Modeling method, parameter adjustment method and modeling system of SPICE life model

Abstract

The present invention provides a modeling method, parameter adjustment method, modeling system and MOS transistor life determination method of a SPICE life model, which are applied to the semiconductor field. Since the functional relationship between each selected parameter in the SPICE life model provided by the present invention and the electrical characteristic decay characteristics of the target MOS transistor includes a time parameter, the electrical characteristic decay characteristics corresponding to the MOS transistor at each moment can be simulated quickly and accurately, thereby achieving the purpose of predicting the degradation effect of the electrical characteristics of the MOS transistor over time.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor technology, and in particular to a modeling method, a parameter adjustment method, a modeling system and a method for determining the life of a MOS transistor of a SPICE lifetime model. Background Art

[0002] SPICE (Simulation Program with Integrated Circuit Emphasis) is a language and simulator software used for circuit description and simulation. It is used to detect the integrity of circuit connections and functions, as well as to predict circuit behavior. SPICE is mainly used for simulation of analog circuits and mixed-signal circuits. If SPICE is to work well, device-level model parameters must be provided. The commonly used SPICE models in the industry include the BSIM series, PSP or empirical models. SPICE modeling engineers rely on device theory and experience to extract model parameters for use in SPICE simulation programs.

[0003] SPICE modeling is a bridge between semiconductor process manufacturing technology and circuit design. It provides circuit designers with device-level models for circuit simulation. A complete SPICE model for a process node generally includes models of MOSFET, BJT and related back-end metal interconnect layer capacitors (MOM capacitors), MOSFET parasitic resistance, MOS varactors, etc. The goal of the SPICE model is to use the BSIM model to accurately fit the IV curve of the device and the trend of electrical parameters changing with size, and extract the correct BSIM model parameter group.

[0004] At present, the modeling process for the MOS life model is as follows. Specifically, select the parameters that need to be life-tested in the BSIM model, that is, assume that the changes in these BSIM model parameters can fully simulate the changes in the electrical characteristics of the MOS transistor; then, based on the electrical parameters of a series of MOS transistors that have been degraded and measured after different periods of time under the conditions of applied electrical stress, a set of BSIM models are obtained by fitting the BSIM model for the measured data of the electrical properties of the MOS transistor at each time point, and each model corresponds to the measured data of the electrical properties of the MOS transistor at a time point. As the electrical properties of the MOS transistor continue to decay, only a few selected parameters of this set of BSIM models change. Establish a functional relationship between each selected parameter in these BSIM models and the attenuation characteristics of the electrical properties of the MOS transistor; based on the functional relationship between these selected parameters and the attenuation of the electrical properties of the MOS transistor, define the sub-circuit model, and the SPICE life model is completed.

[0005] Since the BSIM-based life model of MOS transistors in the existing life model does not directly include the time t factor, it is not possible to directly extrapolate other stress application times t for which there is no measured data. n The corresponding BSIM model value increases the modeling calculation amount, reduces the modeling speed, and ultimately reduces the model's efficiency in predicting the degradation effect of the electrical characteristics of MOS transistors over time and determining the efficiency of device life. Summary of the invention

[0006] The purpose of the present invention is to provide a modeling method of a SPICE lifetime model and a method for determining the lifetime of a MOS transistor, so as to solve the problem of low efficiency in predicting the degradation effect of the electrical characteristics of a MOS transistor over time in the existing lifetime model based on a BSIM model.

[0007] In order to solve the above technical problems, the present invention provides a modeling method of a SPICE life model, comprising:

[0008] Selecting a target MOS transistor, selecting at least one model parameter that needs to be life-tested from the model parameters of the BSIM model as a selected parameter, and establishing a functional relationship between the selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor;

[0009] The functional relationship is transformed into a sub-circuit life model of the selected parameters varying with time, and when the sub-circuit life models of all selected parameters are obtained, the SPICE life model of the target MOS transistor is established.

[0010] Optionally, the selected parameters include at least one of the following model parameters of the BSIM model: low field mobility U0, a first gate voltage dependent parameter UA of the effective mobility model, a second gate voltage dependent parameter UB of the effective mobility model, a body voltage dependent parameter UC of the effective mobility model, a zero body voltage long channel device threshold voltage VTH0, and a simulation parameter PCLM of the channel length modulation effect of channel output resistance degradation.

[0011] Optionally, the functional relationship between the selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor is:

[0012] (ΔIds / Ids0)×%=[(Ids0-Ids1) / Ids0]×100;

[0013] log(ΔP / P0)=α P +β P ×log(ΔIds / Ids0);

[0014] Wherein, ΔIds is the attenuation of the source-drain current of the MOS transistor, Ids0 is the source-drain current of the MOS transistor at time t=0 when no electrical stress is applied, Ids1 is the source-drain current of the MOS transistor after the MOS transistor is subjected to electrical stress of a preset duration, ΔP is the change of the selected parameter P, P0 is the initial value of the selected parameter P at time t=0 when no electrical stress is applied to the MOS transistor, α P and β P is the fitting parameter;

[0015] The sub-circuit life model of the selected parameters varying with time is:

[0016]

[0017] Among them, P i (t) is the parameter P selected at time t i The value of P i (0) is the selected parameter P when t = 0 i The value of A, n and For the selected parameter P i The fitting parameters of the intermediate variable i range from 1 to 7.

[0018] Based on the modeling method of the SPICE life model as described above, the present invention also provides a modeling system of the SPICE life model, comprising:

[0019] A functional relationship determination module is used to select a target MOS transistor, select at least one model parameter that needs to be life-tested from the model parameters of the BSIM model as a selected parameter, and establish a functional relationship between the selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor;

[0020] The life model establishment module is used to transform the functional relationship into a sub-circuit life model in which the selected parameters vary with time, and when the sub-circuit life models of all the selected parameters are obtained, the SPICE life model of the target MOS transistor is established.

[0021] Furthermore, based on the modeling method of the SPICE life model as described above, the present invention also provides a parameter adjustment method of the SPICE life model, comprising:

[0022] Based on the modeling method of the SPICE life model according to any one of claims 1 to 3, a corresponding SPICE life model is formed, wherein the SPICE life model includes a sub-circuit life model of at least one selected parameter;

[0023] For any of the selected parameters, determine the initial measured data of the electrical characteristic attenuation characteristic parameter of the target MOS transistor when no electrical stress is applied, and use the initial measured data as the initial value in the sub-circuit life model of the selected parameter, that is, P i (0);

[0024] Obtaining first measured data of the change of different electrical characteristic attenuation characteristic parameters of the target MOS transistor over time within a first preset time period through actual measurement;

[0025] The sub-circuit life model corresponding to the selected parameters is used to fit the first measured data, and the fitting parameters in the sub-circuit life model are adjusted during the fitting process so that the sub-circuit life model corresponding to the selected parameters can accurately fit the first measured data.

[0026] Optionally, when there are multiple selected parameters, each of the selected parameters corresponds to first measured data of a characteristic parameter of an electrical characteristic attenuation of the target MOS transistor changing over time, and the fitting parameters of the sub-circuit life model corresponding to the multiple selected parameters are correlated with each other;

[0027] The parameter adjustment method may also include: using the sub-circuit life model corresponding to each selected parameter to fit the first measured data corresponding to the selected parameter, and comprehensively coordinating the fitting parameters in the sub-circuit life model corresponding to each selected parameter, so that the sub-circuit life model corresponding to each selected parameter can accurately fit the corresponding first measured data.

[0028] Optionally, there are K selected parameters, which are P1 to P K The step of comprehensively coordinating the fitting parameters in the sub-circuit life model corresponding to each of the selected parameters may include:

[0029] Complete the selected parameters P1~P i-1 The fitting parameters in the corresponding sub-circuit life model are adjusted so that the selected parameters P1 to P i-1 The corresponding sub-circuit life models are all able to accurately fit the corresponding first measured data;

[0030] The currently selected parameter P i The fitting parameters in the corresponding sub-circuit life model are adjusted so that the currently selected parameter P i The corresponding sub-circuit life model can fit the corresponding first measured data to a required degree;

[0031] Unify and coordinate the selected parameters P1~P iThe corresponding fitting parameters in the sub-circuit life model are adjusted to adjust the selected parameters P1 to P i-1 The other fitting parameters in the corresponding sub-circuit life model make the selected parameters P1 to P i The corresponding sub-circuit life models can accurately fit the corresponding first measured data.

[0032] Optionally, the parameter adjustment method may further include:

[0033] Obtaining second measured data of the change of different electrical characteristic attenuation characteristic parameters of the target MOS transistor over time within a second preset time period through actual measurement;

[0034] fitting the second measured data using the sub-circuit life model corresponding to the selected parameters;

[0035] If the sub-circuit life model corresponding to the selected parameter does not match the fitting results of the first measured data and the second measured data, then for the selected parameter, at least one auxiliary parameter is selected from the model parameters of the BSIM model, and a sub-circuit life model of the auxiliary parameter changing with time is further obtained, and the sub-circuit life model corresponding to the auxiliary parameter and the sub-circuit life model corresponding to the selected parameter are uniformly adjusted, so that the sub-circuit life model corresponding to the selected parameter and the sub-circuit life model corresponding to the auxiliary parameter can accurately fit the corresponding first measured data and the second measured data.

[0036] Optionally, the auxiliary parameters include at least one of the following model parameters of the BSIM model: low field mobility U0, first gate voltage dependent parameter UA, second gate voltage dependent parameter UB, body voltage dependent parameter UC, zero body voltage long channel device threshold voltage VTH0, simulation parameter PCLM of channel length modulation effect of channel output resistance degradation, depletion region capacitance dependence coefficient nfactor of subthreshold region slope parameter, channel carrier saturation velocity vsat at standard temperature, gate voltage dependent parameter ags of body charge effect model and channel length dependent parameter a0 of body charge effect model.

[0037] In addition, based on the modeling method of the SPICE lifetime model as described above, the present invention also provides a method for determining the lifetime of a MOS transistor, comprising:

[0038] By the modeling method of the SPICE life model as described above, or by the parameter adjustment method of the SPICE life model as described above, a desired SPICE life model is obtained, wherein the SPICE life model includes at least one sub-circuit life model of a selected parameter;

[0039] Determining, according to the SPICE lifetime model, an initial value of the selected parameter corresponding to the electrical characteristic decay characteristic of the target MOS transistor for fitting when no electrical stress is applied to the target MOS transistor;

[0040] According to the SPICE lifetime model, the time required for the SPICE lifetime model corresponding to the selected parameter to decrease from the initial value to nine-tenths of the initial value after an electrical stress is applied to the target MOS transistor is determined, and the time is used as the lifetime of the MOS transistor.

[0041] Compared with the prior art, the technical solution of the present invention has at least one of the following beneficial effects:

[0042] The present invention provides a new modeling method of SPICE life model, specifically, by selecting at least one model parameter that needs to be life-tested from the model parameters of the BSIM model as a selected parameter, and establishing a functional relationship between the selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor; the functional relationship is transformed into a sub-circuit life model of the selected parameter that changes with time, and when the sub-circuit life models of all the selected parameters are obtained, the SPICE life model of the target MOS transistor is established. Since the functional relationship between each selected parameter in the SPICE life model provided by the present invention and the electrical characteristic attenuation characteristic of the target MOS transistor includes a time parameter, the electrical characteristic attenuation characteristic corresponding to the MOS transistor at each moment can be quickly and accurately simulated, thereby achieving the purpose of predicting the degradation effect of the electrical characteristic attenuation characteristic of the MOS transistor over time.

[0043] Furthermore, since the embodiments of the present invention can accurately simulate the electrical characteristic attenuation characteristics corresponding to the MOS transistor at each moment, the life of the MOS transistor can be ultimately determined by determining the time required for the selected parameter corresponding to the electrical characteristic of the MOS transistor to decrease from the initial value to one tenth of the initial value. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 Schematic diagram of a flow chart of a modeling method of a SPICE life model in one embodiment of the present invention;

[0045] Figure 2 A schematic diagram of the structure of a modeling system for a SPICE life model in an embodiment of the present invention

[0046] Figure 3 Schematic diagram of a flow chart of a parameter adjustment method of a SPICE life model in one embodiment of the present invention;

[0047] Figure 4 An example diagram of a parameter adjustment method of a SPICE life model in an embodiment of the present invention;

[0048] Figure 5 FIG. 4 is another example diagram of a parameter adjustment method of a SPICE life model in an embodiment of the present invention;

[0049] Figure 6 FIG. 4 is another example diagram of a parameter adjustment method of a SPICE life model in an embodiment of the present invention;

[0050] Figure 7 The figure is a flow chart of a method for determining the life of a SPICE life model in one embodiment of the present invention. DETAILED DESCRIPTION

[0051] The specific implementation of the present invention will be described in more detail below in conjunction with the schematic diagram. The advantages and features of the present invention will become clearer based on the following description. It should be noted that the drawings are all in a very simplified form and are not in exact proportions, and are only used to facilitate and clearly assist in explaining the purpose of the embodiments of the present invention.

[0052] As described in the background art, in the existing life model, the life model of MOS transistor based on the BSIM model does not directly include the time t factor, so it cannot directly extrapolate other applied stress time t without measured data. n The corresponding BSIM model value increases the modeling calculation amount, reduces the modeling speed, and ultimately reduces the model's efficiency in predicting the degradation effect of the electrical characteristics of MOS transistors over time and determining the efficiency of device life.

[0053] To this end, the present invention provides a modeling method for a SPICE lifetime model to solve the problem of low efficiency in predicting the degradation effect of the electrical characteristics of MOS transistors over time in the existing lifetime model based on the BSIM model.

[0054] See also Figure 1 , Figure 1 A flow chart of a modeling method of a SPICE life model provided by the present invention, the method comprising the following steps:

[0055] Step S101, select a target MOS transistor, select at least one model parameter that needs to be life-tested from the model parameters of the BSIM model as a selected parameter, and establish a functional relationship between the selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor.

[0056] In this embodiment, the change of the electrical characteristics of the MOS transistor can be simulated by changing one or more model parameters in the BSIM model. For example, the BSIM model can be used to predict the time degradation effect of the electrical characteristics of the MOS transistor caused by the hot carrier effect under the application of a certain electrical stress (also called the attenuation characteristic of the electrical characteristics of the MOS transistor). Afterwards, the time required for the initial value of the MOS transistor saturated source-drain current model through the SPICE lifetime model to decrease to nine-tenths of the initial value (the initial value is reduced to 90% of the initial value) is used as the lifetime of the MOS transistor, thereby determining the lifetime of the MOS transistor.

[0057] Among them, the selected parameters may include at least one of the following model parameters of the BSIM model: low field mobility U0, the first gate voltage dependence parameter UA of the effective mobility model, the second gate voltage dependence parameter UB of the effective mobility model, the body voltage dependence parameter UC of the effective mobility model, the zero body voltage long channel device threshold voltage VTH0 and the simulation parameter PCLM of the channel length modulation effect of channel output resistance degradation.

[0058] Furthermore, the functional relationship between the selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor provided by the present invention may specifically be:

[0059] Formula 1: (ΔIds / Ids0)×%=[(Ids0-Ids1) / Ids0]×100;

[0060] Formula 2: log(ΔP / P0) = α P +β P ×log(ΔIds / Ids0);

[0061] Wherein, ΔIds is the attenuation of the source-drain current of the MOS transistor, Ids0 is the source-drain current of the MOS transistor at time t=0 when no electrical stress is applied, Ids1 is the source-drain current of the MOS transistor after the MOS transistor is subjected to electrical stress of a preset duration, ΔP is the change of the selected parameter P, P0 is the initial value of the selected parameter P at time t=0 when no electrical stress is applied to the MOS transistor, α P and β P is the fitting parameter.

[0062] Step S102, transforming the functional relationship into a sub-circuit life model of the selected parameters varying with time, and when the sub-circuit life models of all selected parameters are obtained, the SPICE life model of the target MOS transistor is established.

[0063] In this embodiment, the functional relationship between each selected parameter established in the above step S100 and the electrical characteristic attenuation characteristic of the target MOS transistor, such as Formula 2, can be converted to obtain the following Formula 3 provided by the present invention, that is, the sub-circuit life model of each selected parameter changing with time. Specifically, log(ΔP / P0)=α P +β P ×log(ΔIds / Ids0) and perform exponential operations on both sides to obtain Formula 4: Then, expand ΔP in Formula 4 to obtain Formula 5 including the time parameter t: Then, use Formula 6: ΔIds / Ids0=At n and formula 7: Substitute it into Formula 5 to obtain the self-circuit life model of each selected parameter changing with time.

[0064] Specifically, as shown in the following formula 3, the sub-circuit life model of the selected parameter changing with time can be:

[0065]

[0066] Among them, P i (t) is the parameter P selected at time t i The value of P i (0) is the selected parameter P when t = 0 i The value of A, n and For the selected parameter P i The fitting parameters of the intermediate variable i range from 1 to 7.

[0067] Since the functional relationship between each selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor in the SPICE lifetime model provided by the present invention, such as Formula 3, includes a time parameter, the electrical characteristics corresponding to the MOS transistor at each moment can be simulated quickly and accurately, thereby achieving the purpose of predicting the degradation effect of the electrical characteristics of the MOS transistor under a certain electrical stress over time.

[0068] Based on the modeling method of the SPICE life model as described above, the present invention also provides a modeling system of the SPICE life model for implementing the modeling method as described above, such as Figure 2 The structure diagram of the modeling system of the SPICE life model shown, the modeling system comprises: a function relationship determination module 10 and a life model establishment module 20;

[0069] The functional relationship determination module 10 is used to select a target MOS transistor, select at least one model parameter that needs to be life-tested from the model parameters of the BSIM model as a selected parameter, and establish a functional relationship between the selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor;

[0070] The life model building module 20 is used to transform the functional relationship into a sub-circuit life model of the selected parameters changing with time, and when the sub-circuit life models of all the selected parameters are obtained, the SPICE life model of the target MOS transistor is established.

[0071] In addition, based on the modeling method of the SPICE life model as described above, the present invention also provides a parameter adjustment method of the SPICE life model, such as Figure 3 The flow chart of the parameter adjustment method of the SPICE life model shown in FIG. 1 is a flow chart of the parameter adjustment method of the SPICE life model, and the parameter adjustment method may include the following steps:

[0072] Step S301 : Based on the modeling method of the SPICE life model as described above, a corresponding SPICE life model is formed, wherein the SPICE life model includes a sub-circuit life model of at least one selected parameter.

[0073] Step S302, for any of the selected parameters, determine initial measured data of the electrical characteristic attenuation characteristic parameter of the target MOS transistor when no electrical stress is applied, and use the initial measured data as the initial value in the sub-circuit life model of the selected parameter.

[0074] Step S303 , obtaining first measured data of different electrical characteristic attenuation characteristic parameters of the target MOS transistor varying with time within a first preset time period through actual measurement.

[0075] In this embodiment, a certain electrical stress of a first preset time length is applied to the target MOS transistor, for example, a voltage of Vgs=Vds=2.2V, and then, a plurality of different electrical characteristic attenuation characteristic parameters of the target MOS transistor are measured at the first preset time length, and the corresponding first measured data thereof.

[0076] Step S304, using the sub-circuit life model corresponding to the selected parameters to fit the first measured data, and adjusting the fitting parameters in the sub-circuit life model during the fitting process, so that the sub-circuit life model corresponding to the selected parameters can accurately fit the first measured data.

[0077] In this embodiment, the first measured data of the attenuation characteristics of different electrical characteristics of the target MOS transistor determined in step S303 changing with time within the first preset time length is used as the formula Medium i (t), the initial value of the sub-circuit life model of the selected parameter obtained in step S302 is taken as P i (0), the first preset time length is used as the value of t, and then the formula is manually adjusted The fitting parameters in A, n and Thereby, the purpose of enabling the SPICE lifetime model corresponding to the selected parameters to accurately fit the electrical characteristic attenuation characteristics of the target MOS transistor is achieved.

[0078] Among them, the fitting parameters A, n and The adjustment range is: 0 <n<1。

[0079] It can be understood that when there are multiple selected parameters, each of the selected parameters corresponds to the first measured data of an electrical characteristic attenuation characteristic parameter of the target MOS transistor changing with time, and the fitting parameters of the sub-circuit life model corresponding to the multiple selected parameters are interrelated. In other words, an electrical characteristic of each of the target MOS transistors corresponds to one or more of the selected parameters, and by adjusting the fitting parameters of the sub-circuit life parameters of the one or more selected parameters, the electrical characteristic corresponding to the target MOS transistor can be effectively fitted. However, the selected parameter corresponding to the electrical characteristic of this MOS transistor may have an impact on another one or more electrical characteristic models of the MOS transistor, so the fitting parameters of the sub-circuit life parameters of all selected parameters need to be coordinated to make the SPICE life model completely fit the first measured data.

[0080] Furthermore, the parameter adjustment method described in the embodiment of the present invention may also include: using the sub-circuit life model corresponding to each selected parameter to fit the first measured data corresponding to the selected parameter, and comprehensively coordinating the fitting parameters in the sub-circuit life model corresponding to each selected parameter, so that the sub-circuit life model corresponding to each selected parameter (i.e., SPICE life model) can accurately fit the corresponding first measured data.

[0081] Specifically, if there are K selected parameters, namely P1 to P K , then the step of comprehensively coordinating the fitting parameters in the sub-circuit life model corresponding to each of the selected parameters provided by the present invention may include:

[0082] First, complete the selection of parameters P1~P i-1 The fitting parameters in the corresponding sub-circuit life model are adjusted so that the selected parameters P1 to P i-1The corresponding sub-circuit life models are all able to accurately fit the corresponding first measured data;

[0083] Next, the currently selected parameter P is performed. i The fitting parameters in the corresponding sub-circuit life model are adjusted so that the currently selected parameter P i The corresponding sub-circuit life model can fit the corresponding first measured data to a required degree;

[0084] Afterwards, the selected parameters P1 to P i The corresponding fitting parameters in the sub-circuit life model are adjusted to adjust the selected parameters P1 to P i-1 The other fitting parameters in the corresponding sub-circuit life model make the selected parameters P1 to P i The corresponding sub-circuit life models can accurately fit the corresponding first measured data.

[0085] In addition, the parameter adjustment method of the SPICE life model provided by the present invention may also include the following steps:

[0086] First, second measured data of the change of different electrical characteristic attenuation characteristic parameters of the target MOS transistor over time within a second preset time period is obtained through actual measurement;

[0087] Next, the sub-circuit life model corresponding to the selected parameters is used to fit the second measured data;

[0088] Secondly, if the sub-circuit life model corresponding to the selected parameter does not match the fitting results of the first measured data and the second measured data, then for the selected parameter, at least one auxiliary parameter is selected from the model parameters of the BSIM model, and a sub-circuit life model of the auxiliary parameter changing with time is further obtained, and the sub-circuit life model corresponding to the auxiliary parameter and the sub-circuit life model corresponding to the selected parameter are uniformly adjusted, so that the sub-circuit life model corresponding to the selected parameter and the sub-circuit life model corresponding to the auxiliary parameter can accurately fit the corresponding first measured data and the second measured data.

[0089] Among them, the auxiliary parameters include at least one of the following model parameters of the BSIM model: low field mobility U0, first gate voltage dependence parameter UA, second gate voltage dependence parameter UB, body voltage dependence parameter UC, zero body voltage long channel device threshold voltage VTH0, simulation parameter PCLM of channel length modulation effect of channel output resistance degradation, depletion region capacitance dependence coefficient nfactor of subthreshold region slope parameter, channel carrier saturation velocity vsat at standard temperature, gate voltage dependence parameter ags of body charge effect model and channel length dependence parameter a0 of body charge effect model.

[0090] The practical application of the above parameter adjustment method is specifically illustrated below to further explain the parameter adjustment method of the present invention.

[0091] Firstly, initial measured data of electrical characteristic attenuation characteristic parameters of the target MOS transistor corresponding to each selected parameter when no electrical stress is applied to the target MOS transistor are determined.

[0092] Then, for each selected parameter, determine the measured data of the electrical characteristic decay characteristic parameter of the target MOS transistor when the electrical characteristic decay characteristic corresponding to the selected parameter is applied at different electrical stress times, and use the SPICE lifetime model to fit the sub-circuit model corresponding to the selected parameter to the measured data. Specifically, if the selected parameter is the zero body voltage long channel device threshold voltage VTH0, for Figure 4 The measured data of the decay of the linear region threshold voltage vtlin of the MOS transistor over time are fitted. The fitting parameters γ(vth0) and β(vth0) of the sub-circuit life model of the BSIM model parameter VTH0 corresponding to the linear region threshold voltage Vtlin and the fitting parameters A and n are used to comprehensively coordinate and adjust the sub-circuit life model so that the SPICE life model can accurately fit the measured data of the decay of the linear threshold voltage Vtlin of the MOS device over time.

[0093] Secondly, if the selected parameter is low-field mobility U0, then Figure 5The measured data of the decay of the source-drain current Idlin in the linear region of the MOS transistor over time are fitted, and the fitting parameters γ(u0) and β(u0) of the sub-circuit life model of the BSIM model parameter U0 corresponding to the source-drain current Idlin in the linear region and the fitting parameters A and n are comprehensively coordinated, so that the SPICE life model can roughly fit the measured data of the decay of the source-drain current Idlin in the linear region of the MOS transistor over time. Since γ(u0) and β(u0) as well as A and n will affect the model value of the threshold voltage Vtlin in the linear region of the MOS transistor, after using these fitting parameters to make the model roughly fit the measured data of Idlin, it is necessary to use several fitting parameters of the sub-life circuit of VTH0 to adjust the Vtlin model, and uniformly coordinate the fitting parameters of the sub-circuit life models of VTH0 and U0, that is, using γ(vth0) and β(vth0) and γ(u0) and β(u0) as well as A and n to make the measured data of Vtlin and Idlin can be accurately fitted by the SPICE life model. Again, if the selected parameter is the channel carrier saturation velocity VSAT at standard temperature, Figure 6 The measured data of the MOS transistor saturated source-drain current Idsat decaying over time are fitted, and the fitting parameters γ(vsat) and β(vsat) of the sub-circuit life model corresponding to vsat and A and n are comprehensively coordinated. The SPICE life model can be adjusted to roughly fit the measured data of the MOS transistor saturated source-drain current Idsat decaying over time. Since A and n will affect the models of Vtlin and Idlin, it is necessary to comprehensively coordinate A and n as well as the parameters γ(vth0) and β(vth0) and γ(u0) and β(u0) of the models that affect Vtlin and Idlin so that the models of Vtlin, Idlin and Idsat can accurately fit their measured data.

[0094] After these steps, the characteristics of the electrical characteristics of the MOS transistor that decay over time after the application of electrical stress can basically be accurately fitted. However, if there is a slight discrepancy between the measured data of the MOS transistor current-voltage IV characteristics and the SPICE life model, the above adjustment steps can be repeated to fine-tune the life model to be more accurate.

[0095] After fitting the SPICE life model in the above embodiment, if it is found that the IV characteristic data of the MOS device measured after applying stress for 10 minutes and 100 minutes have any major inaccuracies in the fitting of the SPICE life model, the corresponding BSIM model parameters can be added and selected in a targeted manner, and formulated into a sub-circuit life model, and then the SPICE life model can be adjusted with the new fitting parameters to conform to the measured IV characteristic data of the MOS device.

[0096] Example 1: If the measured data of the subthreshold region IdVg and transconductance GM are inconsistent with the SPICE lifetime model, the selected parameters UA and UB can be added to perform the same formula operation as VTH0 in the subcircuit model lifetime parameter example to add the time t factor. For example, UA can be formulated in the subcircuit model as follows:

[0097] Ua='Ua0×(Rua×POWER(A,Bua)×POWER(T,N×Bua)+1)'

[0098] In fact, this is the abstract parameter P of formula 3 i Replace it with the specific parameter name UA, and then write it in the specific form in the subcircuit model. Among them, POWER(X,Y) represents X to the power of Y, and Ua0 represents the initial value of UA in the BSIM model parameters extracted when no electrical stress is applied. UB also performs similar operations, so that the SPICE life model can fit the measured data of Idlin's evolution over time and the measured data of Gm's evolution over time.

[0099] Example 2: If the measured data of IdVd and output resistance Rout are inconsistent with the SPICE life model, you can add selected a0, ags, and pclm to perform the same formulaic operation of adding the time t factor as vth0 in the sub-circuit life model example, so as to achieve unified parameter adjustment operations so that the SPICE life model can fit the measured data of Idsat evolving over time and the measured data of Rout evolving over time.

[0100] Example 3: If the measured data of the subthreshold region Id of IdVg (Vd=0.05V) and IdVg (Vd=1.2V) are inconsistent with the SPICE lifetime model, the selected parameter nfactor can be added to perform the same formula operation of adding the time t factor as vth0 in the sub-circuit model lifetime parameter example, so as to uniformly adjust the parameters so that the sub-circuit model lifetime parameters can fit the measured data of the evolution of the subthreshold region Id of IdVg over time and the measured data of the evolution of the threshold voltage Vtlin in the linear region over time.

[0101] Further, based on the modeling method of the SPICE lifetime model as described above, or the parameter adjustment method of the SPICE lifetime model as described above, the present invention also provides a method for determining the lifetime of a MOS transistor, such as Figure 6 FIG. 1 is a flow chart of a method for determining the life of a MOS transistor, wherein the method for determining the life of a MOS transistor comprises the following steps:

[0102] Step S601, obtaining a desired SPICE life model by the SPICE life modeling method as described above, or by the SPICE life model parameter adjustment method as described above, wherein the SPICE life model includes at least one sub-circuit life model of selected parameters.

[0103] Step S602 , determining, based on the SPICE lifetime model, initial values ​​of the selected parameters of the BSIM model for fitting the electrical characteristic degradation characteristics of the target MOS transistor when no electrical stress is applied thereto.

[0104] Step S603, determining, based on the SPICE lifetime model, the time required for the SPICE lifetime model corresponding to the selected parameter to decrease from the initial value to nine-tenths of the initial value after an electrical stress is applied to the target MOS transistor, and taking the time as the lifetime of the MOS transistor.

[0105] In this embodiment, since the embodiment of the present invention can accurately simulate the electrical characteristic attenuation characteristics corresponding to the MOS transistor at each moment, therefore, illustratively, the life of the MOS transistor can be ultimately determined by determining the time required for the initial value of the saturated source-drain current Idsat of the MOS transistor when no electrical stress is applied to decrease to nine-tenths of the SPICE life model corresponding to the saturated source-drain current Idsat when an electrical stress is applied to the target MOS transistor.

[0106] In summary, the present invention provides a new modeling method for SPICE life model, specifically by selecting at least one model parameter that needs to be life-tested from the model parameters of the BSIM model as a selected parameter, and establishing a functional relationship between the selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor; the functional relationship is transformed into a sub-circuit life model in which the selected parameter changes with time, and when the sub-circuit life models of all the selected parameters are obtained, the SPICE life model of the target MOS transistor is established. Since the functional relationship between each selected parameter in the SPICE life model provided by the present invention and the electrical characteristic attenuation characteristic of the target MOS transistor includes a time parameter, the electrical characteristic attenuation characteristic corresponding to the MOS transistor at each moment can be quickly and accurately simulated, thereby achieving the purpose of predicting the degradation effect of the electrical characteristic attenuation characteristic of the MOS transistor over time.

[0107] Furthermore, since the embodiments of the present invention can accurately simulate the electrical characteristic attenuation characteristics corresponding to the MOS transistor at each moment, for example, the life of the MOS transistor can be ultimately determined by determining the time required for the initial value of the saturated source-drain current Idsat of the MOS transistor when no electrical stress is applied to decrease to nine-tenths of the SPICE life model corresponding to the saturated source-drain current Idsat when an electrical stress is applied to the target MOS transistor.

[0108] It should be noted that, although the present invention has been disclosed as a preferred embodiment, the above embodiment is not intended to limit the present invention. For any technician familiar with the art, without departing from the scope of the technical solution of the present invention, the technical content disclosed above can be used to make many possible changes and modifications to the technical solution of the present invention, or modified into equivalent embodiments of equivalent changes. Therefore, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still belongs to the scope of protection of the technical solution of the present invention.

[0109] It should also be understood that, unless otherwise specified or indicated, the terms "first", "second", "third", etc. in the specification are merely used to distinguish between the various components, elements, steps, etc. in the specification, and are not used to indicate the logical relationship or sequential relationship between the various components, elements, steps, etc.

[0110] It should also be recognized that the terms described herein are only used to describe specific embodiments and are not intended to limit the scope of the invention. It should be noted that the singular forms "a" and "an" used herein and in the appended claims include plural references unless the context clearly indicates otherwise. For example, a reference to "a step" or "a device" means a reference to one or more steps or devices, and may include secondary steps and secondary devices. All conjunctions used should be understood in the broadest sense. And, the word "or" should be understood to have the definition of a logical "or", rather than a logical "exclusive or", unless the context clearly indicates otherwise. In addition, the implementation of the method and / or device in the embodiments of the present invention may include performing the selected task manually, automatically, or in combination.

Claims

1. A modeling method for a SPICE life model, characterized in that: include: Selecting a target MOS transistor, selecting at least one model parameter that needs to be life-tested from the model parameters of the BSIM model as a selected parameter, and establishing a functional relationship between the selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor; The functional relationship is transformed into a sub-circuit life model in which the selected parameters vary with time, and when the sub-circuit life models of all selected parameters are obtained, a SPICE life model of the target MOS transistor is established; The functional relationship between the selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor is: (ΔIds / Ids0)×%=[(Ids0-Ids1) / Ids0]×100; log(ΔP / P0)=α P +b P ×log(ΔIds / Ids0); Wherein, ΔIds is the attenuation of the source-drain current of the MOS transistor, Ids0 is the source-drain current of the MOS transistor at time t=0 when no electrical stress is applied, Ids1 is the source-drain current of the MOS transistor after the MOS transistor is subjected to electrical stress of a preset duration, ΔP is the change of the selected parameter P, P0 is the initial value of the selected parameter P at time t=0 when no electrical stress is applied to the MOS transistor, α P and β P is the fitting parameter; The sub-circuit life model of the selected parameters varying with time is: Among them, P i (t) is the parameter P selected at time t i The value of P i (0) is the selected parameter P when t = 0 i The value of γ Pi , A, n and β Pi For the selected parameter P i The fitting parameters of the intermediate variable i range from 1 to 7.

2. The modeling method of the SPICE life model according to claim 1, characterized in that: The selected parameters include at least one of the following model parameters of the BSIM model: low field mobility U0, a first gate voltage dependence parameter UA of the effective mobility model, a second gate voltage dependence parameter UB of the effective mobility model, a body voltage dependence parameter UC of the effective mobility model, a zero body voltage long channel device threshold voltage VTH0, and a simulation parameter PCLM of the channel length modulation effect of channel output resistance degradation.

3. A modeling system for a SPICE life model, characterized in that: The modeling system is used to implement the modeling method of the SPICE life model according to any one of claims 1 to 2, and the modeling system includes: A functional relationship determination module is used to select a target MOS transistor, select at least one model parameter that needs to be life-tested from the model parameters of the BSIM model as a selected parameter, and establish a functional relationship between the selected parameter and the electrical characteristic attenuation characteristic of the target MOS transistor; The life model establishment module is used to transform the functional relationship into a sub-circuit life model in which the selected parameters vary with time, and when the sub-circuit life models of all the selected parameters are obtained, the SPICE life model of the target MOS transistor is established.

4. A parameter adjustment method for a SPICE life model, characterized in that: include: Based on the modeling method of the SPICE life model described in any one of claims 1 to 2, a corresponding SPICE life model is formed, wherein the SPICE life model includes a sub-circuit life model of at least one selected parameter; For any of the selected parameters, determine initial measured data of the electrical characteristic attenuation characteristic parameter of the target MOS transistor when no electrical stress is applied to the target MOS transistor, and use the initial measured data as the initial value in the sub-circuit life model of the selected parameter; Obtaining first measured data of the change of different electrical characteristic attenuation characteristic parameters of the target MOS transistor over time within a first preset time period through actual measurement; The sub-circuit life model corresponding to the selected parameters is used to fit the first measured data, and the fitting parameters in the sub-circuit life model are adjusted during the fitting process so that the sub-circuit life model corresponding to the selected parameters can accurately fit the first measured data.

5. The parameter adjustment method of the SPICE life model according to claim 4, characterized in that: When there are multiple selected parameters, each of the selected parameters corresponds to first measured data of a characteristic parameter of an electrical characteristic attenuation of the target MOS transistor changing with time, and the fitting parameters of the sub-circuit life model corresponding to the multiple selected parameters are correlated with each other; The parameter adjustment method also includes: using the sub-circuit life model corresponding to each selected parameter to fit the first measured data corresponding to the selected parameter, and comprehensively coordinating the fitting parameters in the sub-circuit life model corresponding to each selected parameter, so that the sub-circuit life model corresponding to each selected parameter can accurately fit the corresponding first measured data.

6. The parameter adjustment method of the SPICE life model according to claim 4, characterized in that: There are K selected parameters, which are P1~P K The step of comprehensively coordinating the fitting parameters in the sub-circuit life model corresponding to each of the selected parameters comprises: Complete the selected parameters P1~P i-1 The fitting parameters in the corresponding sub-circuit life model are adjusted so that the selected parameters P1 to P i-1 The corresponding sub-circuit life models are all able to accurately fit the corresponding first measured data; The currently selected parameter P i The fitting parameters in the corresponding sub-circuit life model are adjusted so that the currently selected parameter P i The corresponding sub-circuit life model can fit the corresponding first measured data to a required degree; Unify and coordinate the selected parameters P1~P i The corresponding fitting parameters in the sub-circuit life model are adjusted to adjust the selected parameters P1 to P i-1 The other fitting parameters in the corresponding sub-circuit life model make the selected parameters P1 to P i The corresponding sub-circuit life models can accurately fit the corresponding first measured data.

7. The parameter adjustment method of the SPICE lifetime model according to any one of claims 4 to 6, characterized in that: Also includes: Obtaining second measured data of the change of different electrical characteristic attenuation characteristic parameters of the target MOS transistor over time within a second preset time period through actual measurement; fitting the second measured data using the sub-circuit life model corresponding to the selected parameters; If the sub-circuit life model corresponding to the selected parameter does not match the fitting results of the first measured data and the second measured data, then for the selected parameter, at least one auxiliary parameter is selected from the model parameters of the BSIM model, and a sub-circuit life model of the auxiliary parameter changing with time is further obtained, and the sub-circuit life model corresponding to the auxiliary parameter and the sub-circuit life model corresponding to the selected parameter are uniformly adjusted, so that the sub-circuit life model corresponding to the selected parameter and the sub-circuit life model corresponding to the auxiliary parameter can accurately fit the corresponding first measured data and the second measured data.

8. The parameter adjustment method of the SPICE life model according to claim 7, characterized in that: The auxiliary parameters include at least one of the following model parameters of the BSIM model: low field mobility U0, first gate voltage dependent parameter UA, second gate voltage dependent parameter UB, body voltage dependent parameter UC, zero body voltage long channel device threshold voltage VTH0, simulation parameter PCLM of channel length modulation effect of channel output resistance degradation, depletion region capacitance dependence coefficient nfactor of subthreshold region slope parameter, channel carrier saturation velocity vsat at standard temperature, gate voltage dependent parameter ags of body charge effect model and channel length dependent parameter a0 of body charge effect model.

9. A method for determining the life of a MOS transistor, characterized in that: include: The desired SPICE life model is obtained by the modeling method of the SPICE life model described in any one of claims 1 to 2, or by the parameter adjustment method of the SPICE life model described in any one of claims 4 to 8, wherein the SPICE life model includes at least one sub-circuit life model of a selected parameter; Determining, according to the SPICE lifetime model, initial values ​​of the selected parameters of the BSIM model corresponding to the electrical characteristic degradation characteristics of the target MOS transistor for fitting when no electrical stress is applied to the target MOS transistor; According to the SPICE lifetime model, the time required for the SPICE lifetime model corresponding to the selected parameter to decrease from the initial value to nine-tenths of the initial value after an electrical stress is applied to the target MOS transistor is determined, and the time is used as the lifetime of the MOS transistor.

Images