Abstract
Artificial
intelligence
(AI)
advancements
are
driving
the
need
for
highly
parallel
and
energy‐efficient
computing
analogous
to
human
brain
visual
system.
Inspired
by
brain,
resistive
random‐access
memories
(ReRAMs)
have
recently
emerged
as
an
essential
component
of
intelligent
circuitry
architecture
developing
high‐performance
neuromorphic
systems.
This
occurs
due
their
fast
switching
with
ultralow
power
consumption,
high
ON/OFF
ratio,
excellent
data
retention,
good
endurance,
even
great
possibilities
altering
resistance
biological
counterparts
applications.
Additionally,
advantages
photoelectric
dual
modulation
switching,
ReRAMs
allow
optically
inspired
artificial
neural
networks
reconfigurable
logic
operations,
promoting
innovative
in‐memory
technology
image
recognition
tasks.
Optoelectronic
architectured
can
simulate
functionalities,
such
light‐triggered
long‐term/short‐term
plasticity.
They
be
used
in
robotics
bionic
neurological
optoelectronic
Metal
oxide
(MOx)–polymer
hybrid
nanocomposites
beneficial
active
layer
bistable
metal–insulator–metal
ReRAM
devices,
which
hold
promise
memory
technology.
review
explores
state
art
storage,
advancement
materials,
mechanisms
selecting
appropriate
materials
layers
boost
flexibility,
density
while
lowering
programming
voltage.
Furthermore,
material
design
cum‐synthesis
strategies
that
greatly
influence
overall
performance
MOx–polymer
nanocomposite
performances
highlighted.
recent
progress
multifunctional
composites‐based
explored
synapses
emulate
visualization
memorize
information.
Finally,
challenges,
limitations,
future
outlooks
fabrication
composite
over
conventional
von
Neumann
systems
discussed.
Journal of Semiconductors,
Год журнала:
2025,
Номер
46(2), С. 021403 - 021403
Опубликована: Фев. 1, 2025
Abstract
To
address
the
increasing
demand
for
massive
data
storage
and
processing,
brain-inspired
neuromorphic
computing
systems
based
on
artificial
synaptic
devices
have
been
actively
developed
in
recent
years.
Among
various
materials
investigated
fabrication
of
devices,
silicon
carbide
(SiC)
has
emerged
as
a
preferred
choices
due
to
its
high
electron
mobility,
superior
thermal
conductivity,
excellent
stability,
which
exhibits
promising
potential
applications
harsh
environments.
In
this
review,
progress
SiC-based
is
summarized.
Firstly,
an
in-depth
discussion
conducted
regarding
categories,
working
mechanisms,
structural
designs
these
devices.
Subsequently,
several
application
scenarios
are
presented.
Finally,
few
perspectives
directions
their
future
development
outlined.
Advanced Science,
Год журнала:
2024,
Номер
unknown
Опубликована: Июль 25, 2024
Binocular
stereo
vision
relies
on
imaging
disparity
between
two
hemispherical
retinas,
which
is
essential
to
acquire
image
information
in
three
dimensional
environment.
Therefore,
retinomorphic
electronics
with
structural
and
functional
similarities
biological
eyes
are
always
highly
desired
develop
perception
system.
In
this
work,
a
optoelectronic
memristor
array
based
Ag-TiO
Advanced Functional Materials,
Год журнала:
2025,
Номер
unknown
Опубликована: Янв. 24, 2025
Abstract
Organic
memristors
have
emerged
as
promising
candidates
for
neuromorphic
computing
due
to
their
potential
low‐cost
fabrication,
large‐scale
integration,
and
biomimetic
functionality.
However,
practical
applications
are
often
hindered
by
limited
thermal
stability
device‐to‐device
variability.
Here,
an
organic
polymer‐based
memristor
using
a
thiadiazolobenzotriazole
(TBZ)
2,5‐Dioctyl‐3,6‐di(thiophen‐2‐yl)pyrrolo[3,4‐c]pyrrole‐1,4(2H,5H)‐dione
(DPP)‐based
conjugated
polymer
is
presented
that
exhibits
exceptional
reliable
resistance
switching
behavior
over
wide
temperature
range
(153–573
K).
The
device
leverages
charge‐transfer
mechanism
achieve
gradual
uniform
switching,
overcoming
the
challenges
associated
with
filamentary‐based
mechanisms.
memristor's
consistent
performance
enable
its
integration
into
various
applications,
including
image
processing.
device's
ability
demonstrated
effectively
deblur
images,
even
under
varying
conditions,
showcasing
robust
computing.
This
study
establishes
pathway
toward
high‐performance,
thermally
stable
advanced
artificial
intelligence
applications.
Advanced Materials,
Год журнала:
2025,
Номер
unknown
Опубликована: Март 25, 2025
The
ongoing
global
energy
crisis
has
heightened
the
demand
for
low-power
electronic
devices,
driving
interest
in
neuromorphic
computing
inspired
by
parallel
processing
of
human
brains
and
efficiency.
Reconfigurable
memristors,
which
integrate
both
volatile
non-volatile
behaviors
within
a
single
unit,
offer
powerful
solution
in-memory
computing,
addressing
von
Neumann
bottleneck
that
limits
conventional
architectures.
These
versatile
devices
combine
high
density,
low
power
consumption,
adaptability
positioning
them
as
superior
alternatives
to
traditional
complementary
metal-oxide-semiconductor
(CMOS)
technology
emulating
brain-like
functions.
Despite
their
potential,
studies
on
reconfigurable
memristors
remain
sparse
are
often
limited
specific
materials
such
Mott
insulators
without
fully
unique
reconfigurability.
This
review
specifically
focuses
examining
dual-mode
operation,
diverse
physical
mechanisms,
structural
designs,
material
properties,
switching
behaviors,
applications.
It
highlights
recent
advancements
low-power-consumption
solutions
memristor-based
neural
networks
critically
evaluates
challenges
deploying
standalone
or
artificial
systems.
provides
in-depth
technical
insights
quantitative
benchmarks
guide
future
development
implementation
computing.
Reliable
parylene–PbTe
memristors
controlled
via
electrical
and
optical
stimuli
replicate
key
synaptic
functions
are
applicable
in
neuromorphic
computing
systems.
Advanced Intelligent Systems,
Год журнала:
2025,
Номер
unknown
Опубликована: Март 31, 2025
Data‐intensive
computing
tasks,
such
as
training
neural
networks,
are
fundamental
to
artificial
intelligence
applications
but
often
demand
substantial
energy
resources.
This
study
presents
a
novel
complementary
metal‐oxide‐semiconductor
(CMOS)‐based
memcapacitor
framework
designed
address
these
challenges
by
enabling
efficient
and
robust
neuromorphic
computing.
Utilizing
devices,
crossbar
array
that
performs
parallel
vector‐matrix
multiplication
operations,
validated
through
cadence
simulations
implemented
in
python
for
scalable
accelerator
design,
is
developed.
The
demonstrates
outstanding
performance
across
classification
achieving
98.4%
accuracy
digit
recognition
85.9%
object
recognition.
A
key
aspect
of
this
research
its
focus
on
real‐world
fabrication
nonidealities,
including
up
30%
device
parameter
variations,
ensuring
robustness
reliability
under
practical
deployment
conditions.
results
emphasize
the
effectiveness
capacitance‐based
systems
handling
tasks
while
demonstrating
resilience
fabrication‐induced
variations.
work
establishes
foundation
scalable,
energy‐efficient,
memcapacitor‐based
advancing
potential
intelligent
intelligence‐driven
paving
way
future
innovations
Nanomaterials,
Год журнала:
2024,
Номер
14(19), С. 1573 - 1573
Опубликована: Сен. 29, 2024
The
traditional
computer
with
von
Neumann
architecture
has
the
characteristics
of
separate
storage
and
computing
units,
which
leads
to
sizeable
time
energy
consumption
in
process
data
transmission,
is
also
famous
“von
wall”
problem.
Inspired
by
neural
synapses,
neuromorphic
emerged
as
a
promising
solution
address
problem
due
its
excellent
adaptive
learning
parallel
capabilities.
Notably,
2016,
researchers
integrated
light
into
computing,
inspired
extensive
exploration
optoelectronic
all-optical
synaptic
devices.
These
optical
devices
offer
obvious
advantages
over
all-electric
devices,
including
wider
bandwidth
lower
latency.
This
review
provides
an
overview
research
background
on
discusses
their
implementation
principles
different
scenarios,
presents
application
concludes
prospects
for
future
developments.
Applied Physics Letters,
Год журнала:
2024,
Номер
125(17)
Опубликована: Окт. 21, 2024
Brain-inspired
neuromorphic
sensory
devices
play
a
crucial
role
in
addressing
the
limitations
of
von
Neumann
systems
contemporary
computing.
Currently,
synaptic
rely
on
memristors
and
thin-film
transistors,
requiring
establishment
read
voltage.
A
built-in
electric
field
exists
within
p–n
junction,
enabling
operation
zero-read-voltage
devices.
In
this
study,
we
propose
an
artificial
synapse
utilizing
ZnO
diode.
Typical
rectification
curves
characterize
formation
diodes.
diodes
demonstrate
distinct
properties,
including
paired-pulse
facilitation,
depression,
long-term
potentiation,
depression
modulations,
with
voltage
0
V.
An
neural
network
is
constructed
to
simulate
recognition
tasks
using
MNIST
Fashion-MNIST
databases,
achieving
test
accuracy
values
92.36%
76.71%,
respectively.
This
research
will
pave
way
for
advancing
