Molecules,
Год журнала:
2022,
Номер
27(14), С. 4489 - 4489
Опубликована: Июль 13, 2022
The
aim
of
this
work
was
to
develop
a
simple
and
easy-to-apply
model
predict
the
pH
values
deep
eutectic
solvents
(DESs)
over
wide
range
that
can
be
used
in
daily
work.
For
purpose,
38
different
DESs
were
measured
(ranging
from
0.36
9.31)
mathematically
interpreted.
To
mathematical
models,
first
numerically
described
using
σ
profiles
generated
with
COSMOtherm
software.
After
DESs’
description,
following
models
used:
(i)
multiple
linear
regression
(MLR),
(ii)
piecewise
(PLR),
(iii)
artificial
neural
networks
(ANNs)
link
experimental
descriptors.
Both
PLR
ANN
found
applicable
very
high
goodness
fit
(R2independent
validation
>
0.8600).
Due
good
correlation
predicted
values,
profile
could
as
DES
molecular
descriptor
for
prediction
their
values.
ACS Omega,
Год журнала:
2022,
Номер
7(36), С. 32194 - 32207
Опубликована: Сен. 1, 2022
Studies
on
deep
eutectic
solvents
(DESs),
a
new
class
of
"green"
solvents,
are
attracting
increasing
attention
from
researchers,
as
evidenced
by
the
rapidly
growing
number
publications
in
literature.
One
main
advantages
DESs
is
that
they
tailor-made
and
therefore,
potential
extremely
large.
It
essential
to
have
computational
methods
capable
predicting
physicochemical
properties
DESs,
which
needed
many
industrial
applications
research.
Surface
tension
one
most
important
required
applications.
In
this
work,
we
report
relatively
generalized
artificial
neural
network
(ANN)
for
surface
DESs.
The
database
used
can
be
considered
comprehensive
because
it
contains
1571
data
points
133
different
DES
mixtures
520
compositions
prepared
18
ions
63
hydrogen
bond
donors
temperature
range
277-425
K.
ANN
model
uses
molecular
parameter
inputs
derived
conductor-like
screening
real
(Sσ-profiles).
training
testing
results
show
best
performing
architecture
consisted
two
hidden
layers
with
15
neurons
each
(9-15-15-1).
proposed
was
excellent
R2
values
0.986
0.977
were
obtained
testing,
respectively,
an
overall
average
absolute
relative
deviation
2.20%.
models
represent
initiative
promote
development
robust
based
only
parameters,
leading
savings
investigation
time
resources.
ACS Sustainable Chemistry & Engineering,
Год журнала:
2023,
Номер
11(26), С. 9564 - 9580
Опубликована: Июнь 16, 2023
Deep
eutectic
solvents
(DESs)
are
a
new
class
of
environmentally
friendly
that
have
attracted
the
attention
many
researchers.
Since
DESs
several
practical
applications
in
CO2
capture,
knowledge
their
solubility
is
crucial.
In
this
study,
was
predicted
via
multilayer
perceptron
(MLP)
using
molecular
descriptors
derived
from
Conductor-like
Screening
Model
for
Real
Solvents
(COSMO-RS).
An
extensive
database
2327
data
points
created
94
unique
DES
mixtures
made
2
anions,
17
cations,
and
39
hydrogen
bond
donors
(HBDs)
at
150
different
compositions
operating
conditions
temperatures
pressures.
Several
statistical
tests
were
performed,
after
thorough
hyperparameter
tuning,
it
found
best
MLP
architecture
with
an
R2
value
0.986
±
0.002
average
absolute
relative
deviation
(AARD)
4.504
0.507.
The
has
also
been
loaded
into
accessible
Excel
spreadsheet
included
Supporting
Information.
Thereafter,
order
to
guide
design
achieving
high
solubilities,
utilized
high-throughput
screening
1320
combinations.
This
model
encourages
creation
robust
accurate
models
predict
novel
DESs,
which
will
minimize
need
conducting
costly
time-consuming
experiments.
Materials Science and Engineering R Reports,
Год журнала:
2024,
Номер
159, С. 100798 - 100798
Опубликована: Май 7, 2024
In
the
flourishing
field
of
materials
science
and
engineering,
ionic
liquids
(ILs)
stand
out
for
their
advantageous
features,
unique
tunable
properties,
environmentally
friendly
attributes,
making
them
ideal
candidates
various
applications.
However,
enormous
diversity
ILs
presents
a
challenge
that
has
traditionally
been
addressed
through
extensive
experimental
work.
this
study,
computational
approach
combines
robust
molecular
modeling
advanced
ensemble
deep
learning
is
employed.
This
proof-of-concept
allows
simultaneous
prediction
multiple
properties
ILs,
thereby
enabling
simplified
pathway
to
eco-efficient
inverse
solvent
design.
Based
on
an
dataset
from
ILThermo
with
73,847
data
points
2917
1213
references
using
insightful
features
derived
COSMO-RS,
8
machine
algorithms
were
used
predict
physical
ILs.
Artificial
Neural
Networks
(ANNs)
have
proven
be
optimal
choice
based
results
obtained.
The
ANN
model
was
carefully
tuned,
resulting
in
total
11,241
parameters
exhibited
remarkable
predictive
ability
R2
values
0.993,
0.907,
0.931,
0.875
density,
viscosity,
surface
tension,
melting
temperature,
respectively.
A
feature
study
screening
303,880
obtained
by
combining
all
possible
pairs
set
1070
cations
284
anions
(1070×284).
demonstrates
pragmatic
identifying
different
property
profiles
significantly
narrow
spectrum
validation.
screening,
open-source
"Inverse
Designer
Tool"
developed
as
database
filter
explore
user-defined
criteria,
facilitating
identification
promising
IL
specific
presented
here
open
door
new
exploration
application
catalyze
integration
industrial
fields
potential
solvents.
