WO2021036446A1 - 测试结构和测试方法 - Google Patents

测试结构和测试方法 Download PDF

Info

Publication number
WO2021036446A1
WO2021036446A1 PCT/CN2020/097747 CN2020097747W WO2021036446A1 WO 2021036446 A1 WO2021036446 A1 WO 2021036446A1 CN 2020097747 W CN2020097747 W CN 2020097747W WO 2021036446 A1 WO2021036446 A1 WO 2021036446A1
Authority
WO
WIPO (PCT)
Prior art keywords
relational expression
test
spin
electrical signal
resistive switching
Prior art date
Application number
PCT/CN2020/097747
Other languages
English (en)
French (fr)
Inventor
何世坤
王明
竹敏
Original Assignee
浙江驰拓科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 浙江驰拓科技有限公司 filed Critical 浙江驰拓科技有限公司
Publication of WO2021036446A1 publication Critical patent/WO2021036446A1/zh

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
    • G11C29/54—Arrangements for designing test circuits, e.g. design for test [DFT] tools
    • 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
    • G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing

Definitions

  • the present disclosure relates to the field of memory, and in particular, to a test structure and a test method.
  • FIGS. 1 and 2 are schematic diagrams of writing a memory structure 40 based on the spin-orbit moment supply line 10 and the MTJ. Can be used for the preparation of storage devices.
  • the write flip voltage is an important parameter.
  • test samples In order to obtain the distribution of the flip voltage of the device and ensure the functionality of the product in the mass production stage, more test samples will be selected in the research and development stage. At present, for the multi-sample flip voltage test, the samples are generally tested one by one, and then the test data is tested. Statistical Analysis.
  • the method for measuring the switching voltage of a large number of MTJ devices in the prior art includes the following steps:
  • Step S101 applying a voltage pulse of a certain amplitude and width to a single MTJ device
  • Step S102 measuring the resistance value of the device at a lower voltage
  • Step S103 increase the amplitude of the applied pulse voltage.
  • Fig. 3 is the voltage application timing chart of steps S101 and S102. Write at a high voltage that increases in sequence, and read resistance at a low voltage. Steps S101 and S102 are repeated in sequence. Until the resistance value jumps, as shown in Figure 4, the voltage applied at this time is recorded as the flip voltage of the device;
  • the above-mentioned test method needs to test multiple MTJs one by one, and several IV curve scans are required to obtain statistics of the distribution curve of the inversion voltage, and the test takes a long time.
  • the main purpose of the present disclosure is to provide a test structure and a test method to solve the technical problem of the long time required to obtain the distribution curve of the inversion voltage in the prior art.
  • a test structure including: at least one spin orbital moment providing line; a plurality of resistive switching devices, each of the resistive switching devices is located in the spin orbital moment providing line On the surface of the wire, the resistive switching device includes at least one magnetic layer; a first test electrode is electrically connected to the first end of the at least one spin orbital moment providing wire; a second test electrode is connected to at least one of the self The second end of the spin orbit moment supply line is electrically connected; the third test electrode is respectively electrically connected with one end of the plurality of resistive switching devices far away from the spin orbit moment supply line.
  • the resistive switching devices are arranged on the spin orbit moment supply lines in a one-to-one correspondence.
  • one end of each spin orbit moment is electrically connected to the first test electrode, and the other end of each spin orbit moment is electrically connected to the second test electrode.
  • a plurality of the spin orbit moment providing lines are connected in series to form a spin orbit moment providing group, one end of the spin orbit moment providing group is electrically connected to the first test electrode, and the spin orbit moment providing The other end of the group is electrically connected to the second test electrode.
  • each of the resistive switching devices is arranged on the surface of the spin orbit moment supply line at intervals, and one end of the spin orbit moment supply line is connected to the The first test electrode is electrically connected, and the other end of the spin orbit moment providing line is electrically connected to the second test electrode.
  • the spin orbit moment providing line includes a first spin segment, a second spin segment, and a third spin segment that are sequentially connected, wherein the first spin segment and the third spin segment The segments are parallel, and the first spin segment and the third spin segment have opposite current directions when passing current.
  • each of the resistive switching devices located on the first spin segment is electrically connected to one of the third test electrodes, and is located on the third spin segment
  • Each of the resistive switching devices is electrically connected to the other third test electrode.
  • the resistive switching device is an MTJ.
  • a test method including: step S2, applying an excitation electrical signal between a first test electrode and a second test electrode, or applying an excitation signal between the first test electrode and the second test electrode.
  • the test method further includes: step S1, initializing each of the resistive switching devices so that each of the resistive switching devices is in the same resistance state.
  • the step S2 further includes: in the step S2, applying an excitation electrical signal between the first test electrode and the second test electrode, and the step S2 further includes: A predetermined regulating voltage is applied to the three test electrodes.
  • an excitation electrical signal is applied between the first test electrode and the third test electrode, and the excitation electrical signal is an excitation voltage.
  • step S4 the excitation electrical signal applied last time is an end-point excitation electrical signal
  • the test method further includes: step S5, repeating step S2 and step S3 multiple times in sequence, Until the resistance of the resistive switching device no longer changes, a second relational expression is obtained, which is the relational expression between the resistance and the excitation electrical signal, and at least part of the obtained resistances are different
  • the difference between the subsequent excitation electrical signal and the previous excitation electrical signal is the first difference
  • the subsequent excitation electrical signal in step S4 is the same as
  • the difference between the previous excitation electrical signals is a second difference, and one of the first difference and the second difference is greater than zero, and the other is less than zero.
  • the step S6 includes: obtaining a third relational expression according to the first relational expression, and/or obtaining a fourth relational expression according to the second relational expression, the third relational expression and the fourth relational expression
  • the relational expressions are the relational expressions between the reciprocal of the resistance and the voltage corresponding to the excitation electrical signal; according to the third relational expression and/or the fourth relational expression, a fitting formula is obtained; according to the simulation
  • the electrical performance parameters are obtained by combining formulas.
  • obtaining a fitting formula according to the third relational expression and/or the fourth relational expression includes: obtaining a fifth relational expression according to the third relational expression, and/or according to the fourth relational expression
  • the sixth relational expression is obtained by the formula, and the fifth relational expression and the sixth relational expression are respectively the relational expressions between the switching probability density of the resistive switching device and the voltage corresponding to the excitation electrical signal;
  • the fifth relational expression and/or the sixth relational expression are fitted to obtain the fitting formula.
  • obtaining the fifth relational expression according to the third relational expression includes: taking a voltage corresponding to the excitation electrical signal as a variable to differentiate the third relational expression to obtain the fifth relational expression
  • Obtaining the sixth relational expression by the fourth relational expression includes: taking a voltage corresponding to the excitation electrical signal as a variable to differentiate the fourth relational expression to obtain the sixth relational expression, and to determine the fifth relational expression Fitting formula and/or the sixth relational expression to obtain the fitting formula, including: fitting a normal distribution to the fifth relational expression and/or the sixth relational expression to obtain the ⁇ formula.
  • obtaining a fitting formula according to the third relational expression and/or according to the fourth relational expression includes: accumulating the third relational expression and/or according to the fourth relational expression The probability distribution is fitted to obtain the fitting formula.
  • a plurality of resistive switching devices are arranged on the spin orbit moment supply line in one-to-one correspondence, the first test electrode is electrically connected to one end of the at least one spin orbit moment supply line, and the second test The electrode is electrically connected to one end of the at least one spin orbit moment supply line, and the third test electrode is electrically connected to the end of the resistive switching device far away from the spin orbit moment supply line.
  • the test method adopted in the present disclosure can measure multiple devices at once, thereby effectively saving test time and greatly improving test efficiency.
  • the MTJ is located in the same tiny area, and the test results can directly reflect the array statistical behavior of the device.
  • the present disclosure uses very few test resources and does not require hardware improvements to the existing test machine. At the same time, data processing is performed on a large number of test curves. , No need to remove invalid devices to get an effective flip distribution curve, simple operation.
  • 1 and 2 show the writing principle diagram of the storage structure 40 based on the spin-orbit moment supply line 10 and the MTJ;
  • Figure 3 shows a voltage application timing diagram of a single MTJ device in the prior art
  • FIG. 4 shows the relationship curve between the writing voltage and the resistance of a single MTJ device in the prior art
  • FIG. 5 shows a schematic diagram of the switching voltage distribution of multiple MTJ devices in the prior art
  • Fig. 6 shows a schematic diagram of a test structure according to an embodiment of the present disclosure
  • FIG. 7 shows a schematic diagram of another test structure according to an embodiment of the present disclosure.
  • FIG. 8 shows a schematic diagram of yet another test structure according to an embodiment of the present disclosure.
  • FIG. 9 shows a flowchart of a testing method according to an embodiment of the present disclosure.
  • FIG. 10 shows the resistance distribution of each MTJ in two states in a test structure according to an embodiment of the present disclosure
  • Fig. 11 shows a schematic diagram of an R-V relationship curve according to an embodiment of the present disclosure
  • FIG. 12 shows a schematic diagram of a 1/R-V relationship curve according to an embodiment of the present disclosure
  • FIG. 13 shows a schematic diagram of a normal distribution fitting curve according to an embodiment of the present disclosure.
  • the present disclosure proposes a test structure.
  • the test structure includes the following components: a spin orbit moment supply line 10, a plurality of resistive switching devices 20, a first test electrode 30, a second test electrode 31, and a third test ⁇ 32 ⁇ Electrode 32.
  • each of the above-mentioned resistive switching devices 20 is located on the surface of the above-mentioned spin-orbital moment providing wire 10, and the above-mentioned resistive switching device 20 includes at least one magnetic layer; One end is electrically connected; the second test electrode 31 is electrically connected to the second end of at least one of the above-mentioned spin orbit moment supply lines 10; the third test electrode 32 and the plurality of the above-mentioned resistive switching devices 20 are far away from the above-mentioned spin orbit moment supply lines One end of 10 is electrically connected.
  • Applying a voltage or current between the first test electrode 30 and the second test electrode 31 can write to the resistive switching device, change its state, and then between the second test electrode 31 and the third test electrode 32 or By applying a voltage or current between the third test electrode 32 and the first test electrode 30, the resistance of multiple resistive switching devices 20 can be tested, so that the switching voltage of at least one magnetic layer of each resistive switching device can be obtained. carry out testing.
  • Using the test structure for one test can test multiple resistive switching devices, and there is no need to test multiple resistive switching devices 20 one by one. Therefore, the test structure can effectively save test time and greatly improve test efficiency.
  • spin orbit moment supply lines there may be multiple or one spin orbit moment supply lines in the present disclosure, and those skilled in the art can choose and set an appropriate number of spin orbit moment supply lines according to actual conditions.
  • FIG. 6 and FIG. 7 there are multiple spin orbital moment supply lines 10, and the resistive switching devices 20 are arranged on the spin orbital moment in one-to-one correspondence. Provide on line 10.
  • each of the spin orbital moments is electrically connected to the first test electrode 30, and the other end of each of the spin orbital moments is respectively connected to the first test electrode 30.
  • the two test electrodes 31 are electrically connected.
  • multiple spin orbital moments are connected in parallel, and at the same time, each resistive switching device 20 is arranged on each orbital moment in a one-to-one correspondence, so that multiple resistive switching devices 20 can be measured at a time.
  • a plurality of the above-mentioned spin orbital moment providing lines 10 are connected in series to form a spin-orbital moment providing group, and one end of the above-mentioned spin orbital moment providing group is connected to the first
  • the test electrode 30 is electrically connected, and the other end of the spin orbit moment providing group is electrically connected to the second test electrode 31.
  • the series connection in this embodiment is another connection method of the orbital moment supply line. It should be noted that the connection of the orbital moment supply line in the present disclosure The method is not limited to the above two.
  • each resistive switching device 20 is arranged on the surface of the spin orbit moment providing wire 10 at intervals.
  • One end of the spin orbit moment providing wire 10 is electrically connected to the first test electrode 30, and the other end of the spin orbit moment providing wire 10 is electrically connected to the second test electrode 31.
  • the resistive switching device 20 located on the spin orbit moment supply line 10 can be written.
  • the shape of the spin orbital moment providing line 10 can be designed according to the actual situation, and it can be linear or bent. Those skilled in the art can follow the actual situation. A suitable shape of the spin orbit moment supply line 10 is set in the situation.
  • the shape of the spin orbit moment providing line 10 is a bent shape, and the spin orbit moment providing line 10 includes a first spin segment and a second spin segment connected in sequence.
  • the direction of the current is opposite when the current passes through the first spin segment and the third spin segment.
  • the resistive switching device of the present disclosure may be any resistive switching device including a magnetic layer. In a specific embodiment of the present disclosure, the above-mentioned resistive switching device is an MTJ.
  • test method of the above-mentioned test structure includes:
  • Step S2 applying an excitation electrical signal between the first test electrode 30 and the second test electrode 31, or applying an excitation electrical signal between the first test electrode 30 and the third test electrode 32;
  • Step S3 applying a predetermined electrical signal between the second test electrode 31 and the third test electrode 32 or between the third test electrode 32 and the first test electrode 30 to test the resistance of the plurality of resistive switching devices 20 ,
  • the predetermined electrical signal is smaller than the inverted electrical signal of any one of the magnetic layers of the resistive switching device 20, so as to ensure that the predetermined electrical signal does not change the state of the resistive switching device;
  • Step S4 repeat the above step S2 and the above step S3 several times in sequence until the resistance of the resistive switching device 20 no longer changes, and the first relational expression is obtained.
  • the resistance of the resistive switching device no longer changes that is, the resistance of the resistive switching device no longer changes.
  • the resistance of the device changes from a changing state to a stable state;
  • the above-mentioned first relational expression is the relational expression between the above-mentioned resistance and the above-mentioned excitation electrical signal, and at least part of the above-mentioned resistances obtained are different, during any two repetitions ,
  • the latter excitation electrical signal is greater or less than the previous excitation electrical signal, and during multiple repetitions, the excitation electrical signal changes monotonously; the latter excitation electrical signal is greater or less than the previous excitation electrical signal , Indicates that the excitation electrical signal changes monotonously, that is, the excitation electrical signal monotonically increases or decreases.
  • Step S6 Obtain electrical performance parameters of the resistive switching device at least according to the first relational expression, where the electrical performance parameters include at least one average value of the switching voltage of the magnetic layer and the standard deviation of the switching voltage.
  • an excitation electrical signal is applied between the first test electrode and the second test electrode, and then between the second test electrode and the third test electrode or between the third test electrode and the first test electrode Apply a predetermined electrical signal, test the resistance of multiple resistive switching devices, and repeatedly apply the excitation electrical signal to obtain the relationship between the excitation electrical signal and the resistance. According to the relationship, the electrical including the mean value of the switching voltage and the standard deviation of the switching voltage can be obtained. Performance parameters, these two parameters characterize the writing performance of the device.
  • the mean value and the standard deviation of the inversion excitation parameter corresponding to the data writing can be obtained respectively.
  • the above test method further includes: step S1, initializing each of the above-mentioned resistive switching devices so that each of the above-mentioned resistive switching devices is in the same resistance state.
  • the method of initializing each of the above-mentioned resistive switching devices may be applying voltage, current, or magnetic field.
  • step S2 an excitation electrical signal is applied between the first test electrode and the second test electrode, and the step S2 further includes: applying an excitation signal to the third test electrode.
  • Predetermined regulation voltage In the above step S4, the predetermined regulating voltage on the third test electrode is unchanged.
  • step S2 an excitation electrical signal is applied between the first test electrode and the third test electrode, and the excitation electrical signal is an excitation voltage.
  • the excitation electrical signal applied for the last time is the end excitation electrical signal
  • the test method further includes: step S5, repeating step S2 and step S3 in sequence Many times, until the resistance of the resistive switching device no longer changes, the resistance of the resistive switching device no longer changes; a second relational expression is obtained, and the second relational expression is the relational expression between the resistance and the excitation electrical signal
  • the difference between the above-mentioned excitation electric signal of the last time and the above-mentioned excitation electric signal of the previous time is the first difference
  • the above-mentioned excitation electric signal of the latter time in the above step S4 and the above-mentioned excitation signal of the previous time
  • the difference between the excitation electrical signals is a second difference, and one of the first difference and the second difference is greater than zero, and the other is less than zero.
  • step S5 When the first difference is greater than 0, it means that, in step S5, the excitation electrical signal of the next time increases monotonously on the basis of the electrical excitation signal of the previous time, and at the same time, in step S4, the excitation electrical signal of the next time is The signal decreases monotonously on the basis of the aforementioned electrical excitation signal of the previous time; when the aforementioned first difference is less than 0, it means that in step S5, the aforementioned electrical excitation signal of the subsequent time is monotonous on the basis of the aforementioned electrical excitation signal of the previous time.
  • step S4 the above-mentioned excitation electric signal of the last time monotonously increases on the basis of the above-mentioned excitation electric signal of the previous time.
  • the above step S6 includes: obtaining a third relational expression according to the above-mentioned first relational expression, and/or obtaining a fourth relational expression according to the above-mentioned second relational expression, the above-mentioned third relational expression and
  • the fourth relational expression is the relational expression between the reciprocal of the resistance and the voltage corresponding to the excitation electrical signal; according to the third relational expression and/or the fourth relational expression, a fitting formula is obtained; according to the fitting formula Obtain the above electrical performance parameters.
  • the excitation signal is set to S
  • the first relational expression or the second relational expression in the above embodiment can be an R-S relational curve
  • the third relational expression or the fourth relational expression can be 1/
  • the R-S relationship curve is obtained by fitting the third relationship equation and/or the fourth relationship equation to obtain a fitting formula, and then the electrical performance parameters are obtained according to the fitting formula, and the distribution curve of the inversion voltage is quickly obtained.
  • obtaining the fitting formula according to the third relational expression and/or according to the fourth relational expression includes: obtaining the fifth relational expression according to the third relational expression, and/or The sixth relational expression is obtained according to the above-mentioned fourth relational expression, the above-mentioned fifth relational expression and the above-mentioned sixth relational expression are respectively the relational expressions between the switching probability density of the resistive switching device and the voltage corresponding to the excitation electrical signal; The fifth relational expression and/or the above-mentioned sixth relational expression are fitted to obtain the above-mentioned fitting formula.
  • the excitation signal is set to V
  • the first relational expression in the above embodiment may be an RV relation curve
  • the third relational expression may be a 1/RV relational curve
  • the fifth relational expression is the above-mentioned resistance change
  • the relationship between the switching probability density of the device and the voltage corresponding to the above-mentioned excitation signal is P 12 (V).
  • the probability distribution function is normally fitted, and the fitting formula is:
  • C 0 , ⁇ and ⁇ are parameters obtained by fitting a normal distribution, ⁇ represents the mean value of the critical excitation condition, and ⁇ is used to represent the standard deviation of the critical excitation condition.
  • obtaining the fifth relational expression according to the third relational expression includes: taking the voltage corresponding to the excitation electrical signal as a variable to differentiate the third relational expression to obtain the fifth relational expression.
  • obtaining the sixth relational expression includes: taking the voltage corresponding to the excitation electrical signal as a variable to differentiate the fourth relational expression to obtain the sixth relational expression, and to compare the fifth relational expression and/ Or fitting the aforementioned sixth relational expression to obtain the aforementioned fitting formula includes: performing normal distribution fitting on the aforementioned fifth relational expression and/or the aforementioned sixth relational expression to obtain the aforementioned fitting formula.
  • the first relational expression or the second relational expression in the above embodiment can be an R-V relational curve
  • the third relational expression or the fourth relational expression can be a 1/R-V relational curve.
  • the fitting formula is obtained according to the above third relational expression and/or according to the above fourth relational expression, including: comparing the above-mentioned third relational expression and/or according to the above-mentioned fourth relational expression Perform cumulative probability distribution fitting to obtain the above-mentioned fitting formula, thereby quickly obtaining the distribution curve of the flipping voltage.
  • the test method further includes: determining the distribution relational expression of the write error rate of the resistive switching device according to the fitting formula.
  • the distribution of the write error rate of the resistive switching device helps to quickly obtain the distribution curve of the switching voltage.
  • the fitting formula used to characterize the voltage distribution flip detection results can be used to determine the distribution of the write error rate of the resistive switching device, and the calculation formula of the distribution of the write error rate of the resistive switching device is:
  • C 0 , V 0 , and ⁇ are parameters obtained by fitting a normal distribution
  • V 0 represents the mean value of the critical excitation condition
  • ⁇ is used to represent the standard deviation of the critical excitation condition
  • the write error rate can be obtained by entering the calculation formula Distribution.
  • the fitting method of the present disclosure can also be a variety of deformation forms such as Taylor expansion or Fourier expansion.
  • the resistive switching device of the present disclosure may be an MTJ.
  • the MTJ structure includes but is not limited to a free layer, a tunneling layer, and a reference layer.
  • the MTJ structure may also include a horizontal magnetization layer.
  • the material of the spin-orbit moment providing wire of the present disclosure includes, but is not limited to, heavy metals Pt, Ta, W, Ir, TI, Bi, Au and the like.
  • test structure is shown in Figure 6. The following steps are used in the process of testing with the test structure:
  • Step S201 initialize each resistive switching device 20, so that each resistive switching device 20 is in the same resistance state
  • step S201 a sufficiently large voltage is applied between the first test electrode 30 and the second test electrode 31 so that all the resistive switching devices 20 are in the same state.
  • a magnetic field generating device is used to make all the resistive switching devices 20 in the same state to realize the initialization of each resistive switching device 20.
  • step S202 the first end of all the spin orbit moment providing wires 10 is connected to the first test electrode 30, the second end is connected to the second test electrode 31, and one end of the parallel resistive switching device 20 is connected to the third test electrode 32. Applying an excitation electrical signal between the test electrode 30 and the second test electrode 31;
  • Step S203 between the second test electrode 31 and the third test electrode 32, or between the first test electrode 30 and the third test electrode 32, test the resistance of the multiple resistive switching devices 20, and record the excitation electrical signal and the resistance.
  • Step S204 Step S202 and Step S203 are repeated in sequence for multiple times until at least one magnetic layer of the resistive switching device 20 is inverted, and a first relational expression is obtained.
  • the first relational expression is the relational expression between the resistance and the excitation electrical signal;
  • step S205 the electrical performance parameters of the resistive switching device 20 are obtained at least according to the first relational expression.
  • the electrical performance parameters include the average value of the switching voltage and the standard deviation of the switching voltage of the at least one magnetic layer.
  • the test structure is composed of a plurality of resistive switching devices 20 and a spin orbit moment supply line 10.
  • the first end of the spin orbit moment supply line 10 is connected to the first test electrode 30, and the second end is connected to the second test electrode 30.
  • the test electrode 31 is connected, and one end of the parallel resistive switching device 20 is connected to the third test electrode 32.
  • the two stable states of the plurality of resistive switching devices 20 are state 1 and state 2, respectively.
  • the corresponding resistances of the two states are R1 and R2, when When the applied excitation electrical signal is greater than a certain value, part of the device flips, and the response is that the total parallel resistance changes.
  • the turnover probability distribution of the device can be obtained as
  • the distribution of the switching voltage can be obtained.
  • the switching voltage distribution from state 2 to state 1 can be obtained as
  • V 0 V mean is the mean value of the switching voltage
  • a simulation operation is also performed on the test method. It is assumed that the test structure is a parallel structure of 1000 MTJs, and the distribution of various parameters of each MTJ is shown in Table 1. Then the resistance distribution of each MTJ two states in the test structure is shown in Figure 10. Impulse voltages of different amplitudes are applied to the test structure, and the resistance value R of the test structure is measured, and the resistance change relationship (R-V curve diagram) under different voltages is obtained, as shown in FIG. 11.
  • the normal fitting result is shown in Figure 13.
  • test structure is shown in Figure 7. The following steps are used in the process of testing with the test structure:
  • Step S201 initialize each resistive switching device 20, so that each resistive switching device 20 is in the same resistance state
  • step S201 a sufficiently large voltage is applied between the first test electrode 30 and the second test electrode 31 so that all the resistive switching devices 20 are in the same state.
  • a magnetic field generating device is used to make all the resistive switching devices 20 in the same state to realize the initialization of each resistive switching device 20.
  • step S202 the spin orbit moment supply lines 10 are connected in series to form a spin orbit moment supply group.
  • One end of the spin orbit moment supply group is connected to the first test electrode 30, and the other end is connected to the second test electrode 31.
  • One end of the device 20 is connected to the third test electrode 32, and an excitation electrical signal is applied between the first test electrode 30 and the second test electrode 31;
  • Step S203 between the second test electrode 31 and the third test electrode 32, or between the first test electrode 30 and the third test electrode 32, test the resistance of the multiple resistive switching devices 20, and record the excitation electrical signal and the resistance.
  • Step S204 Step S202 and Step S203 are repeated in sequence for multiple times until at least one magnetic layer of the resistive switching device 20 is inverted, and a first relational expression is obtained.
  • the first relational expression is the relational expression between the resistance and the excitation electrical signal;
  • step S205 the electrical performance parameters of the resistive switching device 20 are obtained at least according to the first relational expression.
  • the electrical performance parameters include the average value of the switching voltage and the standard deviation of the switching voltage of the at least one magnetic layer.
  • test structure is shown in Figure 8. The following steps are used in the process of testing with this test structure:
  • Step S301 initialize each resistive switching device 20 so that each resistive switching device 20 is in the same resistance state
  • step S301 there is one spin orbit moment supply line 10, one end of the spin orbit moment supply line 10 is connected to the first test electrode 30, and the other end is connected to the second test electrode 31.
  • a sufficiently large voltage is applied between the two test electrodes 31 so that all the resistive switching devices 20 are in the same state.
  • a magnetic field generating device is used to make all the resistive switching devices 20 in the same state to realize the initialization of each resistive switching device 20.
  • the spin orbit moment providing line 10 includes a first spin segment, a second spin segment, and a third spin segment that are sequentially connected, wherein the first spin segment and the third spin segment are parallel, and the second spin segment is parallel to the third spin segment.
  • the spin segments are perpendicular to the first spin segment and the third spin segment, respectively.
  • Each resistive switching device 20 on the first spin segment is electrically connected to a third test electrode 32.
  • Each resistive switching device 20 on the third spin segment is electrically connected to another third test electrode 32;
  • Step S303 between the second test electrode 31 and the third test electrode 32, or between the first test electrode 30 and the third test electrode 32, test the resistance of the multiple resistive switching devices 20, and record the excitation electrical signal and The relationship between the resistances of the two sets of resistive switching devices 20;
  • Step S304 Step S302 and Step S303 are repeated in sequence for several times, until at least one magnetic layer of the two sets of resistive switching devices 20 is inverted, and a first relational expression is obtained.
  • the first relational expression is the relational expression between the resistance and the excitation electrical signal ;
  • step S305 the electrical performance parameters of the resistive switching device 20 are obtained at least according to the first relational expression.
  • the electrical performance parameters include the average value of the switching voltage and the standard deviation of the switching voltage of the at least one magnetic layer.
  • test resources used by the present disclosure are few, and there is no need to improve the hardware of the existing test machine;
  • test structure proposed in the present disclosure it is possible to test multiple resistive switching devices at one time, without the need to test multiple resistive switching devices one by one. Therefore, the test structure can effectively save test time and greatly improve test efficiency .

Landscapes

  • Hall/Mr Elements (AREA)

Abstract

本公开提供了一种测试结构和测试方法。该测试结构包括：至少一个自旋轨道矩提供线；多个阻变器件，各阻变器件位于自旋轨道矩提供线的表面上，阻变器件包括至少一个磁性层；第一测试电极，与至少一个自旋轨道矩提供线的一端电连接；第二测试电极，与至少一个自旋轨道矩提供线的一端电连接；第三测试电极，与阻变器件的远离自旋轨道矩提供线的一端电连接。利用本公开的测试结构进行测试，可以一次进行多个器件的测量，获取统计信息进而可以有效节省测试时间，极大提升测试效率。

Description

测试结构和测试方法
本公开以2019年8月30日递交的、申请号为201910819082.5且名称为“测试结构和测试方法”的专利文件为优先权文件，其全部内容通过引用结合在本公开中。
技术领域
本公开涉及存储器领域，具体而言，涉及一种测试结构和测试方法。
背景技术
在具有自旋轨道矩效应(spin orbit torque，SOT)或自旋霍尔效应(spin Hall effect)的材料中通入电流时，会在材料的界面处产生自旋极化的自旋电流，该自旋电流可以用于翻转纳米磁铁，例如磁性隧道结中的自由层，图1和图2为基于自旋轨道矩提供线10和MTJ的存储结构40写入原理图。可以用于存储器件的制备。
作为一种存储器件，写入翻转电压是一个重要参数。
在研发阶段为了拿到器件翻转电压的分布，确保量产阶段产品的功能性，测试样本的选取会较多，目前对多样本翻转电压测试，一般会逐个对样本进行测试，然后对测试数据进行统计分析。
在研发阶段为了得到MTJ器件的翻转电压分布，分析MTJ的性能及失效机理，会测试大量MTJ样品，一般会逐个对样本进行测试，然后对测试数据进行处理分析，得到最终的翻转电压分布。
具体地，现有技术中测量大量MTJ器件的翻转电压的方法包括以下步骤：
步骤S101，施加一定幅值和宽度的电压脉冲作用于单个MTJ器件；
步骤S102，在较低电压下测量器件的电阻值；
步骤S103，增加施加的脉冲电压幅值，图3为步骤S101以及S102的电压施加时序图，在依次递增的高电压下写入，在低电压下读取阻值，依次重复步骤S101和S102，直到电阻值发生跳变，如图4所示，记录此时施加的电压即为该器件的翻转电压；
对多个MTJ器件重复以上步骤，得到每个器件的翻转电压。统计处理数据即可得到器件的翻转电压分布，如图5所示。
上述的测试方法需要对多个MTJ逐个进行测试，需要进行若干次的IV曲线扫描才能统计得出翻转电压的分布曲线，测试花费时间长。
在背景技术部分中公开的以上信息只是用来加强对本文所描述技术的背景技术的理解，因此，背景技术中可能包含某些信息，这些信息对于本领域技术人员来说并未形成在本国已知的现有技术。
发明内容
本公开的主要目的在于提供一种测试结构和测试方法，以解决现有技术中获取翻转电压的分布曲线所需要的时间较长的技术问题。
为了实现上述目的，根据本公开的一个方面，提供了一种测试结构，包括：至少一个自旋轨道矩提供线；多个阻变器件，各所述阻变器件位于所述自旋轨道矩提供线的表面上，所述阻变器件包括至少一个磁性层；第一测试电极，与至少一个所述自旋轨道矩提供线的第一端电连接；第二测试电极，与至少一个所述自旋轨道矩提供线的第二端电连接；第三测试电极，分别与多个所述阻变器件的远离所述自旋轨道矩提供线的一端电连接。
可选地，所述自旋轨道矩提供线有多个，且所述阻变器件一一对应地设置在所述自旋轨道矩提供线上。
可选地，各所述自旋轨道矩的一端分别与所述第一测试电极电连接，各所述自旋轨道矩的另一端分别与所述第二测试电极电连接。
可选地，多个所述自旋轨道矩提供线串联形成自旋轨道矩提供组，所述自旋轨道矩提供组的一端与所述第一测试电极电连接，所述自旋轨道矩提供组的另一端与所述第二测试电极电连接。
可选地，所述自旋轨道矩提供线有一个，各所述阻变器件间隔地设置在所述自旋轨道矩提供线的表面上，所述自旋轨道矩提供线的一端与所述第一测试电极电连接，所述自旋轨道矩提供线的另一端与所述第二测试电极电连接。
可选地，所述自旋轨道矩提供线包括依次连接的第一自旋段、第二自旋段和第三自旋段，其中，所述第一自旋段和所述第三自旋段平行，所述第一自旋段和第三自旋段在通过电流时电流方向相反。
可选地，所述第三测试电极有两个，位于所述第一自旋段上的各所述阻变器件与一个所述第三测试电极电连接，位于所述第三自旋段上的各所述阻变器件与另一个所述第三测试电极电连接。
可选地，所述阻变器件为MTJ。
为了实现上述目的，根据本公开的另一个方面，提供了一种测试方法，包括：步骤S2，在第一测试电极和第二测试电极之间施加激励电信号，或者在所述第一测试电极和第三测试电极之间施加所述激励电信号；步骤S3，在所述第二测试电极和所述第三测试电极之间或者在所述第三测试电极和所述第一测试电极之间施加预定电信号，测试多个阻变器件的电阻， 所述预定电信号小于所述阻变器件的任意一个磁性层的翻转电信号；步骤S4，依次重复执行所述步骤S2和所述步骤S3多次，直到所述阻变器件的电阻不再发生变化，得到第一关系式，所述第一关系式为所述电阻与所述激励电信号之间的关系式，且得到的至少部分所述电阻不同，任意两次重复过程中，后一次的所述激励电信号大于或者小于前一次的所述激励电信号，且多次所述重复过程中，所述激励电信号单调变化；步骤S6，至少根据所述第一关系式获取所述阻变器件的电性能参数，所述电性能参数包括至少一个所述磁性层的翻转电压的均值和所述翻转电压的标准差。
可选地，在所述步骤S2之前，所述测试方法还包括：步骤S1，初始化各所述阻变器件，使得各所述阻变器件处于相同的阻态。
可选地，所述步骤S2中还包括：所述步骤S2中，在所述第一测试电极和所述第二测试电极之间施加激励电信号，所述步骤S2还包括：在所述第三测试电极上施加预定调控电压。
可选地，所述步骤S2中，在所述第一测试电极和所述第三测试电极之间施加激励电信号，所述激励电信号为激励电压。
可选地，所述步骤S4中，最后一次施加的所述激励电信号为终点激励电信号，所述测试方法还包括：步骤S5，依次重复执行所述步骤S2和所述步骤S3多次，直到所述阻变器件的电阻不再发生变化，得到第二关系式，所述第二关系式为所述电阻与所述激励电信号之间的关系式，且得到的至少部分所述电阻不同，多次重复过程中，后一次的所述激励电信号与前一次的所述激励电信号之间的差值为第一差值，所述步骤S4中的后一次的所述激励电信号与前一次的所述激励电信号之间的差值为第二差值，所述第一差值和所述第二差值中的一个大于0，另一个小于0。
可选地，所述步骤S6包括：根据所述第一关系式获取第三关系式，和/或根据所述第二关系式获取第四关系式，所述第三关系式和所述第四关系式分别为所述电阻的倒数与所述激励电信号对应的电压之间的关系式；根据所述第三关系式和/或所述第四关系式，得到拟合公式；根据所述拟合公式获取所述电性能参数。
可选地，根据所述第三关系式和/或所述第四关系式，得到拟合公式，包括：根据所述第三关系式获取第五关系式，和/或根据所述第四关系式获取第六关系式，所述第五关系式和所述第六关系式分别为所述阻变器件的翻转概率密度与所述激励电信号对应的电压之间的关系式；对所述第五关系式和/或所述第六关系式进行拟合，得到所述拟合公式。
可选地，根据所述第三关系式获取第五关系式，包括：以所述激励电信号对应的电压为变量对所述第三关系式取微分，得到所述第五关系式，根据所述第四关系式获取所述第六关系式，包括：以所述激励电信号对应的电压为变量对所述第四关系式取微分，得到所述第六关系式，对所述第五关系式和/或所述第六关系式进行拟合，得到所述拟合公式，包括：对所述第五关系式和/或所述第六关系式进行正态分布拟合，得到所述拟合公式。
可选地，根据所述第三关系式和/或跟据所述第四关系式，得到拟合公式，包括：对所述第三关系式和/或跟据所述第四关系式进行累计概率分布拟合，得到所述拟合公式。
应用本公开的技术方案，将多个阻变器件一一对应地设置在自旋轨道矩提供线上，将第一测试电极，与至少一个自旋轨道矩提供线的一端电连接，第二测试电极，与至少一个自旋轨道矩提供线的一端电连接，第三测试电极，与阻变器件的远离所述自旋轨道矩提供线的一端电连接。相较于现有技术中的需要逐个测试多个器件的测量方法来说，本公开采用的测试方法可以一次进行多个器件的测量，进而可以有效节省测试时间，极大提升测试效率，且所有MTJ位于同一微小区域，测试结果可以直接反应器件的阵列统计行为，本公开利用的测试资源很少，不需要对现有测试机台进行硬件上的改进，同时对大数量个测试曲线进行数据处理，无需剔除无效器件就能得到有效的翻转分布曲线，操作简单。
附图说明
构成本公开的一部分的说明书附图用来提供对本公开的进一步理解，本公开的示意性实施例及其说明用于解释本公开，并不构成对本公开的不当限定。在附图中：
图1和图2示出了基于自旋轨道矩提供线10和MTJ的存储结构40写入原理图；
图3示出了现有技术中的单个MTJ器件的电压施加时序图；
图4示出了现有技术中的单个MTJ器件的写入电压和电阻之间的关系曲线；
图5示出了现有技术中的多个MTJ器件的翻转电压分布情况示意图；
图6示出了根据本公开的实施例的一种测试结构示意图；
图7示出了根据本公开的实施例的另一种测试结构示意图；
图8示出了根据本公开的实施例的再一种测试结构示意图；
图9示出了根据本公开的实施例的测试方法流程图；
图10示出了根据本公开的实施例的测试结构中每个MTJ两状态的电阻分布；
图11示出了根据本公开的实施例的R-V关系曲线示意图；
图12示出了根据本公开的实施例的1/R-V关系曲线示意图；
图13示出了根据本公开的实施例的正态分布拟合曲线示意图。
其中，上述附图包括以下附图标记：
10、自旋轨道矩提供线；20、阻变器件；30、第一测试电极；31、第二测试电极；32、第三测试电极；40、MTJ的存储结构。
具体实施方式
应该指出，以下详细说明都是例示性的，旨在对本公开提供进一步的说明。除非另有指明，本文使用的所有技术和科学术语具有与本公开所属技术领域的普通技术人员通常理解的相同含义。
需要注意的是，这里所使用的术语仅是为了描述具体实施方式，而非意图限制根据本公开的示例性实施方式。如在这里所使用的，除非上下文另外明确指出，否则单数形式也意图包括复数形式，此外，还应当理解的是，当在本说明书中使用术语“包含”和/或“包括”时，其指明存在特征、步骤、操作、器件、组件和/或它们的组合。
应该理解的是，当元件(诸如层、膜、区域、或衬底)描述为在另一元件“上”时，该元件可直接在该另一元件上，或者也可存在中间元件。而且，在说明书以及权利要求书中，当描述有元件“连接”至另一元件时，该元件可“直接连接”至该另一元件，或者通过第三元件“连接”至该另一元件。
正如背景技术所介绍的，现有需要逐个测试多个器件才能获得有效的翻转电压分布曲线，为了解决如上获取翻转电压的分布曲线所需要的时间较长的技术问题，本公开提出了一种测试结构。
如图6、图7和图8所示，该测试结构包括如下组成部分：自旋轨道矩提供线10、多个阻变器件20、第一测试电极30、第二测试电极31和第三测试电极32。其中，各上述阻变器件20位于上述自旋轨道矩提供线10的表面上，上述阻变器件20包括至少一个磁性层；第一测试电极30与至少一个上述自旋轨道矩提供线10的第一端电连接；第二测试电极31与至少一个上述自旋轨道矩提供线10的第二端电连接；第三测试电极32与多个上述阻变器件20的远离上述自旋轨道矩提供线10的一端电连接。
在第一测试电极30和第二测试电极31之间施加电压或者电流，就可以对阻变器件进行写入，改变其状态，然后在上述第二测试电极31和第三测试电极32之间或者在上述第三测试电极32和上述第一测试电极30之间施加电压或电流，就可以测试多个上述阻变器件20的电阻，从而可以获得各阻变器件的至少一个上述磁性层的翻转电压进行测试。利用该测试结构进行一次测试可以测试多个阻变器件，并不需逐个对多个阻变器件20的测试，因此，该测试结构可以有效节省测试时间，极大提升测试效率。
本公开的自旋轨道矩提供线可以有多个，也可以有一个，本领域技术人员可以根据实际情况选择设置合适数量的自旋轨道矩提供线。
在本公开的一种具体的实施方式中，如图6和图7所示，上述自旋轨道矩提供线10有多个，且上述阻变器件20一一对应地设置在上述自旋轨道矩提供线10上。
在本公开的一种具体的实施方式中，如图6所示，各上述自旋轨道矩的一端分别与上述第一测试电极30电连接，各上述自旋轨道矩的另一端分别与上述第二测试电极31电连接。实际为多个自旋轨道矩并联，同时，各阻变器件20一一对应地设置在各轨道矩上，可实现一 次测量多个阻变器件20。
在本公开的一种具体的实施方式中，如图7所示，多个上述自旋轨道矩提供线10串联形成自旋轨道矩提供组，上述自旋轨道矩提供组的一端与上述第一测试电极30电连接，上述自旋轨道矩提供组的另一端与上述第二测试电极31电连接。相对于上一实施方式中的自旋轨道矩提供线10并联，本实施方式中的串联连接为轨道矩提供线的另一种连接方式，需要说明的是本公开中的轨道矩提供线的连接方式并不仅限于以上两种。
如图8所示，本公开的一种实施方式中，上述自旋轨道矩提供线10有一个，各上述阻变器件20间隔地设置在上述自旋轨道矩提供线10的表面上，上述自旋轨道矩提供线10的一端与上述第一测试电极30电连接，上述自旋轨道矩提供线10的另一端与上述第二测试电极31电连接。这样通过对该自旋轨道矩提供线10两端施加电压，就可以对位于该自旋轨道矩提供线10上的阻变器件20进行写入。
只有一个自旋轨道矩提供线10的方案中，自旋轨道矩提供线10的形状可以根据实际情况来设计，可以为直线型的，也可以为折弯型的，本领域技术人员可以根据实际情况设置合适形状的自旋轨道矩提供线10。
本公开的一种实施方式中，图8所示的结构中，自旋轨道矩提供线10的形状为折弯型，其自旋轨道矩提供线10包括依次连接的第一自旋段、第二自旋段和第三自旋段，其中，上述第一自旋段和上述第自旋段平行，上述第二自旋段分别与上述第一自旋段和上述第三自旋段连接，上述第一自旋段和第三自旋段在通过电流时电流方向相反。
由于第一自旋段和第三自旋段中的测试电流方向相反，为了准确测量各阻变器件20两端的测试电压，本公开的一种实施方式中，如图8所示，上述第三测试电极32有两个，位于上述第一自旋段上的各上述阻变器件20与一个上述第三测试电极32电连接，位于上述第三自旋段上的各上述阻变器件20与另一个上述第三测试电极32电连接。由于第一自旋段与第三自旋段中测试电流方向相反，变阻器件的状态变化方向相反，所以可以在施加一次测试激励的过程内同时测得两种状态变化的电性参数，加快了测试效率。本公开的阻变器件可以为任何包括磁性层的阻变器件，在本公开的一种具体的实施方式中，上述阻变器件为MTJ。
在本公开的另一种典型的实施方式中，如图9所示，上述测试结构的测试方法，包括：
步骤S2，在第一测试电极30和第二测试电极31之间施加激励电信号，或者在第一测试电极30和第三测试电极32之间施加激励电信号；；
步骤S3，在上述第二测试电极31和第三测试电极32之间或者在上述第三测试电极32和上述第一测试电极30之间施加预定电信号，测试多个上述阻变器件20的电阻，上述预定电信号小于上述阻变器件20的任意一个上述磁性层的翻转电信号，保证该预定电信号不改变阻变器件的状态；
步骤S4，依次重复执行上述步骤S2和上述步骤S3多次，直到上述阻变器件20的电阻不再发生变化，得到第一关系式，上述阻变器件的电阻不再发生变化，即上述阻变器件的电阻 从一个变化的状态到一个稳定不变的状态；上述第一关系式为上述电阻与上述激励电信号之间的关系式，且得到的至少部分上述电阻不同，任意两次重复过程中，后一次的上述激励电信号大于或者小于前一次的上述激励电信号，且多次上述重复过程中，上述激励电信号单调变化；上述后一次的激励电信号大于或者小于前一次的激励电信号，表示激励电信号单调变化，即激励电信号单调增大或者单调减小。
步骤S6，至少根据上述第一关系式获取上述阻变器件的电性能参数，上述电性能参数包括至少一个上述磁性层的翻转电压的均值和上述翻转电压的标准差。
上述的测试方法中，通过在第一测试电极和第二测试电极之间施加激励电信号，然后在第二测试电极和第三测试电极之间或者在第三测试电极和第一测试电极之间施加预定电信号，测试多个阻变器件的电阻，重复施加激励电信号得到激励电信号和电阻之间的关系式，根据该关系式可以得到包括翻转电压的均值和翻转电压的标准差的电性能参数，这两个参数表征器件的写入性能。通过上述测试方法，可以实现一次进行多个器件的测量，避免了现有技术中一次只能测试一个电阻的问题，进而可以有效节省测试时间，极大提升测试效率。
通过第一关系式或第二关系式的峰值和半高宽，可以分别获取数据写入对应的翻转激励参数的均值和标准差。
为了使在第一测试电极和第二测试电极之间施加激励电信号之前使所有的阻变器件处于相同的阻态，从而进一步保证测试结果的准确性，在本公开的一种具体的实施方式中，在上述步骤S2之前，上述测试方法还包括：步骤S1，初始化各上述阻变器件，使得各上述阻变器件处于相同的阻态。初始化各上述阻变器件的方式可以为施加电压、电流或者磁场。
在本公开的一种具体的实施方式中，上述步骤S2中，在上述第一测试电极和上述第二测试电极之间施加激励电信号，上述步骤S2还包括：在上述第三测试电极上施加预定调控电压。上述步骤S4中，上述第三测试电极上的预定调控电压不变。
在本公开的一种具体的实施方式中，上述步骤S2中，在上述第一测试电极和上述第三测试电极之间施加激励电信号，上述激励电信号为激励电压。
在本公开的一种具体的实施方式中，上述步骤S4中，最后一次施加的上述激励电信号为终点激励电信号，上述测试方法还包括：步骤S5，依次重复执行上述步骤S2和上述步骤S3多次，直到上述阻变器件的电阻不再发生变化，上述阻变器件的电阻不再发生变化；得到第二关系式，上述第二关系式为上述电阻与上述激励电信号之间的关系式，多次重复过程中，后一次的上述激励电信号与前一次的上述激励电信号之间的差值为第一差值，上述步骤S4中的后一次的上述激励电信号与前一次的上述激励电信号之间的差值为第二差值，上述第一差值和上述第二差值中的一个大于0，另一个小于0。
上述第一差值大于0时表示，在步骤S5中，后一次的上述激励电信号在前一次的上述激励电信号的基础上单调增大，同时，在步骤S4中，后一次的上述激励电信号在前一次的上述激励电信号的基础上单调减小；上述第一差值小于0时表示，在步骤S5中，后一次的上述激 励电信号在前一次的上述激励电信号的基础上单调减小，同时，在步骤S4中，后一次的上述激励电信号在前一次的上述激励电信号的基础上单调增大。
在本公开的一种具体的实施方式中，上述步骤S6包括：根据上述第一关系式获取第三关系式，和/或根据上述第二关系式获取第四关系式，上述第三关系式和上述第四关系式分别为上述电阻的倒数与上述激励电信号对应的电压之间的关系式；根据上述第三关系式和/或上述第四关系式，得到拟合公式；根据上述拟合公式获取上述电性能参数。
一种具体的实施例中，将激励信号设为S，上述实施方式中的第一关系式或者第二关系式可为R～S关系曲线，第三关系式或者第四关系式可为1/R～S关系曲线，通过对第三关系式和/或第四关系式进行拟合，得到拟合公式，进而根据拟合公式获取电性能参数，快速得到翻转电压的分布曲线。
在本公开的一种具体的实施方式中，根据上述第三关系式和/或根据上述第四关系式，得到拟合公式，包括：根据上述第三关系式获取第五关系式，和/或根据上述第四关系式获取第六关系式，上述第五关系式和上述第六关系式分别为上述阻变器件的翻转概率密度与上述激励电信号对应的电压之间的关系式；对上述第五关系式和/或上述第六关系式进行拟合，得到上述拟合公式。
一种具体的实施例中，将激励信号设为V，上述实施方式中的第一关系式可为R-V关系曲线，第三关系式可为1/R-V关系曲线，第五关系式为上述阻变器件的翻转概率密度与上述激励电信号对应的电压之间的关系式P 12(V)，对概率分布函数进行正态拟合，拟合公式为：
Figure PCTCN2020097747-appb-000001
其中，C 0、μ以及σ为正态分布拟合得到的参数，μ表示临界激励条件的均值，σ用来表示临界激励条件的标准差。
在本公开的一种具体的实施方式中，根据上述第三关系式获取第五关系式，包括：以上述激励电信号对应的电压为变量对上述第三关系式取微分，得到上述第五关系式，根据上述第四关系式获取第六关系式，包括：以上述激励电信号对应的电压为变量对上述第四关系式取微分，得到上述第六关系式，对上述第五关系式和/或上述第六关系式进行拟合，得到上述拟合公式，包括：对上述第五关系式和/或上述第六关系式进行正态分布拟合，得到上述拟合公式。
将激励信号设为V，上述实施方式中的第一关系式或者第二关系式可为R～V关系曲线，第三关系式或者第四关系式可为1/R～V关系曲线，对第三关系式取微分得到第五关系式
Figure PCTCN2020097747-appb-000002
对第四关系式取微分得到第六关系式
Figure PCTCN2020097747-appb-000003
对上述第五关系式和/或上述第六关系式进行正态拟合，拟合公式为：
Figure PCTCN2020097747-appb-000004
在本公开的一种具体的实施方式中，根据上述第三关系式和/或据上述第四关系式，得到拟合公式，包括：对上述第三关系式和/或据上述第四关系式进行累计概率分布拟合，得到上述拟合公式，从而快速得到翻转电压的分布曲线。
为了更精确地表示测试方法的测试原理，上述实施方式中，测试方法还包括：根据拟合公式确定阻变器件的写入错误率的分布关系式。其中阻变器件的写入错误率的分布，有助于快速得到翻转电压的分布曲线。用于表征电压分布翻转检测结果的拟合公式可用来确定阻变器件的写入错误率的分布情况，其中阻变器件写入错误率的分布情况的计算公式为：
Figure PCTCN2020097747-appb-000005
其中，C 0、V 0、σ为正态分布拟合得到的参数，V 0表示临界激励条件的均值，σ用来表示临界激励条件的标准差，带入计算公式即可得到写入错误率的分布。
当然，除上述的两种正态分布拟合方式外，本公开的拟合方式还可以为泰勒展开式或傅里叶展开式等多种变形形式。
本公开的阻变器件可为MTJ，MTJ结构包含但不限于自由层、隧穿层以及参考层，MTJ结构还可以包括水平磁化层。
本公开的自旋轨道矩提供线其材料包括但不限于重金属Pt，Ta，W，Ir，TI，Bi，Au等。
实施例1
测试结构如图6所示，用该测试结构进行测试的过程中以下步骤：
步骤S201，初始化各阻变器件20，使得各阻变器件20处于相同的阻态；
步骤S201中，通过在第一测试电极30和第二测试电极31之间施加一个足够大的电压使所有阻变器件20全部处于同一状态。或者利用磁场产生装置，使得所有阻变器件20处于同一状态实现各阻变器件20的初始化。
步骤S202，所有自旋轨道矩提供线10第一端与第一测试电极30相连，第二端与第二测试电极31相连，并联阻变器件20一端与第三测试电极32相连，在第一测试电极30和第二测试电极31之间施加激励电信号；
步骤S203，在第二测试电极31和第三测试电极32之间，或者在第一测试电极30和第三测试电极32之间，测试多个阻变器件20的电阻，记录激励电信号与阻变器件20的电阻之间的关系；
步骤S204，依次重复执行步骤S202和步骤S203多次，直到阻变器件20的至少一个磁性层发生翻转，得到第一关系式，第一关系式为电阻与激励电信号之间的关系式；
步骤S205，至少根据第一关系式获取阻变器件20的电性能参数，电性能参数包括至少一个磁性层的翻转电压的均值和翻转电压的标准差。
如图6所示，该测试结构由多个阻变器件20和自旋轨道矩提供线10组成，自旋轨道矩提供线10第一端与第一测试电极30相连，第二端与第二测试电极31相连，并联阻变器件20一端与第三测试电极32相连，多个阻变器件20两个稳定状态分别为状态1和状态2，两种状态对应的电阻分别为R1和R2，当施加的激励电信号大于某一值时，部分器件发生翻转，反应为总并联电阻发生变化。假设初始状态所有的阻变器件20全部处于状态1，在第一测试电极30和第二测试电极31之间施加电压信号为V时，每个阻变器件20由状态1翻转至状态2的概率为P 12(V),则外加电压由0扫至V时，总的并联电阻为
Figure PCTCN2020097747-appb-000006
对上式两边取微分，可得器件的翻转概率分布为
Figure PCTCN2020097747-appb-000007
即通过测量各个电压下并联电阻的大小，即可得到翻转电压的分布。同理可得状态2到状态1的翻转电压分布为
Figure PCTCN2020097747-appb-000008
根据测试得到的R～V曲线，得到1/R～V曲线，对曲线对V取数值微分，得到微分关系曲线d(1/R)/dV与V的曲线关系，对该曲线进行正态分布拟合得到：
Figure PCTCN2020097747-appb-000009
其中拟合参数
Figure PCTCN2020097747-appb-000010
V 0＝V mean即为翻转电压的均值，σ＝V sigma即为翻转电压的标准差。
本实施例对还对测试方法进行了仿真运算，假设测试结构为1000个MTJ的并联结构，其中每个MTJ的各项参数分布如表1所示。那么测试结构中每个MTJ两状态的电阻分布如图 10所示。对测试结构施加不同幅值的脉冲电压，测量测试结构的阻值R，得到不同电压下阻值的变化关系(R-V曲线图)，如图11所示。
表1
  均值(Ω) 变异系数(CV，％)
第一阻态电阻(R1) 36000 6％
第二阻态电阻(R2) 90000 8％
归一化翻转电压 0.5 8％
对如图11所示的R-V曲线进行处理得到1/R-V曲线如图12所示，在1/R-V曲线中使1/R对V取微分得到翻转电压的概率分布函数P 12(V)，对概率分布函数进行正态拟合，拟合公式为：
Figure PCTCN2020097747-appb-000011
正态拟合结果如图13所示，拟合结果为μ＝0.506，σ＝0.04，所以仿真结果翻转电压均值为0.505V，CV为7.99％，与假设在误差允许的范围内一致。
实施例2
测试结构如图7所示，用该测试结构进行测试的过程中以下步骤：
本实施例的测试方法包括以下步骤：
步骤S201，初始化各阻变器件20，使得各阻变器件20处于相同的阻态；
步骤S201中，通过在第一测试电极30和第二测试电极31之间施加一个足够大的电压使所有阻变器件20全部处于同一状态。或者利用磁场产生装置，使得所有阻变器件20处于同一状态实现各阻变器件20的初始化。
步骤S202，自旋轨道矩提供线10依次串联，形成自旋轨道矩提供组，自旋轨道矩提供组的一端与第一测试电极30相连，另一端与第二测试电极31相连，串联阻变器件20一端与第三测试电极32相连，在第一测试电极30和第二测试电极31之间施加激励电信号；
步骤S203，在第二测试电极31和第三测试电极32之间，或者在第一测试电极30和第三测试电极32之间，测试多个阻变器件20的电阻，记录激励电信号与阻变器件20的电阻之间的关系；
步骤S204，依次重复执行步骤S202和步骤S203多次，直到阻变器件20的至少一个磁性层发生翻转，得到第一关系式，第一关系式为电阻与激励电信号之间的关系式；
步骤S205，至少根据第一关系式获取阻变器件20的电性能参数，电性能参数包括至少一个磁性层的翻转电压的均值和翻转电压的标准差。
实施例3
测试结构如图8所示，用该测试结构进行测试的过程中以下步骤：
步骤S301，初始化各阻变器件20，使得各阻变器件20处于相同的阻态；
步骤S301中，自旋轨道矩提供线10为一个，自旋轨道矩提供线10的一端与第一测试电极30相连，另一端与第二测试电极31相连，通过在第一测试电极30和第二测试电极31之间施加一个足够大的电压使所有阻变器件20全部处于同一状态。或者利用磁场产生装置，使得所有阻变器件20处于同一状态实现各阻变器件20的初始化。
步骤S302，自旋轨道矩提供线10包括依次连接的第一自旋段、第二自旋段和第三自旋段，其中，第一自旋段和第三自旋段平行，第二自旋段分别与第一自旋段和第三自旋段垂直，第三测试电极32有两个，位于第一自旋段上的各阻变器件20与一个第三测试电极32电连接，位于第三自旋段上的各阻变器件20与另一个第三测试电极32电连接；
步骤S303，在第二测试电极31和第三测试电极32之间，或者在第一测试电极30和第三测试电极32之间，测试多个阻变器件20的电阻，同时记录激励电信号与两组阻变器件20的电阻之间的关系；
步骤S304，依次重复执行步骤S302和步骤S303多次，直到两组阻变器件20的至少一个磁性层发生翻转，得到第一关系式，第一关系式为电阻与激励电信号之间的关系式；
步骤S305，至少根据第一关系式获取阻变器件20的电性能参数，电性能参数包括至少一个磁性层的翻转电压的均值和翻转电压的标准差。
上述的三个实施例均具有如下的效果：
1)、相较于多个器件多次测量，并联结构同时测量可以有效节省测试时间，极大提升测试效率；
2)、所有MTJ位于同一微小区域，测试结果可以直接反应器件的阵列统计行为；
3)、本公开利用的测试资源很少，不需要对现有测试机台进行硬件上的改进；
4)、同时对大数量个MTJ的曲线进行处理，无需剔除失效器件就能得出有效的翻转分布曲线，操作简单。
从以上的描述中，可以看出，本公开上述的实施例实现了如下技术效果：
1)采用本公开所提出的测试结构，可以实现一次测试多个阻变器件，并不需逐个对多个阻变器件的测试，因此，该测试结构可以有效节省测试时间，极大提升测试效率。
2)采用本公开所提出的测试方法，可以实现一次进行多个器件的测量，进而可以有效节 省测试时间，极大提升测试效率。
以上所述仅为本公开的优选实施例而已，并不用于限制本公开，对于本领域的技术人员来说，本公开可以有各种更改和变化。凡在本公开的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本公开的保护范围之内。

Claims (17)

  1. 一种测试结构，其特征在于，包括：
    至少一个自旋轨道矩提供线；
    多个阻变器件，各所述阻变器件位于所述自旋轨道矩提供线的表面上，所述阻变器件包括至少一个磁性层；
    第一测试电极，与至少一个所述自旋轨道矩提供线的第一端电连接；
    第二测试电极，与至少一个所述自旋轨道矩提供线的第二端电连接；
    第三测试电极，分别与多个所述阻变器件的远离所述自旋轨道矩提供线的一端电连接。
  2. 根据权利要求1所述的测试结构，其特征在于，所述自旋轨道矩提供线有多个，且所述阻变器件一一对应地设置在所述自旋轨道矩提供线上。
  3. 根据权利要求2所述的测试结构，其特征在于，各所述自旋轨道矩的一端分别与所述第一测试电极电连接，各所述自旋轨道矩的另一端分别与所述第二测试电极电连接。
  4. 根据权利要求2所述的测试结构，其特征在于，多个所述自旋轨道矩提供线串联形成自旋轨道矩提供组，所述自旋轨道矩提供组的一端与所述第一测试电极电连接，所述自旋轨道矩提供组的另一端与所述第二测试电极电连接。
  5. 根据权利要求1所述的测试结构，其特征在于，所述自旋轨道矩提供线有一个，各所述阻变器件间隔地设置在所述自旋轨道矩提供线的表面上，所述自旋轨道矩提供线的一端与所述第一测试电极电连接，所述自旋轨道矩提供线的另一端与所述第二测试电极电连接。
  6. 根据权利要求5所述的测试结构，其特征在于，所述自旋轨道矩提供线包括依次连接的第一自旋段、第二自旋段和第三自旋段，其中，所述第一自旋段和所述第三自旋段平行，所述第一自旋段和第三自旋段在通过电流时电流方向相反。
  7. 根据权利要求6所述的测试结构，其特征在于，所述第三测试电极有两个，位于所述第一自旋段上的各所述阻变器件与一个所述第三测试电极电连接，位于所述第三自旋段上的各所述阻变器件与另一个所述第三测试电极电连接。
  8. 根据权利要求1至7中任一项所述的测试结构，其特征在于，所述阻变器件为MTJ。
  9. 一种用于测试权利要求1至8中任一项所述的测试结构的测试方法，其特征在于，包括：
    步骤S2，在第一测试电极和第二测试电极之间施加激励电信号，或者在所述第一测试电极和第三测试电极之间施加所述激励电信号；
    步骤S3，在所述第二测试电极和所述第三测试电极之间或者在所述第三测试电极和所述第一测试电极之间施加预定电信号，测试多个阻变器件的电阻，所述预定电信号小 于所述阻变器件的任意一个磁性层的翻转电信号；
    步骤S4，依次重复执行所述步骤S2和所述步骤S3多次，直到所述阻变器件的电阻不再发生变化，得到第一关系式，所述第一关系式为所述电阻与所述激励电信号之间的关系式，且得到的至少部分所述电阻不同，任意两次重复过程中，后一次的所述激励电信号大于或者小于前一次的所述激励电信号，且多次所述重复过程中，所述激励电信号单调变化；
    步骤S6，至少根据所述第一关系式获取所述阻变器件的电性能参数，所述电性能参数包括至少一个所述磁性层的翻转电压的均值和所述翻转电压的标准差。
  10. 根据权利要求9所述的测试方法，其特征在于，在所述步骤S2之前，所述测试方法还包括：
    步骤S1，初始化各所述阻变器件，使得各所述阻变器件处于相同的阻态。
  11. 根据权利要求9或10所述的测试方法，其特征在于，所述步骤S2中，在所述第一测试电极和所述第二测试电极之间施加激励电信号，所述步骤S2还包括：在所述第三测试电极上施加预定调控电压。
  12. 根据权利要求9或10所述的测试方法，其特征在于，所述步骤S2中，在所述第一测试电极和所述第三测试电极之间施加激励电信号，所述激励电信号为激励电压。
  13. 根据权利要求9所述的测试方法，其特征在于，所述步骤S4中，最后一次施加的所述激励电信号为终点激励电信号，所述测试方法还包括：
    步骤S5，依次重复执行所述步骤S2和所述步骤S3多次，直到所述阻变器件的电阻不再发生变化，得到第二关系式，所述第二关系式为所述电阻与所述激励电信号之间的关系式，且得到的至少部分所述电阻不同，多次重复过程中，后一次的所述激励电信号与前一次的所述激励电信号之间的差值为第一差值，所述步骤S4中的后一次的所述激励电信号与前一次的所述激励电信号之间的差值为第二差值，所述第一差值和所述第二差值中的一个大于0，另一个小于0。
  14. 根据权利要求13所述的测试方法，其特征在于，所述步骤S6包括：
    根据所述第一关系式获取第三关系式，和/或根据所述第二关系式获取第四关系式，所述第三关系式和所述第四关系式分别为所述电阻的倒数与所述激励电信号对应的电压之间的关系式；
    根据所述第三关系式和/或所述第四关系式，得到拟合公式；
    根据所述拟合公式获取所述电性能参数。
  15. 根据权利要求14所述的测试方法，其特征在于，根据所述第三关系式和/或所述第四关系式，得到拟合公式，包括：
    根据所述第三关系式获取第五关系式，和/或根据所述第四关系式获取第六关系式，所述第五关系式和所述第六关系式分别为所述阻变器件的翻转概率密度与所述激励电信号对应的电压之间的关系式；
    对所述第五关系式和/或所述第六关系式进行拟合，得到所述拟合公式。
  16. 根据权利要求15所述的测试方法，其特征在于，
    根据所述第三关系式获取第五关系式，包括：以所述激励电信号对应的电压为变量对所述第三关系式取微分，得到所述第五关系式，
    根据所述第四关系式获取所述第六关系式，包括：以所述激励电信号对应的电压为变量对所述第四关系式取微分，得到所述第六关系式，
    对所述第五关系式和/或所述第六关系式进行拟合，得到所述拟合公式，包括：对所述第五关系式和/或所述第六关系式进行正态分布拟合，得到所述拟合公式。
  17. 根据权利要求14所述的测试方法，其特征在于，根据所述第三关系式和/或跟据所述第四关系式，得到拟合公式，包括：
    对所述第三关系式和/或跟据所述第四关系式进行累计概率分布拟合，得到所述拟合公式。
PCT/CN2020/097747 2019-08-30 2020-06-23 测试结构和测试方法 WO2021036446A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910819082.5A CN112447250B (zh) 2019-08-30 2019-08-30 测试结构和测试方法
CN201910819082.5 2019-08-30

Publications (1)

Publication Number Publication Date
WO2021036446A1 true WO2021036446A1 (zh) 2021-03-04

Family

ID=74684502

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/097747 WO2021036446A1 (zh) 2019-08-30 2020-06-23 测试结构和测试方法

Country Status (2)

Country Link
CN (1) CN112447250B (zh)
WO (1) WO2021036446A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103165581A (zh) * 2011-12-16 2013-06-19 台湾积体电路制造股份有限公司 测试结构、其制造方法、测试方法、以及mram阵列
US20150214274A1 (en) * 2012-09-21 2015-07-30 Korea University Research And Business Foundation Horizontal magnetic memory element using inplane current and electric field
CN107689416A (zh) * 2016-08-04 2018-02-13 株式会社东芝 磁存储器
CN108109668A (zh) * 2017-11-28 2018-06-01 中电海康集团有限公司 一种磁存储器的测试方法、装置、存储介质及电子装置
CN109065707A (zh) * 2018-07-23 2018-12-21 华中科技大学 一种基于铁磁材料的多态存储器及多态存储方法
CN109256160A (zh) * 2018-09-13 2019-01-22 北京航空航天大学 一种自旋轨道矩磁存储器读取方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4100025B2 (ja) * 2002-04-09 2008-06-11 ソニー株式会社 磁気抵抗効果素子及び磁気メモリ装置
US8587993B2 (en) * 2009-03-02 2013-11-19 Qualcomm Incorporated Reducing source loading effect in spin torque transfer magnetoresisitive random access memory (STT-MRAM)
KR102149882B1 (ko) * 2012-01-01 2020-08-31 고쿠리츠다이가쿠호진 도호쿠다이가쿠 집적회로
EP3477647B1 (en) * 2017-10-27 2022-04-27 Karlsruher Institut für Technologie Efficient testing of a magnetic memory circuit
WO2019094864A1 (en) * 2017-11-13 2019-05-16 Massachusetts Institute Of Technology Magneto-ionic devices using a solid state proton pump and methods for using the same

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103165581A (zh) * 2011-12-16 2013-06-19 台湾积体电路制造股份有限公司 测试结构、其制造方法、测试方法、以及mram阵列
US20150214274A1 (en) * 2012-09-21 2015-07-30 Korea University Research And Business Foundation Horizontal magnetic memory element using inplane current and electric field
CN107689416A (zh) * 2016-08-04 2018-02-13 株式会社东芝 磁存储器
CN108109668A (zh) * 2017-11-28 2018-06-01 中电海康集团有限公司 一种磁存储器的测试方法、装置、存储介质及电子装置
CN109065707A (zh) * 2018-07-23 2018-12-21 华中科技大学 一种基于铁磁材料的多态存储器及多态存储方法
CN109256160A (zh) * 2018-09-13 2019-01-22 北京航空航天大学 一种自旋轨道矩磁存储器读取方法

Also Published As

Publication number Publication date
CN112447250B (zh) 2022-09-27
CN112447250A (zh) 2021-03-05

Similar Documents

Publication Publication Date Title
Fieback et al. Device-aware test: A new test approach towards DPPB level
US7158407B2 (en) Triple pulse method for MRAM toggle bit characterization
TWI379096B (en) Digital magnetic current sensor and logic
US7929334B2 (en) In-situ resistance measurement for magnetic random access memory (MRAM)
CN108109668B (zh) 一种磁存储器的测试方法、装置、存储介质及电子装置
JP2004519809A (ja) Ｍｒａｍセル用の基準
US6265885B1 (en) Method, apparatus and computer program product for identifying electrostatic discharge damage to a thin film device
TWI539457B (zh) 電阻式隨機存取記憶體以及其製作方法
US8942032B2 (en) Method for magnetic screening of arrays of magnetic memories
Wu et al. Characterization, modeling and test of synthetic anti-ferromagnet flip defect in STT-MRAMs
WO2021036446A1 (zh) 测试结构和测试方法
CN111223518B (zh) 用于阻性存储单元的测试结构及耐久性测试方法
CN212675922U (zh) 测试结构
US20110025339A1 (en) Magnetoresistive elelctrostatic discharge (esd) detector
US7042783B2 (en) Magnetic memory
US7138797B2 (en) Reverse magnetic reset to screen for weakly pinned heads
CN103117094B (zh) 闪存的测试方法
CN116991646B (zh) 磁存储器寿命预测方法、装置、电子设备及存储介质
US11869612B2 (en) Device aware test for memory units
US9355660B2 (en) Testing method of a magnetic head, and testing apparatus thereof
CN109961811A (zh) 一种自旋转移力矩mram的读电路
WO2020258822A1 (zh) 测试结构与测试方法
CN112444764A (zh) 翻转电压的测试方法
US20230358826A1 (en) Method for measuring an external magnetic field by at least one magnetic memory point
KR102033503B1 (ko) Mtj 소자의 유전체 막질 검증 방법 및 그 시스템

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20857545

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20857545

Country of ref document: EP

Kind code of ref document: A1