Review articleMolecular simulations on graphene-based membranes
Graphical abstract
Introduction
Membrane separation technology is very crucial for many industrial processes, such as water purification and gas separation. Permeance, the mole number of permeated molecules per unit membrane area in per unit time at per unit driving pressure, is an essential parameter for characterizing the permeation abilities of permeating species through separation membranes, such as molecules and ions. For the separation membranes, high permeance is always very important for improving the separation efficiency. It is generally deemed that the membrane permeance is inversely proportional to the membrane thickness. Graphene is a two-dimensional carbon material with a thickness of only one-atom-layer and composed of sp2-bonded and honeycomb-packed atoms. Thus, graphene and its derivatives [[1], [2], [3], [4], [5], [6], [7]], which have the thinnest thickness known to science, are becoming an easy choice as promising separation membranes. Since the born of graphene, scientists put forward ongoing efforts to realize the graphene-based separation membranes; up to now, two totally different concepts of graphene-based membranes have been proposed. For the pristine graphene, any molecules or ions as small as helium cannot permeate through at room temperature, because the graphene sheet is packed so tightly with carbon atoms that only tiny spaces are left among the carbon atoms. In order to be employed as a separation membrane, graphene must be artificially introduced into selective pores to sieve the molecules based on the size-sieving effects. This kind of graphene with introduced selective nanopores, called as nanoporous graphene (NPG), is the first concept of graphene-based membranes. It has been widely demonstrated that the NPG membranes are very effective for the water purification and gas separation. The separation mechanism of NPG membranes is basically the molecular size-sieving effects of the nanopores in graphene sheets, as shown in Fig. 1. Graphene oxide (GO) is a derivative of graphene presenting layered structures and substantial oxidation groups (hydroxyl, carboxyl, epoxy, carbonyl etc.) asymmetrically functionalized on the graphene sheets and edges. GO sheets are the second proposed concept of graphene-based membranes and were also demonstrated to be competent for the water purification and gas separation. The molecular and ionic separation mechanisms of the GO membranes are basically the selective transport through the spacing channels between GO laminate sheets and through the defects and wrinkles in GO sheets (see Fig. 1).
Although the NPG membranes and GO membranes both have the characteristics of good chemical stability [[8], [9], [10], [11], [12]] and strong mechanical strength [[13], [14], [15], [16], [17], [18], [19], [20]] etc., they have their own advantages and disadvantages. From the permeance point of view, the permeances of NPG membranes with an atomic thickness are several orders of magnitude higher than those of GO membranes; from the membrane fabrication point of view, the fabrication of industrial-scale GO membranes is very easy both for the synthesis and scaling up, while the fabrication of NPG membranes is really challenging. For the NPG membranes, the top-down fabrication method which synthesizes the nanostructures by removing crystal planes from the original structures on the substrates includes the complicated processes of graphene transfer and pore generation; while the bottom-up method, which synthesizes the nanostructures on the substrates by stacking atoms, is more concise but very skillful. Because of their distinctive advantages, the NPG and GO membranes have different development paces. The GO membranes seem more promising currently for their easy fabrication; however, the NPG membranes are the ultimate aim of people owing to their ultrahigh permeance brought by the thinnest thickness known to science.
The NPG membranes are proposed early by Jiang et al. [2] in 2009, who firstly theoretically demonstrated that the NPG can be efficiently employed as the gas separation membranes. Afterward, several similar theoretical works [[21], [22], [23], [24]] further showed the possibility of NPG as efficient gas separation membranes. Then, the first experimental measurement on the gas transport through NPG membranes was performed by Koenig et al. [25] in 2012; they measured the transport rates of several different gases through the NPGs with micrometre areas by using the mechanical resonance and pressurized blister method. More excitedly, in 2014 Celebi et al. [26] measured the transport rates of gases across a double-layer NPG with an area of square millimeters. These earlier works coupling with the subsequent outstanding works [[27], [28], [29]] promoted the NPG gas separation membranes from concept to reality step by step. As for the water purification by NPG membranes, they follow closely with the NPG gas separation membranes; in 2012 Cohen-Tanugi et al. [3] firstly showed that the NPG can be applied as the water purification membranes to effectively filter NaCl salt from water. Then, several experimental works were successively conducted; in 2014 O'Hern et al. [30] measured the transport rates of different ions through the NPG membranes; in 2015 Surwade et al. [31] measured the salt rejection rate of the NPG membranes and showed a nearly 100% rejection rate and a high water permeation rate. Namely, the NPG water purification membranes are also developing rapidly although they were studied relatively late. Currently, the NPG membranes are really coming as many advanced experimental works [[32], [33], [34], [35], [36], [37], [38]] are conducted both for water purification and gas separation.
GO-based membranes for water purification and gas separation [[39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49]] were extensively tested, after Nair et al. [50] observed the unimpeded water permeation through the GO laminate sheets with a sub-micrometre thickness. For water purification, in 2013 Hu and Mi [44] showed that the GO membranes can allow the water permeation by flowing through the spaces among the GO sheets while reject the impurities by the mechanisms of size exclusion and charge effects. More detailed investigations were performed subsequently; Joshi et al. [49] reported that the transport rates of small ions through GO sheets were greatly higher than the theoretical values by thousands of times and that the GO laminate sheets can block the solutes with a hydrated radii larger than 4.5 Å; Huang et al. investigated [42] the water permeation rates through the GO sheets, with considering the effects of pH value, salt concentration and water pressure. While for the gas separation, Li et al. [48] showed that the GO membranes prepared by the filtration method presented a high selectivity up to 3400 and 900, respectively, for the separation of gaseous H2/CO2 and H2/N2 mixtures. Since then, many researchers explored the applications of GO-based membranes in other gas separation fields, e.g. CO2 capture [41]. Until now, many experimental works [[51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63]] are conducted to show that the GO membranes are applicable both for water purification and gas separation.
Although many challenges are still faced currently in the device development [64], a great progress has been achieved for the graphene-based membranes. The recent developments in the graphene-based membranes were properly summarized in several excellent review papers [[65], [66], [67], [68], [69]]. For the graphene-based membranes, experimental studies are usually difficult to conduct, but the simulations are always effective. Molecular simulations are the methods probing the transport phenomena from the molecular levels by considering the motion trajectories of various molecules. Hence, molecular simulation is a very effective tool to study the graphene-based membranes involving molecular separation processes in nanospaces. The molecular permeation mechanisms and the effects of various tedious factors, such as chemical functionalization and structural parameter, are especially expected to be revealed by molecular simulations. In this review, the recent advances in the molecular simulations on both NPG-based membranes and GO-based membranes for the applications in water purification and gas separations are summarized. The representative works are highlighted, accompanying with the discussions of the separation mechanisms, optimization measures and so on. It is expected that these highlights can promote people to understand from the molecular insights the current challenges in the applications of graphene-based membranes for water purification and gas separation. Meanwhile, the understanding of these cutting-edge studies can promote the development of other graphene-related carbon-based separation membranes [[70], [71], [72], [73]] and stimulate the applications of two-dimensional atomically thick membranes, especially the graphene-based membranes, in other fields such as isotope separation [[74], [75], [76], [77]], hydrogen storage and production [5,[78], [79], [80], [81], [82], [83]], and DNA sequencing [[84], [85], [86], [87], [88], [89]].
Section snippets
Water purification
The molecular simulation works were conducted considerably on the NPG membranes for water purification. Currently, the molecular simulations on graphene-based membranes are almost all based on the classical molecular dynamics (MD) methods without considering the quantum effects by quantum mechanical calculations. The classical MD method not only is competent for simulating the separation processes of graphene membranes in most cases, but also requires less computational resources comparing to
Water purification
As a derivative of graphene, in the GO membranes the edges and planes of graphene sheets are asymmetrically modified by oxygen-containing chemical functional groups, including hydroxyl, epoxy, carboxyl, carbonyl, phenol etc. The structures of GO membranes are more complex comparing to the NPG membranes, but the large-scale GO membranes can be easily prepared by some skillful processes, such as the filtration process [48,49], the spray-coating process [50] etc. Therefore, plenty of experimental
Conclusions and outlook
NPG and GO are two kinds of graphene-based two-dimensional highly-permeable membranes, presenting great potentials in the applications of water purification and gas separation. Molecular simulation is an effective tool for studying the molecular transport through the graphene-based membranes with nanoscale pores and channels, especially for revealing the underlying mechanisms for water purification and gas separation. In this perspective review, we summarize the recent molecular simulation
Conflicts of interest
There are no conflicts of interest to declare.
Acknowledgement
The financial supports from National Natural Science Foundation of China for general program (No. 51876169) and Distinguished Young Scientists (No. 51425603) are highly acknowledged.
References (193)
- et al.
Efficient helium separation of graphitic carbon nitride membrane
Carbon
(2015) - et al.
Graphene oxide based materials for desalination
Carbon
(2019) - et al.
The mechanism for the stability of graphene oxide membranes in a sodium sulfate solution
Chem. Phys. Lett.
(2013) - et al.
Preparation of chitosan/graphene oxide composite film with enhanced mechanical strength in the wet state
Carbohydr. Polym.
(2011) - et al.
Ultrathin, molecular-sieving graphene oxide membranes for selective hydrogen separation
Science
(2013) - et al.
Thermally-driven isotope separation across nanoporous graphene
Chem. Phys. Lett.
(2012) - et al.
Room-temperature hydrogen storage via two-dimensional potential well in mesoporous graphene oxide
Nano Energy
(2016) - et al.
Hydrogen storage in hybrid of layered double hydroxides/reduced graphene oxide using spillover mechanism
Energy
(2016) Influence of curvature on water desalination through the graphene membrane with Si-passivated nanopore
Comput. Mater. Sci.
(2016)- et al.
Molecular dynamics study on water desalination through functionalized nanoporous graphene
Carbon
(2017)