Scientific Reports,
Год журнала:
2023,
Номер
13(1)
Опубликована: Дек. 2, 2023
The
adsorption
of
carbon
dioxide
(CO2)
on
porous
materials
offers
a
promising
avenue
for
cost-effective
CO2
emissions
mitigation.
This
study
investigates
the
impact
textural
properties,
particularly
micropores,
capacity.
Multilayer
perceptron
(MLP)
neural
networks
were
employed
and
trained
with
various
algorithms
to
simulate
adsorption.
Study
findings
reveal
that
Levenberg-Marquardt
(LM)
algorithm
excels
remarkable
mean
squared
error
(MSE)
2.6293E-5,
indicating
its
superior
accuracy.
Efficiency
analysis
demonstrates
scaled
conjugate
gradient
(SCG)
boasts
shortest
runtime,
while
Broyden-Fletcher-Goldfarb-Shanno
(BFGS)
requires
longest.
LM
also
converges
fewest
epochs,
highlighting
efficiency.
Furthermore,
optimization
identifies
an
optimal
radial
basis
function
(RBF)
network
configuration
nine
neurons
in
hidden
layer
MSE
9.840E-5.
Evaluation
new
data
points
shows
MLP
using
bayesian
regularization
(BR)
achieves
highest
research
underscores
potential
deep
BR
training
process
simulation
provides
insights
into
pressure-dependent
behavior
These
contribute
our
understanding
processes
offer
valuable
predicting
gas
behavior,
especially
scenarios
where
micropores
dominate
at
lower
pressures
mesopores
higher
pressures.
Membranes,
Год журнала:
2023,
Номер
13(12), С. 898 - 898
Опубликована: Дек. 2, 2023
Carbon
dioxide
(CO2),
which
results
from
fossil
fuel
combustion
and
industrial
processes,
accounts
for
a
substantial
part
of
the
total
anthropogenic
greenhouse
gases
(GHGs).
As
result,
several
carbon
capture,
utilization
storage
(CCUS)
technologies
have
been
developed
during
last
decade.
Chemical
absorption,
adsorption,
cryogenic
separation
membrane
are
most
widely
used
post-combustion
CO2
capture
technologies.
This
study
reviews
latest
progress
in
processes
separation.
More
specifically,
objective
present
work
is
to
state
art
membrane-based
flue
focuses
mainly
on
recent
advancements
commonly
employed
materials.
These
materials
utilized
fabrication
application
novel
composite
membranes
or
mixed-matrix
(MMMs),
improved
intrinsic
surface
characteristics
and,
thus,
can
achieve
high
selectivity
permeability.
Recent
described
regarding
metal–organic
frameworks
(MOFs),
molecular
sieves
(CMSs),
nanocomposite
membranes,
ionic
liquid
(IL)-based
facilitated
transport
(FTMs),
comprise
MMMs.
The
significant
challenges
future
prospects
implementing
also
presented.
Global
warming
is
a
crisis
that
humanity
must
face
together.
With
greenhouse
gases
(GHGs)
as
the
main
factor
causing
global
warming,
adoption
of
relevant
processes
to
eliminate
them
essential.
advantages
high
specific
surface
area,
large
pore
volume,
and
tunable
synthesis,
metal–organic
frameworks
(MOFs)
have
attracted
much
attention
in
GHG
storage,
adsorption,
separation,
catalysis.
However,
pool
MOFs
expands
rapidly
with
new
syntheses
discoveries,
finding
suitable
MOF
for
particular
application
highly
challenging.
In
this
regard,
high-throughput
computational
screening
considered
most
effective
research
method
number
materials
discover
high-performance
target
MOFs.
Typically,
generates
voluminous
multidimensional
data,
which
well
suited
machine
learning
(ML)
training
improve
efficiency
explore
relationships
between
data
depth.
This
Review
summarizes
general
process
common
methods
using
ML
screen
field
removal.
It
also
addresses
challenges
faced
by
exploring
space
potential
directions
future
development
screening.
aims
enhance
understanding
integration
various
fields
broaden
ideas
