science.org SCIENCEseparation can drastically cut down on the
energy requirement, compared to more tra-
ditional methods such as distillation ( 5 ).
Furthermore, the membranes can be easily
integrated to create a hybrid process using
existing infrastructures.
In nonaqueous conditions for the separa-
tion of small liquid organic molecules, the
stability required for the polymeric mem-
brane is extremely demanding. Besides hav-
ing to balance between flux and selectivity,
the membranes also need to be cheap and
easy to fabricate. Chisca et al. combined the
classical nonsolvent quenching technique
with a heat treatment to create a 10-nm-
thick polymeric membrane capable of sort-
ing complex hydrocarbon mixtures. Using
this technique, they synthesized hydroxyl-
functionalized polytriazole (PTA-OH)
membranes with a macroporous structure.
Subsequent heat treatment cross-linked the
polymer structure and simultaneously in-
duced densification in the membrane sur-
face. The layered and asymmetric structure
allows fast and selective sorting of complex
organic liquid mixtures on the basis of size
and shape, while the cross-linked structure
provides sufficient stabilities under a wide
range of organic solvents. The proposed
route is an exceptionally simple process re-
sembling the cellulose acetate membrane,
but can be made thinner and tougher.
In creating the membrane, Chisca et al.
focused on the fractionation of light crude
oil, which primarily consists of gasoline,
kerosene, and diesel and accounts for 60% of
global liquid fuel consumption ( 6 ). In their
proof-of-concept experiment, the cross-
linked polytriazole membrane allowed for
up to 95% permeate enrichment of hydro-
carbons with a carbon number lower than
C 10 , which matches that of gasoline. Based
on the tunability of the polytriazole mem-
branes, it is possible to create a cascade of
polytriazole membranes crosslinked at dif-
ferent temperatures to achieve highly selec-
tive enrichment of other components such
as kerosene and diesel.
One of the exciting aspects of this work is
that the membrane performs crude fraction-
ation in the organic solvent nanofiltration
(OSN) pressure range—at <15 bars, which is
lower than that of organic solvent reverse
osmosis (OSRO) (typically over 30 bars).
Other researchers have proposed an “OSRO
trade-off ” curve to compare membrane ma-
terials for organic solvent separation, which
shows the upper limits for membrane per-
meability and selectivity ( 7 ). Although the
comparison depends on the feed compo-
sition and upstream pressure, the cross-
linked polytriazole membrane can reject
up to 60% of one of the standard marker
solutes (1,3-diisopropylbenzene, 162.26
g mol–1). This selectivity is comparable to
those of state-of-the-art OSRO membranes
(8 –10). In addition, the polytriazole mem-
branes of Chisca et al. have permeances for
pure solvents (10 to 30 liter m–2 hour–1 bar–1)
for methanol, acetone, tetrahydrofuran, and
toluene that are similar to those of state-of-
the-art OSN membranes, such as polyamide
and polyarylate membranes (11, 12). These
numbers suggest that the membranes of
Chisca et al. may have both high solvent
flux and high selectivity. By combining ex-
isting technologies and sorting complex hy-
drocarbon mixtures, this research counters
complexity in both membrane manufacture
and membrane separation.
Although membrane-based petroleum
refining is a relatively new technology, it
can potentially change the conventional
separation processes performed in the
chemical industries. Global decarboniza-
tion requirements will only accelerate thewidespread use of this technology. To aid
process engineers in the difficult task of
switching from traditional technologies to
advanced membrane separations, upscaling
from lab-scale to large-scale membrane pro-
cesses in an economical manner is a criti-
cal step. Process modeling, combined with
better thermodynamic models ( 13 ), could
help with these upscaling initiatives by of-
fering predictive outcomes from membrane
processes regarding optimal operations
and techno-economical features. Although
membranes still face practical challenges
before they can fully meet industrial needs,
new potential in present and developing ap-
plications and a growing selection of mem-
brane materials are encouraging. jREFERENCES AND NOTES- K. A. Thompson et al., Science 369 , 310 (2020).
- S. Chisca et al., Science 376 , 1105 (2022).
- US Energy Information Administration (EIA),
International energy outlook (2021); https://www.eia.
gov/outlooks/ieo/.
4 R. P. Lively, AIChE J. 67 , e17286 (2021). - US National Academy of Sciences, Engineering,
and Medicine, A Research Agenda for Transforming
Separation Science (National Academies Press, 2019). - International Energy Agency (IEA), Oil Market
Report, (2021); https://www.iea.org/reports/
oil-market-report-december-2021. - The Lively Lab, OSRO upper bound (2022); https://
lively.chbe.gatech.edu/osro-upper-bound. - H. Jang et al., AIChE J. 65 , 431 (2019).
- E. K. McGuinness, F. Zhang, Y. Ma, R. P. Lively, M. D.
Losego, Chem. Mater. 31 , 5509 (2019). - W. Kushida et al., J. Mater. Chem. A Mater. Energy
Sustain. 10 , 4146 (2022). - S. Karan, Z. Jiang, A. G. Livingston, Science 348 , 1347
(2015). - M. F. Jimenez-Solomon, Q. Song, K. E. Jelfs, M. Munoz-
Ibanez, A. G. Livingston, Nat. Mater. 15 , 760 (2016). - K. P. Bye, M. Galizia, J. Membr. Sci. 603 , 118020 (2020).
ACKNOWLEDGMENTS
D.-Y.K acknowledges funding from the Basic Science
Research Program administered by the Korea National
Research Foundation (NRF-2021R1C1C1012014).10.1126/science.abq3186 PHOTO: JOSHUA HICKS/ISTOCKPHOTO.COMTraditional crude oil refineries consume a lot of energy to power the separation process. Membrane-based crude oil fractionation can reduce the carbon footprint of refineries.
1054 3 JUNE 2022 • VOL 376 ISSUE 6597