Nano-Porous Graphene as Free-Standing Membranes
71
on the graphene membrane by O 2 plasma treatment which led to a pore size of
0.5–1 nm and a density of around 10
12 pore/cm
2 . Over 200 samples were fabricated
in this work, with 20% success, meaning 80% of the samples were ruptured with
either large or small tears and ruled out of the measurements. The tearless samples
gained 93–100% rejection of 6 mM KCl. Their study also provided a reproducible
protocol for pore generation by tuning the power and time of O 2 plasma evidenced
in Raman spectroscopy spectra and analyses of D and G peaks intensities. Water
permeability in Surwade work was exceedingly high, several orders of magnitude
higher than current RO membranes, when gravity-driven configuration was used.
Extremely high permeability was later observed by Kazemi et al., in 2019 [158,
159] when a similar configuration was used. It is likely that the water transport
mechanism in the configuration of water-filled feed-empty permeate is significantly
different from the configuration of water-filled feed-filled permeate [204]. Another
explanation is the role of the SiN/graphene interface and the small aperture size of
the SiN contribution to achieving such high permeance [26, 158, 159]. Note that the
water flux from osmotic diffusion dropped to much lower values due to ions binding
with unsaturated edges around pores that preceded the nanopores blockage [31].
In contrast to previous studies, graphene exfoliated by ribonucleic acid was used
as a raw material by Park et al. [206]. Nanopores with a 5 nm nominal size was
formed with the assistance of Cu nanoparticles as a template during a guided HNO 3
etching procedure. At a feed pressure of 83 kPa, 1.662 × 10
4 L m
−2 h
−1 bar
−1 water
permeance was achieved. The salt rejection was measured at ~25% for Mn
2+ and
~20% for Fe
3+ ions by electron paramagnetic resonance. In 2017, FIB etching was
employed in a few layers of graphene to generate pores by Jang et al. [32]. This
was followed by O 2 plasma treatment to modify the surface chemistry of the porous
structure and further extend the dimension of the pore which generated a wide range
of pore-size from 0.18 nm up to 15 nm with and density of 1.67 × 10
13 pore/cm
2
in 50 s of etching. Interfacial polymerization was used to mitigate large intrinsic
tears in the graphene layer surface. About 98% rejection was obtained for MgSO 4
with water permeance comparable to other nanofiltration membranes. Faster water
transport rate was achieved for the membranes etched by the combination of ion
bombardment and plasma treatment.
On a different approach, Boutilier et al. [30] focused on designing an appropriate
porous support layer to minimize defects side effects. Using a resistance model, they
chose supports with similar resistance as the selective pores in graphene, which was
proved to be effective in isolating small defects and limiting leakages through large
defects [207]. They successfully fabricated a cm-scale membrane which was tested
up to 200 kPa. A year later, Yang et al. [208] employed the interwoven network of
single-walled carbon nanotubes (SWCNT) as a novel support layer to mechanically
reinforce nano-porous graphene membranes. This method assisted in avoiding solute
leakage during water desalination. A SWCNT layer was transferred over graphene,
and then covered by SiO 2 , as a mesoporous template, prior to O 2 plasma treatment. The resulting hybrid membranes exhibited pore sizes of 0.32 nm, 0.55 nm
and 1.14 nm when treated for <5 s to 20 s of O 2 plasma. Using crossflow nanofiltration, water desalination performance was assessed with these membranes. They
71
on the graphene membrane by O 2 plasma treatment which led to a pore size of
0.5–1 nm and a density of around 10
12 pore/cm
2 . Over 200 samples were fabricated
in this work, with 20% success, meaning 80% of the samples were ruptured with
either large or small tears and ruled out of the measurements. The tearless samples
gained 93–100% rejection of 6 mM KCl. Their study also provided a reproducible
protocol for pore generation by tuning the power and time of O 2 plasma evidenced
in Raman spectroscopy spectra and analyses of D and G peaks intensities. Water
permeability in Surwade work was exceedingly high, several orders of magnitude
higher than current RO membranes, when gravity-driven configuration was used.
Extremely high permeability was later observed by Kazemi et al., in 2019 [158,
159] when a similar configuration was used. It is likely that the water transport
mechanism in the configuration of water-filled feed-empty permeate is significantly
different from the configuration of water-filled feed-filled permeate [204]. Another
explanation is the role of the SiN/graphene interface and the small aperture size of
the SiN contribution to achieving such high permeance [26, 158, 159]. Note that the
water flux from osmotic diffusion dropped to much lower values due to ions binding
with unsaturated edges around pores that preceded the nanopores blockage [31].
In contrast to previous studies, graphene exfoliated by ribonucleic acid was used
as a raw material by Park et al. [206]. Nanopores with a 5 nm nominal size was
formed with the assistance of Cu nanoparticles as a template during a guided HNO 3
etching procedure. At a feed pressure of 83 kPa, 1.662 × 10
4 L m
−2 h
−1 bar
−1 water
permeance was achieved. The salt rejection was measured at ~25% for Mn
2+ and
~20% for Fe
3+ ions by electron paramagnetic resonance. In 2017, FIB etching was
employed in a few layers of graphene to generate pores by Jang et al. [32]. This
was followed by O 2 plasma treatment to modify the surface chemistry of the porous
structure and further extend the dimension of the pore which generated a wide range
of pore-size from 0.18 nm up to 15 nm with and density of 1.67 × 10
13 pore/cm
2
in 50 s of etching. Interfacial polymerization was used to mitigate large intrinsic
tears in the graphene layer surface. About 98% rejection was obtained for MgSO 4
with water permeance comparable to other nanofiltration membranes. Faster water
transport rate was achieved for the membranes etched by the combination of ion
bombardment and plasma treatment.
On a different approach, Boutilier et al. [30] focused on designing an appropriate
porous support layer to minimize defects side effects. Using a resistance model, they
chose supports with similar resistance as the selective pores in graphene, which was
proved to be effective in isolating small defects and limiting leakages through large
defects [207]. They successfully fabricated a cm-scale membrane which was tested
up to 200 kPa. A year later, Yang et al. [208] employed the interwoven network of
single-walled carbon nanotubes (SWCNT) as a novel support layer to mechanically
reinforce nano-porous graphene membranes. This method assisted in avoiding solute
leakage during water desalination. A SWCNT layer was transferred over graphene,
and then covered by SiO 2 , as a mesoporous template, prior to O 2 plasma treatment. The resulting hybrid membranes exhibited pore sizes of 0.32 nm, 0.55 nm
and 1.14 nm when treated for <5 s to 20 s of O 2 plasma. Using crossflow nanofiltration, water desalination performance was assessed with these membranes. They