Neuromorphic Computing and Engineering,
Год журнала:
2024,
Номер
4(3), С. 032003 - 032003
Опубликована: Сен. 1, 2024
Abstract
The
growing
demand
for
artificial
intelligence
has
faced
challenges
traditional
computing
architectures.
As
a
result,
neuromorphic
systems
have
emerged
as
possible
candidates
next-generation
systems.
Two-dimensional
(2D)
materials-based
devices
that
emulate
biological
synapses
and
neurons
play
key
role
in
hardware
due
to
their
unique
properties
such
high
strength,
thermal
conductivity,
flexibility.
Although
several
studies
shown
the
simulations
of
individual
devices,
experimental
implementation
large-scale
crossbar
arrays
is
still
unclear.
In
this
review,
we
explore
working
principles
mechanisms
memristive
devices.
Then,
overview
development
based
on
2D
materials
including
transition
metal
dichalcogenides,
graphene,
hexagonal
boron
nitride,
layered
halide
perovskites.
We
also
highlight
requirement
recent
progress
building
by
utilizing
advantageous
materials.
Lastly,
address
currently
face
propose
path
towards
system-level
applications
computing.
Nano-Micro Letters,
Год журнала:
2024,
Номер
16(1)
Опубликована: Июнь 6, 2024
Abstract
Manipulating
the
expression
of
synaptic
plasticity
neuromorphic
devices
provides
fascinating
opportunities
to
develop
hardware
platforms
for
artificial
intelligence.
However,
great
efforts
have
been
devoted
exploring
biomimetic
mechanisms
simulation
in
last
few
years.
Recent
progress
various
modulation
techniques
has
pushed
research
electronics
from
static
dynamic
modulation,
improving
accuracy
computing
and
providing
strategies
implementing
sensing
functions.
Herein,
several
through
chemical
techniques,
device
structure
design,
physical
signal
are
reviewed.
For
underlying
modification
functional
materials
were
clarified
its
effect
on
was
also
highlighted.
Based
reconfigurable
operation
well
demonstrated
achieve
programmable
Besides,
integrating
sensory
units
with
processing
circuits
paved
a
new
way
human-like
intelligent
perception
under
signals
such
as
light,
strain,
temperature.
Finally,
considering
that
relevant
technology
is
still
basic
exploration
stage,
some
prospects
or
development
suggestions
put
forward
promote
devices.
Applied Physics Letters,
Год журнала:
2025,
Номер
126(1)
Опубликована: Янв. 6, 2025
Hafnium
oxide
(HfOx)
films
are
highly
valued
as
functional
layers
in
nonvolatile
resistive
switching
(RS)
memristors
due
to
their
scalability,
compatibility
with
CMOS
technology,
and
high
dielectric
constant.
However,
the
low
reliability
of
HfOx-based
is
key
factor
hindering
widespread
practical
applications.
Herein,
amorphous
HfOx
(a-HfOx)
used
construct
memristors,
nitrogen
treatment
strategy
employed
enhance
characteristics.
All
fabricated
Al/a-HfOx/ITO
demonstrate
bipolar
digital
RS
behaviors,
specifically,
500
°C-treated
a-HfOx
device
exhibits
reliable
performance,
including
cycle-to-cycle
variability,
concentrated
distributions
operating
voltages,
long-term
retention
capacity
(>104
s),
good
cycle
endurance
(>200
cycles).
The
mechanisms
physical
models
for
enhanced
performance
thoroughly
elucidated,
revealing
that
formation
stable
oxygen
vacancy–dinitrogen
complexes
confines
conductive
filament
path
significantly
reduces
randomness
during
rupture.
This
work
renders
an
effective
material
engineering
widening
a
toward
designing
data
storage
devices
striking
performances.
Advanced Materials,
Год журнала:
2025,
Номер
unknown
Опубликована: Янв. 21, 2025
Abstract
Mechanical
information
is
a
medium
for
perceptual
interaction
and
health
monitoring
of
organisms
or
intelligent
mechanical
equipment,
including
force,
vibration,
sound,
flow.
Researchers
are
increasingly
deploying
recognition
technologies
(MIRT)
that
integrate
acquisition,
pre‐processing,
processing
functions
expected
to
enable
advanced
applications.
However,
this
also
poses
significant
challenges
acquisition
performance
efficiency.
The
novel
exciting
mechanosensory
systems
in
nature
have
inspired
us
develop
superior
bionic
(MIBRT)
based
on
materials,
structures,
devices
address
these
challenges.
Herein,
first
strategies
pre‐processing
presented
their
importance
high‐performance
highlighted.
Subsequently,
design
considerations
sensors
by
mechanoreceptors
described.
Then,
the
concepts
neuromorphic
summarized
order
replicate
biological
nervous
system.
Additionally,
ability
MIBRT
investigated
recognize
basic
information.
Furthermore,
further
potential
applications
robots,
healthcare,
virtual
reality
explored
with
view
solve
range
complex
tasks.
Finally,
future
opportunities
identified
from
multiple
perspectives.
As
the
age
of
Internet
Things
(IoTs)
unfolds,
along
with
rapid
advancement
artificial
intelligence
(AI),
traditional
von
Neumann-based
computing
systems
encounter
significant
challenges
in
handling
vast
amounts
data
storage
and
processing.
Bioinspired
neuromorphic
strategies
offer
a
promising
solution,
characterized
by
features
in-memory
computing,
massively
parallel
processing,
event-driven
operations.
Compared
to
rigid
silicon-based
devices,
flexible
devices
are
lightweight,
thin,
highly
stretchable,
garnering
considerable
attention.
Among
materials
utilized
these
transition
metal
carbides/nitrides
(MXenes)
particularly
noteworthy
their
excellent
flexibility,
exceptional
conductivity,
hydrophilicity,
which
confer
remarkable
properties
upon
devices.
Herein,
comprehensive
discussion
is
provided
on
applications
MXenes
memory
This
review
covers
basic
principles
device
structures
common
parameters
emerging
as
well
synthesis,
functionalization
methods,
distinct
MXenes.
The
remaining
future
opportunities
relevant
also
presented.
can
serve
valuable
reference
lay
cornerstone
for
practical
feasible
implementation
technologies.
The Journal of Physical Chemistry Letters,
Год журнала:
2024,
Номер
15(34), С. 8667 - 8675
Опубликована: Авг. 19, 2024
Emerging
optoelectronic
memristive
devices
with
high
parallelism
and
low-power
consumption
have
made
neuromorphic
computing
hardware
a
tangible
reality.
The
coordination
of
conductivity
regulation
through
both
electrical
light
signals
is
pivotal
for
advancing
the
development
synaptic
memristors
brainlike
functionalities.
Here,
an
artificial
visual
synapse
presented
Ti3C2
MXene
memristor
which
demonstrates
not
only
nonvolatile
memory
effect
(Set/Reset:
0.58/–0.55
V;
Retention:
>103
s)
sustained
multistage
conductivity,
but
also
facile
modulation
electrical-
light-stimulated
behaviors.
By
adjusting
stimulus
parameters,
enables
realization
biosynaptic
excitatory
postsynaptic
current,
stable
long-term
facilitation/depression,
paired
pulse
facilitation,
spiking-timing-dependent
plasticity,
experiential
learning.
Particularly,
benefiting
from
distinguishable
photoconductive
effects
multiple
near-infrared
intensities
(7–13
mW/cm2),
potential
applications
in
nociceptive
perception
("threshold",
"noadaption",
"relaxation")
imaging
(e.g.,
"Superman"
cartoon
character)
infrared
environments
are
well
achieved
such
memristors.
These
results
hold
significant
implications
future
advancement
integrated
sensing,
memory,
nociception,
systems.
Research Square (Research Square),
Год журнала:
2025,
Номер
unknown
Опубликована: Янв. 6, 2025
Abstract
Novel
neuromorphic
imaging
arrays
integrate
photonic
perception,
memory,
and
process
capability,
enabling
intelligent
with
efficient
spatial
temporal
data
fusion
for
object
recognition,
enhancement,
motion
perception
in
autonomous
vehicles
surveillance
systems,
surpassing
the
limitations
of
conventional
charge-coupled
device
(CCD)
complementary
metal
oxide
semiconductor
(CMOS)
image
sensors.
Halide
perovskites
hold
significant
promise
due
to
their
capacity
simultaneously
manipulate
photogenerated
ions
electronic
charges,
thereby
facilitating
development
sophisticated
systems
based
on
intrinsic
material
dynamics.
However,
limited
response
range
(ultraviolet-visible
spectrum)
toxic
nature
lead
remain
unresolved
perovskite-based
applications.
Here,
we
present
lead-free
non-toxic
CH(NH2)2SnI3
(FASnI3)
low-toxicity
components,
excellent
optoelectronic
properties,
superior
near-infrared
by
multi-site
chelate
effect
bio-friendly
quercetin
(QR)
molecules.
Coupled
mechanism
non-equilibrium
carrier
strategy,
(NIR)
synapse
FASnI3-QR
perovskite
films
exhibited
key
synaptic
characteristics
practical
applications,
including
quasi-linear
time-dependent
photocurrent
generation,
prolonged
decay,
low
energy
consumption.
Ultimately,
12×12
real-time
NIR
array
was
successfully
constructed
thin-film
transistor
(TFT)
backplanes
through
heterogeneous
integration
devices
Si
circuits,
which
enables
spatiotemporal
information
complex
environments
at
hardware
level.
Abstract
Homeostasis
is
essential
in
biological
neural
networks,
optimizing
information
processing
and
experience‐dependent
learning
by
maintaining
the
balance
of
neuronal
activity.
However,
conventional
two‐terminal
memristors
have
limitations
implementing
homeostatic
functions
due
to
absence
global
regulation
ability.
Here,
three‐terminal
oxide
memtransistor‐based
synapses
are
demonstrated
perform
highly
linear
synaptic
weight
update
enhanced
accuracy
neuromorphic
computing.
Particularly,
leveraging
gate
control
contact‐engineered
indium‐gallium‐zinc‐oxide
(IGZO)
memtransistor,
scaling
enabled
for
high‐linearity
precision
Moreover,
sinusoidal
voltage
demonstrated,
possibly
enabling
emulation
higher‐order
functions.
The
device
structure
IGZO
memtransistor
optimized
regarding
source/drain
electrode
materials
an
interfacial
layer
inserted
between
channel
source
electrode.
As
a
result,
memtransistors
exhibiting
high
current
switching
ratio
>10
4
reliable
endurance
characteristics
obtained.
Furthermore,
through
adaptation
scaling,
emulating
homeostasis,
non‐linearity
values
0.01
−0.01
achieved
potentiation
depression,
respectively,
recognition
91.77%
digit
images.
It
envisioned
that
hold
significant
promise
homeostasis
computing
linearity
efficiency.
physica status solidi (RRL) - Rapid Research Letters,
Год журнала:
2025,
Номер
unknown
Опубликована: Фев. 3, 2025
Reservoir
computing
with
optical
devices
offers
an
energy‐efficient
approach
for
time‐series
forecasting.
Quantum
dot
lasers
feedback
are
modeled
in
this
article
to
explore
the
extent
which
increased
complexity
charge‐carrier
dynamics
within
nanostructured
semiconductor
can
enhance
prediction
performance.
By
tuning
scattering
interactions,
laser's
and
response
time
be
finely
adjusted,
allowing
a
systematic
investigation.
It
is
found
that
both
system
task
requirements
need
considered
find
optimal
operation
conditions.
Further,
pronounced
relaxation
oscillations
outperform
those
strongly
damped
dynamics,
even
if
underlying
more
complex.
This
demonstrates
reservoir
performance
relies
not
only
on
high
internal
phase
space
dimension
but
also
effective
utilization
of
these
through
output
sampling
process,
quantum
laser,
computing,
delay,
rate,
oscillation.
Micromachines,
Год журнала:
2025,
Номер
16(3), С. 269 - 269
Опубликована: Фев. 26, 2025
This
paper
presents
a
novel
passive
floating
memristor
emulator
that
operates
without
an
external
DC
bias,
leveraging
the
DTMOS
technique.
The
design
comprises
only
four
MOSFETs
and
eliminates
need
for
capacitors.
achieves
high
operating
frequency
of
around
250
MHz
consumes
zero
static
power.
A
comprehensive
analysis
simulation,
conducted
using
180
nm
CMOS
technology,
validates
circuit’s
performance.
versatility
effectiveness
proposed
are
demonstrated
through
its
application
in
various
circuits,
including
logic
gates,
ring
oscillator,
analog
filters,
highlighting
potential
diverse
low-power,
high-frequency
applications.
provides
compact,
efficient,
integrable
solution
nanoelectronic
circuit
designs.
