Three-Dimensional and Lamellar Graphene Oxide Membranes …
101
Other computational approaches, such as those based on finite element methods,
have also been employed to study mechanical properties of nanostructures. For
instance, numerical simulations based on finite element methods were used together
with Raman spectroscopy to determine Young’s modulus of single- and bilayer
graphene [100]. Interestingly, electronic structure calculations, Monte Carlo, and
MD simulations have shown that the oxidation process significantly decreases the
in-plane Young’s modulus and fracture strength of GO [101]. Especially nanoporous
GO has shown stress/strain slope about one-half that of ideal graphene, with a fracture strength of about one-third of ideal graphene. This is particularly interesting
because high-resolution TEM observations have shown that reduced GO platelets
present quite large holes [102].
Computer simulations have been at the forefront of nanofluidics, contributing
to the establishment of measurement protocols and the manufacture of new materials—currently two of the most challenging experimental endeavors. Notably, MD
simulations were used over the past decades to predict and explain high water flux
rates and ionic selectivities of few-atom-layered materials. For instance, theoretical
works anticipated the continuous flow of water through graphene with pore diameters smaller than 1 nm [103]. The same computational framework was used to show
that a single-layer graphene with defect pores was able to reject salt and still perform
a water flux two to three orders of magnitude higher than that of traditional RO
membranes [104]. Enhanced water flow in GO and rGO has also been reported in
computational and experimental works [105].
Additionally, the structure of GO membranes has been modified to enhance both
water flux and rejection of salt ions and dye molecules. Further theoretical analysis
revealed that the fast water transport found in GO membranes can be attributed to
the presence of oxidized and pristine parts in GO sheets [106]. The oxidized region
causes adjacent microcrystals to separate and inhibits their reunion, allowing water
to flow between the GO layers, while the non-oxidized region provides a capillary
network that creates high capillary pressure. The smooth surface of the pristine region
enables water to flow under this pressure without friction—previous simulations
already showed the high slip length of graphene nanochannels to largely increase
water molecules’ speed [107]. The underlying mechanism can then be understood
as a capillary-like pressure due to a combination of pristine regions (where water
molecules can move faster) and oxidized regions where hydrogen bond formation
is a driving force for water to move through the membrane. Other structural factors,
such as different kinds of defects, valleys, folds, cavitations, and voids, also subsidize
the reported high water flux through GO films [108].
4.1 Graphyne-Based Membranes
An emblematic contribution of computer simulations to membrane science is the
prediction of graphyne by Baughman et al. [23]. They described a lattice of
benzene rings (sp
2 hybridized) linked together with acetylenic bonds (−C≡C−,
101
Other computational approaches, such as those based on finite element methods,
have also been employed to study mechanical properties of nanostructures. For
instance, numerical simulations based on finite element methods were used together
with Raman spectroscopy to determine Young’s modulus of single- and bilayer
graphene [100]. Interestingly, electronic structure calculations, Monte Carlo, and
MD simulations have shown that the oxidation process significantly decreases the
in-plane Young’s modulus and fracture strength of GO [101]. Especially nanoporous
GO has shown stress/strain slope about one-half that of ideal graphene, with a fracture strength of about one-third of ideal graphene. This is particularly interesting
because high-resolution TEM observations have shown that reduced GO platelets
present quite large holes [102].
Computer simulations have been at the forefront of nanofluidics, contributing
to the establishment of measurement protocols and the manufacture of new materials—currently two of the most challenging experimental endeavors. Notably, MD
simulations were used over the past decades to predict and explain high water flux
rates and ionic selectivities of few-atom-layered materials. For instance, theoretical
works anticipated the continuous flow of water through graphene with pore diameters smaller than 1 nm [103]. The same computational framework was used to show
that a single-layer graphene with defect pores was able to reject salt and still perform
a water flux two to three orders of magnitude higher than that of traditional RO
membranes [104]. Enhanced water flow in GO and rGO has also been reported in
computational and experimental works [105].
Additionally, the structure of GO membranes has been modified to enhance both
water flux and rejection of salt ions and dye molecules. Further theoretical analysis
revealed that the fast water transport found in GO membranes can be attributed to
the presence of oxidized and pristine parts in GO sheets [106]. The oxidized region
causes adjacent microcrystals to separate and inhibits their reunion, allowing water
to flow between the GO layers, while the non-oxidized region provides a capillary
network that creates high capillary pressure. The smooth surface of the pristine region
enables water to flow under this pressure without friction—previous simulations
already showed the high slip length of graphene nanochannels to largely increase
water molecules’ speed [107]. The underlying mechanism can then be understood
as a capillary-like pressure due to a combination of pristine regions (where water
molecules can move faster) and oxidized regions where hydrogen bond formation
is a driving force for water to move through the membrane. Other structural factors,
such as different kinds of defects, valleys, folds, cavitations, and voids, also subsidize
the reported high water flux through GO films [108].
4.1 Graphyne-Based Membranes
An emblematic contribution of computer simulations to membrane science is the
prediction of graphyne by Baughman et al. [23]. They described a lattice of
benzene rings (sp
2 hybridized) linked together with acetylenic bonds (−C≡C−,