to be quite promising in the applications of oil spill cleanup.^16,17 High performance sorbents with excellent oil–water selectivity and high absorption capacity are crucial to the continuous in situ separation techniques. Among the existing sorbents, inorganic mineral sorbents and organic natural sorbents are abundant and cheap but have poor oil–water selectivity. Synthetic polymeric sorbents, with commercial polypropylene fibrous mats as typical examples, have very good oil–water selectivity, but their absorption capacity is quite low. Oil sorbents produced by electrospun polymeric fibers have been reported to exhibit high absorption capacity (up to 145 g g⁻¹) and oil–water selectivity at the same time.^18,19 However, these electrospun oil sorbents cannot stand the suction force of the continuous in situ separation process due to their poor mechanical property. Newly developed three dimensional (3D) porous materials, such as carbon nanotube sponges and graphene aerogels, are reported to have high absorption capacity and recyclability.^20–23 It is unfortunate that carbon nanotube sponges and graphene aerogels can only be prepared at lab scale for the time being with prohibitive high cost. Polyurethane (PU) sponges have excellent elastic property and absorption capacity and can be produced on a large scale with low cost. However, oil–water selectivity of the unmodified PU sponges is not acceptable due to their hydrophilic property.^24–27 Therefore, it is necessary to modify the PU sponge to utilize it as a continuous in situ oil–water separation material for cleanup of oil spills.

PU sponge can be changed from hydrophilic to hydrophobic through certain modification methods. For example, Liu et al. modified the PU sponge by dip-coating the sponge with graphene oxide suspension, and then reduced it with hydrazine to obtain high oil–water selectivity.²⁸ Wang and Lin fabricated CNT/PDMS coated PU sponge by dip-coating PU sponge in CNT/PDMS suspension and then curing at 120 °C for 6 h to obtain superhydrophobic property.²⁹ Zhou et al. fabricated a superhydrophobic sponge by dip-coating the PU sponge with FeCl₃/perfluorooctyltriethoxysilane and successive vapor-phase deposition of a layer of polypyrrole on the sponge surface.³⁰ However, the abovementioned modification methods are relatively complex and costly. In comparison, TiO₂ sol is a cheaper raw material available at large scale. Dip-coating with TiO₂ sol can lead to attachment of TiO₂ nanoparticles on the surface of PU sponges. The nanostructures formed by these TiO₂ nanoparticles may increase the surface roughness and thus change the wettability of the modified sponges. Therefore, it would be interesting to prepare such materials and study their properties as an oil–water separation material. However, to our knowledge, TiO₂ nanoparticle modified PU sponges have not been reported as oil sorbents.

In the current work, we present a preliminary design of a separation technique to demonstrate the possibility of applying a direct, in situ, continuous process for the oil spill cleanup operation. The sorbents used in this technique are surface modified PU sponges, which are treated with TiO₂ sol and n-octadecylthiol successively. We choose n-octadecylthiol because the strong covalent bond between sulfur and the transition-metal element³¹ make thiols the preferred modifying agents for TiO₂-coated PU sponges. Furthermore, the octadecyl can impart the modified sponges with superhydrophobic property. The properties of the modified sponges as oil sorbents, including oil–water selectivity, absorption capacity and recyclability, are also investigated. The entire process, as well as the preparation of the modified PU sponges, is easy to scale up.

2. Materials and methods

2.1. Materials

Original PU sponges were purchased from Duxiu Commodity Factory of Jinhua (Zhejiang province, China). TiO₂ sol was supplied by Fuzhou University (Fujian province, China). Ethanol, n-octadecylthiol (97%), sodium hydrate, hydrochloric acid, liquid paraffin, hexane, chloroform and oil red were obtained from Aladdin Reagent. Gasoline and diesel oil were obtained from Sinopec. Engine oil was obtained from Shell Oil Company. Peanut oil was obtained from the supermarket.

2.2. Preparation and characterization of modified sponge

Before the modification process, the original PU sponge was ultrasonically cleaned in ethanol and deionized water successively for 30 min to remove surface stains and oils. Then, blocks of original PU sponges (45 × 20 × 20 mm) were soaked in 6 mmol L⁻¹ TiO₂ sol (500 mL) for 30 min at room temperature, and then centrifuged at 1000 rpm to remove liquid and dried at 60 °C for 12 h. Subsequently, the as-dried sponges were immersed in 200 mL of 4 mmol L⁻¹ n-octadecyl thiol/ethanol solution for 12 h at 25 °C. The sponges were then centrifuged at 1000 rpm, washed with anhydrous ethanol 3 times, and dried at 60 °C for 12 h to obtain the superhydrophobic and superoleophilic sorbents.

The internal geometric structure and fiber surface morphology of the modified sponges, as well as the original sponges, were studied using a scanning electron microscope (SEM, FESEM-6700). The static contact angles were measured using an OCA20 contact angle system.

2.3. Continuous oil–water separation process and absorption capacity tests

The device used for continuous oil–water separation is presented in Fig. 1. The modified PU sponge was crammed into one end of a hosepipe (inner diameter of 10 mm), and then fixed at the oil–water interface. The other end of the hosepipe was inserted into a filter flask, and the flask was connected to a vacuum pump. In the oil–water separation testing, the vacuum pump was turned on, oil was continuously absorbed and removed from the water surface through the modified PU sponge and finally collected into the flask. The collected oil was then weighed, and the water content was measured based on a titration method according to a study,³² and the detailed experimental procedures are provided in the ESI.†

Fig. 1 The device used for continuous oil–water separation.

Absorption capacity tests were performed at 20 ± 4 °C. For each test, five independent experiments were conducted to obtain the average value. In a typical procedure, a certain amount of oil was poured into a 500 mL beaker containing 250 mL 0.01 mol L⁻¹ NaOH aqueous solution to obtain an oil film of 3–4 mm. Then, a block of modified sponge was weighed and placed into the beaker. For a dynamic test, the system was stirred with a magnetic stirrer at a constant speed of 500 rpm, whereas for a static test, the magnetic stirrer was turned off. After a 60 min absorption process, the sponge was removed and allowed to drain for 2 min, and then squeezed to obtain the liquid. Water content of the liquid was determined based on a titration method according to the (ref. 32) (see ESI†). The oil absorption capacity and water absorption capacity of the sponge was determined by the following equations:

q_o = [m_f − (m₀ + m_w)]/m₀ and q_w = m_w/m₀

where q_o is the oil absorption capacity (g g⁻¹), q_w is the water absorption capacity (g g⁻¹) or water take-up, m_f is the weight of the wet sponge after 2 min of drainage (g), m₀ is the initial weight of the modified sponge (g), and m_w is the weight of the absorbed water (g).

3. Results and discussion

3.1. Characteristics of the original and modified sponges

Fig. 2a shows the SEM image of the original PU sponge. The original PU sponge is composed of three-dimensional hierarchical porous structures with pore sizes in the range of 200–300 μm. The fiber surface is flat and smooth without any observable nanostructure (inset in Fig. 2a). The porous and interconnected framework of the sponge provides an extremely large volume for the entrance, and storage of liquids thus allows a high uptake capacity for oil and/or water. However, the original PU sponge contains lots of ether, carbamate and amide groups (Fig. S1 in ESI†), which results in a remarkable hydrophilic property. When a water droplet or gasoline droplet reaches the surface of the original sponge, both of them could completely spread into the pores of the sponges within several seconds (Fig. 2b). If the PU sponge is immersed in a water bath under an external force, it can be quickly filled with water (Fig. 2c). It is essential to modify the surface of the PU sponge so that they can be used as oil–water separation material for cleanup of oil spills.

Fig. 2 Characteristics of the original and modified PU sponge. (a) SEM images of the original sponge. Inset: enlarged view of the fiber surface. (b) Image of gasoline (dyed with oil-red) and water droplet rapidly absorbed by the original sponge. (c) Image of the original sponge immersed in a water bath under an external force. (d) SEM images of the modified sponge. Inset: enlarged view of the fiber surface. (e) Image of the gasoline droplet (dyed with oil-red) rapidly absorbed by the modified sponge and water droplet with spherical shape standing on the modified sponge. Inset: water contact-angle of the modified sponge. (f) The modified PU sponge exhibiting a silver mirror-like surface when immersed in a water bath under an external force.

After being treated with TiO₂ sol and n-octadecyl thiol successively, the 3D porous structure of the sponge remains unchanged (Fig. 2d). However, the morphology of the fiber surface of the modified sponge (inset in Fig. 2d) is completely different from that of the original sponge (inset in Fig. 2a). It is full of rough nanostructures, whereas the surface of the original sponge is smooth. These rough nanostructures are caused by thin coatings of TiO₂ nanoparticles and n-octadecylthiol, which can be confirmed by the FT-IR spectra (Fig. S1†) and energy dispersive spectra (Fig. S2†). In Fig. S2,† element Ti and S are shown in the EDS spectrum of the modified sponge, indicating that TiO₂ nanoparticles and n-octadecylthiol have been coated on the fiber surface. The changes in the surface structure and the chemical composition of the fiber surface can alter the wettability of the sponge. For example, Feng reported that deposition of TiO₂ nanoparticles, followed by modification with octadecylphosphonic acid, imparted copper mesh with superhydrophobicity and superoleophilicity.³³ The wettability of the modified sponge is shown in Fig. 2e–f. When a gasoline droplet is placed on the surface of the modified sponge, it can completely spread into the sponge within 1 second, and no contact angle can be measured (Fig. 2e). A water droplet can no longer spread into the modified sponge but maintains the spherical shape and stands on the surface (Fig. 2e). The water contact angle of the modified sponge is measured to be 152° (inset in Fig. 2e). When a modified sponge is immersed in a water bath under an external force, the sponge surface is surrounded by air bubbles, exhibiting a silver mirror-like surface (Fig. 2f), and no water can spread into the sponge. The above results indicate that the modified sponge is superhydrophobic and superoleophilic and is ready to be used as the oil sorbent or separation material for cleanup of oil spills.

3.2. Continuous in situ oil–water separation property

A continuous in situ oil–water separation technique requires excellent sorbents that can selectively absorb oil and repel water. Furthermore, the sorbents should have high absorption capacity for oil storage to maintain continuous oil flow. When the sorbent is fixed at the oil–water interface, it soon saturates with oil. Then, the saturated sorbent is subjected to vacuum suction force through a pipe connected to a vacuum pump. The absorbed oil is then extracted from the sorbent and flows through the pipe towards a collector. Oil in the vicinity will then flow towards the sorbent and fill up the voids, and thus continuous oil flow is maintained. As the process continues, the oil film over the water surface becomes thinner and thinner until oil is removed completely and transferred into the collector. Within the entire process, water is completely rejected owing to the excellent oil–water selectivity of the sorbent.

To verify the above design, a testing system has been set up, as illustrated in Fig. 1. The results show that various kinds of oil and organic pollutant can be continuously separated from the water surface with this system. In a typical procedure, the modified sponge is fixed to the interface of gasoline (dyed with oil red) and water, and then it quickly absorbs gasoline and repels water completely due to its superhydrophobic and superoleophilic properties (Fig. 3a). Then, the vacuum pump is turned on and gasoline is continuously absorbed and removed from water surface through the modified sponge. A stream of gasoline is formed in the pipe, and the thickness of the gasoline film gradually decreases (Fig. 3b and Video S1 in ESI†). Finally, all the gasoline on the surface of water is completely cleaned, and the transparent and clean water left in the beaker shows little change in volume (about 150 mL) during the entire process (Fig. 3c). The gasoline recovered is collected into a filter flask without water droplets visible to the naked eye (Fig. 3d). By analyzing the water content in the collected gasoline it is found to be less than 0.004 g g⁻¹ (gram water per gram gasoline), suggesting excellent oil–water selectivity. Using this technique, engine oil, diesel oil, peanut oil, liquid paraffin and n-hexane have been continuously separated from the surface of water. This approach appears to be very suitable for the separation of considerable amount of oil pollutants from water surfaces.

Fig. 3 The process of the continuous absorption and removal of gasoline (dyed with oil red) from the water surface with the water line maintained at 150 ml (a–c) and no obvious water droplet in the collected oil (d).

In the continuous separation process, the separation rate can be influenced by various experimental factors.³⁴ Higher vacuum can increase the separation rate, which is estimated from the time (h) required to collect a certain volume (L) of oil through certain cross-section area (m²) of the sponge. The vacuum pressure applied in the current process is about 10–15 kPa. The separation rate for gasoline increases from about 3 × 10⁵ LMH (liter per square meter per hour) to about 5 × 10⁵ LMH when vacuum pressure decreases from 15 kPa to 10 kPa. Another factor that impacts the separation rate is the viscosity of oil. Higher viscosity results in lower separation rate. For diesel oil, whose viscosity is much higher than that of gasoline, the separation rate is estimated to be about 1 × 10⁵ LMH at 10 kPa. For n-hexane, whose viscosity is lower, the separation rate is estimated to be about 7 × 10⁵ LMH 10 kPa.

3.3. Absorption capacity of the modified sponge

The success of the continuous in situ oil–water separation technique is based on the performance of the sorbents. The excellent oil–water selectivity of the modified sponge has been demonstrated in Sections 3.1 and 3.2. In this section, the absorption process and absorption capacity of the modified sponge will be discussed.

The oil absorption process is illustrated in Fig. 4. Fig. 4a–c show the process of engine oil (dyed with oil-red) being absorbed by the modified sponge from an oil–water mixture in a static system. When the modified sponge is placed on the mixture, it selectively absorbs engine oil from the mixture (Fig. 4a and b). After 60 min, the modified sponge can absorb all the engine oil and still float on the water (Fig. 4c). In dynamic systems, testing carried out under constant stirring (500 rpm) shows that engine oil can also be selectively absorbed by the modified sponge (Fig. 4d–f). The modified sponge can also effectively absorb chloroform (dyed with oil-red), a high-density organic solvent, which is found deep under the water surface (Fig. 4g–i). When a piece of modified sponge is forced into the water, it exhibits a silver mirror-like surface, and water cannot spread into the sponges (Fig. 4h). When the modified sponge comes in contact with chloroform under the water surface, it quickly sucks all chloroform within a few seconds (Fig. 4i). This makes the modified sponge a promising candidate for absorbing and eliminating high-density organic solvents under the water surface.

Fig. 4 Cleanup of engine oil film (dyed with oil red) on water surface at static condition (a–c) and dynamic condition (d–f), as well as cleanup of chloroform (dyed with oil red) under water (g–i), by the modified sponge.

To investigate the absorption capacities of the modified sponges, several oils and organic liquids are tested at static and dynamic conditions according to the procedure presented in Section 2.3 (chloroform is tested by forcing the modified sponge into water to reach the organic liquid at the bottom of the beaker). The results are shown in Fig. 5. At static condition, the absorption capacities of modified sponges for engine oil, gasoline, diesel oil, peanut oil, liquid paraffin, hexane and chloroform are about 87, 94, 90, 89, 84, 87, and 110 g g⁻¹ (the errors are about 2–5 g g⁻¹), respectively. At dynamic condition (not eligible for chloroform), the corresponding absorption capacities are about 102, 99, 109, 106, 109, and 89 g g⁻¹ (the errors are about 2–7 g g⁻¹), respectively. It should be emphasized that the water absorption capacities of the modified sponge in all cases are less than 0.3 g g⁻¹ (gram water per gram sorbent) at both static and dynamic conditions (The results are not shown in Fig. 5). The abovementioned results indicate that the prepared oil sorbents have high oil–water selectivity and high oil absorption capacity for various oils and organic liquids. In real applications, the used sponges can be regenerated through a simple squeezing process. When the modified sponges are saturated, the adsorbed oil can be collected through a mechanical squeezing process, and the sponges are regenerated and ready for the next absorption operation. There is no significant reduction in the oil absorption capacity of the modified sponge after 10 cycles of the absorption-squeezing process.

Fig. 5 Absorption capacity of the modified sponge for various oils and organic liquids.

4. Conclusion

In summary, superhydrophobic and superoleophilic sponges were prepared by modifying commercial PU sponge with TiO₂ sol and n-octadecylthiol successively. The as-prepared sponge shows high oil–water selectivity and high absorption capacity (80–110 g g⁻¹) for a broad variety of oils and organic solvents at both static and dynamic conditions. Most important of all, a continuous in situ oil–water separation technique was designed by using the modified sponges as sorbents. It can continuously separate various kinds of oil or organic pollutant from the surface of water. The technique is very suitable for the separation of a large amount of oil pollutants from water surfaces.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (51172117), the Natural Science Foundation of Shandong Province (ZR2010EM035, ZR2013EMM003) and the Foundation of Qingdao Science and Technology (13-1-4-148-jch).

References

1. M. F. Fingas, The Basics of Oil Spill Cleanup, Lewis Publishers, London, 2001.
2. J. L. Schnoor, Environ. Sci. Technol., 2010, 44, 4833.
3. R. A. Kerr, E. Kintisch, L. Schenkman and E. Stokstad, Science, 2010, 328, 962.
4. F. Aguilera, J. Mendez, E. Pasaro and B. Laffon, J. Appl. Toxicol., 2010, 30, 291.
5. J. Aurell and B. K. Gullett, Environ. Sci. Technol., 2010, 44, 9431.
6. R. Boopathy, S. Shields and S. Nunna, Appl. Biochem. Biotechnol., 2012, 167, 1560.
7. M. Roulia, K. Chassapis, Ch. Fotinopoulos, Th. Savvidis and D. Katakis, Spill Sci. Technol. Bull., 2003, 8, 425.
8. P. Calcagnile, D. Fragouli, I. S. Bayer, G. C. Anyfantis, L. Martiradonna, P. D. Cozzoli, R. Cingolani and A. Athanassiou, ACS Nano, 2012, 6, 5413.
9. M. W. Lee, S. An, S. Latthe, C. Lee, S. Hong and S. Yoon, ACS Appl. Mater. Interfaces, 2013, 5, 10597.
10. Y. Cao, X. Zhang, L. Tao, K. Li, Z. Xue, L. Feng and Y. Wei, ACS Appl. Mater. Interfaces, 2013, 5, 4438.
11. S. Wang, M. Li and Q. Lu, ACS Appl. Mater. Interfaces, 2010, 2, 677.
12. D. Deng, D. P. Prendergast, J. MacFarlane, R. Bagatin, F. Stellacci and P. M. Gschwend, ACS Appl. Mater. Interfaces, 2013, 5, 774.
13. P. Duong, T. S. Chung, S. Wei and L. Irish, Environ. Sci. Technol., 2014, 48, 4537.
14. L. Xu, N. Liu, Y. Cao, F. Lu, Y. Chen, X. Zhang, L. Feng and Y. Wei, ACS Appl. Mater. Interfaces, 2014, 6, 13324.
15. Y. Chen, Z. Xue, N. Liu, F. Lu, Y. Cao, Z. Sun and L. Feng, RSC Adv., 2014, 4, 11447.
16. C. F. Wang and S. J. Lin, ACS Appl. Mater. Interfaces, 2013, 5, 8861.
17. J. Ge, Y. Ye, H. Yao, X. Zhu, X. Wang, L. Wu, J. Wang, H. Ding, N. Yong, L. He and S. Yu, Angew. Chem., Int. Ed., 2014, 126, 3686.
18. H. Zhu, S. Qiu, W. Jiang, D. Wu and C. Zhang, Environ. Sci. Technol., 2011, 45, 4527.
19. J. Lin, Y. Shang, B. Ding, J. Yang, J. Yu and S. Al-Deyab, Mar. Pollut. Bull., 2012, 64, 347.
20. X. C. Gui, H. B. Li, K. L. Wang, J. Q. Wei, Y. Jia, Z. Li, L. L. Fan, A. Y. Cao, H. W. Zhu and D. H. Wu, Acta Mater., 2011, 59, 4798.
21. H. Hu, Z. Zhao, W. Wan, Y. Gogotsi and J. Qiu, Adv. Mater., 2013, 25, 2219.
22. H. Bi, X. Xie, K. Yin, Y. Zhou, S. Wan, L. He, F. Xu, F. Banhart, L. Sun and R. S. Ruoff, Adv. Funct. Mater., 2012, 22, 4421.
23. H. Hu, Z. Zhao, W. Wan, Y. Gogotsi and J. Qiu, Environ. Sci. Technol. Lett., 2014, 1, 214.
24. V. O. A. Tanobel, T. H. D. Sydenstricker, S. C. Amico, J. V. C. Vargas and S. F. Zawadzki, J. Appl. Polym. Sci., 2009, 4, 1842.
25. H. Li, L. Liu and F. Yang, Mar. Pollut. Bull., 2012, 8, 1648.
26. D. Wu, L. Fang, Y. Qin, W. Wu, C. Mao and H. Zhu, Mar. Pollut. Bull., 2014, 1, 263.
27. H. Liu, Y. Wang, M. Yang and Q. He, J. Macromol. Sci., Pure Appl. Chem., 2014, 51, 88.
28. Y. Liu, J. Ma, T. Wu, X. Wang, G. Huang, Y. Liu, H. Qiu, Y. Li, W. Wang and J. Gao, ACS Appl. Mater. Interfaces, 2013, 5, 10018.
29. C. F. Wang and S. J. Lin, ACS Appl. Mater. Interfaces, 2013, 5, 8861.
30. X. Zhou, Z. Zhang, X. Xu, X. Men and X. Zhu, Ind. Eng. Chem. Res., 2013, 52, 9411.
31. B. Wang, J. Li, G. Wang, W. Liang, Y. Zhang, L. Shi, Z. Guo and W. Liu, ACS Appl. Mater. Interfaces, 2013, 5, 1827.
32. H. D. Liu, Z. Y. Liu, M. B. Yang and Q. He, J. Appl. Polym. Sci., 2013, 130, 3530.
33. C. Gao, Z. Sun, K. Li, Y. Chen, Y. Cao, S. Zhang and L. Feng, Energy Environ. Sci., 2013, 6, 1147.
34. P. H. H. Duong and T. S. Chung, J. Membr. Sci., 2014, 452, 117.

---

Footnote

† Electronic supplementary information (ESI) available: FT-IR spectra (Fig. S1) and energy dispersive spectra (Fig. S2) of the original and modified PU sponge. A video clip showing the continuous in situ oil–water separation. Experimental details for the determination of the water content. See DOI: 10.1039/c4ra07583h

---