CN112270949A - Method for testing transient stability of resistive random access memory - Google Patents
Method for testing transient stability of resistive random access memory Download PDFInfo
- Publication number
- CN112270949A CN112270949A CN202011166822.9A CN202011166822A CN112270949A CN 112270949 A CN112270949 A CN 112270949A CN 202011166822 A CN202011166822 A CN 202011166822A CN 112270949 A CN112270949 A CN 112270949A
- Authority
- CN
- China
- Prior art keywords
- random access
- access memory
- resistance
- resistive random
- value
- 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
Images
Classifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
- G11C29/50—Marginal testing, e.g. race, voltage or current testing
Abstract
The invention relates to the technical field of semiconductor memory testing, in particular to a method for testing transient stability of a resistive random access memory. The test method comprises the following steps: activating the resistive random access memory; applying a plurality of working signals to enable the resistive random access memory to reach a stable state; setting a test temperature; setting a conductance G to be measured0And applying an excitation signal to enable the conductance value of the RRAM to reach G0(ii) a And testing the transient stability of the resistive random access memory. The testing method can rapidly and efficiently characterize the transient stability of the resistive random access memory, is beneficial to further restraining the short-time fluctuation of the resistive random access memory in the follow-up process, and improves the design of an optimization strategy of the accuracy of a neuromorphic system.
Description
Technical Field
The invention relates to the technical field of semiconductor memory testing, in particular to a method for testing transient stability of a resistive random access memory.
Background
The rapid development of information technology makes the data volume and complexity of the chip required to be processed exponentially rise, and higher requirements are put forward on the performances of storage, calculation and the like of electronic devices. The resistive random access memory has the advantages of low power consumption, high storage density, simple structure, compatibility with CMOS devices and the like, and is considered to be one of novel non-volatile memories with the most application prospect. In the practical application process, one of the biggest challenges faced by the resistive random access memory is the stability of the resistance state of the device, which affects the accuracy of data storage.
At present, the resistance state stability test of the resistance random access memory mainly comprises the fluctuation and the retention time of the resistance state. These tests mainly focus on the resistance state stability of the resistive random access memory over multiple cycles or over a longer period of time. However, with the gradual application of the resistive random access memory in the field of neuromorphic computing, the resistive state stability in a short time is particularly important. Particularly, in a short time (<1 second) after the resistance state of the resistance random access memory is written, the resistance state of the resistance random access memory fluctuates greatly due to the influence of the conductance relaxation effect and the random telegram noise, as shown in fig. 1. When the RRAM is used for constructing a neural network to participate in calculation, as shown in FIG. 2, the relaxation phenomenon can greatly affect the accuracy of neural network mapping, resulting in the reduction of identification precision, as shown in FIG. 3. Therefore, how to test the transient stability of the resistive random access memory has important significance for inhibiting the resistance state fluctuation of the device and the application of the resistive random access memory in the neural network.
Disclosure of Invention
The invention aims to provide a method for testing transient stability of a resistive random access memory, which is used for testing the transient stability of the resistive random access memory aiming at the phenomenon that the conductance of the resistive random access memory fluctuates greatly in a short time after being written in a resistive state, so that the transient stability of the resistive random access memory at different temperatures can be quickly and conveniently obtained, and the subsequent resistive state stability of a device can be further improved.
The invention provides a method for testing transient stability of a resistive random access memory, which comprises the following steps:
(1) using an activation signal V0To, forActivating the resistive random access memory to form conductive filaments in the resistive random access memory;
wherein the activation signal V0The signal is a direct current signal or a pulse signal, wherein the amplitude range of the pulse signal is 0 to 5 volts, the pulse width range is 100 nanoseconds to 100 microseconds, and the amplitude range of the direct current signal is 0 to 4 volts; the conductive filament is formed, namely the resistance random access memory reaches a low resistance state, so that the resistance value of the resistance random access memory is smaller than 100 kilo-ohm;
(2) applying multiple groups of working signals V to the resistive random access memory1The working signals are square wave pulse signals, each group of working signals comprises a positive voltage and a negative voltage, the amplitude range of the positive voltage is 0-4 volts, the duration range is 1 millisecond-1 second, the negative voltage range is 0-4 volts, the duration range is 1 millisecond-1 second, the resistance value of the resistive random access memory is read after the positive voltage is applied and is recorded as a low-resistance value, the resistance value of the resistive random access memory is read after the negative voltage is applied and is recorded as a high-resistance value, after a group of working signals are applied, the difference value of the resistance values of the high resistance state read for the current time and the next time is less than 5 percent of the resistance value of the high resistance state read for the previous time, and the difference value of the resistance values in the low resistance state read in the previous two times is less than 5% of the resistance value in the low resistance state read in the previous time, and (4) considering the resistive random access memory to be in a qualified state, repeating the step, and continuously applying more than 10 groups of working signals V.1When the resistance random access memory is in a qualified state, the resistance random access memory is considered to reach a stable state;
(3) setting a test temperature T1；
(4) Setting a conductance value G to be measured after the resistance random access memory reaches the stable state in the step (2)0，G0The method comprises the following steps of setting a conductance fluctuation threshold lambda between a high resistance state resistance value and a low resistance state resistance value of the resistive random access memory, and measuring the transient stability of the resistive random access memory, wherein the specific process is as follows:
(4-1) applying an excitation signal V to the resistive random access memory2The excitation signal is a square wave pulse signal, and the amplitude range of the square wave pulse signal0.3-0.7 volt, 10 ms-1 microsecond duration;
(4-2) after 10 milliseconds, applying a read signal V to the RRAM3The reading signal is a square wave pulse signal, the amplitude of the square wave pulse signal is 0.1 volt, and the duration of the square wave pulse signal is 10 milliseconds;
(4-3) reading the conductance value G of the RRAM1According to the value of conductance G0With a conductance fluctuation threshold lambda, to a conductance value G1Make a judgment if G0+λ≥G1≥G0λ, then step (5) is carried out, if G1<G0- λ, then step (4-4); if G is1>G0And + lambda, then carrying out the step (4-6);
(4-4) adding an excitation signal by a difference of 0.1 volt, namely applying an excitation signal to the resistive random access memory, wherein the amplitude of the excitation signal is 0.4-0.8 volt, the duration of the excitation signal is 10 milliseconds, and after 10 milliseconds, applying a square wave pulse with the amplitude of 0.1 volt and the duration of 10 milliseconds; reading conductance value G of resistive random access memory2；
(4-5) according to the conductance value G0With a fluctuation threshold lambda, to a conductance value G2Make a judgment if G0+λ≥G2≥G0λ, then step (5) is carried out, if G2<G0- λ, then returning to step (4-4); if G is2>G0And + lambda, then carrying out the step (4-6);
(4-6) applying a reverse excitation signal-V to the resistive random access memory2The excitation signal is a square wave pulse signal, the amplitude range of the square wave pulse signal is-0.3 to-0.7 volt, the duration is 10 milliseconds to 1 microsecond, and after 10 milliseconds, a reading signal V is applied to the resistive random access memory3Reading the conductance value G of the resistance random access memory, wherein the reading signal is a square wave pulse signal, the amplitude of the square wave pulse signal is 0.1 volt, the duration of the square wave pulse signal is 10 milliseconds, and the conductance value G of the resistance random access memory is read3；
(4-7) according to the conductance value G0With a fluctuation threshold lambda, to a conductance value G3Make a judgment if G0+λ≥G3≥G0λ, then step (5) is carried out, if G3<G0Lambda, returning to step (4-4) if G is3>G0And + lambda, returning to the step (4-6);
(4-8) increasing the reverse excitation signal-V by a difference of-0.1 volt2Namely, an excitation signal is applied to the RRAM, the amplitude range of the excitation signal is-0.4 to-0.8 volt, a square wave pulse signal with the duration of 10 milliseconds is applied, a square wave pulse with the amplitude of 0.1 volt and the duration of 10 milliseconds is applied after the interval of 10 milliseconds, and the conductance value G of the RRAM is read4；
(4-9) according to the conductance value G0With a fluctuation threshold lambda, to a conductance value G4Make a judgment if G0+λ≥G4≥G0λ, then step (5) is carried out, if G4<G0Lambda, returning to step (4-4) if G is4>G0And + lambda, returning to the step (4-8);
(4-10) repeating the steps (4-4) to (4-7) until the electric conductance value of the resistive random access memory is between G0- λ and G0And (5) performing the step (5);
(5) calculating the transient stability of the resistive random access memory, which comprises the following specific steps:
(5-1) applying a plurality of square wave pulse signals with the amplitude of 0.1 volt and the time interval of 10 milliseconds to the resistive random access memory in the step (4), wherein the total duration of the pulse signals is 1 second, and respectively recording all conductance values G of the resistive random access memory within the range of 0 to 1 secondiI represents the serial number of the test;
(5-2) calculating the actual conductance values G in the ranges of 0 to 1 second using the following formulasiAnd (3) the conductance value G to be measured in the step (2)0Standard deviation σ of absolute difference:
wherein G isiRepresentative of the actually measured conductance value of the resistance random access memory during measurement, generation iThe serial number of the table test, N is the number of times of the test of applying the square wave pulse signal in step 5, G0The resistance random access memory is used as a conductance value to be measured of the resistance random access memory;
(5-3) setting a threshold ζ of the standard deviation σ, determining the standard deviation σ in the step (5-2), and determining that the temperature is at the set temperature T when the standard deviation σ is less than or equal to the threshold ζ1The conductance value of the RRAM is G0Good transient stability if the standard deviation sigma is>And determining that the transient stability of the resistive random access memory is poor when the threshold value is zeta.
The method for testing the transient stability of the resistive random access memory has the characteristics and advantages that:
1. the method for testing the transient stability of the resistive random access memory can quickly and efficiently characterize the transient stability of the resistive random access memory, is beneficial to comparing the stability difference between the resistive random access memories, screens out the resistive random access memory with good transient stability, and fills the blank of the existing method for testing the transient stability of the resistive random access memory.
2. The testing method can rapidly and efficiently characterize the transient stability of the resistive random access memory, is beneficial to further restraining the short-time fluctuation of the resistive random access memory in the follow-up process, and improves the design of an optimization strategy of the accuracy of a neuromorphic system.
Drawings
Fig. 1 is a schematic diagram of a resistance state fluctuation phenomenon in a short time after resistance state writing in a conventional resistance random access memory.
Fig. 2 is a schematic diagram of a three-layer neural network for text or picture recognition.
FIG. 3 is a graph of the effect of conductance fluctuation on the recognition rate of a neural network performing a handwritten digit task.
Fig. 4 is a flow chart of a method for testing transient stability of a resistive random access memory according to the present invention.
FIG. 5 is a diagram of pulse signals applied in the programming process and the reading process of the transient stability test according to the present invention.
Fig. 6 is a conductance-time diagram of the resistive random access memory according to the invention after transient stability test.
Detailed Description
The method for testing the transient stability of the resistive random access memory, provided by the invention, has a flow chart as shown in fig. 4, and comprises the following steps:
(1) using an activation signal V0Activating the resistance random access memory to form a conductive filament in the resistance random access memory;
wherein the activation signal V0The signal is a direct current signal or a pulse signal, wherein the amplitude range of the pulse signal is 0 to 5 volts, the pulse width range is 100 nanoseconds to 100 microseconds, and the amplitude range of the direct current signal is 0 to 4 volts; the conductive filament is formed, namely the resistance random access memory reaches a low resistance state, so that the resistance value of the resistance random access memory is smaller than 100 kilo-ohm;
(2) applying multiple groups of working signals V to the resistive random access memory1The working signals are square wave pulse signals, each group of working signals comprises a positive voltage and a negative voltage, the amplitude range of the positive voltage is 0-4 volts, the duration range is 1 millisecond-1 second, the negative voltage range is 0-4 volts, the duration range is 1 millisecond-1 second, the resistance value of the resistive random access memory is read after the positive voltage is applied and is recorded as a low-resistance value, the resistance value of the resistive random access memory is read after the negative voltage is applied and is recorded as a high-resistance value, after a group of working signals are applied, the difference value of the resistance values of the high resistance state read for the current time and the next time is less than 5 percent of the resistance value of the high resistance state read for the previous time, and the difference value of the resistance values in the low resistance state read in the previous two times is less than 5% of the resistance value in the low resistance state read in the previous time, and (4) considering the resistive random access memory to be in a qualified state, repeating the step, and continuously applying more than 10 groups of working signals V.1When the resistance random access memory is in a qualified state, the resistance random access memory is considered to reach a stable state;
(3) setting a test temperature T1；
(4) Setting a conductance value G to be measured after the resistance random access memory reaches the stable state in the step (2)0，G0In the range of resistance change randomSetting a conductance fluctuation threshold lambda between the high resistance state resistance value and the low resistance state resistance value of the memory, and measuring the transient stability of the resistive random access memory, wherein the specific process is as follows:
(4-1) applying an excitation signal V to the resistive random access memory2The excitation signal is a square wave pulse signal, the amplitude range of the square wave pulse signal is 0.3-0.7 volt, and the duration time is 10 milliseconds-1 microsecond;
(4-2) after 10 milliseconds, applying a read signal V to the RRAM3The reading signal is a square wave pulse signal, the amplitude of the square wave pulse signal is 0.1 volt, and the duration of the square wave pulse signal is 10 milliseconds;
(4-3) reading the conductance value G of the RRAM1According to the value of conductance G0With a conductance fluctuation threshold lambda, to a conductance value G1Make a judgment if G0+λ≥G1≥G0λ, then step (5) is carried out, if G1<G0- λ, then step (4-4); if G is1>G0And + lambda, then carrying out the step (4-6);
(4-4) adding an excitation signal by a difference of 0.1 volt, namely applying an excitation signal to the resistive random access memory, wherein the amplitude of the excitation signal is 0.4-0.8 volt, the duration of the excitation signal is 10 milliseconds, and after 10 milliseconds, applying a square wave pulse with the amplitude of 0.1 volt and the duration of 10 milliseconds; reading conductance value G of resistive random access memory2；
(4-5) according to the conductance value G0With a fluctuation threshold lambda, to a conductance value G2Make a judgment if G0+λ≥G2≥G0λ, then step (5) is carried out, if G2<G0- λ, then returning to step (4-4); if G is2>G0And + lambda, then carrying out the step (4-6);
(4-6) applying a reverse excitation signal-V to the resistive random access memory2The excitation signal is a square wave pulse signal, the amplitude range of the square wave pulse signal is-0.3 to-0.7 volt, the duration is 10 milliseconds to 1 microsecond, and the excitation signal is applied to the resistive random access memory after 10 millisecondsA read signal V3Reading the conductance value G of the resistance random access memory, wherein the reading signal is a square wave pulse signal, the amplitude of the square wave pulse signal is 0.1 volt, the duration of the square wave pulse signal is 10 milliseconds, and the conductance value G of the resistance random access memory is read3；
(4-7) according to the conductance value G0With a fluctuation threshold lambda, to a conductance value G3Make a judgment if G0+λ≥G3≥G0λ, then step (5) is carried out, if G3<G0Lambda, returning to step (4-4) if G is3>G0And + lambda, returning to the step (4-6);
(4-8) increasing the reverse excitation signal-V by a difference of-0.1 volt2Namely, an excitation signal is applied to the RRAM, the amplitude range of the excitation signal is-0.4 to-0.8 volt, a square wave pulse signal with the duration of 10 milliseconds is applied, a square wave pulse with the amplitude of 0.1 volt and the duration of 10 milliseconds is applied after the interval of 10 milliseconds, and the conductance value G of the RRAM is read4；
(4-9) according to the conductance value G0With a fluctuation threshold lambda, to a conductance value G4Make a judgment if G0+λ≥G4≥G0λ, then step (5) is carried out, if G4<G0Lambda, returning to step (4-4) if G is4>G0And + lambda, returning to the step (4-8);
(4-10) repeating the steps (4-4) to (4-7) until the electric conductance value of the resistive random access memory is between G0- λ and G0And (5) performing the step (5);
(5) calculating the transient stability of the resistive random access memory, which comprises the following specific steps:
(5-1) applying a plurality of square wave pulse signals with the amplitude of 0.1 volt and the time interval of 10 milliseconds to the resistive random access memory in the step (4), wherein the total duration of the pulse signals is 1 second, and respectively recording all conductance values G of the resistive random access memory within the range of 0 to 1 secondiI represents the serial number of the test;
(5-2) calculating the actual conductance values G in the ranges of 0 to 1 second using the following formulasiAnd (3) the conductance value G to be measured in the step (2)0Standard deviation σ of absolute difference:
wherein G isiRepresenting the actually measured conductance value of the RRAM during measurement, i representing the test sequence number, N being the number of times of the test of applying the square wave pulse signal in step 5, G0The resistance random access memory is used as a conductance value to be measured of the resistance random access memory;
(5-3) setting a threshold ζ of the standard deviation σ, determining the standard deviation σ in the step (5-2), and determining that the temperature is at the set temperature T when the standard deviation σ is less than or equal to the threshold ζ1The conductance value of the RRAM is G0Good transient stability if the standard deviation sigma is>And determining that the transient stability of the resistive random access memory is poor when the threshold value is zeta.
In one embodiment of the invention, the threshold ζ for the standard deviation σ is 0.5 microsiemens.
When the method is used, a plurality of test temperatures can be selected according to the test requirements, and the steps of the method are repeated to improve the test precision.
The following detailed description is provided for the purpose of facilitating understanding of the present invention and for the purpose of making more detailed and clear the objects, solutions and advantages of the present invention, and the specific parameters and details thereof are provided for the purpose of describing the exemplary embodiments only and are not intended to limit the same.
Example 1:
in this embodiment, the method for testing transient stability of the resistive random access memory includes: activating the resistive random access memory; applying a plurality of working signals to enable the resistive random access memory to reach a stable state; setting a test temperature; setting a conductance G to be measured0And applying an excitation signal to enable the conductance value of the RRAM to reach G0(ii) a And testing the transient stability of the resistive random access memory.
Step 1: and (3) performing an activation process on the resistive random access memory, and applying an activation voltage of 3 volts until the device is switched to a low-resistance state (<100 kilo ohms), so that the activation process is completed.
Step 2: applying 50 groups of working signals V to the resistive random access memory1The working signals are square wave pulse signals, each group of working signals comprises a positive voltage and a negative voltage, the amplitude of the positive voltage is 3 volts, the duration time range is 10 milliseconds, the negative voltage range is-3 volts, the duration time range is 10 milliseconds, the resistance value of the resistive random access memory is read after the positive voltage is applied and is recorded as a low-resistance value, the resistance value of the resistive random access memory is read after the negative voltage is applied and is recorded as a high-resistance value, after a group of working signals are applied, the high-resistance value and the low-resistance value read in two times are compared, the difference value of the high-resistance value read in two times is less than 5 percent of the high-resistance value read in the previous time, and when the difference value of the low-resistance value read in two times is less than 5 percent of the low-resistance value read in the previous time, the resistive random, repeating the steps, and continuously applying more than 10 groups of working signals V1When the resistance random access memory is in a qualified state, the resistance random access memory is considered to reach a stable state;
and step 3: the test temperature was set at 300K.
And 4, step 4: setting a conductance value G to be measured after the resistance random access memory reaches a stable state in step 20And (3) performing transient stability measurement after the fluctuation threshold value lambda is 0.5 microSiemens and the specific process is as follows:
step 4-1: applying an excitation signal V to the RRAM2The excitation signal is a square wave pulse signal, the amplitude range of the square wave pulse signal is 0.3 volt, and the duration time is 10 milliseconds;
step 4-2: after 10 milliseconds, a read signal V is applied to the RRAM3The reading signal is a square wave pulse signal, the amplitude of the square wave pulse signal is 0.1 volt, and the duration of the square wave pulse signal is 10 milliseconds;
step 4-3: reading conductance value G of resistive random access memory1According to the value of conductance G0With a fluctuation threshold lambda, to a conductance value G1Judging that if the molecular weight of 20.5 microsiemens is more than or equal to G1If G is more than or equal to 19.5 microsiemens, then step 5 is carried out1<19.5 microsiemens, then carrying out step 4-4; if G is1>20.5 microsiemens, then carrying out the step 4-6; the loading process is shown as the programming process of fig. 5.
Step 4-4: increasing the excitation signal V by a difference of 0.1 volt2Namely, a square wave pulse signal with the amplitude of 0.4 volt and the duration of 10 milliseconds is applied to the resistance random access memory, and after 10 milliseconds, a square wave pulse with the amplitude of 0.1 volt and the duration of 10 milliseconds is applied; reading conductance value G of resistive random access memory2。
And 4-5: if 20.5 microsiemens are not less than G2If G is more than or equal to 19.5 microsiemens, then step 5 is carried out2<19.5 micro siemens, and continuously returning to the step 4-4; if G is2>20.5 microsiemens, then proceed to steps 4-6.
And 4-6: applying an inverse excitation signal-V2The excitation signal is a square wave pulse signal, the amplitude of the square wave pulse signal is-0.3 volt, and the duration time is 10 milliseconds; after 10 milliseconds, a read signal V is applied to the RRAM3Reading the conductance value G of the resistance random access memory, wherein the reading signal is a square wave pulse signal, the amplitude of the square wave pulse signal is 0.1 volt, the duration of the square wave pulse signal is 10 milliseconds, and the conductance value G of the resistance random access memory is read3。
And 4-7: if 20.5 microsiemens are not less than G3If G is more than or equal to 19.5 microsiemens, then step 5 is carried out3<19.5 micro siemens, then returning to the step 4-4; if G is3>20.5 microsiemens, then proceed to steps 4-8.
And 4-8: increasing the reverse excitation signal-V by a difference of-0.1 volt2Namely, a square wave pulse signal with the amplitude of-0.4 volt and the duration of 10 milliseconds is applied to the resistance random access memory, and after 10 milliseconds, a square wave pulse with the amplitude of 0.1 volt and the duration of 10 milliseconds is applied; reading conductance value G of resistive random access memory4。
And 4-9: if 20.5 microsiemens are not less than G4Not less than 19.5 micro-westDoor, go to step 5, if G4<19.5 micro siemens, then returning to the step 4-4; if G is4>20.5 microsiemens, then return to step 4-8.
Step 4-10: repeating the steps 4-4 to 4-9 until the conductance value of the resistive random access memory is between 19.5 microsiemens and 20.5 microsiemens, and performing the step 5;
and 5: the transient stability of the resistive random access memory is tested, and the specific process is as follows:
step 5-1: applying a plurality of square wave pulse signals with the amplitude of 0.1 volt and the time interval of 10 milliseconds to the resistive random access memory in the step 4, wherein the total duration of the plurality of pulse signals is 1 second, and the loading process is shown as the reading process in fig. 5, and respectively recording all conductance values G of the resistive random access memory within the range of 0 to 1 secondiI represents the serial number of the test; the test results are shown in fig. 6.
Step 5-2: the actual conductance values G in the range of 0 to 1 second were calculated respectively using the following formulaiAnd step 2, measuring the conductance value G0Standard deviation σ of absolute difference:
wherein G isiRepresenting the actually measured conductance value of the RRAM during measurement, i representing the test sequence number, N being the number of times of the test of applying the square wave pulse signal in step 5, G0The resistance random access memory is used as a conductance value to be measured of the resistance random access memory;
step 5-3: setting a threshold ζ of one standard deviation σ to 0.5 microsiemens, determining the standard deviation σ in step 5-2, and determining that the temperature T is set at the threshold ζ when the standard deviation σ is less than or equal to the threshold ζ1The conductance value of the RRAM is G0The transient stability is good, and if the standard deviation sigma is larger than the threshold value zeta, the transient stability of the resistive random access memory is judged to be poor.
Claims (1)
1. A method for testing transient stability of a resistive random access memory is characterized by comprising the following steps:
(1) using an activation signal V0Activating the resistance random access memory to form a conductive filament in the resistance random access memory;
wherein the activation signal V0The signal is a direct current signal or a pulse signal, wherein the amplitude range of the pulse signal is 0 to 5 volts, the pulse width range is 100 nanoseconds to 100 microseconds, and the amplitude range of the direct current signal is 0 to 4 volts; the conductive filament is formed, namely the resistance random access memory reaches a low resistance state, so that the resistance value of the resistance random access memory is smaller than 100 kilo-ohm;
(2) applying multiple groups of working signals V to the resistive random access memory1The working signals are square wave pulse signals, each group of working signals comprises a positive voltage and a negative voltage, the amplitude range of the positive voltage is 0-4 volts, the duration range is 1 millisecond-1 second, the negative voltage range is 0-4 volts, the duration range is 1 millisecond-1 second, the resistance value of the resistive random access memory is read after the positive voltage is applied and is recorded as a low-resistance value, the resistance value of the resistive random access memory is read after the negative voltage is applied and is recorded as a high-resistance value, after a group of working signals are applied, the difference value of the resistance values of the high resistance state read for the current time and the next time is less than 5 percent of the resistance value of the high resistance state read for the previous time, and the difference value of the resistance values in the low resistance state read in the previous two times is less than 5% of the resistance value in the low resistance state read in the previous time, and (4) considering the resistive random access memory to be in a qualified state, repeating the step, and continuously applying more than 10 groups of working signals V.1When the resistance random access memory is in a qualified state, the resistance random access memory is considered to reach a stable state;
(3) setting a test temperature T1；
(4) Setting a conductance value G to be measured after the resistance random access memory reaches the stable state in the step (2)0，G0The method comprises the following steps of setting a conductance fluctuation threshold lambda between a high resistance state resistance value and a low resistance state resistance value of the resistive random access memory, and measuring the transient stability of the resistive random access memory, wherein the specific process is as follows:
(4-1) applying an excitation to the resistive random access memorySignal V2The excitation signal is a square wave pulse signal, the amplitude range of the square wave pulse signal is 0.3-0.7 volt, and the duration time is 10 milliseconds-1 microsecond;
(4-2) after 10 milliseconds, applying a read signal V to the RRAM3The reading signal is a square wave pulse signal, the amplitude of the square wave pulse signal is 0.1 volt, and the duration of the square wave pulse signal is 10 milliseconds;
(4-3) reading the conductance value G of the RRAM1According to the value of conductance G0With a conductance fluctuation threshold lambda, to a conductance value G1Make a judgment if G0+λ≥G1≥G0λ, then step (5) is carried out, if G1＜G0- λ, then step (4-4); if G is1＞G0And + lambda, then carrying out the step (4-6);
(4-4) adding an excitation signal by a difference of 0.1 volt, namely applying an excitation signal to the resistive random access memory, wherein the amplitude of the excitation signal is 0.4-0.8 volt, the duration of the excitation signal is 10 milliseconds, and after 10 milliseconds, applying a square wave pulse with the amplitude of 0.1 volt and the duration of 10 milliseconds; reading conductance value G of resistive random access memory2；
(4-5) according to the conductance value G0With a fluctuation threshold lambda, to a conductance value G2Make a judgment if G0+λ≥G2≥G0λ, then step (5) is carried out, if G2＜G0- λ, then returning to step (4-4); if G is2＞G0And + lambda, then carrying out the step (4-6);
(4-6) applying a reverse excitation signal-V to the resistive random access memory2The excitation signal is a square wave pulse signal, the amplitude range of the square wave pulse signal is-0.3 to-0.7 volt, the duration is 10 milliseconds to 1 microsecond, and after 10 milliseconds, a reading signal V is applied to the resistive random access memory3Reading the conductance value G of the resistance random access memory, wherein the reading signal is a square wave pulse signal, the amplitude of the square wave pulse signal is 0.1 volt, the duration of the square wave pulse signal is 10 milliseconds, and the conductance value G of the resistance random access memory is read3；
(4-7) according to the conductance value G0With a fluctuation threshold lambda, to a conductance value G3Make a judgment if G0+λ≥G3≥G0λ, then step (5) is carried out, if G3＜G0Lambda, returning to step (4-4) if G is3＞G0And + lambda, returning to the step (4-6);
(4-8) increasing the reverse excitation signal-V by a difference of-0.1 volt2Namely, an excitation signal is applied to the RRAM, the amplitude range of the excitation signal is-0.4 to-0.8 volt, a square wave pulse signal with the duration of 10 milliseconds is applied, a square wave pulse with the amplitude of 0.1 volt and the duration of 10 milliseconds is applied after the interval of 10 milliseconds, and the conductance value G of the RRAM is read4；
(4-9) according to the conductance value G0With a fluctuation threshold lambda, to a conductance value G4Make a judgment if G0+λ≥G4≥G0λ, then step (5) is carried out, if G4＜G0Lambda, returning to step (4-4) if G is4＞G0And + lambda, returning to the step (4-8);
(4-10) repeating the steps (4-4) to (4-7) until the electric conductance value of the resistive random access memory is between G0- λ and G0And (5) performing the step (5);
(5) calculating the transient stability of the resistive random access memory, which comprises the following specific steps:
(5-1) applying a plurality of square wave pulse signals with the amplitude of 0.1 volt and the time interval of 10 milliseconds to the resistive random access memory in the step (4), wherein the total duration of the pulse signals is 1 second, and respectively recording all conductance values G of the resistive random access memory within the range of 0 to 1 secondiI represents the serial number of the test;
(5-2) calculating the actual conductance values G in the ranges of 0 to 1 second using the following formulasiAnd (3) the conductance value G to be measured in the step (2)0Standard deviation σ of absolute difference:
wherein G isiRepresenting the actually measured conductance value of the RRAM during measurement, i representing the test sequence number, N being the number of times of the test of applying the square wave pulse signal in step 5, G0The resistance random access memory is used as a conductance value to be measured of the resistance random access memory;
(5-3) setting a threshold ζ of the standard deviation σ, determining the standard deviation σ in the step (5-2), and determining that the temperature is at the set temperature T when the standard deviation σ is less than or equal to the threshold ζ1The conductance value of the RRAM is G0The transient stability is good, and if the standard deviation sigma is larger than a threshold value zeta, the transient stability of the resistive random access memory is judged to be poor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011166822.9A CN112270949B (en) | 2020-10-27 | 2020-10-27 | Method for testing transient stability of resistive random access memory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011166822.9A CN112270949B (en) | 2020-10-27 | 2020-10-27 | Method for testing transient stability of resistive random access memory |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112270949A true CN112270949A (en) | 2021-01-26 |
| CN112270949B CN112270949B (en) | 2022-11-15 |
Family
ID=74342786
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011166822.9A Active CN112270949B (en) | 2020-10-27 | 2020-10-27 | Method for testing transient stability of resistive random access memory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112270949B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101882462A (en) * | 2009-05-08 | 2010-11-10 | 复旦大学 | Setting operation method of resistance random access memory |
| CN104335284A (en) * | 2012-05-25 | 2015-02-04 | 科洛斯巴股份有限公司 | High operating speed resistive random access memory |
| CN105869669A (en) * | 2015-01-14 | 2016-08-17 | 财团法人工业技术研究院 | Resistive random access memory and control method thereof |
| CN109147860A (en) * | 2017-06-27 | 2019-01-04 | 华邦电子股份有限公司 | Memory storage apparatus and its test method |
-
2020
- 2020-10-27 CN CN202011166822.9A patent/CN112270949B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101882462A (en) * | 2009-05-08 | 2010-11-10 | 复旦大学 | Setting operation method of resistance random access memory |
| CN104335284A (en) * | 2012-05-25 | 2015-02-04 | 科洛斯巴股份有限公司 | High operating speed resistive random access memory |
| CN105869669A (en) * | 2015-01-14 | 2016-08-17 | 财团法人工业技术研究院 | Resistive random access memory and control method thereof |
| CN109147860A (en) * | 2017-06-27 | 2019-01-04 | 华邦电子股份有限公司 | Memory storage apparatus and its test method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112270949B (en) | 2022-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109817267B (en) | Deep learning-based flash memory life prediction method and system and computer-readable access medium | |
| Wan et al. | A voltage-mode sensing scheme with differential-row weight mapping for energy-efficient RRAM-based in-memory computing | |
| US9047944B2 (en) | Resistance variable memory sensing | |
| KR19990078380A (en) | Reduced signal test for dynamic random access memory | |
| Ntinas et al. | Power-efficient noise-induced reduction of ReRAM cell’s temporal variability effects | |
| TWI670720B (en) | Memory functional tester, and testing method and manufacturing method of memory device | |
| CN112270949B (en) | Method for testing transient stability of resistive random access memory | |
| Nielen et al. | Study of memristive associative capacitive networks for CAM applications | |
| CN106653079A (en) | Resistive memory device and writing method thereof | |
| Shih et al. | Training-based forming process for RRAM yield improvement | |
| Brum et al. | Evaluating the impact of process variation on RRAMs | |
| US8184469B2 (en) | Stored multi-bit data characterized by multiple-dimensional memory states | |
| Copetti et al. | Exploring an on-chip sensor to detect unique faults in RRAMs | |
| CN106571165B (en) | A kind of test method and device of DDR device read-write | |
| CN109584930B (en) | Data reading method and nonvolatile memory device using the same | |
| CN112382331B (en) | Method for extracting characteristic parameters of conductance relaxation effect of resistive random access memory | |
| Seo et al. | Automatic ReRAM SPICE model generation from empirical data for fast ReRAM-circuit coevaluation | |
| CN116151344B (en) | Current compensation method and device for memristor array access resistor | |
| CN116994634B (en) | Memristor array fault test circuit | |
| CN112444765A (en) | Test method of turning voltage | |
| Radakovits et al. | Behavioral leakage and inter-cycle variability emulator model for rerams (BELIEVER) | |
| Li et al. | An FPGA-Based Training System for a 1T1R Memristor Array with 500 nS Conductance Resolution Limit | |
| Mbarek et al. | Design and properties of logic circuits based on memristor devices | |
| US10579759B2 (en) | Apparatus for modeling resistive memory devices | |
| WO2021036446A1 (en) | Test structure and test method |
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 |