CN109449286A - A kind of nitride memristor and preparation method thereof that phase transformation nano particle is inlayed - Google Patents
A kind of nitride memristor and preparation method thereof that phase transformation nano particle is inlayed Download PDFInfo
- Publication number
- CN109449286A CN109449286A CN201811175304.6A CN201811175304A CN109449286A CN 109449286 A CN109449286 A CN 109449286A CN 201811175304 A CN201811175304 A CN 201811175304A CN 109449286 A CN109449286 A CN 109449286A
- Authority
- CN
- China
- Prior art keywords
- nitride
- oxide film
- phase transformation
- hearth electrode
- memristor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N—ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N70/00—Solid-state devices having no potential barriers, and specially adapted for rectifying, amplifying, oscillating or switching
- H10N70/20—Multistable switching devices, e.g. memristors
- H10N70/231—Multistable switching devices, e.g. memristors based on solid-state phase change, e.g. between amorphous and crystalline phases, Ovshinsky effect
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B82—NANOTECHNOLOGY
- B82Y—SPECIFIC USES OR APPLICATIONS OF NANOSTRUCTURES; MEASUREMENT OR ANALYSIS OF NANOSTRUCTURES; MANUFACTURE OR TREATMENT OF NANOSTRUCTURES
- B82Y10/00—Nanotechnology for information processing, storage or transmission, e.g. quantum computing or single electron logic
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B82—NANOTECHNOLOGY
- B82Y—SPECIFIC USES OR APPLICATIONS OF NANOSTRUCTURES; MEASUREMENT OR ANALYSIS OF NANOSTRUCTURES; MANUFACTURE OR TREATMENT OF NANOSTRUCTURES
- B82Y30/00—Nanotechnology for materials or surface science, e.g. nanocomposites
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N—ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N70/00—Solid-state devices having no potential barriers, and specially adapted for rectifying, amplifying, oscillating or switching
- H10N70/011—Manufacture or treatment of multistable switching devices
-
- H—ELECTRICITY
- H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N—ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
- H10N70/00—Solid-state devices having no potential barriers, and specially adapted for rectifying, amplifying, oscillating or switching
- H10N70/801—Constructional details of multistable switching devices
- H10N70/881—Switching materials
- H10N70/883—Oxides or nitrides
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Nanotechnology (AREA)
- Physics & Mathematics (AREA)
- Crystallography & Structural Chemistry (AREA)
- Materials Engineering (AREA)
- General Physics & Mathematics (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- Composite Materials (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Semiconductor Memories (AREA)
- Physical Vapour Deposition (AREA)
Abstract
The invention discloses a kind of nitride memristors and preparation method thereof that phase transformation nano particle is inlayed, memristor includes the hearth electrode, dielectric layer and top electrode set gradually, and dielectric layer is the nitride or nitride oxide film doped with transition metal being arranged in hearth electrode one side.Preparation method includes the following steps: in deposition on substrate inert metal as hearth electrode；One layer of nitride or nitride oxide film are deposited on hearth electrode, while depositing a small amount of transition metal into nitride or nitride oxide film；Mask plate is added on dielectric layer, and inert metal is deposited on mask plate as removing mask plate after top electrode；Back bias voltage is used between top electrode and hearth electrode, so that forming the phase transformation nano particle for running through the dielectric layer and gathering into bundles in dielectric layer.This memristor has encoding function under periodically strong input pulse effect, has memory and learning functionality after strong input pulse, can analog neuron cynapse well calculating and learning functionality.
Description
Technical field
The present invention relates to a kind of nitride memristors and preparation method thereof that phase transformation nano particle is inlayed, and belong to information electronics
Field of material technology.
Background technique
Using memristor analog neuron synaptic plasticity, realize that the calculating of class brain is information, material, computer, Neuscience
The problem of scientists and engineers in field pay close attention to jointly.Many memristors are invented, such as various metal oxides, Neng Goumo
Quasi- synaptic plasticity and learning functionality.However, these devices differ distant with the property of true biological synapse, such as Jie that works
Matter is unclear, mechanism is unintelligible, and operating mode still differs greatly with real synapse.For example, being reported on Nano Letters
And quote extensively inversion of phases memristor (Kuzum, D., Jeyasingh, R.G.D., Lee, B.&Wong,
H.S.P.Nanoelectronic programmable synapses based on phase change materials
For brain-inspired computing.Nano Lett.12,2179-2186 (2012)), the phase-change material energy used
Resistance value gradually changes under outer impulse stimulation, the synaptic plasticity variation observed in similar Neuscience.However this kind of material
And structure, Synaptic plasticity is simulated in terms of electroresponse, but does not reflect the flowing of biological synapse inner ion, nerve
The dynamic process of tramsmitter release and gate ion channel.It is in this way likeness in form rather than alike in spirit, the computing function of real synapse
It can not reflect.
Therefore there is an urgent need to find behavior, efficiency, dynamic process closer to biological synapse property material and device, from
And class brain calculating of really marching toward.
Summary of the invention
The object of the present invention is to provide a kind of nitride memristors and preparation method thereof that phase transformation nano particle is inlayed, this is recalled
Hindering device has coding and learning functionality, it can convert periodical input stimulus signal to another cyclical signal output,
Without any extraneous modulation circuit, after stimulation, change in long term occurs for the whole conductivity (resistivity) of device, i.e., when long
Journey memory and study efficacy.
To achieve the goals above, the present invention adopts the following technical scheme:
A kind of nitride memristor that phase transformation nano particle is inlayed proposed by the present invention, including be located in substrate and set gradually
Hearth electrode, dielectric layer and top electrode, it is characterised in that: the dielectric layer be arranged in the hearth electrode one side doped with
The nitride or nitride oxide film of transition metal；By applying the continuous of 2~6V of amplitude between the hearth electrode, top electrode
Back bias voltage is by pulse shock or continuous scanning mode, so that the transition metal of doping, which is changed into, is embedded in nitride or nitrogen oxidation
Phase transformation nano particle in object film；
The nitride includes gallium nitride, aluminium nitride, silicon nitride, boron nitride, indium nitride；The nitrogen oxides includes nitrogen oxygen
Change gallium, aluminum oxynitride, silicon oxynitride, nitrogen oxidation boron, nitrogen oxidation indium.
Further, the phase transformation nanoparticle aggregate bunchy, and run through the nitride or nitride oxide film, phase transformation
The width of nano particle beam is between 20~100nm.
The present invention also proposes a kind of preparation method of above-mentioned nitride memristor, comprising the following steps:
1) lazy in one layer of any deposition on substrate using electron beam evaporation plating, thermal evaporation, magnetron sputtering or ion sputtering method
Property metal, forms the hearth electrode；
2) magnetron sputtering, ion sputtering, chemical vapor deposition or atomic layer are used on the top surface for be formed by hearth electrode
The method of deposition deposits the nitride or nitride oxide film, using deposition rate a；Depositing the nitride or nitrogen oxidation
It is following any using the method deposition of magnetron sputtering, ion sputtering, chemical vapor deposition or atomic layer deposition while object film
A kind of transition metal element, including vanadium, chromium, tantalum, molybdenum, yttrium, hafnium, tungsten and niobium, using deposition rate b；By the phase for adjusting a and b
To size, the atomic percent of metal ion elements in the transition metal element and nitride or nitride oxide film is controlled
ArateIn 1:100, to making between 1:10, the transition metal element even dispersion is distributed in the nitride or nitrogen oxides is thin
In film；
3) mask plate is added on being formed by nitride or nitride oxide film top surface, using electricity on the mask plate
Beamlet vapor deposition, thermal evaporation, magnetron sputtering or ion sputtering method deposit one layer of inert metal, remove institute after forming the top electrode
State mask plate；
4) the negative pulse bias for applying 2~6V of amplitude between the top electrode and the hearth electrode, continuous 20~200
After a pulse shock, the phase transformation nano particle for running through and gathering into bundles is formed in the nitride or nitride oxide film.
The features of the present invention mainly has:
Memristor of the invention has Signal coding and learning functionality, including but not limited to:
1) it is higher than the recurrent pulses input signal of 2V to voltage magnitude, re-modulates output cyclical signal, output letter
Number waveform, amplitude and frequency change, and with input frequency variation and change；The peak value of output signal is with input signal
Number be in periodic swinging, waveform, amplitude and frequency all change with frequency input signal；
2) it is lower than the recurrent pulses input signal of 1V to voltage magnitude, output signal is with input signal without obvious poor
Not；
3) the recurrent pulses input signal to voltage magnitude between 1V~2V, the variation of output signal is between above-mentioned two
Between person, the variational regularity of output signal is not strong；
4) pulse signal that a voltage magnitude is lower than 1V, the original state of read-out system, with conductivity or electricity are first inputted
Resistance rate indicates, then inputs the recurrent pulses input signal that one group of voltage magnitude is higher than 2V, then inputs a voltage amplitude again
Value is lower than the pulse signal of 1V, and the end-state of read-out system is indicated with conductivity or resistivity；Calculate former and later two low-voltages
The ratio of impulse response show that the variation of the weight of this memristor or long changing course, weight ratio as a percentage, and pass through
State and result after overfitting；Weighted value is greater than 100, and referred to as enhanced plasticity shows to remember and learns to be formed；Weighted value
Less than 100, referred to as suppressive plasticity shows to forget.
The invention has the following advantages that
1) due to being used as matrix using nitride or nitride oxide film, and phase transformation nano particle is oxide, therefore phase
The size for becoming nano particle is restricted, and between 2 nanometers to 10 nanometers, and is not easy to grow up, and can inhibit nano particle company
Lead to form an entirety；Each nano particle can also cooperate with other nano particles with operating alone.
2) under the action of outer making alive, phase transformation nanoparticle aggregate bunchy runs through nitride film, is formed conductive logical
Road.Phase transformation nano particle is not Dispersed precipitate in nitride or nitride oxide film, is gathered into bundles convenient for electric current edge
Channel is flowed through, rather than from film, some position passes through at random, and analog neuron mediator and electric signal are fixed from some site of pre-synapse
To the next site for being transmitted to post-synapse.
3) under the effect of strong input signal, the peak value of output signal is periodic swinging, is suitble to grind using Fourier transformation
Study carefully the signal processing of memristor.All multi informations such as the frequency, amplitude, ingredient of signal, letter can be obtained from Fourier's variation
It number can be analyzed with classical principles of signal processing, obtain the calculating rule or rule of memristor.
4) after the effect of strong input signal, permanent change can occur for the state (such as conductance, resistance) of memristor, can be with weak
Input signal (not changing memristor state) reads the state before and after strong input signal, calculates weight variation, i.e., effective to learn
Practise intensity.By permanent change, memory is formd, so as to analyze its learning process.This memristor can fine simulation
The calculating and learning functionality of nerve synapse.
5) either to strong input or weak input signal, the response of memristor is all to rely on to input frequency, be frequency according to
Rely the mode of learning of plasticity.The mode that frequency relies on can be with the mistake of the theory analysis learning and memory of signal processing
Journey.
Detailed description of the invention
Fig. 1 is the integrally-built cross-sectional view figure of memristor of the embodiment of the present invention.
Fig. 2 is the top view of memristor of the embodiment of the present invention.
Fig. 3 is nitride film layer of the invention and phase transformation nano particle distribution schematic diagram.
Fig. 4 is exemplary voltages current cycle scanning curve figure of the memristor of embodiment of the present invention under the conditions of positive bias.
Fig. 5 is exemplary voltages current cycle scanning curve figure of the memristor of embodiment of the present invention under the conditions of back bias voltage.
Fig. 6 is memristor of the embodiment of the present invention after continuous 100 voltage and current scan round, is taken in Fig. 4 or Fig. 5 respectively
A point and the current-responsive curve drawn of B point.
Fig. 7 is the response that the embodiment of the present invention typically inputs by force (high voltage amplitude) pulse, wherein shows frequency in a
Impulse response under 1Hz, 5Hz, 10Hz, 20Hz shows the impulse response under frequency 40Hz, 60Hz, 80Hz in b.
Fig. 8 is the long time-histories data for embodying memristor of embodiment of the present invention learning functionality characteristic.
Specific embodiment
Experimental method used in following embodiments is conventional method unless otherwise specified.
The materials, reagents and the like used in the following examples is commercially available unless otherwise specified.
Invention is further explained in the following with reference to the drawings and specific embodiments, but the present invention is not limited thereto,
Any modifications, equivalent replacements and improvementsmade within the spirit and principles of the invention, etc. should be included in of the invention
Within protection scope.
A kind of nitride memristor that phase transformation nano particle is inlayed proposed by the present invention, including be located in substrate and set gradually
Hearth electrode, dielectric layer and top electrode；The dielectric layer be arranged in the hearth electrode one side doped with transition metal
Nitride or nitride oxide film；Continuous back bias voltage by applying 2~6V of amplitude between the hearth electrode, top electrode passes through
Pulse shock or continuous scanning mode are embedded in nitride or nitride oxide film so that the transition metal of doping is changed into
Phase transformation nano particle.
The nitride includes gallium nitride, aluminium nitride, silicon nitride, boron nitride and indium nitride；The nitrogen oxides includes nitrogen
Gallium oxide, aluminum oxynitride, silicon oxynitride, nitrogen oxidation boron and nitrogen oxidation indium.Further, nitride or nitride oxide film thickness
It can be 50~150nm.
The hearth electrode and top electrode are respectively adopted platinum, gold or palladium and are made；Wherein, top electrode is as signal input part,
Hearth electrode is as signal output end.
The material of the phase transformation nano particle includes the oxidation of vanadium, chromium, tantalum, molybdenum, yttrium, hafnium, tungsten, niobium these types transition metal
Object.
Further, the average-size of phase transformation nano particle, can be in amorphous phase and any one between 2nm~10nm
Change between crystalline phases.
Further, phase transformation nanoparticle aggregate bunchy, and through nitride or nitride oxide film, phase transformation nanometer
The width of grain beam is between 20~100nm.
Further, the hearth electrode and top electrode is respective with a thickness of 50~300nm.
The present invention also proposes a kind of preparation method of above-mentioned memristor, comprising the following steps:
1) using electron beam evaporation plating, thermal evaporation, magnetron sputtering or ion sputtering method in any substrate (preferably insulation base
Piece) on deposit one layer of inert metal (including Au, Pt or Pd), form the hearth electrode；
2) magnetron sputtering, ion sputtering, chemical vapor deposition or atomic layer are used on the top surface for be formed by hearth electrode
The method of deposition deposits the nitride or nitride oxide film, and the deposition rate used is a；Depositing the nitride or nitrogen
It is as follows using the method deposition of magnetron sputtering, ion sputtering, chemical vapor deposition or atomic layer deposition while sull
Any one transition metal element, including vanadium, chromium, tantalum, molybdenum, yttrium, hafnium, tungsten and niobium, the deposition rate used is b；By adjusting a
With the relative size of b, the transition metal element (including vanadium, chromium, tantalum, molybdenum, yttrium, hafnium, tungsten, niobium) and nitride or nitrogen oxygen are controlled
The atomic percent A of metal ion elements (including gallium, aluminium, silicon, boron, indium) in compound filmrateIt is 1:100 between 1:10,
It is distributed in the transition metal element even dispersion in the nitride or nitride oxide film；
3) mask plate is added on being formed by nitride or nitride oxide film top surface, using electricity on the mask plate
Beamlet vapor deposition, thermal evaporation, magnetron sputtering or ion sputtering method deposit one layer of inert metal (including Au, Pt or Pd), form institute
Top electrode is stated, the mask plate is then removed；
4) the negative pulse bias for applying 2~6V of amplitude between the top electrode and the hearth electrode, continuous 20~200
After a pulse shock, the phase transformation nano particle for running through and gathering into bundles is formed in the nitride or nitride oxide film.
Further, in step 4), the pulse width of the negative pulse bias of application is 20 μ s~100ms, controls pulse frequency
Rate increases in 0.1Hz~1000Hz, the width of phase transformation nano particle beam with the increase of pulse frequency, and quantity is with pulse frequency
Increase and increases.
Further, the mask plate is the metal equipped with circular hole figure, line pattern, camber line figure or broken line figure
Plate, the metal plate are made of stainless steel material or chromium material, when the mask plate is the metal plate equipped with circular hole figure,
The diameter of the circular hole is 100nm~0.5mm.
The following are the embodiment of the present invention: Pd/Nb adulterates AlNO/Pd memristor and preparation method thereof
For the overall structure of memristor of the embodiment of the present invention referring to Fig. 1, Fig. 2, which is included in surface band SiO2Si
Pd hearth electrode, Nb doping AlNO film and the Pd hearth electrode set gradually on substrate；By between Pd hearth electrode, Pd top electrode
Apply the continuous negative pulse bias that amplitude is greater than 2V (specific voltage magnitude used by the present embodiment is 4V), so that the Nb of doping
It is changed into the phase transformation nano particle being embedded in AlNO film.The Pd hearth electrode, Nb doping AlNO film and the bottom Pd of the present embodiment
The thickness of electrode is respectively 150nm, 100nm, 150nm.
The preparation method of the present embodiment memristor the following steps are included:
1) by commercially available surface band SiO2Si substrate acetone, alcohol and deionized water be successively cleaned by ultrasonic 4-8 minutes,
With being dried with nitrogen.Si substrate is put into vacuum chamber chip bench, with electron beam evaporation plating, thermal evaporation, magnetron sputtering or ion sputtering etc.
Method forms Pd hearth electrode in the Pd of one layer of 150nm of deposition on substrate.
2) using magnetron sputtering, ion sputtering, chemical vapor deposition or atomic layer deposition in the one side of Pd hearth electrode
Method deposits 100nm aluminum oxynitride film, using deposition rate a；While depositing silicon oxynitride aluminium film, splashed using magnetic control
Penetrate, ion sputtering, chemical vapor deposition or atomic layer deposition method deposition Nb transition metal element, using deposition rate b.It is logical
The relative size for overregulating a and b, control the atomic percent of aforementioned Al/Nb 1:100 to 1:10 (the present embodiment 1:80) it
Between.The AlNO film of Nb doping is formed by this step.
3) metal of the addition equipped with circular hole figure (diameter 0.25mm) in the AlNO film one side that the Nb of formation is adulterated
Then mask plate deposits the top electrode 150nm using the methods of electron beam evaporation plating, thermal evaporation, magnetron sputtering or ion sputtering
Pd, remove the mask plate, both the memristor initial state.As depicted in figs. 1 and 2.
4) add the negative arteries and veins of amplitude 2V~6V (specific amplitude 4V used by the present embodiment) between top electrode and hearth electrode
Rushing bias, (pulse width is 20 μ s~100ms, and pulse frequency is in 0.1Hz~1000Hz, for example, if pulse width is 10ms, frequency
Rate 100Hz), after continuous 20-200 (such as 100) pulse shocks, form phase transformation nano particle niobium oxide, and niobium oxide
Nanoparticle aggregate bunchy and Nb doping AlNO film, as shown in figure 3, different size of ellipse is nano particle in figure,
When loading continuous back bias voltage, many phase transformation nano particles are gradually appeared, are extended from hearth electrode to top electrode, are formed by phase transformation
The bundles of conductive filament of nanoparticle aggregate.
Or above-mentioned steps 4) can be substituted by following step:
The back bias voltage of amplitude 2V~6V (specific amplitude 4V used by the present embodiment) is used between top electrode and hearth electrode
Continuous scanning after about 4 scanning, forms phase transformation nano particle niobium oxide, and niobium oxide nanoparticle aggregate bunchy and Nb doping
AlNO film, as shown in Figure 3.
The validation verification of the embodiment of the present invention:
In order to verify the electrical properties of the embodiment of the present invention, different amplitudes, frequency and width are applied to hearth electrode and top electrode
The pulse signal of degree observes this memristor to the corresponding of the pulse signal of input, the including but not limited to following:
Property 1, no matter positive bias or back bias voltage, negative differential resistance can all occur on DC voltage and current curve.Such as figure
4, be typical voltage and current scan round curve graph shown in 5: no matter positive bias or back bias voltage, the peak value of electric current all goes out
The voltage of present A point, the point is less than maximum scan voltage；The current value when current value of B point is maximum scan voltage.
The voltage-current curve obtained after property 2, continuous scanning can extract the maximum current value scanned every time out, respectively most
Oscillation of the high current value with the variation of scanning times and in rule.As shown in fig. 6, being the typical change of sweep current peak value: even
After continuous 100 voltage and current scan rounds, the current-responsive curve of A point in Fig. 4 or 5, B point-rendering is taken respectively；Wherein, A point
Rising in pure oscillation, B point is in fall before and then be gradually increasing trend, in Fig. 6, variable IVWhen being each scanning voltage maximum
Accordingly current value, IPIt is maximum current value in each scanning.
Property 3, this memristor have the function of Signal coding.In the response of typical strong input pulse (- 4V square wave), also referred to as
For short term plasticity, weighing computation method are as follows:Wherein IAIt is each pulse
Peak value, N indicate n-th pulse, and -4V is the amplitude of input pulse.It can see that weight (as described in ordinate in figure) is frequency f
With the function of pulse number N (as shown in abscissa in figure), output response is very regular, is cyclically-varying curve.In Fig. 7
A, b shown in, illustrate that this memristor has the function of good Signal coding.
Property 4, this memristor have typical learning functionality.Small voltage pulse (- 1.5V square wave is used before the strong input of Fig. 7
Pulse) this memristor state is read, one group of flash stimulation memristor as used in Fig. 7 is then used, then again with small
Voltage pulse reads this memristor state.Weighing computation method are as follows: WN(I (- 1.5V))=IN(-1.5V)/I0(-1.5V).Wherein
N refers to n-th flash, and -1.5V is the amplitude for reading pulse.By weight (square as shown in Figure 8) tool of peak computational
There is typical frequency selectivity, i.e. low frequency (frequency 10Hz) inhibits (weight is less than 100), high frequency (frequency is greater than 50Hz) enhancing
The characteristic of (weight is greater than 100), meets the typical frequency dependence plasticity (spike-rate-dependent of Neuscience
Plasticity) mode of learning.
In addition, being lower than the recurrent pulses input signal of 1V for voltage magnitude, this memristor output signal and input are believed
Number do not have significant difference.For recurrent pulses input signal of the voltage magnitude between 1V~2V, this memristor output signal
Variation between said two devices, the variational regularity of output signal is not strong.
To sum up, memristor proposed by the present invention has characteristics that
1) it is higher than the recurrent pulses input signal of 2V to voltage magnitude, output cyclical signal is re-modulated, such as property 3
It is described, that is, have the function of Signal coding.Waveform, amplitude and the frequency of output signal change, and with the variation of input frequency
And change.The peak value of output signal is in periodic swinging with the number of input signal, and waveform, amplitude and frequency are all with input
Signal frequency variation.
2) it is lower than the recurrent pulses input signal of 1V to voltage magnitude, output signal is with input signal without obvious poor
Not.
3) the recurrent pulses input signal to voltage magnitude between 1V~2V, the variation of output signal is between above-mentioned two
Between person, the variational regularity of output signal is not strong.
4) pulse signal that a voltage magnitude is lower than 1V, the original state of read-out system, with conductivity or electricity are first inputted
Resistance rate indicates, then inputs the recurrent pulses input signal that one group of voltage magnitude is higher than 2V, then inputs a voltage amplitude again
Value is lower than the pulse signal of 1V, and the end-state of read-out system is indicated with conductivity or resistivity.Calculate former and later two low-voltages
The ratio of impulse response show that the variation of the weight of system or long changing course, weight ratio as a percentage, and are passed through and learned
State and result after habit.Weighted value is greater than 100, referred to as enhanced plasticity；For weighted value less than 100, referred to as suppressive is plastic
Property.As described in property 4.
Claims (10)
1. a kind of nitride memristor that phase transformation nano particle is inlayed, including being located at the hearth electrode, the medium that set gradually in substrate
Layer and top electrode, it is characterised in that: the dielectric layer is the nitrogen doped with transition metal being arranged in the hearth electrode one side
Compound or nitride oxide film；Continuous back bias voltage by applying 2~6V of amplitude between the hearth electrode, top electrode passes through arteries and veins
It hits in a state of excitement or continuous scanning mode, so that the transition metal of doping is changed into the phase being embedded in nitride or nitride oxide film
Become nano particle；
The nitride includes gallium nitride, aluminium nitride, silicon nitride, boron nitride, indium nitride；The nitrogen oxides includes nitrogen oxidation
Gallium, aluminum oxynitride, silicon oxynitride, nitrogen oxidation boron, nitrogen oxidation indium.
2. nitride memristor according to claim 1, which is characterized in that the phase transformation nano-particle material include vanadium,
Chromium, tantalum, molybdenum, yttrium, hafnium, tungsten, niobium these types transition metal oxide.
3. nitride memristor according to claim 1, which is characterized in that the thickness of the nitride or nitride oxide film
Degree is 50~150nm.
4. nitride memristor according to claim 1, which is characterized in that the average-size of shown phase transformation nano particle exists
Between 2~10nm, it can change between amorphous phase and any one crystalline phases.
5. nitride memristor according to claim 1, which is characterized in that the hearth electrode and the top electrode are all made of
Inert metal is made, including platinum, gold or palladium.
6. nitride memristor according to claim 1, which is characterized in that the phase transformation nanoparticle aggregate bunchy, and
Through the nitride or nitride oxide film, the width of phase transformation nano particle beam is between 20~100nm.
7. a kind of preparation method of nitride memristor according to claim 1 to 6, which is characterized in that including
Following steps:
1) using electron beam evaporation plating, thermal evaporation, magnetron sputtering or ion sputtering method in one layer of inertia gold of any deposition on substrate
Belong to, forms the hearth electrode；
2) magnetron sputtering, ion sputtering, chemical vapor deposition or atomic layer deposition are used on the top surface for be formed by hearth electrode
Method deposit the nitride or nitride oxide film, using deposition rate a；It is thin in the deposition nitride or nitrogen oxides
While film, using the method deposition of magnetron sputtering, ion sputtering, chemical vapor deposition or atomic layer deposition it is following any one
Transition metal element, including vanadium, chromium, tantalum, molybdenum, yttrium, hafnium, tungsten and niobium, using deposition rate b；By adjusting the relatively large of a and b
It is small, control the atomic percent A of metal ion elements in the transition metal element and nitride or nitride oxide filmrate?
1:100 is distributed in the nitride or nitride oxide film between 1:10, making the transition metal element even dispersion；
3) mask plate is added on being formed by nitride or nitride oxide film top surface, electron beam is used on the mask plate
Vapor deposition, thermal evaporation, magnetron sputtering or ion sputtering method deposit one layer of inert metal, are formed after the top electrode and are covered described in removing
Diaphragm plate；
4) the negative pulse bias for applying 2~6V of amplitude between the top electrode and the hearth electrode, in continuous 20~200 arteries and veins
After hitting in a state of excitement, the phase transformation nano particle for running through and gathering into bundles is formed in the nitride or nitride oxide film.
8. preparation method according to claim 7, which is characterized in that the step 4 is substituted using following steps:
After the back bias voltage continuous scanning for applying 2~6V of amplitude between the top electrode and the hearth electrode, in the nitride
Or the phase transformation nano particle for running through and gathering into bundles is formed in nitride oxide film.
9. preparation method according to claim 7, which is characterized in that in step 3), the mask plate is equipped with circular hole figure
Shape, line pattern, camber line figure or broken line figure metal plate, the metal plate is made of stainless steel material or chromium material；
When the mask plate is the metal plate equipped with circular hole figure, the diameter of the circular hole is 100nm~0.5mm.
10. preparation method according to claim 7, which is characterized in that in step 4), the pulse of the negative pulse bias of application
Width is 20 μ s~100ms, controls pulse frequency in 0.1Hz~1000Hz.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811175304.6A CN109449286B (en) | 2018-10-10 | 2018-10-10 | Phase-change nanoparticle-embedded nitride memristor and preparation method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811175304.6A CN109449286B (en) | 2018-10-10 | 2018-10-10 | Phase-change nanoparticle-embedded nitride memristor and preparation method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109449286A true CN109449286A (en) | 2019-03-08 |
| CN109449286B CN109449286B (en) | 2020-04-24 |
Family
ID=65546258
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811175304.6A Active CN109449286B (en) | 2018-10-10 | 2018-10-10 | Phase-change nanoparticle-embedded nitride memristor and preparation method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109449286B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111537111A (en) * | 2020-04-29 | 2020-08-14 | 西安交通大学 | Dual-signal output nerve morphology touch sensor and preparation method thereof |
| CN111667059A (en) * | 2020-05-13 | 2020-09-15 | 清华大学 | Artificial neural component based on phase-change nanoparticle inlaid nitride |
| CN113675223A (en) * | 2021-05-17 | 2021-11-19 | 松山湖材料实验室 | A kind of photoelectric synapse device and its application |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7560721B1 (en) * | 2008-02-21 | 2009-07-14 | International Business Machines Corporation | Phase change material with filament electrode |
| CN101681911A (en) * | 2006-11-08 | 2010-03-24 | 思美公司 | Correlated electron memory |
| CN102265397A (en) * | 2008-12-23 | 2011-11-30 | 惠普开发有限公司 | Memristive device and method of making and using same |
| CN102484129A (en) * | 2009-07-10 | 2012-05-30 | 惠普发展公司，有限责任合伙企业 | Memristive junction with intrinsic rectifier |
| CN102484128A (en) * | 2009-09-04 | 2012-05-30 | 惠普发展公司，有限责任合伙企业 | Switchable junction with intrinsic diode formed with voltage dependent resistor |
| CN103797573A (en) * | 2011-08-03 | 2014-05-14 | 惠普发展公司，有限责任合伙企业 | Nitride-based memristors |
| US20160028005A1 (en) * | 2014-01-15 | 2016-01-28 | Hewlett-Packard Development Company, L.P. | Memristor structure with a dopant source |
| CN206163528U (en) * | 2016-11-29 | 2017-05-10 | 东华理工大学 | Vanadium dioxide film is recalled and is hindered memory |
| US20170250340A1 (en) * | 2016-02-29 | 2017-08-31 | Arm Ltd. | Barrier layer for correlated electron material |
-
2018
- 2018-10-10 CN CN201811175304.6A patent/CN109449286B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101681911A (en) * | 2006-11-08 | 2010-03-24 | 思美公司 | Correlated electron memory |
| US7560721B1 (en) * | 2008-02-21 | 2009-07-14 | International Business Machines Corporation | Phase change material with filament electrode |
| CN102265397A (en) * | 2008-12-23 | 2011-11-30 | 惠普开发有限公司 | Memristive device and method of making and using same |
| CN102484129A (en) * | 2009-07-10 | 2012-05-30 | 惠普发展公司，有限责任合伙企业 | Memristive junction with intrinsic rectifier |
| CN102484128A (en) * | 2009-09-04 | 2012-05-30 | 惠普发展公司，有限责任合伙企业 | Switchable junction with intrinsic diode formed with voltage dependent resistor |
| CN103797573A (en) * | 2011-08-03 | 2014-05-14 | 惠普发展公司，有限责任合伙企业 | Nitride-based memristors |
| US20160028005A1 (en) * | 2014-01-15 | 2016-01-28 | Hewlett-Packard Development Company, L.P. | Memristor structure with a dopant source |
| US20170250340A1 (en) * | 2016-02-29 | 2017-08-31 | Arm Ltd. | Barrier layer for correlated electron material |
| CN206163528U (en) * | 2016-11-29 | 2017-05-10 | 东华理工大学 | Vanadium dioxide film is recalled and is hindered memory |
Non-Patent Citations (2)
| Title |
|---|
| ZHONGRUIWANG，ET AL.: "Memristors with di usive dynamics as synaptic emulators for neuromorphic computing", 《NATURE MATERIALS》 * |
| ZIYU LV,ET AL.: "From biomaterial-based data storage to bio-inspired artificial synapse", 《MATERIALS TODAY》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111537111A (en) * | 2020-04-29 | 2020-08-14 | 西安交通大学 | Dual-signal output nerve morphology touch sensor and preparation method thereof |
| CN111667059A (en) * | 2020-05-13 | 2020-09-15 | 清华大学 | Artificial neural component based on phase-change nanoparticle inlaid nitride |
| CN111667059B (en) * | 2020-05-13 | 2022-09-13 | 清华大学 | Artificial neural component based on phase-change nanoparticle inlaid nitride |
| CN113675223A (en) * | 2021-05-17 | 2021-11-19 | 松山湖材料实验室 | A kind of photoelectric synapse device and its application |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109449286B (en) | 2020-04-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yu et al. | Hf0. 5Zr0. 5O2-based ferroelectric memristor with multilevel storage potential and artificial synaptic plasticity | |
| Milano et al. | Self-limited single nanowire systems combining all-in-one memristive and neuromorphic functionalities | |
| Yan et al. | Graphene oxide quantum dots based memristors with progressive conduction tuning for artificial synaptic learning | |
| Choi et al. | SiGe epitaxial memory for neuromorphic computing with reproducible high performance based on engineered dislocations | |
| Waser et al. | Introduction to new memory paradigms: memristive phenomena and neuromorphic applications | |
| Park et al. | Nonideal resistive and synaptic characteristics in Ag/ZnO/TiN device for neuromorphic system | |
| Kuemmeth et al. | Measurement of discrete energy-level spectra in individual chemically synthesized gold nanoparticles | |
| Dongle et al. | Development of self-rectifying ZnO thin film resistive switching memory device using successive ionic layer adsorption and reaction method | |
| Yin et al. | Emulation of learning and memory behaviors by memristor based on Ag migration on 2D MoS2 surface | |
| CN109449286A (en) | A kind of nitride memristor and preparation method thereof that phase transformation nano particle is inlayed | |
| More et al. | Resistive switching and synaptic properties modifications in gallium-doped zinc oxide memristive devices | |
| Jeon et al. | Suppressed stochastic switching behavior and improved synaptic functions in an atomic switch embedded with a 2D NbSe2 material | |
| Yang et al. | Ultra-wide temperature electronic synapses based on self-rectifying ferroelectric memristors | |
| Hadiyal et al. | Bio-inspired artificial synapse for neuromorphic computing based on NiO nanoparticle thin film | |
| Mahata et al. | Synaptic plasticity and quantized conductance states in TiN-Nanoparticles-Based memristor for neuromorphic system | |
| Papakonstantinopoulos et al. | Highly flexible artificial synapses from SiO2-based conductive bridge memristors and Pt nanoparticles through a crack suppression technique | |
| Lee et al. | Crystallinity‐controlled volatility tuning of ZrO2 memristor for physical reservoir computing | |
| KR20250110181A (en) | Synapse device and fabrication method of the same | |
| Min et al. | The effects of Si doping on the endurance and stability improvement of AlN-based resistive random access memory | |
| Kim et al. | Progressive and stable synaptic plasticity with femtojoule energy consumption by the interface engineering of a metal/ferroelectric/semiconductor | |
| Kwon et al. | Comparison of synaptic properties considering dopant concentration and device operation polarity in Cu/SiN/SiO2/p-Si devices for neuromorphic system | |
| Byun et al. | Convolutional neural network for high-performance reservoir computing using dynamic memristors | |
| Choi et al. | Highly-packed self-assembled graphene oxide film-integrated resistive random-access memory on a silicon substrate for neuromorphic application | |
| Bai et al. | High‐Performance GdNiO3 Epitaxial Film Memristor for Neuralactivity Analysis | |
| Jeon et al. | Effect of ITO electrode on conductance quantization and multi-level cells in TiN/SiOx/ITO devices |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |